The Art of Contrarian Trading Part 1

The art of contrarian trading.

How to profit from crowd behavior in the financial markets.

by Carl Futia.

Preface.

Why is it so difficult to beat the stock market? It is easy to see that the market gives us plenty of chances to buy low and sell high. Just look at the history of the past 10 years, 1998 to 2008. During that time the Standard & Poor's (S&P) 500 index has fluctuated between 752 and 1,565. There have been five distinct, substantial swings across this range. The brief panic in 1998 arising from the Russian credit default and the failure of a big hedge fund, Long Term Capital Management, dropped the S&P nearly 20 percent, from 1,187 to 957. Those fears quickly evaporated, and the subsequent climb in prices capped a stock market bubble that was unprecedented in the financial history of the United States. The S&P rose to a high of 1,527 in March 2000, an advance of nearly 1,400 percent from its 1982 low of 102.

The biggest thrills of this stock market roller-coaster ride were yet to come. The S&P dropped nearly 50 percent during the following two years. At the index's October 2002 low of 777, investors gasped at the shocking collapse of the Internet stocks and feared that corporate accounting statements were meaningless. But the downward rush in stock prices during the preceding two years seemed to generate just the momentum needed to push the market back up to the top of its next hill. During the subsequent five years the S&P more than doubled to a closing high in October 2007 of 1,565.

As the stock market began to edge downward from its 2007 peak, no one could imagine the terrors that lay ahead. Within a year the panic of 2008 had destroyed financial inst.i.tutions around the world. The rest teetered on the edge of collapse. On November 20, 2008, the S&P closed at 752 after careening downward 52 percent from its high a year earlier. Many people feared that even worse was to come.

In this book I tell you why it is so difficult for the average investor to profit from these roller-coaster swings in stock prices. I explain why it is almost impossible to consistently buy low and sell high and thus do better than the benchmark strategy of the buy-and-hold investor. Along the way I hope to help you make an informed choice of your personal investment strategy.

You may decide that attempting to beat the market is not really a good choice for you. The emotional strain involved may just not be worth the effort. Self-knowledge like this is invaluable, worth far more than the price of this book.

Or you may choose to learn the art of contrarian trading. If so, I think you have chosen a difficult path, but I also think you have in your hands the only book in print that can help you achieve this goal.

I am a contrarian trader. I have learned my art the hard way, by making plenty of mistakes, by unknowingly becoming part of the crowd at the wrong time. You see, the reason stock prices move up and down so much is that we all like to join crowds, social groups of like-minded people. When such crowds form around investing themes in the stock market, they push stock prices too high or too low relative to fair value. Why? Crowds suppress the dissenting views of nonmembers and amplify the consensus views of their members. Crowd members act together, not independently, and when this happens the market price strays substantially from fair value.

Economic experts believe that stock prices are much more variable than warranted by fluctuations in corporate profits and dividends. I think that the constant formation and disintegration of investment crowds is responsible for this excessive variability and for the wide range over which stock prices tend to fluctuate.

Another way of putting this is to say that investment crowds are responsible for the pricing mistakes made by Mr. Market. Mr. Market is the subject of an investment parable told by the father of value investing, Benjamin Graham. Mr. Market is at your elbow each day telling you what he thinks your investment portfolio is worth. Many days his estimate seems plausible and justified by business conditions. On many other days Mr. Market lets his enthusiasm or fears run away with him, and the value he proposes seems little short of silly. Investment crowds are responsible for Mr. Market's periodic bouts of enthusiasm or fear.

If investment crowds are responsible for the pricing mistakes made by the stock market, then it logically follows that you can do better than a buy-and-hold investor if you can detect those situations in which an investment crowd has driven a stock or the entire market too high or too low relative to fair value. The method for doing this that I propose in this book rests on a simple observation.

Crowds develop and grow during a communication process called an information cascade information cascade. During an information cascade the print and electronic media focus public attention on recent, dramatic movements in markets and the a.s.sociated profits and losses of investors. This in turn encourages people to put aside their natural skepticism and adopt the investment theme the media are highlighting. As the investment crowd thus grows larger, it pushes the market even further away from fair value and toward a substantial valuation mistake.

I think a contrarian trader can learn to take advantage of Mr. Market's periodic bouts of enthusiasm and fear by tracking information cascades in the media. I will show you how to do this by keeping a media diary. In the final third of this book I'll ill.u.s.trate the use of the contrarian trader's media diary during the turbulent years beginning with the 1987 crash and continuing through the panic of 2008.

I warn you that the journey toward becoming a contrarian trader will be a difficult one with an uncertain end. Most people are simply not cut out to be contrarian traders, for they love the companions.h.i.+p and approval of their fellow investors too much. But if you are prepared to step away from the crowd, to make wise investment choices that the crowd will think silly or ill-advised, then this book is for you. You may also want to follow my contrarian trading views in real time. For these you can look to my blog, which currently can be found at www.carlfutia.blogspot.com.

Each of the following 16 chapters begins with a brief outline of its content. Here I want only to give you a general picture of this book's structure and the way in which it explains the process of contrarian trading.

Chapters 1 to 5 develop the foundation on which our method will rest. We answer the question of why some speculators win but most lose, and in so doing we identify the successful speculator's characteristic edge. We will see why investment crowds are responsible for market mistakes and discuss the characteristic behavior of such crowds.

