Was the Russian Financial Crisis Contagious?

Using daily data from the stock markets of nine European transition economies, this paper tests for stock market contagion during the 1998 Russian financial crisis by utilizing both univariate and bivariate correlation analysis. The results of the linear model indicate that there is no evidence of contagion, while the bivariate analysis, based on the newly developed Corsetti-Pericoli-Sbracia (CPS) test, reveals the presence of structural breaks between the Russian and Czech stock markets. Moreover, crisis-post crisis comparison analysis shows that contagion occurred after the Russian crisis. This paper proposes to label such an effect as a 'reverse' contagion. The results of Monte Carlo experiments show that the linear model performs poorly under the null hypothesis of interdependence and systematically under-rejects in the case of small test sizes. In sum, at least for the examined parameter values, it appears that the CPS test has less size distortion than the linear model.