Textbook Cases

9 Market Efficiency and Behavioral Finance

Learning Goals

After studying this chapter, you should be able to:

1. LG 1 Describe the characteristics of an efficient market, explain what market anomalies are, and note some of the challenges that investors face when markets are efficient.

2. LG 2 Summarize the evidence which indicates that the stock market is efficient.

3. LG 3 List four “decision traps” that may lead investors to make systematic errors in their investment decisions.

4. LG 4 Explain how behavioral finance links market anomalies to investors’ cognitive biases.

5. LG 5 Describe some of the approaches to technical analysis, including, among others, moving averages, charting, and various indicators of the technical condition of the market.

6. LG 6 Compute and use technical trading rules for individual stocks and the market as a whole.

In 2013 the Nobel Prize in economics was awarded to three co-recipients: Eugene Fama, Robert Shiller, and Lars Peter Hanson. In giving a shared award to Fama and Shiller, the committee appeared to display a sense of humor because those two scholars are best known for holding opposing views on the efficiency of financial markets. Eugene Fama was among the first to define the term “efficient markets” in his landmark study that concluded that stock prices moved almost at random and that any attempt to earn better-than-average returns by identifying winners and losers in the stock market was a fool’s errand. Fama argued that competition among rational investors resulted in stock prices that accurately reflected all information available to market participants. If market prices reflected all available information, then no single investor could consistently identify overvalued or undervalued stocks, and therefore no investor could earn a return that consistently beat the market average (on a risk-adjusted basis).

Shiller, on the other hand, gained popular notoriety through his book, Irrational Exuberance, which argued that the stock market had become grossly overvalued in the late 1990s due to irrational behavior by investors. Indeed, Shiller’s book was published just before a stock market crash in 2000. Shiller’s message was that the stock market was anything but efficient, and that smart investors could identify times when it would be wiser to sit on the sidelines than to invest in stocks. Less than a decade later, Shiller made headlines again through his warnings that the housing market was becoming overheated, a prediction that the subsequent collapse in housing prices and related financial crisis seemed to confirm.

For many years, academics and investment professionals were on opposite sides of this debate. A broad consensus existed among academics that the market was very efficient and that neither amateur nor professional investors were likely to earn better-than-average returns over time. The professional investment community mostly disagreed with this view, arguing that well-trained investors with access to sophisticated information and trading systems could deliver superior returns to their clients. Over time, the two sides have moved closer together. A growing body of academic research, generally referred to as behavioral finance, has found evidence that the market is not as efficient as scholars once believed and that human cognitive biases place a limit on how efficient the market can be. At the same time, members of the investment community have acknowledged that consistently identifying overvalued or undervalued securities is extremely difficult and that many investors will be better off buying and holding a diversified portfolio of securities rather than paying experts to identify mispriced stocks. Among practitioners, this view has led to the growth in low-cost investment options such as index funds and exchange-traded funds.

Efficient Markets

1. LG 1

2. LG 2

To some observers, the stock market is little more than a form of legalized gambling. They argue that movements in the stock market have no real connection to what is happening in the economy or to the financial results produced by specific companies. In the eyes of people who hold this view, large swings in the market are driven by emotions like greed and fear rather than by business fundamentals. In this chapter we study the connection between prices in the stock market (and other financial markets) and real business conditions, and we ask whether and how stock prices might be affected by human emotions.

To begin, consider  Figure 9.1  , which shows quarterly revenues reported by Walmart from 2000 to mid 2015. A quick glance at the figure reveals two obvious patterns. First, Walmart’s revenues have grown over time. In early 2015 the company reported quarterly revenues of $132 billion, more than double the quarterly revenues that they had generated in early 2000. Perhaps an even more striking pattern is that there is clearly one quarter each year in which Walmart earns higher revenues than any other quarter. Those peaks, marked by red dots in  Figure 9.1  , occur in Walmart’s first quarter, which ends on January 31st each year. In other words, in every year since 2000, Walmart has sold more goods in November, December, and January than in any other quarter, a remarkably stable pattern. When you think about this pattern a little, it should come as no surprise. Nearly every retail company in the United States sells more near the end of the year because of the Christmas season, and Walmart is no exception. Although  Figure 9.1  plots Walmart’s revenues, a plot of the company’s net income would show similar patterns.

Walmart is a huge corporation, and roughly 11% of U.S. retail sales (not counting automobiles) occur in Walmart stores. Partly because it is so large and partly because much of its business focuses on life’s necessities, Walmart’s financial results are not terribly difficult to predict. This is another lesson from  Figure 9.1  . The persistence of the patterns in Walmart’s revenues over a long period of time suggests that forecasts of

Figure 9.1 Walmart Quarterly Revenues

From 2000 to mid 2015 Walmart steadily increased its quarterly revenues from $43 billion to more than $132 billion. The long-term upward trend is marked by a distinct seasonal pattern in which Walmart’s revenues peak in the first quarter each year, marked by red dots in the figure. The peak in revenues is due to the Christmas shopping season and is common in retail companies.

Figure 9.2 Walmart’s Stock Price

From 2000 to mid 2015, Walmart’s stock price rose, but it did not follow a predictable trend. Furthermore, the seasonal pattern in Walmart’s revenues does not appear in its stock price. The stock price in the first quarter of each year is marked by a red dot, and those red dots show no discernible pattern over the past 15 years.

Walmart’s future performance, at least in the not-too-distant future, are likely to be fairly accurate. Is Walmart’s stock price just as predictable?

Figure 9.2  plots Walmart’s stock price at the end of each quarter from 2000 to mid 2015, the same period covered in  Figure 9.1  . Like the company’s revenues, Walmart’s stock price was higher in 2015 than it was in 2000, but it hardly followed the relatively smooth upward trend that revenues did. The striking difference between  Figures 9.1  and  9.2  is the seemingly random movements in Walmart’s stock price, which stand in sharp contrast to the predictable movements in Walmart’s revenues. Clearly there was no tendency for Walmart’s stock to peak at the same time that its revenues did (i.e., at the end of the first quarter each year, marked by the red dots in  Figure 9.2  ). Does this mean that there is no connection between Walmart’s financial performance and the behavior of its stock?

Naturally our answer to that question is a firm no. To understand why, it may be helpful to think about what would happen if Walmart’s stock price moved in sync with its revenues, showing a seasonal peak at the end of each year. Suppose that over many years, Walmart stock displayed a regular, predictable tendency to shoot up every year in the fourth quarter. If that pattern persisted and investors came to expect the pattern to continue, what would they do? Smart investors would buy Walmart’s stock in the third quarter each year, hoping to profit from the fourth quarter runup. But if investors rushed to buy Walmart shares in the third quarter each year, their actions would put upward pressure on the stock price in the third quarter rather than the fourth. In other words, the pattern of fourth-quarter peaks in the stock price would change to a pattern of third-quarter peaks. Pretty soon, investors would see that pattern and begin buying even sooner, perhaps in the second quarter each year. Eventually, the actions of investors trying to buy ahead of any peak in the stock price would cause the seasonal pattern to disappear. So the lesson here is that even if a company’s financial results follow a highly predictable pattern, its stock price will not follow the same pattern (or perhaps even any pattern). If stock prices do exhibit predictable patterns, the actions of investors will tend to eliminate those patterns over time.

A second line of argument helps explain why the seemingly random behavior of Walmart’s stock price (or any stock price) does not imply that the stock market and Walmart’s financial performance are unconnected. Remember that you previously learned that a stock’s price depends on investors’ expectations about the future performance of the company that issued the stock. Prices move up when investors’ expectations become brighter, and prices move down when the opposite occurs. Investors who bought Walmart’s stock way back in 2000 probably expected that over the next 15 years the company’s revenues would grow and that they would peak in the fourth quarter every year. After all, by the year 2000, Walmart had already established a long history of growth, and the seasonal pattern in revenues was well known to the investment community. In other words, much of the performance displayed in  Figure 9.1  would not have surprised investors and therefore would not have moved Walmart’s stock a great deal. What would cause a sudden and potentially large change in Walmart’s stock price is any sign that the firm’s future financial performance would deviate from what investors expected. For example, suppose that in 2015 Walmart’s revenues were not only high in the first quarter (as usual) but that they were even higher than investors had anticipated they would be. In that case, investors would likely raise their expectations about Walmart’s future performance, and the company’s stock price would go up as a result. If Walmart reported financial results that failed to match investors’ expectations, then its stock price would probably fall as investors revised their views about how the company would perform in the future.

The main point here is that stock prices respond to new information. By definition, new information is something that people do not already know and that they do not anticipate. That Walmart’s revenues peak at the end of each year is not new information, so when the peak occurs each year it does not tend to boost the company’s stock price. Only if fourth-quarter revenues are surprising (better or worse than expected) would Walmart’s stock price respond. Because new information is unpredictable, stock price movements are also largely unpredictable. This is the central idea of the  random walk hypothesis , which says that predicting stock price movements is very difficult, if not impossible. We must emphasize here that if stock prices move at random, it is not a sign that the stock market is a casino that lacks any connection to the real business world. Just the opposite is true. The seemingly random behavior of stock prices is a sign that the stock market is processing information quickly and efficiently. In fact, economists say that a market that rapidly and fully incorporates all new information is an  efficient market .

An Advisor’s Perspective

Bob Grace President, Grace Tax Advisory Group

“There is absolutely no connection between what happened yesterday and what will happen tomorrow.”

MyFinanceLab

The Efficient Markets Hypothesis

The notion that stock prices (and prices in other financial markets) rapidly incorporate new information is known formally as the  efficient markets hypothesis (EMH) . An implication of this idea is that it is very difficult for investors, even professional investors, to earn abnormally high returns by identifying undervalued stocks and buying them (or identifying overvalued stocks and selling them). Spotting bargains in the stock market is difficult because if the market is indeed efficient, by the time you have processed the information that leads you to believe that a stock is a good buy, the market has already incorporated that information, and the information is reflected in the stock’s price.

The EMH says that investors should not expect to earn abnormal returns consistently. What constitutes an abnormal return? Previously you learned that there is a positive relation between risk and return. Investments that tend to earn higher returns also tend to be riskier. Therefore, an investment’s expected return is directly related to its risk. An  abnormal return  (also known as alpha) is the difference between an investment’s actual return and its expected return (i.e., the return that it should earn given its risk).

