As resources available for agricultural research and development become increasingly scarce, the Kenya Agricultural Research Institute (KARI) has taken steps to define a clear research agenda. This effort has taken the form of formal priority setting in various research programmes. Priority setting is a key step in the formulation of a research agenda because priority setting, almost by definition, increases the efficiency and relevance of our research programmes by adding valuable structure to our wide information base on agro-ecological and socio-economic conditions in the country, by focusing attention on client constraints and by specifying the potential impact of research on these constraints.

The process of establishing clear and systematic priorities in our research programmes allows us to take a proactive rôle in soliciting government and donor support for areas identified as vital to agricultural development efforts. Perhaps most important, the priority setting processes developed at KARI have helped to broaden participation in formulating our research agenda by including a wide range of stakeholders and clientsfrom outside the Institute and in the process widened and deepened our constituency base.

Within specific programmes, the priority setting process has comprised five interrelated and cumulative steps: (1) compiling the information base on relevant commodities and regions; (2) identifying programme research target zones and research themes; (3) eliciting the potential for technology generation and adoption; (4) estimating research-induced social benefits; and (5) establishing research priorities and medium-term resource allocation guidelines with programme stakeholders.

All but a handful of our commodity research programmes, which include a number of livestock research programmes, have completed this five-step procedure. A similar process has been developed for regional and factor-based research programmes, which should have clear research priorities and resource allocation guidelines in 1998. Once all programmes have completed their respective priority setting activities, we will embark on an Institute-wide priority setting exercise that will pit programme against programme with a view to establishing a coherent set of priorities for the entire organization.

This paper uses the United Kingdom (UK) government's Department for International Development (DFID) centrally funded renewable natural resources research strategy (RNRRS) programme as an example to answer the key questions of: (i) what research; (ii) why conduct research; and (iii) why set an agenda?

Donor research organizations may fund research through various channels. DFID funds research through its multilateral programmes (e.g. to the system of International Agricultural Research Centres which constitute the Consultative Group for International Agricultural Research); as components of its bilateral development programmes with individual countries (e.g. assistance to the development of the Kenya Agricultural Research Institute (KARI) and the Kenyan National Agricultural Research Programme) and sometimes through projects with non-governmental organizations (NGOs), as part of the Joint Funding Scheme. Research funded through these avenues is, however, part of a larger agenda to which the DFID has agreed but not necessarily set. In the interests of pursuing strategic and wider research objectives, the outputs of which may later be applied through development programmes, DFID also has a centrally controlled fund for research, which is not specific to any one country or organization. The part of that research programme relevant to natural resources is called the RNRRS (Research Task Group, 1994). Other donors have similar centrally funded research programmes and it is these to which the title refers and for which it is appropriate that the donor organization sets its own agenda.

The global agenda for livestock research must be led by the requirements for agricultural development, recognizing the integral and complementary rôle livestock play in sustainable agricultural systems. Demand for meat and milk will increase by more than 150 per cent over the next three decades, fuelled by the combination of income growth, population increase and urbanization. Increased demand will be primarily in developing regions, where current consumption per capita is low and livestock production systems are inefficient relative to those in developed countries. This situation encourages development to increase supply of livestock product. Research can facilitate sustainable livestock development to serve the needs of both producers and consumers.

The benefits already realized from past investments in research, primarily in developed countries, help make the case for investing in livestock research. Nevertheless, the resources available for research are limited; choices must be made and expected benefits and costs should guide priorities. These benefits and costs should take into account social and environmental as well as financial values. Procedures for valuing the contribution of livestock research to sustainable development are inadequate and, in themselves, are a priority in the global agenda.

