Weak Form Market Efficiency - Essay Example

Running Head: Weak Test the Weak Form Market Efficiency Hypothesis for a given country's stock index --------------------------- ------------------- Test the Weak Form Market Efficiency Hypothesis for a given country's stock index Introduction Information that is contained in the stock market data is considered by a few to be able to guide investors and analysts about the future course the stock markets might take in terms of predicting equity prices and returns. A lot of technical analysis is based upon this belief. However finance theory assumes idealistic models for the stock markets and formulates the investor utility functions and expectations accordingly. These models are based on perfect competition and passage of information in an unfettered manner. As Wikipedia (2007) seems to point out, "In economics and financial theory, analysts use random walk techniques to model behavior of asset prices, in particular share prices on stock markets, currency exchange rates and commodity prices. This practice has its basis in the presumption that investors act rationally and without bias, and that at any moment they estimate the value of an asset based on future expectations. Under these conditions, all existing information affects the price, which changes only when new information comes out. By definition, new information appears randomly and influences the asset price randomly. Empirical studies have demonstrated that prices do not completely follow random walk. Low serial correlations (around 0.05) exist in the short term; and slightly stronger correlations over the longer term. Their sign and the strength depend on a variety of factors, but transaction costs and bid-ask spreads generally make it impossible to earn excess returns. Researchers have found that some of the biggest prices deviations from random walk result from seasonal and temporal patterns. In particular, returns in January significantly exceed those in other months (January effect) and on Mondays stock prices go down more than on any other day. Observers have noted these effects in many different markets for more than half a century, but without succeeding in giving a completely satisfactory explanation for their persistence. Technical analysis uses most of the anomalies to extract information on future price movements from historical data. But some economists, for example Eugene Fama, argue that most of these patterns occur accidentally, rather than as a result of irrational or inefficient behavior of investors: the huge amount of data available to researchers for analysis allegedly causes the fluctuations. Another school of thought, behavioral finance, attributes non-randomness to investors' cognitive and emotional biases". Taking an apposite viewpoint Leverton () states, "Without market fundamentals being able to predict prices, the investor is forced to learn new ways of investing.. Ratios and trend analysis are important to picking a winning portfolio. Subscribers to the adaptive expectations theory believe investors are backward looking in deciding on the correct price to pay for a stock". Realized and expected rreturns from the stock markets have been the subject of intense debate since a long period of time .Several theories suggesting various constructs and factors responsible for determining the returns from the stocks have been postulated thus far.It was not until the late 1960s and early 1970s that a fully-developed, empirically-supported theory of share prices' behavior emerged in the form of the Efficient Markets Hypothesis (EMH).Prior to the development of the EMH , analysts assumed some degree of dependence across successful price changes. Very many efforts were made towards identifying a predictable trading pattern which could be used for chasing profitable deals. From the mid-1950s to the early 1980s, a random walk theory (RWT) of share prices was developed based on the past empirical evidence of randomness in share price movements. RWT basically stated that speculative price changes were independent and identically distributed, so that the past price data had no predictive power for future share price movements. RWT also stated that the distribution of price changes from transaction to transaction had finite variance. In addition, if transactions were fairly uniformly spread across time and were large in numbers, then the Central Limit Theorem suggested that the price changes would be normally distributed. Kendall (1953) calculated the first differences of twenty-two different speculative price series at weekly intervals from 486 to 2,387 terms. He concluded that the random changes from one term to the next were large and obfuscated any systematic effect which may be present. In fact, he stated that 'the data behaved almost like a wandering series' (random walk). Specifically, an analysis of share price movement revealed little serial correlation, with