CHAPTER - 4 A SCAN ON EFFICIENT MARKET THEORY

Transcription

1 CHAPTER - 4 A SCAN ON EFFICIENT MARKET THEORY 4.1 INTRODUCTION 4.2 MARKET EFFICIENCY 4.3 THEORIES OF SHARE PRICE BEHAVIOUR 4.4 FUNDAMENTAL ANALYSIS 4.5 TECHNICAL ANALYSIS 4.6 EFFICIENT MARKET HYPOTHESIS 4.7 STOCK PRICE BEHAVIOUR IN EFFICIENT MARKETS 4.8 IMPLICATIONS OF EMH 4.9 IMPLICATIONS OF EMH FOR FUNDAMENTAL ANALYSIS 4.10 IMPLICATIONS OF EMH FOR TECHNICAL ANALYSIS SUMMARY

2 4.1 INTRODUCTION Before one starts the process of security analysis* it is important for investors to understand the market environment in which they compete. Behaviour of stock prices is an important indicator of market efficiency. The concept of market efficiency tells us about how effectively investor expectations are translated into security prices. That is, whether the expectations of the investor regarding the future cash flows for a particular stock quickly and accurately reflected in the price of the stock. The most important factor causing securities to be efficiently priced is competition. The more the competitors, the less likely the opportunities of abnormal profit. If the stock market is competitive, well regulated and consists of rational profit seeking investors, its price movements may be explained by the Efficient Market Hypothesis (EMH). Thus, this chapter gives a brief note on the concept of market efficiency and the EMH. 4.2 MARKET EFFICIENCY Market efficiency implies that all known information is immediately discounted by all investors and reflected in share prices in the stock market. In an ideal efficient market, everyone knows all possible to know information simultaneously, interprets it similarly and behaves rationally. But, in reality, it rarely happens.1 In an efficient market, securities are correctly priced so that they will provide a fair or normal level of return, given their risk characteristics. Liquid capital will channel quickly and accurately where it will do the community the most good. Efficient markets will provide ready financing for worthwhile business ventures and drain capital away from corporations, which are poorly managed or producing obsolete products. It is essential that a country has efficient capital markets if it is to enjoy highest possible level of wealth, welfare and education for population. One of the main reasons that some undeveloped countries do not advance is that they have inefficient capital markets.2

3 94 In inefficient capital markets, prices may be fixed or manipulated rather than determined by supply and demand. Capital may be controlled by a few wealthy people and not be fluid. Hence, it may riot flow where it is actually needed. Corruption arid public distrust can cause money to be hoarded rather invested in the capital market. Investors may be ignorant and unable to distinguish between worthwhile business ventures and bad investment.3 In an efficient market, all the relevant information is reflected in the current stock price. Information cannot be used to obtain excess return because the information has already been taken into account and absorbed in the prices. All prices are correctly stated, and there are no bargains in the stock market. Thus efficiency means the ability of the capital market to function so that prices of securities react rapidly to new information. Such efficiency will produce prices that are appropriate in terms of current knowledge, and investors will be less likely to make unwise investments. A corollary is that investors will also be less likely to discover great bargains and thereby earn extraordinary high rates of return.4 Thus in an efficient market, all the relevant information about a firm that is available to market participants is expected to be fully and immediately absorbed in its share prices. The investors try to assess all available information in order to predict the likely movement in share prices, so as to make profits. This implies that if investors recognise that changes in some macro-economic variable would have an impact on share prices, they would react whenever such information is available. Market efficiency assumes that securities are rightly priced and they reflect an unbiased estimate of the intrinsic value! Hence there is no under or over-priced security. All the securities are fairly valued. There is no possibility of extra normal return and the investors earn normal returns on their investments that are appropriate for the level of risk assumed. That is, in a perfect market, the securities are correctly priced. There is no under or over-priced stocks. So, the degree of market s efficiency

4 95 is important for an investor. If the market is perfectly efficient, then the money invested in securities are wasted. If a section of the market is. less efficient than the other, then efforts are taken to identify mis-priced securities.5 Those who believe that security markets are perfectly efficient agree that, to have market efficiency, there must be well-informed analysts who carry on the evaluation of available information on a particular stock. But in an efficient market these analysts compete away the chance of earning abnormal profit. Inefficiency in the market is concerned with the inability of the stock market to process information. Efficiency in this context means, prices in the stock market react rapidly to new information. The degree of market efficiency, thus have important information content for investors. Eugene Fama (1965) first studied this degree of market efficiency. He defines market as a fair game where prices fully reflect all information. He also noted that the hypothesis that securities prices at any point of time fully reflect all available information is obviously an extreme null hypothesis and like any other null hypothesis it is not expected to.be literally true.6 Thus the. requirements for a securities market to be efficient are: Prices must be efficient so that new inventions and better products will cause a firm s security prices to rise and motivate investors to buy its stock. Information must be discussed freely and quickly across the nations so that all investors can react to new information. Transaction costs such as sales commission on securities are ignored. Taxes are assumed to have noticeable effect on investment policy. Every investor is allowed to borrow or lend at the same rate. Investors must be rational and able to recognise efficient assets and that they will invest money in the assets with relatively high returns.