Abnormal return (or alpha)=Actual return − Expected returnAbnormal return (or alpha)=Actual return − Expected returnEquation9.1

One way that investors can estimate the expected return on a stock is to use the capital asset pricing model, or CAPM. Recall that the CAPM says that the expected return on a stock (E(rj )) is equal to the risk-free rate (rrf ) plus the product of the stock’s beta (bj ) and the risk premium on the overall market (rm−rrf)(rm−rrf).

E(rj)=rrf+bj(rm−rrf)E(rj)=rrf+bj(rm−rrf)Equation9.2

Example

Suppose that a particular stock has a beta of 1.0. This means that the stock has average risk and should earn a return that is on average equal to the return on the overall market. Suppose that in a particular year the risk-free rate is 2% and the return on the overall stock market is 10%.  Equation 9.2  tells us that the return that we should expect on this stock is 10%:

E(r)=2% + 1.0(10% − 2%) = 10%E(r)=2% + 1.0(10% − 2%) = 10%

Suppose instead that the stock earned a 12% return. In this case it earns an abnormal positive return (alpha) of 2%:

Abnormal return=Actual return − Expected return=12%−10%=2%Abnormal return= Actual return − Expected return=12% −10%=2%

The EMH says that spotting stocks like this (i.e., stocks that earn positive abnormal returns) on a consistent basis over time is nearly impossible, even for highly sophisticated investors with extensive training.

The efficient markets hypothesis focuses on the extent to which markets incorporate information into prices. The more information that is incorporated into stock prices and the more rapidly that information becomes incorporated into prices, the more efficient the market becomes. One way of characterizing the extent to which markets are efficient is to define different levels of efficiency corresponding to different types of information that prices may reflect. These levels of market efficiency are known as the weak form, the semi-strong form, and the strong form.

Weak Form

The  weak form of the EMH  holds that stock prices fully reflect any relevant information that can be obtained from an analysis of past price movements. If investors study the historical record of stock prices and spot some kind of pattern that seems to repeat, their attempts to exploit that pattern through trading will cause the pattern to disappear over time. We have already described this idea to explain why Walmart’s stock price does not exhibit predictable patterns, even though its revenues show distinct seasonal peaks. In short, the weak form of the EMH says that past data on stock prices are of no use in predicting future price changes. According to this hypothesis, prices follow a random walk, meaning that tomorrow’s price change is unrelated to today’s or yesterday’s price, or that of any other day.

The earliest research on the weak form of market efficiency appeared to confirm the prediction that prices moved at random. Using databases that contained the past prices of listed stocks in the United States, researchers constructed a variety of “trading rules,” such as buying a stock when it hit a 52-week low, and then tested these rules using historical information to see what returns investors following these rules might have earned. The results were encouraging to theorists but not to traders—none of the trading rules earned abnormal returns, but they did generate significant transactions costs. The researchers concluded that investors would do better by purchasing a diversified portfolio and holding it.

Watch Your Behavior

Investors’ Expectations vs. Expected Returns When investors are asked about their future expectations for stock market returns, those expectations are positively correlated with recent market returns. Since we know actual market returns do not have this kind of correlation, it suggests that investors beliefs are not fully rational. In fact, research shows that investors’ expectations are negatively correlated with the predictions of sophisticated financial models.

(Source: Robin Greenwood and Andrei Shleifer, “Expectations of Returns and Expected Returns,” The Review of Financial Studies RFS, 2014, Vol. 27, Iissue 3, pp. 714–746.)

Semi-Strong Form

The  semi-strong form of the EMH  asserts that stock prices fully reflect all relevant information that investors can obtain from any public source. This means that investors cannot consistently earn abnormally high returns using publicly available information such as annual reports and other required filings, analyst recommendations, product reviews, and so on. To illustrate the idea, suppose that you see that a particular firm has just posted its latest financial results online. You read the report and see that the company reported an unexpected surge in profits in the most recent quarter. Should you call your broker and buy some shares? The semi-strong form of the EMH says that by the time you download the annual report, read it, and call your broker, the market price of the stock will have already increased, reflecting the company’s latest good news.

Figure 9.3  comes from a recent research study that tested this form of the EMH. The researchers gathered data on a large number of earnings announcements by

Figure 9.3 Daily Stock Price Reactions Surrounding Positive Earnings News

The figure shows that for a group of companies reporting favorable earnings, abnormal returns are close to 0 leading up to the announcement and beyond 2 days after the announcement. The market responds fully to the new information in 1 or at most 2 days.

(Source: Modified from Andreas Neuhierl, Anna Scherbina, and Bernd Schlusche, “Market Reaction to Corporate Press Releases,” Journal of Financial and Quantitative Analysis, August 2013.)

different companies and tracked the companies’ stock price behavior before and after the announcements. The common factor in all of these announcements was that the companies were reporting good news that their earnings were higher than analysts had expected. In a sense, the question that the researchers were asking was, is it smart to buy the stock of a company that announces this kind of good news?

The horizontal axis of  Figure 9.3  measures time relative to the earnings announcement day. The earnings announcement day is day 0, so day -1 is one day before the announcement and day +1 is one day after the announcement. Keep in mind that many firms release their financial information after the market has closed. This means that the first opportunity for the stock market to incorporate the new information occurs the day after the announcement, on day +1. The vertical axis in the figure measures the average abnormal return exhibited by companies in the sample. The behavior of stock prices exhibited in  Figure 9.3  is very close to what the semi-strong form of the EMH would predict. Observe that leading up to the earnings announcements, the companies in the sample earn returns that are essentially normal (i.e., the abnormal return is 0, so the actual return matches the expected return). However, from day 0 to day +1, the average company in the sample earned an abnormal return of about 2.5%, with an additional 1% abnormal return occurring from day +1 to day +2. Beyond that point, however, abnormal returns quickly revert to 0%. In other words, the market quickly (in a day or at most two days) incorporates the good news from earnings announcements.

Many tests of semi-strong efficiency have examined how stock prices respond before and after particular types of news. One study looked at four companies that were major contractors in the space shuttle program. When the shuttle Challenger exploded shortly after liftoff in 1986, the stock prices of all four companies fell, but the one that fell the most was Morton Thiokol. That company made the booster rockets that lifted the shuttle into orbit, and months after the accident occurred, an investigation concluded that a problem with the O-rings in these rockets had caused the accident. In other words, the market’s initial reaction within minutes of the accident seemed to point to the same conclusion as the subsequent investigation.

Numerous studies have examined the investment performance of professional investors such as mutual fund managers. Some people argue that although the stock market may be efficient enough to prevent individual investors from earning abnormally high returns, surely professional investors who have advanced training in investments and who spend their entire professional lives thinking about investments can perform better. The conclusions from research in this area are not unanimous, but most studies find that even professional investors struggle to earn abnormal returns on a consistent basis. On average, mutual fund managers do not earn returns that beat the market average by a sufficient degree to cover the fees that they charge investors. Furthermore, there is not much persistence in mutual fund returns. In other words, fund managers who have above-average returns one year do not have a very high likelihood of generating above-average returns the next year.

The overwhelming evidence indicates that stock prices react very rapidly to any important new information, which makes it very hard for investors (individuals or professionals) to “beat the market.” Unless you hear about an event almost as soon as it happens, the stock price will adjust to the news before you can trade the stock.

Investor Facts

Robots and Efficient Markets Professional investors have to work even harder these days to trade on information before prices react. For example, the hedge fund Two Sigma Investments, LLC, uses computer programs to sift through real-time data from sources like Twitter to identify emerging news about stocks and execute trades within seconds.

Strong Form

The  strong form of the EMH  holds that the stock market can rapidly incorporate new information even if it is not disseminated through public sources. It states that stock prices rapidly adjust to any information, even if it isn’t available to every investor.

One type of private information is the kind obtained by corporate insiders, such as officers and directors of a corporation. They have access to valuable information about major strategic and tactical decisions the company makes. They also have detailed information about the financial state of the firm that may not be available to other shareholders. Insiders are generally prohibited from trading the shares of their employers prior to major news releases. However, at other times corporate insiders may legally trade shares of stock in their company, if they report the transactions to the Securities and Exchange Commission (SEC). When insiders file the required forms with the SEC, they are quickly made available to the public via the Internet. Several studies of corporate insiders find that their trades are particularly well timed, meaning that they tend to buy before significant price increases and sell prior to big declines. This, of course, is contrary to what you’d expect to find if the strong form of the EMH were true.

Insiders and other market participants occasionally have inside—nonpublic— information that they obtained or traded on illegally. With this information, they can gain an unfair advantage that permits them to earn an abnormal return. Clearly, those who violate the law when they trade have an unfair advantage. Empirical research has confirmed that those with such inside information do indeed have an opportunity to earn an abnormal return—but there might be an awfully high price attached, such as spending time in prison, if they’re caught.

Arbitrage and Efficient Markets

Closely linked to the notion of efficient markets is the concept of arbitrage.  Arbitrage  is a type of transaction in which an investor simultaneously buys and sells the same asset at different prices to earn an instant, risk-free profit. Let us give a simple example to illustrate the concept of arbitrage before examining the concept more closely.

Example

Suppose that banks in New York City will convert dollars into euros (or vice versa) at an exchange rate of one dollar per euro. In London, however, banks are exchanging dollars and euros at the rate of $1.25 dollars per euro. Notice that given these exchange rates, one euro is more valuable in London than in New York. Another way to say this is that euros are relatively cheap in New York and relatively expensive in London. This means that we have the identical asset (the euro) trading in different markets at different prices, so we would say that this presents an arbitrage opportunity. A trader could exploit this opportunity by buying cheap euros in New York and selling them in London as follows:

1. At a New York bank, use $1 million to buy €1 million. Remember, in New York, €1 is worth $1. Of course, if many traders begin buying euros in New York, the price of the euro will tend to rise in this market.

2. At a bank in London, sell the €1 million in exchange for $1.25 million at the prevailing exchange rate of $1.25 dollars per euro. Again, if many investors begin selling euros on the London market, then the price of the euro should begin to fall there.

3. Simply by purchasing euros in New York and selling them in London, the trader makes an instant profit of $250,000. But as the price of the euro rises in New York and falls in London, the opportunity to profit from these transactions will shrink, and ultimately vanish.