the conclusion that there was very little predictability of movements in share prices for a week ahead without extraneous information. In 1959, Roberts generated a pattern of market levels and changes akin to real levels and changes in the Dow Jones Industrial Index. He estimated the probability of different share price movements over time by using a frequency distribution of historical changes in the weekly market index, and assumed weekly changes were independently drawn from a normal distribution with a mean of + 0.5 and a standard deviation of 5.0. He concluded that changes in security prices behaved as if they had been generated by a simple chance model .The fundamental concept behind random walk theory is that competition in perfect markets would remove excess economic profits, except from those parties who exercised some degree of market monopoly. This meant that a trader with specialized information about future events could profit from the monopolistic access to information, but that fundamental and technical analysts who rely on past information should not expect to have speculative gains. It was from the foundations of this empirical evidence and theory of random walks arose the theory of efficient markets. Fama (1970, 1976) gives out the details of the early literature on both the theoretical and empirical foundations of the Efficient Markets Hypothesis, whilst Cuthbertson (1996) summarizes the latest research developments. While EMH has empirical findings in respect of aspects like market perfection and information availability when combined with practices like trading platform and transaction costs may produce only marginal and well calculated opportunities for speculative gains; many other economists have quoted the existence of stock market bubbles. A bubble is generally defined by the economists as a deviation from stock market fundamentals; whereas Kindleberger reckons a bubble as an upward price movement over an extended range that tends to implode (Kindleberger 1996). By the same analogy an extended negative bubble is a crash. The existence of such situations has immediate learning points for an investor. Theory is distinct from market practices and the distinction has to be recognized by all practitioners. Financial theory, taught in finance textbooks the globe over, normally exposes a student of finance to the concepts like the efficient market hypothesis and the economically rational individual. Bubbles and crashes seem to defy these two seminal concepts with an awkwardness equivalent to the awkwardness one would attach to those things on earth that defy gravity. Nevertheless such extreme stock market movements are a reality. Bubbles make investing decisions arduous as stock prices tend to deviate by substantial margins from their fundamental valuations. Investors relying on past company results and technical analysis are equally defeated in such situations as is the EMH.In fact, investors always act on the basis that they have an applicable construct to explain stock price movements and tend to input all available information collected under such constructs in their investment decisions (Poole 2000). Finance research has also held varying opinions on this issue. For instance, Biermann (1995) supports the idea that market prices are determined from backward looking investors than by those that indulge in predictions of all sorts. Others have, for example elaborated on the use of price to earnings ratios to determine excess market valuations. Some technical work has set to rest in a convincing manner the phenomenon of bubbles and bursts. For instance, Graham (1973) describes in details why markets fluctuate and how to deal with the violent fluctuations .Graham discusses five basic points to read into cross sectional view of market bubbles. Most of these points concern factors like growth and earnings and their impact on price movements and price levels. Graham (1973), in fact, provides a much better viewpoint on gauging market bubbles through an adaptive expectations model. Statement of the Efficient Market Hypothesis From the theory of random walks arose the theory of efficient markets. The Efficient Markets Hypothesis (EMH) states that current prices always 'fully reflect' available information, so that the only reason prices change between time t and time t+1 is the arrival of new information. The EMH requires that only two necessary conditions be met. First, the market must be aware of all available information .The type of information available is determined by the strength of the EMH being tested. In a Weak Form EMH, current prices entirely reflect all that can be known from the study of historical prices and trading volumes. If the Weak Form is valid, technical analysis becomes ineffective. Any information contained in past prices has been analyzed and acted on by the market, so that shares are neither under-valued nor over-valued. In a Semi- Strong