We propose a Bayesian approach to identify the excessive comovement of two markets as a contagion. This goal is technically achieved by linking a latent factor model and single equation error correction model and testing the breaks in the short-term and long-term relationships and correlatedness in the linked model. We find that a short-term relationship representing a systematic volatility ratio between two markets plays a key role in contagion dynamics. When long-term relationship or correlatedness is broken, the cause is determined by calculating posterior probabilities. If the cause is a break in the short-term relationship, a contagion is formally declared.