CO-INTEGRATION IN CAPITAL MARKETS OF BRICS NATIONS

CO-INTEGRATION IN CAPITAL MARKETS OF BRICS NATIONS S.V. Phanindra Natha Naidu Research Scholar, Dept. Business Management, Yogi Vemana University, Kadapa Dr. Y. Subbarayudu Asst. Professor, Dept. Business Management, Yogi Vemana University, Kadapa ABSTRACT The study covers the co-integration analysis among the stock markets in group of nations with acronym BRICS comprising of Brazil stock market, Russian stock market, India s stock market, China Stock Market and South Africa stock market. These five countries have taken reforms and have their presence in the globalization process for last two decades or more competing with each other and evolving as Emerging Markets. South Africa joined the group recently, whereas other four nation joined earlier. BRICS recent summit in Brazil comes out with the plans to come up with development Bank which bring these markets come close and compete themselves. In this context there is perception that stock market movement one market has impact on the stock market of other group nations basing on the stock market movement happened among the European markets in recent history. As India being one among the Emerging Markets (BRICS), the study of the market co-integration and spill over of the same from various markets is essential for the better formulation of policies in India. Further, the same helps in understanding international portfolio management cash flows movement among the markets and how to face the future challenges. The present study in this paper concentrates on the co-integration analysis among Ibovespa of Brazil, RTS of Russia, BSE Sensex of India, Shangai s SSE Composite of China and South Africa s Johannesburg stock exchange indices. Introduction India, one among the three countries which were not affected by the recent recession and other two being Iran and China may be because of the unique features these countries having with largest consumption markets and other macro fundamentals. Though this type of uniqueness happens it is necessary for these countries to study how their countries capital markets are related with global markets. www.apjor.com Page 43

From the financial literature in recent times it is overwhelmingly revealed that there is greater close comovement of stock markets of global major countries and group of countries. This co-movement of stock markets in financial terms are said to be integrated or inter linked and is also observed that the degree of integration among the stock markets is increasing from the last two decades. The important contributing factors for this phenomenon are policies relating to liberalization and globalization of the markets, Chou, Ng, Pi (1994). For India as it has undertaken the reforms since 1991, the stock markets have grew to the new heights and presence of Foreign Institutional Investors also increased many folds and thus stock markets are exposed to international stock markets influence. Under these circumstances, it should be emphasized to examine, how India s stock markets are influenced by the major countries stock markets and in particular Indian market with its co-member countries of BRICS. This is so stressed as India being one member country among the new group of nations called BRICS comprising of Brazil, Russia, India, China and South Africa which are going to come up with development bank, from excerpts of BRICS summit, Brazil, 2014. In order to know about the co-integration or co-movement of stock markets indices it is necessary to know the few words or terms used in statistics. Stock market indices data is time series data. In time series literature each time series is said to be integrated of the order 1 or I(1) that is when time series as such considered is non-stationary in nature and when first difference is taken is transformed into stationary series. When two non-stationary time series tend to move together in time are said to be co-integrated. The test procedure for co-integration is very simple. Regress one I(1) variable on another using least squares. Then test the residuals for non-stationarity using the (augmented) Dickey-Fuller test. If the series are co-integrated, the Dickey-Fuller test statistic will be statistically significant. The null hypothesis is that the residuals are non-stationary. Rejection of this leads to the conclusion that the residuals are stationary and the