Relationship between Macroeconomic Factors and ...

New Age Business Strategies in Emerging Global Markets, First Impression: 2015, Excel India Publishers, Page: 104-123.

Relationship between Macroeconomic Factors and Aggregate Stock Returns in BRICS Stock Markets – A Panel Data Analysis 1

Dr. Vanita Tripathi1 and Arnav Kumar2

Associate Professor (Finance) & P.I. (UGC M.R.P.), Department of Commerce, Delhi School of Economics, University of Delhi, India. 2 Research Scholar, Department of Commerce, Delhi School of Economics, University of Delhi, India. Email: Email address: [email protected], [email protected]

Abstract This paper examines the relationship between select macroeconomic factors (i.e., GDP, Inflation, Interest Rate, Exchange Rate and Money Supply) and aggregate stock returns in emerging markets constituting the BRICS block over the period 1995 to 2014 using quarterly panel data. This relationship is also examined during two sub periods viz., a Pre Crisis period (1995:Q1 to 2007:Q2) and a Post Crisis Period (2007:Q3 to 2014:Q4). Robust econometric tests like Panel Granger Causality Test, Pedroni’s Panel Cointegration Test and Panel Auto Regressive Distributed Lag (ARDL) Model has been used. We find that primarily in short run there is unidirectional causality running from stock returns to GDP growth rate, inflation rate, rate of change in exchange rate and money supply. The results are almost similar in pre and post crisis periods, except that in the pre crisis period, there is bidirectional causality between stock returns and inflation, while in the post crisis period it disappears. Long run panel causality results reveals unidirectional causality from stock returns to GDP growth rate in total and post crisis periods. However in pre crisis period, there was no long run causal relationship. Pedroni’s panel cointegration test shows that stock indices are cointegrated with GDP in total period and with GDP, inflation and money supply in post crisis period. Panel ARDL models have explanatory power ranging from 28% in total period to 62% in post crisis period. We find that while current stock returns are negatively linked to rate of change in exchange rate and money supply; they are positively linked to their own lagged values. In pre crisis period, rate of change in money supply significantly explains stock returns while in post crisis period, inflation rate, interest rate and rate of change in exchange rate and money supply negatively affects BRICS panel stock returns. These findings, besides augmenting the empirical literature and knowledge domain on the topic, have significant implications for policy makers, regulators, researchers and investing community in emerging markets. The regulators need to ensure that financial sector reforms agenda consciously considers interlinkages between stock markets and real economy. The investment community can devise investment strategy, using the results of this study to earn arbitrage profits in emerging stock markets. Keywords: Aggregate Stock Returns, BRICS Stock Markets, Macroeconomic Factors, Panel Auto Regressive Distributed Lag (ARDL) Model, Panel Causality, Panel Cointegration. JEL Classification: B26, C23, C58, E44.

Relationship between Macroeconomic Factors and Aggregate Stock Returns in BRICS Stock Markets – A Panel Data Analysis

Introduction In the present-day scenario, where there is an increasing integration of the financial markets and implementation of various stock market reforms, the activities in the stock markets and their relationships with the macro economy have assumed significant importance. As mentioned by Galbraith (1955), “the stock market is but a mirror, which provides an image of the underlying or fundamental economic situation”. Therefore, in the past two decades an increasing attention is being paid to the relationship between share returns and the macroeconomic variables by both economists and finance specialists. Different macroeconomic factors which have been examined for their possible relationship with stock returns are - GDP growth rate, Inflation rate, Interest rate, Foreign Exchange Rate and Money Supply. According to Flannery and Protopapadakis (2002) “there are two direct benefits of identifying the macro variables that influence aggregate equity returns - it may indicate hedging opportunities for investors and if investors as a group are averse to fluctuations in these variables, these variables may constitute priced factors”. Macroeconomic factors influence the stock market performance and particularly stock returns through their effect on future cash flows and the rate at which they are discounted. The relationship between stock prices and macroeconomic variables is well illustrated by theoretical stock valuation models such as Dividend Discount Model (DDM), Free Cash Flow Valuation, and Residual Income Valuation. According to the models, the current price of an equity share is approximately equal to the present value of all future cash flows; thus any economic variable affecting cash flows and required rate of return in turn influences the share value as well. GDP growth rate is typically used as a proxy for the level of real economic activity. It is theoretically shown that the productive capacity of an economy rises during economic growth, which in turn contributes to the ability of firms to generate cash flows. Hence a positive relationship between real economy and stock prices exist [Fama (1981), Mukherjee and Naka (1995)]. In the process of stock valuation, it is important to consider the effects of inflation on stock returns. In theory, stocks should be inflation neutral, and rising inflation should have no impact on stock valuations. A negative relationship between inflation and stock prices is contended in literature because an increase in the rate of inflation is accompanied by both lower expected earnings growth and higher required real returns. [Fisher (1930), Fama (1981), Tripathi and Kumar (2015 a & b)]. Interest rates are expected to be negatively related to market returns either through the inflationary or discount factor effect. An increase in interest rate increases the discount rate (or minimum required rate of return) and hence reduces share prices [Asprem (1989), Mukherjee and Naka (1995)]. There is no theoretical consensus on the existence of relationship between stock prices and exchange rates or on the direction of the relationship. However, in the literature, two approaches have been asserted to establish a relationship between exchange rate and stock prices: The goods market model and the portfolio balance model. Goods market model suggests that changes in exchange rates affect the competitiveness of a firm, which in turn influence the firm’s earnings or its cost of funds and hence its stock price. Thus, goods market models represent a positive relationship between stock prices and exchanges rates with direction of causation running from exchange rates to stock prices. On the other hand, Portfolio balance model assumes a negative relationship between stock prices and exchange rates. A rise in domestic stock prices would increase the demand for domestic currency and


cause exchange rate to appreciate. A rising stock market leads to the appreciation of domestic currency through direct and indirect channels. [Maysami et al. (2004), Mukherjee and Naka (1995)]. Money supply’s net impact on stock returns is also debatable and can be positive or negative. An increase in money supply increases liquidity making more money available for consumption and investments and lowers interest rate in the economy favourably affecting corporate performance and stock returns. But, they also build up substantial inflationary pressure in the economy which could negatively impact stock returns. [Chaudhuri and Smiles (2004), Mukherjee and Naka (1995)]. BRICS is an acronym for a group of five prominent emerging and developing economies of Brazil, Russia, India, China and South Africa. They have big, fast-growing economies and now command significant political and economic influence at global level. In 2014, these five BRICS economies jointly represented about 40% of world’s population and 20% of world’s nominal GDP. Our objective in this paper is to examine both the short term and long term dynamic relationship between aggregate stock returns in BRICS and their major macroeconomic factors, i.e., GDP, Inflation, Interest Rate, Exchange Rate and Money Supply. We also probe whether any of the macroeconomic variables are useful in predicting BRICS stock returns. We also investigate for the presence of any causal (lead-lag) relationship between BRICS stock returns and major macroeconomic variables in the short and long term. We study this relationship for these countries collectively using Panel data. The remaining paper is structured as follows: Section 2 provides review of literature. Section 3 explains the data and methodology. Section 4 elucidates the empirical results. Section 5 provides the conclusions and implications of the study.