Priorities for research differ but are generally linked across the levels of agricultural systems: household, community, landscape, national, regional and international. Specific priorities in the global agenda vary with livestock species (ruminant, non-ruminant), production systems (grazing, mixed, industrial), agro-ecological and socio-economic factors (especially, those differing for developed and developing regions) and whether research will be supported by public or private sector funding. The priorities for developing countries emphasize increasing productivity and efficiency; whereas for developed countries, more emphasis is given to food safety, zoonotic diseases and environmental issues. Non-traditional, but increasingly important priorities derive from the need for information and technologies to improve soil and water management to ensure long-term sustainability of livestock production systems and for socio-economic analysis to provide policy options for decision makers.

The global agenda for livestock research is broad, embracing the full spectrum from basic to adaptive research. The traditional animal sciences will continue to be important but there is increasing need to draw from crop and environment research, from human health and genetics research, inter alia. Results from research in developed regions can be adapted to the needs of livestock systems in developing regions. In return, research on genetic resistance developed through natural selection in livestock populations in developing regions, may provide environmentally friendly means to control livestock disease and parasites in developed regions.

The challenges posed by the global agenda are considerable. To meet these challenges, partnerships are required, linking skills, capacities and access to problems. The successes of these research partnerships will meet the requirements for sustainable agricultural development and ensure support for livestock research in the future.

The seminar, the aim of which was to develop policy recommendations to improve the livestock sector in eastern and southern Africa, was held from 28 July to 1 August 1997 in Mbabane, Swaziland. All countries were represented and a wide range of topics were discussed. These included livestock development in Africa, relative to the rest of the world, and meeting the food security needs of each country, for which production must increase at 4% per annum, rather than 2% as at present.

Topics considered included: constraints to livestock development, current status, global and African trends in the livestock sector, human resource needs, animal genetic resources, animal health constraints, data and information, research and extension, private sector contribution, participation and contribution by donors or international organizations and interactions between livestock, environment and wildlife.

Recommendations were made on: livestock, environment and wildlife interactions for extensive production systems and mixed crop and livestock interactions for intensive production systems, animal health and genetic resource constraints and policy information management and communication.

The region under discussion has a tremendous potential for livestock production, which has not been fully realized, and potential internal markets for livestock and livestock products. At present it lacks cohesive livestock development policies at both national and regional levels. The region carries more than 37%, 72%, 63%, 55%, 33% and 39% of Africa's human, cattle, sheep, goat, pig and chicken populations respectively (FAO, 1988).

The purpose of this paper is an attempt to summarize the proceedings of the seminar, highlighting the issues raised and recommendations made.

People in developed countries currently consume about three to four times as much meat and fish and five to six times as much milk products per capita as in developing Asia and Africa. Meat, milk and fish consumption per capita has barely grown in the developed countries as a whole over the past 20 years. Yet poor people everywhere clearly desire to eat more animal protein products as their incomes rise above the poverty level and as they become urbanized. Growth in per capita consumption and production has in fact occurred in regions such as developing Asia and most particularly China. Per capita consumption of animal proteins and use of cereals for animal food in Asia have both grown in the 3 to 5% per annum range over the past 20 years. By 2020, according to the International Food Policy Research Institute's IMPACT model projections, the share of developing countries in total world meat consumption will expand from 47% currently to 63%. Of the global total projected increase in meat consumption, 40% is from pork, 30% is from poultry and 24% is from beef. The latter helps mitigate the otherwise much larger decline in real beef prices expected through 2020. Projected annual growth in meat consumption in China of 3.2% per annum through 2020, up from 8.3% per annum from the early 1980s to the early 1990s, drives these results.

A rapidly expanding supply of feedgrains will be essential to achieving the desired production increases for livestock products without undue upwards pressure on grain prices, especially in view of the rôle of monogastrics and the relative increase in industrial production in developing countries. IMPACT projections under various technical and economic assumptions suggest that there is enough production supply response in world systems to accomplish these production increases smoothly. Sensitivity analysis of the impact of restrictions on China's ability to produce more feedgrains illustrates that in a system of linked global markets for cereals and livestock products, such restrictions are not effective at lowering Chinese livestock consumption, which is driven by global trade in manufactures, although they do lower Chinese livestock production. The resulting imbalance raises world food costs by one-third in 2020 over anticipated levels, encourages increased livestock exports from Latin America, discourages livestock exports from the USA and reduces meat and cereals imports and consumption in the poorer countries of Africa and Asia.

