Productivity measurement with adjustments for variations in capacity utilization and other forms of temporary equilibrium

Abstract

One of the most persistent puzzles addressed by recent productivity research is the substantial decline in measured productivity growth during the 1970's. In this paper we trace a substantial part of the measured decline to the fact that traditional methods of productivity measurement assume that producers are in long-run equilibrium when in fact they may be in short-run or temporary equilibrium. We utilize the Marshallian framework of a short-run production or cost function with certain inputs quasi-fixed to provide a theoretical basis for accounting for temporary equilibrium. Within this theoretical framework it is the value of services from stocks of quasi-fixed inputs which should be altered rather than the quantity. The empirical application to U.S. manufacturing data 1958–81 shows that, depending on the measurement procedure, one can attribute somewhere between 18% and 65% of the traditionally measured decline in TFP growth between 1965–73 and 1973–81 to the effects of temporary equilibrium.