series are co-integrated. Stock markets indices are time series data generated through the trading in stock market and indices assume value through stochastic process exhibits non-stationary nature and when two or more market movement are in same direction, then they exhibit co-integration when above procedure and tests are conducted and then these markets are said to be co-integrated. Outcome of the co-integration analysis is very much useful for the international fund managers who seek international diversification in their investment portfolio investing across different countries. When markets are not co-integrated there exists more possibility for international diversification to reduce the country specific market risk and if markets are increasingly co-integrated then diversification benefits will be reduced. Therefore, the importance of the co-integration analysis is gaining much importance in the fund managers of FIIs, financial analysts and researchers and academicians. The present paper purely concentrates on the co-integration analysis of the BRICS nations due their nature as Emerging Markets and Size in terms of capitalization and consumption markets point of view and also due to their proposed initiation going to come up. From the factors such as Size and Market Capitalization, these countries stand in the global order among top 21 markets and recently influencing the globe through the association of BRICS and also about to come up with Development Bank from the Association. (South Africa recently joined in 2011). Major stock exchanges (top 21 by market capitalization), as at 31 June 2014 by Monthly reports of World Federation of Exchanges as follows. China through its Shangai Stock Exchange ranks 7 th with a market capitalization of 2,408 USD billion, India through its National Stock Exchange and Bombay Stock exchange ranks 12 th and 13 th with market capitalization of 1,472 USD billion and 1,499 USD billion for NSE and BSE respectively, Brazil through BM and F BOVESPA Sao Paulo exchange ranks 17 th with market capitalization of 1,100 USD billion and Russia through its Moscow Stock Exhange ranks 21 st with market capitalization of 735 USD billion. South Africa ranks through its Johannesburg stock exchange ranks 18 th with market capitalization of 1,028 USD billion. Due to the paucity of data from South African market, its data is used from 2009 whereas other markets are considered from July 1997 to July 2014. www.apjor.com Page 44

Literature Review From the context of global stock market and developed stock markets there are more studies on the stock market indices co-integration among stock markets. According to the Chou, Ng, Pi (1994) the basic reasons for co-integration are liberalization and globalization as said earlier in this paper and from Bachman et. al. (1996), the other factors that indirectly inter link markets are similarity in income patterns, the formation of a currency area strengthening the relationship between domestic economic variables, the role of a dominant financial centre within a multinational area facilitating within area capital flows, a common technological trend assimilating concurrently into different economies, financial deregulations allowing investors to extend their portfolios internationally and significant international trades in general and in capital goods inducing strong economic ties. In search of the reviews on the co-integration, in this study, reviewed from the global and regional level literature. In the literature, the integration of global equity markets has been a well-studied topic since the stock crisis of October 1987. The merits of the international diversification in containing the systematic risk is long recognized in the literature with one of the earliest attempts by Grubel (1968) and origin of studies on contagion and integration stem from Sharpe (1964) and Grubel and Fadner (1971). Taylor and Tonks (1989) studied the market integration among U.S., U.K., Germany, Netherlands and Japan using monthly data on stock price indices for the sub-periods, April 1973 September 1979 and October 1979 June 1986 using Engle and Granger, 1987 bivariate co-integration technique and found that stock price index of the U.K. was co-integrated with the stock price index of the U.S., Germany, Netherlands and that of Japan for the later period but not for the former period. Arshanapalli and Doukas (1993) using unit root and co-integration technique examined relationships and interactions among the stock markets of New York, Japan, Paris, Frankfurt, and London, from January 1980 to May 1990 and concludes that there has been an increasing interdependence among these stock markets after the crash of 1987, except for Japanese stock market. The French, U.K., and German markets are affected by the U.S. market. The Japanese market performance has no links at all with any market in the U.S, France, Germany and U.K. Cheung and Mak (1992) based on weekly return series for the period of 1977 and 1988 investigated the relationship between the two developed markets U.S. and Japan with eight Asia-Pacific markets viz. Australia, Hong Kong, Korea, Malaysia, Philippines, Singapore, Taiwan and Thailand and found that U.S. market leads the stock market of most of these countries with the exception of Korea, Taiwan and Thailand and Japanese market found to have a less influence in this region. Janakiramanan (1998) and Hsiao (2003) have tried to examine the possible linkages between the stock markets in the Pacific-Basin region and Asia-Pacific region respectively with the U.S. The unidirectional linkages from the U.S. market to the others are found to be significant in both the studies. Abhilash (2003) using the Engle-Granger residual based test of co-integration, Hsiao, Granger and Sims tests of causality revealed the presence of long run relationship between the NASDAQ Composite Index and NSE Nifty, indicating a direction of causation from NASDAQ Composite Index to NSE Nifty. Nath and Verma (2003) have examined the interdependence of the three major stock markets in South Asia, viz. India, Singapore and Taiwan using daily stock market data from January 1994 to November 2002 and found no co-integration between the stock market indices for the entire peiod and found mild causality for some years. Bose (2005) examined the co-movement of Indian Stock Market with U.S., Japan and other Asian Markets using daily data for the period January 1999 to June 2004 and found that the nature of co-movement or integration with emerging Asian Markets does not yet warranty any immediate concern regarding possible contagion and the degree of integration shows still much scope for portfolio diversification. The source of scope to present study to investigate in BRICS nations stock market indices is from the study of Ortiz (2006) on NAFTA Capital markets. His study reveals financial integration among NAFTA Capital markets and between these markets and world capital market and found a time-varying integration among NAFTA capital markets, and a mild segmentation and a time-varying integration www.apjor.com Page 45

between these markets and the world capital market. Hande Erdinc and Joniada Milla (2009) also found similar results among the markets of France, Germany and U.K. In this present paper investigation is undertaken to know the nature of co-integration among the member nations stock market indices of BRICS and inter linkage between each other member on timevarying nature to come out with significant outcomes. Objectives of the Study 1. To study the stochastic pattern of the Indian BSE Sensex and other BRICS nations 2. To examine the results from the co-integration analysis in both bi-variate and multivariate environment. 3. To observe whether BRICS nations exhibits co-integration of stock indices or not. Methodology The present paper cover the study on co-integration of stock indices of the Brazil, Russia, India, China and South Africa choosing important indices of respective countries. Data of these indices is collected from yahoo.finance.com and verified from the respective websites of the stock markets. Work mainly concentrates on the Johansen Co-integration among group and between one nation to other nation using Granger causality test. Data set used in the co-integration analysis is monthly series from 01-07-1997 to 30-7-2014. As these countries are from different continents and cultural back drop their respective holidays or nonfunctioning date of markets may be different and respective observation period details are given in the following table 1. South Africa data is from the year 2009 due to non-availability from yahoo.finance.com and further the country recently joined BRICS. Table 1: BRICS Countries and their Indices, Period and Observations Country Index Period