Consider how the definition of arbitrage applies to this example. First, arbitrage occurs when an investor simultaneously buys and sells the same asset. In this example, the underlying asset is just a currency, so the investor is buying euros in New York and selling them in London. The underlying asset is literally the same thing in each market. Furthermore, the purchase in New York and the sale in London can occur simultaneously through electronic transactions. The second part of the definition says that the purchase and sale must occur at different prices, and clearly that is the case here. In New York, €1 is worth $1, but in London it is worth $1.25. Finally, the definition of arbitrage says that the profit earned must be instantaneous and free of risk. Again, this example seems to satisfy those conditions because the trader earns the profit as soon as the currency trades take place, and because they take place essentially at the same time, there would appear to be no risk involved.

In the real world, naturally, we do not see large differences in currency prices in different markets. The price quoted in New York and in London will be virtually the same. If that were not true, arbitragers would exploit the price differences and, through their buying and selling transactions, push the prices closer together until no arbitrage opportunity remained. Economists refer to this as the “no arbitrage” condition, which simply means that prices in financial markets will quickly adjust to eliminate arbitrage opportunities.

Believers in efficient markets often cite arbitrage as a key mechanism that makes markets efficient. For example, suppose that the true intrinsic value of Pepsi stock is $100 per share but for some reason investors have been irrationally pessimistic about the company and have driven its price down to $80. To efficient markets advocates, this represents a kind of arbitrage opportunity. Smart investors will buy the undervalued shares of Pepsi and to hedge their bets they will simultaneously sell shares in another similar company, like Coca-Cola, for example. The buying pressure will cause Pepsi shares to move back toward their intrinsic value of $100, so in the end the market price and the intrinsic value of Pepsi are equal.

Arbitrage is a powerful force, and it plays a very important role in setting the prices of many types of securities, but there are limits to arbitrage. In the Pepsi example, the arbitrage process involves not only buying Pepsi shares but also selling something else that is very similar to Pepsi. Although Pepsi and Coca-Cola are similar stocks, one cannot really argue that they are identical investments. They are imperfect substitutes for one another, so even if Pepsi is mispriced, buying Pepsi and selling Coca-Cola may be risky. In addition, making these trades is costly, especially for an investor who wants to sell Coca-Cola but does not own any shares. That investor must engage in a short sale, which means that the investor must borrow Coca-Cola shares from someone else before selling them. Short sales often carry high transactions costs, and at times, shorting a particular stock is just not possible because there is no one willing to lend the required shares.

Another risk associated with arbitrage has to do with what created the apparent arbitrage opportunity in the first place. We presumed that some investors were irrationally pessimistic about Pepsi, and their pessimism caused Pepsi to be undervalued. It may be true that some smart investors can spot this situation, but what if other traders continue to be pessimistic or become even more pessimistic about Pepsi? In that case, there is no absolute guarantee that the actions of smart traders (who are buying Pepsi) will swamp the trades of irrational traders (who continue to sell Pepsi) and thereby move Pepsi’s stock price toward its intrinsic value. Instead, Pepsi could become more undervalued, which would cause losses for the “smart” traders conducting the arbitrage trades.

To sum up, there is considerable evidence suggesting that the stock market is relatively efficient, and there are compelling reasons to expect that to be the case. Nevertheless, some contrary evidence exists, and it is to that evidence that we now turn.

Watch Your Behavior

It’s Hard to Beat the Market The EMH says that trading in and out of securities doesn’t make much sense. That’s exactly the conclusion from a study of over 66,000 investors, grouped according to how often they traded. Buy-and-hold investors, whose annual trades amounted to just 2% of the value of their portfolios, earned annual returns that were seven percentage points higher than the most active traders. Heavy traders churned through stocks so fast that they replaced their entire portfolios more than twice each year.

(Source: Brad Barber and Terrance Odean, “Trading Is Hazardous to Your Wealth: The Common Stock Investment Performance of Individual Investors,” Journal of Finance, December 2002.)

Market Anomalies

Despite considerable evidence in support of the EMH, researchers have uncovered some patterns that seem inconsistent with the theory. Collectively, this body of puzzling evidence is known as  market anomalies , a name that itself suggests that there is less evidence contradicting the EMH than there is in support of it. What all of these anomalies have in common is that they reveal patterns or trading strategies that, at least in hindsight, earned higher returns than would be expected in efficient markets.

Calendar Effects

One widely cited anomaly is the calendar effect, which holds that stock returns may be closely tied to the time of the year or the time of the week. That is, certain months or days of the week may produce better investment results than others. The most famous of the calendar anomalies is the January effect, which is a tendency for small-cap stocks to outperform large-cap stocks by an unusually wide margin in the month of January. One possible explanation for this pattern has to do with taxes. Under certain conditions, investors can deduct investment losses when calculating their federal income taxes. Thus, there is an incentive for investors to sell stocks that have gone down in value during the year, and investors who recognize that incentive are particularly likely to sell in December as the tax year comes to a close. Think about what happens to the market capitalization of a firm when its stock falls during the year—the market cap gets smaller. Thus, if investors have a tax incentive to sell their loser stocks in December, and if these stocks by definition tend to be smaller than average, then their prices may be temporarily depressed due to December tax selling, and they may rebound in January. As plausible as this explanation may sound, there is at best mixed evidence that it can account for the puzzling behavior of small stocks in January.

Small-Firm Effect

Another anomaly is the small-firm effect, or size effect, which states that small firms tend to earn positive abnormal returns of as much as 5% to 6% per year. Indeed, several studies have shown that small firms (or small-cap stocks) earn higher returns than large firms (or large-cap stocks), even after taking into account the higher betas typical of most small firms. This tendency has been documented in the United States as well as in many stock markets around the world and is not confined to the month of January.

Post Earnings Announcement Drift (or Momentum)

Another market anomaly has to do with how stock prices react to earnings announcements. In  Figure 9.3  we showed the results of a study that tracked stock returns around earnings announcements. In that study, stocks reporting good earnings exhibited abnormal returns for a day or two, but those abnormal returns quickly dissipated. However, several older studies reported a tendency for stocks to “drift” after earnings announcements in the same direction as the initial reaction. In other words, when companies reported better-than-expected earnings, their stock prices jumped immediately, earning positive abnormal returns. But surprisingly, these firms’ stock prices continued to earn positive abnormal returns for weeks or even months after the earnings announcements. Similarly, firms reporting bad earnings earned negative abnormal returns that continued for several months beyond the initial announcement. This seems to indicate that investors underreact to the information in earnings announcements. When firms report good news, investors don’t realize just how good the news is, and similarly, when bad news comes, investors don’t fully appreciate how bad the news is, so stock prices take a long time to fully adjust to a new level. This pattern seems to create an opportunity for investors to earn abnormal returns by purchasing stocks that have recently issued good earnings news or by short selling stocks that have recently delivered poor earnings results.

Figure 9.4  illustrates the post earnings announcement drift pattern. The horizontal axis marks time measured in weeks relative to an earnings announcement, and the vertical axis measures the cumulative abnormal return from 52 weeks prior to the earnings announcement to 52 weeks after the announcement. The earnings announcement occurs at week 0. Two types of companies are tracked in the figure—companies that announce better-than-expected earnings and companies that announce worse-than-expected earnings. The blue line in the figure plots cumulative abnormal returns (i.e., the abnormal return over the entire period) earned by the sample of “good news” stocks, and the red line tracks abnormal returns for the “bad news” stocks. Notice that when firms announce good news, their stock prices react quickly, as indicated by the jump in the blue line at week 0. Similarly, when firms reveal that their earnings are below investors’ expectations, their stock prices move down almost immediately, as shown by the drop in the red line at week 0. That rapid initial reaction is exactly the pattern that an efficient market should produce.

However, it appears that investors underreact to the news contained in earnings announcements. Observe that after the initial reaction to the earnings announcement,

Figure 9.4 Post Earnings Announcement Drift

When firms announce better-than-expected earnings, their stock prices jump quickly, as the EMH would predict, but contrary to the EMH, stock prices continue to drift upward at an abnormally rapid clip over the next year or so. The same thing happens in reverse when firms announce poor earnings.

both the blue and red lines exhibit trends, with the blue line slowly rising and the red line slowly falling. This means that the initial reaction to the earnings announcement was not large enough, and stock prices are adjusting slowly to the information contained in the earnings announcement. The slow adjustment process creates an opportunity for investors. For example, after a company announces positive earnings news (i.e., investors do not have to anticipate what the content of the announcement will be), investors who buy the stock earn significant abnormal returns. Looking at the blue line in  Figure 9.4  , you can see that the amount of drift is roughly 2% over the 52 weeks following the earnings announcement. In other words,  Figure 9.4  suggests that investors who closely monitor earnings announcements and buy stocks after firms announce better-than-expected returns will earn a return that is about 2% above normal (i.e., 2% greater than one would expect given the risk of the stocks being purchased). Investors can also make money by short selling the shares of companies that announce poor earnings results. The drift in stock prices following the earnings announcement is not consistent with the predictions of the EMH.

A slight variation on this story is known as the momentum anomaly. In physics, momentum refers to the tendency of an object in motion to continue moving or the tendency of an object at rest to remain at rest. Applied to stocks, momentum refers to the tendency for stocks that have gone up recently to keep going up or the tendency for stocks that have gone down recently to continue going down. The connection to earnings announcement drift is easy to see. When a company has a particularly good quarter, it is common for some of the good news to leak out into the market before the official earnings announcement. So leading up to the earnings release, it is common to see the stock price moving up, just as the blue line in  Figure 9.4  rises ahead of the earnings news. As we’ve already discussed, when the firm releases the news that it has had a very strong quarter, the price goes up more, but then it continues to drift up for weeks. Taking the entire pattern into account, we observe that before a company releases very good earnings news, its stock price has gone up, and then it keeps going up after the earnings announcement. Hence, these stocks display positive momentum. The same thing happens in reverse for companies that have particularly bad quarters. Some of the bad news leaks out early, and the stock goes down (see the red line in  Figure 9.4  ), but then the stock continues to go down after the announcement.

The Value Effect

According to the value effect, the best way to make money in the market is to buy stocks that have relatively low prices relative to some measure of fundamental value such as book value or earnings. An investor following a value strategy might calculate the P/E ratio or the ratio of market value to book value for many stocks, and then buy the stocks with the lowest ratios (and perhaps short sell the stocks with high P/E or market-to-book ratios). Studies have shown that, on average, value stocks outperform stocks with high P/E or market-to-book ratios (so-called growth stocks). This pattern has repeated itself decade after decade in the United States and in most stock markets around the world.

