The evaluation of alternative decision models: a case of crop rotation in Northern Thailand

Abstract

Economic evaluations of multi-location testing results at the Thailand Farming Systems Research Institute have not been sufficient in terms of whole-farm decision planning. If no risk is involved in farm decision planning, farmers should act as profit maximizers. since risk is inherent in agriculture, farmers should be concerned with risk when they make decisions on farm investments. Therefore, it is hypothesized that risk decision making models predict northern Thai rice farmers' behaviors better than profit maximization models. Two elements in the mean-variance (E-V) decision analysis are farmer's risk preferences and risk perceptions. A quadratic function is used to represent farmer's attitudes to risk. Farmer's risk perceptions concerning crop production are measured by farmer's expectations of yields and prices. The expected mean-variance (E-V) efficiency frontier for each farmer is derived by a quadratic programming algorithm. The E-V efficiency frontier together with the utility function are used to derive optimal farm plans. The Target MOTAD model, using a linear programming algorithm, is included as an alternative decision model, which involves safety-first considerations (i.e., cash cost and variable cost as a target income level). The results show that the expected utility maximization model predicts actual farmers' behavior more accurately than the expected profit maximization model. Moreover, the Target MOTAD model performs quite well and better than the expected profit maximization model in predicting actual farmers' behavior when the selected target income level portrays their risk preferences and goals, such as variable cost of farm investment. The results suggest the importance of risk in farm decision making. Investment decisions of farmers concerning the amount of land allocated to mungbean production preceding the rice crop in the yearly rotation are significantly affected by competition from alternative crops for limited farm resources, such as land, labor and cash capital. Therefore, formulation of programs or policies, such as the integration of a new crop into farmer's existing cropping systems, should take risk and interdependency among cropping enterprises into consideration.