5 96 The fundamental idea behind the theory of efficient markets is that in a free _and competitive market, such as a stock market, the prices of financial assets should reflect all publicly available information and that these prices should adjust very rapidly to new information. The security market in which stock prices always fully reflect available information has been termed as "efficient" by its protagonists.7 The stock market efficiency may be defined as the ability of securities to reflect and incorporate all relevant information in their prices. An efficient stock market is a market where information about securities is available and rapidly reflected in the prices of these securities. This means that when stocks are traded, prices are accurate signals for capital allocation. Thus it plays a vital role in setting the price of its securities. If stock market is efficient, then the current price of a company s share is fair. There is no question of the share price being under or overvalued. This phenomenon of over or under-valuation of share is possible only in an efficient capital market.8 It must be understood, however, that.it requires some necessary conditions to be present for a market to be efficient. Properly anticipated prices vibrate randomly and that securities markets will assuredly be efficient under certain assumptions. These assumptions include: (i) There are no transaction costs in trading securities. (ii) All the market participants have free access to available information. (iii) All the participants have uniform opinion on the implications of current information for the current prices and distribution of future prices of all securities. (iv) No individual player in the market is big enough to influence the market price. (v) I identical time horizon and homogenous expectations of market participants regarding current and future prices.

6 97 However, such stringent conditions are not usually characteristic of the real world markets and that they are not necessary for a market to be efficient and for prices to fluctuate randomly. These conditions are sufficient for a market to be efficient but are not necessary. Even if there are transaction costs involved, provided these are taken into account and they inhibit the flow of transactions, it does not imply that all available information is not fully reflected in the prices. If sufficiently large number of participants have ready access to information, the market may be considered as efficient. As regards the homogenous expectations of all the investors are concerned, so long as investors can make better evaluations of available information and can beat the market consistently, it will imply the market is efficient. There are two aspects of market efficiency - the type of information that the market is reacting to, and the speed with which the market responds to that information. Since in an efficient market at a given point of time, stock prices are assumed to reflect all available information, one would expect the current price of a given security to be a good estimate of its intrinsic value owing to competition among market participants, In an uncertain world, however, the intrinsic value of a security cannot be precisely determined. Hence, one would expect differences of opinion among market participants regarding the value of each share. As a result, the actual prices wander randomly around the intrinsic value. The efficient market theorists, however, argue that competition among investors will ensure that these discrepancies are not too large to be profitably used.9 There are different concepts relating to efficiency like allocational efficiency, operational efficiency and pricing efficiency. A market is said to be allocationally efficient when prices are determined in a way that equates the marginal rates of return for all producers and savers. Funds flowing into the market are allocated where they can be most efficiently used. Source savings are optimally allocated to productive investments in a way that benefits everyone. Thus, in an allocationally efficient market funds are channelised to the most promising investment opportunities.

7 98 Operational efficiency, also called internal efficiency, relates to the cost and speed of operations in the market. Thus, in an operationally efficient market, operations are less costly and more rapid. By pricing efficiency, also called external efficiency, it is meant that information relevant for pricing securities is made public rapidly and theoretically with no cost, to reach all investors at the same time and will be rapidly reflected in the prices of securities in the market. This will at least theoretically assure a fair game for all investors.10 \ Allocational efficiency requires both pricing and operational efficiency. A necessary condition for optimal allocation of resources is that prices always provide accurate signals for investor choices. This will be the case when there is pricing efficiency and prices fully reflect all available information. Most of the studies to test stock market efficiency that has been reported in financial literature investigate pricing efficiency only. This is so because data are not readily available to test allocational or operational efficiency. The present study, too seeks to investigate pricing efficiency, also termed as the Efficient Market Hypothesis (EMH) in modern financial literature.11 Efficiency is associated, therefore, with the degree of market organisation: highly the markets are organised, the better the chances that they are efficient. The New York and other European financial markets can be taken as examples where most necessary condition for efficiency are present, such as, investors, security analysts, dealers, brokers, market makers, motivated and competent market observers as well as governing bodies and financial institutions. These parties in their competition for information enable the current prices to be at least close to the intuitive price and that the absence of perfect condition for efficiency in the real world provides potential sources of inefficiency. In case of developing and thin markets, there is usually little or no evidence of market efficiency. A common justification for the less frequent findings of market efficiency include less stringent information disclosure requirements, lower technical organisation of the smaller

