In this paper, we introduce the concept of causality in the Markov switching framework into the analysis of financial inter-market dependencies. We extend the methodology of testing for financial spillovers between capital markets by explicitly defining contagion, spillovers and independence, and providing statistics to test for the existence of causality. We apply the methodology to stock index returns on the Japanese (Nikkei 225) and the Hong Kong (HSI) markets during the Asian crisis and find no evidence of contagion between the markets, but strong evidence of feedback spillovers between them.