Comment on "The Output Composition Puzzle: A Difference in the Monetary Transmission Mechanism in the Euro Area and U.S."

This paper (Angeloni et al., 2003, this issue of JMCB) deploys a range of VAR specifications to look at the monetary transmission mechanism in Europe and U.S. A feature that I like very much is that the specification keeps track of consumption and investment dynamics as well as GDP so that the composition effects of monetary policy can be studied. The paper also checks for robustness by looking at same phenomena in large-scale structural models.

The paper finds that much more of the U.S. monetary transmission mechanism operates through consumption, while the European transmission mechanism operates much more through investment. This can be seen from Figures 1 and 2 and Tables 3 and 4. For example, in structural models at an eight-quarter horizon, we see that in the U.S., the consumption impulse is 70% of the size of the GDP impulse, while the investment impulse is only 20%. By contrast, in Europe, the investment impulse is 70% of the magnitude of the GDP impulse. The VAR models tend to show even bigger differences.

The paper estimates dynamic structural general equilibrium models to see if they are consistent with VAR evidence. It finds that they can "be tweaked" to fit pattern for Europe (by reducing the investment adjust cost parameter below its estimated level). However, the models apparently cannot be tweaked to fit the pattern in the U.S.