The Art of Contrarian Trading Part 3

This is the essence of herd behavior. The members of a herd find safety in numbers but only so long as the herd stays together in a group and acts together as well. An investment crowd holds together only so long as the price of its a.s.set(s) moves in the direction expected by the crowd. But this can happen only if crowd members respond almost instantly to the appropriate images and their implied suggestions.

A fascinating example of the suggestibility of stock market crowds developed during the late stages of the stock market bubble in the late 1990s. Companies with ".com" in their names were priced at a premium by the market, and several were motivate to change their corporate names for this reason. Of course, no value per se resides in a mere name, but the image image of a dot-com company was strongly a.s.sociated with stock market profits. This made every such stock a favorite of the bubble crowds of the late 1990s. Similar things happen in every stock market boom and ill.u.s.trate the important role that images play in investment crowds. of a dot-com company was strongly a.s.sociated with stock market profits. This made every such stock a favorite of the bubble crowds of the late 1990s. Similar things happen in every stock market boom and ill.u.s.trate the important role that images play in investment crowds.

Once an investment crowd becomes herdlike in its behavior, the price of its a.s.set becomes much more volatile. Where once daily price movements up or down seemed normal in magnitude, one instead sees daily price fluctuations in much wider bands. The market seems to soar or tumble with little rhyme or reason, and every tidbit of news, relevant or not, has an exaggerated effect.

A dramatic increase in price volatility is one of the signs that an investment crowd has reached maturity and that the information cascade supporting the crowd has become very fragile. With members of the crowd at their most suggestible, willing to accept both hopeful and fearful images at face value, every new bit of information has an exaggerated impact on the crowd and on the market price. Because price has become so volatile, the messages crowd members think they see in price changes will have ever more powerful effects on their willingness to stick with the crowd's theme. Any negative news, even if not a.s.sociated directly with the crowd's theme, can bring down its information cascade because it may trigger significant selling or buying of the crowd's a.s.set, perhaps by investors outside the crowd. The crowd's suggestible state guarantees that such price movements are highly likely to s...o...b..ll as members of the crowd successively take flight.

While dramatic increases in price volatility are signs of a mature investment crowd, the increased fragility of its information cascade also means that the disintegration of the cascade is almost impossible to predict ahead of time. In subsequent chapters we will take a closer look at the problem of identifying the imminent disintegration of investment crowds, a problem that every contrarian trader must be able to solve.

CHAPTER 6.

The Historical Context for Market Mistakes Exhausted investment themes * * clues that an investment crowd is mature clues that an investment crowd is mature * * look for the market mistake look for the market mistake * * market mistakes in the context of the market's history market mistakes in the context of the market's history * * only need a guesstimate only need a guesstimate * * look at market's price history and then at media content look at market's price history and then at media content * * this method focuses directly on the market mistake, not on fair value itself this method focuses directly on the market mistake, not on fair value itself * * price history more useful than estimates of earnings price history more useful than estimates of earnings * * regularity in emotional behavior of crowds regularity in emotional behavior of crowds * * data sources for historical prices data sources for historical prices * * beginners' mistakes beginners' mistakes * * how to identify a stock market bubble how to identify a stock market bubble * * examples of 1929, 1966, and 2000 examples of 1929, 1966, and 2000 * * how to identify undervaluation in the stock market how to identify undervaluation in the stock market * * examples of undervaluation examples of undervaluation * * the peak oil bubble the peak oil bubble MATURE INVESTMENT THEMES AND MARKET CROWDS.

Every contrarian trader must learn how to identify the point at which an investment theme has exhausted itself. At this juncture the information cascade has done its work and the market commitments of the investment crowd have grown about as large as is likely based on historical precedents. The a.s.sociated market crowd has reached maturity and is nearing the disintegration a.s.sociated with old age. The crowd has forced the market to make a valuation mistake. Price stands well above or below fair value. Soon economic forces and sharp-eyed value investors will begin the process of correcting the market's mistake. Once this process begins, the crowd will disintegrate and price will return rapidly to fair value, sometimes continuing well past that point to produce another valuation mistake in the opposite direction. This return to fair value is the investment opportunity that a contrarian trader seeks to antic.i.p.ate and exploit. For this reason the contrarian trader must be always vigilant, always on the lookout for mature investment themes and investment crowds.

Are there any clues that can warn us that a market might be the hands of a mature investment crowd? Since mature crowds force markets to make mistakes, to trade well above or below fair value, it makes sense to try to identify these market mistakes directly. I believe that every market mistake can be identified by putting it into the context of the market's historical price fluctuations. I believe that every market mistake can be identified by putting it into the context of the market's historical price fluctuations. I will explain the kind of computations needed to identify this context. I will explain the kind of computations needed to identify this context.

Of course any such method will at best yield only approximate answers. But this is not a problem for the contrarian trader. His a.s.sessments of whether the market is over- or undervalued will be only one input to his decision process. He only needs these guesstimates about possible over- or undervaluation to get him into the right valuation ballpark, to indicate the possibility that a mature market crowd may be causing a valuation mistake.

In this chapter I want to show you how to identify situations in which a market may be trading significantly above or below fair value. The novel aspect of the methods I explain is that they use only the market's price history to estimate the current position of fair value. No information about past dividends or profits or estimates of their prospective levels is needed. In a sense we will be extracting from the market's price history information about how fair value changes over time, as well as information about the size of typical market mistakes. Using this information, we will compute a rough estimate of the current fair value price.