The demand for dairy products is rising steeply in developing countries with increasing per capita incomes, in particular in Asia and Latin America. In contrast consumption is relatively static in developed countries. Only about 6% of global milk production is traded with the European Union (EU) being the major exporter, followed by New Zealand and Australia.

A conceptual model of driving force, state and response (DSR) is used to describe the dynamic in which research agendas have to be set within the milk production and consumption environment. Two countries with contrasting dairy industries, Mexico and the United Kingdom (UK) are discussed in relation to the DSR model and the implications for their research agendas. Mexico is a large importer of milk powder and has a milk production industry which consists of predominantly small- to medium-sized farms in a range of climatic regions, with poor farmer to consumer infrastructure. The UK has a well developed industry from farmer to consumer but has problems of milk quota restrictions, declining subsidies and the impact of BSE. Consumer demand in Mexico is in general for more dairy products at competitive prices, whereas in the UK, consumer demand is declining and there is a greater emphasis on food safety, animal welfare and environmental implications of dairying. Consequently the research agendas for the two countries have to be different.

In both countries, researchers dominate the research agendas, having a substantial influence on research priorities and the topics and modes of research. If the problems relating to the milk production and consumption environment are to be addressed, the relevant stakeholders within the DSR model need to be participants in setting research agendas. This should lead to less inappropriate research being carried out and to improved adoption pathways. Participatory approaches to research agendas involve stakeholders both in the prioritization of research and in its implementation.

In sub-Saharan Africa (SSA), the association of crops and livestock in mixed farming systems generally benefits both enterprises. This paper focuses on the main contributions of livestock to crop production: the use of manure and animal draught power to produce crops and the investment of income from livestock into technologies that benefit crop production. In low-input, grazing-based feeding operations, manure is a vital soil fertility amendment. In these systems, penning livestock overnight on fields, fallow between cropping periods, returns both manure and urine to the soil and results in much higher crop yields than if manure only is gathered from stalls and spread onto fields. However, most farmers have insufficient manure to sustain food production. Nutrient harvests often exceed nutrient inputs, requiring a much greater use of fertilizers to arrest soil nutrient depletion. The opposite may be true for mixed farming where livestock are given food in confinement. In these emerging systems, the continuous importation of food (and fertilizer) can result in nutrient surpluses with subsequent soil nutrient build-up and loss. The contribution of animal power to crop production is relatively new in Africa. Animal power affects the amount of land cultivated by farmers, crop selection, the yield per farm and per ha, and on the participation and work load of people (family members and outside labour) involved in crop production and its associated activities. In addition to the impacts of manure and draught power on crop production, income derived from livestock is often invested in inputs that enhance crop production. At the ‘micro’ level, livestock income influences crop production (1) directly by allowing households to invest in productive inputs such as fertilizer, hired labour, and carts and (2) indirectly by allowing poor households to improve their nutritional status and, therefore, the productivity of their most important resource, their own labour. At the ‘macro’ level, increased livestock exports have a large stimulating effect on the demand for locally produced goods and services, particularly basic food crops. Thus, increasing the productivity of the livestock sector, including an emphasis on the policy and institutional environment influencing marketing and trade, is an important element of a development strategy focused on stimulating economic growth and alleviating poverty.

