Joint Risk Preference‐Technology Estimation with a Primal System

Abstract Applied studies of the firm in a risky environment have concentrated either on the firm's technology or on its risk preferences. These models result in generally inconsistent and inefficient parameter estimates. A primal model is proposed which allows a firm's preferences and technology to be estimated jointly in the presence of risk. The model is applied to Iowa corn production and estimated technology parameters are compared with those from other approaches. Modest risk aversion leads to inelastic (even backbending) per‐acre supplies and input demands. Yield heteroskedasticity in inputs leads to supply heteroskedasticity in prices, especially for risk‐neutral firms.