8 99 markets, less information released by companies and less rigorous accounting information. Further, in small markets, it may be easier for large traders to manipulate the market as privileged information channels exits.12 In an efficient market, the firms can make production-investment decisions and investors can choose among the securities that represent ownership of firms activities under the assumption that security price at any time "fully reflect" all available information. Jensen (1970) stated that the stock prices reflect information to the point where the marginal benefits of acting upon the information do not exceed the marginal costs. Thus, in an efficient market, an investor should expect to make normal rate of return on investment. Market efficiency may be tested in terms of, (i) the information reflected in the market prices, and (ii) the accuracy and speed with which market prices adjust to the new information.13 It is very necessary to systematize the vast amount of publicly available information, both objective and subjective, into a valuable framework so that one can reach a buy, sell or hold decision. Over the years, academicians have been interested in developing and testing models of stock price behaviour. As a result, a number of theories have been formed to describe and predict share price behaviour. i.3 THEORIES OF SHARE PRICE BEHAVIOUR Share prices are affected by a large number of factors, some of which are economic in character while the others are of non-economic - ranging from wealth o war. In fact, information, moods, opinion and guesses as to the future all intermix o cause variation in prices of shares over time. However, the influence of each factor s so small that it would be extremely difficult to single out one factor as the principal determinant of share prices though cumulatively they give rise to the observed movements in share prices. In such a situation, it is exceedingly difficult to search for universally accepted theory of price formation and price changes. Thus, three ifferent schools of thought have emerged in the literature concerning security aluation and the behaviour of security prices.

9 100 (j) (ii) (iii) Fundamental or Intrinsic value Analysis. "Chartist" on Technical Analysis. Random walk or Efficient Market Hypothesis. 4.4 FUNDAMENTAL ANALYSIS In the fundamental approach, the security analyst or the prospective investor is primarily interested in analysing factors such as economic influences, industry factors, and pertinent company information such as company s income, growth, financial status, productive efficiency, product demand, earnings, dividends and management, in order to calculate an intrinsic value for the firms securities. He reaches an investment decision by comparing this value with the current market price of the security.14 Fundamentalists believe that share prices have a tendency to regress towards intrinsic values. At any point in time, an individual security has an intrinsic value, which should, be equal to the present value of the future stream of. income from that security, discounted at an appropriate risk-related rate of interest. The actual price of a security, is therefore, considered to be a function of a set of anticipated returns and anticipated capitalization rates corresponding to future time period. Price of a share changes as anticipations change, which in turn changes as a result of new information. The fundamentalists, thus, argue that in case there is something less than complete dissemination of opinion, the actual price of a security is generally away from its theoretical value. That is, they believe that the market can very often go wrong in appraising the value of a company.15 The fundamentalists attempt to estimate the real worth of a security by considering key economic and financial variables and then decide as to what investment action is called for in a given situation depending upon whether the actual price is above or below its intrinsic value. If the market price of a security is below its intrinsic value i.e. under priced, it should be purchased. When the prevailing price

10 101 is above the intrinsic value i.e. over priced, a selling decision is called for. Therefore, at any time, the security analysts needs to identify the overpriced and under priced securities. He then, replaces overpriced securities with the under priced securities in the portfolio. This will result in improving the performance of the portfolio.16 Thus, in fundamental analysis, the intrinsic value of a security is the true economic worth of a financial asset dependent on earning prospects of the firm in light of anticipated economic conditions. It is unaffected by temporary variations in security prices. Methodology Intrinsic value estimates are determined by capitalising earnings and dividends and is equal to expected EPS (earnings multiplex) and expected DPS (dividends multiplier) respectively. Growth estimations, which affect the multipliers, are made on the basis of the indicators of growth such as sales, total accounting profits, cash flows, dividends and earnings per share. Earnings per share should be the best measure of the earning power behind each share and should be the major factor in determining growth in intrinsic values per share.17 In the evaluation of firm s earnings growth rate and its capitalization rate, there are many subjective considerations, which are difficult to evaluate quantitatively, two of the most important of which are firm s management and its financial situation. Only experience can enable the fundamental analyst to judge the firm s management. Some judgemental criteria can be experienced of management, age, management s ability to react effectively to change in firm s risk, R & D programmes etc. As for assessing the firm s financial situation, fundamental analyst make use of working capital and income ratios and compare them with firm s own historical ratios, and published industry average ratios. Obviously, subjective elements creep in such calculations. It is important for fundamental analysts to predict the course of the national economy and this is done, by predicting GNP through leading

