The author proposes a new test for financial contagion based on a non-parametric measure of the cross-market correlation. The test does not depend on the assumption that the data are drawn from a given probability distribution; therefore, it allows for maximal flexibility in fitting into the data. Simulation studies show that the test has reasonable size and good power to detect financial contagion, and that Forbes and Rigobon's test (2002) is conservative, suggesting that their test tends not to find evidence of contagion when it does exist. The author's new test is applied to investigate contagion from a variety of recent financial crises to the Canadian banking system. Three empirical results are obtained. First, compared to recent financial crises, including the 1987 U.S. stock market crash, 1994 Mexican peso crisis, and 1997 East Asian crisis, the ongoing 2007 subprime crisis has been having more persistent and stronger contagion impacts on the Canadian banking system. Second, the October 1997 East Asian crisis induced contagion in Asian countries, and it quickly spread to Latin American and G-7 countries. The contagion from the East Asian crisis to the Canadian banking system was not as strong or as ersistent as that of the ongoing subprime crisis. However, it had a stronger impact on emerging markets. Third, there is no evidence of contagion from the 1994 Mexican peso crisis to the Canadian banking system. Contagion from that crisis occurred in Argentina, Brazil, and Chile, but the contagion effects of that crisis were limited to the Latin American region.
Financial stability Central bank research Econometric and statistical methods