Locating price relative to fair value is only the first step in the process of spotting a potential contrarian trade. In subsequent chapters I will show you how to combine this valuation information with an a.n.a.lysis of media content. The stories appearing in the electronic and print media give you a way to observe information cascades as they proceed and to a.s.sess the strength of a.s.sociated market crowds. It is the combination of a potential valuation mistake together with solid, affirmative evidence that this mistake is caused by a mature market crowd that is the key to a profitable contrarian trade.

MISTAKES VERSUS FAIR VALUE.

In Chapter 5 I explained why the disintegration of a bullish investment crowd usually results in the birth of bearish one, and vice versa. The cycle of alternating bullish and bearish crowds is a universal feature of financial markets. This cycle of birth and death is responsible for the mistakes markets make-for prices being alternately too high and then too low relative to fair value.

But what is fair value? We have already seen that economists compute the fair value price of an a.s.set by discounting to the present the estimates of the a.s.set's future cash payouts (dividends in the case of common stocks). Sadly, this wonderful theoretical concept is not very useful as a yardstick for the contrarian trader. I have found it more useful to try to identify market mistakes directly instead of focusing my attention on estimates of fair value. In fact, I think this can be done by studying a market's historical price behavior. Why might this be so? The market prices we observe result from the confluence of two kinds of investment activity. The first arises from a rational calculation of value based on purely economic and statistical considerations. The second arises from the psychological and emotional character of investment crowds and information cascades. I think the latter shows far more regularity than the former. Moreover, this regularity in psychological and emotional behavior is very difficult to exploit, because the typical investor is a part of this phenomenon, not apart from it. So we would expect to see such regularity showing more persistence instead of being destroyed by the few traders capable of exploiting it.

The method I have adopted uses historical price data to identify situations in which the market is trading significantly above or below fair value. It does this by tabulating the extent and duration of a market's typical swings from high to low and low to high. My working hypothesis is that at historical high points the market was overvalued and at historical low points it was undervalued. I expect to see consistency in the duration and the extent of the swings from undervalued to overvalued conditions and back again. Such consistency I regard as an empirical fact about the psychological life cycle of crowds. As such, it should be quite independent of the specific economic environment of the time.

Because this method rests on empirical estimates of the effects of crowd psychology on markets, it can be expected to work only if a market has enough partic.i.p.ants to support the formation of large investment crowds, especially crowds of people who are not normally attentive to financial markets. For this reason this approach works best with the stock market averages, the bond markets, and some commodity markets (those that attract public attention). It is much less useful in estimating when the price of an individual stock is high or low relative to fair value.

MARKET DATA SOURCES.

Before going further I think it is useful to identify some sources for the market data that you will need to do the calculations I will describe. These sources have changed a lot during the 40 years I have been partic.i.p.ating in the financial markets. Years ago you had to collect this data from newspapers, or, if you were lucky, you might have been able to purchase some of it already tabulated from some obscure source you happened to hear about. But nowadays computers and the Internet have made the price data aspect of contrarian trading so much easier. The Internet has become a tremendously useful data repository for the contrarian trader. Having computational power at hand on your desktop is a tremendous laborsaving device when a.n.a.lyzing this data.

At present there is no single comprehensive data source for the information you will need, but as computing power and communication resources expand I expect this to change over the next 10 years. Whenever I need data I don't already own I often can locate a source by using Google or some other search engine. Here are three online data sources that I have found very valuable. They will no doubt be either outdated or defunct in a few years. No matter. I am sure you will still be able to use online search tools (like Google) to identify new data sources.

The first one is Yahoo! Finance (http://finance.yahoo.com). This data is comprehensive and free! Once you have identified a stock or stock market index, Yahoo! offers access to daily, weekly, or monthly historical data, which can be downloaded as an Excel spreadsheet. Excel itself has charting capabilities that you can use for a preliminary data a.n.a.lysis.

There are a number of sites that offer downloadable data for stocks, stock market indexes, currencies, and commodities on a subscription or one-time-purchase basis. Here are two of the most widely used: Commodity Systems Inc. (www.csidata.com) and MetaStock (www.equis.com).

THE DEADLY MISTAKE.

The goal of every contrarian trader is to beat the market-to earn an investment return that exceeds the return earned by the buy-and-hold strategy. (Sophisticates might want to make appropriate adjustments for portfolio risk when making this comparison.) There is only one kind of mistake a contrarian trader can make that will prevent him from matching or exceeding the buy-and-hold return: being out of or even short the market when it rising. (Being fully invested when the market is falling does no harm to his performance relative to buy-and-hold.) Sadly, it is precisely this mistake that novice contrarian traders are most p.r.o.ne to make. Why? I think there are two reasons. First, periods of market overvaluation tend to be of longer duration than periods of undervaluation, so there is simply a lot more opportunity to make this kind or error. Second, there is much more variability in the size of the mistake a market can make in the direction of overvaluation. These two factors work together to make the task of identifying periods of overvaluation much more difficult and much more risky than the task of identifying periods of undervaluation. Combine the technical difficulty of identifying periods of overvaluation with the novice's eagerness to show the world how clever he is, and you have a reliable recipe for disaster.