Monthly Observations Brazil Ibovespa 1/7/1997 to 205 31/7/2014 Russia RTS I 1/7/1997 to 205 31/7/2014 India BSE Sensex 1/7/1997 to 205 31/7/2014 China SSE Composite 2/7/1997 to 205 31/7/2014 South Africa JSE 1/2/2009 to 67 8/29/2014 Source: Compiled by the researcher Monthly data is used for co-integration analysis rather than daily data series because co-integration is a long-run phenomena and as such long time spans of data, rather high data frequency, is essential to conduct test for the existence of the co-integration, Hakkio and Rush (1992) and Bailey and Stulz (1990). Further the data is used in local currency value and used in exact values and also by taking logarithm value of each series as it may similar to the return on which investors are interested. Time Series Data Analysis Methodology For the purpose of conducting econometric analysis on time series data, certain precautions and preparation necessary for time series data. Time series data pertaining macroeconomic or stock indices data show trend and along the time, due to this reason they are statistically non-stationary series. The standard classical methods are based on the assumption that the mean and variances are constant and time invariant. In practice time series data of market indices using the unit root test exhibit means and variances change www.apjor.com Page 46

over time. This phenomena happens as time series reflect process that involves trend, cycle and seasonality. The Unit root tests determine the stationarity characteristics of the data. The non-stationary time series if regressed on the other non-stationary time series results into spurious regression. To avoid this problem the non-stationary series should be converted to stationary by removing the deterministic and/or stochastic trends. There are several unit root test methods such as Dickey Fuller test, Augmented Dickey Fuller test, KPSS test etc. For example ADF test of unit root is used in three different distinct models of generating processes of a series such as: p Model 1: Without any Constant and Trend y t = py t 1 + i=1 δ t y t i + μ t p Model 2: With Constant but no Trend y t = α + py t 1 + i=1 δ t y t i + μ t p Model 3: With Constant and Trend y t = α + βt + py t 1 + i=1 δ t y t i + μ t Where μ t is a white noise error term and y t i = y t 1 y t 2 ADF test will test null hypothesis against alternate hypothesis. The null hypothesis is H 0 : p = 0 and alternate hypothesis is H 1 : p 0. If null hypothesis is not rejected it shows unit root present in y, that is y is non-stationary. If a variable is stationary, that is it does not have unit root and is integrated of order zero or I(0). If unit root exists, time series data variable is non-stationary, and is differenced to make series stationary and again tested. If it rejects null hypothesis then, series is integrated of the order one or I(1) and on the other not rejects null hypothesis process is repeated. In general, time series y t will be integrated of order d, that is y t I(d) which means series stationary after differencing d times. Time series y t contains d unit roots, Dickey and Fuller (1981). When all variables are found to be integrated of the same order the co-integration test is conducted to determine whether long-term relationship exists among the variables. The co-integration test evaluates the co-movement of a long term asset prices under equilibrium approach. Concept of the Co-integration is that if all the components of a vector time series process y 1 have a unit root, or y 1 is multivariate I(1) process then the series are said to co-integrated when linear combination of them is stationary. The cointegration regression produces an error term with I(0). There are different co-integration methods like Engle Granger method, Johansen s method and Gregory Hansen method etc. Particularly, the Engle Granger method is bivariate in nature, Johansen s method is multivariate in nature and Gregory Hansen consider the structural breaks in the relation among variable and is multivariate in nature. Though directly Gregory Hansen method is not used in this paper, the concept of structural break is consider through the division of the data series into three different time periods and also with total time length. This approach is used to know whether their co-integration is time varying or is same across all time periods. The Johansen s method is used in this