SDR/TAC:IAR/96/6
TAC MEMBERS ONLY
CONSULTATIVE GROUP ON INTERNATIONAL AGRICULTURAL RESEARCH
TECHNICAL ADVISORY COMMITTEE
Sixty-Ninth Meeting, IRRI Hqs. Los Banos (Philippines), 25-30 March 1996
DRAFT
CGIAR PRIORITIES AND STRATEGIES
FOR RESOURCE ALLOCATION
DURING 1998 - 2000
(Agenda Item 3)
TAC SECRETARIAT
FOOD AND AGRICULTURE ORGANIZATION OF THE UNITED NATIONS
March 1996
CGIAR PRIORITIES AND STRATEGIES
FOR RESOURCE ALLOCATION
DURING 1998-2000
Table of Contents
Page
1. Introduction 1
l.l. Background 1
1.2. Implementation of 1992 Recommendations 1
1.2.1. Priorities 1
1.2.2. Strategies 2
1.3. Outline of Report 3
2. The Expected Development Scenario 4
2.1. The Socioecological Environment 4
2.1.1. Demographic Changes and Expected Responses 4
2.1.2. The Changing Patterns of Demand 6
2.1.3. Perspectives in the Productive Management 
of Resources 6
2.1.4. The Impact on Natural Resources 9
2.2. The Institutional Environment 10
2.2.1. The Surrounding Environment 10
2.2.2. The Internal Environment 11
2.3. Possible Items for a Discussion on CGIAR Strategies 13
2.3.1. Research Strategies 13
2.3.2. Organizational Strategies 13
ii
3. Analytical Framework 15
3.1. Introduction 15
3.2. The Framework 16
3.3. The Process 16
3.4. Agricultural Research, Productivity, and Income Growth 18
3.5. Economic Growth and Poverty Alleviation 21
3.6. Agricultural Research and the Conservation of
Natural Resources 22
3.7. Methods for Setting Priorities for
Agricultural Research 25
3.8. Poverty as an Element in Priority Setting 26
3.9. Other Elements Considered in Priority Setting 29
3.10. Conclusions 32
4. Overview of Outcome of Recent Studies Relevant to 
CGIAR Priorities and Strategies 33
4.1. Background 33
4.2. The Lucerne Declaration and Action Programme 33
4.3. World Agriculture Towards 2010 34
4.4. 2020 Vision for Food, Agriculture and the Environment 37
4.5. Genetic Resources Policy Committee 38
4.6. Task Force on Sustainable Agriculture 39
4.7. Task Force on Ecoregional Approaches 39
4.8. Action Plan to Strengthen NARS-CGIAR Partnership 40
4.9 TAC Commissioned Studies 40
4.9.1. Stripe Study of Genetic Resources in the CGIAR 40
4.9.2. Inter-Centre Review of Rice 41
4.9.3. Livestock 42
4.9.4. Review of CGIAR Commitments in West Africa 44
4.9.5. Inter-Centre Review of Root and Tuber Crops 45
4.9.6. Soil and Water Aspects of Natural Resources 
Management Research 46
4.9.7. Public Policy and Public Management Research 47
4.9.8. Institution Strengthening Research and Services 48
4.9.9. ICLARM Mid-Term Review 49
5. Analytical Process: Activities 51
5.1. Increasing Productivity 52
5.1.1. Germplasm Enhancement and Breeding 52
5.1.2. Production Systems Development and Management 53
iii
5.2. Protecting the Environment 55
5.3. Saving Biodiversity 56
5.4. Improving Policies 57
5.5. Strengthening National Programmes 58
6. Analysis By Production Sector and Commodity 60
6.1. Production Sectors 60
6.2. Analysis by Commodity 61
7. Systemwide Programmes and the Ecoregional Approach 69
7.1. Introduction 69
7.2. Background 69
7.2.1. Terminology 70
7.3. An Overview of CGIAR Experience in the Initiation and 
Implementation of Systemwide Programmes and the 
Ecoregional Approach 71
7.4. Analysis and Recommendations 74
7.4.1. A Classification of the CGIAR’s Systemwide 
Activities 74
7.4.2. TAC’s Proposals for the Management of 
Systemwide Activities 75
7.4.3. Will New Systemwide Activities be Required 
During the Period 1998-2000 77
8. Recommendations on Future Priorities and Strategies 80
8.1. By Activity 80
ANNEXES
I. Activities and their Definitions 
II. Methodology
III. Crops, Livestock, Forestry and Agroforestry, Fisheries
IV. Selected Tables of Quantitative Analysis
V. Synthesis Tables on Systemwide Initiatives/Programmes
CHAPTER 1 - INTRODUCTION
1.1. Background
This report presents TAC’s current views on CGIAR priorities and strategies for 
resource allocation in 1998-2000. Priority setting is essential in order to ensure efficient use 
of increasingly scarce resources by directing them to activities that most effectively help to 
meet the CGIAR’s goals. As formulated in the Lucerne Declaration, "the mission of the 
CGIAR is to contribute, through its research, to promoting sustainable agriculture for food 
security in developing countries".
TAC produces an updated report on CGIAR priorities and strategies every five 
years. The last report was completed in 1992 and previous reports have been prepared in 
1973, 1976, 1979 and 1986. The attached report has been prepared a year early because 
there was a perceived need for an updated report after the Lucerne Consultation on the 
renewal of the CGIAR, and the realization that there was a need to advance the medium-term 
resource allocation cycle by one year. The 1992 report provided the basis for resource 
allocation during 1994-98 but, as proposed by TAC in its 1996 Research Agenda paper, this 
cycle would be completed a year early during 1997 and the new three-year cycle would start 
in 1998. The Committee’s recommendations on priorities, as reported in this paper, are 
made at the broad System level. TAC will consider the programme priorities of the centres 
when it assesses their medium-term plans. In its considerations on future priorities, the 
Committee has given substantial attention to the fact that the CGIAR System is only one 
component in the global agricultural research system for developing countries and commands 
less than 4% of its resources. TAC also carefully considered the outcome of each of the 
regional fora of national research systems where NARS expressed their views on future 
CGIAR priorities.
1.2. Implementation of 1992 Recommendations
In its 1992 report, TAC made recommendations on future CGIAR priorities by 
region, activity, commodity, agroecological zone, and production sector.
1.2.1. Priorities
• Region
TAC considered that the share of resources allocated to sub-Saharan Africa (SSA) 
and to WANA should be modestly reduced while the share allocated to Asia in particular, 
and Latin America should be increased. By 1996 the allocation of CGIAR resources had 
decreased from 43% to 40% in SSA, and from 13% to 11% in WANA. The allocation to 
Asia and Latin America had increased from 29% to 31 % and from 15% to 18% respectively. 
Overall, the broad directions recommended by TAC have been followed although the 
increased share of CGIAR resources to Asia is still well below that recommended by TAC.
2
• Activities
TAC recommended a significant increase in research efforts on natural resources 
conservation and management and on socioeconomics, public policy and public management. 
It also recommended a reduction in investment in production systems work and in 
strengthening NARS. These recommendations were implemented broadly although TAC 
notes with concern the erosion of support to research on germplasm enhancement and 
breeding.
• Commodities
TAC had reaffirmed the priority given to cereal and root and tuber crops. Among 
food legumes, it proposed a reduction in the level of priority to the phaseolus bean and 
pigeonpea, while increasing the priority to soybean and groundnut. Overall, TAC’s 
recommendations have been broadly implemented except for the reduction in investment in 
pigeonpea research. Investment in livestock research was well below the level that TAC had 
recommended.
• Agroecological Zones
TAC recommended a shift towards a greater effort in tropical areas and on the basis 
of available data this recommendation has been followed up on.
• Production Sectors
TAC did not make a recommendation on the level of priority across production 
sectors but noted that the proposed new programmes in forestry and fisheries should not be 
funded at the expense of critical research needs in crops and livestock. While these new 
production sectors brought some new resources to the CGIAR, given the funding crisis in 
the CGIAR during 1992-94, it is also clear that there has been some competition for funding 
between the different production sectors.
1.2.2. Strategies
• Global and Ecoregional Activities
TAC envisaged the CGIAR as having two major types of activities: global and 
ecoregional. TAC developed the basic framework for what it called the ecoregional approach 
to research which had three aspects: conducting applied and strategic research on the basis 
of sustainable production systems; improving productivity in the ecoregion by drawing in 
appropriate global research activities; and strengthening the cooperation with national 
partners and developing transnational mechanisms for collaboration.
It was considered that the global community did not yet have an effective paradigm 
for sustainable improvement of productivity. Identifying such a paradigm and making it 
operational was seen as a goal of truly international relevance and significance. By 1996, 
the CGIAR was well underway in implementing TAC’s views on this matter. Programmes
3
have now been set up in each of the major ecoregions and there has been a rapid expansion 
of the CGIAR’s work in this area.