11 102 economic indicators. They must also be able to forecast economic changes with emphasis on its timing and direction. This involves assumption about international, national and industrial conditions.18 Regardless how sophisticated the technique used, a complete painstaking fundamental analysis based on relevant facts, is a logical way to estimate the true value of a growing concern. However, it must be remembered that even under ideal conditions, fundamental analysis can suggest only a range of prices rather than a specific value. 4.5 TECHNICAL ANALYSIS Technical analysis involves studying the history of share price with an objective of tracking the future behaviour of a share or a group of shares. Technical analysts or charists, as they are commonly called, believe that by observing and studying the past behaviour patterns of given stocks, they can predict the future price movements in the security. Thus the future share prices may be well projected if we understand the past history of share price behaviour.19 Technical analysis is based on the premise that security prices depend on supply of and demand for securities in the market and has no relationship with intrinsic value, as the fundamentalists believe it to be. A technician believes that supply and. demand for shares in the market governs the price behaviour and the share price fluctuates according to the whims and fancies of the crowd and successful investing relies upon guessing how the investing public is going to behave in the future and trying to flow the trend today.20 The basic assumption of technical theorists is that history tends to repeat itself. In other words, past patterns of market behaviour will recur in the future and can thus be used for predictive purposes. In statistical terminology, the stock market technician relies upon the dependence of successive price changes.

12 103 Technical analysis comprises many different subjective approaches, but all have one thing in common - a belief that past patterns of market action will recur in the future and thus can be used in predicting future movements of security price. The technical approach to investment is essentially a reflection of the idea that the stock market moves in trends, which are determined by changing attitudes of investors to a variety of economic, monetary, political and psychological forces. The art of technical analysis is to identify changes in such trends at an early stage and to maintain an investment posture until a reversal trend is indicated. By studying the nature of previous market turning points, it is possible to develop some characteristics, which can help identify major market tops and bottoms. Technical analysis is therefore based on the assumption that people will continue to make the same mistake that they made in the past. According to the technicians, all data of importance are internal to the stock market. To them, it is an exercise to evaluate accurately a host of detailed information as the fundamentalists attempt to do. They choose not to engage themselves in this type of activity, but rather to allow others to do it for them. They believe that diverse opinions held by analysts and investors, who are supposed to evaluate these factors, will be reflected in the price of shares and the volume traded. They claim that their analysis of price and volume incorporates one important factor, which is not in the domain of fundamental analysis, namely the investor psychology. Since technical analysis involves an attempt to measure the psychology of the people, Malkiel (1975) coined the term "Castle-in-the-air" theory in order to describe the propensity to built castles in the air. According to him technical analysis is analogous to newspaper beauty judging contest in which you have to select the six prettiest faces out of 100 photographs. The prize goes to the person whose section most nearly confirm to those of the group as a whole. The smart players recognise that their personal criteria of beauty are irrelevant in determining the contest winner. A better strategy is to select those faces the other players are likely to fancy. The beauty contest analogy represents the form of Castle-in-the-air theory of public determination.21

13 104 But the technicians find it difficult to answer one question, i.e., why is Chartism supposed to work? The explanations as to why Chartism is supposed to work, if at all it really works, are - there may be a trend in the price movement for either of the two reasons. Firstly, the crowd instinct of mass psychology makes it so, when investors see the price of a speculative favourite going higher and higher, they want to jump on to the bandwagon and join the rise. Secondly, there may be unequal access to fundamental information about a company when some favourable pieces of news come to the market. This leads to the professionals getting the information first and when they react to the information, price rises. The common investors get the information a bit late and buy, pushing the price still higher. The technicians, observing the price movements pick up the clue and get into action before the common investors and there-by make a profit.22 Arguments against the validity of technical analysis point out that buying shares based on charts takes place only after price trends have been established and selling only after trends have been broken. Thus, the technician must (1) identify the trend, and (2) recognize when one trend comes to an end and prices start in the opposite direction. The main problem is to distinguish between reversals within a trend and real changes in the trend itself. Since sharp reversals in market may occur quite suddenly, the chartist will often miss the boat. Moreover, these techniques, besides being unscientific, are self-destructive. As more and more people start using it, any pattern will be ironed out.23 Methodology The main body of technical analysis is based on charts and graphs. The technician views price changes and their significance mainly through price and volume statistics. They use an array of methods to assess the market conditions. These include Dow theory, variety of moving averages methods, high-low-close-open analysis, odd lot theory, relative strength measures, candle stick analysis, methods based on Fibonacci series, advance - decline ratios, and numerous other methods.

