LAUSR

Purpose This paper aims to test whether the latest global financial crisis propagated contagiously from the USA to the rest of the world. Design/methodology/approach If the reason of the propagation of a crisis is a normal time interdependence with the crisis origin country due to real linkages, the spread of crisis can be limited by implementing well-defined preventive policies. On the other hand, if a crisis propagates because of the speculative attacks or irrational behaviors, the “national policymakers will face difficulties in protecting their markets from such a crisis” (Kleimeier et al., 2003, p. 2). Therefore, separation of contagion and interdependence may provide crucial insights for policymakers to implement appropriate policies to prevent and/or stop the financial crisis. Hence, this paper compares the heteroscedasticity-corrected conditional correlations and dynamic conditional correlations in the tranquil and shock periods. Findings The findings were quite straightforward and consistent for both Forbes and Rigobon heteroscedasticity correction technique and dynamic conditional correlation (DCC) model. The Forbes and Rigobon technique failed to reject the null of no contagion for 25 countries in our data sample, while the DCC model failed to reject the null of no contagion for 21 countries. While heteroscedasticity-corrected correlation technique confirmed the presence of a contagion for six countries, the DCC technique confirmed the presence of a contagion for ten countries. Originality/value This study particularly investigates whether the subprime mortgage crisis spilled over contagiously to the rest of the world. To investigate whether there is a significant increase in the cross-market correlations between the crisis origin country, the USA and the rest of the world markets during the latest financial crisis, both heteroscedasticity-corrected correlation technique and DCC model are used.Therefore, this study possibly contributes well to the literature using a large country set and conducting the analysis from different angles for important properties.