• Collaboration
TAC considered that new modes of operation were needed for both the ecoregional 
approach and for closer collaboration with other international sources of expertise. 
Substantial progress has since been made on the development of consortia and on much 
closer collaboration with national research systems and advanced research institutes.
• Institutional Options
At the request of the CGIAR, TAC carefully considered the optimal future structure 
of the CGIAR. The Group chose, however, not to proceed with the restructuring of the 
CGIAR while it was launching its process of renewal. As indicated in its 1992 report, TAC 
also proceeded with stripe reviews of rice, genetic resources, roots and tubers, livestock, 
policy and management, institution strengthening, and postharvest technology research.
• Systemwide Programmes
In 1992, TAC had recommended that in addition to support of CGIAR Centres, the 
Group should also support Systemwide programmes which addressed issues of high priority 
to the System as a whole but not necessarily to the individual centres concerned. Several 
such programmes were developed and implemented as reported upon elsewhere in this paper.
1.3. Outline of Report
(TO FOLLOW)
4
CHAPTER 2 - THE EXPECTED DEVELOPMENT SCENARIO
As a background for TAC’s decision-making on priorities and strategies for the 
coming years, this chapter attempts to discuss development trends in agriculture1, with a 
special focus on developing countries. The first section on the socioecological environment 
highlights demographic trends and associated demands, with special reference to the poor; 
to then discuss trends in the productive management of resources and corresponding research 
strategies, including environmental impacts. The second section deals with latest changes in 
the institutional environment, both outside and inside the CGIAR. Finally, regarding TAC’s 
discussion on future CGIAR strategies, in the last section a set of possible items for the 
agenda is raised.
2.1. The Socioecological Environment
Population growth, agricultural performance and environmental degradation are the 
overarching concerns in agricultural development. Poverty and environmental degradation 
are closely linked, often in a self-perpetuating negative spiral in which poverty accelerates 
degradation, and the latter exacerbates the former.
2.1.1. Demographic Changes and Expected Responses
In the coming 25 years the world population is expected increase by 2.3 billion 
people, 93% of them in the developing world. Absolute population increase will be highest 
in Asia (1.6 bln) and lowest in Latin America and the Caribbean (240 min)2. However, in 
relative terms, Africa will more than double its present population, going from 0.6 to 1.4 
bln. Furthermore, projections indicate that this time no significant progress will be made on 
the incidence of poverty, that had declined in the two decades to the mid-1980s. The absolute 
number of 1.1 bln people that are poor today will remain the same in the year 20003, as 
shown in Table 1.
The highest incidence of poverty is encountered in South-Asia, where close to 50% 
of the population is below the poverty line, followed by 19% in Sub-Saharan Africa, 15% 
in East Asia, and 10% in Latin America and the Caribbean. But they are projected to 
increase by 40% in Africa, which will then account for 27% of the developing world’s poor. 
Rural poor make up more than 75% of the poor in many Sub-Saharan and South Asian 
countries. Urban poor are a slight majority in Latin America, although the poorest of the 
poor are still found in rural areas. Studies on rural poverty identify the small farmers, the
Agriculture and natural resources use are inextricably related. Hereafter, therefore, agriculture is understood 
throughout to mean the development of agriculture and natural resources, including forestry and fisheries.
2 Pinstrup-Andersen, P. and R. Pandya-Lorch, 1994. Alleviating Poverty, Intensifying Agriculture, and 
Effectively Managing Natural Resources. IFPRI - Food, Agriculture and the Environment, Discussion Paper 1. 
Washington, D.C.
3 World Bank, 1992. World Development Report. New York, Oxford University Press.
5
landless, the women, the nomadic pastoralists, the artisanal fishermen, the indigenous ethnic 
groups and the displaced persons as the functionally vulnerable groups in the rural sector4.
Table 1. Number of People Below the Poverty Line5, 1990 and 2000
REGIONS 1990 (millions) 2000 (millions)
Latin America and the Caribbean 108 126
Middle East and North Africa 73 89
Sub-Saharan Africa 216 304
East Asia 169 73
South Asia 562 511
TOTAL 1128 1103
There is a general consensus that population growth and raising incomes will 
increase the global demand for food by 2025 to more than double current production levels6. 
There are diverging views on the capacity to mobilize resources to meet such demands. 
Conventional estimates give reasonable hope that they can be met at the global level without 
price increases, while others say that this can result only from the mining of natural capital, 
i.e., at the cost of future production. Most agree, however, that meeting demands in 
developing countries will require them to significantly increase their capacity to feed 
themselves. Nonetheless, food shortages are expected to occur in both Asia and Africa, and 
may be especially severe in the latter. If present trends continue, by 2025 Africa could well 
have an annual food gap of 214 min tons, as compared to the current 11 min. Estimates for 
Asia show the expected gap to reach 255 min.
Present production levels of captured and cultured fish amount to 90 and 12 min tons 
respectively. In developing countries fish make up 19% of animal protein consumption (4% 
of total), while they provide only 0.006% of dietary energy supplies. As total marine catches 
are not likely to exceed 100 min tons by 20107, there will be severe constraints to increasing 
aggregate fish production. Aquaculture production may, however, reach the 15-20 min tons 
level in 2010, assuming that the prevailing growth rate is maintained.
4 FAO, 1995. World Agriculture: Towards 2010, An FAO Study. Ed. by N. Alexandratos
5 Taken as below US$ 370 of 1985 in purchasing power parity dollars
6 McCalla, A.F., 1994. Agriculture and Food Needs to 2025: Why We Should be Concerned. CGIAR - Sir John 
Crawford Memorial Lecture, October 27 - Washington, D.C.
7 FAO, 1995. World Agriculture: Towards 2010, An FAO Study. Ed. by N. Alexandratos
6
The quantity of wood used directly in generating energy is about 1.8 bln m3, around 
half of all the wood consumed. This, and the 300 min m3 of recovered residues from the 
manufacturing industry, make a total equivalent to 0.52 bln tons of oil, or approximately 5% 
of the world energy consumption. Fuelwood consumption in developing countries represents 
80% of their annual wood production, and is expected to grow at 1.4% per year, lower than 
population growth. It would then reach the 2.4 bln m3 in 2010.
2.1.2. The Changing Patterns of Demand
Given rapid population and income growth, market demands for foodgrains and 
livestock products are expected to grow much faster in developing than developed countries8. 
It is estimated that average per capita demand for foodgrains in developing countries will 
grow 0.4% per year between 1990 and 2020, and demand for livestock products by 1.5%. 
As a consequence, over that period countries market demand for foodgrains in such countries 
will grow by 75% to 1.7 bln tons, and that for livestock by 155% to 110 min tons.
Urban population of least-developed countries is projected to increase by 4.6% per 
year, 43% expected to be urbanized by 20259. There will be more than 200 cities with 
populations over 1 min, but the greater part of the urbanized people will live in cities of less 
than 0.5 min. This urbanization trend will increase problems of food supply and distribution, 
compounded by income differences. Incomes of certain segments of urban populations are 
rising rapidly, leading to increasing demand for more expensive and diversified foods, such 
as high quality cereal, livestock, fish and horticultural, and forest (spices) products. 
However, the majority of urban dwellers in most developing countries remain with limited 
purchasing power, making the supply of low-cost but nutritious food a pressing concern.
2.1.3. Perspectives in the Productive Management of Resources
The goal of the "renewed" CGIAR is to conduct research that will help liberate the 
deprived and disadvantaged from the grip of extreme poverty and hunger. The central themes 
of the CGIAR Vision are then: less poverty; a healthier, better nourished human family; 
reduced pressure on fragile natural resources; and people centred policies for sustainable 
development10. The emphasis on poverty would indicate a shift from the one on 
sustainability of natural resources that guided the past TAC analysis of priorities and 
strategies for the CGIAR. It should not be seen, however, as a deviation from such a goal. 
It is a new stage building upon the knowledge and insights of the preceding ones, in the 
evolutionary path towards a broader view of agricultural development in response to changing 
socioecological environments.
8 IFPRI, 1995. A 2020 Vision for Food, Agriculture, and the Environment: The Vision, Challenge, and 
Recommended Action. Washington, D.C.
9 UN Economic and Social Council, 1995. Concise Report on the Monitoring of World Population Trends and 
Policies - Report of the Secretary General, January.
10 Serageldin, I., 1995. Foreword. In Background Documents on Major Issues. Ministerial-Level Meeting, 
Lucerne - 9-10 February. CGIAR.
