EMH

 The Efficient Market HypothesisONE OF THE early applications of computers in economics in the 1950s was to analyze economic time series. Business cycle theorists felt that tracing the evolution of several economic variables over time would clarify and predict the progress of the economy through boom and bust periods. A natural candidate for analysis was the behavior of stock market prices over time. Assuming that stock prices reflect the prospects of the firm, recurrent patterns of peaks and troughs in economic performance ought to show up in those prices.Maurice Kendall examined this proposition in 1953.1 He found to his great surprise that he could identify no predictable patterns in stock prices. Prices seemed to evolve randomly. They were as likely to go up as they were to go down on any particular day, regardless of past performance.At first blush, Kendall’s results were disturbing to some financial economists. They seemed to imply that the stock market is dominated by erratic market psychology, or “animal spirits”—that it follows no logical rules. In short, the results appeared to confirm the irrationality of the market. On further reflection, however, economists came to reverse their interpretation of Kendall’s study.It soon became apparent that random price movements indicated a well-functioning or efficient market, not an irrational one. In this chapter we explore the reasoning behind what may seem a surprising conclusion. We show how competition among analysts leads naturally to market efficiency, and we examine the implications of the efficient market hypothesis for investment policy. We also consider empirical evidence that supports and contradicts the notion of market efficiency.page 332 11.1 Random Walks and Efficient MarketsSuppose Kendall had discovered that stock price changes are predictable. What a gold mine this would have been. Investors who could predict stock prices would reap unending profits simply by purchasing stocks that the computer model implied were about to increase in price and selling those stocks about to fall.A moment’s reflection should be enough to convince yourself that this situation could not persist for long. For example, suppose that the model predicts with great confidence that XYZ stock price, currently at $100 per share, will rise dramatically in three days to $110. What would all investors with access to the model’s prediction do today? Obviously, they would place a great wave of buy orders to cash in on the prospective increase in stock price. No one holding XYZ, however, would be willing to sell. The net effect would be an immediate jump in the stock price to $110 as the market digests and reflects the “good news” implicit in the model’s forecast.This simple example illustrates why Kendall’s attempt to find recurrent patterns in stock price movements was likely to fail. A forecast about favorable future performance leads instead to favorable current performance, as market participants all try to get in on the action before the price jump.More generally, one might say that any information that could be used to predict stock performance should already be reflected in stock prices. As soon as there is any information indicating that a stock is underpriced, investors flock to buy the stock and immediately bid up its price to a fair level, where only ordinary rates of return can be expected. These “ordinary rates” are simply rates of return commensurate with the risk of the stock.However, if prices are bid immediately to fair levels, given all available information, it must be that they increase or decrease only in response to new information. New information, by definition, must be unpredictable; if it could be predicted, then the prediction would be part of today’s information. Thus, stock prices that change in response to new (i.e., previously unpredicted) information also must move unpredictably.This is the essence of the argument that stock prices should follow a random walk, that is, that price changes should be random and unpredictable.2 Far from a proof of market irrationality, randomly evolving stock prices would be the necessary consequence of intelligent investors competing to discover relevant information on which to buy or sell stocks before the rest of the market becomes aware of that information.Don’t confuse randomness in price changes with irrationality in the level of prices. If prices are determined rationally, then only new information will cause them to change. Therefore, a random walk would be the natural result of prices that always reflect all current knowledge. Indeed, if stock price movements were predictable, that would be damning evidence of stock market inefficiency because the ability to predict prices would indicate that all available information was not already reflected in stock prices. Therefore, the notion that stocks already reflect all available information is referred to as the efficient market hypothesis (EMH).3page 333 Figure 11.1 illustrates the response of stock prices to new information in an efficient market. The graph plots the price response of a sample of firms that were targets of takeover attempts. In most takeovers, the acquiring firm pays a substantial premium over current market prices. Therefore, an announcement of a takeover attempt should cause the stock price to jump. The figure shows that stock prices jump dramatically on the day the news becomes public. However, there is no further drift in prices after the announcement date, suggesting that prices reflect the new information, including the likely magnitude of the takeover premium, by the end of the trading day.Figure 11.1 Cumulative abnormal returns before takeover attempts: target companiesSource: This is an update of a figure that appeared in Arthur Keown and John Pinkerton, “Merger Announcements and Insider Trading Activity,” Journal of Finance 36 (September 1981). Updates courtesy of Jinghua Yan.Even more dramatic evidence of rapid response to new information may be found in intraday prices. For example, Patell and Wolfson show that most of the stock price response to corporate dividend or earnings announcements occurs within 10 minutes of the announcement.4 A nice illustration of such rapid adjustment is provided in a study by Busse and Green, who track minute-by-minute stock prices of firms that are featured on CNBC’s “Morning” or “Midday Call” segments.5 Minute 0 in Figure 11.2 is the time at which the stock is mentioned on the midday show. The top line is the average price movement of stocks that receive positive reports, while the bottom line reports returns on stocks with negative reports. Notice that the top line levels off, indicating that the market has fully digested the news within 5 minutes of the report. The bottom line levels off within about 12 minutes.Figure 11.2 Stock price reaction to CNBC reports. The figure shows the reaction of stock prices to on-air stock reports during the “Midday Call” segment on CNBC. The chart plots cumulative returns beginning 15 minutes before the stock report.Source: J. A. Busse and T. C. Green, “Market Efficiency in Real Time,” Journal of Financial Economics 65 (2002), p. 422.Competition as the Source of EfficiencyWhy should we expect stock prices to reflect “all available information”? After all, if you are willing to spend time and money on gathering information, it might seem reasonable that you could turn up something that has been overlooked by the rest of the investment community. When information is costly to uncover and analyze, one would expect investment analysis calling for such expenditures to result in an increased expected return.This point has been stressed by Grossman and Stiglitz.6 They argue that investors will have an incentive to spend time and resources to analyze and uncover new information page 334only if such activity is likely to generate higher investment returns. Thus, in market equilibrium, efficient information-gathering activity should be fruitful. Moreover, it would not be surprising to find that the degree of efficiency differs across various markets. For example, emerging markets that are less intensively analyzed than U.S. markets or in which accounting disclosure requirements are less rigorous may be less efficient than U.S. markets. Small stocks that receive relatively little coverage by Wall Street analysts may be less efficiently priced than large ones. Still, while we would not go so far as to say that you absolutely cannot come up with new insights, it makes sense to consider and respect your competition.Example 11.1 Rewards for Incremental PerformanceConsider an investment management fund currently managing a $5 billion portfolio. Suppose that the fund manager can devise a research program that could increase the portfolio rate of return by one-tenth of 1% per year, a seemingly modest amount. This program would increase the dollar return to the portfolio by $5 billion × .001, or $5 million. Therefore, the fund should be willing to spend up to $5 million per year on research to increase stock returns by a mere tenth of 1% per year. With such large rewards for such small increases in investment performance, it should not be surprising that professional portfolio managers are willing to spend large sums on industry analysts, technical support, and research effort, and therefore that price changes are, generally speaking, difficult to predict.With so many well-backed analysts willing to spend considerable resources on research, easy pickings in the market are rare. Moreover, the incremental rates of return on research activity may be so small that only managers of the largest portfolios will find them worth pursuing.Although it may not literally be true that “all” relevant information will be uncovered, it is likely that there are many investigators hot on the trail of most leads that seem likely to improve investment performance. Competition among these many well-backed, highly paid, aggressive analysts ensures that, as a general rule, stock prices ought to reflect available information regarding their proper levels.Information is often said to be the most precious commodity on Wall Street, and the competition for it is intense. Consider the industry of so-called alternative data firms that has emerged to uncover and sell to large investors information that might shed light on corporate page 335prospects. For example, these firms use satellite imagery to estimate the number of cars parked outside big retailers such as Walmart, thereby getting a sense of daily sales. Other firms use satellite imagery to estimate the height of oil tanks and thus the size of oil stocks. One firm, Thanos, specializes in insights from cell phones. For example, in 2018, it tracked pings from cell phones at a Tesla factory and concluded that the overnight shift had increased by 30%. This was evidence of a soon-to-be-announced increase in production, news of which Thanos sold to its hedge fund clients. When the production increase was formally announced to the public, Tesla shares increased by 9%. Thanos reportedly will use similar techniques to track (and sell to clients) indicators of foot traffic at shopping malls.7WORDS FROM THE STREETMatchmakers for the Information AgeThe most precious commodity on Wall Street is information, and informed players can charge handsomely for providing it. An industry of so-called expert network providers has emerged for selling access to experts with unique insights about a wide variety of firms and industries to investors who need that information to make decisions. These firms have been dubbed matchmakers for the information age.