Credit shocks and cycles: a Bayesian calibration approach

This paper asks how well a general equilibrium agency cost model describes the dynamic relationship between credit variables and the business cycle. A Bayesian VAR is used to obtain probability intervals for empirical correlations. The agency cost model is found to predict the leading, countercyclical correlation of spreads with output when shocks arising from the credit market contribute to output fluctuations. The contribution of technology shocks is held at conventional RBC levels. Sensitivity analysis shows that moderate prior calibration uncertainty leads to significant dispersion in predicted correlations. Most predictive uncertainty arises from a single parameter. JEL classification: C11, C32, E32, E37, E44 ∗Nuffield College, University of Oxford. Email: [email protected]. Financial support from the Economic and Social Research Council and the British Academy Post Doctoral Fellowship is gratefully acknowledged.