The empirical performance of the financial accelerator since 2008

Abstract

We evaluate the empirical performance of financial frictions à la Bernanke et al. (1999) during and after the Global Financial Crisis. We document that in an ex-post analysis based on nonlinear Bayesian methods, these frictions do not improve the standard medium-scale DSGE model's ability to explain the macroeconomic dynamics during the Great Recession. The reason is that in the estimated model with financial frictions, the drastic post-2008 collapse of investment causes firms' leverage to decline. Taking the model at face value, this would trigger a narrowing of the credit spread, contradicting the observed persistently large credit spread throughout the post-2008 period. Additionally, the estimated model attributes only a minor role to risk shocks à la Christiano et al. (2014). These findings are confirmed independently for US and euro area data.