What Is Driving The TFP Slowdown? Insights From a Schumpeterian DSGE Model

In this paper, I incorporate a Schumpeterian mechanism of creative destruction in a medium-scale DSGE framework. In the model, a sector of profit-maximizing innovators invests in R&D and endogenously gen- erates productivity gains, ultimately determining the economy's growth rate. I estimate the model using Bayesian methods on U.S. data of the last 25 years (1993q1-2018q4) in order to disentangle the key forces underlying the productivity slowdown experienced by the US economy since the early 2000s. In contrast with the previous literature, I exploit Fernald (2014) data on TFP, factor utilization and labour quality to discipline the production function, and find that the bulk of the TFP slowdown is due to a decrease in innovation's ability to generate TFP gains. These findings challenge the view of a large part of the literature, according to which the recent TFP dynamics in the US are mostly driven by demand slumps and/or liquidity crunches.