TFP, Labor Productivity and the (Un)Observed Labor Input: Temporary Agency Work

Abstract

The study focuses on the question of whether productivity estimates are biased due to the emergence of a new input that is usually omitted: temporary agency worker (TAW). The study analyzes labor productivity and TFP by means of a structural approach using a representative dataset of German manufacturing firms. The empirical results show, once TAW is taken into account, that: i) labor productivity in most manufacturing sectors is significantly lower; ii) average TFP differs significantly in most sectors; but iii) the coefficients for regular labor are not significantly different between estimations with and without TAW.

Keywords: Temporary agency work, total factor productivity, labor productivity, omitted variable bias, structural productivity estimation

JEL Classification: D24, J24, L24, L60