Review of Literature A plethora of studies have examined the relationship between macroeconomic variables and stock returns in developed markets of US, UK and other European markets. However the literature on such a relationship in emerging markets has been limited and is growing only recently especially in the context of India. Fama (1981, 1990) reported a strong relationship is present between stock returns and macroeconomic variables, notably, inflation, national output and industrial production. Stock returns are determined by forecasts of more relevant real variables and negative stock returnsinflation relations are induced by negative relationships between inflation and real activity. Chen et al. (1986) were the first to explore a set of economic state variables as systematic influences on stock market returns and have examined their influence on asset pricing. Macro-economic variables that systematically affect stock market returns are- spread between long and short interest rates, expected and unexpected inflation, industrial production, and the spread between high- and low-grade bonds. Chang and Pinegar (1989) affirmed that there exists a close relationship between stock market performance and the domestic economic activity. They also report unidirectional Granger causality running from large firms' stocks returns to future growth rates in industrial production at least six months in advance. Mukherjee and Naka (1995) suggested that cointegration relation existed and positive relationship was found between the Japanese industrial production and stock return.


Maysami and Koh (2000) reported that changes in Singapore’s stock market levels do form a cointegrating relationship with changes in price levels, money supply, short- and long-term interest rates, and exchange rates. Abugri (2008) found that Interest rates and exchange rates are significant in three out of the four Latin American markets examined. The performance of money supply and industrial production is generally weak. Gay (2008) reported that , though not significant, but the relationship between exchange rates and stock prices was positive while, the relationship between respective stock market prices and monthly oil prices was negative but insignificant. Agarwalla and Tuteja (2008) revealed causality running from economic growth proxied by industrial production to share price index and not the other way round which shows that stock markets in India are still demand driven and industry led. Singh (2010) indicated that IIP is the only variable having bilateral causal relationship with BSE Sensex. WPI is having unilateral causality with BSE Sensex. Hsing (2011) finds that South Africa’s stock market index is positively influenced by the growth rate of real GDP, the ratio of the money supply to GDP and the U.S. stock market index and negatively affected by the ratio of the government deficit to GDP, the domestic real interest rate, the nominal effective exchange rate, the domestic inflation rate, and the U.S. government bond yield. Dasgupta (2012) found one cointegration vector and long-run relationships between BSE SENSEX with index of industrial production and call money rate. They further found no short-run unilateral or bilateral causal relationships between BSE SENSEX with the macroeconomic variables. Tripathi and Seth (2014) conveyed a significant correlation among stock market indicators and macroeconomic factors and identified Inflation, Interest rate and Exchange rate as three principal factors through Factor analysis. They also reported presence of five co-integrating relationships between stock market and macro-economic variables. Tripathi and Kumar (2015 a & b) used granger causality and panel cointegration on BRICS market to conclude that while inflation rate may be significantly related to stock returns in the short run, they do not seem to move together in the long run. Tripathi and Kumar (2015 c) used ARDL model and reported that Stock returns generally lead rather than follow GDP and Inflation. Also, they find significant negative relationship of stock returns with Interest Rate, Exchange Rate and Oil Prices and a positive relationship with money supply. Overall, it can be said that, the studies have comprehensively analysed the developed markets and arrived at some common ground. But for developing markets, the consensus is largely lacking both due to varying results for most macroeconomic variables and paucity of research.

Data and Methodology Data The period of present study is 1995: Q1 to 2014: Q4. Frequency of all data is quarterly. The data comprises of macroeconomic variables and stock indices values for all BRICS nations. We have considered five prominent macroeconomic variables, i.e., GDP, Inflation, Interest Rate, Exchange Rate and Money Supply. The operational definitions, time period of availability, source and symbol of each macroeconomic variable for each country is provided


in Table 1. Using individual country data, we have constructed a panel data of BRICS stock index and macroeconomic variables. Table 1: Data Description (Macroeconomic Variables) S. No.

Country

Macroeconomic Variables

Operational Definition

1.

Brazil

GDP

Fixed PPP, 2005 Prices

2.

Brazil

Inflation

Consumer Price Index, Base 2010

3.

Brazil

Interest Rate

Brazil Selic Target Rate

4.

Brazil

Exchange Rate

1 USD in Brazilian Real(BRL)

5.

Brazil

Money Supply

Broad Money Supply (M3)

6.

Russia

GDP


7.

Russia

Inflation


8.

Russia

Interest Rate

Russia Refinancing Rate

9.

Russia

Exchange Rate

1 USD in Russian Ruble (RUB)

10.

Russia

Money Supply

Narrow Money Supply (M1)

11.

India

GDP


12.

India

Inflation


13.

India

Interest Rate

Weighted Average Call Money Rates

14.

India

Exchange Rate

1 USD in Indian Rupees

15.

India

Money Supply

Broad Money (M3)

16.

China

GDP

GDP at current prices

17.

China

Inflation


18.

China

Interest Rate

1 Year Benchmark Lending

Time Period 1996: Q1 -2014: Q3 1995: Q1 -2014: Q4 1999: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q3 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4 2002: Q2 -2014: Q4 1996: Q2 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q3 1995: Q1 -2014: Q4 1996: Q2

Source

Symbol

OECD

BGDP

OECD

BINF

Bloomberg

BIR

Bloomberg

BER

Central Bank of Brazil

BMS

OECD

RGDP

OECD

RINF

Bloomberg

RIR

Bloomberg

RER

Bloomberg

RMS

OECD

IGDP

OECD

IINF

RBI

IIR

RBI

IER

RBI

IMS

National Bureau of Statistics

CGDP

OECD

CINF

Bloomberg

CIR

Relationship between Macroeconomic Factors and Aggregate Stock Returns in BRICS Stock Markets – A Panel Data Analysis Rates

19.

