We develop testable hypotheses for utility maximization given risk averse producers based on a general specification of the utility function. This is a direct expansion of the model posed by Pope (1978). Empirical tests using production data with a translog specification indicate that utility maximization does not always hold.
