International transmission of domestic and foreign structural shocks: A new perspective

The increased globalization of the world led to a greater emphasis on the importance of foreign events in understanding business cycles and macroeconomic developments in home countries. This dissertation introduces new aspects in the analysis of the international transmission of domestic and foreign structural shocks in a selected sample of industrialized countries using structural vector autoregression models.;In the first article, we analyze the transmission of global versus country-specific shocks on major bilateral macroeconomic variables between Germany, Japan and the United States by adopting an identification scheme that allows global shocks to affect all countries asymmetrically. This is in contrast to the standard assumptions of open-economy macroeconomics, which state that global shocks affect all nations equally, and as such have no effect on relative prices, output levels, and current account balances. Interestingly, we find that global shocks play an important role in explaining fluctuations in bilateral real exchange rates and real current accounts.;In the second article, we investigate the transmission of U.S. supply and demand shocks to the Canadian economy using different identification methods to assess the robustness of the results. The methods applied are the standard Blanchard-Quah decomposition invoking long run neutrality, the Sims-Bernanke decomposition accounting for price inertia, and a recent decomposition method developed by Cover, Enders and Hueng (2005), which allows for nonzero cross-correlations between the shocks.;Our main findings are robust across identifications. We show that U.S. shocks tend to intensify both Canadian booms and recessions, while they have an overall stabilizing effect on Canadian prices over the flexible exchange rate period. We also show that, contrary to the findings of earlier studies, Canadian shocks account for most of the variations in Canadian variables, with U.S. shocks playing a significant but not predominant role. Another interesting result related to the identification methods is that, when the correlations between the shocks are not restricted to zero, Canadian shocks explain close to 20% of U.S. real GDP growth long run forecast error variance. This result suggests that the effect of Canadian shocks on U.S. variables may be larger than what is generally assumed.