EMH, current prices efficiently adjust to information that is publicly available. If this form of the hypothesis holds true, then fundamental as well as technical analysis is also ineffective because all publicly available information has been thoroughly analyzed, assessed and acted upon. Finally, in a Strong Form EMH, current prices fully reflect all information, and not just that information which is to be found in the historical trading pattern or available publicly. Thus, if the Strong Form holds true, any attempt to make profitable use of monopolistic access to information is useless because this information has already been incorporated into the market price of the share. The second necessary condition is that of the rational expectations element of the EMH, or informational efficiency. As Leverton () states, "With rational expectations, investors focus on the future. If a company has hired a top-notch management staff, then they should be profitable in the future. Economic agents predict future events that are not falsified by actual events. Investors will construct their opinions in such a way that on the average, they are correct. It means that real returns can be randomly higher or lesser than expected returns, but on average, unexpected returns must be zero". The main implication of the rational expectations is that no system of trading rules can have higher expected returns than the equilibrium expected returns derived by the market. In other words in a perfect market scenario of RWT and varying degrees of market perfection in 3 forms of hypotheses under EMH the prices determined are efficient and there is no incentive on the market to either change prices or move away from equilibrium prices until unless some new unabsorbed information/event comes about making it necessary to move towards a new equilibrium. Testing the weak EMH The efficient-market hypothesis says that there is no easy way to make money. Thus, when such an opportunity seems to present itself, we should be very skeptical. For example, in the case of short- versus long-term rates, and borrowing short-term versus long-term, there are different risks involved. For example, suppose that we need the money long-term but we borrow short-term. When the short-term note is due, we must somehow refinance. However; this may not be possible, or may only possible at a very high interest rate. In the case of Japanese versus United States interest rates, there is the risk that the Japanese yen - U.S. dollar exchange rate will change during the period of time for which we have invested. Empirical data can help test out a theory and the EMH is amenable to easy testing as the stock market generates not only substantial amount of data on stock prices but also a mind boggling magnitude of information that keeps on affecting stock markets almost 24 hours a day. In order to test the three forms of the EMH, in particular the weak hypothesis, this paper analyzed a cross sectional data set from the London Stock Exchange. In order for the weak form of hypothesis to hold any stock should not exhibit greater movement or variability than a bench mark market indices, which works as a proxy for the market returns. Since in weak form EMH stock prices reflect all the available information, therefore, specific movements and variability in any particular stock cannot be more than the movement and variability in the market proxy indices. This paper chose to analyse the stock prices of the UK retailer TESCO for the months of March and April 2007 and the FTSE 100 index for these two months. The data on TESCO stock prices was obtained from www.yahoo/Finance.com and the data on FTSE100 was obtained from the relative months' fact sheet published on the website of the London Stock Exchange. Both data sets were day close figures and excluded banking holidays/non trading days. A simple statistical test was conducted to analyse the movement and variability magnitudes in TESCO stock prices and the FTSE100 indices for the chosen period. The test comprised in calculating percentage change over the previous day's close ,both for the stock prices and the indices and then calculating the respective means and standard deviations of the two series of percentage changes. If weak form EMH was to hold true then the average and standard deviations of the two series of percentage changes were expected to be near in absolute values to each other or fall within a very narrow range. The analyzed data and the results are carried in Table 1 at Annexure A .The results clearly reflect that the TESCO stock had greater average movement as well as variability when compared to the market proxy indices- FTSE100. This average of percentage changes in the TESCO stock was 0.23 as compared to FTSE100 which had an average of just 0.14.Similarly the standard deviation of the TESCO percentage changes was as high as 1.06 whereas the same was a mere 0.82 for the FTSE100. It appeared