14 105 These methods appear obscure to the layman and the judgement is extremely important both in extracting the meaning of the information from patterns as well as the choice of the method to be used. The development and usage of hundreds of indicators, patterns, etc, make the job of interpretation difficult. There is likelihood that different conclusions will be reached depending on the indicator used, i.e. one indicator may predict the continuation of the present trend, whereas another may predict the reversal of the present trend. Technicians in India, however, don t use all these tools except the moving averages. The method of Moving Average Convergence Divergence (MACD) seems to be popular in India, particularly on the Bombay Stock Exchange.24 Dow Theory The Dow theory, proposed by Charles Dow, at the turn of the twentieth century is one of the oldest technical, methods still widely followed. There are many versions of the theory, but essentially it consists of three types of market movements - the major market trend, which can often last a year or more; a secondary intermediate trend, which can move against the primary trend for one to several months; and minor movements lasting only for hours to a few days. The major trend or the primary trend is the long-range cycle that carries the entire market up or down. The secondary trend acts as a restraining force on the primary trend, tending to correct deviations from its general boundaries. The minor trends have little analytic value because of its short duration and variations in amplitude.25 Dow theorists use three basic types of charts, i.e., line charts, bar charts and point charts. Line charts are used to connect successive days prices Bar charts have vertical bars representing each day s price movements. The bar spans the distance from the day s highest price to the day s lowest price. A small cross on the bar marks the closing price. Point and figure charts (PFCs) are made of X s and O s and is more complex than line and bar charts. In essence, only significant changes are posted to a PFC. Thus for high priced securities, three or five point price changes are

15 106 posted and for low priced securities only one point changes are posted. PFCs do not have a time dimension unlike line and bar charts since determining the days become, sometimes impossible because eaich column represents a significant reversal instead of a trading day.26 A rationale for the use of various technical theories results from the fact that market agents are unable to learn the true model characterising stock price behaviour. When the ability of market agents falls short of what is required to competently analyse and process information, given the degree of complexity inherent in a particular decision problem, they tend to fall back on relatively uncomplicated rule of thumb. This lack of ability is referred to as a competence - Difficulty (C-D) gap. In responses, to a C-D gap, the market agents restrict their behaviour to only a limited repertoire of actions. When the operators in the stock market are unable to learn the true model governing the stock prices, and consequently are incapable of processing information optimally, they will develop rule-based strategies. The rulebased methods followed by technicians are examples of this phenomenon.27 Being highly subjective in nature, the technical theories come under severe criticism from academicians, particularly by the efficient market theorists who hold the view that past price patterns could not be meaningfully used to predict the future prices. To quote Malkiel On close examination, technicians are often seen with holes in their shoes and frayed shirt collars. I personally have never known a successful technician, but I have seen the wrecks of several unsuccessful ones. Curiously, however the broken technician is never apologetic about his method EFFICIENT MARKET HYPOTHESIS A French Mathematician, Louis Bachelier in 1900, first proposed the Efficient Market Hypothesis. The empirical research conducted by Bachelier (1900) to study the share price behaviour in the beginning of 20th century showed randomness in share price behaviour. Several other empirical studies were undertaken after a gap of

16 107 five decades that further supported the randomness but failed to provide any iheoretical framework to explain the behaviour. The further researches, however, formalised the EMH. This hypothesis, also known as Random Walk Hypotheses, is n sharp contrast with both, the technical and fundamental schools of thought for iescribing and predicting share price behaviour. The empirical evidence of this theory n the random - walk literature existed even before the theory was established. That s to say, empirical results were discovered first, and then an attempt was made to ievelop a theory. that could possibly explain the results. After these initial jccurrences, more results and more theories were uncovered. This has led to a liversity of theories, which are generically called the random - walk theory.29 The EMH states that current stock prices fully reflect all available information. -Iowever it has no empirically testable implications. Most of the researches done in his area were under the assumption that the conditions of market equilibrium can be ;tated in term of expected returns.30 A number of empirical studies have been conducted on different set of data or varying time periods to test the appropriateness of the model for describing stock >rice behaviour. Being a purely empirical phenomenon in the beginning, it lacked conomic content although the outcome was mathematically predictable. Later, when ests showed that changes in stock prices were largely random, attempts were made o clothe the empirical results with economic content, which led to the development if the theory of efficient capital markets. The EMH is concerned with the question of whether a series of historical rices or rates of return can help in predicting future stock prices or rates of return, 'his hypothesis refers to pricing efficiency of stock markets. It states that in an fficient market, securities are correctly priced and the prices fully reflect all vailable information. Fama has expressed the above definition of efficiency in terms f a fair game model according to which if security prices fully reflect all available