7
A brief overview of CGIAR shifts in emphasis may serve to trace the evolutionary 
path, and to show how the various "paradigms" complement each other11. The first stage 
emphasized production aspects, focusing on crop improvement through "seed-embodied" 
technologies that resulted in higher "piles" of rice and wheat, albeit with uneven adoption 
rates. These triggered the poor farmer-centred socioeconomic stage focused on "constraints 
research" in small-scale systems, and broaden the analysis to the human-ecological 
environment where technologies had to fit. This wider analytical context, and the impact on 
society of an increasing environmental degradation, shifted the emphasis towards natural 
resources, with special focus on their conservation. In recognition that the sustainability of 
natural resources could not be pursued independently or in opposition to the interests of the 
poor, and particularly to the rural poor, the present emphasis is now shifting to the stage of 
environment-agriculture-poverty "integration", with poverty alleviation as the central nexus 
between agricultural production and environmental degradation.
As indicated by the overview, research strategies have been adapted to prevailing 
circumstances influencing farmers decision-making. In a schematic way, two are the 
prevailing resource management "macro" trends for the coming 20 years: "market"- and 
"consumption"-oriented. The former includes producers that manage resources primarily to 
supply growing demands by an increasing urbanized population. Their behaviour is driven 
by external changes in demands and technologies, following the "induced-innovation" that 
drove the green revolution to its great success12. These market-driven producers will 
probably continue to focus on tradable commodities, requiring a re-invented green revolution 
that integrates "seed-embodied" technologies with "management" of the resource base. The 
former should further increase the yield potential of new varieties, and the latter should allow 
the sustainable expression of such potentials; moving further the "production" strategy to the 
first and the second post-green revolution phases13.
Producers closer to the "consumption" paradigm exploit land heterogeneity primarily 
to meet such needs, and complementarily to serve special market "niches". Their 
"intensification" behaviour responds to increasing population pressures through labour- 
intensive landscape arrangements and increasing cropping frequencies14, making them prone 
to an "integration" strategy. Their intensification is induced largely by endogenous 
mechanisms, derived from experimentation and accumulated wisdom. Market ventures are 
based on high-value, marketable products, as those produced from trees/shrubs and
11 Rhoades, R., 1989. Evolution of Agricultural Research and Development Since 1950: Toward an Integrated 
Framework. IIED, Gatekeeper Series No. SA12.
12 Hayami, Y. and V. Ruttan, 1985. Agricultural Development: An International Perspective. Baltimore, Md; 
John Hopkins University Press.
13 Byerlee, D., 1994. Technology Transfer Systems for Improved Crop Management: Lessons for the Future. 
In Agricultural Technology: Policy Issues for the International Community. Ed. by J. Anderson. CAB 
International/World Bank
14 Kleiver, K.M. and G. A. Schreiber, 1994. Reversing the Spiral: The Population, Agriculture, and 
Environment Nexus in Sub-Saharan Africa. The World Bank.
Tiffen, M., M. Mortimore and F. Gichuki, 1994. More People, Less Erosion: Environmental Recovery 
in Kenya. John Wiley & Sons.
8
horticulture crops. Such "non-basic" commodities facilitate a dynamic diversification of their 
economy, and are pliable to the landscaping approach to overcome constraints in natural 
resources. The full exploitation of this model, however, seems to be linked to the promotion 
of rural manufacturing. The processing of primary products plays an incentive role for rural 
development through synergistic effects between agricultural and non-agricultural growth, 
which promise to alleviate poverty beyond the effects of production increases alone15.
Studies in the last years have documented the intensification effect of population 
pressure on land use, largely through the labour-intensive creation of "resource" capital. 
What this model requires for further development is the critical link between intensification 
and material well-being16. It has been suggested that this link is still weak because 
smallholders, especially in Africa, have not found technological breakthroughs that shift per 
capita production and incomes to new levels. Given the nature of the model, such innovations 
might benefit from both a "pluralistic" approach (combining traditional and "modern" 
technologies) and the "blending" of new information into old practices17. Their generation 
would require the contribution of multiple sources through a participatory approach18.
The "production" and the "integration" perspectives should not be seen as exclusive 
of each other, but rather as strongly complementary ways to achieve food security through 
sustainable agricultural development. Actually, a basic condition for the diversification from 
staple food crops to highly-valued commercial commodities advocated in the integration 
strategy is a sustainable increase in crop production to outgrow demand, so that such 
diversification is profitable. Efforts should be intensified to maintain growth trends in the 
staple food crops perspective, together with the search for resource management systems that 
combine consumption crops with marketable commodities. "Otherwise, agricultural 
development through significant diversification will soon be trapped by the tradeoff between 
food and cash crops, as was observed in the past...with consequent stagnation, poverty and 
inequality"19.
15 Von Braun, J. and R. Pandya-Lorch, 1991. Income Sources of Malnourished People in Rural Areas: 
Microlevel Information and Policy Implications. IFPRI - Working Papers on Commercialization of Agriculture and 
Nutrition No. 5. Washington, D.C.
16 Hyden, G., R.W. Kates, and B.L. Turner II, 1993. Beyond Intensification. In B.L. Turner II, G. Hyden and 
R. Kates (Eds) Population Growth and Agricultural Change in Africa.
17 Gallopin, G.C., M. Winograd, and I. A. Gomez, 1991. Ambiente y Desarrollo en America Latina y el Caribe: 
Problemas, Oportunidades y Prioridades. Grupo de Analisis de Sistemas Ecologicos. Bariloche, Argentina.
18 Biggs, S.D., 1989. A Multiple Source of Innovation Model of Agricultural Research and Technology. 
Agricultural Administration (Research and Extension) Network Paper 6, Overseas Development Institute - London.
19 Hayami, Y. and K. Otsuka, 1994. Beyond the Green Revolution: Agricultural Development Strategy into the 
New Century. In Agricultural Technology: Policy Issues for the International Community. CAB/World Bank
9
2.1.4. The Impact on Natural Resources
At present rates of population growth, developing countries will depend on 
dwindling area of cropland per person and declining access to forests, rangelands and 
fisheries. In Asia, for example, the current 0.15 hectares of available cropland per capita will 
fall to a mere 0.09 hectares by 202520. Moreover, it is estimated that 5-12 min has are lost 
annually to severe degradation, defined as attainable annual output at a fixed level of non­
land inputs. This means that by 2020 losses will amount to 1.4-2.8% of the total potential 
land. Globally, GLASOD estimates that of the 8.7 bln has of the potential agricultural land, 
half is under forest, of which 18% is degraded; 3.2 bln has are under pasture, of which 21% 
is degraded; and nearly 1.5 bln has are in crops, of which 37% is degraded21.
Decreasing crop yields are often mentioned as an important indicator of land 
degradation, but there are few quantitative studies on the impact of degradation on 
production, especially in developing countries. Land degradation is being perceived as either 
a potential threat or as relatively unimportant to global food supplies. Detailed studies for 
irrigated systems in South Asia estimate that salinization and waterlogging are responsible 
for a 50% decline in yields over eight years. In China, erosion in rainfed lands lowered 
yields by 19%, and multiple cropping intensity by 11%22. Global desk studies indicate, 
however, that the impact of present and prospective losses of productivity because of land 
and water degradation are small relative to future global demands, amounting to no more 
than 10-15% of expected increases23.
These uncertainties about the extent and causes of land degradation and its impact 
on agriculture and forestry production may stem from analyses that rarely see it in terms of 
decreased production, but as environmental processes assumed to negatively affect 
productivity. There is an emerging consensus, however, that the productive capacity of lands 
is not inherent in ecosystems, but is rather a latent biophysical quality realized through 
human use24. It is then necessary to know who makes decisions about land use, and to 
understand why they make such decisions and how decisions are made25. Interpretations of 
such decisions and impacts may differ as much between different resource users as between
20 Renewal of the CGIAR: An Overview, 1995. In Background Documents on Major Issues. Ministerial-Level 
Meeting, Lucerne, 9-10 February. CGIAR
21 Scherr, S.J., B. Barbier, L.A. Jackson, and S. Yadav, 1995. Land Degradation in the Developing World: 
Implications for Food, Agriculture and Environment to the Year 2020: A Synthesis of Recommendations from an 
International Workshop. IFPRI - Discussion Paper.
22 Joshi and Jha, 1991; and Rozelle et. al., 1995; cited in (15)
23 Crosson, P., 1994. Degradation of Resources as a Threat to Sustainable Agriculture. Paper prepared for the 
First World Congress of Professionals in Agronomy, Santiago - Chile.
24 Turner II, B.L. and P.A. Benjamin, 1994. Fragile Lands: Identification and Use for Agriculture. In 
Agriculture, Environment and Health: Sustainable Development in the 21st Century. V.W. Ruttan (Ed). University 
of Minnesota Press, London.
25 Biot, Y., P.M. Blaikie, C. Jackson, and R. Palmer-Jones, 1995. Rethinking Research on Land Degradation 
in Developing Countries. World Bank Discussion Papers, No. 289.