that during this period he TESCO stock was subject to a process of discounting and assimilating some market information and that the same was not already discounted fully by the market and reflected in the TESCO stock prices-resulting in higher movements and variability in the TESCO stock. It may also be observed from the table 1 at Annexure A that the TESCO stock prices rose by 2% or more during the chosen period whereas the FTSE100 indices did so only on one occasion. Investor psychology is a slippery concept, more often than not used to explain price movements that the individual invoking it cannot personally explain. Even if it exists, is there any way to make money from it If investor psychology drives up the price one day, will it do so the next day also Or will the price drop to a 'true'level Almost no one can tell you beforehand what 'investor psychology' will do. Theories based on it have no content. Events in stock markets and related to stocks do mould the investor psychology. A close examination of the TESCO stock watch revealed that certain positive events and information about them had crept in the market very gradually. These events began with the interest of Warren Buffet in the TESCO stock and the information that Buffet was building positions in TESCO was out in the markets by early 2007 and in the months of March and April 2007 this information was being built into the stock price of the TESCO-giving enough justification for the above noted movements and variability in the TESCO stock in these months. Yahoo (2007) carried this story substantiating the above event linked behaviour of the TESCO stock, "Investors also bid up Tesco (TEO.TO - news), which gained 2.1% on the London Stock Exchange (LSE: LSE.L - news), after Buffett reported holding a 2.9% stake in the British retailer. That was more than double the stake that was disclosed in previous regulatory filings. The famed investor first bought shares in Tesco last year after the Cheshunt, England-based retailer said it would open "Fresh & Easy" convenience stores in several western U.S. states. The initial stores will be in Phoenix (Xetra: PHO.DE - news), Las Vegas, Los Angeles and San Diego, with Tesco planning to spend $2 billion through 2011 on the concept. The company already operates in 12 markets outside the U.K. Tesco shares also were boosted by a Mar. 1 upgrade by Morgan Stanley (NYSE: MS - news) , (COP), with analyst Nick Coulter arguing that Tesco's international operations are undervalued. Tesco also faces favorable prospects for its U.S. venture, according to Morgan Stanley. "What s not to be excited about One of the best global retailers is entering the largest consumer market in the world with a concept designed to fall between the existing food retail categories," Coulter wrote. Tesco and Posco are among the most recently known to make the list of Buffett's major holdings, or those with market value of more than $700 million. Of those companies identified in Buffett's annual letter for 2006 that had not been major holdings in 2005, most derived a large share of their business outside the U.S". It may be observed that March 1,2007 Morgan and Stanley upgrade of TESCO share was also in the process of being discounted by bidders and sellers in the TESCO stock. Validity The Efficient Markets Hypothesis (EMH) posits that present stock prices always 'fully reflect' available information, so that the only possibility or factor which makes for a price change of stock, between time t and time t+1, is the arrival of 'news' or unanticipated events. However this is assuming away a lot of inconvenient eality.In real stock market situations there are transactions costs, information is collected and analyzed by an expensive system of analysts and information is priced and carries a cost and the reactions to a piece of information vary across a wide range of investors. However the EMH is based on the assumptions of zero transaction costs, freely available information and an agreement among investors on the implications of information on the share price. As Fama (1970) states, these conditions do not hold in the real stock market situations. However the fact that these assumptions may not come true in real life situations does not affect the validity of the EMH. For instance, the market can still be efficient if a sufficiently large number of traders have access to the necessary information. This is quite true in active lobbies of traders. A deviation from this access pattern-often results in the notorious insider trading charges. Insider trading is a situation wherein select market operatives have access to sensitive information pertaining to a stock and they use it to reap speculative gains. This is often seen in situations where a market operator has built sufficient and controlling buys in the market of a takeover target on the sly. In such a scenario the market is not aware of the takeover by a highly rated