Sources of cross-sectional variations in stock returns and risk: an empirical analysis of emerging markets

It is well established in the financial economics literature that potential gains from international diversification are generated from the imperfect correlation between national stock market returns. This empirical study explores the factors that impede perfect integration among national equity markets by examining emerging markets data. The first major topic of the dissertation is to re-visit the debate on the relative importance of country and industry effects in the cross-sectional variation of stock returns. By applying the standard Heston and Rouwenhorst (1994) dummy variable decomposition method to $U. S. nominal returns from 11 industry sectors of 13 emerging markets from 1984 to 2004, this work confirms that country effects do play a dominant role in determining the cross-sectional variation in stock returns in emerging markets but since late 1990s, the industry effects have become increasingly important. This conclusion is robust even after the removal of three potential biases: inflation rate, exchange rate and interest rate effects, all of which may amplify the country effects. The second topic is to investigate the debate from the perspective of stock risk. Stock risk is modeled and calculated independently from a return model with ARCH type errors. By applying the standard dummy variable decomposition method to stock risks, the empirical evidence is found to support the conclusions drawn on stock return decompositions. Finally, in order to find the fundamental sources of the country and industry factors, pure country and industry effects are then regressed on fundamental characteristics of country and industry. The findings show that the change in the variables representing the exchange rate can explain a substantial amount of the country effect variations, while at the same time, banking and stock markets development also contribute to the variations. The regressions also find evidence that the legal origin of the market does matter to stock returns. Regressions on industry effects are not as promising as the results of the country effects regression. Only the geographical concentration of industries is found to explain a small amount of the industry effects.