DSGE model with financial frictions over subsets of business cycle frequencies

Abstract

We estimate a DSGE model with financial frictions and banks on subsets of frequency bands corresponding to higher and lower business cycle frequencies. Our results show that goodness of fit and parameter estimates associated with real and financial frictions vary substantially across business cycle frequency bands, with the estimated coefficients related to financial adjustment costs and for the parameter governing the costs associated to capital requirements displaying the largest variations. Moreover, the comparison of the marginal likelihoods with a frictionless framework suggests that the introduction of financial frictions provides a significant improvement at lower business cycle frequencies. This is consistent with the presence of macro-financial linkages where credit market frictions affect spending decisions over longer time horizons.