* Experts can range from doctors who help predict the release of blockbuster drugs to meteorologists who forecast weather that can affect commodity prices to business executives who can provide specialized insight about companies and industries.The risk is that these experts may peddle inside information. For example, in 2011, a consultant for Primary Global Research was convicted of selling information to the hedge fund SAC Capital Advisors.Expert firms are supposed to provide only public information, along with the expert’s insights and perspective. But the temptation to hire experts with inside information and charge handsomely for access to them is obvious. The SEC has raised concerns about the boundary between legitimate and illegal services.In the wake of increased scrutiny, compliance efforts of both buyers and sellers of expert information have mushroomed. Expert firms now maintain detailed records of which experts have talked to whom, when those conversations took place, and what was discussed. These records can be released to authorities in the event of an insider trading investigation.Even with these safeguards, however, there remains room for trouble. For example, an investor may meet an expert through a legitimate network and then the two may establish a consulting relationship on their own. This legal matchmaking becomes the precursor to the illegal selling of insider tips. Where there is a will to cheat, there usually will be a way.*See, for example, “Expert Networks Are the Matchmakers for the Information Age,” The Economist, June 16, 2011.Sometimes the quest for a competitive advantage can tip over into a search for illegal inside information. For example, in 2011, Raj Rajaratnam, the head of the Galleon Group hedge fund, which once managed $6.5 billion, was convicted for soliciting tips from a network of corporate insiders and traders. In 2014, another successful hedge fund, SAC Capital Advisors, paid $1.8 billion to settle an insider trading probe. While these firms clearly crossed the line separating legitimate and prohibited means to acquire information, that line is often murky. For example, a large industry of expert network firms connects (for a fee) investors to industry experts who can provide unique perspective on a company. As the nearby box discusses, this sort of arrangement can easily lead to insider trading and, in fact, was a key component of the case against SAC Capital.Versions of the Efficient Market HypothesisIt is common to distinguish among three versions of the EMH: the weak, semistrong, and strong forms of the hypothesis. They differ by their notions of what is meant by the term “all available information.”The weak-form hypothesis asserts that stock prices already reflect all information that can be derived by examining market trading data such as the history of past prices, trading volume, or short interest. This version implies that trend analysis is fruitless. Past stock price page 336data are publicly available and virtually costless to obtain. The weak-form hypothesis holds that if such data ever conveyed reliable signals about future performance, all investors already would have learned to exploit them. Ultimately, the signals lose their value as they become widely known because a buy signal, for instance, would result in an immediate price increase.The semistrong-form hypothesis states that all publicly available information regarding the prospects of a firm must be reflected already in the stock price. Such information includes, in addition to past trading data, fundamental data on the firm’s product line, quality of management, balance sheet composition, patents held, earnings forecasts, and accounting practices. Again, if investors have access to such information from publicly available sources, one would expect it to be reflected in stock prices.Finally, the strong-form version of the efficient market hypothesis states that stock prices reflect all relevant information, even including information available only to company insiders. This version of the hypothesis is quite extreme. Few would argue with the proposition that corporate officers have access to pertinent information long enough before public release to enable them to profit from trading on it. Indeed, much of the activity of the Securities and Exchange Commission is directed toward preventing insiders from profiting by exploiting their privileged position. Rule 10b-5 of the Security Exchange Act of 1934 sets limits on trading by corporate officers, directors, and substantial owners, requiring them to report trades to the SEC. These insiders, their relatives, and any associates who trade on information supplied by insiders are considered in violation of the law.Defining insider trading is not always easy, however. After all, stock analysts are in the business of uncovering information not already widely known to market participants. As we saw in Chapter 3 as well as in the nearby box, the distinction between private and inside information is sometimes murky.Notice one thing that all versions of the EMH have in common: They all assert that prices should reflect available information. We do not expect traders to be superhuman or market prices to always be right. We will always wish for more information about a company’s prospects than will be available. Sometimes market prices will turn out in retrospect to have been outrageously high, at other times absurdly low. The EMH asserts only that at the given time, using current information, we cannot be sure if today’s prices will ultimately prove themselves to have been too high or too low. If markets are rational, however, we can expect them to be correct on average.Concept Check 11.1Suppose you observed that high-level managers make superior returns on investments in their company’s stock. Would this be a violation of weak-form market efficiency? Would it be a violation of strong-form market efficiency?If the weak-form version of the efficient market hypothesis is valid, must the strong-form version also hold? Conversely, does strong-form efficiency imply weak-form efficiency?11.2 Implications of the EMHTechnical AnalysisTechnical analysis is essentially the search for recurrent and predictable patterns in stock prices. Although technicians recognize the value of information regarding future economic prospects of the firm, they believe that such information is not necessary for a successful page 337trading strategy. This is because whatever the fundamental reason for a change in stock price, if the price responds slowly enough, the analyst will be able to identify a trend that can be exploited during the adjustment period. The key to successful technical analysis is a sluggish response of stock prices to fundamental supply-and-demand factors. This, of course, is diametrically opposed to the notion of an efficient market.Technical analysts are sometimes called chartists because they study records or charts of past stock prices, hoping to find patterns they can exploit to make a profit. As an example of technical analysis, consider the relative strength approach. The chartist compares stock performance over a recent period to performance of the market or other stocks in the same industry. A simple version of relative strength takes the ratio of the stock price to a market indicator such as the S&P 500 index. If the ratio increases over time, the stock is said to exhibit relative strength because its price performance has been better than that of the broad market. Such strength presumably may continue for a long enough period of time to offer profit opportunities.One of the most commonly heard components of technical analysis is the notion of resistance levels or support levels. These values are said to be price levels above which it is difficult for stock prices to rise, or below which it is unlikely for them to fall, and they are believed to be levels determined by market psychology.Example 11.2 Resistance LevelsConsider stock XYZ, which traded for several months at a price of $72 and then declined to $65. If the stock price eventually begins to increase, $72 will be considered a resistance level (according to this theory) because investors who bought originally at $72 will be eager to sell their shares as soon as they can break even on their investment. Therefore, at prices near $72 a wave of selling pressure will exist. Such activity imparts a type of “memory” to the market that allows past price history to influence current prospects.The efficient market hypothesis implies that technical analysis should be fruitless. The past history of prices and trading volume is publicly available at minimal cost. Therefore, any information that was ever available from analyzing past trading has already been reflected in stock prices. As investors compete to exploit their common knowledge of a stock’s price history, they necessarily drive stock prices to levels where expected rates of return are exactly commensurate with risk. At those levels one cannot expect abnormal returns.As an example of how this process works, consider what would happen if the market believed that a level of $72 truly was a resistance level in Example 11.2. No one would be willing to purchase the stock at a price of $71.50 because it would have almost no room to increase in price, but ample room to fall. However, if no one would buy it at $71.50, then $71.50 would become a resistance level. But then, using a similar analysis, no one would buy it at $71, or $70, and so on. The notion of a resistance level poses a logical conundrum. Its simple resolution is the recognition that if the stock is ever to sell at $71.50, investors must believe that the price can as easily increase as fall. The fact that investors are willing to purchase (or even hold) the stock at $71.50 is evidence of their belief that they can earn a fair expected rate of return at that price.Concept Check 11.2If everyone in the market believes in resistance levels, why don’t these beliefs become self-fulfilling prophecies?page 338 An interesting question is whether a technical rule that seems to work will continue to work once it becomes widely recognized. A clever analyst may occasionally uncover a profitable trading rule; the real test of efficient markets is whether the rule itself becomes reflected in stock prices once its value is discovered. Once a useful technical rule is discovered, it ought to be invalidated when many traders attempt to exploit it. In this sense, price patterns ought to be self-destructing.Thus, the market dynamic is one of a continual search for profitable trading rules, followed by destruction by overuse of those rules found to be successful, followed by more searching for yet-undiscovered rules. The following example illustrates this process.Example 11.3 Moneyball and the Efficient Market HypothesisIn his famous book Moneyball Michael Lewis explores an inefficient market—not a financial market but a sports market.8 He begins with the observation that baseball players at the turn of the century were systematically “mispriced” by team management and scouts, whose conventional wisdom about how best to evaluate players led them to ignore better gauges of “intrinsic value.” Billy Beane, the manager of the Oakland Athletics, facing an extremely limited payroll compared to better-funded teams in the league, adopted sabermetrics (statistical analysis applied to baseball) as a means to identify the best talent. He was forced to take what were then unconventional tactics to identify the players most undervalued by the league. Sabermetrics was a resounding success: Despite a salary payroll of only $44 million in 2002 (compared to more than $125 million for the N.Y. Yankees!), Oakland made it to the playoffs, on the way setting an American League record with a 20-game win streak.Beane took advantage of what appears to have been a grossly inefficient market, but the success of his innovations, not surprisingly, led to imitation. Part of the inefficiency in the sports market reflected a lack of relevant data about players; once that became clear, considerable resources were employed to gather more information and analyze it better. Other teams hired their own sabermetricians, player assessment across the league improved, and Oakland’s analytical advantage dissipated. This feels a bit unfair to the innovators, but it is the inevitable dynamic of a market groping its way to greater efficiency. Competition among market participants ultimately forced market prices to better reflect players’ intrinsic value. Useful information may occasionally be ignored or poorly used, but once a better means of analysis is devised, it is rarely long before it spreads widely.Fundamental AnalysisFundamental analysis uses earnings and dividend prospects of the firm, expectations