I am convinced that the difficulty of identifying periods of market overvaluation means that the contrarian trader, especially the novice, should focus his efforts on identifying and exploiting periods of market undervaluation. I'll have more to say about this matter in Chapter 11, in which I discuss "The Grand Strategy of Contrarian Trading." In the meantime, I think the way to proceed is to adopt an asymmetrical approach to this problem by developing criteria for identifying overvaluation that are more stringent than those for identifying undervaluation.

WHEN IS THE STOCK MARKET (EXTREMELY) OVERVALUED?.

I take it as axiomatic that the contrarian trader wants to avoid being caught in the stock market debacles of the sort that followed the bubbles of the 1920s and the 1990s. These were two of the three distinct instances of extreme stock market overvaluation during the past 100 years. The a.s.sociated stock market tops occurred in 1929, 1966, and 2000. Notice that these market tops were separated by 37 and 34 years, almost one and a half generations in each case. This sort of extreme overvaluation does not occur very frequently, perhaps once or at most twice in the lifetime of a contrarian trader. In these three cases the subsequent drop in the inflation-adjusted Dow Jones Industrial Average amounted to 87 percent to the 1932 low, 62 percent to the 1974 low, and 42 percent to the 2002 low. The 2007 high in the Dow was substantially above its 2000 peak but according to my a.n.a.lysis (see chapters 14 and 15) was not a.s.sociated with a stock market bubble. Nonetheless, from its 2007 high to its 2008 low the Dow dropped an inflation-adjusted 51 percent. Being able to sidestep declines of this magnitude should be a goal of every contrarian trader. But how might this be done? Are there any common features in the price fluctuations preceding these important stock market tops that might have warned of gross overvaluation?

Here is the way I try to answer this sort of question. Each of the first three market tops ended a bull market that had not been interrupted by any decline of as much as 30 percent in nominal terms or by any decline that lasted as much as eight months from high to low. These were the bull markets of 1921-1929, of 1949-1966, and of 1987-2000. In inflation-adjusted terms, prices advanced 496 percent, 334 percent, and 346 percent respectively during these bull markets. These advances lasted 8, 17, and 13 years respectively. Moreover, only the first of these three bull markets began from the preceding crash low. The start of the 1949-1966 bull market occurred after 17 years had elapsed from the 1932 low, while the 1987-2000 bull market started 13 years after the 1974 low.

Let's see if we can combine these observations to make some rough-and-ready guesses about the top of the next bubble. One possibility is that the low we are now seeing develop in 2008 will be historically a.n.a.logous to 1921. I think it far more likely that the 2008 low will play this role than will the 2002 low. Why? As we will see later, the bearish crowds of 2008 are very powerful, indeed much more well-developed than were the bearish crowds of 2002. I don't think we would see such strong bearish information cascades occur in the midst of a bull market leading to a bubble top. So let's suppose that 2008 is a.n.a.logous to 1921. In this case we would expect a top to develop around 2016 after an advance of at least 250 percent in real terms from the 2008 low, which I will take as 7,552 in the Dow Industrials. a.s.suming inflation of 3 percent per year for the next eight years, the bubble top would occur near 33,000 in the Dow around 2016.

Why did I use 250 percent instead of the larger numbers a.s.sociated with the three previous bubble bull markets? I wanted to be conservative, yet stay in the same ballpark. Half of the biggest advance is 248 percent, and this is not too much less that the other two. But feel free to do your own guesstimating!

Why 2016? Well, the obvious answer is that if 2008 is like 1921, then since the 1921-1929 bull market lasted eight years this one should, too. But actually I would be very surprised if the next bubble top occurred so soon after the 2000 top. I think it more likely that if a bubble bull market is actually starting in 2008, then it will probably last between 13 and 17 years, putting the top sometime during the years 2021 to 2025. This would be long enough after the 2000 top for a new generation of investors to grow up and make the same mistakes their parents and grandparents made during the bubble of the 1990s.

If 2008 is not the start of a bubble bull market, then the top of the next bubble will be pushed out further in time and much higher in price. The next bubble bull market would then have to start some number of years after 2008 and from a price level higher than the 2008 low. Once you think you have seen the starting point of the bubble bull market, you can use the precedents cited earlier to guesstimate the level and timing of its ultimate top.

I have just ill.u.s.trated how a contrarian trader can identify situations of extreme overvaluation in the stock market. But such bubble tops are infrequent events. There are many less dramatic bull market tops occurring in the intervals between major bubble tops. These less dramatic bull market tops are seen on average every four or five years. Can the method I have just ill.u.s.trated be adapted so as to help identify these situations of less extreme overvaluation? I believe it can. In fact, I'll tell you about two simple rules of thumb I use to do this. But first a warning: I think only an expert, experienced contrarian trader should attempt to identify these less extreme periods of overvaluation. For the reasons I cited earlier, the costs of making a mistake-of antic.i.p.ating a top when none occurs-can be very substantial. I don't think the effort is worthwhile for a newcomer to contrarian trading.

That being said, here are my two rules of thumb for identifying moderate periods of overvaluation. First, a typical bull market in the U.S. stock market averages about 24 months from low to high. Second, a typical bull market carries the averages upward about 65 percent from low to high. These are two useful facts, but of course the real difficulty lies is determining whether any particular bull market will be typical. I'm afraid I will have to leave this as the subject of another book.

WHEN IS THE STOCK MARKET UNDERVALUED?.

