Market Efficiency In An Emerging Market
This unprecedented rise of China and India to global economic superpowers has defied all expectation and shows no signs of waning. Even in the tough economic condition of the past few years both countries have registered growth rates which are close to the highest in the world. But has this come at the cost of market efficiency? This report aims to test empirical evidence and run other tests to find if both the markets are weak form efficient.
Tests are conducted over a period of 4 years, covering the global recession, to see if market efficiency has been affected in any way. The empirical analysis is done by performing various unit root tests and the auto correlation tests. Event study is conducted on the two largest oil and gas companies in India and China, by studying the market reaction to the announcement of dividends during the sample period. Market efficiency can also be affected by seasonal anomalies. These anomalies are caused by the local holidays or other factors. This research aims to study the effect of these holidays on the returns to investors.
Efficient markets will help attract domestic investments and significant foreign investments. The next decade will be a vital phase for both countries as they vie for natural resources and investments. An efficient and mature market might be an indicator of which of the two neighbours will emerge as a true global superpower at the turn of this decade.
Introduction of the Research Problem
According to the Efficient Market Hypothesis proposed by Eugene Fama in 1970, an efficient market is one in which the true value of an investment is reflected in its market price. The market price is an aggregate of all the historical and present information available. This research aims to comparatively test the market efficiency in the two emerging markets of India and China, with focus on finding the effects of the Global Recession of 2008-2009 on it. The research will also explore the possibility of extrapolating the findings to other emerging economies.
A brief study of the origins and development of the two markets after the colonial rule and economic reforms in the mid 1990, which resulted in the rapid growth of the two economies, will be undertaken. Both India and China are heavily influenced by their rich cultures and religion plays an important role, it is thus important to test if they have any influence on returns and efficiency in the markets. This will be done by conducting holiday effect tests during the periods of Dusshera and Diwali in India and Chinese New Year and National Day in China. Another important method for testing market efficiency is by conducting event studies. The market reaction to announcement of dividends by companies will be studies to comment upon the informational efficiency and overall market efficiency.
According to the Efficient Market Hypothesis, which is based on the theory first forwarded by Samuelson (1965) and later formally reviewed and presented by Fama (1970), no one can consistently make gains in excess of the average market returns. It further states that at any point in time, the security prices fully reflect all the information available and no information or analysis can be expected outperform benchmark market returns. There are three forms of efficiency according to the hypothesis:
Weak Form efficiency: Prices reflect all past publicly available information and past market data. No form of technical analysis will result in abnormal gains.
Semi-Strong Form efficiency: Prices reflect all past publicly available information and past market data and also instantly change to reflect new information.
Strong Form efficiency: Prices reflect all past and new information and also reflect any hidden (insider) information.
In case of semi-strong efficiency, in addition to technical analysis, fundamentals analysis is also of no use and in strong form; even insider information does not result in extraordinary gains. Random Walk Theory was the forerunner to the EMH. It was first hypothesized by French mathematician Louis Bachelier and later developed by Cootner (1964). According to it, even though the markets are inefficient, it is impossible to consistently beat the returns due to its inherent unpredictability, akin to drunk trying to make his way home.
“An ‘efficient’ market is defined as a market where there are large numbers of rational, profit-maximizers 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, 1965)
“This implies that in an efficient market, equity research and valuation would be a costly task that provided no benefits. The odds of finding an undervalued stock should be random (50/50). Also, a strategy of randomly diversifying across stocks or indexing to the market, carrying little or no information cost and minimal execution costs, would be superior to any other strategy, that has larger information and execution costs. There would be no value added by portfolio managers and investment strategists.In an efficient market, a strategy of minimizing trading, i.e., creating a portfolio and not trading unless cash was needed, would be superior to a strategy that required frequent trading.” (Damodaran, A, 2010)
However, the paradox is that if investors find no undervalued stock (i.e. inefficiencies), they would stop looking for them, thus will stop being “profit maximizing investors”, which in turn makes the markets inefficient. It is therefore argues that an efficient market is a self-correcting mechanism where inefficiencies occurred occasionally.
