An analysis of factors which contribute to differences between actual and programmed optimum organization on individual farm units

The purpose of this study was to investigate factors which may limit linear programming as a predictive tool. It has specifically centered attention on components of linear programming models and characteristics of farm operators which create differences between actual and linear programmed farm organizations. Twenty farms in Wasco County, Oregon, were selected for the empirical investigation. Data on enterprise costs, technical coefficients, and restraints were obtained from each farmer. Additional information on age, education, farming experience, family size, and income were also obtained for each farmer. Three programming models were constructed for each farm. Model I represented year-to-year choices among alternative levels of participation in government wheat and feed grain programs. Separate models were developed for each of the years from 1963 to 1966 which represented the program alternatives in these four years. The purpose of Model I was to calculate profit maximizing solutions for individual farms in this short-run context, to make comparisons with actual decisions of the farmers within the same framework, and to isolate factors which created differences between the actual and programmed solutions. Model II was constructed to predict individual farm organization where the planning horizon was sufficiently long for changes to occur in resource use patterns and enterprise combinations. Land, family labor, and in some cases operating capital, were treated as fixed resources. The objectives of the Model II analysis were to evaluate the degree of association between the actual and profit maximizing programmed farm organizations and to aid in determining factors which caused differences between the two solutions. Model III was the same as Model II except that its objective function required the least-cost organization for obtaining a given level of income. Its purpose was to provide an alternative representation of the objectives which guided the farmers in their management decisions. The Model I analysis indicated that farmers did not make the profit maximizing decisions with respect to choices of government programs. The main factors which hindered them were continual changes in programs and associated incentives and inability to individually assess the economic consequences of alternative program participation. The more educated farmers made the better decisions while poorer decisions were made on highly productive farina. The Model II analysis indicated that maximum profit models did not accurately predict the decisions of farmers. However, they did perform better in the short run than in the longer run situations. They also predicted production for major enterprises better than for supplementary enterprises. Model III with its minimum cost objective described the organization of more than 50 percent of the farmers more completely than the maximum profit model. Errors in specification of enterprises to include in the models were a major source of differences between actual and programmed organization. Enterprises must not only be physically and economically feasible but also psychologically acceptable to the farmer. Characteristics such as education, family size, age, and experience were found to be associated with the farmer's management objectives.