Does the central bank respond to credit market factors? A Bayesian DSGE approach

This paper estimates a version of a New Keynesian Dynamic Stochastic General Equilibrium model with financial frictions for the United States using Bayesian techniques. Various Henderson-McKibbin-Taylor style monetary policy rules are examined, which react to inflation, output and credit market factors including credit spreads, financial leverage and credit growth. The central question is whether the central bank responds to credit market factors in setting the policy interest rate, which is investigated using posterior odds tests. The paper explores whether there is evidence of stabilization, if indeed the central bank is responding to credit market factors. This is conducted using impulse response analysis and an examination of parameter posterior distributions. The most compelling result during the period under study is the US Fed responded to credit spreads in setting the policy rate. The empirical results also confirm that credit spreads offer stabilization benefits. This result is robust to variations in the policy rule. It is also found that while financial leverage improves model fit when included in the policy rule, the response is pro-cyclical, which would unlikely be a feature of stabilization policy. Finally, there is no evidence that the policy interest rate responded to credit growth.