10
those based in indigenous and scientific knowledge. There may be a need, then to conduct 
research under "living conditions", and involving a wide range of actors relevant to both 
technological and institutional innovations.
2.2. The Institutional Environment
The last years have seen the consolidation of changes leading to an increasingly 
interdependent world, as reflected in the enforcement of global trade agreements, the 
conceptual move from "food self sufficiency" to "food security", and the agreements reached 
at the Earth Summit. In the agriculture environment, however, most developing countries and 
the majority of the development assistance community do not yet fully recognize the role of 
agriculture as an engine of economic growth.
2.2.1. The Surrounding Environment
At the Uruguay Round of trade negotiations, countries undertook commitments to 
reduce domestic support to agriculture. These market changes are expected to boost the value 
of world agricultural trade significantly, offering higher prices and some increase in import 
possibilities. Low-income, food importing countries, may loose from higher food prices, but, 
if properly situated, they may gain access to developed-country markets. At the Earth 
Summit it was made plain that the world could no longer think of environment and economic 
and social development as isolated fields. It emphasized the role of agriculture and 
agricultural research in feeding the world, while avoiding further encroachment by poor 
people into marginal lands. It is worth noting that the thrust of the discussions on agricultural 
issues in Agenda 21 focuses on their environmental impact, with less emphasis given to 
poverty alleviation than in the CGIAR Agenda.
Given the lack of recognition to the role of agriculture and to structural adjustment 
policies, public investment in agriculture fell from 7.8% in the decade of the ’70s to 7.0% 
in the ’80s. This lack of interest is also present among development assistance agencies, 
which have lowered the share of agriculture from 20% in 1980 to 14% in 1990. In this 
context, support for agricultural research has also fell down in developing countries, slowing 
growth of public spending in agricultural research from 7% a year in the 60s to 2.7% in the 
past decade26. On the other side, the emergence of regional groupings of national systems 
is seen as a potentially positive development that can help in both linking complementary 
strengths among developing countries and better channelling the efforts of international 
centres27.
26 Pinstrup-Andersen, P., 1995. The Challenge for a 2020 Vision: Extent of Today’s Human Suffering and a 
View Toward 2020. Speech made at an International Conference on ”A 2020 Vision for Food, Agriculture, and the 
Environment". July 13-15, Washington, D.C.
27 TAC Secretariat, 1995. The Future Role of the CGIAR in Development of National Agricultural Research 
Systems: A Strategic Study of Institution Strengthening Research and Services. SDR/TAC:IAR/95/12.1 - Working 
Document.
11
Studies indicate that structural adjustment has had mixed effects on agricultural 
research systems28. In a few countries, adjustment policies have resulted in more 
competitive and market-responsive agricultural research. In many, however, the pace of 
change in policies has exceeded the capacity of research systems to change course and 
address new concerns, seriously affecting their capacity to generate public goods. And the 
private sector is yet in the process of getting organized, but it seems clear that it will not take 
up much of the gap in "public goods" work.
In the profit-seeking sector, most of the private activities are carried out by the seed, 
pesticide and livestock feed industries29. They conduct mainly applied research, spending 
more money on linking research and technology. Among the non-profit organizations, NGOs 
are playing an increasingly important role in developing countries, especially in activities 
related to both introducing a user perspective into downstream adaptive research and 
organizing local communities for the processing of primary products and/or the management 
of natural resources.
The weakening of national research systems and the increasing diversification of its 
members pose important managerial challenges to the envisaged catalytic role of the CGIAR 
in collaborative activities.
2.2.2. The Internal Environment
The last few years have witnessed important changes in the System, embodied in the 
process known as the "Renewal of the CGIAR". This is based on the clear commitment by 
the world community to promote sustainable agriculture for a food secure world. It 
materialized in the "Declaration and Action Program" adopted by participants in the Lucerne 
Ministerial-Level Meeting, which demonstrated a clear commitment to addressing the 
challenges of promoting a people-centred sustainable development that helps feed the hungry, 
reduces poverty, and protects the environment, in the context of a rapidly expanding global 
population that places increasing demands on the Earth’s fragile and finite natural resources. 
In this context they:
placed agriculture at the heart of the development paradigm, reaffirming the role of 
agriculture as both a catalyst and an integral component of development;
identified research as a fundamental precondition for sustainable agricultural 
development, and recognized the CGIAR as a valuable and vital contributor to 
international efforts in this field;
28 Tabor, S.R., 1995. Structural Adjustment and Agricultural Research: Summing Up. In Agricultural Research 
in an Era of Adjustment: Policies, Institutions and Progress. S.R. Tabor (Ed). World Bank - Economic Development 
Institute, Seminar Series.
29 Pray, C. and R. Echeverria, 1990. Private Sector Agricultural Research and Technology Transfer Links in 
Developing Countries. In Making the Link: Agricultural Research and Technology Transfer in Developing Countries. 
Westview Press, London.
12
enjoined the CGIAR to continue efforts to nurture a dynamic South-North 
partnership in the interests of the world poor and marginalized;
endorsed current CGIAR emphasis on natural resource management, twinned to its 
continuing goal of increased food productivity; and
encouraged the CGIAR to complete its reform and renewal process, setting down 
guidelines for action in four areas: broader partnerships, the research agenda, 
governance, and finance.
Broader Partnerships. Activities in this field aim at including more members from the South 
and incorporating the perspective of national systems into the CGIAR research agenda. 
Initially they are focusing on the development of an Action Plan (AP), guided by the regional 
representatives of NARS to the CGIAR and evolving in a participatory framework30. It 
started with a meeting in May ’95 to outline an AP, and will culminate in a detailed one 
(DAP), to be presented at ICW ’96. The AP is seen as an exercise to strengthen the global 
technology development and transfer system, and has four main goals:
ensuring a stronger reflection of the collective views of NARS in CGIAR priority 
setting;
improving modalities for developing formal partnerships between IARCs and NARS;
strengthening NARS’ collaboration and representation in the CGIAR through the 
establishment of strong regional and sub-regional fora; and
enhancing institutional capacity building at both the regional and national levels.
The AP will be facilitated by three specific sets of activities: i) the effective use of 
information technologies; ii) improving the economic analysis of investment in research; and 
iii) developing case analyses of effective institutional arrangements and participatory 
mechanisms.
Deliberations of the four regional fora are taking place, and issues will be taken 
forward by their chosen representatives to the MTM’96 in Jakarta in May, where plans for 
each of the four regions will be consolidated into the DAP. Case studies will proceed 
concurrently, and their lessons incorporated into a revised version by August ’96, to be 
reviewed at ICW’96.
To institutionalize the dialogue between the CGIAR and other partners, special 
committees have been established to seek the views of NGOs and the private sector.
The Research Agenda. As part of the Renewal process a new framework for decision-making 
was endorsed at Lucerne, that features greater openness and broader participation, and which
30 Outline Action Plan to strengthen NARS-CGIAR partnership, 1995. Prepared by Members of NARS Working 
Group, Chaired by C. Ndiritu. Washington, D.C. - Revised 14/11/95.
13
is more transparently related to the goals of its stakeholders. It is based on five undertakings, 
each relating directly to the overarching goals of the CGIAR: i) increasing productivity; 
ii) protecting the environment; iii) saving biodiversity; iv) socioeconomic, public policy, and 
public management research; and v) strengthening NARS. The agenda is then built around 
twelve sets of activities, each of which can be related to one of the five undertakings. In 
addition, there are activities related directly to Systemwide and ecoregional programmes, 
specifically focused on the objectives of each initiative.
Among the various activities which might be pursued, the CGIAR has concluded that 
its strength lies in work on important (vis-à-vis poverty and the environment) international 
(because of mandates and economies of size) public goods (non-appropriable and that involve 
non-rivalrous consumption). Among the international public goods that the System might 
consider, TAC focuses its attention on those in which the System has either a cost or an 
apparent reliability advantage.
Governance. TO FOLLOW
Finance. TO FOLLOW
2.3. Possible Items for a Discussion on CGIAR Strategies
Regarding an agenda for TAC’s discussion on future CGIAR strategies, the 
preceding sections would suggest the following tentative items, which can be grouped into 
those related to research and organizational strategies.
2.3.1. Research Strategies
Widen the Production Strategy: to increase the yield potential in new varieties, and to manage 
the resource base for a sustainable expression of such potentials.
Modernize the Integration Strategy: by combining traditional and scientific knowledge in the 
blending of new information into old practices.
Evaluate the Impact of Degradation: to dissipate uncertainties about the extent and causes of 
land degradation and its impact on agriculture and forestry production.
Expand the Productivity Horizons: to look at the efficiency of product utilization in the 
production-to-consumption continuum
2.3.2. Organizational Strategies
Integrate the Research Dimensions: bringing to bear agroecological and socioeconomic 
perspectives to a logical framework linking agricultural production to poverty alleviation and 
environmental degradation.
