3 The Efficient Markets Hypothesis (EMH)
The classic statements of the Efficient Markets Hypothesis (or EMH for short) are to be found in Roberts (1967) and Fama (1970).
An ‘efficient’ market is defined as a market where there are large numbers of rational, profit ‘maximisers’ actively competing, with each trying to predict future market values of individual securities, and where important current information is almost freely available to all participants. In an efficient market, competition among the many intelligent participants leads to a situation where, at any point in time, actual prices of individual securities already reflect the effects of information based both on events that have already occurred and on events which, as of now, the market expects to take place in the future. In other words, in an efficient market at any point in time the actual price of a security will be a good estimate of its intrinsic value.
Fama identified three distinct levels (or ‘strengths’) at which a market might actually be efficient.
In its strongest form, the EMH says a market is efficient if all information relevant to the value of a share, whether or not generally available to existing or potential investors, is quickly and accurately reflected in the market price. For example, if the current market price is lower than the value justified by some piece of privately held information, the holders of that information will exploit the pricing anomaly by buying the shares. They will continue doing so until this excess demand for the shares has driven the price up to the level supported by their private information. At this point they will have no incentive to continue buying, so they will withdraw from the market and the price will stabilise at this new equilibrium level. This is called the strong form of the EMH. It is the most satisfying and compelling form of EMH in a theoretical sense, but it suffers from one big drawback in practice. It is difficult to confirm empirically, as the necessary research would be unlikely to win the cooperation of the relevant section of the financial community – insider dealers.
In a slightly less rigorous form, the EMH says a market is efficient if all relevant publicly available information is quickly reflected in the market price. This is called the semi-strong form of the EMH. If the strong form is theoretically the most compelling, then the semi-strong form perhaps appeals most to our common sense. It says that the market will quickly digest the publication of relevant new information by moving the price to a new equilibrium level that reflects the change in supply and demand caused by the emergence of that information. What it may lack in intellectual rigour, the semi-strong form of EMH certainly gains in empirical strength, as it is less difficult to test than the strong form.
One problem with the semi-strong form lies with the identification of ‘relevant publicly available information’. Neat as the phrase might sound, the reality is less clear-cut, because information does not arrive with a convenient label saying which shares it does and does not affect. Does the definition of ‘new information’ include ‘making a connection for the first time’ between two pieces of already available public information?
In its third and least rigorous form (known as the weak form), the EMH confines itself to just one subset of public information, namely historical information about the share price itself. The argument runs as follows. ‘New’ information must by definition be unrelated to previous information, otherwise it would not be new. It follows from this that every movement in the share price in response to new information cannot be predicted from the last movement or price, and the development of the price assumes the characteristics of the random walk. In other words, the future price cannot be predicted from a study of historic prices.
Each of the three forms of EMH has different consequences in the context of the search for excess returns, that is, for returns in excess of what is justified by the risks incurred in holding particular investments.
If a market is weak-form efficient, there is no correlation between successive prices, so that excess returns cannot consistently be achieved through the study of past price movements. This kind of study is called technical or chart analysis, because it is based on the study of past price patterns without regard to any further background information.
If a market is semi-strong efficient, the current market price is the best available unbiased predictor of a fair price, having regard to all publicly available information about the risk and return of an investment. The study of any public information (and not just past prices) cannot yield consistent excess returns. This is a somewhat more controversial conclusion than that of the weak-form EMH, because it means that fundamental analysis – the systematic study of companies, sectors and the economy at large – cannot produce consistently higher returns than are justified by the risks involved. Such a finding calls into question the relevance and value of a large sector of the financial services industry, namely investment research and analysis.
If a market is strong-form efficient, the current market price is the best available unbiased predictor of a fair price, having regard to all relevant information, whether the information is in the public domain or not. As we have seen, this implies that excess returns cannot consistently be achieved even by trading on inside information. This does prompt the interesting observation that somebody must be the first to trade on the inside information and hence make an excess return. Attractive as this line of reasoning may be in theory, it is unfortunately well-nigh impossible to test it in practice with any degree of academic rigour.
Critics of EMH
For about ten years after publication of Fama's classic exposition in 1970, the Efficient Markets Hypothesis dominated the academic and business scene. A steady stream of studies and articles, both theoretical and empirical in approach, almost unanimously tended to back up the findings of EMH. As Jensen (1978) wrote: ‘There is no other proposition in economics which has more solid empirical evidence supporting it than the EMH.’
However, as Shleifer (2000) put it, ‘strong statements portend reversals’ – and in the two decades following Jensen's statement, a growing volume of theoretical and empirical work either contradicted the EMH outright or sought at least to show that its case was ‘not proven’.
Critics of EMH have produced a wide range of arguments, of which the following is a summary.
The assumption that investors are rational and therefore value investments rationally – that is, by calculating the net present values of future cash flows, appropriately discounted for risk – is not supported by the evidence, which shows rather that investors are affected by:
a tendency to ‘churn’ their portfolios
a tendency to under-react or over-react to news (Sheifer, 2000; Barber and Odean, 2000)
asymmetrical judgements about the causes of previous profits and losses.
Furthermore, many alleged anomalies have been detected in patterns of historical share prices. The best known of these are the ‘small firm’ effect, the January effect and the mean reversion.
The ‘small firm’ effect. Banz (1981), in a major study of long-term returns on US shares, was the first to systematically document what had been known anecdotally for some years – namely, that shares in companies with small market capitalisations (‘small caps’) tended to deliver higher returns than those of larger companies. Banz's work was followed by a series of broadly corroborative studies in the US, the UK and elsewhere. Strangely enough, the last twenty years of the twentieth century saw a sharp reversal of this trend, so that over the century as a whole the ‘small cap’ effect was much less marked. Whatever the reason or reasons for this phenomenon, clearly there was a discernible pattern or trend that persisted for far too long to be readily explained as a temporary distortion within the general context of EMH.
The January effect. Following on from the ‘small firm’ effect, it was also observed that nearly all of the net outperformance by small cap stocks was achieved in successive Januarys. Again, this was a discernible trend that under EMH should have been arbitraged away. As one commentator rather acidly remarked, it was not as if the annual coming of January could be characterised as entirely fresh news!
Mean reversion. This is the name given to the tendency of markets, sectors or individual shares following a period of sustained under-or out-performance to revert to a long-term average by means of a corresponding period of out- or under-performance. This was picked up in detailed research by De Bondt and Thaler (1985), who showed that, if for each year since 1933 a portfolio of ‘extreme winners’ (defined as the best-performing US shares over the past three years) was constructed, it would have shown poor returns over the following five years, while a portfolio of ‘extreme losers’ would have done very well over the same period.
And finally, a word of caution about the EMH debate
EMH states that an investor cannot make excess returns out of stale information. It is not difficult to define stale information, but the calculation of an excess return depends also on an accurate assessment of the risk associated with holding a share. Despite all the work done in this area since the 1960s, there is still no single, universally accepted or objectively verifiable measure of risk in the context of investment holdings. Supporters of EMH can plead with some justification that pricing anomalies may be more apparent than real, as they may be based on inaccurate measures of risk.