China

Exchange Rate

1 USD in Chinese Yuan (CNY)

20.

China

Money Supply

Money Supply (M2)

21.

South Africa

GDP


22.

South Africa

Inflation


23.

South Africa

Interest Rate

Average Repo Rate

24.

South Africa

Exchange Rate

1 USD in South African Rand

25.

South Africa

Money Supply

Money Supply (M2)

26.

Panel

GDP

-

27.

Panel

Inflation

-

28.

Panel

Interest Rate

-

29.

Panel

Exchange Rate

-

30.

Panel

Money Supply

-

-2014: Q4 1995: Q1 -2014: Q4 1996: Q1 -2014: Q4 2002: Q1 -2014: Q4 2002: Q1 -2014: Q4 2002: Q1 -2014: Q4 2002: Q1 -2014: Q4 2002: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4 1995: Q1 -2014: Q4

Bloomberg

CER

Bloomberg

CMS

OECD

SAGDP

OECD

SAINF

Bloomberg

SAIR

Bloomberg

SAER

Bloomberg

SAMS

-

PGDP

-

PINF

-

PIR

-

PER

-

PMS

The detailed description of stock market variables of each country is given in Table 2. Table 2: Data Description (Stock Market Variables) S.No.

Country

1.

Brazil

2.

Russia

3.

India

4.

China

5. 6.

South Africa Panel (Index)

Stock Exchange Sao Paulo Stock Exchange Moscow Stock Exchange Bombay Stock Exchange Shanghai Stock Exchange Johannesburg Stock Exchange

Stock Index Ibovespa RTSI INDEX BSE SENSEX Shanghai SE Composite FTSE-JSE All Share Index -

Time Period 1995: Q1 to 2014: Q4 1995: Q3 to 2014: Q4 1995: Q1 to 2014: Q4 1995: Q1 to 2014: Q4 2002: Q1 to 2014: Q4 1995: Q1 2014: Q4

Source Yahoo Finance Yahoo Finance Yahoo Finance Yahoo Finance Yahoo Finance

Symbol

-

PINDEX

BINDEX RINDEX IINDEX CINDEX SAINDEX


Methodology 1. Panel Unit Root Test If the mean, variance and auto-covariance of a time series data is time invariant, it is said to be stationary. Following Panel unit root tests have been applied. a. Levin, Lin, and Chu (LLC) Test IHS (2013): “Levin, Lin, and Chu test assume that there is a common unit root process so that is identical across cross-sections. LLC consider the following basic ADF specification: 𝑝𝑖 ∆𝑦𝑖𝑡 = 𝛼𝑦𝑖𝑡−1 + ∑𝑗=1 𝛽𝑖𝑗 ∆𝑦𝑖𝑡−𝑗 + 𝑋′𝑖𝑡 𝛿 + 𝜖𝑖𝑡 (1) Where, we assume a common α = ρ - 1, but allow the lag order for the difference terms 𝑝𝑖 , to vary across cross-sections. The null and alternative hypotheses for the tests may be written as: 𝐻0 : 𝛼 = 0 (unit root) and 𝐻1 : 𝛼 < 0 (no unit root).” (p. 488). b. Im, Pesaran and Shin (IPS) Test IHS (2013): “The Im, Pesaran, and Shim test allow for individual unit root processes so that 𝑝𝑖 may vary across cross-sections. IPS begins by specifying a separate ADF regression for 𝑝𝑖 each cross section: ∆𝑦𝑖𝑡 = 𝛼𝑦𝑖𝑡−1 + ∑𝑗=1 𝛽𝑖𝑗 ∆𝑦𝑖𝑡−𝑗 + 𝑋′𝑖𝑡 𝛿 + 𝜖𝑖𝑡 (2) The null hypothesis may be written as, 𝐻0 ∶ 𝛼𝑖 = 0, 𝑓𝑜𝑟 𝑎𝑙𝑙 𝑖. 𝛼 = 0 𝑓𝑜𝑟 𝑖 = 1, 2, … , 𝑁1 While the alternative hypothesis is given by: 𝐻1 ∶ { 𝑖 ” 𝛼𝑖 < 0 𝑓𝑜𝑟 𝑖 = 𝑁 + 1, 𝑁 + 2, … , 𝑁 (p. 491-492). c. Fisher-ADF and Fisher-PP Test IHS (2013): “The Fisher-ADF and PP tests allow for individual unit root processes so that 𝑝𝑖 may vary across cross-sections. The tests are characterized by the combining of individual unit root tests to derive a panel-specific result. The null and alternate hypotheses are same as for IPS” (p. 492-493).

2. Panel Stacked Granger Causality Test IHS (2013): “In general, the bivariate regressions in a panel data context take the form: 𝑦𝑖,𝑡 = 𝛼0,𝑖 + 𝛼1,𝑖 𝑦𝑖,𝑡−1 + ⋯ + 𝛼𝑙,𝑖 𝑦𝑖,𝑡−𝑙 + 𝛽1,𝑖 𝑥𝑖,𝑡−1 + ⋯ + 𝛽𝑙,𝑖 𝑥𝑖,𝑡−𝑙 + 𝜖𝑖,𝑡 ……..... (3), and 𝑥𝑖,𝑡 = 𝛼0,𝑖 + 𝛼1,𝑖 𝑥𝑖,𝑡−1 + ⋯ + 𝛼𝑙,𝑖 𝑥𝑖,𝑡−𝑙 + 𝛽1,𝑖 𝑦𝑖,𝑡−1 + ⋯ + 𝛽𝑙,𝑖 𝑦𝑖,𝑡−𝑙 + 𝜖𝑖,𝑡 ……… (4). Where t denotes the time period dimension of the panel, and i denotes the cross-sectional dimension. This test treats the panel data as one large stacked set of data, and then performs the Granger Causality test in the standard way, with the exception of not letting data from one cross-section enter the lagged values of data from the next cross-section. This method assumes that all coefficients are same across all cross-sections, i.e.: 𝛼0,𝑖 = 𝛼0,𝑗 , 𝛼1,𝑖 = 𝛼1,𝑗 , … , 𝛼𝑙,𝑖 = 𝛼𝑙,𝑗 , ∀ 𝑖, 𝑗 and 𝛽1,𝑖 = 𝛽1,𝑗 , … , 𝛽𝑙,𝑖 = 𝛽𝑙,𝑗 , ∀ 𝑖, 𝑗” (5) Granger causality test establishes short run causality if we take stationary values. “Causality tests by the level Vector Auto Regression (VAR) (non-stationary) can complement the result of the cointegration tests in terms of long-run information” [Worthington & Higgs, 2007]. So,