of future interest rates, and risk evaluation of the firm to determine proper stock prices. Ultimately, it represents an attempt to determine the present value of all the payments a stockholder will receive from each share of stock. If that “intrinsic value” exceeds the stock price, the fundamental analyst would recommend purchasing the stock.Fundamental analysts usually start with a study of past earnings and an examination of company balance sheets. They supplement this analysis with further detailed economic analysis, ordinarily including an evaluation of the quality of the firm’s management, the firm’s standing within its industry, and the prospects for the industry as a whole. The hope is to attain insight into future performance of the firm that is not yet recognized by the rest of the market. Chapters 17, 18, and 19 provide a detailed discussion of the types of analyses that underlie fundamental analysis.page 339 Once again, the efficient market hypothesis predicts that most fundamental analysis also is doomed to failure. If the analyst relies on publicly available earnings and industry information, his or her evaluation of the firm’s prospects is not likely to be significantly more accurate than those of rival analysts. Many well-informed, well-financed firms conduct such market research, and in the face of such competition it will be difficult to uncover data not also available to other analysts. Only analysts with a unique insight will be rewarded.Fundamental analysis is more difficult than merely identifying well-run firms with good prospects. Discovery of good firms does an investor no good in and of itself if the rest of the market also knows those firms are good. If the knowledge is already public, the investor will be forced to pay a high price for those firms and will not realize a superior rate of return.The trick is not to identify firms that are good, but to find firms that are better than everyone else’s estimate. Similarly, troubled firms can be great bargains if their prospects are not quite as bad as their stock prices suggest.This is why fundamental analysis is difficult. It is not enough to do a good analysis of a firm; you can make money only if your analysis is better than that of your competitors because the market price will already reflect all commonly recognized information.Active versus Passive Portfolio ManagementBy now it is apparent that casual efforts to pick stocks are not likely to pay off. Competition among investors ensures that any easily implemented stock evaluation technique will be used widely enough so that any insights derived will be reflected in stock prices. Only serious analysis and uncommon techniques are likely to generate the differential insight necessary to yield trading profits.Moreover, these techniques are economically feasible only for managers of large portfolios. If you have only $100,000 to invest, even a 1% per year improvement in performance generates only $1,000 per year, hardly enough to justify herculean efforts. The billion-dollar manager, however, reaps extra income of $10 million annually from the same 1% increment.If small investors are at a disadvantage in active portfolio management, what are their choices? They probably are better off investing in mutual funds or exchange-traded funds. By pooling resources in this way, they can gain from economies of scale.More difficult decisions remain, though. Can investors be sure that even large funds have the ability or resources to uncover mispriced stocks? Furthermore, will any mispricing be sufficiently large to repay the costs entailed in active portfolio management?Proponents of the efficient market hypothesis believe that active management is largely wasted effort and unlikely to justify the expenses incurred. Therefore, they advocate a passive investment strategy that makes no attempt to outsmart the market. A passive strategy aims only at establishing a well-diversified portfolio of securities without attempting to find under- or overvalued stocks. Passive management is usually characterized by a buy-and-hold strategy. When stock prices are at fair levels, it makes no sense to buy and sell frequently, which generates large trading costs without increasing expected performance.One common strategy for passive management is to create an index fund, which is a portfolio designed to replicate the performance of a broad-based index of stocks. For example, Vanguard’s 500 Index Fund holds stocks in direct proportion to their weight in the Standard & Poor’s 500 stock price index. Investors in this fund obtain broad diversification with low management fees. The fees can be kept to a minimum because Vanguard does not need to pay analysts to assess stock prospects and does not incur transaction costs from high portfolio turnover. Indeed, while the typical annual charge for an actively managed equity fund is almost 1% of assets, the expense ratio of the 500 Index Fund is only .04%. page 340Vanguard’s 500 Index Fund is among the largest equity mutual funds, with over $430 billion of assets in late 2018, and between 20% and 25% of assets in equity funds are indexed.Indexing need not be limited to the S&P 500, however. For example, some of the funds offered by the Vanguard Group track the broader-based CRSP9 index of the total U.S. equity market, the Barclays U.S. Aggregate Bond Index, the CRSP index of small-capitalization U.S. companies, and the Financial Times indexes of the European and Pacific Basin equity markets. Several other mutual fund complexes offer indexed portfolios, but Vanguard dominates the retail market for indexed mutual funds.Exchange-traded funds, or ETFs, are a close (and often lower-expense) alternative to indexed mutual funds. As described in Chapter 4, these are shares in diversified portfolios that can be bought or sold just like shares of individual stock. ETFs matching several broad stock market indexes such as the S&P 500 or CRSP indexes and dozens of international and industry stock indexes are available to investors who want to hold a diversified sector of a market without attempting active security selection.Concept Check 11.3What would happen to market efficiency if all investors attempted to follow a passive strategy?The Role of Portfolio Management in an Efficient MarketIf the market is efficient, why not pick stocks by throwing darts at The Wall Street Journal instead of trying rationally to choose a stock portfolio? This is a tempting conclusion to draw from the notion that security prices are fairly set, but it is far too facile. There is a role for rational portfolio management, even in perfectly efficient markets.You have learned that a basic principle in portfolio selection is diversification. Even if all stocks are priced fairly, each still poses firm-specific risk that can be eliminated through diversification. Therefore, rational security selection, even in an efficient market, calls for the construction of an efficiently diversified portfolio providing the systematic risk level that the investor wants.Rational investment policy also requires that tax considerations be reflected in security choice. High-tax-bracket investors generally will not want the same securities as low-bracket ones. At an obvious level, high-bracket investors find it advantageous to buy tax-exempt municipal bonds despite their relatively low pretax yields, whereas those same bonds are unattractive to low-tax-bracket or tax-exempt investors. At a more subtle level, high-bracket investors might want to tilt their portfolios in the direction of capital gains as opposed to interest income because capital gains are taxed less heavily and because the option to defer the realization of capital gains income is more valuable the higher the current tax bracket. They also will be more attracted to investment opportunities for which returns are sensitive to tax benefits, such as real estate ventures.A third argument for rational portfolio management relates to the particular risk profile of the investor. For example, a Toyota executive whose annual bonus depends on Toyota’s profits generally should not invest additional amounts in auto stocks. To the extent that his or her compensation already depends on Toyota’s well-being, the executive is already overinvested in Toyota and should not exacerbate the lack of diversification. This lesson was learned with considerable pain in September 2008 by Lehman Brothers employees page 341who were famously invested in their own firm when the company failed. Roughly 30% of the shares in the firm were owned by its 24,000 employees, and their losses on those shares totaled around $10 billion.Investors of varying ages also might prefer different portfolio policies with regard to risk bearing. For example, older investors who are essentially living off savings might choose to avoid large equity investments, where market values fluctuate dramatically. Because these investors are living off accumulated savings, they require conservation of principal. In contrast, younger investors might be more inclined toward long-term inflation-indexed bonds. The steady flow of real income over long periods of time that is locked in with these bonds can be more important than preservation of principal to those with long life expectancies.In conclusion, there is a role for portfolio management even in an efficient market. Investors’ optimal positions will vary according to factors such as age, tax bracket, risk aversion, and employment. The role of the portfolio manager in an efficient market is to tailor the portfolio to these needs, rather than to beat the market.Resource AllocationWe’ve focused so far on the investment implications of the efficient market hypothesis. Deviations from efficiency may offer profit opportunities to better-informed traders at the expense of less-informed ones.However, deviations from informational efficiency would also result in a large cost that would be borne by all citizens, namely, inefficient resource allocation. Recall that in a capitalist economy, investments in real assets such as plant, equipment, and know-how are guided in large part by the prices of financial assets. For example, if the value of telecommunication capacity reflected in stock market prices exceeds the cost of installing such capacity, managers might justifiably conclude that telecom investments seem to have positive net present value. In this manner, capital market prices guide allocation of real resources.If markets were inefficient and securities commonly mispriced, then resources would be systematically misallocated. Corporations with overpriced securities would be able to obtain capital too cheaply, and corporations with undervalued securities might forgo investment opportunities because the cost of raising capital would be too high. Therefore, inefficient capital markets would diminish one of the most potent benefits of a market economy. As an example of what can go wrong, consider the dot-com bubble of the late 1990s, which sent a strong but, as it turned out, wildly overoptimistic signal about the immediate prospects for Internet and telecommunication firms and ultimately led to substantial overinvestment in those industries.Before writing off markets as a means to guide resource allocation, however, one has to be reasonable about what can be expected from market forecasts. In particular, you shouldn’t confuse an efficient market, where all available information is reflected in prices, with a perfect-foresight market. As we said earlier, “all available information” is still far from complete information, and generally rational market forecasts will sometimes be wrong; sometimes, in fact, they will be very wrong.11.3 Event StudiesThe notion of informationally efficient markets leads to a powerful research methodology. If security prices reflect all currently available information, then price changes must reflect new information. Therefore, it seems that one should be able to measure the importance of an event of interest by examining price changes during the period in which the event occurs.page 342 An event study describes a technique of empirical financial research that enables an observer to assess the impact of a particular