At first glance, technical analysis seems so appealing. You need only obtain a set of charts and follow the signal given by the analysis. You, however, must realize that the efficient market hypothesis suggests that technical analysis will not lead to superior in- vestment results. In addition, technical analysis may require frequent buying and selling,
which generates commissions and short-term capital gains. The returns must obviously more than cover these costs for you to outperform the market on a risk-adjusted basis.
Several methods may be used to test the validity of technical analysis. Consider the following returns for stocks A and B.
Period A B
1 15% 12%
2 10 10
3 6 8
4 24 29
5 28 26
An obvious positive correlation exists between the two series of returns. In this il- lustration, the numerical value of the correlation coefficient is 0.96. Such correlation coefficients appear frequently in this text to illustrate the potential for diversification.
If the returns on various assets lack strong positive correlation, combining the assets in a portfolio reduces risk, because the lack of correlation reduces the variability of the portfolio’s return.
Serial correlation measures the correlation between the data for one of the sets of variables. The individual returns for A appear to be serially correlated since the return in each subsequent period is smaller than the return in the previous period (e.g., 15 per- cent in period 1 and 10 percent in period 2). Serial correlation may be used to test tech- nical analysis, especially the assertion that prices move in trends or have momentum in a particular direction. If a stock follows the particular pattern, the individual observa- tions should be serially correlated.
The Stocks, Bonds, Bills, and Inflation Yearbook, however, reported that serial correlation between stock returns was virtually nil. Such a result is consistent with the weak form of the efficient market hypothesis. That is, technical analysis does not lead to superior investment returns. Other studies also tend to support the lack of serial cor- relation between returns. However, while the returns for large-company stocks exhibit serial correlation, the return on small-company stocks may lack this correlation, which would suggest an investor may be able to earn excess returns through a trading strategy involving small stocks.
The majority of research on the various technical indicators has failed to verify the technical approach to investing. (See Burton G. Malkiel, A Random Walk Down Wall Street, 10th ed. [New York: W.W. Norton, 2011] for a summary of this empirical evidence.) This large body of evidence has resulted in a general rejection of technical analysis by many academically trained teachers of finance. In addition, many investors have also concluded that technical analysis does not lead to superior investment perfor- mance. An investor may do just as well by acquiring a randomly selected portfolio and holding it indefinitely!
One major reason why technical approaches may not lead to superior investment results is the speed with which securities prices change. Information is readily dissemi- nated among the investors, and prices adjust accordingly. Thus, if an investor were to develop an approach that outperformed the market, it would only be a matter of time
before the technique would be learned by others. The method would no longer achieve the initial results as additional investors applied it. A system that works (if one can be found) can succeed only if it is not known by many investors. Thus, it is naive for an in- vestor to believe that he or she can use a known technical approach to beat the market.
A new and unknown system is needed. However, when you realize that many investors are looking for and testing various approaches, it is hard to believe that the individual investor will find a technical approach that can beat the market.
Even though empirical results do not favor the use of technical analysis, some in- vestors and portfolio managers continue to use this type of analysis. This usage has the potential to affect securities prices. For example, breaking a trend line may suggest a buying (or selling) opportunity. Heavy buying (or selling) could occur even though the firm’s fundamentals have not changed. By knowing technical trading rules, an investor may avoid buying when the technicians are buying and perhaps artificially raising the stock’s price.
Even if investors and portfolio managers do not employ technical analysis as the sole criterion for investment decisions, they may apply the analysis to confirm deci- sions based on fundamental analysis. One possible explanation for the continued use is the accuracy of the empirical tests. These tests must specify a confidence level, such as 95 percent. Consider a technical approach that generates a return of 12.2 percent when the average return is 12 percent. Can the investor assert with a 95 percent level of confidence that the 0.2 percent difference is the result of the approach’s ability to outperform or is the difference the result of chance? (An analogy with batting averages may help clarify the point. A player with a batting average of .256 has a .298 season.
Since baseball is a game of streaks, is the higher average the result of improved skills or chance, that is, a lucky streak during the season? The answer is obviously important since management may pay for improved skills but trade the player if the improved average is the result of chance.) Even if the returns had been 15 percent versus 12 per- cent and the probability of the difference being statistically significant were higher, the 3 percent difference could still be the result of chance.
Empirical tests often use 95 percent as the level of confidence, with 90 percent being the lowest acceptable level. If it cannot be shown with at least a 90 percent level of confidence that the results are attributable to the technical indicator, the empirical test concludes that the difference is the result of chance. Supporters of technical analysis may argue that 95 percent or even the less rigorous 90 percent is too high a level of confidence. If a technique works only 70 percent of the time, it still generates a higher return. If this return is 0.2 percent greater than the average return, then over a period of years the difference will generate a higher terminal value (i.e., in 20 years, $100,000 grows to $333,035 at 6.2 percent but only $320,714 at 6.0 percent). Even if the addi- tional return is the result of chance, it is doubtful the investor would say, “I don’t want the additional $12,321. It was not earned but was the result of luck!”
The debate concerning the efficacy of technical analysis will continue, and the In- ternet will increase access to technical analysis by the individual investor. Data are readily available that permit you to track stocks and apply technical analysis. Even if you do not use the analysis, its jargon permeates the popular, if not the academic, press on investments. Thus, you need to be aware of technical analysis even if you never use it as part of an investment strategy.