Possible Explanations

Each new discovery of an anomaly that appears to violate the EMH prompts a flurry of research that offers rational explanations for the pattern observed. The most common explanation for market anomalies is that the stocks that earn abnormally high returns are simply riskier than other stocks, so the higher returns reflect a risk premium rather than mispricing by the market. For example, most academics and practitioners would agree that small firms are riskier than large firms, so it is not surprising that small stocks earn higher returns. The real question is, how much riskier are small firms, and how large should the risk premium be on those securities? According to the CAPM, if a small stock has a beta of 2.0 and a large stock has a beta of 1.0, the small stock should earn roughly twice the risk premium (over Treasury bills) that the large stock earns. The reason that the small-firm effect is known as an anomaly is that small stocks seem to earn higher returns than their betas can justify. Believers in the EMH argue that beta is an imperfect measure of risk and that if a better risk measure were available, the difference in returns between small and large stocks could be fully attributed to differences in risk.

Another explanation for market anomalies is that even in an efficient market where prices move essentially at random, some trading rules may appear to earn abnormally high returns simply as a matter of chance. For example, one of the more amusing market anomalies is known as the Super Bowl anomaly. This anomaly says that if the team winning the Super Bowl in a particular year is one of the original National Football League teams (prior to the merger with the old American Football League), then the stock market will rise. Otherwise, the stock market will fall. This “trading rule” correctly predicted the direction of the market more than 80% of the time in the last 48 years. But should investors rely on it in the future? Most people would agree that the connection between the Super Bowl winner and the stock market is purely a matter of chance and is unlikely to exhibit a similar track record in the next 48 years. Some EMH advocates believe that most market anomalies are similarly just an artifact of random chance. However, this explanation is less persuasive in the face of evidence that anomalies such as the small-firm effect, momentum, and the value effect appear in most markets around the world.

The discovery of these and other anomalies led to the development of an entirely new way of viewing the workings of financial markets that has come to be known as  behavioral finance . In contrast to traditional finance, which starts with the assumption that investors, managers, and other actors in financial markets are rational, behavioral finance posits that market participants make systematic mistakes and that those mistakes are inextricably linked to cognitive biases that are hard-wired into human nature. We now turn to a discussion of the basic tenets of behavioral finance and how they may help explain market anomalies.

Concepts in Review

Answers available at  http://www.pearsonhighered.com/smart

1. 9.1 What is the random walk hypothesis, and how does it apply to stocks? What is an efficient market? How can a market be efficient if its prices behave in a random fashion?

2. 9.2 Explain why it is difficult, if not impossible, to consistently outperform an efficient market.

a. Does this mean that high rates of return are not available in the stock market?

b. How can an investor earn a high rate of return in an efficient market?

3. 9.3 What are market anomalies and how do they come about? Do they support or refute the EMH? Briefly describe each of the following:

a. The January effect

b. The size effect

c. The value effect

Behavioral Finance: A Challenge to the Efficient Markets Hypothesis

1. LG 3

2. LG 4

For more than 40 years, the efficient markets hypothesis has been an influential force in financial markets. The notion that asset prices fully reflect all available information is supported by a large body of academic research. In practitioner circles, supporters of market efficiency include John Bogle of Vanguard, who helped pioneer the development of a special type of mutual fund known as an index fund. Managers of index funds don’t try to pick individual stocks or bonds because they assume that the market is efficient. They recognize that any time and energy spent researching individual securities will merely serve to increase the fund’s expenses, which will drag down investors’ returns.

Although considerable evidence supports the concept of market efficiency, an increasing number of academic studies have begun to cast doubt on the EMH. This research documents various anomalies and draws from research on cognitive psychology to offer explanations for the anomalies. One notable event that acknowledged the importance of this field was the awarding of the 2002 Nobel Prize in economics to Daniel Kahneman, whose work integrated insights from psychology and economics. In addition to academic studies, some professional money managers are also incorporating concepts from behavioral finance into their construction and management of portfolios.

Investor Facts

Behavioral Funds Underperform Too A recent study tracked the performance of 22 U.S. mutual funds that claimed to use the findings from behavioral finance to guide their stock selections. From 2007 to 2013, these funds as a group performed slightly worse than average, generating negative abnormal returns of less than 0.20% per month.

(Source: Nikolaos Philippas, “Did Behavioral Funds Exploit Market Inefficiencies during or after the Financial Crisis?” Multinational Finance Journal 2014, Vol. 18, Iss.1/2, pp. 85–138.)

Investor Behavior and Security Prices

Researchers in behavioral finance believe that investors’ decisions are affected by a number of psychological biases that lead investors to make systematic, predictable mistakes in certain decision-making situations. These mistakes, in turn, may lead to predictable patterns in stock prices that create opportunities for other investors to earn abnormally high profits without accepting abnormally high risk. Let’s now take a look at some of the behavioral factors that might influence the actions of investors.

Overconfidence and Self-Attribution Bias

Research in psychology provides overwhelming evidence that, on average, people tend to exhibit  overconfidence , putting too much faith in their own ability to perform complex tasks. Try this experiment. The next time you are in a large group, ask people to indicate whether they believe they have above average, average, or below average skill in driving a car. What you will probably find is that a majority of the group believes that they have above-average ability, and almost no one will lay claim to having below-average skill. But simply by the definition of average, some people have to be above average and some must be below average. Therefore, at least some people in the group are overconfident in their driving ability.

Closely linked to overconfidence is a phenomenon known as self-attribution bias.  Self-attribution bias  roughly means that when something good happens, individuals attribute that outcome to actions that they have taken, but when something bad happens, they attribute it to bad luck or external factors beyond their control. The connection to overconfidence is straightforward. An individual takes an action or makes a decision that leads to a favorable outcome. Self-attribution bias causes the individual to discount the role that chance may have played in determining the outcome and to put too much emphasis on his or her actions as the cause. This causes the individual to become overconfident.

Watch Your Behavior

Overconfidence and Acquisitions Warren Buffett summarized the role of overconfidence in acquisitions in one of his famous letters to shareholders: “Many managements apparently were overexposed in impressionable childhood years to the story in which the imprisoned handsome prince is released from a toad’s body by a kiss from a beautiful princess. Consequently, they are certain their managerial kiss will do wonders for the profitability of Company Target. . . . We’ve observed many kisses but very few miracles. Nevertheless, many managerial princesses remain serenely confident about the future potency of their kisses—even after their corporate backyards are knee-deep in unresponsive toads.” (Source: http://www .berkshirehathaway.com/owners.html.)

What effects do overconfidence and self-attribution bias have in the investments realm? Consider an individual investor, or even a professional money manager, who analyzes stocks to determine which ones are overvalued and which are bargains. Suppose in a particular year the investor’s portfolio earns very high returns. Perhaps the high returns are largely due to a booming stock market, but perhaps in addition the investor’s stock picks performed even better than the overall market. Is this the result of good fortune or good analysis? It’s not easy to separate the roles of skill and luck, but most investors would probably attribute the favorable outcome to their own investing prowess. What is the consequence if investors mistakenly attribute investment success to their own skill? One study found that investors whose portfolios had outperformed the market in the past subsequently increased their trading activity. After beating the overall market average by 2% per year for several years, these investors increased their trading activity more than 70%. The increase in trading led to much higher transactions costs and much lower returns. The same group of investors trailed the market by 3% per year after increasing their trading activity.

This tendency is not confined to individual investors. A recent study found that CEOs exhibit similar behavior when they undertake acquisitions of other firms. When a CEO acquires a firm and the acquisition target performs well, the CEO is more likely to acquire a second firm. The CEO is also more likely to buy more shares in his or her employer’s stock prior to the next acquisition. But these second acquisitions actually destroy shareholder value on average. In other words, it appears that CEOs become overconfident regarding their ability to acquire other firms and run them profitably.

Loss Aversion

Here’s an interesting series of questions. Suppose you have just won $8,500 in a game of chance. You can walk away with your winnings or you can risk them. If you take the risk, there is a 90% chance that you will win an additional $1,500, but there is a 10% chance that you will lose everything. Would you walk away or gamble? Most people who are asked this question say that they would take the $8,500—the sure thing. They say this even though the expected value from the additional gamble is $500. That is,

Expected value = (Probability of gain) × (Amount of gain) − (Probability of loss) × (Amount of loss ) = 0.90 × $1,500 − 0.10 × $8,500 = $500––––––––––Expected value = (Probability of gain) × (Amount of gain) − (Probability of loss) × (Amount of loss ) = 0.90 × $1,500 − 0.10 × $8,500 = $500__

In this case, the decision to take the $8,500 indicates that the individual making that choice is risk averse. The risk of losing $8,500 isn’t worth the expected $500 gain.

However, if we reframe the question, most people respond differently. Suppose you have already lost $8,500 in a game of chance. You can walk away and cut your losses or you can gamble again. If you gamble, there is a 90% chance that you will lose $1,500 more, but there is a 10% chance that you will win $8,500, thus entirely reversing your initial loss. When confronted with this choice, most people say that they will take the risk to try to “get even,” even though the expected value of this gamble is −$500.

Expected value = 0.10 × $8,500 − 0.90 × $1,500 =−$500––––––––––Expected value = 0.10 × $8,500 − 0.90 × $1,500 =−$500__

In this case, people are exhibiting risk-seeking behavior. They are accepting a risk that they do not have to take, and it is a risk that has a negative expected return.

Famous Failures in Finance Loss Aversion and Trading Volume

When people are loss averse, they are reluctant to sell investments that have lost value because doing so forces them to realize the loss. But if investors are reluctant to sell when prices are falling, trading activity can dry up. That was a finding from a study of residential real estate activity over several market cycles in Boston. Researchers found that when market prices were rising, homeowners were generally willing to sell their properties at market value. But when price declines left homeowners in a position such that the market value of their home was less than what they had paid for it, homeowners exhibited a tendency to set asking prices above the true market value. For these homeowners, selling at the current market price would mean recognizing a loss, something homeowners were very averse to do. As a consequence, overpriced homes sat on the market month after month, with very few transactions taking place.

(Source: David Genesove and Christopher Mayer, “Loss Aversion and Seller Behavior: Evidence from the Housing Market,” Quarterly Journal of Economics, 2001, Vol. 116, No. 4, pp. 1233–1260.)