event on a firm’s stock price. A stock market analyst might want to study the impact of dividend changes on stock prices, for example. An event study would quantify the relationship between dividend changes and stock returns.Analyzing the impact of any particular event is more difficult than it might at first appear. On any day, stock prices respond to a wide range of economic news such as updated forecasts for GDP, inflation rates, interest rates, or corporate profitability. Isolating the part of a stock price movement that is attributable to a specific event is not a trivial exercise.The general approach starts with a proxy for what the stock’s return would have been in the absence of the event. The abnormal return due to the event is estimated as the difference between the stock’s actual return and this benchmark. Several methodologies for estimating the benchmark return are used in practice. For example, a very simple approach measures the stock’s abnormal return as its return minus that of a broad market index. An obvious refinement is to compare the stock’s return to those of other stocks matched according to criteria such as firm size, beta, recent performance, or ratio of price to book value per share. Another approach estimates normal returns using an asset pricing model such as the CAPM or one of its multifactor generalizations such as the Fama-French three-factor model.To illustrate, we use a “market model” to estimate abnormal returns. This approach is based on the index models we introduced in Chapter 9. Recall that a single-index model holds that stock returns are determined by a market factor and a firm-specific factor. The stock return, rt, during a given period t, would be expressed mathematically as(11.1)where rMt is the market’s rate of return during the period and et is the part of a security’s return resulting from firm-specific events. The parameter b measures sensitivity to the market return, and a is the average rate of return the stock would realize in a period with a zero market return.10 Equation 11.1 therefore provides a decomposition of rt into market and firm-specific factors. The firm-specific or abnormal return may be interpreted as the unexpected return that results from the event.Determination of the abnormal return in a given period requires an estimate of et. Therefore, we rewrite Equation 11.1:(11.2)Equation 11.2 has a simple interpretation: The residual, et, that is, the component presumably due to the event in question, is the stock’s return over and above what one would predict based on broad market movements in that period, given the stock’s sensitivity to the market.The market model is a highly flexible tool because it can be generalized to include richer models of benchmark returns, for example, by including industry as well as broad market returns on the right-hand side of Equation 11.1, or returns on indexes constructed to match characteristics such as firm size. However, one must be careful that regression parameters in Equation 11.1 (the intercept a and slope b) are estimated properly. In particular, they page 343must be estimated using data sufficiently separated in time from the event in question that they are not affected by event-period abnormal stock performance. In part because of this vulnerability of the market model, returns on characteristic-matched portfolios have become more widely used benchmarks in recent years.Example 11.4 Abnormal ReturnsSuppose that the analyst has estimated that a = .05% and b = .8. On a day that the market goes up by 1%, you would predict from Equation 11.1 that the stock should rise by an expected value of .05% + .8 × 1% = .85%. If the stock actually rises by 2%, the analyst would infer that firm-specific news that day caused an additional stock return of 2% − .85% = 1.15%. This is the abnormal return for the day.We measure the impact of an event by estimating the abnormal return on a stock (or group of stocks) at the moment the information about the event becomes known to the market. For example, in a study of the impact of merger attempts on the stock prices of target firms, the announcement date is the date on which the public is informed that a merger is to be attempted. The abnormal returns of each firm surrounding the announcement date are computed, and the statistical significance and magnitude of the typical abnormal return indicate the impact of the newly released information.One concern that complicates event studies arises from leakage of information. Leakage occurs when information regarding a relevant event is released to a small group of investors before official public release. In this case the stock price might start to increase (in the case of a “good news” announcement) days or weeks before the official announcement date. Any abnormal return on the announcement date is then a poor indicator of the total impact of the information release. A better indicator would be the cumulative abnormal return (CAR), which is simply the sum of all abnormal returns over the time period of interest. The cumulative abnormal return thus captures the total firm-specific stock movement for the entire period that the market might be responding to new information.Figure 11.1 (earlier in the chapter) presents the results from a fairly typical event study. The authors of this study were interested in leakage of information before merger announcements and constructed a sample of firms that were targets of takeover attempts. In most takeovers, stockholders of the acquired firms sell their shares to the acquirer at substantial premiums over market value. Announcement of a takeover attempt is good news for shareholders of the target firm and therefore should cause stock prices to jump.Figure 11.1 confirms the good-news nature of the announcements. On the announcement day, called day 0, the average cumulative abnormal return (CAR) for the sample of takeover candidates increases substantially, indicating a large and positive abnormal return on the announcement date. Notice that immediately after the announcement date the CAR no longer increases or decreases significantly. This is exactly what the efficient market hypothesis would predict. Once the new information became public, the stock prices jumped almost immediately in response to the good news. With prices once again fairly set, reflecting the effect of the new information, further abnormal returns on any particular day are equally likely to be positive or negative. In fact, for a sample of many firms, the average abnormal return should be extremely close to zero, and thus the CAR will show neither upward nor downward drift. This is precisely the pattern shown in Figure 11.1.The pattern of returns for the days preceding the public announcement date yields some interesting evidence about efficient markets and information leakage. If insider trading rules were perfectly obeyed and perfectly enforced, stock prices should show no abnormal page 344returns on days before the public release of relevant news because no special firm-specific information would be available to the market before public announcement. Instead, we should observe a clean jump in the stock price only on the announcement day. In fact, Figure 11.1 shows that the prices of the takeover targets clearly start an upward drift 30 days before the public announcement. It appears that information is leaking to some market participants, who then purchase the stocks before the public announcement. Such evidence of leakage appears almost universally in event studies, suggesting at least some abuse of insider trading rules.Nevertheless, the SEC also can take some comfort from patterns such as that in Figure 11.1. If insider trading rules were widely and flagrantly violated, we would expect to see abnormal returns earlier than they appear in these results. For example, in the case of mergers, the CAR would turn positive as soon as acquiring firms decided on their takeover targets because insiders would start trading immediately. By the time of the public announcement, the insiders would have bid up the stock prices of target firms to levels reflecting the merger attempt, and the abnormal returns on the actual public announcement date would be close to zero. The dramatic increase in the CAR that we see on the announcement date indicates that a good deal of these announcements are indeed news to the market and that stock prices do not already reflect complete knowledge about the takeovers. It would appear, therefore, that SEC enforcement does have a substantial effect on restricting insider trading, even if some still persists.Event study methodology has become a widely accepted tool to measure the economic impact of a wide range of events. For example, the SEC regularly uses event studies to measure illicit gains captured by traders who may have violated insider trading or other securities laws.11 Event studies are also used in fraud cases, where the courts must assess damages caused by a fraudulent activity.Example 11.5 Using Abnormal Returns to Infer DamagesSuppose the stock of a company with market value of $100 million falls by 4% on the day that news of an accounting scandal surfaces. The rest of the market, however, generally does well that day. The market indexes are up sharply and, on the basis of the usual relationship between the stock and the market, one would have expected a 2% gain on the stock. We would conclude that the impact of the scandal was a 6% drop in value, the difference between the 2% gain that we would have expected and the 4% drop actually observed. One might then infer that the damages sustained from the scandal were $6 million because the value of the firm (after adjusting for general market movements) fell by 6% of $100 million when investors became aware of the news and reassessed the value of the stock.Concept Check 11.4Suppose that we see negative abnormal returns (declining CARs) after an announcement date. Is this a violation of efficient markets?page 345 11.4 Are Markets Efficient?The IssuesNot surprisingly, the efficient market hypothesis does not exactly arouse enthusiasm in the community of professional portfolio managers. It implies that a great deal of the activity of these managers—the search for undervalued securities—is at best wasted effort, and quite probably harmful to clients because it costs money and leads to imperfectly diversified portfolios. Consequently, the EMH has never been widely accepted on Wall Street, and debate continues today on the degree to which security analysis can improve investment performance. However, the following issues imply that the debate probably never will be settled: the magnitude issue, the selection bias issue, and the lucky event issue.The Magnitude Issue We noted that an investment manager overseeing a $5 billion portfolio who can improve performance by only .1% per year will increase investment earnings by .001 × $5 billion = $5 million annually. This manager clearly would be worth her salary! Yet can we, as observers, statistically measure her contribution? Probably not: A .1% annual contribution would be swamped by the volatility of the market. Remember, since 1926 the annual standard deviation of the well-diversified S&P 500 index has been around 20%. Against these fluctuations, a small increase in performance would be hard to detect.All might agree that stock prices are very close to fair values and that only managers of large portfolios can earn enough trading profits to make the exploitation of minor mispricing worth the effort. According to this view, the actions of intelligent investment managers are the driving force behind the constant evolution of market prices to fair levels. Rather than ask the qualitative question, Are markets efficient?, we ought instead to ask a more quantitative question: How efficient are markets?The Selection Bias Issue Suppose you discover an investment scheme that could really make money. You have two choices: either publish your technique in The Wall Street Journal to win fleeting fame or keep your technique secret and use it to earn millions of dollars. Most investors would choose the latter option, which presents us with a conundrum. Only investors who find that an