After Fama found that the American Dow Jones Index was weak form efficient; Solnik (1973) tested various European markets and found that they were efficient, but to a lesser degree than of the DJI. However, an analysis of some of the smaller European indices by Jennergren and Korsvold (1975) showed inconclusive results. Tests on some under developed markets in 1985 showed that they were inefficient. Empirical studies on Asian stock markets also have been conducted on a large scale. Research conducted by Yu and Mukherjee (1999) and Groenewold (2003) found that the SSE and SZSE were weak form inefficient. They found that the B share index of both the indices do not follow a random walk. However, Seddighi and Nian (2004) found that the SSE was weak form efficient based on returns of a one year period in 2000. Similarly, research in the late 90s shows that the Taiwanese exchange was indeed weak form efficient. Moreover, a comprehensive study by Ma and Barnes (2001) conducted on both the SSE and SZSE could not comprehensively accept or reject the null hypothesis.
There have been a few studies on the Indian Market too. Using the BSE data for a period seven years from 1987 to 1994, Poshakwale (1996), concluded that the Indian market was inefficient. His findings were later supported by the studies of Pant and Bishnoi (2002), Pandey (2003) and Gupta and Basu (2007), using different tests over different time scales. However, earlier studies conducted in 1997 by Sharma & Kennedy, Barua (1987) and Ramachandran (1985) found evidence that the Indian Stock Markets were in fact weak form efficient.
Studies on Day of Week effect have been fewer in the Indian and Chinese markets. Nath and Dalvi (2004) found that after the introduction of rolling settlements in the NSE (an effect of Market Reforms), the returns on Mondays and Fridays had a high deviation. This shows the presence of inefficiency in the maket. Similarly, Shiguang Ma (2004) also found the presence of inefficiency in the Shanghai and Shenzhen stock markets. Both studies used the Regression with bitweights and dummy variables to perform the tests.
The literature reviewed shows that it still cannot be completely said if stock markets follow a random walk, especially in emerging markets. There is an abundance of literature but the results have been mixed. It is important to conduct timely and accurate studies to determine the efficiency of these markets because of the potential reward to investors. There have always been studies and debates about the applicability of the random walk model but this is now especially important because of the recent global recession.
Despite being classed as emerging markets, the two largest stock exchanges of both the countries are among the top 15 stock exchanges in the world, based on their market capitalization. India’s Bombay Stock Exchange (BSE) and National Stock Exchange of India (NSE) are at numbers 8 and 9 respectively, with a combined market capitalization of $ 3227 billion, while China’s Shanghai Stock Exchange (SSE) and Shenzhen Stock Exchange (SZSE) are at numbers 5 and 13, with a total market capitalization of $ 4048 billion. (IMF’s World Development Indicators, 2010).
In recent years, despite the global turndown, India and China have been powerhouses of growth, with figures of 10% and 9% respectively. They also have the highest levels of domestic savings in the world, which can be turned to investments in the presence of an efficient market. The two neighbours can also be an attractive destination for Foreign Investments for this reason because of the effect of the slowdown in developed markets. A market following random walk is in equilibrium as equity is appropriately priced, while the absence of randomness skews the risk-return characteristics. This in turn effects the appropriate allocation of capital, which can be detrimental to overall economic growth. Capital markets are important to decision makers as the signals the investors send can affect the policy of the company. Undervalued stock can trigger buybacks, which prevents allocation of capital to growth and investment activities.
India and China’s growth and histories have always intertwined and faced a lot of similar problems, but answered with very different responses. There has always been hostility between the two since a war in 1954 and the geo-political situation in Southeast Asia. The two have been vying for dominance in Asia and now for raw material to fuel their rapid growth. They are now competing to attract investors and also to free up their significant domestic savings. The results of this research may shed more light on who is winning that competition as efficient markets gain investor confidence. This research also will try to fill the void that exists with the present literature by updating it to include the effects of the recession on the markets and the lack of comparative quantitative analysis of the two markets.
Research Aims & Objectives
The research aims to test the efficiency of secondary markets in India and China.
Study the origins and structure of the stock exchanges in both countries.
Test weak form efficiency by Random Walk Behavior and conducting ADF test and Philips-Perron test.
Undertake an event study by studying the effect of dividend announcement by Reliance Industries in India and PetroChina in China.
Test the effect of local holidays and/or festivals on returns.
Explore the possibility of extrapolating the findings to other emerging markets.