Chapters 6 to 11 explain a practical approach to contrarian trading. Here we learn about the contrarian trader's princ.i.p.al tool, his media diary. We discover how the information contained in a media diary can be interpreted and then coordinated with a statistical view of a market's current and past swings. We also develop specific contrarian trading strategies, one for a conservative and another for an aggressive contrarian trader.

In Chapters 12 to 15 we apply the techniques explained in the preceding chapters to the stock market. I kept my own media diary in real time during the years 1987 to 2008. I think you will be surprised to see how effectively it identified the many valuation mistakes the stock market made during those years.

Chapter 16 contains a small number of very brief essays and notes I wrote for my own benefit as I learned to be a contrarian trader. It explains the development of the theory of contrary opinion, highlights the contributions key individuals made to the theory, discusses briefly several books every contrarian should read, and offers comments on back-of-the-envelope value investing for the contrarian trader.

You will notice that this book contains not a single stock market chart. There is a good reason for this. When you see a chart accompanying an explanation of a stock market technique, you also generally also see how things turned out. This makes the result of a good investment decision seem inevitable and obvious. But every real-life decision is made under conditions of great uncertainty, at a time when it is not at all obvious whether your choice will yield a subsequent profit or instead yield a loss. To convey more of the feeling of uncertainty that accompanies real investment decisions, I have chosen to focus attention only on the facts that could be known at the time the investment choice was made. This is best done without the use of charts as ill.u.s.trations.

There is another reason to exclude charts from the discussion. When presented with a chart, the human eye is naturally drawn to its salient features. For a stock market chart these are usually the high and low points of prices. But one important message of this book is that the contrarian trader is not in the business of predicting stock market highs and lows or of making correct forecasts of any kind. Instead his focus is on a single objective, that of achieving a higher return than that earned by the buy-and-hold strategy. He does this by adopting an investment strategy that leans against the crowd. This does not require him to buy near low points or sell near high points in prices. It only demands that his average selling price exceeds his average buying price by an amount sufficient to compensate for risk and for the time value of money.

Writing this book has been an adventure and a pleasure. If you can take from it even one idea that improves your investment results, then I shall be doubly rewarded. Now let us begin our uncertain journey.

CARL FUTIA.

November 26, 2008

CHAPTER 1.

Can You Beat the Market?

The speculator's edge * * traits needed for an edge traits needed for an edge * * the right stuff the right stuff * * markets need speculators markets need speculators * * lending a helping hand lending a helping hand * * fair value fair value * * market mistakes market mistakes * * uncovering mistakes uncovering mistakes * * corporate profits and fair value corporate profits and fair value * * statistical models for profit forecasting statistical models for profit forecasting * * such models are useless such models are useless * * investors don't live in Lake Wobegon investors don't live in Lake Wobegon * * evidence from mutual fund performance evidence from mutual fund performance * * technical a.n.a.lysis and market timing won't give you an edge technical a.n.a.lysis and market timing won't give you an edge * * the catch-22 of investing the catch-22 of investing * * the No Free Lunch principle the No Free Lunch principle * * the art of speculation the art of speculation * * most people should not speculate most people should not speculate * * but if you have the right stuff, read on! but if you have the right stuff, read on!

THE SPECULATOR'S EDGE Can you beat the market? I'm going to do my best to convince you that the answer to this question is I'm going to do my best to convince you that the answer to this question is no no. This surely is a novel way to start a book about speculation! Of course the name of the speculative game is beating the market. And, yes, I want you to read this book about beating the market from cover to cover and tell all your friends to do the same. But I also want you to read these chapters with your eyes wide open to the dangers and pitfalls of speculation. There is no easy money waiting for you in the financial markets. So here, right up front, is the most important thing I have to say to you: Don't speculate unless you are sure you have an edge Don't speculate unless you are sure you have an edge. Without an edge you can't beat the market. Without an edge you can't beat the market.

What do I mean by an edge? An edge is a talent or skill or some specific knowledge that will give you an advantage over other investors and speculators. Sad to say, a high IQ, great educational credentials, or a substantial net worth are not not edges in the game of speculation. Neither is the willingness to work hard and to keep trying after repeated failures. These things may make you a success in your profession or trade and a valued member of your community. But they won't guarantee you success in the world of speculation. edges in the game of speculation. Neither is the willingness to work hard and to keep trying after repeated failures. These things may make you a success in your profession or trade and a valued member of your community. But they won't guarantee you success in the world of speculation.

You should know that the biggest part of any speculator's edge does not not come from a superior scientific or statistical knowledge of market behavior. If it did, you could build your edge the same way you acquire skills in any profession-by study and practice. But have you noticed that no college or university offers a major in speculation? There is a good reason for this. A speculator's edge arises from two personal traits that can't be taught and that people either have or don't have. The first is flexibility of mind and spirit, the ability to adapt easily and quickly to changes in market conditions and habits. The second is the willingness to think for oneself and to risk hard-earned money by "fading" (investing opposite to) popular opinion. This means that you will usually take market positions that most people (your husband or wife especially!) will see as unwise or even foolish. Doing this day in and day out requires emotional toughness that few people can muster. It also requires a certain arrogance-a firm conviction that you know what you are doing and that most other people in the market don't. Do you have the right stuff to be a successful speculator? come from a superior scientific or statistical knowledge of market behavior. If it did, you could build your edge the same way you acquire skills in any profession-by study and practice. But have you noticed that no college or university offers a major in speculation? There is a good reason for this. A speculator's edge arises from two personal traits that can't be taught and that people either have or don't have. The first is flexibility of mind and spirit, the ability to adapt easily and quickly to changes in market conditions and habits. The second is the willingness to think for oneself and to risk hard-earned money by "fading" (investing opposite to) popular opinion. This means that you will usually take market positions that most people (your husband or wife especially!) will see as unwise or even foolish. Doing this day in and day out requires emotional toughness that few people can muster. It also requires a certain arrogance-a firm conviction that you know what you are doing and that most other people in the market don't. Do you have the right stuff to be a successful speculator?

