Credit cycles and labor market slacks: predictive evidence from Markov-switching models

We model unemployment and credit cycle dynamics as a Markov-switching process with two states to identify labor market slacks i.e., periods of unemployment above its natural rate. Our results for the US economy between 1955 and 2015 show that credit contractions improve the identification of high unemployment states. Moreover, we find that credit cycles have a sizable out-of-sample predictive power on labor market slacks. This implies that the evolution of credit can be used as a leading indicator for economic policies.