In behavioral finance, the tendency to exhibit risk-averse behavior when confronting gains and risk-seeking behavior when confronting losses is called  loss aversion . Loss aversion simply means that people feel the pain of loss more acutely than the pleasure of gain. In an investments context, loss aversion can lead people to hold onto investments that have lost money longer than they should. In fact, numerous studies have documented that when investors want to sell a stock in their portfolio, they are much more likely to sell a stock that has gone up in value than one that has fallen. Other studies have documented a tendency for the stocks that investors sell (i.e., stocks that have gone up) to perform better than the stocks that they choose to hold (i.e., stocks that have lost value).

Representativeness

Overreaction

In an interesting experiment, six people were asked to flip a coin 20 times and count the number of heads that came up. Six others were asked to imagine flipping a coin 20 times and write down the sequence of heads and tails that might occur. The table below shows the results reported by each group.

Group	Subject	Number of Heads	Group	Subject	Number of Heads
A	1	10	B	1	6
	2	10		2	13
	3	8		3	7
	4	10		4	11
	5	10		5	8
	6	10		6	14
	Average	9.7		Average	9.8

Looking at the responses from individuals in each group, which group do you think actually flipped coins, and which imagined doing so?

The answer is that Group A only imagined flipping coins. Notice that almost everyone in the group said they expected to obtain 10 heads in 20 flips, but in the group that actually tossed the coins, the number of heads varied widely, from 6 to 14. What accounts for the differences between the two groups?

Representativeness  refers to cognitive biases that occur because people have difficulty thinking about randomness in outcomes. Subjects in Group A assume (correctly) that the probability of obtaining a heads on any single flip of a coin is 50%, but they also assume (incorrectly) that this means that in 20 flips of a coin, it is very likely that heads will come up exactly 10 times. It is true that 10 is the average number of heads that one should expect, and notice that the average number of heads flipped by both groups was about 10. However, individual results vary quite a bit around that average. As the results of Group B’s coin flips clearly show, it is rather unusual to obtain exactly 10 heads in 20 flips. Lots of other outcomes are quite likely.

Consider this analogy. Suppose picking stocks is like flipping coins in the sense that if markets are efficient, when you buy a stock there is about a 50% chance that it will do better than average (let’s call that outcome heads) and a 50% chance that it will do worse than average (call that tails). Investors in Group A would appear to believe that if one buys 20 stocks, it is very likely that the outcome of that portfolio will be average because 10 stocks will do better than average and 10 will perform worse than average. However, we know from Group B that it is quite likely that a portfolio of 20 stocks could perform much better (more than 10 heads) or much worse (fewer than 10 heads) than average. In other words, even in an efficient market, some portfolios will do very well while others will lag behind.

Subjects in this experiment were also asked to report whether they obtained a “string” of five heads or five tails in a row in the course of flipping a coin 20 times. Here are their answers to that question.

Group	Subject	Five Heads or Tails in a Row?	Group	Subject	Five Heads or Tails in a Row
A	1	no	B	1	yes
	2	no		2	yes
	3	no		3	no
	4	no		4	yes
	5	no		5	no
	6	no		6	yes

Notice that among the subjects in Group B, those who actually flipped coins 20 times, obtaining a string of five flips in a row with the same outcome (either five heads or five tails in a row) was relatively common. But subjects in Group A did not imagine that they would see a string of five consecutive identical outcomes. Why not? These subjects know that there is a 50% chance of getting heads (or tails) in every flip, so they imagine that on a series of flips they will see a kind of oscillation in outcomes. That is, they appear to believe that a sequence of alternating heads and tails is more likely than a sequence that has several heads (or tails) in a row. This is representativeness at work again. Subjects in Group A dramatically underestimate the likelihood of getting the coin to come up heads or tails several times in a row because they think a 50-50 gamble is much more likely to result in alternating heads and tails.

Now consider how this feature of representativeness can influence the behavior of investors. Think about investors who are trying to decide which mutual fund to invest in. The EMH says that for a mutual fund to earn an above-average return is more a matter of luck than of skill, so any particular fund manager has roughly a 50% chance of beating the market in a particular year. There are thousands of mutual funds to choose from, so even if mutual fund performance is as much due to luck as it is to skill, there will be some fund managers who “beat the market” several years in a row, just as there were several coin flippers in Group B who flipped five heads in a row. However, if investors misinterpret randomness like the subjects in Group A did, they will believe that it is very unlikely for a fund manager to have a string of several good years in a row if the market is efficient. Put another way, these investors will interpret a string of good years as a sign that the market is not efficient, at least not for the fund manager achieving that string of good performance. Therefore, when investors see a manager who has delivered better-than-average returns for several years in a row, they may mistakenly attribute that record to skill. Research shows that investors overreact to a string of good performance and pour money into successful funds, enriching the fund managers but not necessarily themselves. Apparently, many investors see a string of good performance and overestimate the likelihood that the trend will continue. Investors overreact to the past performance of funds, even though there is little objective evidence that past performance is a good predictor of future success.

Watch Your Behavior

Chasing Returns Hurts Investors Individual investors tend to “chase” mutual fund returns in the sense that they buy funds that have exhibited recent good performance and sell funds that have poor recent performance. A recent study found that this behavior reduces the average investor’s return by 1.5% per year.

This logic may provide a behavioral explanation for the value phenomenon cited earlier. Recall that value stocks are stocks that have low prices relative to earnings or book value. These stocks generally display rather poor past performance—several years of declining prices is what puts these stocks in the value category. Similarly, growth stocks, stocks with high prices relative to earnings or book value, generally have very good past performance. One of the earliest studies of the value effect studied the results of a very simple trading rule. Each year, researchers sorted all stocks based on their cumulative performance in the previous three years. The trading rule was to buy the stocks that had performed worst (the value stocks) and sell short the stocks that had performed best (the growth stocks). Researchers discovered that this strategy earned returns that beat the market by 8% per year! Why would such a simple trading rule that anyone could follow work so well?

An Advisor’s Perspective

Tom Riquier Owner, The Retirement Center

“When you’re most frightened is probably the best time to be buying.”

MyFinanceLab

The researchers argued that it was due to representativeness. To be specific, they proposed that investors who watched particular stocks decline in value for three years in a row overreacted to those events by deciding that the trend would continue indefinitely, so they bid the prices of these stocks below their true values. Similarly, after watching other stocks do very well several years in a row, investors overreacted to that trend by naively assuming that this excellent performance would continue, and they bid up the prices of these stocks above their true values. Over time, the firms that had been performing poorly surprised investors by rebounding, and the firms that had been earning spectacular returns failed to sustain that performance. As a result, past price trends reversed themselves, and value investors made money.

Individual investors are not the only participants in markets likely to be affected by representativeness. Consider a firm that is looking to make an acquisition. What makes an acquisition target attractive? One criterion might be recent increases in sales and earnings. Would acquirers be wise to pay a premium to acquire a firm that has been growing faster than its competitors in recent years? The research evidence says no. There is almost no correlation between how fast firms have grown in the past and how fast they will grow in the future. In fact, that is a fundamental prediction of basic economic theory. When one firm enjoys great success in a particular market, other firms will enter the industry. Competition makes it more difficult for firms to sustain the high growth that

Famous Failures in Finance Buying High and Selling Low

Research by the Federal Reserve and the University of Michigan suggests that individual investors, particularly those with lower incomes and wealth, displayed particularly poor timing with their investment decisions before, during, and after the sharp market downturn in 2008. Data from the Fed’s triennial Survey of Consumer Finance shows that as the stock market rose from 2004 to 2007, the percentage of lower-income households who owned stocks climbed. However, from 2007 to 2010, a period containing a steep drop in stock values, the percentage of households owning stocks dropped, and that drop was steepest among households with lower incomes and wealth. The percentage of lower-income households owning stocks continued to fall from 2010 to 2013, while the stock market boomed. In contrast, the percentage of households with higher incomes and greater wealth who owned stocks rose from 2010 to 2013. In other words, the rich got richer, in part because the slump in stocks in 2008 did not deter them from continuing to invest in the market. Less wealthy households bought stocks when market values were high, sold them when the market crashed, and failed to benefit from the subsequent stock market recovery.

(Source: Josh Zumbrun, “Bad Stock-Market Timing Fueled Wealth Disparity,”  http://www.wsj.com/articles/bad-stock-market-timing-fueled-wealth-disparity-1414355341 , accessed 6/26/2015.)

attracted new entrants in the first place. Yet there is ample evidence that managers do pay a larger premium when they acquire firms that experienced rapid growth prior to the acquisition, even though the prospect of sustaining the growth is low.

Underreaction

In certain instances, representativeness can cause investors to underreact to new information. Consider this problem from statistics. On a table are 100 sacks, each of which contains 1,000 poker chips. Forty-five of these sacks contain 70% black chips and 30% red chips. The other 55 bags hold 70% red chips and 30% black chips. If you pick one bag at random, what is the likelihood that it will contain mostly black chips?

Most people get this answer right. If 45 out of 100 bags contain mostly black chips, then the probability of picking a bag at random that has mostly black chips is 45%. Here is a much harder problem. Suppose you choose one bag at random and then take out 12 chips, without looking at the others. Of the 12 chips that you pull out, 8 are black and 4 are red. What is the probability that the bag you picked contains mostly black chips?

Watch Your Behavior

Who Underreacts to News? A recent study found that it is primarily individual investors who underreact to information such as earnings announcements. For example, after firms release good earnings news, individuals tend to sell their shares too quickly before prices have risen high enough to incorporate the new information. Who’s buying these shares from individuals? Professional investors like mutual fund managers.

Intuitively, people know that if the sample of 12 chips taken from the bag has a majority of black chips, then that means the probability that the bag has mostly black chips is higher than in the first problem where we select a bag at random and learn nothing more about it. But how much higher? Few people come close to guessing that the probability is over 95%! In other words, people tend to underreact to the new information they obtain in the second version of the question.

Let’s make an analogy between drawing poker chips out of a bag and reading firms’ earnings announcements. Earnings announcements contain a mix of good and bad news that varies over time. When a company announces particularly good (or bad) news, representativeness may cause investors to underreact to the new information. That is, investors may not appreciate that very good earnings news this quarter probably means the likelihood of good news next quarter has gone up (and vice versa for bad news this quarter). When the firm announces the next quarter’s earnings, investors are surprised by how positive the news is, and the firm’s stock price goes up again. That could explain the post earnings announcement drift (or momentum) phenomenon discussed earlier.

