Money and output causality: A structural approach

Using a new methodology, this paper revisits the question of whether money causes output and inflation. The causality between money and output is studied conditional on estimated DSGE models which allow for consideration of various channels where money and output interact. The causality tests are also conditioned on the structural shocks affecting both money and output. The results show that causality from money to either inflation or output at various horizons and for various specifications can be influenced by the preference, technological or interest rate shocks. It further shows that causality between money and output is not dependent on money demand shocks, though when money demand shocks are absent, there is no causality between money and inflation. Further robustness tests indicate that the results are also influenced by the filtering method used, although most of them are robust to the alternative filtering methods used before estimating the models.