This paper draws on a recent multi-donor study on livestock and the environment and highlights how its findings might impact on research agendas worldwide. The soaring demand for meat and milk puts great pressure on the global natural resource base. About one-quarter of the world's total land area is used for grazing livestock. In addition, about one-fifth of the world's arable land is used for growing cereals for livestock food. Livestock production is the world's largest land user and may soon he its most important agricultural activity in terms of economic output. Livestock produce 13 billion tons of waste each year. A large part of this is recycled but, where animal concentrations are high, waste poses an environmental hazard. Water, already scarce in many parts of the world, is required not only for animals' drinking but also to grow food crops and for waste disposal. Pollution of land and water is another concern as is the impact by livestock, directly or indirectly, on biodiversity. Greenhouse gases from livestock and livestock waste contribute to global warming.

The paper provides an overview of critical livestock environment interactions, so-called hotspots, notably the overgrazing/degradation issue, deforestation, the ‘involution’ of mixed farming systems and the waste problem in areas of high animal concentrations. Measures that tackle only the superficial effects of environmental damage will never be as effective as a policy which attacks the underlying causes. Those causes are often deeply entrenched in what has become an almost universal fact: those who gain benefits from over-exploitation and degradation of the environment have not paid the full cost. Incorporating the environmental cost into the price of livestock products is critical to achieving sustainable development. The technological opportunities, and, therefore, the scope, for increasing livestock production, while simultaneously reducing the use of natural resources per unit of product, are enormous. Research needs to be channelled to reflect real scarcities of natural resources while respecting the need to maintain and improve livelihoods.

The main uses to which marginal and arid rangelands are put involve livestock production, tourism based on wildlife and ethno-tourism, and agriculture, i.e. crop production. There is minimal dry land forestry, sometimes as agro-forestry. The emphasis placed on these three main uses varies according to the ecological potential (i.e. climate, topography and soils) and accessibility to the areas.

Taking the Kenyan example, approximately 20% of the land is arid and used almost exclusively for livestock production while ethno-tourism runs a poor second in dry seasons because of inaccessibility. Current technology in Africa precludes extensive irrigation. Peak production of livestock is in the late wet season and early dry with marketing mostly in dry seasons. Over 50% of the land is semi-arid where all three uses are practised. Livestock production is still the most important and agriculture the least important, because rainfall is unreliable and erratic, wildlife populations are larger and so tourism is more important (e.g. Amboseli, Isiolo, Samburu). Agriculture occurs particularly in wet years and wet seasons.

Although land is only very locally suited to agriculture, permanent water sources, rivers and springs may enable year round settlement. Farmers of non-pastoral backgrounds (and even some pastoralists) wish to follow their traditions and attempt cultivation. This is occasionally successful in above average years of rainfall (two years in five) on good soils but fails in dry years when it also deprives livestock of essential traditional dry season grazing reserves.

Marginal areas occupy perhaps 12% of the land but are in high demand for all three use categories. Pastures are ideal for fattening livestock bred in more arid areas and they have a rapid turn-over. Wildlife populations are often at their highest in these areas, e.g. Laikipia, Mara and Nairobi park. Areas are relatively accessible on tarmac roads for year round viewing of wildlife. Agricultural resettlement has spilled over from higher potential lands where human populations are exceeding the carrying capacity.

Increasing food requirements have led to a greater demand for efficient land use and to diversification into new areas, e.g. eco-tourism, ostrich farming or the intensification of traditional uses such as camel rearing.

Lailipia District, situated mostly in marginal and semi-arid land is used as a case study. Here, successful conservation measures on mostly private land, which was formerly used by Maasai for subsistence pastoralism, has led to the largest population of wildlife in Kenya outside parks and reserves. At the same time land is used in part for crop production especially in the higher potential areas, but also wherever land is available for co-operative arable farmers to purchase. Livestock production remains however, the most widespread form of land use. The main seasonal variation in use is with crop production in the rains and game viewing in the dry seasons but extremes are less than in the lower rainfall areas.

