The Shimer puzzle(s) in a New Keynesian framework

In this paper I shed light on the issues of the (low) volatilities of labor market variables implied by the search and matching model and the (high) values of the correlations between these variables and labor productivity. On the one hand, Shimer (2005) claims that “Not only there is little amplification, but there is also no propagation of the labor productivity shock in the [search and matching] model.” On the other, starting from Galì (1999) empirical evidence about the reaction of employment to a neutral positive technological shock seems to indicate a recessionary effect in the short term, thus casting doubts about the whole transmission mechanism as described by Shimer (2005) in line with a RBC framework. I claim that a New Keynesian model with nominal rigidities is able to replicate the set of moments of both volatilities and correlations; the model presents two distinctive features: employment decreases after a positive technological shock and the calibration strategy in choosing the vacancy posting cost is different with respect of Shimer (2005) and in line with the RBC tradition. I show also that the use of the traditional separable preferences in consumption and leisure worsens the Shimer's critique, via the consequences of wealth effects on labor supply.