14
Catalyze Collaborative Modalities: to make full use of comparative advantages among 
institutional partners in the development of a common agenda.
Monitor the Transaction Costs: to ensure the effective and efficient management of public 
resources by global/regional research systems.
15
For TAC Members: What follows is a draft of the chapter dealing with 
considerations behind the character and choice of various elements of the 
priority setting process. It is written as if the answers were already 
available. In most cases, the reactions can be inferred from earlier TAC 
discussions. In some cases they cannot be and the draft can be used to 
orient discussion. In either case, nothing is set in concrete, all (including, 
of course, the draft itself) is up for discussion and subject to change. Note 
that the first paragraph of each section describes (briefly) the findings or 
conclusions or characteristics of the section.
CHAPTER THREE - ANALYTICAL FRAMEWORK
3.1. Introduction
TAC’s framework for priority setting, the principal theme of this chapter, rests 
squarely on the principal concerns of the Group. Those concerns, now well established 
through discussion at Lucerne and reinforced at ICW95, emphasize poverty alleviation and 
natural resource conservation in developing countries; both concerns are said to be people 
centred. Combined, the two promote sustainable food security.
Beyond these overarching goals, there are other considerations important to decision 
making. Given commitments to their own constituencies, members of the Group seek the 
effective and efficient pursuit of its goals. TAC has reviewed various considerations and has 
concluded that, at its level of priority setting, the most important to efficiently achieving the 
CGIAR’s objectives are the likely success of particular efforts and the availability of 
alternative sources of supply for the products of such efforts. These two considerations are 
also significant in TAC’s framework for setting priorities.
Finally, in the course of recent discussions about the research agenda the Group has 
evidenced its sense that current priorities are roughly consistent with the stated goals. Given 
that perception, TAC has employed current priorities as a point of departure in its review of 
priorities for 1998-2000.
These elements - the specific goals of the System and the considerations that emerge 
from their efficient pursuit - have been widely discussed and solidly endorsed. What follows 
here is a brief review of their connection to TAC’s priority setting. This preceded by a brief 
discussion of the framework for priority setting and changes from 1992, the process for 
shaping this round of priority setting, and the link with recommended resource allocations.
16
3.2. The Framework
TAC’s framework for priority setting features activities, sectors, commodities, and 
Systemwide work. This represents a change from the framework for the 1992 undertaking 
in that TAC has dropped the regional dimension (see below) as a separate dimension of 
decision making.
The activities considered number 30 and are made up of the activities described by 
TAC for the 1992 effort. In the course of its deliberations, TAC has noted some potential 
improvements in the activities categories; TAC expects to discuss these potential 
improvements with the Group at ICW96.
Unlike 1992 TAC considered four sectors, having added fisheries to the three sectors 
considered then. The commodity portfolio reviewed in this round of priority setting is like 
that utilized in 1992. In a sense, and as in 1992, the sector allocations are the sum of the 
judgments pertaining to the commodities. Systemwide activities played a role in the 1992 
effort and, of course, do so in this undertaking. However, TAC has modified the 
terminology and has recommended a modification in the way that Systemwide Programs, in 
particular, are to be funded. In the 1992 recommendations TAC proposed to allocate specific 
amounts to such work in addition to the amounts recommended for centre budgets, per se. 
At this time TAC proposes that Systemwide Programs be financed through centre budgets, 
in effect leaving the commitment to such enterprises in the hands of centre management and 
boards (as in the case of many of the SWPs currently underway), while Systemwide 
Initiatives - the design phase - be financed through special allocations recommended by TAC.
As for regional allocations, TAC proposes that these emerge from the primary 
considerations in priority setting rather than be imposed apriori. TAC notes that this 
approach assures a more even handed treatment of the overarching goals of the System and 
of the opportunities open through the System’s activities. Furthermore, TAC notes that the 
internal information system, especially with the addition of the standardized project reporting, 
can accurately report the System’s investment by region.
3.3. The Process
Notable about this round of priority setting is the fact that TAC will not have the 
opportunity to submit a draft version of its work to the Group as it did in 1992. As it stands, 
and given the commitments made by the System as a part of the renewal process, TAC has 
been compelled to telescope notably its deliberations and interactions with the Group and 
with others so as to present to the Group its first round of recommendations at MTM96. 
Given the results of MTM96 deliberations, TAC will immediately set out to work with the 
centres on the MTPs for 1998-2000 so that these can be reviewed by the Group at MTM97.
Looking to the past, TAC notes that the 1992 effort, aimed at MTPs for 1994-1998, 
was itself reviewed by specialists both at a workshop organized by the TAC Secretariat in 
April 1994, and at a meeting of the American Association of Agricultural Economists. The 
period from 1992 to 1994 saw the continuation of two standing committees within TAC
17
whose purpose was to monitor the degree to which priorities identified by TAC were being 
integrated by the centres and supported by the Group. During the period TAC met with 
others as a member of the Joint CDC/CBC/TAC Committee on Priorities and Strategies. 
The pace of these meetings increased from 1995 onwards and 1996 opened with a specially 
convened meeting occupying 1.5 days. Finally TAC has been engaged since mid 1994 in 
meetings with representatives of NARS aimed at encouraging greater interaction with TAC 
on priority setting, among other themes.
TAC’s work on the 1998-2000 priorities moved forward at TAC 66 in March of 
1995 with emphasis on the major dimensions of priority setting. This was followed by 
TAC67, July 1995, which included specialists in several areas identified as especially 
important by TAC 66. At ICW TAC invited the Group to comment on its interpretations of 
the Group’s important underlying value judgments. Elements of that discussion are evident 
in many of the points that follow here. At TAC 68, December 1995, TAC integrated the 
findings of ICW95 and laid out the elements of a critical path towards MTM96.
Comparison of the 1992 mechanisms for priority setting with those employed in the 
1996 effort show similarities and differences. Similar is the emphasis on poverty, the mix 
of quantitative and qualitative judgments, and a point of departure based on value of 
production. Differences are evident in the approach to quantitative work - this round features 
but one modifier while the last included several, the inclusion of probabilities of success and 
alternative sources of supply as more explicit considerations in this round, and the explicit 
use in this round of a base year (1997) as a point of departure.
Quite obviously the ultimate purpose of priority setting is that of allocating 
resources. TAC’s sense of the first, conditioned by the considerations that make up this 
chapter, is that the System requires a sense of relative changes in priorities in each of the 
four dimensions of its analysis. Each time that TAC perceives a need for a modification in 
priority it will signal the considerations that occasioned the change. In this way, the Group 
will have not only the outcome of rebalancing but the logic or reasons that rebalancing was 
deemed to be warranted. Quite clearly the set of recommendations must be consistent from 
one dimension to the next, That achieved TAC will frame an analogous set of
recommendations about resource allocations.
Before moving on, TAC notes its problem with vocabulary. In much of what 
follows the term agriculture is used to mean crops, fisheries,forests, and livestock. This is 
not a happy term for this broad conjunction of activities because, for many, agriculture refers 
to crops, for some to crops and livestock, but to few as a name for the four sectors in which 
the CGIAR works. A second possibility was to write out the four each time and a third was 
to frame an acronym, CFFL. TAC decided to stay with agriculture and asked the reader to 
appreciate always the expanded meaning of the term in this report.
18
3.4. Agricultural Research, Productivity, and Income Growth
How do the CGIAR’s activities in agricultural research and related themes pertain 
to productivity increases and to income growth? Especially in the past two years, TAC has 
written much about these themes. TAC has concluded that, especially for poorer countries 
where much of the labour force is in agriculture and much of the average budget is spent on 
food, productivity increases in agriculture offer the best opportunity to stimulate economic 
growth, and improved technologies resting on research are a major source of increased 
productivity in agriculture.
First, TAC notes that, for most developing countries, especially the poorest, a large 
portion of the population works in agriculture and a high proportion of the average family 
budget is spent on foodstuffs. It seems intuitively obvious, then, that increased productivity 
in agriculture is a necessary condition to achieving income growth in poorer economies. As 
income levels increase the role of agriculture decreases (see Table 1), offering less scope 
for agriculturally driven growth.
Table 1: Agriculture’s Share in National Income, Employment, and Export Earnings
by Country Income Groups, 1989
Average Per Capita Agriculture’s Share % in 
Country Group* Income 
GNP Employment ExportUS$ Earnings
Low Income
Less than US$ 250 per capita 190 49.3 76.5 65.0
US$ 251-500 per capita 380 33.1 68.2 44.3
Middle Income
US$ 501-1,000 per capita 800 22.8 52.7 40.5
US$ 1,001-2,400 per capita 1,580 12.8 29.8 39.8
Upper-Middle Income
US$ 2,401-6,000 per capita 3,530 8.7 21.8 16.5
High Income
US$ 6,001-16,000 per capita 11,520 4.6 8.8 18.9
Greater than US$ 16,001 20,460 2.8 4.4 10.8
* Country income groupings are taken from the 1991 WDR but with subdivisions within groups that give roughly equal numbers of 
countries in each.
