بازارهای صاحبان سهام شرق آسیا، بحران مالی، و ارز ژاپن

The paper studies the interactions between the US and four East Asian equity markets. The focus is on the change in the information structure/flow between these markets triggered by the 1997 Asian financial crisis. It is shown that the information structure during the crisis period is different from that in the non-crisis periods. While the US market leads the four East Asian markets before, during, and after the crisis, it is Granger-caused by these markets during the financial crisis period but not in the post-crisis sample. Further, in accordance with concerns reported in the market, the Japanese currency is found to affect these equity markets during the crisis period. The Japanese yen effect, however, disappears in the post-crisis sample. The Japanese currency effect is quite robust as it is found from both local currency and US dollar return data and in the presence of Japanese stock returns. J. Japanese Int. Economies21 (1) (2007) 138–152.

The interaction of national equity markets is an active research area. Early studies usually focus on the comovement of national equity indexes; see, for example, Granger and Morgenstern (1970), Grubel and Fadner (1971), and Ripley (1973). Some of these studies are motivated by the benefits of international portfolio diversification (Grubel, 1968, Levy and Sarnat, 1970 and Solnik, 1974). In addition to portfolio diversification, the pattern of interactions provides evidence on information flows between national markets and the relative dominance of individual markets. Hamao et al. (1990), for instance, examine the interactions between the US, Japan, and UK stock markets and infer that information flow is unidirectional from New York to the other two markets. The leading role of the US in transmitting information to both developed and emerging markets is documented in numerous studies, which usually consider both returns and volatility interactions.1 To the extent that information flow and market dominance have implications for market stability and real economic activity, policymakers are interested in the patterns of equity market interactions. The recurrence of financial crises has spurred a literature on contagion in international markets. Volatility spillover is commonly used to capture the contagion effect (King and Wadhwani, 1990 and King et al., 1994).2 Some recent studies on contagion include Bae et al. (2003) and Forbes and Rigobon (2002). It is noted that the unusual volatility observed during crisis periods can affect the measurement of the intensity of markets interactions and, when is used properly, help identify the transmission mechanism.3 During the 1980s the gradual liberalization of financial markets in Asia, including Korea, Taiwan and other emerging markets, has fostered considerable investment interests in the East Asian equity markets. The creation of various mutual funds that have an investment focus on individual East Asian equity markets and on the region is an evidence of the growing popularity of investing in these markets. For international investors, apart from sharing the growth prospect of the region, diversification is another reason for investing in these East Asian equity markets. Even though these East Asian equity markets suffered a major setback during the recent Asian financial crisis, these markets have come back quite strongly and still represent good investment opportunities for international investors. The current study investigates the interactions between the equity markets in the US and four East Asian economies. Specifically, the study compares the interaction patterns before, during, and after the 1997 Asian financial crisis. Financial crises are characterized by extreme market conditions that may signal a different information transmission mechanism between financial markets during a crisis. Even after the financial crisis, informational linkages between markets can assume a different pattern depending on how the crisis is resolved. For instance, King and Wadhwani (1990) suggest that contagion effects lead to shock transmission during financial crises. On the other hand, Malliaris and Urrutia (1992) assert that there is no lead–lag relationship among the major national equity indexes during the October 1987 crash period. Jeon and von Furstenberg (1990) report that the comovement between international equity indexes is stronger after October 1987, while Cha and Cheung (1998) show that the US exerts a more pronounced effect on Asian markets after the 1987 crash. Tuluca and Zwick (2001) investigate 13 Asian and non-Asian equity markets before and after the 1997 Asian financial crisis and find that these markets experience a stronger comovement after the crisis. Besides the interaction between stock indexes, the current study also investigates the impact of Japanese currency movements on these markets. McKinnon and Schnabl (2003) show that fluctuations in the dollar–yen exchange rate have significant repercussion effects on other East Asian economies.4 It is argued that the dollar–yen exchange rate variation impacts other East Asian economies through, for example, changes in export competitiveness, and foreign direct investment flows. To the extent that Japanese currency movements affect the East Asian economies, these movements can affect the East Asian equity markets. One predominant feature of the Asian crisis is its rippling effects on economies both within and outside the region. Given its economic dominance and trade and financial ties in the region, Japan was closely scrutinized during the crisis period. Specifically, a weak yen was conceived as a threat to the recovery in the troubled region and the turnaround of the depressed stock markets. During that time, officials from various East Asian economies were quite vocal about the possibility that a sharp yen depreciation might lead to a new round of ‘competitive depreciation’ and trigger another wave of financial crises. Equity market traders were looking to the yen exchange rate for clues on the stock market movements.5 The market's acute concern about the Japanese currency suggests a possible change of the fundamental information structure in equity markets during the crisis period. International investment risk has two components—the price risk in the local market movement and the exchange rate risk. Under normal circumstances, investors concern only the market movement because they can hedge the exchange rate risk in the currency market by taking a short position in the forward market. During the Asian crisis, the Japanese yen exchange rate variability represented not only the usual standard exchange rate risk but also the likelihood that it could trigger another wave of financial disturbances in the region. Thus, the Japanese currency variation can assume a different role for the East Asian equity markets during the crisis. To empirically document such a phenomenon, the Japanese yen exchange rate will be explicitly included in our analysis of stock market interactions. Daily equity returns on Hong Kong, Korea, Singapore, Taiwan and the US markets from three sample periods are considered. The sample from January 1995 to June 1997 constitutes the pre-crisis period. The crisis period extends from July 1997 to June 2000. The post-crisis period is from July 2000 to July 2001. The causality test confirms the prominent role of the US in the international equity market. It is found that the US index leads the East Asian market indexes in all the sample periods under consideration. The East Asian market indexes, on the other hand, have the strongest effect on the US market during the crisis period but have no effect after the crisis. Most interestingly, our empirical results attest the effect of the Japanese currency on equity markets. The Japanese currency effect is widely felt during the crisis period but is not found in the post-crisis sample—even after controlling for the returns on the Japanese Nikkei 225 index and the currency conversion effect. As conceived by market participants, yen depreciation is found to induce a downward drift in these equity markets during the crisis. The remainder of the paper is organized as follows. Section 2 presents the basic data analysis. The causal relationships between the stock markets are examined in Section 3. Section 4 considers the Japanese yen effects. Additional analyses including the regression results from dollar-based return data are reported in Section 5. Section 6 offers some concluding remarks.

