Recovery rates are negatively related to default probabilities (Altman et al., 2005). This paper proposes and estimates a model in which this dependence is the result of an unobserved credit cycle: When times are bad, the default probability is high and recovery rates are low; when times are good, the default probability is low and recovery rates are high. The proposed dynamic model is shown to produce a better fit to the data than a standard static approach. It indicates that ignoring the dynamic nature of credit risk could lead to a severe underestimation of credit risk (e.g. by a factor of up to 1.7 in terms of the 95% VaR). Also, the model indicates that the credit cycle is related to but distinct from the business cycle as e.g. determined by the NBER, which might explain why previous studies have found the power of macroeconomic variables in explaining default probabilities and recoveries to be low.