Exchange Rate Changes and Stock Returns in China: A Markov Switching SVAR Approach

This study empirically investigates the spillover effects between exchange rate changes and stock returns in China. Evidenced by multivariate Granger causality tests, stock returns Granger-cause exchange rates changes, but exchange rate changes exhibit little effect on stock returns. As the conventional structural VAR (SVAR) approach fails to examine the contemporaneous effects, we apply the Markov switching SVAR model to allow the coefficients and variances of endogenous variables to be state-dependent. The regimeswitching estimates indicate that the fluctuation in Shanghai B-share returns has positive effects on the remaining stock markets, but a negative impact on foreign exchange markets. This also reveals that the spillovers have longer durations during two financial crisis periods. Finally, this paper suggests investors to pay attention to systematic risks from RMB policy changes, which might alter the current unidirectional causality in the Chinese financial market.