non-stationary level time series data of variables have been used to determine long run causality. Optimal lag length for conducting granger causality test (both short and long run) has been determined as per the Akaike Information Criterion (AIC) within the VAR framework. 3. Pedroni’s Panel Cointegration Test (Engle-Granger Based) Pedroni Cointegration test (Engle-Granger Based) has been applied on panel data of stock index values and macroeconomic variables to determine whether a long term cointegrating or equilibrium relationship exists between stock return and macroeconomic variables for BRICS stock markets when taken together as a panel. The Engle-Granger (1987) cointegration test is based on an examination of the residuals of a spurious regression performed using I(1) variables. Pedroni proposes several tests for cointegration that allow for heterogeneous intercepts and trend coefficients across cross-sections. Consider the following regression: yit = αi + δi t + β1i x1i,t + β2i x2i,t + ⋯ + βMi xMi,t + ei,t …………………….. (6) for t = 1,…….,T; i = 1,……, N; m = 1, ……., M; where y and x are assumed to be integrated of order one, e.g. I(1). The parameters αi and δi are individual and trend effects which may be set to zero if desired. Under the null hypothesis of no cointegration, the residuals ei,t will be I(1). The general approach is to obtain residuals from Equation 1 and then to test whether residuals are I (1) by running the auxiliary regression, ei,t = ρi eit−1 + uit ………………………………………………………… (7) for each cross-section. Pedroni describes various methods of constructing statistics for testing for null hypothesis of no cointegration (ρi = 1 ). There are two alternative hypotheses: the homogenous alternative, (ρi = ρ) < 1 for all i (which Pedroni terms the within-dimension test or panel statistics test), and the heterogeneous alternative, ρi < 1 for all i (also referred to as the between-dimension or group statistics test). The Pedroni panel cointegration statistic ℵN,T is constructed from the residuals from Equation 7. A total of eleven statistics with varying degree of properties (size and power for different N and T) are generated. Pedroni shows that the standardized statistic is asymptotically normally distributed, ℵN,T − μ√N

→ N(0, 1) ………………………………………………………………….. (8) Where μ and v are Monte Carlo generated adjustment terms. √v

4. Panel ARDL Model The Autoregressive Distributed Lag (ARDL) approach was introduced by Pesaran et al. (1996). ARDL model has been used here for the analysis of both short-run dynamic and long run relationship between Stock returns and Macroeconomic variables in BRICS markets. An autoregressive distributed lag model is considered as ARDL (1, 1) model: yt= μ + α1yt-1 + β0xt + β1xt-1 + ut. ………………………… (9) Where yt and xt are stationary variables, and ut is a white noise. Our ARDL model regresses panel stock index variable on their own lagged values; on stationary (short run) contemporary and lagged values of panel macroeconomic variables and on non-stationary (long run) values of panel macroeconomic variables. Thus, while the stationary contemporaneous and lagged values will determine the short run relationship between macroeconomic variables and stock returns, the non-stationary ones will establish the long run relationship. Optimal AIC Lags for Panel ARDL model is 5 in Total and Post Crisis Periods and 4 in Pre Crisis Period.


Empirical Results and Discussion 1. Panel Unit Root Test Results We applied four tests to check whether our panel data is stationary or not. These tests are Levin, Lin& Chu Test, Im, Pesaran & Shin Test, ADF fisher test and PP-Fisher Test. The results are presented in Table 3 and 4 for at level and first differenced series. The results reveal that all the panel series are non stationary at level in all the three time periods and their log of first difference is stationary in all the three time periods. This shows that our data series are I (1) and hence can be used in further analysis without worrying about emergence of any spurious relationship. Table 3: Panel Unit Root Tests Results (At Level) A. Panel Unit Root Test Results (Total Period – At Level) Panel Levin, Lin & Im, Pesaran & ADF-Fisher PP-Fisher Variables Chu Shin t-stat. Prob. WProb. ChiProb. ChiProb. Stat. square square INDEX 0.75 0.77 1.05 0.85 7.79 0.65 6.55 0.77 GDP 2.64 0.99 4.51 0.99 2.63 0.98 18.63 0.05 INF 5.71 0.99 8.79 0.99 0.03 0.99 0.03 0.99 IR -4.93* 0.00 -5.48* 0.00 56.07* 0.00 70.74* 0.00 ER 1.23 0.88 1.91 0.97 3.52 0.97 3.57 0.97 MS 14.79 0.99 15.42 0.99 0.004 0.99 0.001 0.99 *Denotes significant at α = 0.05. B. Panel Unit Root Tests Results (Pre Crisis Period - At Level) Panel Levin, Lin & Im, Pesaran & ADF-Fisher Variables Chu Shin t-stat. Prob. WProb. ChiProb. Stat. square INDEX 7.83 0.99 7.47 0.99 0.07 0.99 GDP 7.20 0.99 7.07 0.99 12.94 0.23 INF 0.60 0.73 2.95 0.99 1.69 0.99 IR -4.10* 0.00 -4.03* 0.00 37.97* 0.00 ER 1.25 0.90 2.55 0.99 10.29 0.42 MS 15.01 0.99 16.95 0.99 0.00 0.99 *Denotes significant at α = 0.05. C. Panel Unit Root Tests Results (Post Crisis Period - At Level) Panel Levin, Lin & Im, Pesaran & ADF-Fisher Variables Chu Shin t-stat. Prob. WProb. ChiProb. Stat. square INDEX 1.06 0.86 -0.62 0.27 15.17 0.13 GDP 1.26 0.90 0.97 0.84 15.51 0.12 INF 1.09 0.86 3.95 0.99 0.88 0.99 IR -2.29* 0.01 -2.17* 0.02 20.15* 0.03 ER 0.05 0.52 1.99 0.98 4.40 0.93 MS 3.17 0.99 6.29 0.99 0.07 0.99 *Denotes significant at α = 0.05.

