An analysis of market efficiency in the South Asian emerging stock markets:Bangladesh, India, Pakistan and Sri Lanka

摘要

This thesis investigates the weak-form of the Efficient Market Hypothesis (EMH) in the South Asian region. In particular, the emerging market countries of Bangladesh, India, Pakistan and Sri Lanka are considered. According to the weak-form of the EMH, current share prices reflect all available historical information such that investors should not be able to outperform the market on a consistent basis by trading on past information. It is an important topic for investigation given the economic growth as well as the financial development which have taken place in the region over the last two decades (South Asian Financial Markets Review, 2010). Moreover, most previous studies have investigated the topic for developed or other emerging markets; the South Asian region has largely been ignored. Prior studies which have investigated the South Asian markets have either focused on each country separately, or included one or two countries from the region as part of a broader sample. This thesis tries to fill this gap in the literature by investigating market efficiency in the South Asian markets as a regional grouping. In the first part of the analysis the long- and short-run relationships among the four stock markets are examined by employing a multivariate cointegration framework, the Vector Error Correction Model (VECM) approach, the Granger Causality test, Impulse Response Function analysis and Variance Decomposition analysis. A large sample of weekly stock index data is used in the analysis covering the 18-year period January 1993 - December, 2010. To analyse the effect of important global events on market integration, the data are split into the two sub-periods of pre- and post-September 11, 2001. The results suggest that linkages exist among the markets in both the long- as well as in the short-run. These findings imply that share price changes may be predicted from historical information not only in the market itself but from the changes in the other three markets as well. In addition, international portfolio diversification into the region may have limited benefits in the long-run as equity prices in all four countries move together in an equilibrium fashion over the longer run. In the second empirical analysis, relationship between the equity returns and macroeconomic variables is investigated. The research examines the EMH by investigating whether lagged shocks to macroeconomic variables are important in explaining equity returns. Both local and global macroeconomic variables are used and their importance in predicting the equity returns for each of the region’s markets is analysed. In particular, 12 macroeconomic variables were investigated, including seven local and five global measures being employed. Principal Components Analysis (PCA) is used to narrow down the most relevant factors. Principal Components (PCs) are then extracted and used as inputs into regressions explaining future returns. The resulting findings show that local economic factors are important in explaining share returns in the South Asian emerging stock markets. The findings support the notion that historical macroeconomic information may be used to predict share price changes in the regional markets.Finally, to investigate market linkages in greater depth, the thesis studies volatility and return interactions among the markets simultaneously. A multivariate GARCH-BEKK model is used to investigate return and volatility spillovers in own as well as in cross-markets. Results from the analysis indicated that the four markets of Bangladesh, India, Pakistan and Sri Lanka are linked not only by the news transmission about the share returns but also by the transmission of volatility. The evidence supports the notion that ‘news’ in one market influences not only the returns in that market but also the variance of price changes in other markets. These findings imply that equity returns in the South Asian stock markets are predictable from historical share price changes in their own, as well as from the other markets of the region; this result calls the weak form of the EMH into question since it suggests that an investor could outperform by studying historic return and volatility data in the region.