What benefits are to be derived from improved farm program planning approaches? - The role of time series models and stochastic optimization.

摘要

This paper examines whether there is room for improvement in farm program decisions through the incorporation of formal mathematical optimization in the practical planning process. Endeavoring to estimate the potential for improvement, we investigate four cases of German cash crop farms over the last six years. The formal planning approach includes a systematic time series analysis of farm-specific single gross margins and a stochastic optimization model. To avoid solutions that would possibly exceed the respective farmer's risk tolerance, the apparently accepted variance of the observed program's total gross margin, which represents an observable reflection of the individual farmer's risk attitude, is used as an upper bound in the optimization. For each of the 24 farm/year combinations, the formal model is used in a quasi ex-ante approach that provides optimized alternative programs. The total gross margins that could have been realized if the formally optimized programs had been implemented are then ex-post compared to those that were actually realized. We find that the farmers could have increased their total gross margins noticeably if they had used the more sophisticated formal planning model. However, we also find that the superiority of formalized planning depends on the quality of statistical analysis and the resulting forecasting model. Using our approach for practical decision support would require that farmers first specify their "own" production programs to reveal the variance they accept. Alternative programs could then be provided which would lead to increased expected total gross margins without exceeding the respective farmer's accepted total gross margin variance.