Measurement Error and Time Aggregation: A Closer Look at Estimates of Output-Labor Elasticities

Abstract

This paper analyzes the effect of time aggregation on estimates of the elasticities of output with respect to employment and to average hours of work. The main goal is to get accurate estimates of production function parameters. Low frequency data generate better estimates of output-employment elasticity while high frequency data generate better estimates of output-average hours elasticity. This result comes from the fact that time aggregation increases (decreases) the bias in the estimate of the elasticity with respect to average hours (employment). Estimations of these elasticities at different data frequencies and numerical simulations illustrate this point. In addition, this estimation methodology shows that the elasticity of output with respect to employment is bigger than the elasticity of output with respect to average hours, as theory predicts, contradicting an established result in the literature.