The past several decades has witnessed immense change in the world’s financial markets, especially in the Pacific-Basin (PB) region. Changes have stemmed from the removal of capital restrictions, the floating of currencies, improvements in information technology, the enhanced level of investor education, the increased mobility of labour, the escalation, and participation in trade organisations and a general shift towards a global economy and community. These structural changes are likely to propagate a greater level of market linkage, which may reduce potential benefits garnered through international diversification. Moreover, Cooper and Kaplanis (2000) argue that the level of financial market linkage influences optimal corporate capital structure through differing costs of capital. Bekaert, Harvey and Lundblad (2002) show that equity market linkage resulted from the market liberalization helps spur real economic growth. Extensive research has been done on equity market linkage. However, three important questions in this area of research still remain inconclusive or unanswered.
The first question is: what are the right measures of equity market linkage? Ammer and Mei (1996) use the contemporaneous covariance, whereas Bekaert and Harvey (1995), among others, use the conditional covariance. Forbes and Rigobon (2002) and Hon, Strauss and Yong (2004) address the issue of short-term contagion following the East Asian currency crisis using the correlation coefficients corrected for heteroskedasticity. However, Kasa (1992) and Darrat and Zhong (2005) point out that contemporaneous correlation may not reflect genuine information on markets interrelationship and could be misleading if investors have long holding periods. Bracker, Docking and Koch (1999) use Geweke measures of feedback to measure the extent of financial market linkage for nine equity markets. These and other studies such as Dekker, Sen and Young (2001) and Johnson and Soenen (2002) all point to considerable linkage over the short run.
We argue that international diversification is a long-term investment strategy. Hence, from the diversification perspective, it is more meaningful to study equity market linkage over a long run. Further, the effects of equity market linkage on the optimal corporate capital structure of firms and on real economic growth should bebetter gauged from a long-run point of view. These may be the reasons for a popular use of cointegration as a measure of market linkage as cointegration represents the long-run equilibrium relation between equity markets. A partial list of studies include Kasa (1992), Hung and Cheung (1995), Phylaktis (1999), Masih and Masih (2001), Manning (2002), Darrat and Zhong (2002, 2005), Leong and Felmingham (2003), and Bessler and Yang (2003).
The second question points to the time variation of equity market linkage. Using a regime-shifting model, Bekaert and Harvey (1995) find that the level of world’s market integration to fluctuate through out time. Longin and Solnik (1995) find that conditional correlations between seven developed stock markets have strengthened over time and that correlations rise during periods of high conditional volatilities. Silkos and Ng (2001) apply the rank constancy test developed by Qunitos (1993) to the cointegration relation between six PB countries and the US and find that in almost every case that the null hypothesis of rank constancy is rejected. Moreover, there is evidence that equity market linkage tends to change in response to major events such as the 1987 US market crash and the 1990 Gulf War. Our fifteen year plus data sample for the PB markets provides a unique opportunity to examine the effects of East Asian currency (EA) crisis and 9/11on equity market linkage.
The third question is: what are the factors that determine the market linkage? A growing literature suggests that the evolution of equity market linkage across national borders tends to be associated with international trade and other macroeconomic factors. For example, Roll (1992) argues that the industrial structure significantly affects the behaviour of international stock prices. Bracker, Docking and Koch (1999) highlight the importance of interdependence among international goods markets when discussing equity market linkages. Phylaktis and Ravazzolo (2002) using return decomposition find significant evidence to suggest that the countries in the PB region are integrated in an economic sense between themselves and with Japan and the US, and economic integration is the main source of stock market integration. Darrat and Zhong (2005) suggest that equity market linkage strengthens as a result of closer goods market linkage induced by trade accords.
In this paper, we aim to address all three questions for all the equity markets with data availability in the PB region. Our study differs from previous literature in three aspects. First, while we rely on the long-run linkage concept—cointegration, we pay attention to the potential effects of the US and Japanese markets on the PB market linkage. We establish that the Pacific-Basin markets exhibit some level of market linkage in the long run, which is primarily driven by the US market as opposed to Japan. Moreover, the degree of linkage varies over time and changes considerably in response to major events such as the East Asia currency crisis and 9/11. Second, we study the time evolution of equity market linkage using recursive trace test and the persistence profile analysis. The former presents the overall variation of cointegration between the PB markets, and the latter shows the change of equity market linkage before and after the EA crisis and 9/11. We attempt to identify a set of factors that can possibly explain the level of market linkage. We find that bilateral exchange rate volatility, interest rate differentials, market return volatility, and market development affect the level of equity market linkage. Moreover, the ASEAN countries have significantly higher level of equity market linkage between each other.
The rest of this paper is organized as follows. Section 2 presents the data and summary statistics. Section 3 discusses the methods for the analyses. Section 4 proposes a set of variables to explain the equity market linkage in cross-sectional regressions. The last section concludes the paper.
Download
PDF Ebook Analyses of Equity Market Linkage in the Pacific Basin