17 108 information in an efficient market, then any trading system based on the information already impounded in prices will be a fair game. No expected speculative gain is therefore, possible by trading in an efficient market.31 I Fama specified the process of price formation as (I) Where E = Expected value operator Pj[ = Price of security j at the end of the period t Rjt t+1 = One period return (Py.f+1 - P.) f Pj( on security j during periods t+1. Qt = the set of information that is assumed to be fully reflected in the price at time t. The tildes (~) indicate that P;-,+I and /?./+1 are random variables at time t. The value of the equilibrium expected return («+I Qt) estimated on the basis of the information set Qc is determined from the particular expected return model used. The conditional expectation notation (I) implies that whatever expected return model is assumed to apply the information set Qt is fully utilized in determining equilibrium expected returns. Fama s assumption that the conditions of market equilibrium stated in terms of expected returns are formed on the basis of the information set Qt and have a major empirical implication i.e., they rule out the possibility of trading systems based only on the information set Q, and have expected profits or returns in excess of equilibrium expected profits on returns. (II) Then, (Z.f+1 <2,)=0 (III)

18 109 In other words, the sequence (Zjt) is a fair game with respect to the 'ormation sequence (Qt). In economic terms, Z} (t+1) is the excess return minus the Terence between observed return of security j at time t + l and the equilibrium pected return of security j that was projected at time t on the basis of the ormation set Qt. Under the fair game model, the expected value of the excess urn would be zero. However, Fama notes that the hypothesis that security prices my point of time fully reflect all available information is obviously an extreme null pothesis. And like any other extreme null hypothesis, it is not expected to be Tally true.32 The fundamental ideas behind the EMH are that successive price changes in individual security are independent over time and that its actual price fluctuates domly about its intrinsic or theoretical value. Moreover, price changes occur hout any discernible trends or patterns and that past prices contain no useful >rmation as to their future price behaviour. Thus the EMH or the theory of random ks implies that a series of stock price changes has no memory i.e., the past history he series cannot be used to predict the future in any meaningful way. This means at a given point of time the size and direction of the next price change is random l respect to the information available at that point of time and that the best lictor of tomorrow s price is today s price. The past history of stock price 'ements, and the history of stock trading volume, do not contain any information will allow the investor to do consistently better than a buy -and -hold strategy in aging a portfolio. This implies that past history of share prices will not enable an stor to obtain an above-average return.33 The strict form of the random walk model states that (a) successive price ges or changes in returns are independent, and (b) price changes or return ges are identically distributed according to some stationary distribution. The strict of the random walk model stated in terms of returns may be written as34 I rj,t rj,t-1 * rj,t-2 * " ) re f Oj [+1) = the.probability distribution of returns for security j at period t + l.

19 110 The above expression implies that the conditional and marginal probability distributions of an independent random variable are identical. In other words, above "equation says that the entire distribution of returns is independent of the preceding sequence of returns. However, the shape of the probability distribution is not specified for the general theory of random walks. It only assumes that price changes confirm to some given probability distribution. Any distribution is consistent with the theory as long as it correctly characterizes the process generating the price change. The knowledge of the. form of distribution of price changes is important for both investors and researchers for determining riskiness of investment in common stocks. But the model when stated in terms of predictability of price changes from earlier changes need not be specific about this distribution.35 Though the fandom walk theorists say that stock price movements are not strictly independent, they, argue that a small degree of dependence in successive price changes doesn t actually reject the practical utility of the theory of random walks. Thus, for the stock market, investor, the assumption of the random walk model is valid as long as knowledge of the past behaviour of the series of price changes cannot be used to increase expected gains. That is, for practical purposes, the model may be considered to be appropriate as long as the degree of dependence in a series of price changes is not sufficient to predict future price changes from past price changes so as to make expected profits greater than they, would be under a buy - and - hold policy.36 The random character of stock prices does not imply that the changes in the prices of common stocks just happen by chance or are uncaused, nor does it imply that the market is senseless or irrational in determination of stock prices. The explanation of the randomness of stock prices lies in understanding the market making mechanisms, the quest for which has led to the development of the theory of efficient markets.

