یک بازار سرمایه چگونه یک رابطه نزدیک با دیگر بازارهای مشابه خود دارد؟بررسی مورد تایوان درگیر در بحران مالی آسیا

In assessing how far and how close the relationships are between the Taiwan capital market and other international capital markets in Asian financial case, we examine the co-movement patterns by developing the “unequal variance test”. We find that a closer relationship exists between Taiwan and Hong Kong throughout the sample period than between Taiwan and other Asian countries and the US. It thus appears that adjacent regions with similar backgrounds in terms of their capital markets will reflect price patterns that are more similar to those of Taiwan than those of countries with which Taiwan frequently trades or cooperates.

Owing to reasons attributed to close interaction in economic trade or the sudden dispersion of international information, capital markets in different countries may frequently affect each other. The past literature has provided extensive discussions and findings related to the issue of linkages among stock markets using the cointegration approach. Similar examples related to the co-movement patterns in international capital markets abound in the literature and the reasons for these relationships have been investigated (Malliaris and Urrutia, 1992, Chan et al., 1992, Arshanapalli and Doukas, 1993, Arshanapalli et al., 1995, Najand, 1996, Uri et al., 1996, Wang et al., 2003, Atteberry and Swanson, 1997, Ng, 2002 and Yang et al., 2003). What is implied in the cointegration of international capital markets is that there exist cross-country spillover effects and transmissions of shocks. Some of the empirical results suggest that co-movements in the international stock markets could be observed following the occurrence of a financial crisis. Two famous financial crises have often been mentioned in the past: the October 1987 global stock market crash and the July 1997 Asian financial crisis that emerged in Thailand. The two cases appear to exhibit similar spillover effects. Arshanapalli and Doukas (1993) regarded the state of the stock markets from 1980 to 1990 and the crash of October 1987 as a watershed. By taking into consideration the Dow-Jones Industrial index, Frankfurt's Faz index, the London FT-SE 100 index, Japan's Nikkei index, and Paris's CAC index, they found that there was an increased tendency for co-movement among most international markets after the crash. Arshanapalli et al. (1995) examined the cointegration effect after the crash in October 1987 in spite of no relationship being found among these markets including the US and various Asian countries (Japan, Malaysia, Singapore, Thailand, the Philippines and Hong Kong) before the crash. Several studies have suggested that a co-movement pattern exists following the Asian financial crisis, for instance, Arshanapalli, Doukas, and Lang (1995). Ng (2002) pointed out that the linkages among the stock market returns of Indonesia, the Philippines and Thailand exhibited a long-term relationship following the Asian financial crisis in 1997 although no such evidence was found before the crisis. Yang et al. (2003) found similar evidence regarding the transmission mechanism in the 1997 Asian financial crisis among stock markets that included those of the United States, Japan, and Asian emerging countries, showing that country-by-country stock market integration tends to change the leader–follower relationships around the time of the financial crisis period. Although these past studies have yielded common results with regard to the cointegration effects, the evidence regarding the relationships is still somewhat contradictory. Chan, Gup, and Pan (1992) analyzed the degree of integration among securities markets in Hong Kong, South Korea, Singapore, Taiwan, Japan, and the United States and found no close relationships among any of these markets. Some natural questions that arise are as follows. First, “how close” are the relationships between the stock markets? Although past studies point out whether or not the interconnections may exist by applying cointegration models, little is known about the how close the co-movement relationship is at present. Second, what is the main reason the spillover effect exists in the first place? Is it because of the adjacent region and the similarities of background of the capital markets, or the international trade and the business cooperation that take place among countries? While the closeness of the relationship could be observed, the answers to the factors behind the spillover effects are much clearer. Third, have the relationships changed because of the financial crisis? It is reasonable to consider that the cointegration relationships will change over time. Regarding the first question, as to “how close” the relationship is between capital markets has become an interesting and important challenge to traditional thinking that is based on forming an international portfolio for investors. In assessing how far and close the relationships are between international capital markets, we examine the co-movement patterns by developing the “unequal variance test”. This paper builds a statistical test referred to as “the unequal variance test” which, as will be seen later, seeks to find which are the closest linkages. As for the second issue which relates to why spillover effects exist, we specifically focus on the linkages between each pair of stock markets and include Taiwan in each pair, such as East Asia and Taiwan, and the US and Taiwan. The country of most interest to us in this paper is Taiwan. The other countries related to Taiwan include Australia, Hong Kong, Malaysia, the Philippines, Singapore, South Korea, Thailand, and the US. The main reason why we choose Taiwan is that much trade has taken place with Taiwan with countries other than those in the Asian region over the past several decades, for instance with the US. If the crisis, for example, emerges in Asia, the influence of the spillover effect will be slight for the US and, consequently, for Taiwan. Therefore, this paper attempts to use Taiwan as its central focus to examine the long-run relationships among stock markets including the US market and other Asian markets. The links in terms of stock price movements are closer between the Taiwan and US markets owing to their great trading activities, or between Taiwan and other Asian markets owing to their similar backgrounds. We are able to observe the different linkages with Taiwan and to include the possible reasons for the spillover effect. While “how close” the relationship is can be observed, it is much more clear with regard to the answer to the factors driving the spillover effect. With regard to the last question and whether the relationships have changed, we compare three different sub-periods, namely, the pre-crisis, crisis, and post-crisis periods, in order to explore the changing impacts of the 1997 Asian financial crisis. To sum up, we adopt two viewpoints to observe the impacts of the co-movements of international stock markets in this paper. First, the empirical results are compared for three periods, i.e., before, during, and after the Asian crisis based on the unequal variance test that this paper develops to find out how close the relationship is. Second, this paper contributes to the literature by explaining the factors underlying the spillover effect and the co-movement of capital markets. We find that Taiwan's relationships with Hong Kong and the US are closer than those between Taiwan and other countries for the whole sample period. If different sub-periods are considered, during the 1997 financial crisis, there is no special relationship between Taiwan and other countries although a cointegration relationship does exist between Taiwan and the other countries. Besides, there is much closer relationship between Taiwan and Malaysia after the 1997 financial crisis. It appears that the similarity of background of the capital market will reflect the more similar price pattern after the financial crisis. The findings of this study have implications for a closer relationship between similar and adjacent countries than major trading partners under the cointegration relationship.

This paper focuses on how close the linkages were between Taiwan and other countries in the event of the 1997 Asian financial crisis. The novelty of our paper rests mainly on the unequal variance test that is performed to point out the different relationship conditions. The unequal variance test could help us to find the countries which are closest to the goal country, Taiwan, in international capital markets. Ever since the Asian crisis erupted in 1997, there has been a significant impact on the whole of Asia. The purpose of this study is to ascertain the co-movement patterns of Taiwan and other countries before the 1997 Asian financial crisis as compared with both during and after the crises. This evidence shows that the co-movement patterns are found to exist during the crisis period but the conclusion that there exists an especially close linkage between Taiwan and any other country cannot be reached. Many investors focus solely on their imagination when diversifying risks, such as countries exhibiting a leader–follower relationship while ignoring the similarities in their stock markets. The co-movements between Taiwan and the US are not as strong as investors imagine and we find the relationship between Taiwan and Hong Kong to be stronger than that between Taiwan and the US for the whole sample period. The financial crisis is found to lead to a greater change in the regional economies, for example, the cointegrating relationship that formerly did not exist has now been switched to one that does exist. It appears that the adjacent regions or similarity of background of the capital markets will be reflected by similar price patterns.