Source: World Development Report 1991, except agricultural employment data which 
were obtained from the 1991 FAO Production Yearbook-, as quoted in "Agricultural Sector 
Review," World Bank’s Agriculture and Natural Resources Department, July 1993, p. 3.
19
Analysis supports the intuition of the previous paragraph. IFPRI studies show that 
an increase of US$ 100 in agricultural income gives rise to more than US$ 50 of added 
income in the rest of the economy. Moreover, current IFPRI and other work shows that, 
especially for poorer countries, the multiplier effects of agriculture on the remainder of the 
economy are greater than those for investment in any other sector. More emphatic is the 
statement in the Agricultural Sector Review, July 1993, ANRD, World Bank ,to the effect 
that “broad-based agricultural growth, involving small and medium-size farms and driven 
by productivity-enhancing technological change, offers the only way to create productive 
employment and alleviate poverty on the scale required.” (See Box)
Historically, productivity increases in agriculture come from improved technologies, 
from education for producers, from better infrastructure, from improved institutions, and 
from more effective policy. Improved technologies have provided the most noteworthy and 
the most reliable sources of increased productivity in agriculture (see the World Bank 
observation in the previous paragraph); improved policies have also proven fruitful. Both 
are the products of research, the raison d’etre of the CGIAR.
How does increased productivity lead to greater incomes? The increased returns 
to those who hold the factors engaged in agriculture stimulate increased spending on
consumption and on investments. This increased spending, in turn, augments the incomes 
of others, who, in turn, increase their own spending. The result is rising rounds of spending 
across the economy and the effects described by IFPRI. This scenario reflects agriculture’s 
role as an engine of growth.
Beyond that, the lower food prices accompanying increased output serve as a 
lubricant to growth in the remainder of the economy, usually through their effects on wages 
and, via wages, on savings and further investment. That investment, in turn, gives rise to 
an ever larger capital stock, again with further implications for growth, that ultimately 
draws workers into non-agricultural pursuits through higher wages. Many specialists in 
economic growth have dealt with this phenomenon. As well, the lower food prices are a 
gain to rural consumers, especially poor rural consumers.
Clearly some of the consequences for the poor are direct and more immediate while 
others, especially those associated with non-agricultural employment, are more indirect and 
longer term in nature. (See also Box)
TAC notes that these considerations throw light on the poverty/food insecurity 
nexus. Food security rests on the availability of food and access to it. Globally, there is 
no question about the availability of food. However, as argued so cogently in (NA-2010), 
this is not the critical question. For even with food available, there are millions across the 
developing world who, because they have inadequate incomes, do not have access to 
adequate food. (“It is now widely recognized that failure to alleviate poverty is the main 
reason why under-nutrition persists.” p.2.) This leads (NA-2010) to emphasize that the 
production of food serves two important functions which contribute to food security, one 
is to assure availability while the other, and for the developing countries the more 
important, is that the production of food is a major source of income for the poor. In any
20
long run, food security for the family rests on higher incomes or, said another way, on the 
alleviation of poverty.
BOX: Linking Agricultural Research to Growth (Based on materials from 
Peter Hazell, IFPRI)
This theme warrants treatment at three levels: that of the farm, of the region, and 
of the nation. At the farm level, growth permits an increase of spending in several 
dimensions. Clearly it can lead to higher levels of consumption. More important for the 
long term, it can lead to on-farm investments in future productivity growth including those 
in NRM, to investments into non-farm income diversification (including small business 
activities), and to investments in education. . Beyond that, through increased savings, farm 
level growth augments the capacity and willingness to accept risk.
At the regional level, agricultural growth leads to powerful demand linkages to the 
non-farm economy . These include consumption demands, demands for farm inputs, 
demands for marketing and agricultural processing services, and demands for investment 
goods. Empirical studies show that the first three lead to income multipliers of between 
US$ 0.5 and US$ 0.8 per dollar of additional value generated in agriculture with most of 
that coming in rural towns. Of the three, consumption demands play the largest role. More 
recent work suggests that demand for investment goods are more powerful even than 
consumption spending when their impact on future productivity and income generation is 
considered, i.e., there are large dynamic multipliers.
Agricultural growth in high potential areas often generates additional labour 
demands and higher wages that attract seasonal migrant farmer workers from poorer 
regions, leading to important income multiplier elsewhere. Examples include the migrations 
from eastern India to Punjab and Haryana and the Burkina Faso labour migrations to the 
cocoa farms in Cote d’Ivoire. While these payments tend to reduce the host region’s 
multipliers, experience shows that they increase multipliers in the home region, as income 
from non-local sources can be an important catalyst for investment and subsequent growth 
in output in the home region.
Regional multipliers can vary with, e.g., farm size. Medium-sized family farms 
generate the largest regional multipliers. Growth via smaller farms usually has larger 
regional multipliers than for large farmers because the income generated per hectare is 
larger, more labour intensive practices are used, and small farms use a larger share of 
locally produced goods in production, consumption, and investment. Very small farms, 
however, generate small multipliers because they have few cash transactions. Regional 
multipliers also tend to reflect the state of infrastructure. With poorer roads, for example, 
much of an income increase is spent on diversifying the diet, a result starkly shown when 
comparing villages with good versus bad road connections. As well, the level of income 
tends to influence the multipliers. They tend to become larger with regional economic 
growth, at least through middle income status. Favoured environments well served by 
infrastructure, then, would be expected to have larger regional multipliers than would 
marginal environments with less infrastructure.
21
For national multipliers and with the neo-classical conditions satisfied, the 
differences in demand linkages for the reasons described above would not operate. All 
investments would generate the same long-run income multipliers and these would be small 
or even zero because all resources remain fully employed. Markets and conditions are
rarely so efficient, and labour and capital in particular are not notably mobile across regions 
and sectors. The targeting of agricultural growth by region and type of farm does matter. 
In the more real world, the regional multipliers are probably good guides to the relative 
differences in multipliers.
It should be noted that where labour is immobile, investments in poor regions may 
generate smaller regional and national income multipliers than investments in high potential 
areas, but may be necessary to reach the very poorest of the nation. In countries with large 
poor urban populations, however, the larger income multipliers from investments in high 
potential areas may have a bigger impact on poverty. These issues need further research.
A number of macro or national level benefits accompany agricultural growth. 
Among these are: capital flows out of agriculture through the banking system add to 
investment potential elsewhere in the economy; increased entrepreneurial activity in rural 
areas, whether on the farm or in non-farm pursuits, augments managerial skills which can 
be applied in other parts of the economy; lower food prices have implications for wage 
rates that make the economy more competitive in international markets; balance of 
payments are strengthened through increased exports or through reduced imports; and the 
higher incomes in agriculture provide a broader tax base for the public development of 
infrastructure and education.
As for comparisons with growth linkages through other strategies, recent IFPRI 
work shows that development strategies based on agriculture are superior to industrial-led 
strategies for the poorer developing countries and, as in the earlier discussion, that targeting 
the small farm sector is better than targeting the large farm sector. These studies measure 
success in terms of growth rates but also show that the income distribution shifts towards 
the poor. As expected, these results are sensitive to assumptions about the mobility of 
capital and labour.
Importantly for the CGIAR, the superiority of agricultural-led growth diminishes 
with economic development. These results suggest that efforts aimed at poverty alleviation 
should be via productivity growth in agriculture, focus on poorer countries, and emphasize 
rural-labour intensive activities.
(End Box)
3.5. Economic Growth and Poverty Alleviation
What about the evidence that economic growth leads to poverty alleviation? TAC 
has concluded that there is strong empirical evidence that economic growth in poorer 
countries leads to poverty alleviation, even in cases where growth leads to less equal 
distributions of income. Even so, TAC concludes that growth alone is not sufficient to
22
achieve the Group’s pro-poor goals, implying a further emphasis on poverty in TAC’s 
priority setting.
For a brief, lucid, and penetrating treatment of over 200 years of research on 
poverty and economic growth one can do not better than ML/MR (with its 580 citations). 