I think that the typical contrarian trader would do well to specialize in identifying and exploiting periods of market undervaluation. Periods of undervaluation tend to be of much shorter duration than periods of overvaluation. Moreover, bearish market crowds tend to be more vocal, emotional, shorter lived, and easier to identify that bullish crowds. Finally, the relentless upward path of economic progress in free market economies stacks the deck in favor of trying to exploit undervalued situations. The secular upward trend works to mitigate any mistakes the contrarian trader may make in his effort to identify periods of undervaluation and the a.s.sociated bearish crowds.

How frequently do the stock market averages in the United States become undervalued? The answer depends on just how big a valuation mistake we are talking about.

The really big undervaluation mistakes in the U.S. stock markets occurred in 1932 and 1982. I say this because the 1932 low a.s.sociated with the Great Depression ended a drop in stock prices of 85 percent in real terms from the 1929 high. The 1982 low ended a drop of about 74 percent in real terms from the 1966 high. You might wonder about the value of Tobin's q ratio at these lows (see Chapter 5 for a discussion of Tobin's q ratio). At both of these lows the q ratio was 0.5 or below and thus indicated that the stock market was 50 percent undervalued relative to the replacement value of the real a.s.sets owned by the companies in the market average.

Was either the 2002 low or the 2008 low a comparable valuation mistake? I don't think so, and here are my reasons. First, the drop in the inflation-adjusted Dow was just 42 percent from the 2000 high and 51 percent from the 2007 high. These numbers are relatively high in a historical context, but neither is high enough in my judgment to make them comparable to the 1932 and 1982 situations. Second, at the 2002 low the q ratio was 1.5 and at the 2008 low the q ratio stood near 1.0. Neither reading is comparable to the levels of 0.5 seen in 1932 and 1982. (See Chapter 16 for more discussion of stock market valuations at the 2008 low.) I think the combination of these two facts means that at the 2002 and 2008 lows, the stock market was not nearly as undervalued as it had been at the 1932 and 1982 lows.

We see then that valuation mistakes of the 1932 and 1982 variety may occur only once every 50 years, so the next one may not be due until 2032. Obviously, the contrarian trader has to look for less extreme examples of undervaluation.

For more than 100 years the U.S. stock market has established bear market low points on average every four years, with perhaps 70 percent of these successive low points being separated by an interval of three to five years. Thus during a typical investing lifetime of 40 years an individual would on average encounter about 10 of these opportunities. I like to think of bear markets as falling into three categories. A short bear market would typically drop the stock market averages 20 to 25 percent and last about eight or nine months from high to low. A normal bear market would drop the averages about 35 percent and last about 18 months from high to low. An extended bear market would drop the averages about 45 percent or more and last 30 months from high to low.

Extended bear markets are rare. The 2000-2002 drop in prices was an extended bear market lasting 31 to 33 months, depending on which average you use as a yardstick. The Dow dropped 39 percent (no inflation adjustment) during that time, while the S&P 500 dropped 50 percent. The bear market that began in 1937 didn't end until 1942, a span of 61 months during which the Dow dropped 52 percent. The preceding extended bear market was the 1929-1932 event, during which the Dow fell 89 percent in nominal terms in a 34-month period.

Normal bear markets were also rather unusual events during the twentieth century. The 1981-1982 drop in the Dow lasted 16 months but carried the average down only 25 percent. The 1976-1978 bear market dropped the Dow 28 percent over an 18-month period. The 1973-1974 bear market dropped the Dow 46 percent over a 21-month period. The 1968-1970 bear market saw the Dow drop 36 percent over 17 months. Notice that all of these bear markets occurred during the adjustment period in which the stock market moved from an extremely overvalued condition in 1966 to an extremely undervalued one in 1982.

The most common bear market-and the one I think contrarian traders should become skilled in exploiting-is the short bear market that lasts about eight or nine months and drops prices about 20 to 25 percent. Short bull markets are a phenomenon a.s.sociated with extended stock market advances, such as those of 1921-1929, 1942-1966, and 1982-2000. I'll have a lot more to say about how these short bull markets can be exploited when I describe "The Grand Strategy of Contrarian Trading" in Chapter 11. Here let me cite some typical instances of short bear markets.

There was an unusually brief bear market a.s.sociated with the Long Term Capital Management crisis of 1998. Then the Dow dropped 22 percent in barely two months. Another unusually brief bear market was a.s.sociated with preparations for the first Gulf War of 1991. From a high in July 1990, the Dow dropped 23 percent in about two and a half months. A third unusually brief bear market occurred in 1987 for no obvious reason. Then the Dow dropped 37 percent in three and a half months, with most of the decline occurring on a single day, October 19, 1987.

Notice that these were three consecutive short bear markets, and each was of very subnormal duration compared with the eight- to nine-month average for the species. Perhaps this was a clue that a bubble bull market was under way during 1987-2000. A similar situation developed during the bubble bull market of 1921-1929, which was interrupted by subnormal bear markets in 1923 and 1926. This may prove to be a clue that a bubble bull market is under way the next time we see such a phenomenon develop.

Short bear markets were prevalent during the 1942-1966 stock market advance. There was a short bear market in 1946, which dropped the Dow 25 percent in five months. Another occurred in 1956-1957. It lasted 18 months (with 99 percent of the decline occurring during the final three months) and dropped the Dow 20 percent. The short bear market of 1960 dropped the Dow 18 percent in nine months, while the short bear market of 1962 dropped that average 29 percent in six months. Finally, the 1966 short bear market carried the Dow down 27 percent in eight months.

