Productivity, Market Power and Capacity Utilization When Spot Markets are Complete

Our test of price-taking behavior looks at the choice of capacity rather than the choice of output. It is motivated by a complete spot markets model in which goods are distinguished by the selling probabilities in addition to other characteristics. When output is explained by total man-hours and a capacity utilization proxy, the coefficient of the first variable is the elasticity of capacity with respect to fixed labor. Under competition and risk neutrality this coefficient is equal to an average labor share. We use this observation to interpret Abbot-Griliches-Hausman's regressions and to argue that once the capacity utilization proxy is included in the regression, Hall's data at the manufacturing level fail to reject the joint hypothesis of competition and risk neutrality. It is also argued that the coefficient of total man-hours does not tell us anything about monopoly power once the capacity utilization proxy is omitted from the regression.