The Empirical Dynamics of Misallocation

Abstract

Factor misallocation has been emphasized as one of the main sources of differences in aggregate TFP. This paper investigates the empirical dynamics of both capital and labor misallocation. Exploiting a balanced firm-level panel dataset covering manufacturing and services industries in several European countries, the analysis relies on the combination of three well-known and complementary econometric approaches: the (conditional) beta-convergence framework of Barro and Sala-i-Martin (1991, 1992), sigma-convergence, and the distribution dynamics methodology developed by Quah (1993a, 1996). Even though beta-convergence in both inputs allocative efficiency holds conditionally as well as unconditionally, and is consistent to a wide range of robustness checks, full distribution convergence in terms of capital misallocation does not occur at all and it proceeds at a low pace with respect to labor misallocation.