Informations et ressources scientifiques
sur le développement des zones arides et semi-arides

Accueil du site → Doctorat → Allemagne → Integration of risk and multiple objectives in priority setting for agricultural research : the case of the national dairy research program in Kenya

Humboldt-Universität zu Berlin (1999)

Integration of risk and multiple objectives in priority setting for agricultural research : the case of the national dairy research program in Kenya

Gierend, Albert Johannes

Titre : Integration of risk and multiple objectives in priority setting for agricultural research : the case of the national dairy research program in Kenya

Auteur : Gierend, Albert Johannes

Université de soutenance : Humboldt-Universität zu Berlin

Grade : Doctor rerum agriculturarum (Dr. rer. agr.) 1999

Résumé partiel
Public agricultural research systems in developing countries have entered an era of resource scarcity. Funds from national governments and the international donor community do not increase as much as in the past decades. There are many examples where research funding has been significantly reduced, e.g., in the international research centres of the CGIAR group. At the same time agricultural research must continue with its efforts to contribute to food security, poverty alleviation and sustainable production systems by investing in new and expensive research areas such as biotechnology, resource conservation and the like. These developments have put agricultural research systems under increasing pressure to undertake long-term planning and make a more selective use of their available resources. Priority setting has become a key word to research management which subsumes a diverse set of planning and decision making tasks. Priorities must be set for the allocation of research resources to different countries, regions, research programs, commodities and factors, as well as for different research institutions. A variety of formal priority setting methods exist in practical applications. The fundamental part of making decisions on the allocation of resources is the assessment and comparison of the likely impact of research activities on pursued research objectives. Planning and decision making in priority setting of agricultural research can be characterised as a complex task. The complexity is due to the broad mandate of agricultural research in agricultural development including a variety of social and economic objectives. Yet such enormous responsibilities do not only make the assessment of the impact of research an onerous task but also complicates decision making on the type of research to fund and the future directions of a research program. Further complexity is added through uncertainty which is notorious in the planning environment of agricultural research. Unstable agricultural markets, exposure of production to climatic hazards and several internal sources of uncertainty in the research system, such as long planning horizon, the risks in the development, dissemination and adoption of new technologies, make research planning a highly conjectural and uncertain venture. Examples from the literature show some evidence for the recognition of decision complexity and uncertainty in priority setting but methodological approaches to these problems are yet not satisfactory. The aim of this study is to apply formal economic methods for an improved treatment of decision complexity and uncertainty in Priority setting. These methods are illustrated by using a priority setting example from Kenya. In 1996, the Kenyan Agricultural Research Institute (KARI) conducted a priority setting exercise for its national dairy research program where a set of 19 proposed dairy research projects had to be evaluated and prioritised. A prerequisite for the application of the economic methods is the development of a stochastic evaluation system. In a preliminary step the deterministic economic surplus framework within which the welfare effects of the 19 dairy research activities are calculated across different regions in Kenya is transformed into a stochastic system. Numerical simulation is used to reproduce stochastic input variables. For the dairy case study this is done by explicit incorporation and reproduction of the probability distributions of the research projects’ yield increase parameters. Evaluation outcomes are probability distributions (risk profiles) of the net present value and cost-benefit ratios as the two major economic indicators for the research projects. In a next step stochastic dominance analysis is employed as a decision rule for uncertain prospects to compare and rank the set of research alternatives based on their stochastic returns to research

Annonce

Version intégrale

Page publiée le 11 juin 2008, mise à jour le 2 novembre 2018