Explaining the durable goods co-movement puzzle with non-separable preferences: a bayesian approach.

A standard two-sector sticky price model with flexibly priced durables depicts negative co-movement between durables and non-durables after a monetary policy shock, which is at odds with the empirical evidence. This paper proposes a new channel, non-separable preferences with a small wealth effect on labor hours, as a solution to the co-movement puzzle. In contrast to the standard model where the aggregate hours remain relatively unchanged after the contractionary policy shock, aggregate labor hours fall along with the fall in the labor wage, thereby discouraging production in both the durable and non-durable goods sectors. We further compare our model's explanatory power with two other alternatives that can resolve the puzzle by using a Bayesian approach. Based on the log marginal likelihood and cross-correlation function comparison exercises, we find evidence that the data strongly favor both the alternative specifications over our baseline model. More specifically, the model with a working capital channel and habit formation gives the best fit to the data, especially for cross-correlations between durable and non-durable consumption.