Identification in Structural Vector Autoregressions Through Graphical Modelling and Monetary Policy: A Cross-Country Analysis

There is an ongoing debate on how to identify monetary policy shocks in SVAR models. Graphical modelling exploits statistical properties of data for identification and offers a data based tool to shed light on the issue. We conduct a cross-country analysis, considering European Monetary Union (EMU), Japan and US. We obtain some important results. The information set of the monetary authorities, which is essential for the identification of the monetary shock seems to depend on availability of data in terms of higher frequency with respect to the policy instrument (US and Japan). Moreover, there is not yet a widespread consensus on whether or not the European Monetary Union should be considered as a closed economy. Our results indicate that EMU official interest rate depends on the US federal funds rate.