I think you will agree that this is an unusual explanation of the nature of the speculator's edge. In our technological society, it's natural for people to believe that speculative profits arise from the use of superior methods or from some arcane knowledge of market behavior. But this isn't true. The essence of successful speculation cannot be found in specialized knowledge of market behavior or of trading technique. You can't learn to be a successful speculator by reading books (this one included!), by taking courses, or by attending seminars.

However, if you do have the right stuff to be a speculator, then you can move your game to a higher level by applying the methods I explain in the following chapters. The financial markets need skilled speculators. Capitalism couldn't survive without them. To see why, just keep reading.

LENDING A HELPING HAND TO INVESTORS.

What is a speculator? What is his mission on capitalism's battlefield of creative destruction? Lewis and Short (my always-at-hand Latin dictionary) defines the verb speculor speculor to mean "the action of watching, observing, examining or exploring." So a speculator is a lookout, a scout, an explorer, and an investigator. to mean "the action of watching, observing, examining or exploring." So a speculator is a lookout, a scout, an explorer, and an investigator.

A financial speculator explores the terrain ahead of the army of long-term investors. This army is advancing toward a very uncertain future, a consequence of Joseph Schumpeter's perennial gale of creative destruction gale of creative destruction , which always accompanies the development of a capitalist economy. Long-term investors must be a.s.sured of being able to buy and sell at a , which always accompanies the development of a capitalist economy. Long-term investors must be a.s.sured of being able to buy and sell at a fair price fair price, and this despite the enormous uncertainty that is in capitalism's very nature. If long-term investors believe that markets won't give them a fair shake, they will lock up their investment capital, and the machinery of capitalism would then grind to an impoveris.h.i.+ng halt.

How does a speculator help ensure that long-term investors get a fair shake? Every speculator is on the lookout for mistakes the market has made in pricing a stock, bond, or commodity. A market mistake is a situation where the current market fails to accurately reflect all that is known about the probable earning power of the company or the supply-demand balance for the commodity. A speculator profits by spotting market mistakes and helping to correct them by buying when the price is too low and selling when it is too high.

A market mistake is a deviation from the fair value price fair value price. The phrase fair value fair value is a plain-and-simple term for what economists call the is a plain-and-simple term for what economists call the equilibrium price equilibrium price (i.e., the price that will equate supply with demand). Economics teaches that the equilibrium price is an accurate reflection of what is known about the prospects of the stock or commodity in question. As such, the equilibrium price is a very good thing. People who buy or sell at the equilibrium price are getting a fair shake; they aren't being unfairly exploited by more knowledgeable investors. (i.e., the price that will equate supply with demand). Economics teaches that the equilibrium price is an accurate reflection of what is known about the prospects of the stock or commodity in question. As such, the equilibrium price is a very good thing. People who buy or sell at the equilibrium price are getting a fair shake; they aren't being unfairly exploited by more knowledgeable investors.

It is important to remember that the concept of fair value can be difficult to pin down. In the next chapter we briefly discuss one method for calculating fair value: discounted future dividends discounted future dividends. In Chapter 5 we discuss another: the q ratio q ratio, first developed by the economist James Tobin. Both of these methods are designed to give very long-term, multiyear estimates of the fair value price. But generally both methods are too unwieldy to be of much use to a professional speculator. We discuss more practical ways to estimate fair value in Chapter 6.

It should come as no surprise that markets make mistakes. Usually these mistakes are only short-lived, minor ones, but on occasion a market makes a really big, long-lasting mistake. Mistakes can take the form of a shortsighted reaction to a surprising corporate or economic development. Or a mistake can arise because of a ma.s.s delusion or mania. In either case, the price of the stock or commodity rises too high or falls too low relative to any reasonable a.s.sessment of fair value.

A speculator's economic function is to be on the lookout for these market mistakes and to help correct them. He does this by buying when the price is below fair value and by selling when it is above. The speculator's buying and selling thus helps to nudge the market price closer to fair value. In this way speculators perform a valuable service for longer-term investors. They help ensure that market prices more often and more closely reflect the best possible a.s.sessment of future economic prospects.

UNCOVERING MARKET MISTAKES.

How does a speculator know that the market is making a mistake? You can be sure that there is no neon sign to that effect posted in front of the stock exchange. You won't see "XYZ ON SALE TODAY" or "ABC NOT WORTH AN ARM AND A LEG" running across the message board at 42nd and Broadway in New York City.

