[An updated version of this article can be found at Efficient Capital Markets in the 2nd edition.]
Shortly after the Constitution went into effect, Secretary of the Treasury Alexander Hamilton proposed that Congress redeem at face value securities that had been issued by the states and the federal government. At the time, these securities were selling for much less than face value because people were uncertain whether they would ever be redeemed. After Hamilton's proposal was made public but before it was adopted, however, congressmen and others who knew of the redemption plan made large profits by sending their agents into the countryside to buy the securities at depressed prices before most security holders heard of the plan.
Contrast this scenario with security markets today, in which the prices of securities react very quickly to new information about their value. In fact, the market often anticipates and reacts to news before it is officially made public. For example, General Motors announced a major restructuring in December 1991, closing twenty-one factories and cutting seventy-four thousand jobs. On the day of the announcement GM's stock price fell by only 0.4 percent because the market had already incorporated expectations about the restructuring into its price. The market reacted only to the difference between the anticipated news and what was actually announced.
To an economist the difference between the market in the late 1700s and today is that today's market is more "efficient" at incorporating information into security prices. Efficient capital markets are commonly thought of as markets in which security prices fully reflect all relevant information that is available about the fundamental value of the securities. Because a security is a claim on future cash flows, this fundamental value is the present value of the future cash flows that the owner of the security expects to receive. The cash flows anticipated for stocks consist of the stream of expected dividends paid to stockholders plus the expected price of the stock when sold. In the present value calculation, future cash flows are discounted by an interest rate that is a function of the riskiness of those cash flows. The riskier the cash flows, the higher is the rate used in discounting.
Theoretically, the profit opportunities represented by the existence of "undervalued" and "overvalued" stocks motivate competitive trading by investors that moves the prices of stocks toward the present value of the future cash flows. For example, new information about the fundamental values of securities will be reflected in prices through competitive trading. Thus, the search for mispriced stocks by investment analysts and their subsequent trading make the market efficient and make prices reflect fundamental values.
Due to technological innovation and organized markets such as the New York Stock Exchange, information is now relatively cheap to obtain and process. Thus, we can see why securities markets today are more efficient than in the late 1700s. It is in this environment of relatively low-cost information and active security analysis that the theory of efficient capital markets has developed.
The study of capital market efficiency examines how much, how fast, and how accurately available information is incorporated into security prices. Financial economists often classify efficiency into three categories based on what is meant as "available information"—the weak, semistrong, and strong forms. Weak-form efficiency exists if security prices fully reflect all the information contained in the history of past prices and returns. (The return is the profit on the security calculated as a percentage of an initial price.) If capital markets are weak-form efficient, then investors cannot earn excess profits from trading rules based on past prices or returns. Therefore, stock returns are not predictable, and so-called technical analysis (analyzing patterns in past price movements) is useless.
Under semistrong-form efficiency, security prices fully reflect all public information. Thus, only traders with access to nonpublic information, such as some corporate insiders, can earn excess profits. Under weak-form efficiency, some public information about fundamentals may not yet be reflected in prices. Thus, a superior analyst can profit from trading on the discovery of, or a better interpretation of, public information. Under semistrong-form efficiency, the market reacts so quickly to the release of new information that there are no profitable trading opportunities based on public information.
Finally, under strong-form efficiency, all information—even apparent company secrets—is incorporated in security prices; thus, no investor can earn excess profit trading on public or nonpublic information.
Why does informational efficiency matter? The capital markets channel funds from savers to firms, which use the funds to finance projects. Informational efficiency is necessary if funds, allocated through the capital market, are to flow to the highest-valued projects. Shareholders want management to maximize stock prices and thus will attempt to ensure that their managements undertake only projects (decisions) that increase the value of their stock. Management compensation packages tied to stock performance are one way in which stockholders align management's interests with their own. However, maximization of stock prices can result in the capital market directing funds to the most valuable projects only if stocks are efficiently priced, in the sense of accurately reflecting the fundamental value of all future cash flows. Thus, for example, if capital markets are efficient, there is no reason to expect managements to emphasize the short run at the expense of long-term projects. Additionally, efficient capital markets make it easier for firms to raise capital because the markets determine the prices at which existing and potential security holders are willing to exchange claims on a firm's future cash flows.
