'Some contagion, some interdependence': More pitfalls in tests of financial contagion

This paper builds on a standard factor model of stock market returns to reconsider recent empirical literature on contagion in financial markets based on bivariate correlation analysis. According to this literature, contagion is defined as a structural break in the linear transmission mechanism of financial shocks. Using our framework, we show that the result of 'no contagion, only interdependence' stressed by recent contributions is due to arbitrary and unrealistic restrictions on the variance of country-specific shocks. We focus on the international effects of the Hong Kong stock market crisis of October 1997 as a case study. For plausible values of the variance of country-specific shocks in Hong Kong, current tests cannot reject the null of interdependence for 16 countries out of a sample of 17. Our analysis strongly questions such conclusion, finding evidence of 'contagion' for at least five countries.