A paper by Ravallion, “Growth and Poverty: Evidence from the Developing Countries in 
1980s”, cited in Lipton and Ravallion, reports on sixteen studies chosen because their data 
made possible a comparison of economic growth with the reduction in the number of 
people below a commonly defined poverty line (more on poverty lines later). The studies 
showed that the number of such people decreased with economic growth. Moreover, 
higher-order measures of poverty also showed declines in poverty, implying that “benefits 
from growth are typically felt well below the poverty line”. (ML/MR, p. 2607.) ML/MR 
go on to say that “...even when growth has been associated with rising inequality, it 
appears that poverty has typically fallen “ All of this is not to suggest that government 
should concentrate its attentions solely on promoting growth. A number of factors 
influence the relationship between growth and the reduction of poverty, and government, 
through targeting and policy, may have a role in promoting patterns of growth more 
congruent with poverty alleviation (Ibid., p.2607). Among such strategies is that of 
“shifting investments towards rural, labour-intensive...activities”. (Ibid., p.2608.)
It should be noted that some argue that strategies that slow the rate of growth will, 
sooner or later, impede the reduction of poverty, that growth itself is the best pro-poor 
strategy. (See Pardey, et. al.) This point of view is, to an extent, consistent with an 
observation by ML/MR that a key question in assessing the effects on poverty of macro- 
economic policy is: “did adjustment raise or lower the rate of growth?” (ML/MR, p. 
2613.) However, many opine that barriers to the effective functioning of markets—whether 
because of policy, shortage of information, or inertia in labour markets—argue for the 
implementation of pro-poor strategies focused on the contemporary poor, These
considerations make up one strand of the rationale underlying the emphasis on labour- 
intensive activities in rural areas, and support TAC’s decision, discussed later, to include 
a modifier for poverty in its analysis of priorities.
To here, and said briefly, the logic emphasizes poverty alleviation via income 
growth resting on increases in productivity in agriculture, whose effects are felt directly 
(mostly near term) and indirectly (mostly longer term).
3.6. Agricultural Research and the Conservation of Natural Resources
Turning now to natural resource conservation, TAC agrees that poverty underlies 
much of the threat to natural resources in developing countries. TAC has concluded that 
emphasis be given first to conservation’s implications for present and future productivity, 
broadly defined to include on-farm and off-farm benefits and costs, and then to broader, 
global, environmental effects. TAC notes the importance of the positive linkages between 
productivity and conservation as well as potentially negative linkages, where opportunities 
to reduce trade-offs between development and the environment must be pursued. TAC 
notes, too, the now broader view of conservation guiding the CGIAR System.
23
The connection between the CGIAR’s activities and the health of the environment 
is more self-evident than in the case of poverty alleviation. Conservation will require 
technologies that directly serve the various purposes of producers, including those 
pertaining to natural resources, or which can be made to do so through the implementation 
of policy (but see below). Such technologies - productivity-increasing, resource-conserving 
technologies - and policies will rest on the findings of research. The CGIAR is much 
involved in research relevant to conservation. Indeed much of what is done by the CGIAR 
contributes, directly or indirectly, to that aim.
Over the years TAC has examined the conservation theme on several occasions. 
N kinds of questions have motivated the Committee’s concerns. One had to do with the 
resources to be conserved and the question, roughly, was the extent to which the System 
thinks beyond the gains that conservation brings to producers to the off-farm effects of 
practices on , e.g., soil erosion or global green house gases. Another had to do with the 
nature of the System’s role in resource conservation—developing more suitable technologies, 
developing effective research paradigms through the experience of developing more suitable 
technologies, leading the way on the science of resource conservation, or advocacy for 
conservation itself.
Sustainability was a major organizing theme for TAC and has become one for the 
many now involved with natural resource issues. TAC’s approach to the theme has varied 
over time, from in its initial treatment, through “...” in 199X, to the stance reflected
in the current paper on soil and soil water management. By now, of course, the meanings 
ascribed by others to sustainability vary so much, e.g., from maintaining nature’s 
endowment intact to ensuring “resiliency”, that the term has lost much of its force, and 
longer expressions are required to convey meaning. TAC’s 1996 paper has those longer 
expressions. Note, too, that the 1996 paper, consistent with the System’s concern for the 
poor and poverty alleviation, focuses on the conservation of natural resources of concern 
to present and future generations of poor producers and consumers and to their health, and, 
to borrow a term from FAO (NA-2010), in the context of returns and “... timescales which 
meet their differing circumstances or risk perceptions”, p. 395. With these criteria
satisfied, TAC introduces the criteria for involvement in such global environmental 
concerns as green house gases, lifesystems, ethical, aesthetic, and amenity considerations. 
TAC also notes that much CGIAR work on natural resources, indeed much of the System’s 
work, has simultaneous implications for both the local and the global environment.
Earlier reference was made to policy as an agent in promoting the use of 
environmentally friendly technologies. One example of this strategy is in GEF’s policy of 
offsetting the added costs of carbon efficient power plants so as to limit green house gases 
without raising the cost of power to users. This is an industrial example of what (NA- 
2010) calls “reducing the trade-offs between development and the environment.” (See, 
again, NA-2010, Ch 13). Certainly a major challenge for policy in agriculture is its need 
to influence the behaviour of many thousands of largely independent agents. Doing so in 
the name of reducing trade-offs between development and the environment but with limited 
institutional capital and while restricting rent seeking behaviour is one of policy’s daunting 
tasks. TAC notes its awareness of this challenge and its continuing concern for the 
orientation of CGIAR research on policy issues.
24
TAC also notes that increases in productivity themselves have implications for the 
environment. Some of those have been negative in the further past. (See WFS/T6 and 
FAO 2010 for reasoned discussions of the issues). Much, however, has had a positive, 
albeit indirect, affect on the environment. In particular, it is frequently observed that the 
lower prices accompanying increases in productivity in favoured areas have the affect of 
reducing movement onto fragile lands and forest margins.
TAC will have more to say on this theme in Chapter 5, where priorities for 
activities are discussed. Before moving on, however, TAC notes that the two primary 
concerns of the System are imbedded in a three part nexus of problems, That nexus
features poverty, environmental degradation, and population growth. Poverty, the pivotal 
element, has immediate consequences for the environment as the poor are compelled to 
focus on the near term with less attention to the long run when the fruits of conservation 
come into play. Poverty is also strongly associated with population growth, either because 
of the direct benefits from large families as seen by the individual or because poverty limits 
what society can do to reduce growth rates through investment in education (especially for 
women) and better health. In turn, rapid population growth combined with poverty 
promises more damage to natural resources in the future. And, through time, degradation 
of natural resources ultimately threatens productivity, further reinforcing poverty. The 
pivotal element in this nexus is poverty and the way forward is through its alleviation. In 
this context, FAO-2010 (p. 400) opines that “...much of the mismanagement of natural 
resources in developing countries relates to poverty and to the lack of economic growth to 
provide better and sustainable livelihoods outside of subsistence agriculture.”
TAC further notes today’s broader view of the conservation challenge, a challenge 
that not only pertains to land and water, but to such far reaching concerns as global 
warming. In another dimension, this broader view has included the recognition that the 
environment itself is one of four, interacting classes of capital on which humankind relies. 
(Reference one of the Serageldin-Bank publications.) This expanded view of capital 
emphasizes that one generation’s obligations to future generations includes maintaining or 
augmenting the total capital stock, while ensuring that no one of the four classes falls below 
minimum critical levels. It notes that institutional capital may be critical in implementing 
strategies aimed at conserving natural capital, that human capital has been substituted 
effectively for other forms of capital, and is sensitive to the possibility that natural capital, 
declining in physical terms over time while the other classes of capital are increasing, is 
relatively undervalued.
Unlike its stand with respect to poverty, TAC has not included in its priority setting 
a special modifier for degradation of soil and water. As Chapter 5 will evidence, the 
priority accorded to research in this area rests first on its implications for present and future 
productivity and second on its implications for the global environment. To go beyond that 
would require, among other things, data and measures which are not yet available. 
(Citation) TAC is working with those who are engaged in improving such measures.
25
3.7. Methods for Setting Priorities for Agricultural Research
In setting priorities, TAC has relied most on structured discussion supported by 
quantitative and qualitative information. TAC has used its earlier recommendations and 
current budget allocations as a point of departure, altering those given compelling 
information - especially about science, alternative sources of supply, recent experience, 
prospects beyond research - or analysis. Congruence analysis modified by the locus and 
level of poverty has been applied to commodities and sectors as an input to structured 
discussion on priorities in those dimensions.
To set priorities is to order activities in ways that are most consistent with the 
objectives that motivate the effort. Priority setting is always difficult because of the 
unknowns or uncertainties that must be taken into account. Agricultural research is an 
especially difficult arena for convincing priority setting. Beyond uncertainties other issues 
also intervene. (SUS) reviews the various elements that challenge the setting of priorities 
for agricultural research. The book is the latest, and arguably the most complete, 
presentation of the issues influencing priority setting in research and includes a presentation 
of the advantages and disadvantages of various approaches. A prime point is the problem 
of dealing with multiple objectives. The authors recommend that a single objective, 
national income growth, orient priority setting for agricultural research and go on to 
describe ways to set priorities given that goal (but see below for TAC’s view of the 
CGIAR’s objectives).