A related reason for caring about efficiency is that investors who do not have the time or the resources to do extensive analysis will be more willing to invest their savings in the market if they believe the securities they trade are accurately priced. This, in turn, helps the capital market to perform its function of translating savings into productive projects. Finally, there are policy implications of evidence on market efficiency. If capital markets are efficient, then the government's role in capital markets should be very limited. If security prices do not accurately reflect fundamentals, however, there might be a case for regulating both the operation of the securities markets and the capital-allocation process itself.
A large amount of empirical research has been directed at answering whether capital markets are efficient. Most research has used stock price data, for two reasons. First, stock prices are easily available. Second, the stock market is likely to be less efficient than other securities markets (such as the bond market) because cash flows paid to stockholders are relatively uncertain, and there is no terminal payoff as in a bond. Therefore, stocks are relatively difficult to value, and evidence of stock market efficiency would be compelling evidence of efficiency in securities markets in general.
An overwhelming amount of empirical evidence shows that stock prices react quickly, in the expected direction, to the release of information. Stock prices react within ten minutes to an earnings announcement, for example. This evidence is consistent with weak and semistrong efficiency. Such evidence, however, does not show that the amount of price reaction accurately reflects fundamentals or, by extension, that security prices accurately reflect the fundamental value of the securities. Other evidence shows that corporate insiders have earned excess profits trading on inside information. This evidence means that capital markets are not strong-form efficient. Today, the empirical debate on market efficiency centers on whether future returns are predictable.
The empirical tests of capital market efficiency began even before Eugene Fama of the University of Chicago offered a theory in 1970. The early tests hypothesized that if prices fully reflected available information, if information arrives randomly, and if expected returns are constant, then stock returns from one period to the next should be statistically independent. That is, they should follow what has loosely been referred to as a "random walk." This implies that historical returns are useless for predicting future returns, which is consistent with weak-form market efficiency.
The early tests, using various statistical methods, generally conclude that the past short-horizon (daily and weekly) returns of individual stocks are economically insignificant for predicting future returns. Consequently, the joint hypothesis of market efficiency and constant expected—but not actual—returns was generally accepted. Fama later refined the definition of capital market efficiency so that prices must not only fully, but correctly, reflect all available information. This implies that the market price should be a reasonable estimate of the rationally determined fundamentals.
By the early eighties the near consensus among academics in finance that capital markets are efficient started to fade for two reasons. First, researchers found anomalies in stock returns. One anomaly was that firms with low P/E ratios (ratios of stock prices to annual earnings per share) earn higher-than-normal returns. Researchers also found so-called January and day-of-the-week effects: stocks of small firms tend to earn excess returns in January, while Monday returns tend to be low. However, these anomalies could be due to misspecification of the models used in the tests, or to institutional factors (such as the impact of taxes), rather than market inefficiency. Consequently, they represent only an indirect attack on efficiency.
A second kind of evidence was a more direct challenge to market efficiency. Robert Shiller and others argued that the aggregate stock market has been much more volatile than can be justified by actual dividend changes (which represent fundamentals). Lawrence Summers shows that this evidence may indicate that stock prices take long slow swings away from fundamental values that would not be detectable in the early short-horizon return tests.
Shiller, Summers, and others assert that a deviation of prices from fundamental values may be caused by, or persist because of, fads or other manifestations of irrational behavior. In their models, unlike in traditional financial theory, the marginal trader who moves prices may not be rational or may not trade based on fundamentals. Therefore, competition does not necessarily eliminate mispricing because the rational trader cannot be certain that prices will converge on fundamental values, especially in the short term.
Consistent with these assertions, Fama and Kenneth French and, separately, James Poterba and Summers report that long-horizon (two- to ten-year) stock index returns tend to follow what is called a mean-reverting pattern through time. That is, periods of relatively high returns tend to be followed by periods of relatively low returns and vice versa. Summers, Poterba, and Shiller conclude from this evidence that prices often move away from their fundamentals and that markets are, therefore, inefficient. But Fama and French suggest another explanation consistent with market efficiency—that actual returns are mean reverting because rationally determined expected returns are mean reverting.
The evidence of mean reversion—and therefore predictable long-term patterns—focuses on long-horizon index or portfolio returns rather than the returns of individual stocks. There is little evidence of mean reversion in the returns of individual stocks beyond what can be attributed to transaction costs. This suggests that mean-reverting return patterns are systematic across stocks, such that the general level of expected returns may change through time depending on macroeconomic conditions. During economic declines, for example, demanders of capital may need to offer higher levels of expected return to induce individuals to save. Consequently, the new evidence of predictability in index and portfolio returns amounts to a rejection of the constant expected returns model that was implicit in definitions of weak-form efficiency. Predictability in stock market indexes alone, however, is not enough evidence to reject the more basic implication of market efficiency that the market price should be a reasonable estimate of the rationally determined fundamentals.
