Previous Page Table of Contents Next Page

Priority setting for small ruminant research

L. Reynolds

International Livestock Centre for Africa
P.O. Box 80147, Mombasa, Kenya

Priority setting: the key issue
The overall research agenda
Choosing priorities
Principles to follow


The limited resources available to national agriculture research and extension systems in Africa will allow only top priority areas to be funded. Priority identification must take into account farmers' objectives and those of national planners. A farming systems approach is essential to determine farmers' objectives and constraints and to identify appropriate solutions. Such information is an essential prerequisite in the formulation of research programmes, which must be backed up by a thorough review of the literature. Small ruminants may or may not be a national priority.

Définir les axes prioritaires de recherche sur les petits ruminants


L'insuffisance des ressources dont ils disposent oblige les services nationaux de recherche et de vulgarisation agricoles en Afrique de ne financer que les domaines revêtant une priorité absolue. Pour identifier ces priorités, les objectifs des petits exploitants et des planificateurs nationaux doivent être pris en considération. L'adoption d'une approche systèmes agraires est essentielle si l'on veut déterminer les objectifs des paysans et les contraintes auxquelles ils se heurtent, et identifier des solutions appropriées. Ce type d'information est un préalable indispensable à la formulation de programmes de recherche, qui devront être appuyés par une étude bibliographique approfondie. Les petits ruminants peuvent ou non figurer parmi les priorités nationales.


Africa is experiencing political and economic changes, and national research and extension organisations must adapt to survive in the face of these changes. Across sub-Saharan Africa (SSA) the level of funding from governments for agricultural research has fallen below the level needed for active and productive use of trained manpower. While the present level of funding is a reflection of the economic realities for the Governments of SSA, it also accounts for the limited progress that has been made in African agriculture. Donors, be they national or international are more than ever before concerned about impact. Technology developed from agricultural research has not always been adopted in SSA over the past 20 years. Promises of technological breakthroughs and of adoptable packages have not been kept. Scepticism on the part of paymasters must be overcome, realistic objectives set and adoptable technology produced if financial support is to be sustained, let alone increased. Sustainability as an issue has been accepted as an objective for biological research, but less attention has been paid to the sustainability of the research institutions and the accompanying infrastructural support required for research and extension.

Populations in SSA are expected to expand threefold by 2025 from the present level of 500 million. Over the same period, urbanisation is expected to rise from 31 to 57% of the population (Winrock International, 1992). Urban consumers tend to require more processed foods, high value cereals, livestock products and vegetables. To meet this demand agricultural production must increase and become more market oriented to supply urban areas. Market forces can stimulate production, as seen in the variations in cereal sales in Zimbabwe in response to changes in the level of guaranteed prices, but market forces alone will not be enough. More land will be cropped to supply the requirements of higher populations, but uncultivated productive land is already limited. Unit area productivity must therefore increase. Storage, processing and transport must become more efficient. Research, in collaboration with producers and advisory services, has a critical role to play in these developments.

Nations must decide on their priorities and ensure that they are providing enough funds for essential research. Donor-driven research programmes are affected by the availability of funds. They may not conform to national priorities, and may divert the limited number of trained personnel away from more important work. In most countries in SSA, very few activities can be funded solely from national resources. External funds should therefore be directed towards supporting the priority research areas. Only after those areas have been adequately covered should other work be considered. If and when external support declines, high priority areas will be less affected because core funding will already have been assured from the national exchequer. This is a risk-avoidance strategy against variations in donor countries' economic and political climates, similar to that adopted by farmers in minimising annual differences in the biophysical climate. However, this strategy will be effective only if focussed on the highest priorities.

The ratio between staffing and operational costs is 9:1 in many countries. This ratio is unsustainable and must therefore be narrowed. A ratio nearer 3:2 would produce sufficient operational funding to allow useful research to continue. When operational funds are in short supply staff still receive salaries, but cannot carry out any useful function. Unpalatable though it may seem, closing research stations and reducing staff levels to allow those remaining to operate effectively is essential, given that additional funding is unlikely (Etcher, 1989).