investment scheme cannot generate abnormal returns will be willing to report their findings to the whole world. Hence, opponents of the efficient markets’ view of the world always can disregard evidence that various techniques do not provide investment rewards and argue that the techniques that do work simply are not being reported to the public. This is a problem in selection bias; the outcomes we are able to observe have been preselected in favor of failed attempts. Therefore, we cannot fairly evaluate the true ability of portfolio managers to generate winning stock market strategies.The Lucky Event Issue In virtually any month it seems we read an article about some investor or investment company with a fantastic investment performance over the recent past. Surely the superior records of such investors disprove the efficient market hypothesis.Yet this conclusion is far from obvious. As an analogy to the investment game, consider a contest to flip the most number of heads out of 50 trials using a fair coin. The expected outcome for any person is, of course, 50% heads and 50% tails. If 10,000 people, however, compete in this contest, it would not be surprising if at least one or two page 346contestants flipped more than 75% heads. In fact, elementary statistics tells us that the expected number of contestants flipping 75% or more heads would be two. It would be silly, though, to crown these people the “head-flipping champions of the world.” Obviously, they are simply the contestants who happened to get lucky on the day of the event. (See the nearby box.)WORDS FROM THE STREETHow to Guarantee a Successful Market NewsletterSuppose you want to make your fortune publishing a market newsletter. You first need to convince potential subscribers that you have talent worth paying for. But what if you have no talent? The solution is simple: Start eight newsletters.In year 1, let four of your newsletters predict an up-market and four a down-market. In year 2, let half of the originally optimistic group of newsletters continue to predict an up-market and the other half a down-market. Do the same for the originally pessimistic group. Continue in this manner to obtain the pattern of predictions in the table that follows (U = prediction of an up-market, D = prediction of a down-market).After three years, one of the newsletters would have had a perfect prediction record. This is because after three years there are 23 = 8 possible outcomes for the market, and we have covered all eight possibilities with the eight newsletters. Now, we simply slough off the seven unsuccessful newsletters and market the eighth newsletter based on its perfect track record. If we want to establish a newsletter with a perfect track record over a four-year period, we need 24 = 16 newsletters. A five-year period requires 32 newsletters, and so on.After the fact, the one newsletter that was always right will attract attention for your uncanny foresight and investors will rush to pay large fees for its advice. Your fortune is made, and you have never even researched the market!WARNING: This scheme is illegal! The point, however, is that with hundreds of market newsletters, you can find one that has stumbled onto an apparently remarkable string of successful predictions without any real degree of skill. After the fact, someone’s prediction history can seem to imply great forecasting skill. This person is the one we will read about in The Wall Street Journal; the others will be forgotten.Newsletter PredictionsYear 1 2 3 4 5 6 7 81 U U U U D D D D2 U U D D U U D D3 U D U D U D U DThe analogy to efficient markets is clear. If any stock is fairly priced given all available information, any bet on a stock is simply a coin toss with equal likelihood of winning or losing the bet. However, if many investors using a variety of schemes make fair bets, statistically speaking, some of those investors will be lucky and win a great majority of the bets. For every big winner, there may be many big losers, but we never hear of them. The winners, though, turn up in The Wall Street Journal as the latest stock market gurus; then they can make a fortune publishing market newsletters.Our point is that after the fact there will have been at least one successful investment scheme. A doubter will call the results luck; the successful investor will call it skill. The proper test would be to see whether the successful investors can repeat their performance in another period, yet this approach is rarely taken.With these caveats in mind, we turn now to some of the empirical tests of the efficient market hypothesis.Concept Check 11.5Legg Mason’s Value Trust, managed by Bill Miller, outperformed the S&P 500 in each of the 15 years ending in 2005. Is Miller’s performance sufficient to dissuade you from a belief in efficient markets? If not, would any performance record be sufficient to dissuade you? Now consider that in the next three years, the fund dramatically underperformed the S&P 500; by the end of 2008, its cumulative 18-year performance was barely different from the index. Does this affect your opinion?page 347 Weak-Form Tests: Patterns in Stock ReturnsReturns over Short Horizons Early tests of efficient markets were tests of the weak form. Could speculators find trends in past prices that would enable them to earn abnormal profits? This is essentially a test of the efficacy of technical analysis.One way of discerning trends in stock prices is by measuring the serial correlation of stock market returns. Serial correlation refers to the tendency for stock returns to be related to past returns. Positive serial correlation means that positive returns tend to follow positive returns (a momentum type of property). Negative serial correlation means that positive returns tend to be followed by negative returns (a reversal or “correction” property). Both Conrad and Kaul12 and Lo and MacKinlay13 examine weekly returns of NYSE stocks and find positive serial correlation over short horizons. However, the correlation coefficients of weekly returns tend to be fairly small, at least for large stocks for which price data are the most reliably up-to-date. Thus, while these studies demonstrate weak price trends over short periods,14 the evidence does not clearly suggest the existence of trading opportunities.While broad market indexes demonstrate only weak serial correlation at short horizons, for example, a month or so, there appears to be stronger momentum at longer horizons. In an investigation of intermediate-horizon stock price behavior (using 3- to 12-month holding periods), Jegadeesh and Titman15 found a momentum effect in which good or bad recent performance of particular stocks continues over time. They conclude that while the performance of individual stocks is highly unpredictable, portfolios of the best-performing stocks in the recent past appear to outperform other stocks with enough reliability to offer profit opportunities.Returns over Long Horizons Although studies of short- to intermediate-horizon returns have detected momentum in stock market prices, tests of long-horizon returns (i.e., returns over multiyear periods) have found suggestions of pronounced negative long-term serial correlation in the performance of the aggregate market.16 The latter result has given rise to a “fads hypothesis,” which asserts that the stock market might overreact to relevant news. Such overreaction leads to positive serial correlation (momentum) over short time horizons. Subsequent correction of the overreaction leads to poor performance following good performance and vice versa. The corrections mean that a run of positive returns eventually will tend to be followed by negative returns, leading to negative serial correlation over longer horizons. These episodes of apparent overshooting followed by correction give the stock market the appearance of fluctuating around its fair value.page 348 These long-horizon results are dramatic but still not conclusive. An alternative interpretation of these results holds that they indicate only that the market risk premium varies over time. For example, when the risk premium and the required return on the market rise, stock prices will fall. When the market then rises (on average) at this higher rate of return, the data convey the impression of a stock price recovery. In this view, the apparent overshooting and correction are in fact no more than a rational response of market prices to changes in discount rates.In addition to studies suggestive of overreaction in overall stock market returns over long horizons, many other studies suggest that over long horizons, extreme performance in particular securities also tends to reverse itself: The stocks that have performed best in the recent past seem to underperform the rest of the market in following periods, while the worst past performers tend to offer above-average future performance. DeBondt and Thaler17 and Chopra, Lakonishok, and Ritter18 find strong tendencies for poorly performing stocks in one period to experience sizable reversals over the subsequent period, while the best-performing stocks in a given period tend to follow with poor performance in the following period.For example, the DeBondt and Thaler study found that if one were to rank the performance of stocks over a five-year period and then group stocks into portfolios based on investment performance, the base-period “loser” portfolio (defined as the 35 stocks with the worst investment performance) outperformed the “winner” portfolio (the top 35 stocks) by an average of 25% (cumulative return) in the following three-year period. This reversal effect, in which losers rebound and winners fade back, suggests that the stock market overreacts to relevant news. After the overreaction is recognized, extreme investment performance is reversed. This phenomenon would imply that a contrarian investment strategy—investing in recent losers and avoiding recent winners—should be profitable. Moreover, these returns seem pronounced enough to be exploited profitably.Thus, it appears that there may be short-run momentum but long-run reversal patterns in price behavior both for the market as a whole and across sectors of the market. One interpretation of this pattern is that short-run overreaction (which causes momentum in prices) may lead to long-term reversals (when the market recognizes its past error).Predictors of Broad Market ReturnsSeveral studies have documented the ability of easily observed variables to predict market returns. For example, Fama and French19 showed that the return on the aggregate stock market tends to be higher when the dividend/price ratio, the dividend yield, is high. Campbell and Shiller20 found that the earnings yield can predict market returns. Keim and Stambaugh21 showed that bond market data such as the spread between yields on high- and low-grade corporate bonds also help predict broad market returns.Again, the interpretation of these results is difficult. On the one hand, they may imply that abnormal stock returns can be predicted, in violation of the efficient market hypothesis. page 349More probably, however, these variables are proxying for variation in the market risk premium. For example, given a forecast of dividends or earnings, stock prices will be lower and dividend and earnings yields will be higher when the risk premium (and therefore the expected market return) is higher. Thus, a high dividend or earnings yield will be associated with higher market returns. But rather than a violation of market efficiency, the predictability of market returns is due to predictability in the risk premium.Fama and French22 showed that the yield spread between high- and low-grade bonds has greater predictive power for returns on low-grade bonds than high-grade bonds, and greater predictive power for stock rather than bond returns, suggesting that the predictability in returns is in fact a risk premium rather than evidence of market inefficiency. Similarly, the fact that the dividend yield on stocks helps to predict bond market returns suggests that the yield captures a risk premium common to both markets rather than mispricing in the equity market.Semistrong Tests: Market AnomaliesFundamental