20 Ill The definition of marketing efficiency is so general that it has no empirically testable implications. The information set relevant to a market may be general and wide and hence testing market efficiency on this Universal" set involves practical difficulties. Hence, researchers in this context decomposed the universal set into three sub-sets, viz (a) information on past price sequence (b) publicly available information and (c) information by monopolistic access or privately held information. Accordingly Fama categorized EMH into weak form, semi-strong form and strong form respectively. The distinction between these three forms of EMH is related to whether prices fully reflect particular sets of available information. The efficiency of the market is viewed as lying on a continuum, random walk on the one end and the strong form of market efficiency on the other end Weak Form of EMH The weak form of EMH asserts that current prices fully and instantaneously reflect all information implied by historical sequence of prices. Therefore, contrary to what is claimed by technical analysts, it is of no use to study the previous stock price behaviour. Such instantaneous price adjusting mechanism will not ensure consistent extra-normal return to market participants. In other words, own past price information will not be of any use in predicting own future price. This implies that the past price series are serially independent. The weak form of efficiency also says that stock prices have no memory and cannot be used as a basis of future forecast. Weak form of market efficiency is synonymous with random walk model. A random walk model is often compared with a drunkard s walk. Leaving the bar the drunkard moves a random distance U, at time t. If he or she continues to walk indefinitely, he or she will eventually drift further and further away from the bar. Thus, as per weak form of market efficiency stock prices would behave like random walk.38 The weak form can be symbolically represented as: j,t*\ I j,t-\ i > A,- A

21 112 R.{ is the return for security j at period t. In simpler terms, it states that knowledge of all historical returns of a security suggest only that next period s return is expected to equal last period s return.39 In the weak form of EMH, the, type of information being considered is restricted to only historical prices. If the weak form of EMH is correct, then investors should not be able to consistently earn abnormal profits by simply observing the historical prices of securities. Thus, the knowledge of historical prices and returns can t enable investors to make excess returns. Technical analysis, which relies on charts of stock prices over time, is particularly vulnerable to the weak from of EMH, as are techniques, which rely on moving average of prices. Fama (1991) considered the weak form as " Test for Return Predictability" which also includes works on forecasting returns with variables like dividend yields and interest rates SEMI - Strong Form of EMH The semi-strong form of EMH asserts that security prices adjust rapidly and correctly to the release of all publicly available information Thus, under the semistrong form, current prices fully reflect not only all past price data but also information about the corporation such as earnings reports, dividend announcements, annual and quarterly reports and news items in the financial press. The efforts by analysts and investors to acquire and analyse public information will not yield consistently superior returns to the analyst.41 Thus, in semi-strong form of EMH, any information that is available to the public is quickly reflected in security prices and the investors are unable to earn abnormal returns consistently by acting on such public information. Symbolically E (Rj t+1 { Nt) = Rjt N, denotes all the publicly available information at period t. This equation implies that all publicly available information is already reflected in prices.42

22 113 The semi-strong form of EMH maintains that as soon as the information becomes publicly available, it is absorbed and reflected in stock prices. Even if this adjustment is not the correct one immediately, it will in a very short time be properly analysed by/the market. Thus, the analyst would have great difficulty trying to make profit using, fundamental analysis. Furthermore, even while the correct adjustment is taking place, the analyst cannot obtain consistent superior returns because the adjustments will not take place consistently. That is, sometimes, the adjustments will be over adjustments and sometimes, they will be under adjustments. Therefore, an analyst will not be able to develop a trading strategy based on these quick adjustments to new publicly available information Strong Form of EMH While the semi-strong form of efficiency deals only with publicly known information, strong form of efficiency deals with all information. The storm form of EMH represents the most extreme case of market efficiency. It asserts that security prices fully reflect all available information including both public and private or inside information. Private information is defined as that information to which the market would react if it were available to the public. Any investor or groups such as mutual funds and institutional investors who may have monopolistic access to any information relevant for price decisions will not be in advantageous position by virtue of possessing such information since it is already reflected in the current prices. Thus, the strong form of EMH maintains that the publicly available information is useless to the investor or analyst but all information is useless. No information that is available whether it is public or private can be used to earn consistently superior investment returns. However, being an extreme hypothesis, it is not expected to be true.43 It is noteworthy that both the semi-strong and strong forms of the EMH cannot be tested directly. One can do so indirectly by accumulating evidence, which contradicts this hypothesis. Thus, examining whether share prices react correctly to