Priority setting: the key issue

Agricultural research has largely focussed on biological problems, and there is a shortage of economists and social scientists in NARS (National Agricultural Research Systems). However, policy and socio-economic considerations must be considered in order to increase the probability of developing adoptable technology. Technology must be appropriate to the prevailing policies and economic conditions of the country, or to those expected when the research matures. A valid research objective would be to model the effect of alternative policies on agricultural production, so that policy makers can make informed choices. It follows that adaptive research, which must include on-farm studies in collaboration with development/extension agencies and farmers, must try to anticipate the national environment at the time when the research comes to fruition. While it can be argued that "on-the-shelf" technologies and production systems are a valid output from research, they may not fit the present or projected economic and political environment. They may therefore be a luxury when current needs are so pressing.

The overall research agenda

A research organisation must decide where to focus its work and how to allocate resources between commodities, species and research areas, including policy to fulfill its mandate. It must decide how important small ruminants are, and where and for which section of society research could have the most impact. Numbers of animals, distribution and values of production are required to quantify importance. The first stage in deciding on the allocation of resources is to consider the following elements: the estimated value of production, adjusted by modifying factors to allow for national research objectives; the probability of research success; efficiency in the use of research resources; and distribution of impact.

Table 1 shows the value of production of the most important commodities in sub-Saharan Africa. Sheep and goat meat is the least important livestock item, after milk and beef.

Table 2 shows the output of meat and milk from ruminants in the regions of SSA. Of the small ruminants mutton is more important in East Africa, while goat meat predominates in West Africa. From the limited data available, small ruminant milk appears more important in East Africa than in West Africa, but much of the production is consumed by young stock.

Table 1. Gross value of annual production of the twenty most important commodities in SSA, 1983-85 (US$ million).

Fuel (non-coniferous)




Sawlogs (non-coniferous)


Sweet potato










Other industrial wood


Beef/buffalo meat








Sheep/goat meat












Inland fish


A summary of methods for priority setting has been presented by Norton and Pardy (1987), and Contant and Bottomley (1988). Checklists and scoring systems are the most practical methods for setting priorities in agricultural research. However, there are practical difficulties in attempting to rank programmes dealing with aspects of the natural resource base (e.g. soil fertility) alongside those dealing with commodities (e.g. small ruminant meat). Benefits from commodity research programmes can be expressed directly in monetary terms. Benefits from research on factors such as soil can only be estimated from assumptions regarding increased productivity over a range of commodities.

Table 2. Output ('000 mt) of meat and milk in the regions of SSA, 1987.










West Africa







Central Africa







East Africa







Southern Africa







* Data not available from some countries in each region.

ISNAR (International Service for National Agricultural Research) have assisted NARS in a number of countries to prioritise their research programmes. These countries include the Dominican Republic, Ecuador, Uruguay, Peru, Gambia, Zimbabwe and Kenya (Cessay et al, 1989; Norton and Pardy, 1987; Norton, 1991).

Choosing priorities

There are so many scientifically valid research activities that some formal way is required to identify policies for technological and socio-economic research that will have real impact. Selected research activities should stimulate action by policy makers or development and extension colleagues, and be capable of adoption by farmers.

Identifying problems worthy of research can be started by researchers or by farmers. An example of the first route is goal-oriented programme planning (GOPPing). GOPPing is a component of the ISNAR approach to long-term programme design as described by Collion and Kissi (1991). It was used by ILCA, assisted by national scientists, when it produced its long-term strategy document (ILCA, 1987). A multidisciplinary group, including representatives from different research, development and extension groups participate in the exercise. Farmers are usually not included because it is assumed that the group already knows the problems that limit on-farm production, and hence can identify research routes to improve output.

The second approach involves researchers meeting farmers and extension staff to identify problems and propose solutions (Chambers et al, 1989). Farmers are the key players in this approach as they know the problems and will have to work with any proposed solutions.

However, merely identifying a problem says nothing about its importance in relation to other problems or constraints. As indicated earlier, problems are many but funds are limited, therefore priorities must be set. Priorities across agro-ecological zones need to be considered, as these will affect the balance of work between regions. If a weighting method is used, issues of efficiency, equity and security need to be considered. Criteria are required to measure the contribution of each of these issues toward the fulfillment of the goals of the institution. Efficiency criteria should include: importance of the problem; adoption rate; probability of research success; environmental impact; research costs; comparative advantage; and compatability with national goals. Equity criteria will cover: farm scale; producer/consumer benefits; gender; and agro-ecological zones. Security issues would include production stability and food security.

