The Global Financial Crisis and the Integration of Emerging Stock Markets in Asia

This study investigates the effects of volatility spillovers among five Asian stock markets (China, Hong Kong, Korea, Singapore, and Taiwan) and examines how the global financial crisis of 2008 has influenced volatility transmission among Asian stock markets. The results from a VAR(1)-bivariate GARCH model indicate strong volatility linkages between the Chinese stock market and the four emerging stock markets since the global financial crisis, suggesting the intensification of stock market integration in Asia since the crisis increases the integration of Chinese stock market in Asia. This strong integration of the markets is important in that the intensified linkages can reduce potential gains from the diversification of international equity portfolios.


I. Introduction
The issue of interdependence among equity markets has attracted increasing attention from researchers, particularly since the October 1987 crash, which brought about correlated stock price movements across stock markets worldwide (Kanas 1998). Early research focused exclusively on the spillover of the first moment of stock prices among major stock markets (Eun & Shim 1989;Jeon & von Furstenberg 1990;Cumby 1990).
Recent studies have investigated the interdependence of equity markets in terms of the conditional second moment of the distribution of returns, which refers to the volatility spillover effect. The existence of volatility spillovers implies that one large shock in a country can increase the volatility of not only equity market in that country but also that in others. Volatility is often related to the rate of information flow (Ross 1989). If information comes in clusters, then asset returns or prices may exhibit volatility even if the market adjusts to the news both perfectly and instantaneously. Thus, examining volatility spillovers can enable a better understanding of how information is transmitted across equity markets.
Recently, however, a number of studies have examined international stock market linkages among developed stock markets and some emerging stock markets in Asia, South America, and emerging Europe (Ng 2000;Edwards & Susmel 2001;Tse, Wu & Young 2003;Kanokwan & Dibooglu 2006;Chuang, Lu & Tswei 2007;Li & Majerowska 2008;Beirne, Schulze-Ghattas & Spagnolo 2010;Wang & Wang 2010). Uncovering the extent of international linkages among emerging stock markets can provide important implications for investors with international equity portfolios. It is clear that stronger international stock market linkages or co-movements can reduce the independence of emerging stock markets from external shocks, thereby reducing the potential 52 對外經濟硏究 제15권 제4호 2011년 겨울호 benefits of diversification into emerging stock markets.
A number of previous studies have focused on the impact of financial crises on volatility spillovers and provided support for the hypothesis that financial crises lead to market liberalization, market integration, and volatility transmission. In et al. (2001) found that Hong Kong played an important role in the transmission of volatility to other Asian markets during the Asian financial crisis of 1997. Nam, Yuhn, and Kim (2008) investigated how the Asian financial crisis changed emerging markets in Asia by focusing on volatility spillovers and found that the influence of U.S. innovations on stock prices in Asia increased after the crisis. Saleem (2009)  Previous studies have suggested a weak relationship between the Chinese stock market and other stock markets in Asia (Wang & Firth 2004;Groenewold, Tang & Wu 2004;Johansson & Ljungwall 2009). However, such findings may provide outdated guidance on international portfolio strategies in that they do not reflect recent data.
Because of China's large economic scale and impressive economic growth, the Chinese stock market has become one of the most important sources of information in Asia. In this context, this study contributes to research on emerging stock markets by focusing on the Chinese stock market's association, interaction, and integration with other Asian stock markets. The results suggest that stronger international linkages between the Chinese stock market and other stock markets in Asia may limit market independence and reduce potential gains from the diversification of equity portfolios for international investors.
The rest of this paper is organized as follows. Section 2 presents the econometric methodology of the bivariate GARCH-BEKK model. Section 3 provides the descriptive statistics of the sample data. Section 4 discusses the empirical results, and Section 5 concludes.

II. Methodology
Substantial attention has been focused on how news from one market affects the volatility process of another market. The univariate GARCH of Bollerslev (1986) has been extended to the multivariate GARCH model with a cross conditional variance equation. In this study, we analyze the volatility transmission by using a VAR(1)-bivariate GARCH(1,1) model with the BEKK parameterization (Engle & Kroner 1995).
Firstly, we consider the bivariate conditional mean model, namely VAR(1) process: The standard BEKK parameterization for the bivariate GARCH(1,1) model is written as: where t H is a 2 2 × conditional variance-covariance matrix at time t ; C is a 2 2 × lower triangular matrix with three parameters; A is a 2 2 × square matrix of parameters and measures the extent to which conditional variances are correlated past squared errors; and B is a 2 2 × squared matrix of parameters and shows the extent to which current levels of conditional variances are related to past conditional variances.
where the parameters 12 (4) and (5)         that is, "large changes tend to be followed by large changes, of either sign, and small changes tend to be followed by small changes" (Mandelbrot 1963). Table 3 shows the descriptive statistics and the results of the unit root tests for all sample returns in both subperiods. As shown in Panel A of Table 3, the range between the maximum and the minimum indicates that the return series for the post-crisis period was more volatile than that for the pre-crisis period and that all standard deviations for the post-crisis period were higher than those for the pre-crisis period because of the impact of the global financial crisis. All skewness and kurtosis values were high, and the calculated values of J-B test statistic were highly significant at the l% level, indicating that all the returns were not normally distributed.
We also examined the null hypothesis of a white-noise process for sample returns by using the Ljung-Box test statistic of the returns ( ( )   MacKinnon's (1991) 1% critical value was -3.435 for the ADF and PP tests. *** indicates the rejection of the null hypothesis at the 1% level of significance.

IV. Empirical Results
To assess the impact of the global financial crisis on the volatility spillover effect, we examined the direction of volatility spillovers and then compared the (1,1) model.

Volatility Spillovers in the Pre-crisis Period
For the analysis of the effects of volatility spillovers in the pre-crisis period, S.E. Coef.

Volatility Spillovers in the Post-crisis Period
There is a broad consensus that linkages among stock markets are likely to strengthen after a financial crisis. In this context, we considered the post-crisis period to determine the impact of the global financial crisis of 2008 on volatility spillovers among the Asian stock markets by paying close attention to the direction of shocks and volatility spillovers between the Chinese stock market and the other four Asian stock markets. A comparison of the estimation results for the pre-crisis period with those for the post-crisis period indicates that before the global financial crisis, the volatility of the Chinese stock market was largely independent of the volatility of the other Asian stock markets but that after the crisis, the volatility of the