PP-Fisher Chisquare 0.03 20.53 2.86 52.16* 8.80 0.00

Prob. 0.99 0.03 0.99 0.00 0.55 0.99

PP-Fisher Chisquare 24.48* 18.05 1.00 8.74 7.45 0.01

Prob. 0.01 0.05 0.99 0.56 0.68 0.99


Table 4: Panel Unit Root Tests Results (at Log of First Difference) A. Panel Unit Root Tests Results (Total Period – Log of First Difference) Panel Levin, Lin & Chu Im, Pesaran & Shin ADF-Fisher PP-Fisher Variables t-stat. Prob. W-Stat. Prob. ChiProb. ChiProb. square square INDEX -11.77* 0.00 -11.05* 0.00 120.19* 0.00 143.11* 0.00 GDP -7.73* 0.00 -11.55* 0.00 84.03* 0.00 114.58* 0.00 INF -9.68* 0.00 11.91* 0.00 121.98* 0.00 120.58* 0.00 IR -2.63* 0.00 -10.24* 0.00 100.07* 0.00 141.86* 0.00 ER -7.21* 0.00 -9.48* 0.00 101.58* 0.00 146.85* 0.00 MS -9.72* 0.00 -11.02* 0.00 121.18* 0.00 140.90* 0.00 *Denotes significant at α = 0.05. B. Panel Unit Root Tests Results (Pre Crisis Period – Log of First Difference) Panel Levin, Lin & Im, Pesaran & ADF-Fisher PP-Fisher Variables Chu Shin t-stat. Prob. WProb. ChiProb. ChiProb. Stat. square square INDEX -7.19* 0.00 -7.09* 0.00 71.11* 0.00 102.92* 0.00 GDP -5.02* 0.00 -7.73* 0.00 77.03* 0.00 89.74* 0.00 INF -7.58* 0.00 -8.92* 0.00 91.15* 0.00 76.26* 0.00 IR 0.86* 0.80 -6.36* 0.00 60.27* 0.00 112.87* 0.00 ER -2.64* 0.00 -5.25* 0.00 49.97* 0.00 108.41* 0.00 MS -6.75* 0.00 -7.54* 0.00 77.91* 0.00 142.62* 0.00 *Denotes significant at α = 0.05. C. Panel Unit Root Tests Results (Post Crisis Period – Log of First Difference) Panel Levin, Lin & Im, Pesaran & ADF-Fisher PP-Fisher Variables Chu Shin t-stat. Prob. WProb. ChiProb. ChiProb. Stat. square square INDEX -5.85* 0.00 -5.66* 0.00 49.97* 0.00 52.96* 0.00 GDP -3.74* 0.00 -6.11* 0.00 54.87* 0.00 74.73* 0.00 INF -8.81* 0.00 -8.43* 0.00 77.98* 0.00 67.14* 0.00 IR -2.68* 0.00 -3.40* 0.00 28.84* 0.00 42.64* 0.00 ER -2.89* 0.00 -5.28* 0.00 46.92* 0.00 60.91* 0.00 MS -6.04* 0.00 -6.75* 0.00 61.34* 0.00 89.33* 0.00 *Denotes significant at α = 0.05.

2. Panel Stacked Granger Causality Results (a) Short Run Panel Causality Results The panel data short run Granger Causality results presented in Table 5 show unidirectional causality from stock return to four macroeconomic factors viz. GDP growth rate, Inflation rate, changes in exchange rate and money supply in the total period. On the other hand, interest rate granger causes stock return in total period. In the pre crisis period, there is bidirectional causality between stock returns and inflation rate and unidirectional causality


from stock return to GDP growth rate and changes in exchange rate. In the post crisis period, there is bi directional causality between stock returns and interest rate. We also find unidirectional causality from stock returns to GDP growth rate and changes in money supply in this period. Table 5: Short Run Stacked Panel Causality Test Results Total Period FProb. Statistic

Null Hypothesis

DLOG(PGDP) does not Granger 0.50 Cause DLOG(PINDEX) DLOG(PINDEX) does not Granger 11.35* Cause DLOG(PGDP) DLOG(PINF) does not Granger 1.95 Cause DLOG(PINDEX) DLOG(PINDEX) does not Granger 9.10* Cause DLOG(PINF) DLOG(PIR) does not Granger Cause 2.50* DLOG(PINDEX) DLOG(PINDEX) does not Granger 0.56 Cause DLOG(PIR) DLOG(PER) does not Granger 2.07 Cause DLOG(PINDEX) DLOG(PINDEX) does not Granger 10.00* Cause DLOG(PER) DLOG(PMS) does not Granger 1.05 Cause DLOG(PINDEX) DLOG(PINDEX) does not Granger 6.09* Cause DLOG(PMS) Note: * Denotes Significant at 5% Level.

Pre crisis FProb. Statistic

Post Crisis FProb. Statistic

0.77

0.61

0.66

0.70

0.60

0.00

7.77*

0.00

22.70*

0.00

0.09

3.12*

0.02

1.51

0.20

0.00

11.94*

0.00

1.04

0.39

0.03

1.78

0.14

4.37*

0.00

0.73

1.31

0.27

2.66*

0.04

0.07

1.77

0.14

0.50

0.74

0.00

8.13*

0.00

2.03

0.09

0.39

0.38

0.82

1.27

0.29

0.00

1.34

0.26

6.65*

0.00

(b) Long Run Panel Causality Results Long run Stacked Panel causality test results as presented in Table 6 show that in the total period stock prices granger causes GDP growth rate in the long run. No long run causal relationship existed in pre crisis period. However post crisis, stock market is granger causing GDP and Interest rates while there is unidirectional causality from money supply to stock prices in the long run. Table 6: Long Run Stacked Panel Causality Test Results

Null Hypothesis PGDP does not Granger Cause PINDEX PINDEX does not Granger Cause PGDP PINF does not Granger Cause PINDEX

Total Period FProb. Statistic

Pre Crisis FProb. Statistic

Post Crisis FProb. Statistic

0.31

0.90

0.00

0.95

0.08

0.99

10.57*

0.00

0.52

0.47

6.29*

0.00

0.43

0.83

0.60

0.44

0.30

0.91


PINDEX does not Granger Cause 0.53 PINF PIR does not Granger Cause 0.11 PINDEX PINDEX does not Granger Cause 0.18 PIR PER does not Granger Cause 0.27 PINDEX PINDEX does not Granger Cause 0.19 PER PMS does not Granger Cause 0.66 PINDEX PINDEX does not Granger Cause 2.12 PMS Note: * Denotes Significant at 5% Level.