Consideration must also be given to whether research leads to an increase in the use of abundant resources and a saving of scarce resources, and also the number and severity of research problems. Other areas are non-duplication with transferable research from outside, and the current emphases in the research programme. The process is consultative and iterative. The criteria and rankings are produced in discussions with policy makers, researchers, development/extension workers and producers.

The priorities of other countries in the region and of IARCs (International Agricultural Research Centres) such as ILCA, and their compatibility with the priorities of the NARS should also be considered. If, for example, we consider that priority should be given to improving the nutritional value of crop residues, we would need to collaborate with crop breeding institutions and ensure that their priorities reflected our concerns. There is a vast quantity of published information on most subjects, therefore any research process must start with a literature search. All involved in research planning should be aware of the results of past research and development projects, so that we build on success and avoid repeating the same mistakes. Researchers in Africa often have only limited access to adequate library facilities and are unaware of published and ongoing work inside their own country, let alone outside. The ILCA library provides a literature search service for livestock researchers.

Most research activities undertaken by NARS in SSA will be applied or adaptive in nature, as that is where their comparative advantage lies. Traditionally, researchers have set the agenda, but, given the lack of adoptable output, we need to explore other ways of planning research to ensure applicable results. Output from applied and adaptive research that cannot be adopted by farmers is valueless. Strategic research that is not based on problems identified on-farm and which does not have a target group of farmers in mind as beneficiaries should not be started. Sumberg and Okali (1989) go so far as to propose that adoption by farmers is sufficient validation of a technology, even if we are unable to identify or quantify the effects of the technology.

Principles to follow

All applied and adaptive research must have a strong on-farm base. For the humid and subhumid zones this means that the low priority given to small ruminants by farmers must be taken into account. In mixed crop/livestock farming systems where crops are the more important, small ruminant research will be only one component of multidisciplinary on-farm studies. Monodisciplinary small ruminant research on-station can be justified only when the need has arisen from on-farm observations and the results can be taken back for application on-farm. This is particularly relevant to universities where suitable component work for student projects can be identified from on-farm studies.

As small ruminants are of secondary importance to small-scale mixed farmers, they will only become prominent if justified by market demand and price. Projects in which small ruminant production is the central aspect have little chance of success if imposed on farmers unless their objectives have been incorporated into the design. With small farmers, risk avoidance usually takes precedence over increased production.

Little progress has been made in humid or subhumid Africa by research or development projects for increasing small ruminant meat production on small-scale farms. In general, the disease risks associated with raising youngstock are too high to attract farmers. Youngstock mortality rates of 40% are common. Thus for increased productivity the first priority should be to reduce the death rate. Unfortunately, small farmers do not have ready access to veterinary services and governments cannot afford the cost of expanding the service. Improved management, including better health through improved feeding, can reduce susceptibility to disease and improve survival (Reynolds and Adediran, 1988; Reynolds and Ekwuruke, 1988). The integration of small ruminants into alley farming is one such improved management option (Kang and Reynolds, 1989; Kang et al, 1990).

Market importance can be illustrated by comparing two dairy goat schemes, both started about 10 years ago in East Africa, based on crossbred exotic-indigenous animals. In Burundi around 10,000 crossbreds have been accepted by farmers and the milk produced is purchased by a factory for processing into cheese, which is then sold to urban consumers (Rey, personal communication). Regular income from the sale of milk is the key factor encouraging adoption. In Western Kenya a much smaller number of animals have been distributed to farms, but only 10% of the does are milked. All the milk is used for household consumption (Mbabu et al, 1991). The main attraction of the crossbred animals to Kenyan farmers is reported to be their large body frame. At both locations surplus male kids are purchased by the project. Time will tell which of the two approaches will be sustainable when external funding ceases, but so far the market-led scheme in Burundi has had more impact. The scheme in Kenya provides no clear-cut economic advantage to farmers that would encourage their continued interest when external project funding ends and staff are withdrawn.

