This paper studies the existence of changes in the transmission mechanism of monetary policy on 110 US monthly macroeconomics variables during the Great Recession. Impulse Response Functions for this large dataset are estimated for different states of the economy. I combine three different techniques to deal with the dimensionality problems which emerge from an estimation procedure of this magnitude: (i) factor decomposition, (ii) an identification strategy independent of the number of variables included in the dataset and (iii) a blockwise optimization algorithm for the correct selection of the Bayesian priors. Results show the presence of structural breaks in the forces driven the economy as well as qualitative differences in the reaction of all the variables to monetary policy decisions depending on these changes.
Abstract
Download pdf
Leave Comment