@article{alston_scobie_1983, title={DISTRIBUTION OF RESEARCH GAINS IN MULTISTAGE PRODUCTION SYSTEMS - COMMENT}, volume={65}, ISSN={["1467-8276"]}, DOI={10.2307/1240890}, abstractNote={In a recent paper, Freebairn, Davis, and Edwards (hereafter FDE) have examined an interesting and neglected aspect of the distribution of benefits from research. Agricultural production is seen to comprise three stages: input supply, farm production, and farm to retail marketing. FDE have explored the distribution of the benefit stemming from a downwards shift in the supply function at any one of the stages. They conclude these benefits are distributed according to the elasticities of consumer demand and supply facing each stage. Furthermore, under the special assumptions they employ, this distribution is independent of which of the three supply functions is shifted by research. They conclude, albeit with due caution, that the choice of agricultural research projects should not be limited to farm production research alone, especially in view of the cost shares of purchased inputs and marketing services in the retail value of food. "In fact," they argue "they [farmers] may receive greater benefits from off-farm-oriented research than from farm-oriented research" (p. 39). This finding has potentially important implications for agricultural research policy. It immediately spurs one to ask whether research funded by the farmer or the fisc might have neglected opportunities for research related to farm inputs or marketing. To reach their result FDE assume (a) all supply and demand curves are linear, (b) research induces parallel shifts in supply curves, (c) farmers cannot substitute nonfarm inputs for inputs supplied by farmers in production of the farm product, and (d) marketing firms cannot substitute marketing inputs for raw farm products in the production of retail commodities. The purpose of this note is to emphasize the importance of these elasticities of factor substitution in determining the distribution of research benefits between the three stages. To do this we use a simple but general two-factor model in which we focus attention on the elasticity of factor substitution between marketing services and raw farm products in the third stage, marketing. This model is not restricted to linear supply and demand curves, nor to an elasticity of substitution of zero. We maintain the assumptions of parallel shifts of supply curves induced by research and a perfectly elastic supply of marketing services. In a revision of FDE's application to the U.S. hog industry, the distribution of research benefits is shown to depend crucially on the elasticity of factor substitution. Their result, that such distribution is independent of the stage in which research shifts the supply curve downwards, follows only for the special case in which that elasticity is zero. This is clearly a restrictive assumption. A reexamination of some of the policy issues raised by FDE, once the assumption is relaxed, would seem to be warranted.}, number={2}, journal={AMERICAN JOURNAL OF AGRICULTURAL ECONOMICS}, author={ALSTON, JM and SCOBIE, GM}, year={1983}, pages={353–356} }