Research into improved feeding must be linked to a strong market demand for the animal product, and its profitability demonstrated by ex-ante economic analysis. Research into forage production has only produced adoptable technology when linked to milk production. Specialised short-term fattening schemes, designed to produce animals for festival peaks, may be viable when the feed sources are crop residues or agro-industrial by-products. The economic viability of meat production based on planted pasture is poor, as shown by Doppler (1980) for beef cattle in Togo.

Over most of Africa, small ruminants are kept as free roaming scavengers. Many flocks in the humid zone of West Africa have no adult male, relying on those in other flocks in the village. The more active males will sire the most offspring, from which replacement sires will be drawn. At village level there is therefore the danger of inbreeding. In addition the most rapidly growing males are sold or slaughtered first, so the few males that reach breeding status have been selected for slow growth (Reynolds and Adediran, 1993). The selection, distribution and use of superior sires may be a valid research and development objective. However, maintaining improved genes in village herds will require infrastructural support superior to that which present governmental organisations have been able to maintain. Farmer demand for improved breeding stock will be generated if there is a market demand for the eventual product, as shown by smallholder dairy schemes with cattle in Kenya (Walshe et al, 1991).

Much of the day to day management of small ruminant herds is in the hands of women, even when the flock owner is male. Research is needed to determine the flexibility that women have to alter flock management, use different feed stuffs and to market animals at different times or in different ways. Will the woman benefit directly as a result of the changes, or will the benefits accrue to the owner? The contribution to household income from small ruminants is less than 5% of farm income, and probably half that figure for household income. However, their role as an investment for savings may account for the wide ownership pattern (ILCA, 1979). Improved animal productivity is equivalent to a higher rate of interest on the savings account.

Small-scale farming is a high risk business, and with few capital reserves to fall back on in bad years farmers minimise risks by undertaking a wide range of activities. Lack of appreciation of this aspect of the strategy of small farmers is one reason why the traditional biological research approach has failed. Increasingly, farmers are looking to off-farm activities and involvement in the monetary economy as risk avoidance strategies, in addition to or in place of established farming practices (Downing, 1989). In much of Africa small farm households derive over half of their income from off-farm sources such as remittances (von Braun and Pandya-Lorch, 1991). The positive side of this picture is that the extended family can generate funds off-farm for investment in farming. The negative side is that the more innovative members of the household move away from the farm unless the returns from farming appear attractive in comparison to employment prospects in urban areas.

It is important that all small ruminant researchers, biological, social and economic, combine to provide data and analyses to policy makers so that the consequences of alternative courses of action can be predicted. Modelling is the way this objective can be achieved. Piecemeal research will not provide the depth and breadth needed for modellers to construct and test their models. Research and development groups must cooperate within and across both national and research borders to prevent unnecessary duplication of effort.


In the past, livestock research has been ineffective in producing technologies that can be adopted by small-scale farmers. NARS have insufficient resources to stretch further than top priority areas. In any one country these may or may not include small ruminants. A top-down allocation of resources for small ruminant research is not effective; appropriate research can only be identified in consultation with farmers, developers and extensionists.

For humid and subhumid zones integrating sheep and goats more closely into the farming system should be a long-term objective, taking into account the emphasis placed by small-scale farmers on risk avoidance. Market demand for the product and the economics of production should be important considerations in determining the priority of biological research. Another key factor is clarifying who will implement the technology at farm level, and who will reap the benefits.

The need for multidisciplinary work at farm level is overwhelming, but the costs of such work can be high. Cooperation between research, development/extension and farmers, within and across national borders will be necessary to achieve the conditions for success. Networking can help to build these links.


von Braun J and Pandya-Lorch R. 1991. Income sources of malnourished people in rural areas: microlevel information and policy implications. IFPRI Working Papers on Commercialisation of Agriculture and Nutrition 5. IFPRI (International Food Policy Research Institute), Washington, DC, USA. 206 pp.

