SOAS Research Online

Abstract

The proximate objective of this study is to explore the power and limitations of linear programming as an analytical guide in the approach to agricultural policy formulation. We postulate that the role of farm policy is to change opportunities open to farmers in ways which induce desired changes in their behaviour. Further we postulate, tentatively, that farmer behaviour is governed by material constraints on their production and exchange possibilities and that changes in these constraints will be followed by changes in farmer behaviour. Agricultural policy can affect choices particularly by affecting market opportunities, resource availabilities and the availability of knowledge of new technologies. By assessing the consequences of such policies for the alternatives open to farmers it is argued that it is possible to improve the basis for policy decision making. To predict farmers' behaviour it is important that their choice boundaries and production/exchange opportunities be correctly identified. We develop the concept of the farmers' decision making environment as the context in which these choice boundaries must be defined. We articulate this concept, and the approach to the definition of choice boundaries, by applying it to a case study farmer in the Syrian village of Khaldeyeh. Having thus defined the farmer's choice boundaries we examine the farmer's present behaviour to determine whether it is consistent with our postulate that it is constrained by present opportunity and we find that it is. The nature and adequacy of consistency tests for this purpose present problems which are the subject of a prior discussion. Supported by our conclusions with regard to the validity of our approach to this point the analysis proceeds to explore the implications of a variety of policy measures for the farmer's production and exchange opportunities. We conclude on the relative significance of these possible measures for the production and exchange behaviour of our case study farmer. But for policy making the implications of alternative measures need to be generalized beyond the individual case study and the problems of such generalizations are thus also explored. Our approach looks at the problem of generalizing at the village level and uses a classification of village farmers according to the proportionality of their resource availabilities. The analysis of the implications of specific proposed policies reveals, especially, the expectation of an excess demand for labour. The discussion of generalization at the village level therefore focuses upon the problem of predicting the general equilibrium of farming output patterns and resource demands as this is governed by the workings of the labour market and by the process of innovation and adaptation in Khaldeyeh. Both of these are seen to be related to social characteristics of Khaldeyeh village which were not quantitatively investigated during the field work period and whose investigation would in any case have required competence outside the discipline of economics. Our model is therefore shown to be unable to predict the general equilibrium of the village. Our analysis concludes with an evaluation of linear programming in the context of this study.