Recent preliminary analysis of the economics of various forms of land use in Laikipia indicate that in those limited areas where agriculture is reliable (e.g. irrigated areas near rivers) returns may be as high as US$ 132 to 166 per ha per annum. Wildlife tourism which prevails in less well watered areas may yield US$ 4 to 5 per ha, while conventional livestock rearing yields from US$ 0.2 to 1.4 per ha per annum. Game cropping is the least well developed and the least productive but is accepted as a necessity by the Kenya Wildlife Service, particularly with regard to zebra which compete with livestock for resources. It yields only US$ 0.2 to 0.4 per ha per annum.

Wildlife and livestock occur together, except where there has been considerable outlay on electric fencing. Predators, especially lions and hyenas, are incompatible with livestock and together with certain wildlife which may act as disease vectors (e.g. buffalo) reduce income by US$ 0.5 per ha per annum. By contrast, the addition of camels, which are eco-friendly milk and meat producers, with no reduction of conventional stock, may increase livestock yields by US$ 0-4 per ha per annum.

Combined wildlife tourism, cropping and livestock, including camels, may yield US$ 4.7 to 6.4 per ha per annum, which although still less than 5% of agricultural yield, is the best that may be achieved at present on a sustainable basis. Crop production is highly dependent on rainfall which becomes less predictable the more arid the land. It may not be sustainable in the long term in its present form.

Current returns on investment are low for all forms of land use. Constraints to increasing returns are outlined. Research agendas need to be tailored to provide answers which could help minimize them. In particular, we need to refine our knowledge concerning the economics of the different options, both conventional and non-conventional.

This paper discusses some of the unique aspects of the interactions of livestock with the natural resources base. It argues the nature of these interactions make the use of a systems approach of particular importance when planning and executing livestock research if it is to be of any relevance to farmers in developing countries. The key issues are illustrated in two contrasting case studies.

In Nepal, shifts in land use patterns have led to marked changes in the availability and use of fodder resources. Farmers indicate that they are no longer able to use these optimally. Fodder from at least 90 different types of tree (some of which have yet to be properly classified botanically) are used to supplement the diets of buffalo, cattle and goats during dry seasons when food is in short supply. The nutritive value of each of these is affected by a wide range of management and environmental factors. Furthermore, diet selectivity means that fractions consumed differ markedly amongst both tree species and classes of livestock. It is clearly not feasible for researchers to evaluate this diversity using existing in vitro or in vivo indicators of nutritive value. Initial studies suggest that this variability is implicitly catered for in farmers' own assessments of relative nutritive values. A mechanistic understanding of this indigenous technical knowledge and the development of appropriate techniques for integrating it with models of the biological responses of animals to nutrients might, therefore, allow more effective assessment of strategies for tree fodder utilization.

Highland dairy systems in the humid tropics of Costa Rica (3000+ mm/year rainfall) could be very productive but suffer from a substantial dependence on imported inputs. Concentrate use is very high causing substantial underutilization of grazed pasture (the local resource), which is available year-round under these high-rainfall conditions. Over-supplementation with protein, which is the most expensive nutrient in the diet, at levels in excess of 200% of the requirement has been reported widely in these regions. Farmers want new strategies that enable them to manage their land and other resources in an alternative way, in order to be less dependent on grains and other imports. A decision-support system based on simulation and multiple criteria models representing a dairy farm has enabled the design and development of such strategies. Using this approach, farmers' objectives can be incorporated with a holistic understanding of the farming system in terms of management practices that will permit demand-driven animal production research at the farm level.

Livestock make a significant contribution to the world's supply of protein and energy. They occupy dry and cold areas where crop farming is not possible, as well as integrating with crops in warmer, wetter zones. In developing countries they play a major part in household dynamics and family social status. This implies that they should be given serious consideration when agricultural research agendas are set and resources allocated. This paper examines some of the processes used for research priority setting and comments on the likely impact of these processes in setting research agendas for livestock. One case study is considered in detail, that of the Kenya Agricultural Research Institute (KARI).

