Ana Beatriz Galvão , University of Warwick Liudas Giraitis , Queen Mary University of London George Kapetanios , Queen Mary University of London Katerina Petrova , Queen Mary University of London
December 20, 2015
Download full paper
We build a time varying DSGE model with financial frictions in order to evaluate changes in the responses of the macroeconomy to financial friction shocks. Using US data, we find that the transmission of the financial friction shock to economic variables, such as output growth, has not changed in the last 30 years. The volatility of the financial friction shock, however, has changed, so that output responses to a one-standard deviation shock increase twofold in the 2007-2011 period in comparison with the 1985-2006 period. The time varying DSGE model with financial frictions improves the accuracy of forecasts of output growth and inflation during the tranquil period of 2000-2006, while delivering similar performance to the fixed coefficient DSGE model for the 2007-2012 period.
J.E.L classification codes: C11, C53, E27, E52
Keywords:DSGE models, Financial frictions, Local likelihood, Bayesian methods, Time varying parameters