A careful reader may object that we have asserted that representativeness can lead to both overreaction (in the case of value stocks) and underreaction (in the case of momentum). Keep in mind that there are important differences in the nature of the information that investors are reacting to in each case. In the value phenomenon, investors see a common string of information—several good years or several bad years in a row. This causes them to discount the role of chance in the outcome and overreact to the series of events. In the case of earnings announcement drift, investors are responding to a single new piece of information that is extreme—particularly good or particularly bad. In that case, representativeness may lead investors to underreact to the new information they’ve received.

Narrow Framing

Many people tend to analyze a situation in isolation, while ignoring the larger context. This behavior is called  narrow framing . A common example in investments relates to the asset allocation decisions that investors make in their retirement plans. The table below summarizes the retirement savings plans offered to employees of two firms. Firm A offers its employees two options for investing retirement savings—a stock fund and a bond fund. Firm B also offer two options—a stock fund and a blended fund that holds 50% stocks and 50% bonds.

Fund Offered	Company A	Company B
Stock fund (100% stocks)	Yes	Yes
Bond fund (100% bonds)	Yes	Not available
Blended fund (50% stocks, 50% bonds)	Not available	Yes

Research shows that many investors view this decision through the narrow frame of two choices, and they follow a simple guideline—put 50% into one fund and 50% into the other. It is as if investors know that they should diversify, so they divide their investments equally between the available options. However, investors seemingly fail to recognize how the asset allocation of the individual funds influences the resulting composition of their overall portfolios. The narrow frame (splitting money evenly between two funds) combined with the options offered by each company produces an odd outcome. Employees of Company A who divide their money between the stock fund and the bond fund will wind up with portfolios containing 50% stocks and 50% bonds. Employees of Company B also divide their money equally between the two funds, but in this case the two funds are the stock fund and the blended fund. Splitting money equally between those options results in an overall portfolio allocation of 75% stocks and 25% bonds. The retirement portfolios held by employees of Company B are much riskier than those held by workers at Company A, but not necessarily because Company B’s employees prefer to take more risk. Instead, framing influences the risk of their portfolios.

Belief Perseverance

People typically ignore information that conflicts with their existing beliefs, a phenomenon called  belief perseverance . If they believe a stock is good and purchase it, for example, they later tend to discount any signs of trouble. In many cases, they even avoid gathering new information for fear it will contradict their initial opinion. It would be better to view each stock owned as a “new” stock when periodically reviewing a portfolio and to ask whether the information available at that time would cause you to buy or sell the stock.

Anchoring

Anchoring  refers to a phenomenon in which individuals attempting to predict or estimate some unknown quantity place too much weight on information that they have at hand, even when that information is not particularly relevant. For example, it is reasonably well known that a firm’s past rate of growth in revenues is a very poor predictor of its future growth rate. Even so, when individuals are ask to predict the sales growth rate for a firm, if they are given information about the firm’s past growth rate, that information appears to influence their projections. Specifically, individuals tend to predict faster (slower) sales growth when they know that a firm’s past growth rate has been high (low).

A key component of the capital asset pricing model is the expected return on the market. To use the CAPM, an investor must form an expectation for the market’s future return. How do investors estimate future returns? It appears in part that they anchor on the market’s recent past returns. More specifically, surveys of investors reveal that when the previous year’s stock market return was high, investor’s expect a higher return in the subsequent year compared to cases in which the previous market return was low. In fact, high past returns are generally not a reliable signal for high future returns, so when investors based their forecast on recent past returns (i.e., when they anchor on last year’s market return), they were overestimating the market’s return, and that in turn would lead them to overestimate returns on specific stocks via the CAPM.

Familiarity Bias

In this text we have discussed a number of analytical methods that investors can use to decide whether they want to purchase a particular investment. It turns out that in many cases people simply invest in things that are familiar to them, a behavior called  familiarity bias . Research has shown that investors tend to invest in stocks located close to their homes. Even professional investors are not immune to this bias. A recent study found that mutual fund managers tend to invest more heavily in stocks located in their home states.

Investing something familiar is not necessarily a bad thing. Perhaps being more familiar with a company helps investors determine whether that company’s stock is a good buy. However, if familiarity helps give investors an information edge, then investors should earn higher returns on the investments that they make based on familiarity (e.g., investments in companies located nearby). Even among professional investors, the evidence on this question is mixed. One study found that mutual fund managers earned unusually high returns on their investments in nearby firms, but other studies found that investing in companies based on familiarity influenced fund managers to form portfolios that were not fully diversified. As a result, those funds did not earn higher returns, but they did experience higher risk.

Investing heavily in familiar stocks does have one serious potential drawback. Industries are often concentrated in specific geographic areas. Think of the concentration of high-tech firms in Silicon Valley, for example. If investors in northern California invest mostly in companies from that region, they will form portfolios that are heavily weighted in tech firms, neglecting other sectors of the economy. Thus, familiarity bias may lead investors to hold underdiversified portfolios. Investors who do not take full advantage of diversification opportunities bear more risk than they need to without necessarily earning higher returns.

Implications of Behavioral Finance for Security Analysis

Our discussion of the psychological factors that affect financial decisions suggests that behavioral finance can play an important role in investing. Naturally, the debate on the efficiency of markets rages on and will continue to do so for many years. The

Table 9.1 Using Behavioral Finance to Improve Investment Results

Studies have documented a number of behavioral factors that appear to influence investors’ decisions and adversely affect their returns. By following some simple guidelines, you can avoid making mistakes and improve your portfolio’s performance. A little common sense goes a long way in the financial markets!

· Don’t hesitate to sell a losing stock. If you buy a stock at $20 and its price drops to $10, ask yourself whether you would buy that same stock if you came into the market today with $10 in cash. If the answer is yes, then hang onto it. If not, sell the stock and buy something else. · Don’t chase performance. The evidence suggests that past performance is at best a very noisy guide to future performance. For example, the best performing mutual funds in the last year or even the last five years are not especially likely to perform best in subsequent years. Don’t buy last year’s hottest mutual fund based solely on its performance. Always keep your personal investment objectives and constraints in mind. · Be humble and open-minded. Many investment professionals, some of whom are extremely well paid, are frequently wrong in their predictions. Admit your mistakes and don’t be afraid to take corrective action. The fact is, reviewing your mistakes can be a very rewarding exercise—all investors make mistakes, but the smart ones learn from them. Winning in the market is often about not losing, and one way to avoid loss is to learn from your mistakes. · Review the performance of your investments on a periodic basis. Remember the old saying, “Out of sight, out of mind.” Don’t be afraid to face the music and to make changes as your situation changes. Nothing runs on “autopilot” forever—including investment portfolios. · Don’t trade too much. Investment returns are uncertain, but transaction costs are guaranteed. Considerable evidence indicates that investors who trade frequently perform poorly.

contribution of behavioral finance is to identify psychological factors that can lead investors to make systematic mistakes and to determine whether those mistakes may contribute to predictable patterns in stock prices. If that’s the case, the mistakes of some investors may be the profit opportunities for others. See  Table 9.1  for our advice on how to keep your own mistakes to a minimum.

Concepts in Review

Answers available at  http://www.pearsonhighered.com/smart

1. 9.4 How can behavioral finance have any bearing on investor returns? Do supporters of behavioral finance believe in efficient markets? Explain.

2. 9.5 Briefly explain how behavioral finance can affect each of the following:

a. The trading activity of investors

b. The tendency of value stocks to outperform growth stocks

c. The tendency of stock prices to drift up (down) after unusually good (bad) earnings news

Technical Analysis

1. LG 5

2. LG 6

In the first section of this chapter we introduced the idea of market efficiency and suggested that there are many good reasons to believe that stock prices (and prices in other financial markets) are inherently unpredictable. The second section presented the behavioral finance challenge to market efficiency and discussed the evidence that there is at least some predictability in stock returns. In this section we introduce  technical analysis , which is the practice of searching the historical record of stock prices and returns for patterns. If these patterns repeat, investors who know about them and can spot them early may have an opportunity to earn better-than-average returns.

Because it focuses on using past price movements to predict future returns, technical analysis is fundamentally at odds with even the weak form of market efficiency. For this reason, the practice of technical analysis remains controversial. For some investors, it’s another piece of information to use when deciding whether to buy, hold, or sell a stock. For others, it’s the only input they use in their investment decisions. Still others regard technical analysis as a waste of time.

Analyzing market behavior dates back to the 1800s, when there was no such thing as industry or company analysis. Detailed financial information about individual companies simply was not made available to stockholders, let alone the general public. About the only thing investors could study was the market itself. Some investors used detailed charts to monitor what large market operators were doing. These charts were intended to show when major buyers were moving into or out of particular stocks and to provide information useful for profitable buy-and-sell decisions. The charts centered on stock price movements. These movements were said to produce certain “formations,” indicating when the time was right to buy or sell a particular stock. The same principle is still applied today. Technical analysts argue that internal market factors, such as trading volume and price movements, often reveal the market’s future direction long before it is evident in financial statistics.

Measuring the Market

If using technical analysis to assess the overall market is a worthwhile endeavor, then we need some sort of tool or measure to do it. Charts are popular with many investors because they provide a visual summary of the behavior of the market and the price movements of individual stocks. As an alternative or supplement to charting, some investors prefer to study various market statistics. They might look at trends in market indexes or track other aspects of market behavior such as trading volume, short selling, or trading behavior of small investors (e.g., odd-lot transactions).

Technical analysis addresses those factors in the marketplace that can (or may) have an effect on the price movements of stocks in general. The idea is to understand the general condition (or “tone”) of the market and to gain some insights into where the market may be headed over the next few months. Several approaches try to do just that, and we summarize some of the more common approaches below.

The Confidence Index

One measure that attempts to capture the tone of the market is the  confidence index , which deals not with the stock market but with bond returns. Computed and published by Barron’s, the confidence index is a ratio that reflects the spread between the average yield on high-grade corporate bonds relative to the yield on average- or intermediate-grade corporate bonds. Technically, the index is computed by relating the average yield on 10 high-grade corporate bonds to the yield on 10 intermediate-grade bonds. The formula is as follows:

Confidence index=Average yield on 10 high-grade corporate bondsAverage yield on 10 intermediate-grade bondsConfidence index = Average yield on 10 high-grade corporate bondsAverage yield on 10 intermediate-grade bondsEquation9.3

Thus, the index measures the yield spread between high-grade bonds and intermediate-grade bonds. Because the yield on high-grade bonds should always be lower than the average yield on a sample of intermediate-grade bonds, the confidence index should never exceed 1.0. Indeed, as the measure approaches 1.0 (or 100%), the spread between the two sets of bonds will get smaller and smaller, which, according to the theory, is a positive sign. The idea is that as investors become more confident about the economy, they will be willing to invest in riskier bonds, driving down their yields and pushing up the confidence index. Those who follow the confidence index interpret a rise in the index as a positive sign for future stock returns.