What do these statistics tell us? Every time the stock market averages drop 20 to 25 percent from a high and three to five years have elapsed since the preceding bear market low point, we must suspect that the market has entered a zone of undervaluation. This is the time to look closely for evidence that a bearish market crowd has become mature and is about to begin its process of disintegration. I'll have much more to say about how to do this in subsequent chapters.

THE PEAK OIL BUBBLE.

Just so you won't think that the tabulation method I have ill.u.s.trated for the stock market averages can't be applied elsewhere, let's take a look at a completely different market, that for crude oil. In July 2008 crude oil sold at $147 per barrel. (This is being written in August 2008.) I think a very mature investment crowd has grown around the theme of peak oil and is about to start its process of disintegration. Recall that believers in the peak oil theme a.s.sert that worldwide production of crude oil soon will or already has reached the highest level it will ever attain, and supplies of petroleum will soon start contracting. The implication is that crude oil prices have nowhere to go but up.

Let's take a look at what a tabulation of the historical data on oil prices can tell us. Before the first Arab oil embargo in 1973, crude sold for $2.35 per barrel. By 1981 it had risen to a high of $39 before declining. In 1998 it reached a low price of $11 and, as mentioned, it traded at $147 in July 2008. Here is the way I look at this price history.

In eight years from 1973 to 1981 crude oil rose 710 percent in inflation-adjusted terms. From 1981 to 1998, an interval of 17 years, it fell 84 percent in inflation-adjusted terms. In 10 years from 1998 to 2008 crude oil advanced 910 percent in inflation-adjusted terms. Note that the duration of this 910 percent advance exceeds the duration of the 1973-1981 advance by only two years. The inflation-adjusted price rises are comparable also: 910 percent versus 710 percent. This is a good reason to think oil was very overvalued at $147, and, as I have mentioned, a very vocal, bullish investment crowd has formed around the peak oil theme.

The prognosis is that crude oil is about to begin a drop in its inflation-adjusted price that may well last 17 years. The 1981-1998 drop carried the price down 84 percent. If inflation averages 3 percent over the next 17 years, then an inflation-adjusted drop of 84 percent from the $147 level would put the price of crude at $38 in 2025. Remember that all of these numbers are adjusted for an expected level of inflation that I have guessed will be 3 percent per year for the next 17 years. You will have to update this guess with actual inflation numbers to get updated versions of these price projections.

CHAPTER 7.

How Crowds Communicate The information conveyed by information cascades * * two types of information two types of information * * beliefs and stories that make up the investment theme beliefs and stories that make up the investment theme * * the role of the media the role of the media * * the confirming action of market prices the confirming action of market prices * * 1994-2000 bubble as an example 1994-2000 bubble as an example * * the potent brew of emotional persuasion the potent brew of emotional persuasion * * the role of the ma.s.s media the role of the ma.s.s media * * telling people what they want to hear telling people what they want to hear * * media amplification media amplification * * monitor the media midwives monitor the media midwives * * what media sources to use what media sources to use * * print media print media * * Internet media Internet media * * blogs blogs * * television and talk radio television and talk radio * * flipping and the real estate bubble flipping and the real estate bubble * * Jay Leno Jay Leno * * tomorrow's media will be different from today's tomorrow's media will be different from today's * * be flexible and alert to changes be flexible and alert to changes * * monitoring the markets monitoring the markets * * data sources I use data sources I use * * here, too, things will change here, too, things will change * * the value of historical accounts the value of historical accounts WHAT DO INFORMATION CASCADES TELL INVESTORS?.

In the first part of this book we developed a theory of investment crowds. Investment crowds are responsible for the many mistakes markets make. Crowds are an inevitable feature of our social environment. They arise naturally from our desire to form social bonds and make use of the information about our world that is transmitted via these human connections.

To recognize an investment crowd, one has to understand the communication processes that enable the investment crowd to form and the nature of the information these processes convey. Remember that an investment crowd arises from an information cascade. We want to learn how to recognize a cascade in action and to identify the sort of information it conveys. If we can do this, we have a good chance of being able to identify an ongoing cascade and to watch the a.s.sociated crowd as it grows and eventually forces a market mistake.

The information transmitted in a cascade is of two types. First are the specific beliefs and facts conveyed by the founding members of the crowd to potential new members. This is the stuff of the crowd's investment theme. It is a logically coherent story but one intended to trigger an emotional response rather than scientific agreement. Typically, this first sort of information is transmitted through the electronic and print media, although person-to-person contact often plays a role as well. This transmission mechanism is open to public view. The transparent nature of the media gives us the opportunity to watch the information cascade develop simply by keeping track of the number and intensity of the messages.

However, information that merely elaborates on an investment theme would not by itself be persuasive were it not accompanied by a second, more dramatic piece of persuasion: a big change in the market price in the direction predicted by the theme that has enriched a small but visible group of investors (or made them a lot poorer!). Indeed, it is the prospect of getting rich or the fear of getting poor that creates the emotional power the investment theme needs to attract a crowd of investors.

