Abstract
We contribute to an emerging literature that brings the constant elasticity of substitution (CES) speci cation of the production function into the analysis of business cycle uctuations. Using US data, we estimate by Bayesian-Maximum-Likelihood methods a standard medium-sized DSGE model with a CES rather than Cobb-Douglas (CD) technology. We estimate a elasticity of substitution between capital and labour well below unity at 0.15-0.18. In a marginal likelihood race CES decisively beats the CD production and this is matched by its ability to t the data better in terms of second moments. We show that this result is mainly driven by the implied uctuations of factor shares under the CES speci cation. The CES model performance is further improved when the estimation is carried out under an imperfect information assumption. Hence the main message for DSGE models is that we should dismiss once and for all the use of CD for business cycle analysis.