Using Time-Series and Cross-Section Data to Estimate a Production Function with Positive and Negative Marginal Risks

Abstract

Two production-function models with error components for time and firms and a heteroscedastic disturbance are proposed. Unlike previous models, these two permit the variance of output to increase or decrease as one of the inputs is increased. When the models and some suggested estimators are applied to the pastoral zone of eastern Australia, the results indicate that labor, water, and possibly fencing, are likely to reduce the variance of wool production; and that sheep, and buildings and land, are likely to increase the variance.