Most investors approach the problem of identifying market mistakes from an economic and statistical perspective. Basic economic considerations suggest that the fair value price for a company's stock should be determined by discounting to the present the profits the company is likely to earn over some reasonable time interval, say 10 years. You can try to estimate of these profits by modeling the industry and the economy using state-of-the-art statistical and economic tools. Or you could buy this information from someone who can do this modeling for you. In either case this profit estimate will determine an estimate of the fair value price for the stock. Detecting a market mistake is then just a matter of comparing this estimate of the fair value price with the current market price.

This certainly is a logical approach to the problem of uncovering market mistakes, at least in the stock market. Economists agree that the fair value for a corporation's common stock is the price that reflects all the information currently available about the company's future earning power, dividends, general economic conditions-everything that might be relevant to estimating the likely future dividends and capital gains an investor could expect. Investors who adopt this approach will purchase stocks that are trading below their estimates of fair value and sell stocks if they are trading above such estimates. Here is the key question: Is there any reason to believe that this method for detecting market mistakes will allow an investor to earn above-average returns?

You may find my answer to this question shocking. I believe that it is impossible impossible to earn above-average returns on your investment portfolio by using statistical estimates of economic fair value. Why? Well, the key phrase is to earn above-average returns on your investment portfolio by using statistical estimates of economic fair value. Why? Well, the key phrase is above-average returns above-average returns. One can certainly use statistical and business knowledge to construct models for estimating fair value of a common stock that have some reliability. But you must keep in mind that speculation is a very compet.i.tive business. Many investors, money managers, and economic consultants are doing this same thing. They are all competing for the profits that can be earned by making superior estimates of a stock's fair value price.

Sadly, unlike the children of Lake Wobegon, who are all above average, investors cannot all achieve above-average investment results. Remember that lots of people have the knowledge and statistical skills to build good corporate earnings forecasting models. If building such models led to superior investment results, people would rush in and adopt this methodology. But by doing so they would collectively move market prices in the direction of their fair value estimates. This would narrow the deviation of the market price from the fair value estimates to the point where this investment technique would yield only average results. There is so much compet.i.tion among model builders and the investors who pay for these models' forecasts that neither group can earn above-average returns, either by building models or by using the forecasts the models produce to guide their investment strategy!

LOOKING AT THE EVIDENCE.

Perhaps I have already convinced you that compet.i.tion makes it hard to speculate successfully by doing corporate profit modeling. But if not, you might counter by saying that the world in which we live is nothing like the freely compet.i.tive world of theoretical economics. Perhaps all that is needed is to build the better mousetrap, the super-duper, high-tech profit-forecasting model that will beat all others to the pot of gold. I think there is very good reason to be skeptical of this possibility. If resources and technical skills would guarantee success in the battle for investment profits, we should find that investment professionals, those who ought to have access to the best profit-forecasting models, produce better than average investment results. So let's look at the actual investment results achieved by professional money managers to see if this is true.

In a 2005 article in the Financial Review, Financial Review, "Reflections on the Efficient Market Hypothesis: 30 Years Later," volume 40, pp. 1-9, Burton Malkiel examined the performance of professional money managers in the United States and other developed countries. His data on mutual fund performance reveal three important facts. First, most actively managed stock market mutual funds underperform their benchmark index, the Standard & Poor's (S&P) 500. Over a single-year time span, 73 percent do worse than the index, and this percentage increases to 90 percent if one considers performance over a 20-year time span. Second, pa.s.sively managed S&P 500 index funds do about 2 percent better per year than do actively managed stock market mutual funds. Most of this difference is accounted for by the higher fees actively managed mutual funds charge their shareholders. Finally, there is little consistency from year to year in performance relative to the benchmark by any given mutual fund. So it is impossible to tell in advance which mutual funds will do better than the benchmark using only their past performance as a guide. "Reflections on the Efficient Market Hypothesis: 30 Years Later," volume 40, pp. 1-9, Burton Malkiel examined the performance of professional money managers in the United States and other developed countries. His data on mutual fund performance reveal three important facts. First, most actively managed stock market mutual funds underperform their benchmark index, the Standard & Poor's (S&P) 500. Over a single-year time span, 73 percent do worse than the index, and this percentage increases to 90 percent if one considers performance over a 20-year time span. Second, pa.s.sively managed S&P 500 index funds do about 2 percent better per year than do actively managed stock market mutual funds. Most of this difference is accounted for by the higher fees actively managed mutual funds charge their shareholders. Finally, there is little consistency from year to year in performance relative to the benchmark by any given mutual fund. So it is impossible to tell in advance which mutual funds will do better than the benchmark using only their past performance as a guide.

Malkiel's conclusions are typical of those reached by financial economists when they examine the performances of professional money managers. From this body of research I think we must conclude that models that estimate fair value using economic and business data will not not give you any advantage over other investors. If they did, we would expect to see above-average investment performance by stock market mutual funds, because their managers have access to the best earnings forecast models. We would also expect such market-beating performance to persist from year to year for specific mutual funds, because it is the mutual fund management firm that pays for the models and these models would be available to any manager who works for the management firm. give you any advantage over other investors. If they did, we would expect to see above-average investment performance by stock market mutual funds, because their managers have access to the best earnings forecast models. We would also expect such market-beating performance to persist from year to year for specific mutual funds, because it is the mutual fund management firm that pays for the models and these models would be available to any manager who works for the management firm.

