Structural Break and Regime-specific Causality between Stock Prices and Exchange Rates in Asian Countries

Using the recently developed structural break threshold VAR, the main purpose of this study is to test for the structural break and regime dependent Granger causality between stock prices and exchange rates for six Asian countries of China, Japan, Korea, India, Indonesia and Taiwan including the linear Granger causality tests from the traditional VAR model for comparison. This study notes that causality results based on a constant-parameter linear model are likely to be biased in the presence of structural breaks and regime changes. In the case of the unidirectional Granger causality running from an exchange rate to a stock index for China, Korea and Indonesia in regime 2-break 1 and Taiwan in regime 1-break 2, the policy-maker should be careful to implement intervention policies to retain an exchange rate because they could lead to a fall in a stock index. On the other hand, in the case of the unidirectional causality running from a stock index to an exchange rate for India and Indonesia both in regime 1-break 1 and regime 1-break 2, it may be implied that more aggressive intervention policies may be implemented with little adverse or no effects on a stock index. If there exists no causality in either direction, say in regime 2-break 1 for Japan, it would imply that intervention policies to exchange markets do not affect a stock index.