To ill.u.s.trate this last point, let's recall the dot-com bubble of 1994-2000. Does anyone imagine that investment themes like the new economy and the transforming power of the Internet would have been taken seriously if the stocks of America Online (AOL), Yahoo!, and other dot-coms hadn't first staged impressive advances, if a steady stream of initial public offerings hadn't made their lucky buyers rich? No, I don't think so, either. But once investors saw how much money they might have made had they picked up these themes early enough, they were hooked. The logic of the investment theme was simply icing on the cake, a convenient justification for joining the crowd.

We see then that economic and business information and forecasts, the dreams and fears of investors, and the movements of prices in the financial markets all have important roles to play in the creation of an investment crowd. They all are part of a potent brew of persuasion that manifests itself in an information cascade. But a dramatic change in the price of some a.s.set always comes first. This is the prime mover and seed of the information cascade that will ultimately create an investment crowd. Only after the fact is the price change ever explained by appeal to economic and business developments. Even then, these explanations are only dry, unemotional facts. By themselves they will not persuade an individual who has no time or skill for scientific a.n.a.lysis. To do their work, these explanations are leavened with emotional appeals to greed, to fear, and to the natural human desire to get ahead of one's fellows. Then and only then will the information cascade gather momentum and create a new investment crowd.

THE ROLE OF THE Ma.s.s MEDIA.

There would be no investment crowds in the modern world without the print and electronic media. Indeed there is a fascinating symbiotic relationship between the media and investment crowds. When people get rich or poor because of a big change in the price of some a.s.set or commodity, the media have a story to tell and one they can sell. Remember that the media business is all about telling people what they want to hear so that advertisers will be willing to pay to reach an attentive audience. Stories like this attract readers and thus advertisers. But then an interesting thing happens. Because the media business is very compet.i.tive, the same story is picked up by other media outlets and in this way spreads far and wide. These human stories of success or failure become widely known. Success breeds imitators while failures encourage others to share their stories of distress. This process in turn generates even more stories of the same sort, and these stories in their turn encourage more imitators. The entire process amplifies the effects of the original story. What begins as a whisper, barely audible even to the careful listener, eventually becomes a cacophony of communication among the media and its newborn offspring, the investment crowd.

This is indeed a fortunate situation for the contrarian trader. Because the print and electronic media are midwives to the birth of investment crowds, we have the opportunity to watch crowds develop from toddlers to mature adults just by monitoring media content. The particular methods I use to interpret the significance of this content will be the subjects of subsequent chapters. For now I want to discuss the media sources I monitor to help me identify information cascades in operation.

As I write this in 2008, the print media are still the most important lines of communication for the information cascades that build investment crowds. I check the New York Times New York Times and the and the Chicago Tribune Chicago Tribune every morning for interesting stories about the economy, finance, and business. Front page stories are especially noteworthy. Any major metropolitan newspaper can be used in the same way. In fact, if you read a local, small market paper you will occasionally find stories on these topics that have a more personal, local flavor and that can be just as significant as a similar one in a major newspaper. every morning for interesting stories about the economy, finance, and business. Front page stories are especially noteworthy. Any major metropolitan newspaper can be used in the same way. In fact, if you read a local, small market paper you will occasionally find stories on these topics that have a more personal, local flavor and that can be just as significant as a similar one in a major newspaper.

After daily newspapers, the next most important means of monitoring information cascades are weekly and monthly newsmagazines. Those of general interest like Time Time and and Newsweek Newsweek are the most important ones to watch, because they generally do not devote much s.p.a.ce to economic and business subjects. When they do, especially if the story appears on the cover, you have a significant and unusual situation, which tells you that the a.s.sociated investment crowd is probably a very important and substantial one. are the most important ones to watch, because they generally do not devote much s.p.a.ce to economic and business subjects. When they do, especially if the story appears on the cover, you have a significant and unusual situation, which tells you that the a.s.sociated investment crowd is probably a very important and substantial one.

Other general interest and politically oriented magazines can also be used in this way. Over the years I have used stories from U.S. News & World Report U.S. News & World Report, the New Yorker New Yorker, New York New York magazine, the magazine, the New Republic New Republic, and Harper's Harper's (to name just a few). Any magazine of significant circulation can be used for this purpose. In fact, for this reason I periodically visit my local big-box bookstore (in my case Barnes & n.o.ble) because it has a large magazine section. I check out the cover stories for all the magazines displayed, even the ones I don't normally read. I do this whenever I think an information cascade has reached critical ma.s.s. It is amazing how easily cascades can be identified using just this simple device. (to name just a few). Any magazine of significant circulation can be used for this purpose. In fact, for this reason I periodically visit my local big-box bookstore (in my case Barnes & n.o.ble) because it has a large magazine section. I check out the cover stories for all the magazines displayed, even the ones I don't normally read. I do this whenever I think an information cascade has reached critical ma.s.s. It is amazing how easily cascades can be identified using just this simple device.

So far I have left out business-oriented magazines like Fortune Fortune, BusinessWeek BusinessWeek , and the , and the Economist Economist. These can be very valuable, too, especially when a cover story is devoted to a recent market event. But you must keep in mind that these sources generally have many cover stories devoted to the economy and business, so such covers don't carry the same weight and significance as do the covers of Time Time or or Newsweek Newsweek.