paper. The Johansen s co-integration estimation is based on the following model, Y t = A 1 Y t 1 + e t In the above model, Y t 1 and e t are the vector of (P x 1) and A is considered as an (P x P) matrix form and the 1 is considered as the identity matrix form. Johansen and Juselius (1990) used two likelihood ratio test viz. Trace test and Maximum Eigen value test to determine r, the number of co-integrating vectors. Trace Test The null hypothesis is that there are r or fewer co-integrating vectors in the system. The Statistic is g λ max r, n = T i=r+1 ln (1 λ i ). To determine the number of co-integrating vectors, r, test the sequence of null hypothesis r=0, r 1, r 2,.. r (q-1). If r q is the first null accepted then it is to conclude that there are r=q co-integrating vectors. www.apjor.com Page 47

Maximum Eigen Value Test The null hypothesis of the test is that the number of co-integrating vectors is r versus the alternative hypothesis that the number is r+1. The Statistic is λ max r, r + 1 = T ln (1 λ r+1 ). To determine the number of co-integrating vectors, r, test the sequence of null hypothesis r=0, r=1,..,r=p-1. If r=q is the first null accepted then it is to conclude that there are r=q co-integrating vectors. Co-integration regressions show long run or equilibrium relationships between economic variables. Cointegration analysis can be conducted at both bivariate and multivariate levels in order to know relationship among the stock market indices. Basing on only bivariate or multivariate co-integration analysis there is chance to miss the significant information about relationships among variable, some variables may not be co-integrated at bivariate level may be co-integrated collectively along with third or group of other variables. Data Analysis In order to analyse the behaviour of the indices of BRICS nations, the time period considered is from July 1997 to July 2014 for Brazil, Russia, India, China and for South Africa is from the Jan 2009 to July 2014. South Africa is considered from 2009 due to paucity of data and also because of fact that country has joined BRIC group in 2011 and from then onwards group of nations is BRICS. For the first step in study of integration of the market first time series trend has been observed in graph as shown in the Figure 1 and Figure 2. The visual inspection shows similar trend in time series. Then to know the integration of markets traditional correlation coefficient also used, though they may not be enough to understand integration of markets. Then analysed the stationary character of the time series using Augmented Dickey-Fuller test and results shown in Table 5. Results show that all time series of the market indices are non-stationary in nature and first difference of the series are stationary indicating that all time series are stationary of the order I(1). Having all the time series integrated of the same order, the Johansen co-integration test is applied to know the nature of the integration among the stock indices. Johansen and Juselius (1990) and Kasa (1992) revealed that in case of conflict between Trace and Maximum Eigen value tests, the former is superior. In this study also the similar approach is used to analyse the co-integration among the time series data of stock indices. The Complete Data set of the study is described in Table 2 through their summary statistics. The results of the ADF test are shown in Table 5. Complete Data set is from July 1997 to 2014, using this first Johansen Co-integration conducted on complete data set of BRIC s, the results were shown in the Table 6. Trace and Eigen value tests show no integration among the stock markets. Though Gregory Hansen approach to regime shift is not used, similar to this the approach in this study the complete data set is divided into three distinct time periods and again the Johansen Co-integration test is conducted on BRICs data from 1997 2005 and 2006 2009 and including South Africa in BRICS from 2009 to July 2014 and results shown in the Tables 7, 8, and 9 respectively. The result for 1997 2005 shown through Eigen value test indicate one co-integration equation among the BRICs, but as mentioned earlier in case of conflict between Trace and Eigen Value tests, Trace test result is considered superior and hence there is no integration among variables. For time period 2006 to 2009 also results reveal no integration among variables. For the time period between 2009 to July 2014 which also includes South Africa indices, the result of test shown through both the Trace and Eigen Value tests show one co-integration vector among the BRICS indices. Thus these markets exhibit the equilibrium relationship between them exists in the longrun. As there is expected co-operation among the member countries of the BRICS due to the upcoming development bank from these nations there exists more scope for integration among the stock markets in near future. www.apjor.com Page 48