But we see none of these things. The conclusion to be drawn from this evidence is simple enough. If you are trying to identify the market's mistakes by using statistical models to estimate future profits, you are barking up the wrong tree. Models that forecast corporate profits can't help you beat the market, because everyone uses them. After all, this sort of approach to stock market valuation is taught in every business school. How could it give you a chance to earn above-average returns if every professional money manager knows and uses it?

MARKET TIMING.

There is an even more striking conclusion to be drawn from the persistent underperformance of mutual fund money managers as a group. The logic that leads one to conclude that statistical forecasting models that forecast corporate profits can't be used to achieve market-beating investment performance has to apply to other approaches as well.

This broader category of essentially valueless methodologies includes what are popularly known as technical a.n.a.lysis and market timing. The idea behind technical a.n.a.lysis is that a market's price action reveals to the careful observer what other investors have learned about fair value. For example, investors who estimate fair value using economic and business data (so-called fundamentalist investors) reveal these estimates to the watchful and skilled market technician via the buying and selling they do to take advantage of their models' estimates. In this way a market technician believes he can piggyback his a.n.a.lysis upon the efforts of the fundamentalist investors. When he does this, he amplifies the effects of fundamentalists' buy and sell decisions.

In the standard technical a.n.a.lyst tool kit one finds various forms of price chart interpretation, momentum and moving average trading strategies, and overbought-oversold oscillator methods. These tools are too widely known and studied to help you earn above-average returns on your investments. Any advantage they might confer is soon competed away in the profit-seeking rush of technical a.n.a.lysts to adopt them. Of course one cannot rule out the possibility that there are are market-beating technical methods. One can only deduce that you will not read about them in a book! market-beating technical methods. One can only deduce that you will not read about them in a book!

A market timer is someone who attempts to beat the market by predicting the swings in market prices ahead of time and acting on these predictions. Technical a.n.a.lysis typically plays a big role in most market timers' decision processes. But market timing is in general a fruitless activity for the same reason that technical a.n.a.lysis fails.

Think about market timing like this. If a market timer is to be successful he must be right twice in a row-he must first buy low and then sell high. So let's suppose our hypothetical market timer is quite skilled and has developed a method that predicts and takes advantage of the direction of a market's upcoming move 70 percent of the time (most methods I have seen don't come close to this success rate). The probability that this market timer is right two consecutive times is .70 .70 = .49. So even if his method guesses right 70 percent of the time, only 49 percent of the time will he improve his position over the alternative of doing nothing. For this reason the odds are that a market timer's efforts will simply make his portfolio more volatile without increasing his average returns. Even a skilled market timer will have difficulty beating the market.

CATCH-22.

We have just encountered what I call the catch-22 of investing: Any statistical methodology that directly (fundamentalist approach) or indirectly (technical a.n.a.lysis approach) estimates fair value and that is widely used cannot help you beat the market. Economists call this the No Free Lunch principle No Free Lunch principle. Compet.i.tion among investors leads to a situation in which knowledge in the public domain can't lead to above-average investment returns. There is no information you can find in a book on investing or trading (this one included!) or you can learn at an investment seminar that will by itself help you do better than the market. Notice that this also means that it is even impossible for you, an average investor, to purchase superior investment performance by entrusting your money to a professional money manager.

So where does the No Free Lunch principle leave us? One thing seems obvious. Developing an edge over other investors cannot simply be a matter of reading some books, getting a good education, or having a high IQ. An edge cannot arise from mastery of statistical and a.n.a.lytical skills you can learn from books or in business school courses. The reason is simple: Lots of people do this, and what lots of people do won't make you an above-average investor.

I think that to develop an edge you must start by abandoning preconceived ideas. You are making a good start if you have read this far. You must learn to be suspicious of popular opinion and conventional wisdom. Indeed, the entire art of speculation consists of choosing the right moment to invest in a way opposite to that suggested by popular opinion. An edge can be found only by living by one's wits in a world where survival and prosperity are daily question marks. I have seen firsthand that maintaining this kind of edge extracts an emotional and mental toll that very few people can pay. For the vast majority, an investment edge is just not worth the effort it takes to acquire and maintain it. It makes no sense for most people to speculate in the financial markets.

Well, I have done my best to convince you that you should not speculate. But if you still think you have the right stuff, you will find a lot of useful information and suggestions in the rest of this book. I'll try to explain what you must do to hone your speculative skills. I can summarize my message very simply: Become a professional contrarian!

CHAPTER 2.

Market Mistakes Efficient markets? not! * * markets make mistakes markets make mistakes * * "no free lunch" means that mistakes are hard to identify "no free lunch" means that mistakes are hard to identify * * roller coasters and stock markets roller coasters and stock markets * * do stock prices fluctuate too much? do stock prices fluctuate too much? * * s.h.i.+ller's work on stock market volatility s.h.i.+ller's work on stock market volatility * * efficient markets theory can't explain volatility efficient markets theory can't explain volatility * * behavioral finance behavioral finance * * why might market mistakes persist? why might market mistakes persist? * * the unknown unknown the unknown unknown * * the last one to know the last one to know * * the lunatic factor the lunatic factor * * markets do not make systematic mistakes markets do not make systematic mistakes * * Fama's work Fama's work * * No Free Lunch redux No Free Lunch redux * * looking for the edge looking for the edge EFFICIENT MARKETS.