Consider, for example, a point in time where high-grade bonds are yielding 4.50%, while intermediate-grade bonds, on average, are yielding 5.15%. This would amount to a yield spread of 65 “basis points,” or 65/100 of 1% (i.e., 5.15% − 4.50% = 0.65%), and a confidence index of 4.50 ÷ 5.15 = 87.38%. Now, look what happens when yields (and yield spreads) fall or rise:

	Yields (Yield Spreads)
	Fall	Rise
Yields on high-grade bonds	4.25%	5.25%
Yields on average bonds	4.50%	6.35%
Yield spread	0.25%	1.10%
Confidence index	94.44%	82.68%

Lower-yield spreads, in effect, lead to higher confidence indexes. These, in turn, indicate that investors are demanding a lower premium in yield for the lower-rated (riskier) bonds and in so doing are showing more confidence in the economy. This theory implies that the trend of “smart money” is usually revealed in the bond market before it shows up in the stock market, meaning that a rise in the confidence index today foreshadows a rise in the stock market.

Market Volume

Market volume is an obvious reflection of the amount of investor interest in stocks. As a rule, technical analysts who follow market volume say that increasing volume during a rising market is a positive sign that the upward movement in stocks will continue. On the other hand, when stocks are falling, a decline in volume may suggest that the decline in stock prices is approaching an end. In a similar vein, when stocks have been moving up and volume begins to drop off, that may signal the end of the bull market. Numerous financial periodicals and websites report total market volume daily, so it is an easy statistic to track.

Breadth of the Market

Each trading day, some stocks go up in price and others go down. In market terminology, some stocks advance and others decline. Breadth of the market deals with these advances and declines. The principle behind this indicator is that the number of advances and declines reflects the underlying sentiment of investors.

Watch Your Behavior

Plane Crashes and Sentiment Investor sentiment is a tricky thing to define, and it’s even harder to quantify. One study looked at how major airline disasters affected investor sentiment and stock returns. The author of the study found that the average one-day return on the U.S. stock market is about 4 basis points (0.04%), but the average return on a day with a major airline disaster was negative 32 basis points (-0.32%). That one-day dip represented an aggregate market value loss of $60 billion per airline disaster, but over the next two weeks as sentiment returned to normal, the market recovered most of its losses.

(Source: Guy Kaplanski, “Sentiment and Stock Prices: The Case of Aviation Disasters,” Journal of Financial Economics, 2010, Vol. 95, pp. 174–201.)

Analysts who use market breadth to help guide their investment decisions interpret the numbers as follows. As long as the number of stocks that advance in price on a given day exceeds the number that decline, the market is strong. The extent of that strength depends on the spread between the number of advances and declines. For example, if the spread narrows (the number of declines starts to approach the number of advances), market strength deteriorates. Similarly, the market is weak when the number of declines repeatedly exceeds the number of advances. When the mood is optimistic, advances outnumber declines. Again, data on advances and declines

Figure 9.5 Basic Market Statistics

Here is an example of the kind of information on market volume, advances, and declines that is easily accessible on the web.

(Source:  http://finance.yahoo.com/advances , accessed August 12, 2015.)

are widely available.  Figure 9.5  illustrates data on market volume, advances, and declines taken from Yahoo! Finance.

Short Interest

When investors anticipate a market decline, they sometimes sell a stock short. That is, they sell borrowed stock. The number of shares of stocks sold short in the market at any point in time is known as the  short interest . The more stocks that are sold short, the higher the short interest. Because all short sales must eventually be “covered” (the borrowed shares must be returned), a short sale in effect ensures future demand for the stock. Thus, the market is viewed optimistically when the level of short interest becomes relatively high by historical standards. The logic is that as shares are bought back to cover outstanding short sales, the additional demand will push stock prices up. The amount of short interest on the NYSE, the Amex, and Nasdaq’s National Market is published in the Wall Street Journal, Barron’s, and other sources.

Keeping track of the level of short interest can indicate future market demand, but it can also reveal present market optimism or pessimism. Knowledgeable investors usually do short selling, and a significant buildup or decline in the level of short interest hints at the sentiment of sophisticated investors about the current state of the market or a company. For example, a significant shift upward in short interest might indicate pessimism concerning the current state of the market, even though it may signal optimism with regard to future levels of demand.

An Advisor’s Perspective

Ryan McKeown Senior VP–Financial Advisor, Wealth Enhancement Group

“When the economy is going great, we get a little greedy.”

MyFinanceLab

Odd-Lot Trading

A rather cynical saying on Wall Street suggests that the best thing to do is just the opposite of whatever the small investor is doing. The reasoning behind this is that as a group, small investors exhibit notoriously bad timing. The investing public usually does not come into the market in force until after a bull market has pretty much run its course, and it does not get out until late in a bear market. Although its validity is debatable, this is the premise behind a widely followed technical indicator and is the basis for the  theory of contrary opinion . This theory uses the amount and type of odd-lot trading as an indicator of the current state of the market and pending changes.

Because many individual investors deal in transactions of fewer than 100 shares, their combined sentiments are supposedly captured in odd-lot figures. The idea is to see what odd-lot investors “on balance” are doing. So long as there is little or no difference in the spread between the volume of odd-lot purchases and sales, the theory of contrary opinion holds that the market will probably continue along its current line (either up or down). A dramatic change in the balance of odd-lot purchases and sales may be a signal that a bull or bear market is about to end. For example, if the amount of odd-lot purchases starts to exceed odd-lot sales by an ever-widening margin, speculation on the part of small investors may be starting to get out of control—an ominous signal that the final stages of a bull market may be at hand.

Two trends have diminished the usefulness of odd-lot trading as a market indicator. First, transactions costs have fallen dramatically in recent decades, so the cost advantage of trading in round lots rather than odd lots has diminished. Second, it has become more common for larger traders to break their orders into smaller parts to disguise their activities. For both of these reasons, it is less clear today than it used to be that an individual investor is behind an odd-lot trade. If the purpose of watching odd-lot trades is to assess the trading behavior of individuals rather than professionals, that purpose is harder to achieve today than it once was.

Trading Rules and Measures

Market technicians —analysts who believe it is chiefly (or solely) supply and demand that drive stock prices—use a variety of mathematical equations and measures to assess the underlying condition of the market. These analysts often use computers to produce the measures, plotting them on a daily basis. They then use those measures as indicators of when to get into or out of the market or a particular stock. In essence, they develop trading rules based on these market measures. Technical analysts almost always use several of these market measures, rather than just one (or two), because one measure rarely works the same way for all stocks. Moreover, they generally look for confirmation of one measure by another. In other words, market analysts like to see three or four of these ratios and measures all pointing in the same direction.

Although dozens of these market measures and trading rules exist, we’ll confine our discussion here to some of the more widely used technical indicators: (1) advance-decline lines, (2) new highs and lows, (3) the Arms index, (4) the mutual fund cash ratio, (5) on-balance volume, and (6) the relative strength index (RSI).

Advance-Decline Line

Each trading day, the NYSE, Amex, and Nasdaq publish statistics on how many of their stocks closed higher on the day (i.e., advanced in price) and how many closed lower (declined in price). The advance-decline (A/D) line is simply the difference between these two numbers. To calculate it, you take the number of stocks that have risen in price and subtract the number that have declined, usually for the previous day. For example, if 1,000 issues advanced on a day when 450 issues declined, the day’s net number would be 550 (i.e., 1,000 - 450). If 450 advanced and 1,000 declined, the net number would be -550. Each day’s net number is then added to (or subtracted from) the running total, and the results are plotted on a graph.

If the graph is rising, the advancing issues are dominating the declining issues, and the technical analysts conclude that the market is strong. When declining issues start to dominate, the graph will turn down as the market begins to soften. Technicians use the A/D line as a signal for when to buy or sell stocks.

New Highs–New Lows

This measure is similar to the advance-decline line but looks at price movements over a longer period of time. A stock is defined as reaching a “new high” if its current price is at the highest level it has been over the past year (sometimes referred to as the “52-week high”). Conversely, a stock reaches a “new low” if its current price is at the lowest level it has been over the past year.

The new highs–new lows (NH-NL) indicator equals the number of stocks reaching new 52-week highs minus the number reaching new lows. Thus, you end up with a net number, which can be either positive (when new highs dominate) or negative (when new lows exceed new highs), just like with the advance-decline line. To smooth out the daily fluctuations, the net number is often added to (or subtracted from) a 10-day moving average and then plotted on a graph.

As you might have guessed, a graph that’s increasing over time indicates a strong market, where new highs are dominating. A declining graph indicates a weak market, where new lows are more common than new highs. Technicians following a momentum-based strategy will buy stocks when new highs dominate and sell them when there are more new lows than new highs. Alternatively, they might use the indicator to rotate money into stocks when the market looks strong and to rotate money out of stocks and into cash or bonds when the market looks weak.

The Arms Index

This indicator, also known as the TRIN, for trading index, builds on the advance-decline line by considering the volume in advancing and declining stocks in addition to the number of stocks rising or falling in price. The formula is

TRIN=Number of up stocksNumber of down stocks÷Volume in up stocksVolume in down stocksTRIN=Number of up stocksNumber of down stocks ÷Volume in up stocksVolume in down stocksEquation9.4

For example, suppose we are analyzing the S&P 500. Assume on a given day 300 of these stocks rose in price and 200 fell in price. Also assume that the total trading volume in the rising (“up”) stocks was 400 million shares, and the total trading volume in the falling (“down”) stocks was 800 million shares. The value of the TRIN for the day would be

TRIN=300200÷400 million800 million=3.0TRIN=300200÷400 million800 million=3.0

Alternatively, suppose the volume in up stocks was 700 million shares, and the volume in down stocks was 300 million. The value of the TRIN then would be

TRIN=300200÷700 million300 million=0.64TRIN=300200÷700 million300 million=0.64

Higher TRIN values are interpreted as being bad for the market because even though more stocks rose than fell, the trading volume in the falling stocks was much greater. The underlying idea is that a strong market is characterized by more stocks rising in price than falling, along with greater volume in the rising stocks than in the falling ones, as in the second example.

