The Drivers of Revenue Productivity: A New Decomposition Analysis with Firm-Level Data

Abstract

This paper aims to derive a methodology to decompose aggregate revenue TFP changes over time into four different components – namely physical TFP, mark-ups, quality and production scale. The new methodology is applied to a panel of EU countries and manufacturing industries over the period 2006-2012. In summary, patterns of measured revenue productivity have been broadly similar across EU countries, most notably when we group them into stressed (Italy, Spain and Slovenia) and non-stressed countries (Belgium, Finland, France and Germany). In particular, measured revenue productivity drops for both groups by about 6 percent during the recent crisis. More specifically, for both stressed and non-stressed countries the drop in revenue productivity was accompanied by a substantial dip in the proxy we use for TFP in quantity terms, as well as by a strong reduction in mark-ups. Demand also suffered a conspicuous decline. Our results suggest that non-stressed countries seem to enjoy a stronger recovery in terms of fundamentals like quantity TFP, demand and mark-ups than stressed countries. Yet, their overall performance in terms of revenue TFP recovery does not necessarily align with the above analysis which is due to some possible deterioration in the resource reallocation, signaled in our framework from the lower covariance between the two components we split revenue TFP.

Keywords: decomposition, production function estimation, demand, productivity, markups

JEL Classification: E32, O47, D24