0.76

1.70

0.19

0.23

0.95

0.99

0.05

0.82

1.69

0.14

0.97

0.00

0.96

3.63*

0.00

0.93

0.00

0.98

0.49

0.78

0.96

3.06

0.08

0.05

1.00

0.65

1.03

0.31

4.39*

0.00

0.06

0.58

0.45

1.10

0.37

3. Pedroni Panel Cointegration Results The results regarding panel data are provided in Tables 7 to 11. These tables show that there is cointegrating relationship between stock prices and GDP in total period and post crisis period. There is cointegrating relationship between stock prices and inflation as well as between stock prices and money supply in post crisis period. The panel data shows that there is absolutely no cointegration of stock prices with interest rate and exchange rate. (I) GDP Table 7: Pedroni Panel Cointegration Test Results (GDP) Pedroni Panel Statistic

Total Period Simple Stat. Prob.

Weighted Stat. Prob.

Pre Crisis Period Simple Weighted Stat. Prob. Stat. Prob.

Post Crisis Period Simple Weighted Stat. Prob. Stat. Prob.

4.75*

0.00

4.29*

0.00

-0.28

0.61

-0.29

0.62

3.00*

0.00

2.76*

0.00

Panel rhoStatistic

-2.93*

0.00

-3.23*

0.00

2.05

0.98

0.25

0.60

-1.38

0.08

-2.04*

0.02

Panel PPStatistic

-1.95*

0.03

-2.17*

0.02

3.28

1.00

0.86

0.80

-1.60

0.05

-2.84*

0.00

Panel ADFStatistic

-2.12*

0.02

-2.79*

0.00

3.14

1.00

1.29

0.90

-2.36*

0.01

-3.32*

0.00

Group rhoStatistic

-1.78*

0.04

NA

NA

0.40

0.66

NA

NA

-0.77

0.22

NA

NA

Group PPStatistic

-1.68

0.05

NA

NA

2.44

0.99

NA

NA

-2.95*

0.00

NA

NA

NA

2.73

1.00

NA

NA

-3.50*

0.00

NA

NA

Panel Statistic

v-

Group ADF-2.40* 0.01 NA Statistic * Denotes Significant at 5% level.


(II) Inflation Table 8: Pedroni Panel Cointegration Test Results (Inflation) Pedroni Panel Statistic





1.19

0.12

1.90*

0.03

0.60

0.27

0.86

0.20

2.03*

0.02

2.45*

0.01

Panel rhoStatistic

0.11

0.54

-0.85

0.20

1.54

0.94

1.39

0.92

-1.36

0.09

-1.61

0.05

Panel PPStatistic

0.29

0.62

-0.62

0.27

2.01

0.98

2.09

0.98

-1.46

0.07

-2.04*

0.02

Panel ADFStatistic

-0.18

0.43

-1.41

0.08

2.27

0.99

2.42

0.99

-2.40*

0.01

-2.60*

0.00

Group rhoStatistic

0.21

0.58

NA

NA

1.86

0.97

NA

NA

-0.57

0.28

NA

NA

Group PPStatistic

-0.01

0.50

NA

NA

2.73

1.00

NA

NA

-1.85*

0.03

NA

NA

NA

2.65

1.00

NA

NA

-2.40*

0.01

NA

NA

Panel Statistic

v-

Group ADF-0.86 0.19 NA Statistic * Denotes Significant at 5% level.

(III) Interest Rate Table 9: Pedroni Panel Cointegration Test Results (Interest Rate) Pedroni Panel Statistic





-0.95

0.83

-0.64

0.74

-0.95

0.83

-0.48

0.69

0.95

0.17

0.86

0.20

Panel rhoStatistic

-0.56

0.29

-0.18

0.43

0.95

0.83

2.37

0.99

-0.01

0.50

0.15

0.56

Panel PPStatistic

-1.24

0.11

-0.41

0.34

1.59

0.94

3.73

1.00

0.14

0.56

0.21

0.58

Panel ADFStatistic

0.14

0.56

0.02

0.51

4.40

1.00

4.91

1.00

-0.06

0.48

0.26

0.60

Group rhoStatistic

1.07

0.86

NA

NA

3.31

1.00

NA

NA

0.25

0.60

NA

NA

Group PPStatistic

-0.02

0.49

NA

NA

5.70

1.00

NA

NA

-0.28

0.39

NA

NA

NA

5.91

1.00

NA

NA

0.16

0.56

NA

NA

Panel Statistic

v-

Group ADF0.36 0.64 NA Statistic * Denotes Significant at 5% level.


(IV) Exchange Rate Table 10: Pedroni Panel Cointegration Test Results (Exchange Rate) Pedroni Panel Statistic





-1.39

0.92

-0.78

0.78

-1.13

0.87

-0.76

0.78

0.88

0.19

1.25

0.11

Panel rhoStatistic

1.16

0.88

0.21

0.58

3.67

1.00

3.17

1.00

-1.37

0.08

-0.67

0.25

Panel PPStatistic

0.58

0.72

-0.11

0.45

6.79

1.00

5.66

1.00

-2.80*

0.00

-1.91*

0.03

Panel ADFStatistic

0.40

0.66

-0.40

0.34

5.89

1.00

5.13

1.00

-3.20*

0.00

-2.60*

0.00

Group rhoStatistic

0.44

0.67

NA

NA

3.47

1.00

NA

NA

0.32

0.62

NA

NA

Group PPStatistic

-0.12

0.45

NA

NA

6.91

1.00

NA

NA

-1.39

0.08

NA

NA

NA

6.01

1.00

NA

NA

-2.27*

0.01

NA

NA

Panel Statistic

v-


(V) Money Supply Table 11: Pedroni Panel Cointegration Test Results (Money Supply) Pedroni Panel Statistic