analysis uses a much wider range of information to create portfolios than does technical analysis. Investigations of the efficacy of fundamental analysis ask whether publicly available information beyond the trading history of a security can be used to improve investment performance; as such, they are tests of semistrong-form market efficiency. Surprisingly, several easily accessible statistics, for example, a stock’s price–earnings ratio or its market capitalization, seem to predict abnormal risk-adjusted returns. Findings such as these, which we will review in the following pages, are difficult to reconcile with the efficient market hypothesis and therefore are often referred to as efficient market anomalies.A difficulty in interpreting these tests is that we usually need to adjust for portfolio risk before evaluating the success of an investment strategy. Some tests, for example, have used the CAPM to adjust for risk. However, we know that even if beta is a relevant descriptor of stock risk, the empirically measured quantitative trade-off between risk as measured by beta and expected return differs from the predictions of the CAPM. (We review this evidence in Chapter 13.) If we use the CAPM to adjust portfolio returns for risk, inappropriate adjustments may lead to the conclusion that various portfolio strategies can generate superior returns, when, in fact, the risk adjustment procedure has simply failed.Another way to put this is to note that tests of risk-adjusted returns are joint tests of the efficient market hypothesis and the risk adjustment procedure. If it appears that a portfolio strategy can generate superior returns, we must then choose between rejecting the EMH and rejecting the risk adjustment technique. Usually, the risk adjustment technique is based on more-questionable assumptions than is the EMH; by opting to reject the procedure, we are left with no conclusion about market efficiency.An example of this issue is the discovery by Basu23 that portfolios of low price–earnings (P/E) ratio stocks have provided higher returns than high P/E portfolios. The P/E effect holds up even if returns are adjusted for portfolio beta. Is this a confirmation that the market systematically misprices stocks according to P/E ratio? This would be an extremely surprising and, to us, disturbing conclusion because P/E ratios are so page 350simple to observe. Although it may be possible to earn superior returns through unusual insight, it hardly seems plausible that such a simplistic technique is enough to generate abnormal returns.Another interpretation of these results is that returns are not properly adjusted for risk. If two firms have the same expected earnings, the riskier stock will sell at a lower price and lower P/E ratio. Because of its higher risk, the low P/E stock also will have higher expected returns. Therefore, unless the CAPM beta fully adjusts for risk, P/E will act as a useful additional descriptor of risk and will be associated with abnormal returns if the CAPM is used to establish benchmark performance.The Small-Firm Effect The so-called size or small-firm effect, originally documented by Banz,24 is illustrated in Figure 11.3. It shows the historical performance of portfolios formed by dividing the NYSE stocks into 10 portfolios each year according to firm size (i.e., the total value of outstanding equity). Average annual returns between 1926 and 2018 are consistently higher on the small-firm portfolios. The difference in average annual return between portfolio 10 (with the largest firms) and portfolio 1 (with the smallest firms) is 7.32%. Of course, the smaller-firm portfolios tend to be riskier. But even when returns are adjusted for risk using the CAPM, there is still a consistent premium for the smaller-sized portfolios.Figure 11.3 Average annual return for 10 size-based portfolios, 1926–2018Source: Authors’ calculations, using data obtained from Professor Ken French’s data library at http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html.Imagine earning a premium of this size on a billion-dollar portfolio. Yet it is remarkable that following a simple (even simplistic) rule such as “invest in low-capitalization stocks” should enable an investor to earn excess returns. After all, any investor can measure firm size at little cost. One would not expect such minimal effort to yield such large rewards.page 351 The Neglected-Firm and Liquidity Effects Arbel and Strebel25 gave another interpretation of the small-firm effect. Because small firms tend to be neglected by large institutional traders, information about smaller firms is less available. This information deficiency makes smaller firms riskier investments that command higher returns. “Brand-name” firms, after all, are subject to considerable monitoring from institutional investors, which promises high-quality information, and presumably investors do not purchase “generic” stocks without the prospect of greater returns.Merton26 provides a rationale for this neglected-firm effect. He shows that neglected firms might be expected to earn higher equilibrium returns as compensation for the risk associated with limited information. In this sense the neglected-firm premium is not strictly a market inefficiency, but is in fact a type of risk premium.Work by Amihud and Mendelson27 on the effect of liquidity on stock returns might be related to both the small-firm and neglected-firm effects. As we noted in Chapter 9, investors will demand a rate-of-return premium to invest in less-liquid stocks that entail higher trading costs. In accord with this hypothesis, Amihud and Mendelson showed that these stocks have a strong tendency to exhibit abnormally high risk-adjusted rates of return. Because small and less-analyzed stocks as a rule are less liquid, the liquidity effect might be a partial explanation of their abnormal returns. However, exploiting these effects can be more difficult than it would appear. The high trading costs on small stocks can easily wipe out any apparent abnormal profit opportunity.Book-to-Market Ratios Fama and French28 showed that a powerful predictor of returns across securities is the ratio of the book value of the firm’s equity to the market value of equity. Fama and French stratified firms into 10 groups according to book-to-market ratios and examined the average monthly rate of return of each of the 10 groups. Figure 11.4 is an updated version of their results. The decile with the highest book-to-market ratio had an average annual return of 16.5%, while the lowest-ratio decile averaged only 11.0%. The dramatic dependence of returns on book-to-market ratio is page 352independent of beta, suggesting either that high book-to-market ratio firms are relatively underpriced, or that the book-to-market ratio is serving as a proxy for a risk factor that affects equilibrium expected returns.Figure 11.4 Average return as a function of book-to-market ratio, 1926–2018Source: Authors’ calculations, using data obtained from Professor Ken French’s data library at http://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html.In fact, Fama and French found that after controlling for the size and book-to-market effects, beta seemed to have no power to explain average security returns.29 This finding is an important challenge to the notion of rational markets because it seems to imply that a factor that should affect returns—systematic risk—seems not to matter, while a factor that should not matter—the book-to-market ratio—seems capable of predicting future returns.Post–Earnings-Announcement Price Drift A fundamental principle of efficient markets is that any new information ought to be reflected in stock prices very rapidly. A puzzling anomaly, therefore, is the apparently sluggish response of stock prices to firms’ earnings announcements, as uncovered by Ball and Brown.30 Their results were later confirmed and extended in many other papers.31The “news content” of an earnings announcement can be evaluated by comparing the announcement of actual earnings to the value previously expected by market participants. The difference is the “earnings surprise.” (Market expectations of earnings can be roughly measured by averaging the published earnings forecasts of Wall Street analysts or by applying trend analysis to past earnings.) Rendleman, Jones, and Latané32 provide an influential study of sluggish price response to earnings announcements. They calculate earnings surprises for a large sample of firms, rank the magnitude of the surprise, divide firms into 10 deciles based on the size of the surprise, and calculate abnormal returns for each decile. Figure 11.5 plots cumulative abnormal returns by decile.Figure 11.5 Cumulative abnormal returns in response to earnings announcementsSource: R. J. Rendleman Jr., C. P. Jones, and H. A. Latané, “Empirical Anomalies Based on Unexpected Earnings and the Importance of Risk Adjustments,” Journal of Financial Economics 10 (1982), pp. 269–287.Their results are dramatic. The correlation between ranking by earnings surprise and abnormal returns across deciles is unsurprising. There is a large abnormal return (a jump in cumulative abnormal return) on the earnings announcement day (time 0). The abnormal return is positive for positive-surprise firms and negative for negative-surprise firms.The more remarkable result of the study concerns stock price movement after the announcement date. The cumulative abnormal returns of positive-surprise stocks continue to rise—in other words, exhibit momentum—even after the earnings information becomes public, while the negative-surprise firms continue to suffer negative abnormal returns. The market appears to adjust to the earnings information only gradually, resulting in a sustained period of abnormal returns.Evidently, one could have earned abnormal profits simply by waiting for earnings announcements and purchasing a stock portfolio of positive-earnings-surprise companies. page 353These are precisely the types of predictable continuing trends that ought to be impossible in an efficient market.Other Predictors of Stock ReturnsOur list of anomalies could go on and on. We will close with just a brief mention of some other characteristics that seem to have predicted stock returns.Volatility. While the CAPM predicts that idiosyncratic volatility should not be related to stock returns, it appears that at intermediate horizons of 3–12 months, volatility is negatively associated with returns.33Accruals and earnings quality. Accruals measure the component of earnings that does not reflect actual cash flows. For example, if a firm sells an item on credit, it may report a profit, but the immediate impact of the sale is an increase in accounts receivable (an accrual), not cash. High accruals have predicted low future returns.34 This is sometimes viewed as an earnings quality factor as it appears that management can and sometimes does manipulate accruals to paint a rosy view of earnings.35Growth. More rapidly growing firms, for example, with high capital investments, asset growth, or high recent share issuance, tend to have lower future returns.36Profitability. Gross profitability seems to predict higher stock returns.37 Gross profitability is computed by adding back some items conventionally treated as expenses, for page 354example, advertising or research and development. The idea is that these expenditure may be better viewed as investments than expenses if they will enhance future earnings.Strong-Form Tests: Inside InformationIt would not be surprising if insiders were able to make superior profits trading in their firm’s stock. In other words, we do not expect markets to be strong-form efficient; we regulate and limit trades based on inside information. The ability of insiders to trade profitably in their own stock has been documented in studies by Jaffe,38 Seyhun,39 Givoly and Palmon,40 and others. Jaffe’s was one of the first to document the tendency for stock prices to rise after insiders intensively bought shares and fall after insiders intensively sold shares.Can other investors benefit by following insiders’ trades? The Securities and Exchange Commission requires insiders to register their trading activity, and these trades become