Collection of relevant data is vital for accurate findings. Selection of sample size and data source is discussed in this section as well as the quantitative techniques to be used. A time plan is also set. Since this research depends more on quantitative data and numbers, data collection would primarily involve secondary research. Due to the abundance of general research in the field of random walk theory and market efficiency, the quality of information available is generally good. However, only recognized sources will be used to ensure accuracy of the information.
Primary data will be collected through market data aggregators like Datastream International and Digitallook. Future literature review will be done through access various available databases and through requests to the library.
(i) Data Collection and Sampling:
As stated earlier, the stock markets selected will be the Bombay Stock Exchange’s BSE-100 and the NSE’s CNX 100 for the analysis of India’s Markets and Shanghai Stock Exchange’s SSE Composite and Shenzhen Stock Exchange’s SZSE 100 Index will be used for China. This research will use the daily stock return data, computed from the closing prices of the listings of the selected indices. This will be done over a period of January 2007 to December 2010. This period has been selected to fully study the before and after effect of the global recession of 2008-2009. The weekly returns will be based on the closing value on Friday of every week and monthly returns on the closing value of the last working day of the month. The data will be sourced from DataStream.
The daily return is calculated using the formula –
(ii) Quantitative Analysis:
The random walk behavior can be tested by performing unit root tests. The tests which will be used are
Auto Correlation FunctionTest:
“The autocorrelation function (ACF) test is examined to identify the degree of autocorrelation in a time series. It measures the correlation between the current and lagged observations of the time series of stock returns, which is defined as:
where k is the number of lags, and Rt represents the real rate of return calculated as:
Two important elements for estimating of autocorrelation are the standard error test and the Box Pierce Q (BPQ) test. The standard error test measures the autocorrelation coefficient for individual lags and identifies the significant one, while the Box Pierce Q test, measures the significant autocorrelation coefficients at the group level.
The standard error k is defined as:
where N is the total number of observations and k is the autocorrelation at lag (k).
Box Pierce Q is identified as:”
(Islam and Watanapalachaikul, 2005)
Dickey – Fuller Test
“Dickey-Fuller statistic test for the unit root in the time series data Pt is regressed against Pt-1 to test for the unit root in a time series random walk which is given as:
If ρ is significantly equal to 1, then the stochastic variablr Pt is said to have a unit root. A series with unit root is said to be un-stationary and does not follow the random walk. There are three Dickey-Fuller tests for testing unit root in a series.
The above equation is written as
Where, δ = (p-1) and here if δ is equal to 0, Pt is a random walk.
To test the validity of market efficiency, random walk hypothesis may be tested through this test. Unit root test has been conducted on the index return time series function, by running the following regression equations
where, is constant term and β is the coefficient of trend term. The null hypothesis for each is
The null hypothesis that Pt is a random walk can be rejected if calculated τ is greater than the tabulated τ. Calculation of τ is similar to the estimation of τ -statistic but this value is compared with tabulated τ statistic, whose critical values have been tabulated by Dickey & Fuller on the basis of Monte Carlo simulations. The null hypothesis that Pt is a random walk can be rejected if calculated τ is greater than the tabulated τ.” Pant and Bishnoi (2003).
Philips – Perron Unit Root Test:
“Phillips-Perron (1988) developed a number of unit root tests that have popular in the analysis of Financial time series. The Phillips-Perron (PP) unit root tests differ from the ADF tests mainly in how they deal with serial correlation and heteroskedasticity in the errors. The test regression for the PP tests is given by:
Where μt is I(0) and may be heteroskedastic. The PP tests correct for any serial correlation and heteroskedasticity in the errors μt by using OLS estimation and modifying the test statistics tλ=0 and ˆ T . These modified statistics, denoted Zt and Zλ are given by:
Estimated values of λ and its standard errors obtained from OLS results from the ADF tests. The sample variance of the least squares residual u ˆ is a consistent estimate of σ2, and the Newey-West long-run variance estimate of u using u ˆ is a consistent estimate of w2.
Under the null hypothesis that λ=0, the Zt and Zλ statistics of the PP test have the same asymptotic distribution as ADF t-statistic and normalized bias statistics. One advantage of the PP tests over the ADF tests is that the PP tests are robust to general forms of heteroskedasticity in the error terms ut.” (Onour, 2009)
(iii) Event Analysis:
Event analysis is another important test for efficiency. The market model is used to test for abnormal returns. Under normal conditions, a stock is expected to give a return as the market portfolio. Only in the light of unexpected news, a stock’s return will fall outside the market return This is called the confidence interval. Thus an abnormal return AR is defined as the after-event return that falls outside the confidence interval.
