A substantial literature deals with the problem of allocating resources for agricultural research. Ruttan traces the literature to work by Fishel in the early 1970s.1-2 In recent years, the literature on research planning has experienced a minor boom, perhaps driven in part by decreases in funding for public sector agricultural research. Two recent books on the subject are Alston, Norton and Pardey and Evenson, Herdt, and Hossain.3-4 Numerous articles and manuscripts address specific applications of the priority setting literature.
The literature distinguishes between "priority setting" and ex ante evaluation" This paragraph, and the following one, drew heavily from work by Evenson, Herdt, and Hossain.4 Taking these categories in reverse order, the latter involves relatively detailed accounting of the expected costs and benefits of narrowly defined research projects. For example, it might be possible to conduct an ex ante evaluation of a project aimed at breeding for powdery mildew resistance in bread wheat. By contrast, "priority setting" is usually taken to be the process by which resources are allocated across broad problem areas. For example, priority setting techniques might be used to allocate research funds across different crops or regions, or between plant breeding and entomological studies.
In principle, it would be possible to conduct priority setting exercises simply by aggregating information from ex ante evaluations of specific projects. In practice, however, there is seldom enough data to proceed in this fashion. Moreover, ex ante evaluations are perhaps too narrowly focused to be useful for priority setting. They typically ignore the interactions among different research projects; in some cases, there may be several projects that could achieve a particular objective.
Considered on their own, all of them may look like sensible investments, but at the aggregate level they may be redundant. Priority setting thus requires a certain breadth of perspective; it involves a kind of macro analysis that cannot be simply developed from the microanalysis of ex ante evaluations.
Was this article helpful?