In Chapter 1 I explained that a speculator's job is to discover market mistakes and then help to nudge the market price back to fair value. I boldly a.s.serted that markets make lots of mistakes. But you should know that there are respected theories in financial economics that a.s.sert that markets never make mistakes. These theories say that markets are strongly efficient, that at every moment the market accurately reflects all that can be deduced from economic a.n.a.lysis and technical market methods about today's fair value market price.

If markets really were efficient in this strong sense, there would be no room for speculators to make a profit. In this chapter I want to explain why I think markets are not not efficient in this strong sense. I want to argue that, as a matter of fact and not opinion, markets rarely trade at the fair value, economic equilibrium price. I want to convince you that at least in principle there are many opportunities for speculators to make a profit. To do this, we'll have to take a close look at the reasons markets spend extended periods of time trading well above or well below fair value. This investigation will carry us into the realm of behavioral finance, investor sentiment, and crowd behavior. To keep things simple, I am going to focus on the mistakes made by markets for common stocks in the United States, although all financial markets worldwide are mistake-p.r.o.ne as well. efficient in this strong sense. I want to argue that, as a matter of fact and not opinion, markets rarely trade at the fair value, economic equilibrium price. I want to convince you that at least in principle there are many opportunities for speculators to make a profit. To do this, we'll have to take a close look at the reasons markets spend extended periods of time trading well above or well below fair value. This investigation will carry us into the realm of behavioral finance, investor sentiment, and crowd behavior. To keep things simple, I am going to focus on the mistakes made by markets for common stocks in the United States, although all financial markets worldwide are mistake-p.r.o.ne as well.

Keep in mind that the No Free Lunch principle tells us that markets do not make lots of statistically exploitable statistically exploitable mistakes. In other words, financial markets' mistakes show no statistical regularities and can be used to predict and identify these mistakes as they are happening. The reason for this bears repeating: There are large numbers of speculators and investors on the lookout for market mistakes, and they are willing to use any plausible statistical method to uncover these mistakes. Compet.i.tion among speculators will reduce the returns earned by their statistical methodologies to the returns earned by the buy-and-hold strategy. The No Free Lunch principle predicts that every mistake a market makes will be a surprise at the time it is happening. To the vast majority of investors, market mistakes will be visible only in hindsight. So market mistakes will be very hard to detect and exploit at the time they occur. Sadly, a speculator cannot profit from 20/20 hindsight, so the No Free Lunch principle is telling us that successful speculators are a rare breed-people who have skills and characteristics that cannot be taught or easily discovered. mistakes. In other words, financial markets' mistakes show no statistical regularities and can be used to predict and identify these mistakes as they are happening. The reason for this bears repeating: There are large numbers of speculators and investors on the lookout for market mistakes, and they are willing to use any plausible statistical method to uncover these mistakes. Compet.i.tion among speculators will reduce the returns earned by their statistical methodologies to the returns earned by the buy-and-hold strategy. The No Free Lunch principle predicts that every mistake a market makes will be a surprise at the time it is happening. To the vast majority of investors, market mistakes will be visible only in hindsight. So market mistakes will be very hard to detect and exploit at the time they occur. Sadly, a speculator cannot profit from 20/20 hindsight, so the No Free Lunch principle is telling us that successful speculators are a rare breed-people who have skills and characteristics that cannot be taught or easily discovered. It is my thesis that the successful speculator's edge rests on his ability to stand apart from the crowd and to act opposite to the crowd's beliefs and expectations. It is my thesis that the successful speculator's edge rests on his ability to stand apart from the crowd and to act opposite to the crowd's beliefs and expectations.

ROLLER COASTERS AND STOCK MARKETS.

When the financier John Pierpont Morgan (1837-1913) was asked for his stock market forecast he replied, "Stocks will fluctuate." There is more wisdom in his reply than meets the eye. Morgan understood the implications of the No Free Lunch principle, because he made no attempt to predict whether the stock market was about to move higher or lower. He spoke as if he believed that all the factors that might influence the market's movements were already reflected in the level of current prices. Future price changes would therefore be responses to new and currently unknowable information. But he also emphasized that stocks would "fluctuate." We shall soon see that this is an enduring characteristic of the stock market, and indeed of all financial markets.

Do you have fun riding roller coasters? Frankly, they are not my cup of tea, but my children love them. I think stock markets are a lot like roller coasters. Prices always seem to be in motion-up or down. At the start of a bull market, stock prices, like the start of a roller coaster's ride, move slowly but steadily higher. They seem to be climbing a steep hill pulled by improvements in underlying economic conditions and corporate profits. As the market reaches the top of the hill, its advance slows, flattens out, and then prices gradually soften. Soon the market is hurtling downhill as if pulled by gravity, its riders screaming in panic. Fear replaces optimism, but just when the stock prices appear ready to plunge to zero and economic depression seems imminent, the market springs upward. It is as if the very momentum of the previous stomach-churning drop has combined with an invisible turn of the track to push prices higher once more. The path to the next bull market peak will no doubt take the market through several hair-raising twists, turns, and loops. It will be lots of fun, but everyone knows that another terrifying drop lies dead ahead. Its all part of the roller-coaster experience.

Imagine having to ride a roller coaster whenever you had to travel to work from home. How would you feel if a roller coaster was the only available mode of transportation you had? I suppose you could get used to it. But you would be traveling many more miles (many of them in the vertical direction!) than you do now in your car. Not only would your travel times lengthen, but (at least for me) just getting to the grocery store would be a nerve-racking experience.