Methods for prioritization of research range from the very informal, where priorities are determined by discussion in small expert groups, to the comparatively formal and quantitative estimation of economic surpluses. Between those extremes are a variety of methodologies for selecting and ranking research programmes and projects. In a liberalized economy, market forces will play a large part in determining the research agenda; in a centrally planned economy, the national research agenda will be determined by the government, although the trend is towards focusing on the needs of the end user. No one method can guarantee results, since effective research prioritization depends on accurate prediction of future demand. All methods have a degree of subjectivity and may be biased by the selection of stakeholders involved in the debate.

This paper proposes that more rigorous methodology will tend to make results more objective, more transparent and by introducing an explicit market orientation will facilitate the transition from central planning to competitive bidding, but users of any method should be aware of its limitations. The economic surplus method, possibly the most rigorous currently available for setting a national research agenda, is limited by thefact that it does not require measures of social or environmental impact, and to include these requires an additional weighting process. Any trend towards methodological rigour, whether quantitative or qualitative, has costs in terms of data gathering, time and the necessary training to carry out the analyses.

KARI has over the last 10 years moved through the full spectrum of priority-setting methods from informal to formal. It is at present engaged in setting priorities for the next 5-year span, using an economic surplus approach. Some of its experiences and lessons are described in this paper, with particular reference to livestock programmes. The authors conclude that a systematic process of setting research agendas will, on the whole, be favourable to livestock. There has in some cases been a tendency to exclude them because they are harder to work with than annual crops, research can be more costly, their value is harder to estimate and benefits take longer to accrue. A rigorous process of’ estimating benefits from research, with a reasonably long time horizon, should provide a realistic assessment of the value of livestock to an economy and is likely to encourage investment in livestock research.

Livelihoods of most farmers in different ecological settings are local, extremely complex, diverse, dynamic and unpredictable. The situation of low-income farmers is even more complex: poor infrastructure; diverse and risk-prone agro-ecological conditions; strong interactions between crop, livestock, tree and fodder components of the farming system, between agriculture and the management of common resources, combining cropping and herding with off-farm activities; and diverse socio-economic conditions.

Until recently, agricultural research in developing countries had predominantly been conducted under conditions ‘controlled’ by researchers, on topics chosen predominantly by researchers and in ways largely determined by researchers. Concern over the complexity and diversity of situations faced, especially by low-income farmers, paved a way to the steady development of a range of participatory approaches which, in turn, led to the increasing recognition of farmers’ participation (though in varying degrees) in the research process, if research were to generate and spread appropriate agricultural technologies with some prospect of uptake by farmers.

The purpose of this paper is to provide information on ‘Farmers’ Participatory Research’ (FPR), i.e. research in which smallholders play leading roles, from diagnosis through implementation to assessment of results and dissemination. With particular emphasis on recent experiences in Ethiopia, the rationale, techniques, application and some results of FPR, carried out in different settings at different times of the year, are reviewed.

The paper highlights local peoples’ abilities in identifying and prioritizing their own problems; significant rôles, if allowed to participate, in identifying and prioritizing the ‘best-bets’ and, or, researchable issues, carrying out and evaluating research. The degree of smallholders’ participation in setting the research agenda, their involvement in the research process and ultimately success or failure depend on various factors. Of these: nature and state of problems limiting agricultural productivity; types and stages of research; expected duration of the research; time and resource availability; and favourable policy issues are considered to be amongst the most important.

In determining research priorities, as well as in evaluating the relative merits of different disease control options, there is an increasing demand for impact assessment, in order to justify and support research and development investments. In this process, quantitative epidemiology has a significant role to play. This paper first outlines the categories of impacts that diseases have on society and discusses how these can be measured. Using the example of tick-borne infections of livestock, the paper describes five quantitative epidemiological tools that can be used in impact assessment. These are: geographical information systems (GIS) to estimate disease distribution; mathematical models to quantify infection dynamics; observational field studies to link infection status with production loss; statistical models to enhance study design; and spreadsheet models for economic impact assessment. In each category, examples are given of recent studies.