Table 2: Summary Statistics of Close Price of Indices of BRICS nations LNSENSEX LNSHANG LNIBOVESPA LNRTS LNJSE Mean 9.0235 7.562307 10.24145 6.48236 8.853662 Median 9.115229 7.560247 10.4977 6.657921 8.881142 Maximum 10.1618 8.691948 11.19262 7.807868 9.220489 Minimum 7.941175 6.966722 8.77524 3.779862 8.217439 Std. Dev. 0.723262 0.372967 0.727189 1.001726 0.206451 Skewness -0.075363 0.568164-0.275483-0.719708-1.1206 Kurtosis 1.363489 2.990515 1.55274 2.459631 4.640517 Jarque-Bera 23.07007 11.03012 20.48398 20.19182 21.53578 Probability 0.00001 0.004026 0.000036 0.000041 0.000021 Sum 1849.818 1550.273 2099.498 1328.884 593.1954 Sum Sq. Dev. 106.7141 28.37731 107.876 204.7049 2.813062 Observations 205 205 205 205 67 Source: Compiled by the researcher Figure 1 : Indices of the BRICs nation from 1997 to 2014 12 11 10 9 8 7 6 5 4 3 1998 2000 2002 2004 2006 2008 2010 2012 2014 Ln(Ibovespa) Ln(Shang) Ln(Sensex) Ln(RTS) www.apjor.com Page 49

Figure 2 : Indices of the BRICS s nation from 2009 to 2014 12 11 10 9 8 7 6 I II III IV I II III IV I II III IV I II III IV I II III IV I II III 2009 2010 2011 2012 2013 2014 Ln(Sensex) Ln(RTS) ln(jse) Ln(Shang) Ln(Ibovespa) Table 3: Correlation of Logarithmic Close Price of Indices of BRICs nations July 1997 July 2014 LNIBOVESPA LNSENSEX LNSHANG LNRTS LNIBOVESPA 1.000000 0.970670 0.723414 0.908087 LNSENSEX 0.970670 1.000000 0.689082 0.872327 LNSHANG 0.723414 0.689082 1.000000 0.611080 LNRTS_ 0.908087 0.872327 0.611080 1.000000 Source: Compiled by the researcher Table 4: Correlation of Logarithmic Close Price of Indices of BRICs nations Jan 2009 July 2014 LNIBOVESPA LNRTS LNSENSEX LNSHANG LNJSE LNIBOVESPA 1.000000 0.783338 0.382267 0.660674 0.249232 LNRTS 0.783338 1.000000 0.683325 0.288306 0.625096 LNSENSEX 0.382267 0.683325 1.000000-0.146846 0.897354 LNSHANG 0.660674 0.288306-0.146846 1.000000-0.346160 LNJSE_ 0.249232 0.625096 0.897354-0.346160 1.000000 Source: Compiled by the researcher www.apjor.com Page 50

Table 3: Augmented Dickey-Fuller (ADF) test for 1 st Differences of Logs of Indices Variable Test statistic BSE Sensex of India -13.55071* Ibovespa of Brazil -14.34602* Shangai Stock Exchange Composite of China -8.142943* RTS of Russia -11.41978* Johannesburg Stock Exchange index of South Africa -9.613445* Source: Compiled by the researcher Note: The critical values for the case with no trend are -3.462 for 1%, -2.875 for 5% and -2.575 for 10% respectively. For JSE index -3.534 for 1%, -2.906 for 5% and -2.591 for 10% respectively * Rejected null of unit at 1% significance level. Time series are stationary Table 4: Johansen Co-integration result for Indices of BRIC nations July 1997 to July 2014 No. of CE(s) Eigenvalue Trace Statistic None 0.081398 34.73653 47.85613 0.4620 At most 1 0.063078 17.75603 29.79707 0.5841 At most 2 0.022171 4.725027 15.49471 0.8373 At most 3 0.001204 0.240911 3.841466 0.6235 No. of CE(s) Eigenvalue Max-Eigen Statistic None 0.081398 16.98050 27.58434 0.5816 At most 1 0.063078 13.03100 21.13162 0.4494 At most 2 0.022171 4.484116 14.26460 0.8050 At most 3 0.001204 0.240911 3.841466 0.6235 * denotes rejection of the hypothesis at the level **MacKinnon-Haug-Michelis(1999) p-values Table 5: Johansen Co-integration result for Indices of BRIC nations July 1997 to 2005 No. of CE(s) Eigenvalue Trace Statistic None 0.248308 47.72983 47.85613 14 At most 1 0.128308 20.04330 29.79707 0.4200 At most 2 0.066074 6.723322 15.49471 0.6100 At most 3 0.000954 0.092573 3.841466 0.7609 No. of CE(s) Eigenvalue Max-Eigen Statistic None * 0.248308 27.68653 27.58434 0.0485 At most 1 0.128308 13.31998 21.13162 0.4232 At most 2 0.066074 6.630748 14.26460 0.5337 At most 3 0.000954 0.092573 3.841466 0.7609 * denotes rejection of the hypothesis at the level **MacKinnon-Haug-Michelis(1999) p-values www.apjor.com Page 51