You have no doubt noticed that I have left the Wall Street Journal Wall Street Journal off my list of print media sources. There is a good reason for this. The off my list of print media sources. There is a good reason for this. The WSJ WSJ's primary focus is on business and finance. Consequently it is difficult to identify the importance of a story from its positioning on the newspaper's front page. The only WSJ WSJ stories I pay attention to are those describing in some detail the workings of an investment crowd that I have already identified from other sources. I use the stories I pay attention to are those describing in some detail the workings of an investment crowd that I have already identified from other sources. I use the WSJ WSJ as a corroborating source, not a primary one. as a corroborating source, not a primary one.

Over the past 10 years, an important compet.i.tor to the print media has arisen. I call these the Internet media because they are delivered primarily through the Web browser you use to access online resources. Every one of the print media nowadays has an electronic edition that can be accessed over the Internet. The Web content published by the print media changes from hour to hour and from day to day, so it is more difficult to track and record. The same is true of content that appears on sites like Market.w.a.tch or Bloomberg. The basic thing to keep in mind is that the changeability of this content makes it relevant to very short-term (days, not weeks) crowd behavior and thus is of limited use for gauging information cascades. However, one can get a sense of what these sites regard as news over periods of months just by paying daily attention to them. For example, stock market industry groups come into and fall out of favor and so are mentioned more or less frequently over a period of months. The same is true of individual stocks or commodities. Remember that the Internet media want to keep people's interest and therefore highlight stories they think will interest the biggest segment of the public. So their editorial choices tell you something about their editors' perceptions of the markets that hold the most interest for their readers.

I also follow many blogs devoted to investing, finance, and economics. I won't name them here, simply because my list changes frequently-the lifetime of a typical blog is only a couple of years (or even less). However, general interest blogs like Instapundit as well as blogs with a strong political point of view can be very useful reading. You can be sure that any economic or financial market that attracts intense public interest will find mention on these blogs, together with links to other relevant Internet material. Keeping up with the content of a diversified list of blogs is an essential activity for the contrarian trader.

There are other electronic media that are important but not distributed on the Internet (at least not yet). These are network and cable television, talk radio, and the movies. Here, too, the basic rule that the media are profit-seeking ent.i.ties that have to cater to people's interests and prejudices serves the contrarian trader well. Generally, programming specials can tell you a lot about what people hope for or fear. Even more significant are new TV series that appear just as an a.s.sociated crowd is about to begin its disintegration process. The most recent example of this sort occurred in 2005-2006 with the appearance of several cable shows devoted to house flipping (i.e., buying a house, fixing it up, and then selling it for a profit, all within a couple of months). The first episodes of these shows marked almost the exact top of the real estate market and preceded the subprime market debacle of 2007-2008. As I write this (July 2008), a new reality series has appeared on cable TV. It's called Black Gold Black Gold and depicts the day-to-day activities of the roughnecks who man the oil rigs on three separate West Texas drilling projects. For future reference, note that the price of crude oil is currently $143 per barrel. and depicts the day-to-day activities of the roughnecks who man the oil rigs on three separate West Texas drilling projects. For future reference, note that the price of crude oil is currently $143 per barrel.

Another ill.u.s.tration of how TV shows can give useful information to the contrarian trader occurred in July 2002, just as the S&P 500 index was approaching its bear market low at 768 (its low that month was 771). I am a fan of the Tonight Show with Jay Leno Tonight Show with Jay Leno. For the first time in my memory Leno started telling jokes about the stock market crash and the economic recession (which had in fact ended in late 2001) in his monologues that month. At the time I found this strong corroborating evidence that the bear market crowd of 2002 was about to disintegrate and that the stock market's low was at hand.

A WORD ABOUT PERSONAL FLEXIBILITY AND THE FUTURE OF MEDIA.

I am writing this chapter in July 2008, but I hope that this book will be of interest to future generations of investors as well. The phenomenon of investment crowds is a timeless one. One can expect the nature of ma.s.s media and their means of delivery to evolve in unpredictable and surprising ways over the years ahead. I don't doubt that my future readers will find the preceding discussion antiquated in its specific details.

However, the general principles by which one identifies information cascades are timeless. To apply them, one needs to monitor the specific forms of ma.s.s communication of one's era. The forms and technology used by media will evolve, so a contrarian trader must be constantly involved in a media watch and be flexible and willing to adjust his procedures as needed. Some media will grow in popularity, whereas others will decline. It is the contrarian trader's job to keep abreast of this evolutionary process, of the constantly changing nature of modes of ma.s.s communication. It will be the content of these media that enables the contrarian trader to identify ideas and themes that drive information cascades.

MONITORING THE MARKETS.

While the ma.s.s media play an essential role in helping the contrarian trader identify information cascades and the investment crowds they create, these media are not the only grist the trader can or should use in his trading mill. A contrarian trader must have extensive knowledge of the historical behavior of financial markets. He must be able to put the markets being pushed by investment crowds into a quant.i.tative context. Ideally, he wants to be able to estimate how big a mistake a market may be making and to get some sense of when or whether the a.s.sociated investment crowd is still growing stronger or is instead on its last legs.

This is not easy to do for even the most skilled contrarian trader. But it is impossible to do at all without a substantial body of historical market data. In subsequent chapters I'll discuss specific ways of using this historical data to monitor the status of an investment crowd. In this chapter I want to confine my attention to sources of data and to the kind of data one wants to monitor and use.