Mutual Fund Cash Ratio

This indicator looks at the cash position of mutual funds as an indicator of future market performance. The mutual fund cash ratio (MFCR) measures the percentage of mutual fund assets that are held in cash. It is computed as follows:

MFCR=Mutual fund cash position÷Total assets under managementMFCR=Mutual fund cash position ÷  Total assets under managementEquation9.5

The assumption is that the higher the MFCR, the stronger the market. Indeed, the ratio is considered very bullish when it moves to abnormally high levels (i.e., when mutual fund cash exceeds 10% to 12% of assets). It is seen as bearish when the ratio drops to very low levels (e.g., less than 5% of assets). The logic goes as follows: When fund managers hold a lot of cash (when the MFCR is high), that’s good news for the market because they will eventually have to invest that cash, buying stocks and causing prices to rise. If fund managers hold very little cash, investors might be concerned for two reasons. First, there is less demand for stocks if most of the cash is already invested. Second, if the market takes a downturn, investors might want to withdraw their money. Fund managers will then have to sell some of their stocks to accommodate these redemptions (because they don’t have much accumulated cash), putting additional downward pressure on prices.

On-Balance Volume

Technical analysts usually consider stock prices to be the key measure of market activity. However, they also consider trading volume as a secondary indicator. On-balance volume (OBV) is a momentum indicator that relates volume to price change. It uses trading volume in addition to price and tracks trading volume as a running total. In this way, OBV indicates whether volume is flowing into or out of a security. When the security closes higher than its previous close, all the day’s volume is considered “up-volume,” all of which is added to the running total. In contrast, when a stock closes lower, all the day’s volume is considered “down-volume,” which is then subtracted from the running total.

The OBV indicator is used to confirm price trends. According to this measure, you want to see a lot of volume when a stock’s price is rising because that would suggest that the stock will go even higher. On the other hand, if prices are rising but OBV is falling, technical analysts would describe the situation as a divergence and interpret it as a sign of possible weakness.

When analyzing OBV, it is the direction or trend that is important, not the actual value. To begin the computation of OBV, you can start with an arbitrary number, such as 50,000. Suppose you are calculating the OBV for a stock that closed yesterday at a price of $50 per share, and you start with an OBV value of 50,000. Assume that the stock trades 80,000 shares today and closes at $49. Because the stock declined in price, we would subtract the full 80,000 shares from the previous balance (our starting point of 50,000); now the OBV is 50,000 - 80,000 = -30,000 (Note that the OBV is simply the trading volume running total.) If the stock trades 120,000 shares on the following day and closes up at $52 per share, we would then add all of those 120,000 shares to the previous day’s OBV: -30,000 + 120,000 = +90,000. This process would continue day after day. The normal procedure is to plot these daily OBVs on a graph. As long as the graph is moving up, it’s bullish; when the graph starts moving down, it’s bearish.

Relative Strength

One of the most widely used technical indicators is the relative strength index (RSI), an index measuring a security’s strength of advances and declines over time. The RSI indicates a security’s momentum and gives the best results when used for short trading periods. It also helps identify market extremes, signaling that a security is approaching its price top or bottom and may soon reverse trend. The RSI is the ratio of average price change on “up days” to the average price change on “down days” during the same period. The index formula is

RSI=100−[100÷(1+Average price change on up daysAverage price change on down days)]RSI=100−[100÷ (1+Average price change on up daysAverage price change on down days)]Equation9.6

The average price change in this formula is usually calculated over a 9-, 14-, or 25-day period. In the RSI calculation, both price increases and price decreases are treated as positive values. In other words, if a stock fell by $0.05 for 14 days in a row, then the average price change on down days would be 0.05, and the same would hold if a stock rose by $0.05 for 14 days in a row.

The RSI ranges between 0 and 100, with most RSIs falling between 30 and 70. Generally, values above 70 or 80 indicate an overbought condition (more and stronger buying than fundamentals would justify). RSI values below 30 indicate a possible oversold condition (more selling than fundamentals may indicate). When the RSI crosses these points, it signals a possible trend reversal. The wider 80–20 range is often used with the 9-day RSI, which tends to be more volatile than longer-period RSIs. In bull markets, 80 may be a better upper indicator than 70; in bear markets, 20 is a more accurate lower level. Different sectors and industries may have varying RSI threshold levels.

To use the RSI in their own trading, investors set buy and sell ranges—such as sell when the RSI crosses above 70 and buy when it moves below 30. Another strategy is to compare RSIs with stock charts. Most of the time both move in the same direction, but a divergence between RSI and a price chart can be a strong predictor of a changing trend.

Practice Your Charting Skills

Charting

Charting  is perhaps the best-known activity of the technical analyst. Indeed, technical analysts use various types of charts to plot the behavior of everything from the Dow Jones Industrial Average and share price movements of individual stocks to moving averages (see below) and advance-decline lines. In fact, as noted above, just about every type of technical indicator is charted in one form or another.

Charts are popular because they provide a visual summary of activity over time. Perhaps more important (in the eyes of technicians, at least), they contain valuable information about developing trends and the future behavior of the market or individual stocks. Chartists believe price patterns evolve into chart formations that provide signals about the future course of the market or a stock.

Chart Formations

A chart by itself tells you little more than where the market or a stock has been. But to chartists, those price patterns yield formations that tell them what to expect in the future. Chartists believe that history repeats itself, so they study the historical reactions of stocks (or the market) to various formations, and they devise trading rules based on these observations. It makes no difference to chartists whether they are following the market or an individual stock. It is the formation that matters, not the issue being plotted. Chartists believe that they can see formations building and recognize buy and sell signals. These chart formations are often given exotic names, such as head and shoulders, falling wedge, scallop and saucer, ascending triangle, and island reversal, to name just a few.

Figure 9.6  shows six of these formations. The patterns form “support levels” and “resistance lines” that when combined with the basic formations, yield buy and sell signals. Panel A is an example of a buy signal that occurs when prices break out above a resistance line in a particular pattern. In contrast, when prices break out below a support level, as they do at the end of the formation in panel B, a sell signal is said to occur. Supposedly, a sell signal means everything is in place for a major drop in the market (or in the price of a share of stock). A buy signal indicates that the opposite is about to occur.

Figure 9.6 Some Popular Chart Formations

To chartists, each of these formations has meaning about the future course of events.

Unfortunately, one of the major problems with charting is that the formations rarely appear as neatly and cleanly as those in  Figure 9.6  . Rather, identifying and interpreting them often demands considerable imagination.

Moving Averages?

One problem with daily price charts is that they may contain a lot of short-term price swings that mask the overall trend in prices. As a result, technical analysts often use moving averages not only to eliminate those minor blips but also to highlight underlying trends. A  moving average  is a mathematical procedure that records the average value of a series of prices, or other data, over time. Because they incorporate a stream of these average values, moving averages will smooth out a data series and make it easier to spot trends. The moving average is one of the oldest and most popular technical indicators. It can, in fact, be used not only with share prices but also with market indexes and even other technical measures.

Moving averages are computed over time periods ranging from 10 to 200 days—meaning that from 10 to 200 data points are used in each calculation. For example, a series of 15 data points is used in a 15-day moving average. The length of the time period has a bearing on how the MA will behave. Shorter periods (10 to 30 days) are more sensitive and tend to more closely track actual daily behavior. Longer periods (say, 100 to 200 days) are smoother and only pick up the major trends. Several types of moving averages exist, with the most common (and the one we’ll use here) being the simple average, which gives equal weight to each observation. In contrast, there are other procedures that give more weight to the most recent data points (e.g., the “exponential” and “weighted” averages) or apply more weight to the middle of the time period (e.g., “triangular” averages).

Using closing share prices as the basis of discussion, we can calculate the simple moving average by adding up the closing prices over a given time period (e.g., 10 days) and then dividing this total by the length of the time period. Thus, the simple moving average is nothing more than the arithmetic mean. To illustrate, consider the following stream of closing share prices:

Using a 10-day moving average, we add up the closing prices for days 1 through 10 ($4 + $5 + ...+ $8 + $9 = $58)($4 + $5 + ...+ $8 + $9 = $58) and then divide this total by 10($58 ÷10 = $5.8)10($58 ÷10 = $5.8). Thus, the average closing price for this 10-day period was $5.80. The next day, the process is repeated once again for days 2 through 11; that turns out to be $60÷10 = $6.00$60÷10 = $6.00. This procedure is repeated each day, so that over time we have a series of these individual averages that, when linked together, form a moving-average line. This line is then plotted on a chart, either by itself or along with other market information.

Figure 9.7  shows a 100-day moving average (i.e., the red line) plotted against the daily closing prices for Facebook (i.e., the blue line) starting with its May 2012 IPO and continuing through June 2015. In contrast to the actual closing prices, the moving average provides a much smoother line, without all the short-term fluctuations; it clearly reveals the general trend in prices for this stock.

Technicians often use charts like the one in  Figure 9.7  to help them make buy and sell decisions about a stock. Specifically, if the security’s price starts moving above the moving average, they read that situation as a good time to buy because prices should be drifting up (e.g., see the buy signal). In contrast, a sell signal occurs when the security’s price moves below the moving-average line (e.g., see the sell signal). A problem arises when volatility in the stock price leads to repeated buy and sell signals. For example, for Facebook, the red and blue lines cross 11 times between April 4, 2014 and May 6, 2014, resulting in six sell signals and five buy signals all within a single month. Trading based on the moving-average indicator during that period would result in a lot of transactions costs, but not much profit.

Figure 9.7 Daily Closing Prices and 100-Day Moving-Average Line for Facebook

Moving-average lines are often plotted along with the actual daily closing prices for a stock. They’re also widely used with market indexes, such as the S&P 500, and with a variety of technical indicators, including the advance-decline line.

Concepts in Review

Answers available at  http://www.pearsonhighered.com/smart

1. 9.6 What is the purpose of technical analysis? Explain how and why it is used by technicians; note how it can be helpful in timing investment decisions.

2. 9.7 Can the market really have a measurable effect on the price behavior of individual securities? Explain.

3. 9.8 Describe the confidence index, and note the feature that makes it unique.

4. 9.9 Briefly describe each of the following and explain how it is used in technical analysis:

a. Breadth of the market

b. Short interest

c. Odd-lot trading

5. 9.10 Briefly describe each of the following and note how it is computed and how it is used by technicians:

a. Advance-decline lines

b. Arms index

c. On-balance volume

d. Relative strength index

e. Moving averages

6. 9.11 What is a stock chart? What kind of information can be put on charts, and what is the purpose of charting?