The paper presents an empirical study on the interactions between the US and four East Asian markets. The main empirical insights are the changes in the information structure/flow and in the role of the Japanese yen around the 1997 Asian financial crisis. The results are not contaminated by conditional heteroskedasticity because our testing procedure explicitly accounts for GARCH effects in the data. Our results document that the information structure during the crisis period is different from the non-crisis periods. The empirical evidence confirms the dominant role of the US market—the US index leads these East Asian markets before, during, and after the crisis. The influence of these East Asian markets on the US, however, is mainly found during the crisis. Specifically, in the post-crisis sample these markets do not affect the US market. An interesting result is the effect of the Japanese currency. While the reported yen effect on equity markets corroborates with the finding that fluctuations in dollar–yen affect the East Asian economies (McKinnon and Schnabl, 2003), we find that the yen effect is subsample specific. Consistent with concerns reported in the market, the Japanese currency is found to affect these equity markets only during the crisis period but not in the post-crisis sample. The Japanese yen effect does not appear spurious. The same currency effect is uncovered from both local currency and US dollar return data and is detected in the presence of the Japanese stock return variable. The results are suggestive of exchange rate risk is not overwhelmingly important under normal market conditions possibly because the risk can be hedged in the forward market. On the other hand, Japanese yen volatility appears a risk factor during the Asian crisis that is priced in the equity market. The findings corroborate the contention that financial crises endure extreme market conditions. These extreme market conditions can lead to changes in the channel via which information is incorporated and transmitted across markets. Policymakers, practitioners and researchers are aware of the interaction between equity markets. However, the nature and extent of market interaction are not very well understood. Our results on the change in information flows have some interesting implications. For policymakers, our results suggest that measures to stabilize and insulate the domestic market for external influences have to account for possible changes in the pattern and strengthen of market interdependence. Specifically, a measure that is effective during the tranquil market condition may not deliver the expected results in a turmoil and crisis period. For practitioners, information on market interactions helps improve hedging and managing portfolios that contain foreign equities. The documented changes in the causal relationship suggest that different investment strategies should be pursued under different market conditions. Further, if changes are not allowed for, the use of long sample data may yield obscure and even erroneous information on market interactions. For researchers, the reported varying causation pattern warrants a detailed study on information flow and propagation mechanisms under different market conditions. The upsurge and disappearance of the Japanese yen effect is an intriguing phenomenon. In addition to the usual question of what are the economic variables that help us understand the interactions between equity markets, one has to ask: “What are the factors that trigger the currency effect and through what channels the currency affects the equity market interdependence?” The channel and transmission mechanism with which the yen exchange rate affects the East Asian economies is an interesting future research area.