1.44

0.08

3.01*

0.00

0.65

0.26

1.27

0.10

2.24*

0.01

2.68*

0.00

Panel rhoStatistic

0.33

0.63

-1.20

0.11

0.08

0.53

-0.68

0.25

-1.55

0.06

-1.87*

0.03

Panel PPStatistic

0.91

0.82

-0.61

0.27

0.30

0.62

-0.36

0.36

-1.61

0.05

-2.24*

0.01

Panel ADFStatistic

0.55

0.71

-1.14

0.13

-0.19

0.43

-1.31

0.10

-2.56*

0.01

-2.93*

0.00

Group rhoStatistic

-0.46

0.32

NA

NA

0.31

0.62

NA

NA

-0.81

0.21

NA

NA

Group PPStatistic

-0.36

0.36

NA

NA

0.18

0.57

NA

NA

-2.16*

0.02

NA

NA

NA

-0.89

0.19

NA

NA

-2.76*

0.00

NA

NA

Panel Statistic

v-



4. Panel ARDL Model Results Finally, we run the Panel ARDL models to see the short and long run contemporary and leadlag relationships between stock returns and macroeconomic variables of BRICS as one collective group. We find that while current stock returns are negatively linked to rate of change in exchange rate and money supply; they are positively linked to their own lagged values. In pre crisis period, rate of change in money supply significantly explains stock returns while in post crisis period, inflation rate, interest rate and rate of change in exchange rate and money supply negatively affects BRICS panel stock returns (Table 12). Panel ARDL Models have explanatory power ranging from 28% in total period to 62% in post crisis periods. Also, while the Total Period & Post Crisis ARDL models are significant at 5%, the Pre Crisis ARDL model is significant at 10% (Table 13). Table 12: Panel ARDL Model Results- Coefficients of Model Variable C DLOG(PINDEX(-1)) DLOG(PINDEX(-2)) DLOG(PINDEX(-3)) DLOG(PINDEX(-4)) DLOG(PINDEX(-5)) DLOG(PGDP) DLOG(PGDP(-1)) DLOG(PGDP(-2)) DLOG(PGDP(-3)) DLOG(PGDP(-4)) DLOG(PGDP(-5)) DLOG(PINF) DLOG(PINF(-1)) DLOG(PINF(-2)) DLOG(PINF(-3)) DLOG(PINF(-4)) DLOG(PINF(-5)) DLOG(PIR) DLOG(PIR(-1)) DLOG(PIR(-2)) DLOG(PIR(-3)) DLOG(PIR(-4)) DLOG(PIR(-5)) DLOG(PER) DLOG(PER(-1)) DLOG(PER(-2)) DLOG(PER(-3)) DLOG(PER(-4)) DLOG(PER(-5)) DLOG(PMS)

Total Period 0.18 0.20* -0.04 0.05 -0.09 0.00 0.11 -0.31 -0.06 -0.10 -0.24 0.22 -0.68 -1.53 0.56 -0.67 0.72 0.79 -0.03 -0.04 -0.09 0.07 -0.03 -0.06 -0.65* 0.34* 0.16 -0.03 0.13 0.19 0.00

Pre Crisis Period -0.27 0.00 -0.04 -0.02 0.02 NA 1.47 0.80 0.83 0.76 -0.69 NA -0.35 0.32 1.82 0.96 1.02 NA -0.05 -0.14 0.03 -0.02 -0.13 NA -0.37 -0.03 0.15 -0.19 -0.04 NA -0.25

Post Crisis Period 0.90 -0.16 -0.29* -0.07 -0.11 -0.10 0.19 -0.34 0.33 0.37 0.18 0.72 -0.13 -1.11 -2.53* -0.47 -0.13 0.64 0.02 0.12 -0.25* -0.02 -0.15 -0.08 -0.47* 0.13 0.22 0.17 0.18 0.23 0.18


DLOG(PMS(-1)) -0.34 -0.94 DLOG(PMS(-2)) 0.81* 0.10 DLOG(PMS(-3)) -0.36 -0.60 DLOG(PMS(-4)) -0.75* -1.07* DLOG(PMS(-5)) 0.31 NA LOG(PGDP) 0.00 -0.05 LOG(PINF) -0.04 0.03 LOG(PIR) -0.01 0.02 LOG(PER) 0.00 0.03 LOG(PMS) 0.00 -0.03 * Denotes significant at 5% level. Values are regression coefficients.

0.38 -1.06* -0.51 -0.04 -0.24 0.00 -0.26* 0.01 0.03 -0.01

Table 13: Panel ARDL Model Summary Panel ARDL Model F-Stat. Total Period 2.33* Pre Crisis Period 1.46** Post Crisis Period 2.92* * Significant at 5% level. ** Significant at 10% level.

R2 0.28 0.32 0.62

Probability 0.00 0.08 0.00

Figure 1-3 presents graphic representation of actual, fitted & residuals of Panel ARDL Models in the total period, pre crisis period and the post crisis periods respectively. Figure 1: Panel ARDL Model Graph (Total Period) 0.50 0.25 0.00 .6

-0.25

.4

-0.50

.2

-0.75

.0 -1.00

-.2 -.4 -.6 -.8 Residual

Actual

Fitted


Figure 2: Panel ARDL Model Graph (Pre Crisis Period) .6 .4 .2 .0

.4

-.2 .2 -.4 .0 -.2 -.4 Residual

Actual

Fitted

Figure 3: Panel ARDL Model Graph (Post Crisis Period) .6 .4 .2 .0 .2 -.2 .1 -.4 .0 -.1 -.2 -.3 Residual