public. If markets are efficient, fully and immediately processing that information, an investor should not be able to profit from following those trades. Several Internet sites contain information on insider trading.The study by Seyhun found that following insider transactions would be to no avail. Although there was some tendency for stock prices to increase even after the report of insider buying, the abnormal returns were not large enough to overcome transaction costs.Interpreting the AnomaliesHow should we interpret the ever-growing anomalies literature? Does it imply that markets are grossly inefficient, allowing for simplistic trading rules to offer large profit opportunities? Or are there other, more-subtle interpretations?Risk Premiums or Inefficiencies? The small-firm, book-to-market, momentum, and long-term reversal effects are currently among the most puzzling phenomena in empirical finance. There are several interpretations of these effects. First note that to some extent, some of these phenomena may be related. One feature that small firms, high-book-to-market firms, and recent “losers” seem to have in common is a stock price that has fallen considerably in recent months or years. Indeed, a firm can become a small firm or a high-book-to-market firm by suffering a sharp drop in price. These groups therefore may contain a relatively high proportion of distressed firms that have suffered recent difficulties.Fama and French41 argue that these effects can be explained as manifestations of risk premiums. Using their three-factor model, introduced in the previous chapter, they show that stocks with higher betas (also known as factor loadings in this context) on size or book-to-market factors have higher average returns; they interpret these returns as evidence of a risk premium associated with the factor. This model does a much better job than the page 355one-factor CAPM in explaining security returns. While size or book-to-market ratios per se are obviously not risk factors, they perhaps might act as proxies for more fundamental determinants of risk. In this regard, it is noteworthy that returns to “style portfolios,” for example, the return on portfolios constructed based on the ratio of book-to-market value (specifically, the Fama-French high-minus-low book-to-market portfolio) or firm size (the return on the small-minus-big firm portfolio), do indeed seem to predict business cycles in many countries. Figure 11.6 shows that returns on these portfolios tend to have positive returns in years prior to rapid growth in gross domestic product.Figure 11.6 Return to style portfolios as predictors of GDP growth. Average difference in the return on the style portfolio in years before good GDP growth versus in years with bad GDP growth. Positive value means the style portfolio does better in years prior to good macroeconomic performance. HML = high-minus-low portfolio, sorted on ratio of book-to-market value. SMB = small-minus-big portfolio, sorted on firm size.Source: J. Liew and M. Vassalou, “Can Book-to-Market, Size, and Momentum Be Risk Factors That Predict Economic Growth?,” Journal of Financial Economics 57 (2000), pp. 221–45.The opposite interpretation is offered by Lakonishok, Shleifer, and Vishny,42 who argue that these phenomena are evidence of inefficient markets, more specifically, of systematic errors in the forecasts of stock analysts. They believe that analysts extrapolate past performance too far into the future, and therefore overprice firms with recent good performance and underprice firms with recent poor performance. Ultimately, when market participants recognize their errors, prices reverse. This explanation is consistent with the reversal effect and also, to a degree, with the small-firm and book-to-market effects because firms with sharp price drops may tend to be small or have high book-to-market ratios.If Lakonishok, Shleifer, and Vishny are correct, we ought to find that analysts systematically err when forecasting returns of recent “winner” versus “loser” firms. A study by La Porta43 is consistent with this pattern. He finds that shares of firms for which page 356analysts predict low growth rates of earnings actually perform better than those with high expected earnings growth. Analysts seem overly pessimistic about firms with low growth prospects and overly optimistic about firms with high growth prospects. When these too-extreme expectations are “corrected,” the low-expected-growth firms outperform high-expected-growth firms.Anomalies or Data Mining? We have covered several of the so-called anomalies cited in the literature, but our list could go on and on. Some wonder whether these anomalies are really unexplained puzzles in financial markets, or whether they instead are an artifact of data mining. After all, if one reruns the computer database of past returns over and over and examines stock returns along enough dimensions, simple chance will cause some criteria to appear to predict returns.Still, even acknowledging the potential for data mining, a common thread seems to run through many of these anomalies, lending support to the notion that there is a real puzzle to explain. Value stocks—defined by low P/E ratio, high book-to-market ratio, or depressed prices relative to historic levels—seem to have provided higher average returns than “glamour” or growth stocks.One way to address the problem of data mining is to find a dataset that has not already been researched and see whether the relationship in question shows up in the new data. Such studies have revealed size, momentum, and book-to-market effects in security markets around the world. While these phenomena may be a manifestation of a systematic risk premium, the precise nature of that risk is not fully understood.Anomalies over Time We pointed out earlier that while no market can be perfectly efficient, in well-functioning markets, anomalies ought to be self-destructing. As market participants learn of profitable trading strategies, their attempts to exploit them should move prices to levels at which abnormal profits are no longer available.McLean and Pontiff44 test this dynamic. They identify 97 characteristics identified in the academic literature as associated with abnormal returns and track the publication date of each finding. This allows them to break the sample for each anomaly at dates corresponding to when that particular finding became public. They conclude that the post-publication decay in abnormal return is about 60% (e.g., a 5% abnormal return prepublication falls on average to 2% after publication).45 They show that trading volume and variance in stocks identified with anomalies increase, as does short interest in “overpriced” stocks. These patterns are consistent with informed participants attempting to exploit newly recognized mispricing. Moreover, the decay in alpha is most pronounced for stocks that are larger, are more liquid, and have low idiosyncratic risk. These are precisely the stocks for which trading activity in pursuit of reliable abnormal returns is most feasible. Thus, while abnormal returns do not fully disappear, these results are consistent with a market groping its way toward greater efficiency over time.Chordia, Subrahmanyam, and Tong46 find evidence that liquidity and low trading costs facilitate efficient price discovery. They focus on abnormal returns associated with several page 357characteristics including size, book-to-market ratio, momentum, and turnover (which may be inversely related to the neglected-firm effect). They break their sample at 1993, shortly before tick sizes began their rapid decline in U.S. markets, and show that the abnormal returns associated with many of these characteristics in the pre-1993 period largely disappear in the post-1993 period. Their interpretation is that the market became more efficient as the costs of taking advantage of anomalies declined.Bubbles and Market EfficiencyEvery so often, asset prices seem (at least in retrospect) to lose their grounding in reality. For example, in the tulip mania in 17th-century Holland, tulip prices peaked at several times the annual income of a skilled worker. This episode has become the symbol of a speculative “bubble” in which prices appear to depart from any semblance of intrinsic value. Bubbles seem to arise when a rapid run-up in prices creates a widespread expectation that they will continue to rise. As more and more investors try to get in on the action, they push prices even further. Inevitably, however, the run-up stalls and the bubble ends in a crash.Less than a century after tulip mania, the South Sea Bubble in England became almost as famous. In this episode, the share price of the South Sea Company rose from £128 in January 1720 to £550 in May and peaked at around £1,000 in August—just before the bubble burst and the share price collapsed to £150 in September, leading to widespread bankruptcies among those who had borrowed to buy shares on credit. In fact, the company was a major lender of money to investors willing to buy (and thus bid up) its shares. This sequence may sound familiar to anyone who lived through the dot-com boom and bust of 1995–200247 or, more recently, the financial turmoil of 2008, with origins widely attributed to a collapsing bubble in housing prices.It is hard to defend the position that security prices in these instances represented rational, unbiased assessments of intrinsic value. And, in fact, some economists, most notably Hyman Minsky,48 have suggested that bubbles arise naturally. During periods of stability and rising prices, investors extrapolate that stability into the future and become more willing to take on risk. Risk premiums shrink, leading to further increases in asset prices, and expectations become even more optimistic in a self-fulfilling cycle. But in the end, pricing and risk taking become excessive and the bubble bursts. Ironically, the initial period of stability fosters behavior that ultimately results in instability.But beware of jumping to the conclusion that asset prices may generally be thought of as arbitrary and obvious trading opportunities abundant. First, most bubbles become “obvious” only in retrospect. At the time, the price run-up often seems to have a defensible rationale. In the dot-com boom, for example, many contemporary observers rationalized stock price gains as justified by the prospect of a new and more profitable economy, driven by technological advances. Even the irrationality of the tulip mania may have been overblown in its later retelling.49 In addition, security valuation is intrinsically difficult and estimates of intrinsic value are inevitably imprecise. Given this, large bets on perceived mispricing may entail hubris.page 358 Moreover, even if you suspect that prices are in fact “wrong,” taking advantage of them can be difficult. We explore these issues in more detail in the following chapter, but for now, we simply point out some impediments to making aggressive bets against an asset, among them the costs of short selling overpriced securities as well as potential problems obtaining the securities to sell short, and the possibility that even if you are ultimately correct, the market may disagree and prices still can move dramatically against you in the short term, thus wiping out your capital.11.5 Mutual Fund and Analyst PerformanceWe have documented some of the apparent chinks in the armor of efficient market proponents. For investors, the issue of market efficiency boils down to whether skilled investors can make consistent abnormal trading profits, so we will compare the performance of market professionals to that of a passive index fund. We will look at two facets of professional performance: that of stock market analysts who recommend investment positions and that of mutual fund managers who actually manage portfolios.Stock Market AnalystsStock market analysts historically have worked for brokerage firms, which presents an immediate problem in interpreting the value of their advice: Analysts have tended to be overwhelmingly positive in their assessment of the prospects of firms.50 For example, on a scale of 1 (strong buy) to 5 (strong sell), the average recommendation for 5,628 