At the moment the single most useful source for historical stock market data is the free Yahoo! Finance web site, from which one can download data in spreadsheet format. As the years pa.s.s this source will grow in value. But remember that the Internet and information storage technologies are evolving rapidly. So 10 years from now the situation may change and Yahoo! may no longer be a good data source. It is the job of the contrarian trader to stay abreast of informational resources so that when he needs historical information he knows where to look.

Another good source of historical information in chart-based format currently is the online Chart Store web site. For an annual subscription fee one can view historical charts of stock market averages, interest rates, foreign markets, currencies, and commodities.

A third source for chart-based stock market data that I use at the present time is StockCharts.com, another subscription-based web site; it has a vast number of charts kept up to date in various user-controlled formats.

The only thing I am sure about is that 10 years from now I will probably be using different sources for my historical market data than I use now. The financial world is constantly evolving and with it the nature of the information resources one must use to keep abreast of events. It is the job of a contrarian trader to know where he can obtain relevant information even as the world changes around him.

STUDYING THE HISTORY OF BUBBLES AND CRASHES.

There is another source of information that can help the contrarian trader place current information cascades and investment crowds in their proper historical context. These are the historical accounts of past market bubbles and crashes. These historical accounts can be entertaining, but they always provide food for thought and opportunities for critical a.n.a.lysis as well. Later in Chapter 16 I shall suggest a number of books that I have found useful and interesting for this purpose. No doubt many more will appear as time pa.s.ses. History is always being written and often revised in the light of new information.

When I got my start in the stock market in 1965 the great crash of 1929 had occurred 36 years earlier. It seemed to me then as distant as the time of the Roman Empire. I like to think that even as a novice I was influenced by the experiences of other speculators, which I had learned about from reading historical market accounts. As I write these words, the market events of the intervening 43 years have also been recorded for posterity by many authors. To today's novice contrarian trader these events no doubt seem to be ancient history, too. But my personal experience as a market partic.i.p.ant during the past 43 years informs everything I do as a contrarian trader. It is essential for every contrarian trader to build his technique upon a body of experience, even if at the beginning of his career it is only secondhand. The only access the novice has to the experience of these market events is through these historical accounts. Make use of them.

CHAPTER 8.

Constructing Your Media Diary Skill of a contrarian trader * * from deduction to action from deduction to action * * media diary aids a.n.a.lysis and strengthens trader's belief in his deductions media diary aids a.n.a.lysis and strengthens trader's belief in his deductions * * a trading tool and a training tool a trading tool and a training tool * * your diary will be your anchor to windward your diary will be your anchor to windward * * how my diary made a difference in 2002 how my diary made a difference in 2002 * * a wonderful learning tool a wonderful learning tool * * put your diary in a notebook put your diary in a notebook * * cut and paste cut and paste * * what media content is relevant what media content is relevant * * excerpts from my diary in 2005 excerpts from my diary in 2005 * * excerpts from my diary in 2006 excerpts from my diary in 2006 * * magazine covers magazine covers * * examples from 2006 examples from 2006 GAINING THE EDGE.

A contrarian trader's edge arises from two unusual skills. The first is his ability to consistently identify investment crowds and their investment themes. The second is his capacity for making rational deductions from these observations and translating these deductions into buying or selling for his portfolio in the markets. It is relatively easy to acquire the first skill. But it is the second that is the more difficult and yet the more essential one. It is his ability to translate observation and knowledge into profitable investment actions that distinguishes the contrarian trader from the run-of-the-mill devil's advocate or naysayer.

Your media diary plays an essential role in helping you develop both these skills. It will be a contrarian trading tool as well as a contrarian training tool. It will contain all the material you need to identify investment crowds, their investment themes, and the emotional state of the crowd. More important, by studying your diary in the light of subsequent market action, you will gain confidence in your deductions and will strengthen your ability to act contrary to the theme of the crowd at the right time. Your diary will be your anchor to windward on the sea of finance against its constant turbulence of boom and bust.

Let me remind you again that it is very, very, very very hard to be a contrarian trader. If you are to succeed, you will need the help of a well-maintained, up-to-date media diary. It will help you resist the siren songs of the media, which are the princ.i.p.al means of communication for investment crowds. It will teach you to float above the panic that develops during plunging markets and the euphoria during soaring ones. It will help you to shrug off the scorn and derision of crowd members who discover you are betting against them. The rest of the world will think you are at best rash and at worst stupid; and on those occasions when your trades don't work out, the "I told you so's" will be deafening and very hard to endure. But your media diary will be there to offer objective evidence that you made the reasoned choice despite any unfavorable result. The emotional world of any contrarian trader is a turbulent one, and emotional turbulence is the enemy of clear thinking. It will be your media diary that will help you to navigate these turbulent emotional waters successfully. hard to be a contrarian trader. If you are to succeed, you will need the help of a well-maintained, up-to-date media diary. It will help you resist the siren songs of the media, which are the princ.i.p.al means of communication for investment crowds. It will teach you to float above the panic that develops during plunging markets and the euphoria during soaring ones. It will help you to shrug off the scorn and derision of crowd members who discover you are betting against them. The rest of the world will think you are at best rash and at worst stupid; and on those occasions when your trades don't work out, the "I told you so's" will be deafening and very hard to endure. But your media diary will be there to offer objective evidence that you made the reasoned choice despite any unfavorable result. The emotional world of any contrarian trader is a turbulent one, and emotional turbulence is the enemy of clear thinking. It will be your media diary that will help you to navigate these turbulent emotional waters successfully.