Pro-cyclical Solow Residuals without Technology Shocks

Most Real Business Cycle models have a hard time jointly explaining the twin facts of strongly pro-cyclical Solow residuals and extremely low correlations between wages and hours. We present a model that delivers both these results without using exogenous variation in total factor productivity (technology shocks). The key innovation of the paper is to add learning-by-doing to firms technology. As a result firms optimally vary their prices to control the amount of learning which in turn influences future productivity. We show that exogenous variation in labour wedges (preference shocks) measured from aggregate data deliver around fifty percent of the standard deviation in the efficiency wedge (Solow residual) as well as realistic second moments for key aggregate variables which is in sharp contrast to the model without learning-by-doing.