Total-factor-productivity measurement when equilibrium is temporary: A Monte Carlo assessment

Abstract

The use of total-factor-productivity ( TFP) indices to estimate rates of technical change is predicated on the assumption of long-run equilibrium in all markets. Unfortunately, temporary equilibrium is a ubiquitous feature of the real world. It is therefore important to know what sort of errors are made when short-run equilibrium prevails but long-run equilibrium is assumed. When we know how sensitive calculated TFP indices are to different sorts of misspecification, we will know when there is no simple alternative to econometric estimation. This paper reports the results of Monte Carlo experiments designed to assess the ability of temporary-equilibrium TFP indices to capture the properties of the underlying long-run technology as the features of this technology change. The experiments include varying the cost share of the fixed factor, the degree of substitutability among the fixed and variable factors, the degree of returns to scale of the long-run technology, and the variances of the market price of the fixed factor and of output.