The CI used to define statistical power of an AR is:
CI = r^ i,t+e ± tcv * se(r^ i,t+e)
Where r^ i,t+e is the after event return
&, tcv is the critical value
(iv) Time Frame:
Following is the monthly distribution of time, with a weekly breakdown:
Begin Secondary Research.
Review texts and other thesis
Refine Objectives and Thesis Methodology.
Discuss new objectives and methodology with supervisor.
Implement Feedback from Supervisor.
Prepare rough draft of literature review and methodology.
Complete essential data collection.
Present findings for review by Supervisor.
Complete literature review and methodology.
Run empirical analysis on the data collected.
Present findings to Supervisor.
Make correction and re-run the tests.
Start rough draft of final report.
Finish rough draft of final report.
Present the draft to supervisor for review and make necessary changes.
Finish final report and send for binding.
Presentation of Final Dissertation.
Since the thesis mainly involves secondary research, access to the library catalogue and other databases is of vital importance. Access to Datastream International is required, which is available at the library. Also access to computer labs is required to run the quantitative tests.
The aim of the research is to comparatively test the market efficiency of two of the world’s fastest growing economies. This will be done by taking empirical analysis and also conducting an event study. The empirical analysis is done by performing various unit root tests and the auto correlation tests. Event study is conducted on the two largest oil and gas companies in India and China, by studying the market reaction to the announcement of dividends during the sample period. Market efficiency can also be affected by seasonal anomalies. These anomalies are caused by the local holidays or other factors. This research aims to study the effect of these holidays on the returns to investors.
Samuelson, P. (1965). Proof that Properly Anticipated Prices Fluctuate Randomly. Industrial Management Review. Spring 6, 41-49.
Fama, E. (1970). Efficient Capital Markets: A Review of Theory and Empirical Work, Journal of Finance, 25, 283-306.
Cootner, Paul H. (1964). The Random Character of Stock Market Prices. MIT Press.
Solnik, H. B. 1973, “Note on the Validity of the Random Walk for European Stock Prices,”Journal of Finance, 28(5), 1151-1159.
Groenewold, N., Tang, S. H. K. and Y. Wu (2003), “The Efficiency of the Chinese Stock Market and the Role of the Banks”, Journal of Asian Economies, Vol.14, pp. 593-609.
Mookerjee, R. and Yu, Q. (1999), “An Empirical Analysis of the Equity Markets in China”, Review of Financial Economics, Vol.8, pp. 41-60.
Seddighi, H. R. and Nian W. (2004), “The Chinese Stock Exchange Market: Operations and Efficiency”, Applied Financial Economics, Vol.14, pp. 785-797.
Ma, S and Barnes, M (2001), “Are China’s Stock Markets Really Weak form Efficient?”, Adelaide University.
Barua, S. K. (1980), “Valuation of Securities and Inflence of Value on Financial Decision of a Firm”, Doctoral Dissertation, Indian Institute of Management, Ahmedabad.
Gupta, R., and Basu, P. K. (2007), “Weak Form Efficiency in Indian Stock Markets”, IBER Journal , Vol.6, No.3, pp. 57-64.
Pandey, A. (2003), “Efficiency of Indian Stock Market”, A Part of Project Report, Indira Gandhi Institute of Development and Research (IGIDR), Mumbai, India.
Pant, B., and Bishnoi, T. R. (2002), “Testing Random Walk Hypothesis for Indian Stock Market Indices”, Working Paper series, IIM, Ahmadabad.
Poshakwale, S. (1996), “Evidence on weak Form Efficiency and Day of the Week effect in the Indian Stock Market”, Finance India, Vol.10, No.3, pp. 605-616.
Ma, S. (2004), “The Efficiency of China’s Stock Market”, Ashgate Publishing.
Onour,I. (2009), “Testing Efficiency Performance of Saudi Arabia Market” JKAU: Econ. & Adm., Vol. 23 No. 2, pp: 15-27
Islam, S and Watanapalachaikul, S (2005), “Are Emerging Financial Markets Efficient? Some Evidence from the Models of the Thai Stock Market” Victoria University, Financial Modelling Program.