بازگشت و نوسانات سرریز در میان بازارهای سهام شرق آسیا

This article examines the extent of contagion and interdependence across the East Asian equity markets since early 1990s and compares the ongoing crisis with earlier episodes. Using the forecast error variance decomposition from a vector autoregression, we derive return and volatility spillover indices over the rolling sub-sample windows. We show that there is substantial difference between the behavior of the East Asian return and volatility spillover indices over time. While the return spillover index reveals increased integration among the East Asian equity markets, the volatility spillover index experiences significant bursts during major market crises, including the East Asian crisis. The fact that both return and volatility spillover indices reached their respective peaks during the current global financial crisis attests to the severity of the current episode.

The financial crisis that started in the US sub-prime mortgage market in February 2007 reached its climax in mid-September 2008 with the disastrous collapse of the Lehman Brothers. As the global financial crisis have unfolded in several stages, financial markets all around the world went through wild fluctuations, with volatility spreading across markets at an unprecedented speed. The current financial crisis is not the first of its kind. Following the globalization wave of the early 1990s, financial market crises have become a more frequently observed phenomena, especially in the emerging market economies. During these crises, volatility in financial markets has increased sharply as the stock returns moved into negative territory. As the initial tremors of each of these crises are not confined to the originator country but spread to other countries as well, it is important to obtain a measure of return and volatility spillovers across countries during financial crises. Early work on contagion dated back to the aftermath of the October 1987 U.S. stock market crash. However, it was not until after the East Asian and Russian crises of 1997–1998 that financial contagion and spillovers had become a major area of research.1 From the beginning on, the empirical literature on contagion focused on stock returns, and the possibility of volatility contagion has mostly been ignored in the literature. Departing from the rest of the empirical literature, Edwards (1998), Edwards and Susmel (2001) and Baur (2003) are the only papers on the possibility of contagion taking place through spillovers of volatility across stock markets. Recently, there have been scores of new research papers mostly focusing on how the current financial crisis has spread around the globe.2 Among these Diebold and Yilmaz (2009a) proposed a new approach to the analysis of contagion and interdependence across markets. In this paper, we follow in their footsteps. Using separate vector autoregression of returns and range-based volatility estimates for 10 East Asian stock markets, we analyze the differences in the dynamics that drive return and volatility spillovers over time. Variance decomposition analysis of the VAR model allows us to identify spillovers of return and volatility shocks from the indigenous shocks. In order to measure volatility we use efficient range-based volatility estimate that was first proposed by Garman and Klass (1980). In this paper, we focus on major East Asian stock markets only. Over the last two decades, East Asian economies and markets have developed into a powerhouse in the global economy. In addition to attaining a growth rate well above the world average, with their rapidly developing financial markets, the East Asian economies started to play an increasingly influential role in the global financial system. As a consequence, it is interesting to study how the region's markets are affected during different financial crisis episodes since early 1990s and especially during the current global financial crisis. We apply VAR model and the variance decomposition analysis to 100-week long rolling windows of East Asian stock returns and volatility measures separately. For each window we calculate the contribution of spillovers across markets to the variance of forecast errors. Plotting the total contribution of spillovers in all markets across time we obtain a measure of spillovers across markets. Our approach differs from the main contributions to the literature on financial contagion (such as Forbes & Rigobon, 2002, and papers in Claessens & Forbes, 2001) in several respects. We do not test for contagion before and/or after major crisis episodes, the beginning and ending dates of which are determined exogenously. Instead, using a rolling window framework enables us to account for major changes in the return and volatility spillovers separately by plotting the return and volatility spillover indices. Our empirical results show that there is substantial difference between the behavior of the East Asian return and volatility spillover indices over time. While the return spillover index reveals increased integration among the East Asian equity markets, the volatility spillover index experiences significant bursts during major market crises, including the East Asian crisis. The fact that during the current global financial crisis the return spillover index experienced its most significant burst since 1990s along with the volatility spillover index and both indices reached their respective peaks attests to the severity of the current financial crisis episode. Section 2 briefly motivates and describes the spillover index methodology, which is based on variance decompositions of forecast errors obtained from a vector autoregression. In Section 3 we use the spillover index methodology to assess East Asian stock return and volatility spillovers since 1992. In this section, we showed that our results are robust to alternative orderings and also to the inclusion of Chinese, Indian and American equity markets in the analysis. In Section 4 we summarize our results.

We applied the Diebold and Yilmaz (2009a) spillover index methodology to 10 major East Asian stock markets to study the behavior of return and volatility spillovers across the region over the 1992–2009 period. Using rolling sub-sample windows we show that volatility and return spillovers behave very differently over time, during crisis and non-crisis episodes. Plots of volatility spillovers leave no doubt that it is the burst in volatility spillovers across markets rather than the return spillovers that takes place during the major crises. As a result of increased market integration throughout the 1990s East Asian stock markets had become more interdependent as captured by the increase in return spillovers in the mid-1990s. Even after the major emerging market crises the return spillovers had not declined to the levels in the early 1990s. With the global financial crisis of 2008 return spillovers in the East Asia region also reached the highest level. The burst in the return spillover index reflects the systemic nature of the current global financial crisis.