covered firms in 1996 was 2.04.51 As a result, we cannot take positive recommendations (e.g., to buy) at face value. Instead, we must look at either the relative enthusiasm of analyst recommendations, compared to those for other firms, or at the change in consensus recommendations.Womack52 focuses on changes in analysts’ recommendations and finds that positive changes are associated with increased stock prices of about 5%, and negative changes result in average price decreases of 11%. One might wonder whether these price changes reflect the market’s recognition of analysts’ superior information or insight about firms or, instead, simply result from new buy or sell pressure brought on by the recommendations themselves. Womack argues that price impact seems to be permanent and therefore consistent with the hypothesis that analysts do in fact reveal new information. Jegadeesh, Kim, Krische, and Lee53 also find that changes in consensus recommendations are reliably associated with price changes, but that the levels of recommendations are not.Barber, Lehavy, McNichols, and Trueman54 focus on the level of consensus recommendations and show that firms with the most-favorable recommendations outperform those page 359with the least-favorable recommendations. However, they note that portfolio strategies based on analyst recommendations would result in heavy trading activity with associated costs that probably would wipe out the potential profits from the strategy.In sum, the literature suggests that some value is added by analysts, but questions remain. Are superior returns following analyst upgrades due to revelation of new information or due to changes in investor demand in response to the changed outlook? Also, are these results exploitable by investors who necessarily incur trading costs?Mutual Fund ManagersAs we pointed out in Chapter 4, casual evidence does not support the claim that professionally managed portfolios can consistently beat the market. Figure 4.2 in that chapter demonstrated that between 1972 and 2018 the returns of a passive portfolio indexed to the Wilshire 5000 typically would have been better than those of the average equity fund. Moreover, there was no evidence of consistency in performance.However, the analyses cited in Chapter 4 were based on total returns, without adjustment for exposure to systematic risk factors. In this section we revisit the question of mutual fund performance, paying more attention to the benchmark against which performance ought to be evaluated.As a first pass, we might examine the risk-adjusted returns (i.e., the alpha) of a large sample of mutual funds. But the market index may not be an adequate benchmark against which to evaluate mutual fund returns. For example, suppose mutual funds tend to maintain considerable holdings in equity of small firms, whereas the capitalization-weighted index is dominated by large firms. Then funds as a whole will tend to outperform the index when small firms outperform large ones and underperform when small firms fare worse.The importance of the benchmark can be illustrated by examining the returns on small stocks in various subperiods.55 In the 20-year period between 1945 and 1964, for example, a small-stock index underperformed the S&P 500 by about 4% per year (i.e., the alpha of the small-stock index after adjusting for systematic risk was −4%). In the following 20-year period between 1965 and 1984, small stocks outperformed the S&P index by 10%. Thus, if one were to examine mutual fund returns in the earlier period, they would tend to look poor, not necessarily because fund managers were poor stock pickers, but simply because mutual funds as a group tended to hold more small stocks than were represented in the S&P 500. In the later period, funds would look better on a risk-adjusted basis relative to the S&P 500 because small stocks performed better. The “style choice,” that is, the exposure to small stocks (which is an asset allocation decision) would dominate the evaluation of performance even though it has little to do with managers’ stock-picking ability.56The conventional performance benchmark today is a four-factor model, which employs the three Fama-French factors (the return on the market index, and returns to portfolios based on size and book-to-market ratio) augmented by a momentum factor (a portfolio constructed based on prior-year stock return). Alphas constructed using an expanded index model using these four factors control for a wide range of style choices that may page 360affect returns, for example, an inclination to growth versus value or small- versus large-capitalization stocks. Figure 11.7 shows a frequency distribution of four-factor alphas for U.S. domestic equity funds.57 The results show that the distribution of alpha is roughly bell shaped, with a slightly negative mean. On average, it does not appear that these funds outperform their style-adjusted benchmarks.Figure 11.7 Mutual fund alphas computed using a four-factor model of expected return, 1993–2007. (The best and worst 2.5% of observations are excluded from this distribution.)Source: Professor Richard Evans, University of Virginia, Darden School of Business.Consistent with Figure 11.7, Fama and French58 use the four-factor model to assess the performance of equity mutual funds and show that, while they may exhibit positive alphas before fees, after the fees charged to their customers, average alpha is negative. Likewise, Wermers,59 who uses both style portfolios as well as the characteristics of the stocks held by mutual funds to control for performance, also finds positive gross alphas but negative net alphas after controlling for fees and risk.Carhart60 reexamines the issue of consistency in mutual fund performance and finds that, after controlling for these factors, there is only minor persistence in relative performance across managers. Moreover, much of that persistence seems due to expenses and transactions costs rather than gross investment returns.However, Bollen and Busse61 do find evidence of performance persistence, at least over short horizons. They rank mutual fund performance using the four-factor model over a page 361base quarter, assign funds into one of ten deciles according to base-period alpha, and then look at performance in the following quarter. Figure 11.8 illustrates their results. The dark line is the average alpha of funds within each of the deciles in the base period (expressed on a quarterly basis). The steepness of that curve reflects the considerable dispersion in performance in the ranking period. The light line is the average performance of the funds in each decile in the following quarter. The shallowness of this curve indicates that most of the original performance differential disappears. Nevertheless, the plot is still clearly downward-sloping, so it appears that, at least over a short horizon such as one quarter, there is some performance consistency. However, that persistence is probably too small a fraction of the original performance differential to justify performance-chasing by mutual fund customers.Figure 11.8 Risk-adjusted performance in ranking quarter and following quarterThis pattern is actually consistent with the prediction of an influential paper by Berk and Green.62 They argue that skilled mutual fund managers with abnormal performance will attract new funds until the additional costs and challenges of managing those extra funds drive alphas down to zero. Thus, skill will show up not in superior returns, but rather in the amount of funds under management. Therefore, even if managers are skilled, alphas will be short-lived, as they seem to be in Figure 11.8.Del Guercio and Reuter63 offer a finer interpretation of mutual fund performance. They split mutual fund investors into those who buy funds directly for themselves versus those who purchase funds through brokers, reasoning that the direct-sold segment may be more financially literate while the broker-sold segment is less comfortable making financial decisions without professional advice. Consistent with this hypothesis, they show that page 362direct-sold investors direct their assets to funds with positive alphas (consistent with the Berk-Green model), but broker-sold investors generally do not. This provides a greater incentive for direct-sold funds to invest relatively more in alpha-generating inputs such as talented portfolio managers or analysts. Moreover, they show that the after-fee performance of direct-sold funds is as good as that of index funds (again, consistent with Berk-Green), while the performance of broker-sold funds is considerably worse. It thus appears that the average underperformance of actively managed mutual funds is driven largely by broker-sold funds and that this underperformance may be interpreted as an implicit cost that less-informed investors pay for the advice they get from their brokers.In contrast to the extensive studies of equity fund managers, there have been few studies of the performance of bond fund managers. Blake, Elton, and Gruber64 examined the performance of fixed-income mutual funds. They found that, on average, bond funds underperform passive fixed-income indexes by an amount roughly equal to expenses, and that there is no evidence that past performance can predict future performance. More recently, Chen, Ferson, and Peters (2010)65 found that, on average, bond mutual funds outperform passive bond indexes in terms of gross returns but underperform once the fees they charge their investors are subtracted, a result similar to those others have found for equity funds.Thus, the evidence on the risk-adjusted performance of professional managers is mixed at best. We conclude that the performance of professional managers is broadly consistent with market efficiency. The amounts by which professional managers as a group beat or are beaten by the market fall within the margin of statistical uncertainty. In any event, it is quite clear that performance superior to passive strategies is far from routine. Studies show either that most managers cannot outperform passive strategies or that if there is a margin of superiority, it is small.On the other hand, a small number of investment superstars—Peter Lynch (formerly of Fidelity’s Magellan Fund), Warren Buffett (of Berkshire Hathaway), John Templeton (formerly of Templeton Funds), and Mario Gabelli (of GAMCO), among them—have compiled career records that show a consistency of superior performance hard to reconcile with absolutely efficient markets. In an analysis of mutual fund “stars,” Kosowski, Timmerman, Wermers, and White66 conclude that the stock-picking ability of a minority of managers is sufficient to cover their costs, and that their superior performance tends to persist over time. However, Nobel Prize–winner Paul Samuelson67 reviewed this investment hall of fame and pointed out that the records of the vast majority of professional money managers offer convincing evidence that there are no easy strategies to guarantee success in the securities markets.So, Are Markets Efficient?There is a telling joke about two economists walking down the street. They spot a $20 bill on the sidewalk. One stoops to pick it up, but the other one says, “Don’t bother; if the bill were real someone would have picked it up already.”page 363 The lesson is clear. An overly doctrinaire belief in efficient markets can paralyze the investor and make it appear that no research effort can be justified. This extreme view is probably unwarranted. There are enough anomalies in the empirical evidence to justify the search for underpriced securities that clearly goes on.The bulk of the evidence, however, suggests that any supposedly superior investment strategy should be taken with many grains of salt. The market is competitive enough that only differentially superior information or insight will earn money; the easy pickings have been picked. In the end it is likely that the margin of superiority that any professional manager can add is so slight that the statistician will not easily be able to detect it.We conclude that markets are generally very efficient, but that rewards to the especially diligent, intelligent, or creative may in fact be waiting.