The decision to double crop: an application of expected utility theory using Stein's theorem.

A nonlinear, single-equation acreage allocation model derived from expected utility theory using Stein's theorem for covariance decomposition is developed. This decomposition allows the covariance of utility parameters and revenues to be expressed in terms of measurable variables. The model is applied to state-level data to estimate the relative importance of various economic factors in determining the pattern of double cropping wheat and soybeans in the southeastern United States. The hypothesis that the riskiness of returns is important in the aggregate is rejected for some of the states, although the constrained expected utility model without the effect of risk performs better than a standard wheat acreage response model.