corporate-whereas a select band of market operators are aware of it and build buy positions to take advantage of the future rise in stock prices. Thus, whilst these conditions are sufficient, they are not necessary. The EMH requires that only two necessary conditions be met. First, the market must be aware of all available information. Theoretically stated, this means that the information set used by the market in time t to determine the price of security at time t is equivalent to the true information set. The type of information contained in the information set is determined by the strength of the EMH being tested. The second necessary condition states that the market correctly uses the available information in assessing the expected return of the share in the future period. These aspects pertain to rational expectations and informational efficiency. However the EMH has been stated to contain a logical flaw. The American economists Grossman and Stiglitz explored the issue in 1980 .Their findings were published in an article "On the Impossibility of Informationally Efficient Markets" in the American Economic Review. Their point was that markets cannot be efficient by accident. If they are efficient, then it is because of information and research. These activities have costs, but there is no incentive to pay for this research unless it can be used to make higher returns. In a truly efficient (frictionless and costless) market all information would be incorporated in the market price eliminating any profit opportunities. As Leverton() states, "With both rational expectations and adaptive expectations, investors base the price of a stock off of some expected future profits discounted for the time value of money". The long run theories of adaptive expectations and rational expectations do offer sufficient explanation of the wild fluctuations often seen in stock prices as they dangerously defy their fundamental foundations and worth. Simply put, one can ask a question -if the stock market has been observed to have a long term average growth rate of 6% per year, then under what conditions do indices like S & P 500 exhibit stock prices rise many multiple times this long term average growth rate These staggering variations in price valuation are so astounding that traditional models, seeped in neat theory, have a difficulty explaining them (Poole 2000).A plausible explanation of such volatile market behaviour has been offered through the theoretical construct of herd mentality. Herd mentality essentially refers t the investor behaviour pattern wherein the customers tend to develop a core market sentiment and attract most part of the market in following this core sentiment like sheep in herds. In such situations naturally a downward or an upward swing gets accentuated and theorists keep opining that random walk seems to be drifting one way or the other. Thus if most investors followed what others did, the volatility in stock valuation might be explained. As Keynes said: A conventional valuation which is established as the outcome of the mass psychology of a large number of ignorant individuals is liable to change violently as the result of the sudden fluctuation of opinion due to factors which do not really make much difference in the prospective yield (Shiller 2000). As the confidence in the market increases, investors will drive the prices away from the fundamentals (Diba 1990). In short, while the early literature contained potent arguments in favour of market efficiency; the latter market developments and theoretical probing as questioned the neatness inherent in the EMH .This literature ,while acknowledging departures from the random walk with drift and explaining with relevant drift models of stock price movements, concluded that in economic terms these deviations were insignificant, so that the 'fair game' property of the EMH was not disturbed. Fama (1991) state that ," any test of asset pricing models runs into the joint-hypothesis problem. Thus we can never know whether the market is inefficient or the model is wrong. Obviously, the choice of model may influence the findingsand that. For US stocks the relation between Beta and expected returns is feeble even when Beta is the only explanatory variable. This is less so when the data is expanded to include bonds." Conclusion Discounting and leveraging information in stock markets has been eagerly debated. Weak form seems to be supported in literature. Analysts indicate that predictions based on such a formulation have presented money making opportunities. As Berko(2007) states, " Here's how it buries. Since 1950, there have been 35 times when the S&P's averages were in positive territory during the first five days of January. Following of those five up days, the S&P had solid gains for the remaining 11 months, which represents an accurate predictability ratio of nearly 88 percent. In fact, the average gain for those 35 years (which