THE DOGS OF THE DOw
One investment strategy that has come into prominence is the Dogs of the Dow.2 (Weak stocks or low-priced stocks are sometimes referred to as “dogs.”) This simple strategy is neither a technical approach nor a fundamental approach to the selection of securities. Since it requires no analysis of past stock prices, volume of trading, or any other method of technical analysis, it is not readily classifiable as a technical approach. The Dogs of the Dow, however, also avoids the fundamental analysis of financial statements, the valuation of cash flows, and the estimation of future growth rates. Since the Dow dog strategy is mechanical, it is more comparable to technical approaches than to valuation methods for selecting stocks and is included in this chapter.
The Dogs of the Dow strategy requires the investor to rank all 30 stocks in the Dow Jones Industrial Average from highest to lowest based on their dividend yields (dividend divided by the price of the stock). The investor then buys an equal dollar amount of the ten stocks with the highest dividend yields. (An alternative strategy is to buy the five lowest-priced “small dogs” of the ten highest-yielding dividend stocks.) After one year, the process is repeated. The Dow stocks are once again ranked, and, if a stock contin- ues to be among the ten highest dividend yields, it is retained. If the stock is no longer among the ten, it is sold and replaced by a new Dow dog that is one of the ten stocks with the highest dividend yields.
This strategy has obvious appeal. First, since it is rebalanced only once a year, commission costs are modest. Second, by waiting one additional day so the portfolio adjustments occur after a year, all capital gains are long-term. (The dividend payments are also taxed.) Third, by buying the Dow stocks with the highest dividend yields, this yield may offer some downside protection from further price declines. Fourth, buying the Dow dogs is acquiring the stocks in the Dow that are currently out of favor and is consistent with a contrarian strategy.
Does the system work? There is evidence that the Dow dividend strategy produces higher returns than the Dow itself.3 The evidence, however, also shows that the stan- dard deviations of the returns on the Dow dogs exceeded the standard deviations of the returns on the Dow Jones Industrial Average and the S&P 500 stock index. (A Dow dog portfolio is less diversified, so the expectation would be for greater variability in the returns.) This result is, of course, consistent with efficient markets: More risk-taking generates higher returns. The empirical results also suggest that over long periods, such as a decade, a strategy of buying and holding all the Dow stocks was a better alternative after considering risk, taxes, and transaction costs.
2The Dow dividend strategy was popularized in Michael O’Higgins, Beating the Dow (New York: Harper Perennial, 1992).
Information concerning the Dow dogs, such as which stocks would currently compose a Dow dog portfolio, may be found at www.dogsofthedow.com.
3Evidence that the strategy generates larger returns but the returns are more variable may be found in George Wunder and Herbert Mayo, “Study Supports Efficient Market Hypothesis,” Journal of Financial Planning (July 1995): 128–135; and Grant McQueen, Kay Shields, and Steven R. Thorley, “Does the Dow-10 Investment Strategy Beat the Dow Statistically and Economically?” Financial Analysts Journal (July–August 1997): 66–72.
SUMMARY
Behavioral finance combines psychology and finance and identifies human traits that affect investment decisions. These emotions include being overconfident, feeling regret when investment decisions generate losses, and perceiving gains as the “house’s” money.
Individuals tend to acquire assets with which they are familiar; they isolate (mentally budget) individual investment decisions and selectively remember investment results.
Investors also follow a herd mentality. These are some of the personal traits that often lead to poor investment decisions.
Technical analysis seeks to identify potential investments by examining the past performance of the market or individual securities. Technical analysts or “chartists”
stress the past as a means to predict the future. Technical analysis removes emotion and is diametrically opposed to the fundamental analysis that stresses future earnings and dividends (i.e., cash flows) appropriately discounted back to the present.
Several technical approaches such as the Dow Theory and Barron’s confidence index attempt to identify changes in the direction of the market. Because individual securities prices move together, the determination of a change in the market’s direction should identify future movements in individual stock prices. Other technical indicators such as point-and-figure charts, bar graphs, and moving averages may be applied to the market and to individual securities. By constructing various charts, the technical analyst determines when specific securities should be bought or sold.
Whether technical approaches to market timing and stock selection lead to su- perior returns is an empirical question. With some exceptions, academic research has produced little support for technical analysis. These results suggest that investors may achieve similar or even superior results by purchasing and holding a well-diversified portfolio of securities.
QUESTIONS
1. What are several human traits that tend to affect investment decisions?
2. Why do the supporters of behavioral finance suggest that emotions lead to inferior investment decisions?
3. What is the purpose of technical analysis, and why are those who use technical analysis referred to as chartists?
4. What changes produce a sell signal in the Dow Theory and Barron’s confidence index?
5. What is a moving average? What is the significance when a stock’s price crosses a mov- ing average of the stock’s price?
6. What is the problem with time lags in technical analysis and why may the analysis lead to self-fulfilling predictions?
7. What is the difference between “support” and “resistance” in technical analysis?
8. Why does technical analysis receive little support from academically oriented students of investments?
9. Which Dow Jones Industrial Average stocks would be considered “dogs”? Determine the Dow dogs as of January 1; invest $1,000 in each dog. At the end of a time period such as the semester or year, compare the dogs’ performance with the performance of the Dow. Be certain to remember to include the dividends in your calculation.
10. Locate graphs of moving averages for International Business Machines (IBM) and Cisco (CSCO). Based on the moving averages, should you be long or short in each of these stocks? After answering this question, continue to follow the stocks’ prices for a period of time. Did your position prove to be profitable? Various moving averages are available at many of the Internet sites given throughout this text.