Actual

Fitted


Conclusion and Implications This paper examines the relationship between select macroeconomic factors (i.e., GDP, Inflation, Interest Rate, Exchange Rate and Money Supply) and aggregate stock returns in emerging markets constituting the BRICS block over the period 1995 to 2014 using quarterly panel data. This relationship is also examined during two sub periods viz., a Pre Crisis period (1995:Q1 to 2007:Q2) and a Post Crisis Period (2007:Q3 to 2014:Q4). Robust econometric tests like Panel Granger Causality Test, Pedroni’s Panel Cointegration Test and Panel Auto Regressive Distributed Lag (ARDL) Model has been used. We find that primarily in short run there is unidirectional causality running from stock returns to GDP growth rate, inflation rate, rate of change in exchange rate and money supply. The results are almost similar in pre and post crisis periods, except that in the pre crisis period, there is bidirectional causality between stock returns and inflation, while in the post crisis period it disappears. Long run panel causality results reveals unidirectional causality from stock returns to GDP growth rate in total and post crisis periods. However in pre crisis period, there was no long run causal relationship. Pedroni’s panel cointegration test shows that stock indices are cointegrated with GDP in total period and with GDP, inflation and money supply in post crisis period. Panel ARDL models have explanatory power ranging from 28% in total period to 62% in post crisis period. We find that while current stock returns are negatively linked to rate of change in exchange rate and money supply; they are positively linked to their own lagged values. In pre crisis period, rate of change in money supply significantly explains stock returns while in post crisis period, inflation rate, interest rate and rate of change in exchange rate and money supply negatively affects BRICS panel stock returns. Results indicate that Stock Markets already discount the GDP and Inflation data and hence stock prices tend to lead rather than follow GDP and Inflation. However, Money Supply leads Stock Prices. The causal, led-lag & Cointegrating relationships have significantly increased in the Post crisis period indicating the impact of Global Financial Crisis in deepening this relationship. These findings, besides augmenting the empirical literature and knowledge domain on the topic, have significant implications for policy makers, regulators, researchers and investing community in emerging markets. Policy makers and regulators should watch out for impact of fluctuations in exchange rate, interest rate, money supply and oil prices on volatility in their stock markets. The regulators need to ensure that financial sector reforms agenda consciously considers interlinkages between stock markets and real economy. Investor can search for presence of exploitable arbitrage opportunities in BRICS markets to earn above normal returns on the basis of these variables especially GDP and Money Supply.

Acknowledgement: This paper is based on a comprehensive study undertaken under UGC Major Research Project (M.R.P.) titled “Relationship between Macroeconomic Factors and aggregate stock returns in Emerging Markets- An Empirical Study of BICS stock Markets” of which Dr. Vanita Tripathi is the Principal Investigator (P.I.). The authors gratefully acknowledge the financial support provided by University Grants Commission, New Delhi for this study.


References Agarwalla, R.K., & Tuteja, S.K. (2008). Share Prices and Macroeconomic Variables in India: An Approach to Investigate the Relationship between Stock Markets and Economic Growth. Journal of Management Research, Vol. 8, No. 3, pp. 136-146. Asprem, M. (1989). Stock prices, asset portfolios and macroeconomic variables in ten European countries. Journal of Banking and Finance, 13(4/5), 589-612. Benjamin, A. (2008). Empirical relationship between macroeconomic volatility and stock returns: Evidence from Latin American markets. International Review of Financial Analysis, Vol. 17, pp.396–410. Chang, E. C., & Pinegar, J.M. (1989). Seasonal Fluctuations in Industrial Production and Stock Market Seasonals. Journal of Financial and Quantitative Analysis, 59-74. Chaudhuri, K., & Smiles, S. (2004). Stock Market and aggregate economic activity: evidence from Australia. Applied Financial Economics, 14(2), 121-129. Dasgupta, R. (2012). Long-Run and Short-Run Relationships between BSE SENSEX and Macroeconomic Variables. International Research Journal of Finance and Economics, Issue 95, pp. 135-150. Engle, R., & Granger, C. (1987). Co-Integration, error correction Representation, estimation and testing. Econometrica, 55:1251-1276. http://dx.doi.org/10.2307/1913236. Fama, E. (1981). Stock Returns, Real Activity, Inflation and Money. American Economic Review, 71, 545-564. Fama, E. (1990). Stock returns, expected returns, and real activity. Journal of Finance, 45(4), 1089-1108. http://dx.doi.org/10.2307/2328716. Fisher, I. (1930). The theory of Interest. Macmillon, Newyork. Flannery, M. J., & Protopapadakis, A. A. (2002). Macroeconomic Factors do influence Aggregate Stock Returns. The Review of Financial Studies, Vol. 15, No. 3, pp. 751-782. Galbraith, J.K. (1955). The Great Crash 1929. Penguin, UK. Gay, R.D. (2008). Effect of Macroeconomic Variables on Stock Market Returns for Four Emerging Economies: Brazil, Russia, India, and China. International Business and Economics Research Journal, Vol. 7, No.9, pp. 1-33. Hsing, Yu (2011). The Stock Market and Macroeconomic Variables in a BRICS Country and Policy Implications. International Journal of Economics and Financial Issues, Vol. 1, pp. 1218. IHS Global Inc. (2013). Eviews 8 User’s Guide II. Web-link: www.eviews.com.


Maysami R., Howe, L.C., & Hamzah, M. A. (2004). Relationship between Macroeconomic Variables and Stock Market Indices: Cointegration Evidence from Stock Exchange of Singapore’s All-S Sector Indices. Jurnal Pengurusan, Issue 24, pp. 47-77. Maysami, R. C., & Koh, T. S. (2000). A vector error correction model of the Singapore stock market. International Review of Economics and Finance, 9, 79-96. Mukherjee, T. K., & Naka, A. (1995). Dynamic relations between macroeconomic variables and the Japanese stock market: an application of a vector error correction model. Journal of Financial Research, 18, 223-237. Nai-Fu, C., Richard, R., & Stephen, R. (1986). Economic Forces and the Stock Market of US. Journal of Business, Vol. 59, No.2, pp.383-403. Singh, D. (2010). Causal Relationship between Macro-Economic Variables and Stock Market: A Case Study for India. Pakistan Journal of Social Sciences (PJSS), Vol. 30, No. 2, pp. 263-274. Tripathi, V., & Kumar, A. (2015a). Short Run Causal Relationship between Inflation and Stock Returns - An Empirical Study of BRICS markets. Asian Journal of Management Applications and Research, Vol.05, No.01. Tripathi, V., & Kumar, A. (2015b). Relationship between Inflation and Stock Returns – Evidence from BRICS markets using Panel Cointegration Test. International Journal of Accounting and Financial Reporting, Macro Think Institute, USA, Vol. 4, No. 2, 647-658. Tripathi, V., & Kumar, A. (2015c). Do Macroeconomic Variables affect stock returns in BRICS Markets? An ARDL Approach. Journal of Commerce & Accounting Research, Macro Think Institute, USA, Vol. 4, Issue 2. Tripathi, V., & Seth, R. (2014). Stock Market Performance and Macroeconomic Factors: The Study of Indian Equity Market. Global Business Review, Vol. 15, No. 2, 291-316. Worthington, A.C., & Higgs, H. (2007). Assessing financial integration in European Union equity markets, 1990-2006: Panel unit root and multivariate cointegration and causality evidence. University of Wollongong, School of Accounting and Finance, Working Paper Series No. 07/10, 2007.