Fama and French provide support for their argument with evidence on dividend yields and the "default spread." (The default spread is the premium that compensates for the risk of default.) They use dividend yields as a rough measure of expected returns on stocks, and the default spread as a rough measure of expected returns on bonds. They show that both are high during periods of economic decline and low during economic booms. In addition, the common variation in expected returns across securities, explained by the dividend yield and default spread, increases from low-risk to high-risk stocks and from low-grade to high-grade bonds, respectively. This is as would be anticipated in an efficient market, where expected returns vary with economic conditions. On the other hand, this common variation in expected returns may simply indicate that mispricing is systematic. For example, high dividend yields may indicate that stocks, in general, are temporarily undervalued rather than that expected returns are relatively high. Consequently, it may never be possible to precisely determine if the stock market rationally reflects fundamental values.
The main event that gained support for the view that capital markets are inefficient was the 22 percent drop in the Dow-Jones stock index on Monday, October 19, 1987. This happened even though little news about fundamentals was released over the weekend before the crash. The crash in the United States, however, actually began the Wednesday through Friday of the week before the Monday crash (October 14 through 16), when the Standard and Poor's 500 index had fallen 10.44 percent. This decline was the largest one-, two-, or three-day drop in the market in more than forty-five years (since May 13-14, 1940, when German tanks unexpectedly broke through French armies, sealing France's fate in World War II).
Mark Mitchell and Jeffry Netter present evidence that the large decline in the U.S. market from October 14 through 16 was largely a rational reaction to an unanticipated tax proposal by the House Ways and Means Committee limiting the deductibility of interest expense on corporate debt, especially in takeovers. This decline may have triggered portfolio insurance sales on October 19 that the exchanges were not prepared to handle. This liquidity crunch may have furthered depressed the market on that day. Thus, efficient markets theory is consistent with at least part of the market decline from October 14 through October 19, 1987. It may also be that the efficiency of capital markets varies through time. For instance, lessons learned in the 1987 crash by traders, regulators, and the exchanges may have resulted in more efficient capital markets.
The debate on how well security prices reflect fundamental values remains unsettled. There is, however, overwhelming evidence that on average the initial stock price response to new information is at least in the correct direction. This means that the theory of efficient capital markets provides a useful framework for analyzing many problems.
Steven L. Jones is an associate professor of finance at Indiana University's Kelley School of Business. He was formerly a senior financial analyst at Amoco Corporation. Jeffry M. Netter is a finance professor and an adjunct law professor at the University of Georgia. He was formerly a senior financial economist with the U.S. Securities and Exchange Commission.
Ball, Ray. "What Do We Know about Stock Market Efficiency?" In A Reappraisal of the Efficiency of Financial Markets, edited by Rui Guimaraes. 1989.
Brealey, Richard, and Stewart Myers. Principles of Corporate Finance, 4th ed., 287-314. 1991.
Elton, Edwin, and Martin Gruber. Modern Portfolio Theory and Investment Analysis, 399-448. 1991.
Fama, Eugene. Foundations of Finance. 1976.
Fama, Eugene. "Efficient Capital Markets II." Journal of Finance (1991): 1575-1617.
Fama, Eugene., and Kenneth French. "Dividend Yields and Expected Stock Returns." Journal of Financial Economics 22 (1988): 3-25.
Gaines, Sally. "Founding Fathers, First Inside Traders," Chicago Tribune, May 17, 1987, 7-1.
Grossman, Sanford. The Informational Role of Prices. 1989.
Mitchell, Mark, and Jeffry Netter. "Triggering the 1987 Stock Market Crash: Antitakeover Provisions in the Proposed House Ways and Means Tax Bill?" Journal of Financial Economics 24 (1989): 37-68.
Patell, James M., and Mark Wolfson. "The Intraday Speed of Adjustment of Stock Prices to Earnings and Dividend Announcements." Journal of Financial Economics 13 (1984): 223-52.
Poterba, James, and Lawrence Summers. "Mean Reversion in Stock Prices: Evidence and Implications." Journal of Financial Economics 22 (1988): 27-59.
Shiller, Robert. Market Volatility. 1989.
Summers, Lawrence. "Does the Stock Market Rationally Reflect Fundamental Values?" Journal of Finance 41: 591-601.