includes the five years when the S&P was up during the first five days but closed lower at year's end) was 13.7 percent. That's not unimpressive! That 10-year, 13.7 percent average return exceeds the performance of the Dow and S&P since the "lame duck" amendment in 1934 changed the political calendar. (Prior to 1934 the newly elected Senators and Representatives didn't take office 'til December of the following year, 13 months later except when a new President was inaugurated. So, defeated Congressmen remaining in office for all of the following sessions were known as "lame ducks"). So having a darn good handle on how the S&P will perform the remainder of the year can provide investors with a moneymaking opportunity. Now a down January seems to be a precursor of trouble, politically, militarily and economically". While there is wide acceptance that the stock market is weak form and strong form efficient, it has not been feasible or possible to produce specific theoretical constructs or models of the behaviour of share prices across time. Most theorists also agree that information affecting stock price movements is rather randomly generated and reaches the markets equally randomly-until unless vaulted and planned in insider trading situations or with some strategic intents. Given these features determining the efficiency level of the market leaves one with ,at best, a fundamental random model. The swings or drift characteristic is a little more difficult to explain; however, it just goes on to point the fact that values increase across time to reflect the return element that is, in turn, determined by the choice to invest in a risky asset. Works Cited Wikipedia.(2007).Retrieved on July 24, 2007 from http://en.wikipedia.org/wiki/Share_price. Leverton, Justin.(). Bubble Mania. The Park Place Economist Volume X. Kendall, M.G. (1953), The Analysis of Economic Time-Series - Part 1: Prices. Journal of Royal Statistical Society, vol.96, pp.11-25. Fama, E.F. (1970).Efficient Capital Markets: A Review of Theory and Empirical Work. The Journal of Finance. vol.25, no.2, pp.383-417. Fama, E.F. (1976). Foundations of Finance, New York: Basic Books. Cuthbertson, K. (1996) .Quantitative Financial Economics: Stocks, Bonds and Foreign Exchange, England: John Wiley and Sons. Kindleberger, Charles P. (1996).Manias, Panics, and Crashes. 4ed. New York: John Wiley and Sons, Inc. Bierman Jr, Harold.(1995).Bubbles, Theory, and Market Timing. Journal of Portfolio Management 22.1 : 54-61. Graham, Benjamin.(1973). The Intelligent Investor. 4 ed. New York: Harper & Row. Yahoo,Finance.(2007). Buffett Boosts Shares of Steelmaker, Grocer. Retrieved on May 22, 2007 from http://in.finance.yahoo.com/. Poole, William.(2000).Expectations. Twenty-second Henry Thorton Lecture. 28 November 2000. Shiller, Robert J.(2000). Irrational Exuberance. New York: Broadway Books. Diba, Behzad T.(1990).Bubbles and Stock-Price Volatility.. The Stock Market: Bubbles, Volatility, and Chaos. Norwel, MA: Kluwer Academic Publishers. Eugene Fama.(1991).Efficient Capital Markets II.Journal of Finance. December 1991. Berko,Malcolm.(2007). These theories relatively useless. Retrieved on July 24, 2007 from https://www.copleynews.com . Annexure A Table 1: Analysis of the TESCO stock Prices and FTSE100 indices for March/April 2007 Trading Date FTSE100 Close % Change Over the Day TESCO Close % Change Over the Day 4/30/2007 6449 0.48 462.5 0.16 4/27/2007 6418 -0.79 461.75 -1.23 4/26/2007 6469 0.11 467.5 0.00 4/25/2007 6462 0.51 467.5 0.38 4/24/2007 6429 -0.77 465.75 -0.16 4/23/2007 6479 -0.11 466.5 0.65 4/20/2007 6486 0.71 463.5 0.00 4/19/2007 6440 -0.14 463.5 1.26 4/18/2007 6449 -0.74 457.75 -0.87 4/17/2007 6497 -0.29 461.75 1.32 4/16/2007 6516 0.84 455.75 -0.11 4/13/2007 6462 0.72 456.25 0.44 4/12/2007 6416 0.05 454.25 -0.38 4/11/2007 6413 -0.06 456 0.11 4/10/2007 6417 0.31 455.5 -0.60 4/5/2007 6397 0.52 458.25 0.27 4/4/2007 6364 -0.03 457 1.22 4/3/2007 6366 0.81 451.5 0.56 4/2/2007 6315 0.11 449 1.07 3/30/2007 6308 0.65 444.25 -0.17 3/29/2007 6267 0.00 445 1.08 3/28/2007 6267 -0.40 440.25 -0.51 3/27/2007 6292 0.02 442.5 0.91 3/26/2007 6291 -0.76 438.5 -0.17 3/23/2007 6339 0.33 439.25 -0.34 3/22/2007 6318 0.99 440.75 -0.45 3/21/2007 6256 0.58 442.75 1.61 3/20/2007 6220 0.50 435.75 0.00 3/19/2007 6189 0.96 435.75 0.93 3/16/2007 6130 -0.05 431.75 1.11 3/15/2007 6133 2.22 427 2.34 3/14/2007 6000 -2.61 417.25 -1.82 3/13/2007 6161 -1.16 425 -1.45 3/12/2007 6233 -0.19 431.25 -2.32 3/9/2007 6245 0.00 441.5 1.67 3/8/2007 6245 1.45 434.25 -0.29 3/7/2007 6156 0.29 435.5 1.10 3/6/2007 6138 1.32 430.75 1.29 3/5/2007 6058 -0.95 425.25 -1.51 3/2/2007 6116 0.00 431.75 2.07 3/1/2007 6116 423 Average% Change 0.14 0.23 SD of % Change 0.82 1.06 Data Sources: TESCO stock prices from http://in.finance.yahoo.com/. FTSE100 indices from http://www.londonstockexchange.com/NR/rdonlyres/832FA36C-78DF-41D3-AD60-4DF5DDA2EF3E/0/SecondaryMarketStatistics0703/04.pdf Read More