Investors have no choice but to ride the market roller coaster. Its ups and downs seem designed to defeat all but those with the strongest stomachs. And, just like a roller coaster, the market's movements are in a certain sense artificial. At least from an economic perspective, they are not justified by ups and downs in underlying economic conditions and corporate profits. Instead they reflect something that seems inherent in the very nature of the process that prices corporate a.s.sets.

Professor Robert J. s.h.i.+ller is among the word's experts in the study of the behavior of speculative markets. In March 2000, with an exquisite sense of market timing in the publis.h.i.+ng world, the book for which he is justifiably famous, Irrational Exuberance Irrational Exuberance (Princeton University Press), shook the world of Wall Street. In it s.h.i.+ller argued that the great stock market boom that had started in 1982 was displaying all the signs of a bubble, that every indicator of economic value showed that investors were (Princeton University Press), shook the world of Wall Street. In it s.h.i.+ller argued that the great stock market boom that had started in 1982 was displaying all the signs of a bubble, that every indicator of economic value showed that investors were irrationally exuberant irrationally exuberant. They were placing unreasonably high valuations on prospective corporate profits and dividends. We all know what happened next. In March 2000 the S&P 500 began a drop that eventually carried it 50 percent lower by October 2002. The NASDAQ Composite index, home to the Internet and technology highfliers of 1998-2000, dropped 80 percent during the same time. Some fluctuation!

Yet the bull market of 1982-2000 and the ensuing three-year bear market were by no means unprecedented events, as s.h.i.+ller shows in Irrational Exuberance Irrational Exuberance. He is in a position to know. Indeed, Professor s.h.i.+ller's entire career had been devoted to the study of speculative markets and to investigating whether market fluctuations are in any sense determined and justified by economic developments and changes in corporate profits. The results of his investigations were published 11 years earlier in his book Market Volatility Market Volatility (MIT Press, 1989). They provide important information about the frequency and extent of the stock market's mistakes. (MIT Press, 1989). They provide important information about the frequency and extent of the stock market's mistakes.

DO STOCK PRICES FLUCTUATE TOO MUCH?.

In Chapter 4 of Market Volatility Market Volatility s.h.i.+ller adopted the standard economic hypothesis for determining the fair value of the stock market. This says that the long-term fair value price for the Standard & Poor's Composite 500 stock market index should be taken to be the present discounted value of future dividend payouts. This standard approach to a.s.set valuation would apply not just to common stocks but to any other a.s.set (e.g., bonds, real estate. etc.) that is expected to yield a regular sequence of cash payouts over the foreseeable future. Of course, at any point in time one doesn't actually know what these future dividends will be, say over the next 30 years, but s.h.i.+ller's economic model a.s.sumes that you do. You might object that such an a.s.sumption is ridiculous on its face. But it turns out that the discounted stream of dividends is very predicable and grows at a nearly constant rate. So in this sense s.h.i.+ller's "perfect foresight" a.s.sumption is a reasonable one because dividend payouts for the stock market as a whole are quite predictable. s.h.i.+ller adopted the standard economic hypothesis for determining the fair value of the stock market. This says that the long-term fair value price for the Standard & Poor's Composite 500 stock market index should be taken to be the present discounted value of future dividend payouts. This standard approach to a.s.set valuation would apply not just to common stocks but to any other a.s.set (e.g., bonds, real estate. etc.) that is expected to yield a regular sequence of cash payouts over the foreseeable future. Of course, at any point in time one doesn't actually know what these future dividends will be, say over the next 30 years, but s.h.i.+ller's economic model a.s.sumes that you do. You might object that such an a.s.sumption is ridiculous on its face. But it turns out that the discounted stream of dividends is very predicable and grows at a nearly constant rate. So in this sense s.h.i.+ller's "perfect foresight" a.s.sumption is a reasonable one because dividend payouts for the stock market as a whole are quite predictable.

Having this estimate of the stock market's fair value price in hand, s.h.i.+ller then compares it with the actual level of the S&P Composite 500 index. Both these numbers change from year to year, and s.h.i.+ller compares the variability of the S&P Composite over time with the variability of the discounted stream of futures dividends. He finds that prices are in the long run about four times as volatile as the discounted stream of dividends. Moreover, the correlation between the changes in the level of the S&P Composite index and the changes in the value of the long-run stream of dividends is very low. This is surprising, at least to economists. Efficient markets theory predicts that stock market prices should show less variability than discounted dividends, and that changes in stock market prices should be highly correlated with changes in the discounted value of dividends.

