Shocking stuff: technology, hours, and factor substitution

The reaction of hours worked to technology shocks represents a key controversy between RBC and New Keynesian explanations of the business cycle. It sparked a large empirical literature with contrasting results. We demonstrate that, with a more general and data coherent supply and production framework (“normalized” factor-augmenting CES technology), both models can plausibly generate impacts of either sign. We develop analytical expressions to establish the threshold between positive and negative contemporaneous correlations for both models. These will crucially depend on the factor-augmentation nature of the shock, the elasticity of factor substitution, the capital income share, and the reaction of consumption. The impact of technology on hours can thus hardly be taken as evidence in support of any particular business-cycle model. Our results are also important as: i) we introduce the concept of normalization for DSGE models and, ii) they may help interpret possible time-variation in technology and hours correlations over time.