Table 6: Johansen Co-integration result for Indices of BRIC nations 2006 to 2009 No. of CE(s) Eigenvalue Trace Statistic None 0.288932 35.41633 47.85613 0.4262 At most 1 0.222185 19.73091 29.79707 0.4412 At most 2 0.152854 8.172671 15.49471 0.4471 At most 3 0.011715 0.542093 3.841466 0.4616 No. of CE(s) Eigenvalue Max-Eigen Statistic None 0.288932 15.68542 27.58434 0.6923 At most 1 0.222185 11.55824 21.13162 0.5916 At most 2 0.152854 7.630578 14.26460 0.4175 At most 3 0.011715 0.542093 3.841466 0.4616 * denotes rejection of the hypothesis at the level **MacKinnon-Haug-Michelis(1999) p-values Table 7: Johansen Co-integration result for Indices of BRIC nations 2009 to July 2014 No. of CE(s) Eigenvalue Trace Statistic None * 0.549692 82.61518 69.81889 0.0034 At most 1 0.209126 31.55442 47.85613 0.6366 At most 2 0.152210 16.53892 29.79707 0.6738 At most 3 0.082717 5.971089 15.49471 0.6990 At most 4 0.006934 0.445348 3.841466 0.5046 No. of CE(s) Eigenvalue Max-Eigen Statistic None * 0.549692 51.06077 33.87687 0.0002 At most 1 0.209126 15.01549 27.58434 0.7470 At most 2 0.152210 10.56783 21.13162 0.6901 At most 3 0.082717 5.525742 14.26460 0.6745 At most 4 0.006934 0.445348 3.841466 0.5046 * denotes rejection of the hypothesis at the level **MacKinnon-Haug-Michelis(1999) p-values In order to obtain inference on the integration between pair of nations the Granger s Causality test is conducted among the variable of the BRICS and results tabulated in the Table 10. The results clearly indicate no causation between pair of nations. www.apjor.com Page 52

Table 10: Granger s Causality Test for Pair of Stock Indices: Null Hypothesis: Observations F- Statistic Prob. LN_SENSEX_ does not Granger Cause LN_SHANG_ 65 3.62439 0.0327 LN_SHANG_ does not Granger Cause LN_SENSEX_ 2.84756 0.0658 LN_RTS_ does not Granger Cause LN_SHANG 65 3.25037 0.0457 LN_SHANG_ does not Granger Cause LN_RTS 0.76244 0.4710 LN_JSE_ does not Granger Cause LN_SHANG_ 65 3.11189 18 LN_SHANG_ does not Granger Cause LN_JSE_ 3.29664 0.0438 LN_IBOVESPA_ does not Granger Cause LN_SHANG_ 65 2.03998 0.1390 LN_SHANG_ does not Granger Cause LN_IBOVESPA_ 2.86148 0.0650 LN_RTS_ does not Granger Cause LN_SENSEX_ 65 7.98498 0.0008 LN_SENSEX_ does not Granger Cause LN_RTS_ 0.33894 0.7139 LN_JSE_ does not Granger Cause LN_SENSEX_ 65 4.15578 0.0204 LN_SENSEX_ does not Granger Cause LN_JSE_ 3.47891 0.0372 LN_IBOVESPA_ does not Granger Cause LN_SENSEX_ 65 3.82328 0.0274 LN_SENSEX_ does not Granger Cause LN_IBOVESPA_ 3.42291 0.0391 LN_JSE_ does not Granger Cause LN_RTS_ 65 0.08975 0.9143 LN_RTS_ does not Granger Cause LN_JSE_ 5.21735 0.0081 LN_IBOVESPA_ does not Granger Cause LN_RTS_ 65 2.70370 0.0751 LN_RTS_ does not Granger Cause LN_IBOVESPA_ 5.10452 0.0090 LN_IBOVESPA_ does not Granger Cause LN_JSE_ 65 11.0003 9.E-05 LN_JSE_ does not Granger Cause LN_IBOVESPA_ 2.14031 0.1265 Source: Compiled by the researcher Conclusion From Size and Market Capitalization, BRICS countries stand in the global order among top 21 markets and recently influencing the globe through the association of BRICS and about to come up with Development Bank from the Association. (South Africa recently joined in 2011). Due to their ranking among the global countries and also on-going decisions among BRICS associate member countries it is highly significant to know the relation among the stock markets of these countries for the future economic development in their domestic land and for the other countries. Basing on these fact the present study conducted to throw light on the nature of stock markets. The study also significant for the reason that European Union countries in recent times shown the European Crisis which is in one way related with the close and greater financial integration among the countries of Europe and similar conditions may arise in distant future for the BRICS nations as these member countries are willing to come closure and integrate the financial market by a way of bilataeral and multilateral relations to easy and faster conduct of the business, in particular, in their currencies etc. Under the influence of these, the study of equity markets integration is taken up and the results illustrates that as of the present time the integration of the BRICS equity markets is significantly no integration among the BRIC nations based on data set from 1997 to 2014. Whereas when the integration among BRICS nation using the data set from 2009 to 2014 is subjected to Johansen co-integration