What should we make of these facts? At an intuitive level at least, most investors would find s.h.i.+ller's conclusions unsurprising. Indeed, they might marvel that anyone would suggest that the discounted stream of future dividends should determine fair value in the stock market. "After all," they would say, "we care far more about earnings growth than dividend payouts, and our profit forecasting models take this difference into account." But such a response misses the point. The fact is that economy-wide earnings and economy-wide dividends move in lockstep over time. True, any individual company may be seen as a "growth opportunity" or as a "steady dividend payer" by individual investors, but this is a distinction without a difference when looking at the economy as a whole and when trying to understand the source of price fluctuations in the S&P Composite index. If stock prices are not not determined by the long-run dividends and earnings generated by corporations in the U.S. economy, determined by the long-run dividends and earnings generated by corporations in the U.S. economy, what does determine them? what does determine them? In this situation I believe that the sage of Omaha, Warren Buffett, would accept the broad outline of s.h.i.+ller's approach. Buffett has often quoted Benjamin Graham's saying that in the short run the stock market is a voting machine, but in the long run it is a weighing machine. By this he means that the value of a company's common stock is determined in the long run by the company's ability to earn profits and pay out dividends to its shareholders. In this situation I believe that the sage of Omaha, Warren Buffett, would accept the broad outline of s.h.i.+ller's approach. Buffett has often quoted Benjamin Graham's saying that in the short run the stock market is a voting machine, but in the long run it is a weighing machine. By this he means that the value of a company's common stock is determined in the long run by the company's ability to earn profits and pay out dividends to its shareholders.

Some economists have criticized s.h.i.+ller's calculations by observing that the growth of dividends from year to year is better described as a random walk than as a process that fluctuates around a known trend. If this is true, one would in general expect more volatility in stock prices than s.h.i.+ller's model would predict. (In technical terms these economists are saying that the series of dividends is not stationary, contrary to s.h.i.+ller's a.s.sumption.) But it turns out that even when this different statistical model of the behavior of dividends is adopted, the stock prices still fluctuate far too much relative to long-run dividend fluctuations.

The basic conclusion of s.h.i.+ller's research is that the efficient markets model does not do well in explaining stock market volatility. This no doubt cheers investors who have grown weary of economists' employing efficient markets theory as a club to beat them into accepting the futility of trying to beat the market. It is clear from s.h.i.+ller's data that the stock market as a whole makes lots of mistakes-the S&P Composite index fluctuates in a wide range around fair value as measured by discounted future dividends. But those who would take satisfaction from this defeat of the efficient markets theory should be wary. The No Free Lunch phenomenon still rules the investment world. Stock markets may indeed make many mistakes, but it is generally impossible to recognize these mistakes when they occur! But might thinking outside the box of standard financial theory and looking at different sorts of data help us identify and exploit the mistakes of the sort s.h.i.+ller detected?

A LOOK AT BEHAVIORAL FINANCE.

To answer this last question we have to take a look at the reasons stock market prices might deviate from fair value. This will take us into the realm of what is known as behavioral finance, a branch of economics. Cla.s.sical economics studies the behavior of markets in which people buy and sell using all the information they have in a statistically appropriate way. But what if human psychology or computational limitations prevents this? What if a significant number of investors, for whatever reason, fail to use information available to them correctly? How would this affect the behavior of the stock market? Must it inevitably cause the market to make mistakes? Behavioral economics addresses these sorts of questions.

Richard Thaler is a professor at the University of Chicago and a widely recognized expert in behavioral finance. Together with his co-author, Nicholas Barbaris, Thaler wrote a recent survey of behavioral finance that appears as Chapter 1 in Advances in Behavioral Finance, Volume 2 Advances in Behavioral Finance, Volume 2 (Richard Thaler, ed.; Princeton University Press, 2005). Thaler and Barbaris discuss the many reasons why investors might psychologically be unable to make good statistical use of the information they have. Indeed, they point to the ample experimental evidence showing that this is a pervasive phenomenon, that people fail to use a sound statistical approach when making choices under uncertainty. (Richard Thaler, ed.; Princeton University Press, 2005). Thaler and Barbaris discuss the many reasons why investors might psychologically be unable to make good statistical use of the information they have. Indeed, they point to the ample experimental evidence showing that this is a pervasive phenomenon, that people fail to use a sound statistical approach when making choices under uncertainty.

However, as Barbaris and Thaler point out, the fact that a segment of the investor population acts irrationally in this sense does not necessarily imply that markets make mistakes. After all, if some people make statistical errors and consequently drive the price of a stock away from its fair value price, shouldn't a rational investor be able to take advantage of this mistake by taking the opposite action and profiting as the price returns to fair value?

To this question the behavioral economist answers "perhaps," but the cla.s.sical economist answers "always." Here lies the essential difference between behavioral theories of financial markets and the cla.s.sical efficient markets hypothesis. Let's examine the reasons why rational investors might not be able to correct the mistakes made by irrational investors.

Every trading and investment strategy has its a.s.sociated costs and risks. Costs arise from the process of buying and selling itself (brokerage commissions, bid-ask spreads, etc.) as well as from the expense of gathering and processing information correctly. Risks arise from the usual uncertainties a.s.sociated with a.s.set pricing as well as from inst.i.tutional factors resulting from the way markets are organized. A simple example of this is short selling. If a rational investor recognizes that a stock is selling above its fair market price and he wants to partic.i.p.ate in this selling, he may not be able to sell the stock if he does not already own it. To do so he must borrow the shares from a willing lender, and this not only requires the cooperation of two parties but the payment of interest on the loan of the shares. Moreover, the market for borrowing these shares may be very thin and may expose the short seller to added risk of a squeeze by the lender. Thus, even if the shares can be borrowed, the costs and risks of short selling are not symmetrically placed with those of buying the stock in question. For this reason alone, one might expect the stock market to be more p.r.o.ne to make mistakes of overvaluation than of undervaluation.

But there are three other kinds of risk a.s.sociated with attempts to correct market mistakes-risks that are rarely considered, probably because experienced speculators don't usually put their thoughts on paper.