Cessay S. Gilbert E, Mills B. Rowe J and Norton G W. 1989. Analysis of agricultural research priorities in the Gambia. Project Paper 16. ACIAR/ISNAR (Australian Centre for International Agricultural Research/International Service for National Agricultural Research), The Hague, Netherlands.

Chambers R, Pacey A and Thrupp L A (eds). 1989. Farmer first. Intermediate Technology Publications, London, UK. 218 pp.

Collion M H and Kissi A. 1991. An approach to long-term program design. Working Paper 37. ISNAR (International Service for National Agricultural Research), The Hague, Netherlands. 53 pp.

Contant R B and Bottomley A. 1988. Priority setting in agricultural research. Working Paper 10. ISNAR (International Service for National Agricultural Research), The Hague, Netherlands. 19 pp.

Doppler W. 1980. The economics of pasture improvement and beef production in semi-humid West Africa. GTZ (German Agency for Technical Cooperation), Eschborn, Germany. 195 pp.

Downing T E. 1989. Climatic variability, food security and smallholder agriculturists in six districts of central and eastern Kenya. Disssertation, Clark University. 276 pp. In: Dissertation Abstracts International. 1989,49: 2347.

Eicher C K. 1989. Sustainable institutions for African agricultural development. Working Paper 19. ISNAR (International Service for National Agricultural Research), The Hague, Netherlands. 38 pp.

ILCA (International Livestock Centre for Africa).1979. Small ruminant production in the humid tropics. ILCA Systems Study 3. ILCA, Addis Ababa, Ethiopia. 122 pp.

ILCA (International Livestock Centre for Africa). 1987. ILCA's strategy and long-term plan. ILCA, Addis Ababa, Ethiopia. 99 pp.

Kang B T and Reynolds L. 1989. Alley farming in the humid and subhumid tropics. Proceedings of an international workshop held at Ibadan, Nigeria, 10-14 March 1986. IDRC (International Development Research Centre), Ottawa, Canada. 251 pp.

Kang B T. Reynolds L and Atta-Krah A N. 1990. Alley farming. Advances in Agronomy 43:315-359.

Mbabu A N. Nolan M F and Gilles J. 1991. An evaluation of SR-CRSP's techpacks in western Kenya: a year-long study. In: Proceedings of the Ninth S R-CRSP Scientific Workshop. Held on 27-28 February 1991 at Ketri-Muguga. KARI (Kenya Agricultural Research Institute)/SR-CRSP (Small Ruminant Collaborative Research Support Programme), Nairobi, Kenya. pp. 35-48.

Norton G W. 1991. Methods for agricultural research priority setting: Implications for research planning in Zimbabwe. ISNAR (International Service for National Agricultural Research), The Hague, Netherlands.

Norton G W and Pardey P G. 1987. Priority-setting mechanisms for National Agricultural Research Systems: Present experience and future needs. Working Paper 7. ISNAR (International Service for National Agricultural Research), The Hague, Netherlands. 33 pp.

Reynolds L and Adediran S. 1988. The effects of browse supplementation on the productivity of West African Dwarf sheep over two reproductive cycles. In: Bosman H and Smith O B (eds), Goat production in the humid tropics. PUDOC (Centre for Agricultural Publishing and Documentation), Wageningen, The Netherlands. pp. 83-91.

Reynolds L and Adediran S. 1993. Composition of village goat herds in southwest Nigeria. Small Ruminant Research (in press).

Reynolds L and Ekwuruke J O. 1988. The effect of Trypanosoma vivax infection in West African Dwarf sheep at two planes of nutrition. Small Ruminant Research 1:175-188.

Sumberg J and Okali C. 1989. Farmers, on-farm research and new technology. In: Chambers R. Pacey A and Thrupp L A (eds), Farmer first. Intermediate Technology Publications, London, UK. pp. 109-114.

Walshe M J, Grindle J, Nell R and Bachmann M. 1991. Dairy development in sub-Saharan Africa. World Bank Technical Paper 135. The World Bank, Washington DC, USA. 94 pp.

Winrock International. 1992. Assessment of Animal Agriculture in sub-Saharan Africa. Winrock International Institute for Agricultural Research, Morrilton, Arkansas, USA. 125 pp.

Previous Page Top of Page Next Page