By looking at changes in markups, crucial for productivity understanding, this study suggests that structural shifts in manufacturing dynamics may significantly contribute to the UK's productivity challenges.
Abstract
UK manufacturing firms have experienced sharp declines in productivity since 2008, whether measured by real revenue per worker or estimated total factor productivity. Less is known about trends in firms’ markups, which is important for understanding productivity dynamics. The estimation of markups is challenging without direct access to price and cost data, but they can be inferred using microdata on firms’ revenues and input use. In this paper we use two approaches to infer the evolution of aggregate markups for UK manufacturing firms. Both use estimates of the elasticity of substitution within industry subsectors.
Our principal approach involves assumptions about the structure of competition between firms at various levels of industry aggregation, while the second tests the robustness using the influential Hall/DeLoecker approach to infer the markup based on variable input cost shares. Both approaches show large declines in estimated UK manufacturing markups since the financial crisis, estimating a decrease in industry-level gross markups of approximately eight percentage points between 2008 and 2019. There are significant contributions from both within-firm declines and declines due to reallocation. As markup declines are associated with an adverse shift in the distribution of firm-level manufacturing productivity, our results indicate that structural dynamics in manufacturing industry likely play a large part in the UK’s productivity puzzle.
Keywords: Markups, monopolistic competition, manufacturing
JEL: D22, D24, D42, D43, E24, O47
This work contains statistical data from ONS which is Crown Copyright. The use of the ONS statistical data in this work does not imply the endorsement of the ONS in relation to the interpretation or analysis of the statistical data. This work uses research datasets which may not exactly reproduce National Statistics aggregates.
