Interpreting the Hours-Technology time-varying relationship

We investigate the time variation in the correlation between hours and technology shocks using a structural business cycle model. We propose an RBC model with a Constant Elasticity of Substitution (CES) production function that allows for capital- and labor-augmenting technology shocks. We estimate the model using US data with Bayesian techniques. In the full sample, we find (i) evidence in favor of a less than unitary elasticity of substitution (rejecting Cobb-Douglas) and (ii) a sizable role for capital augmenting shock for business cycles fluctuations. In rolling sub-samples, we document that the impact of technology shocks on hours worked varies over time and switches from negative to positive towards the end of the sample. We argue that this change is due to the increase in the elasticity of factor substitution. That is, labor and capital became less complementary throughout the sample inducing a change in the sign and size of the the response of hours. We conjecture that this change may have been induced by a change in the skill composition of the labor input.