analysis, resulted with one co-integrating vector from the trace and eigen value tests. The co-integration vector obtained indicates there exists the equilibrium relationship among these market in long-run. This phenomena nullifies the chances of international portfolio diversification for companies among these countries. On the other hand the results from the Granger s causality test for the same time period data set indicates no causal effects among these markets on the pair countries in bivariate setting. Hence, it is concluded that these countries as group exhibit the integration among the financial markets though not in country to country financial integration and unique finding from the study is the www.apjor.com Page 53

correlation co-efficient between the Sensex of India and SSE composite of China indicate negative correlation this shows their independent nature. In the light of the recent developments from the summits of the BRICS nations, facts of the present study greatly help in strategic studies. References: 1. Abhilash Nair S. and Dr. A. Ramanathan( 2003), Analysis of Co Movements of Select US& Indian Stock Price Indexes, pp. 1-19. 2. Arshanapalli, B., and Doukas, J. 1993, International Stock Market Linkages: Evidence from the Pre- and Post- October 1987 period. Journal of Banking and Finance, 17(1), pp. 193-208. 3. Bachman, D., Choi, J.J., Jeon B.N., and Kopecky, K.J., (1996), Common Factors in international Stock Prices: Evidence from a Cointegration Study, International Review of Financial Analysis, Vol.5, pp.39-53. 4. Bailey, W., Stultz, R.M., (1990), Benefits of International Diversification: The Case of Pacific Basin Stock Markets, Journal of Portfolio Management. 5. Suchismita Bose and Paramita Mukherjee, (2005) A Study of interlinkages between the Indian Stock Market and Some other Emerging and Developed Markets, Working Paper, ICRA Ltd. 6. Cheung, Y.L. and Mak, S.C. (1992), The international transmission of stock market fluctuations between the developed markets and the Asian-Pacific markets, Applied Financial Economics, Vol. 2, March pp. 43-47. 7. Dickey, D.A., and W.A. Fuller (1979), Distribution of the Estimators for Autoregressive Time-series with a Unit Root, Journal of the American Statistical Association, Vol.74, pp.427-431. 8. Dickey, D.A., and W.A. Fuller (1981), Likelihood Ratio Statistics for Autoregressive Time-series with a Unit Root, Econometrica. 9. Grubel, H.G., (1968), Internationally diversified portfolios: Welfare gains and capital flows, American Economic Review, Vol. 58, December, pp. 1299-1314. 10. Grubel, H.G. and Fadner r. (1971), Contagion and Trade: Why are Currency Crises Regional?, Journal of International Money and Finance, 18(4), 603-617. 11. Hsiao Frank S.T., Hsiao M.W. and Yamashita A., The Impact of theu.s. Economy on the Asia-Pacific Region: Does it Matter?, Journal of Asian Economics, Vol. 14, 2003, pp 219 241. 12. Hande Erdinc and Joniada Milla (2009), Analysis of Co-integration in Capital Markets of France, Germany and United Kingdom, Economic & Business Journal: Inquires and Perspectives, Vol. 2 No.1, October 2009, pp.109-123. 13. Janakiramanan S. and Lamba A.S., An Empirical Examination of Linkages Between Pacific-Basin Stock Markets, Journal of International Financial Markets, Institutions and Money, Vol. 8, 1998, pp 155 173. 14. Johansen, S., and Juselius, K., (1990) Maximum Likelihood Estimation and Inference on Co-integration with Applications to the Demand for Money, Oxford Bulletin of Economics and Statistics, Vol.52, pp. 169-210. 15. Kasa Kenneth (1992), Common Stochastic trends in international stock markets, Journal of Monetary Economics 29: pp. 95-124. 16. Nath G. C. and Verma S., Study of Common Stochastic Trend and Co-Integration in the Emerging Markets: A Case Study of India, Singapore and Taiwan, 2003. 17. Ortiz, E. and Lopez Herrara, F., (2006), Patterns of Co-integration at the NAFTA Capital Markets. 18. Pi, L., Chou,R., Ng, V., (1994) Co-integration of International Stock Market Indices, Proceedings of American Statistics Association annual meetings. 19. Sharpe, W. (1964), Capital Asset Prices: A Theory of Market Equilibrium Under Conditions of Risk, Journal of Finance, Vol.19, pp.425-442. 20. Taylor, M.P. and Tonks, I. 1989. The internationalisation of stock markets and abolition of U.K. exchange control, The Review of Economics and Statistics 71: 332 336 www.apjor.com Page 54