News and Labour Market Dynamics in the Data and in Matching Models

This paper uses a vector autoregression model estimated with Bayesian methods to identify the effect of productivity news shocks on labour market variables by imposing that they are orthogonal to current technology but they explain future observed technology. In the aftermath of a positive news shock, unemployment falls, whereas wages and the job finding rate increase. The analysis establishes that news shocks are important in explaining the historical developments in labour market variables, whereas they play a minor role for movements in real activity. We show that the empirical responses to news shocks are in line with those of a baseline search and matching model of the labour market and that the job destruction rate and real wage rigidities are critical for the variables' responses to the news shock