23 114 new information that become available can test the semi-strong form of EMH. Likewise, the strong form of EMH can be tested by determining whether any investor, appear to have gained and used, superior information and whether any investors have earned above-average returns.. ' / The above categorization of EMH serves a useful purpose. It allows the researcher to pinpoint the level of information at which the hypothesis breaks down, 4.7 STOCK PRICE BEHAVIOUR IN EFFICIENT MARKETS The efficient market hypothesis implies a particular way in which stock prices fluctuate. In efficient markets, share prices follow a random walk. In such a market, all available information has already been built into the price of a share. It is the arrival of new information relevant to the intrinsic value of a security that would affect its price. Since, stock prices move in response to relevant but erratic events, stock prices themselves move erratically, like a random walk. Hence, a test of the EMH may be conducted by examining the behaviour of stock prices - whether they follow a random walk on not. The random walk model is, in fact, a more rigorous and narrowly defined case of a martingale model used in the weak form EMH. Any evidence supporting the random walk will provide a robust proof of the weak form of EMH.44 However, any evidence supporting randomness in stock price fluctuations doesn t provide a proof that the fluctuations are around the intrinsic value of the security. They may be irrational without any relationship to intrinsic value. Hence, a test of stock market efficiency must involve an investigation into stock price reaction to specific value relevant information. Such investigation would also provide evidence on the semi-strong form EMH that states that stock price adjustment to new relevant information to the public would be instantaneous and accurate.

24 IMPLICATIONS OF EMH There is a great deal of controversy over the efficient market theory. Statisticians, on one hand, provide evidence in favour of the theory while the economists and financial analysts, on the other hand, do not believe in the cqrrectness of the theory. Examining the evidence relating to stock market efficiency, implications in three different areas for share market investment strategies can be observed. The most general implication of the EMH is that security analysis is logically incomplete and valueless. An analyst s recommendations to buy or sell must be predicted on a significant difference between the analyst s views and those of other investors whose opinions have established the stock s current market price. The second implication of EMH is that analysis of securities costs the same, whatever may be the amount available for investment. Therefore security research might make sense for large financial institutions having crores of rupees to manage, while it would not make sense for investors with smaller sums. Another implication of the efficient market theory is that it is extremely unlikely that one can consistently earn superior rate of return by analysing public information in conventional ways. The only way to earn superior results is by seeking unique ways of forming expectations about the prospects for individual companies. 4.9 IMPLICATIONS OF EMH FOR FUNDAMENTAL ANALYSIS The relationship between EMH and fundamental analysis is not so straightforward. EMH talks about randomness of price changes consequently return changes, since it implies that short-run price changes are random about the true intrinsic value of the security. Fundamentalists operating in a random walk world must be truly exceptional so. that they can seize upon opportunities when security prices differ somewhat significantly from their intrinsic value. This means that the fundamentalist can be successful only in those instances when he either possesses superior insight into the company s future prospects or possesses inside information. It is clear that, even under random walk, such superior fundamental analysis will lead to superior profits for the astute security analyst or investor who projects his own data and doesn t merely rely on already publicly available historical data.

25 IMPLICATIONS OF EMH FOR TECHNICAL ANALYSIS The EMH is inconsistent with technical analysis. While the EMH states that successive price changes are independent, the technical analysts claim that they are dependent i.e. the historical price behaviour of a stock will repeat itself into the future and that by studying this past behaviour, it is impossible to predict the future. EMH or the Random walk theorists however opposes this line of reasoning. They believe that the direction and magnitude of stock price movements cannot be forecasted accurately on a consistent basis to attain superior gains except due to the chance element. They contend that the market prices adjust instantaneously to any information whether it is past price and volume data, publicly available information or privileged insider information. The technicians however deny the findings of researches in this area. They have insisted that the evidence for the EMH doesn t imply that their technical theories are invalid and argued that the statistical procedures employed in the literature of the past were too simple to detect complex, historical price relationships. This means that the trading rules used in technical analysis are so complex that statistical testing cannot be thought of as a valid method of testing. The technical analysts also state that statistical tests assuming a linear relationship are inappropriate for commenting upon the validity of technical analysis as many technical theories can look for non-linear patterns in price data. Hence, the dependencies as seen by technical theories are not the same dependencies that are to be detected by statistical tests. Moreover, different technical analysts often apply same :echnical theories differently. Therefore, the success of the strategy is often dependent m the man rather than the method alone. Thus, merely by using statistical tests, the validity or invalidity of technical theorists cannot be tested.. The many years of statistical testing of the EMH or random walk using 'arious tools with varying power for data on various countries has led to mixed onclusions, both for and against the hypothesis. The EMH s contentions are neither enied nor confirmed conclusively by existing evidence. The conclusions of research tudies regarding the EMH was considered to have serious implications for the