There are several approaches for those who would set public research priorities and 
allocate public resources so as to promote economic efficiency. (SUS) advocates a model 
based on economic surplus and lays out the advantages, disadvantages, and challenges of 
its utilization. As an alternative, (SUS) includes congruence allowance, which posits that 
research be “funded in proportion to the value of production”. Calling it a “crude 
procedure”, but one requiring only “minimal information”, (SUS) points out the conditions 
for this approach to yield maximum economic surplus and notes that these are unlikely to 
be realized. (SUS) notes, however, that “there might be insufficient information available 
to justify a more complete analysis”, going to say that in such cases congruence analysis 
“might be consistent with an ultra-simplified economic surplus approach in that at least 
some account is taken of the scale of the industry.” TAC believes that the ‘insufficient 
information case’ holds in setting priorities for international public goods research on a 
global basis and so has used congruence analysis as the point of departure for one element 
in its work on priorities.
(SUS) also notes that there are many situations in which insufficient information 
limits the use of quantitative methods, that quantitative estimates of criteria be obtained 
where available and that it should be combined with structured discussion where 
quantitative approaches are impossible. (SUS, p479) TAC has followed this approach.
In setting priorities TAC has adopted the position that there should be no change 
in the priorities now in play unless there is compelling reason to rebalance. For the most 
part these reasons will emerge from new information about science, about progress in 
current work, about sources of supply, or about expectations.
26
Finally, and returning briefly to the admonitions in (SUS), TAC notes that the 
CGIAR objectives do not feature income growth over time but rather poverty alleviation. 
Two views on how effectively the one leads to the other were reflected earlier, but a third 
view requires consideration here. The Group gives stronger weight to the current poor than 
to the future poor. This preference might rest on the judgment that catering to the current 
poor will more quickly affect population growth rates and more quickly limit the effects of 
poverty on the environment. The second simply delays the day of reckoning but the first 
promises smaller ultimate populations and less ultimate pressure on the environment. TAC 
agrees with this judgment. The locus and level of poverty must come into the analysis to 
modify strict congruence analysis.
TAC further notes that in the previous round of priority setting a multi-dimensional 
model brought several modifiers into the analysis. While this approach is not without its 
shortcomings, it was felt that, on balance, it was the preferred approach, given the System’s 
desire for a more quantitative orientation to priority setting than in the past. Whatever its 
qualities, the effects of that effort shaped the priorities which guided CGIAR resource 
allocations from 1994 to 1997. With 1997 as the base year for the new round of priority 
setting and to the extent that the 1994 TAC recommendations were adhered to in the 1997 
allocations, that previous round of priority setting will influence the current effort because 
of its influence on the base. And, beyond the effects of the modifiers, there will still be 
visible the residual affects of the 1992 apriori judgments about regional priorities.
3.8. Poverty as an Element in Priority Setting
With the decision to feature poverty in priority setting, TAC sought a measure that 
would permit comparison over space and time, that would be sensitive to differences in the 
inequality of income distributions among countries, that would give increasing weight to 
the poorest within a country, that would facilitate comparison of different ethical views (or 
value judgments) about the importance of the degree of poverty, that assured consistency 
with good practice, and that accommodated the limitations imposed by the data. TAC 
believes that the measure described below has those qualities. (See Appendix 1 for the 
formulation.)
Having related poverty alleviation to the CGIAR’s objectives and to its activities, 
how has TAC taken poverty into account in dealing with priorities. TAC notes that the 
final goods that make up the CGIAR portfolio were selected because they are important to 
poor producers and consumers. Straightway, then, the CGIAR portfolio itself reflects a 
concern for poverty. What follows here deals with the further integration of poverty into 
TAC’s analysis. TAC starts with a notably brief review of measurement themes and 
concludes with the measure it applied in modifying congruence analysis and in guiding its 
structured discussions.
Poverty is a state that most recognize and feel quite competent to label. Even so, 
its definition, effect on behaviour, amelioration, and measurement, among other things, have 
motivated discussion and study for over 200 years. Again, for a brief, lucid, and 
penetrating treatment of the evolution of ideas on the theme see ML/MR. Many of the
27
elements that follow here were drawn from their piece. Those who want the argument 
whole can do no better than ML/MR.
Quite evidently, poverty imposes severe limits on behaviour in various dimensions 
as it limits economic welfare. Efforts to describe poverty and to measure it are made more 
difficult by the fact that the perception of poverty and of welfare varies from place to place. 
“However, it is not controversial that inadequate command over commodities is the most 
important dimension of poverty, and a key determinant of other aspects of welfare, such as 
health, longevity, and self-esteem.” (p2553)
Spurred on by the growing emphasis on more efficiency in reaching the poor, 
development assistance agencies sought a pragmatic measure of poverty, one clearly related 
to well-being and with tolerable costs of implementation. The better known attempts 
include assessing “basic needs” (health, food, education, water, shelter, transport), 
“capabilities” (which sees consumption as permitting desired activities), and human 
development. These, and others, fail the test of efficacy in implementation. What has 
emerged are measures based on control over commodities, resting on consumption itself or 
on income, themselves with limitations but nonetheless evidently correlated with the main 
indicators of such basic needs as health, longevity, and self-esteem and comparable over 
time and space.
Initially, the featured measure of poverty was the number of people or the 
proportion of people below a poverty line, defined in terms of income adjusted for 
purchasing power parity. This was followed by a measure that integrates the gap between 
the poverty line and those below it, a measure that recognizes the depth of poverty. That, 
in turn, was augmented by a measure that reflects the severity of poverty, the squared 
poverty gap measure. Each of these measures is “exclusive”, in the sense that all weight is 
given to those below a poverty line.
Some would argue that all members of the society, the poorest as well as the less 
poor, should be included in a measure of the degree of poverty within a country. (The 
country’s mean income is a simple measure that does that.) Most would agree that an 
“inclusive” measure should, somehow, reflect the severity of poverty.. (The country mean 
does not do that.)
Ravallion has demonstrated that the theoretical differences between inclusive and 
exclusive measures are blurred by the accommodation of measurement errors and 
uncertainties about what should be brought into a measure of well-being. Then, with data 
from 40 developing countries, he goes on to show that there is little empirical difference 
between exclusive poverty measures like the three referred to above and an inclusive 
measure which incorporates aversion to income inequality by giving increasing weight, even 
moderately increasing weight, to the poorest. Given the caveat, Ravallion concludes that 
the choice among such measures will make little difference to the perception of relative 
poverty. This conclusion facilitates the work of TAC. (See Box 2.)
28
Box 2: A Measure of Poverty
In reviewing and rebalancing CGIAR priorities, TAC must ensure that poverty 
plays a central role. The instrument for incorporating poverty must be consistent with the 
concerns and pronouncements of the Group, with the level at which TAC sets priorities, 
with priority setting at the Centre and program level, and with good practice. The key 
principles are that:
1. the measure for any country should reflect both the average income in the country 
(negatively) and the degree of inequality (positively);
2. the weight attached to the gains in any one country is higher the lower the value 
of the measure for that country;
3. beyond some point the weight falls to zero;
4. the rate at which the weight increases as the measure of poverty declines and the 
value of the point beyond which the weight is zero are both matters of judgment, 
for which it should be easy to assess the extent to which alternative judgments 
influence the results.
To satisfy these principles, the following elements were incorporated:
1. The poverty indicator rests on the distribution corrected mean income at purchasing 
power parity. The poverty indicator is then given by p=(l-G)y, where G is the Gini index 
of income or expenditure inequality and y is mean income at purchasing power parity.
2. The indicator has value zero whenever the value of p exceeds some critical value 
z, for which alternative values are assessed.
3. Whenever p is at or below z, the weight is given by (1-p/z) raised to the power 
alpha, where alpha is set at a value of one or higher, and for which alternative values are 
chosen.
In its assessments TAC has set z, measured in estimated 2003 income at purchasing 
power parity and adjusted for the degree of income inequality, at $6001 and at $9000, or 
50% above. (The World Bank describes incomes greater than $6000 as high incomes. 
Incomes above $6000 after adjustment for distribution, as here, would all fall in that 
category.) TAC has set alpha at 2, a commonly applied value, and at 3, which gives even 
greater weight to the severity of poverty.
(See Appendix 2 for implications.)
(End Box 2)
29
3.9. Other Elements Considered in Priority Setting
TAC concluded that incomes, but not prices because of data limitations, should be 
set in 2003; that given the level of generality at which TAC shapes priorities added weight 
to female farmers and to rural populations makes little quantitative difference but that, 
through the internal review process, centre priority setting must be carefully monitored to 
account for these considerations; that the balance between favoured and marginal 
environments is best achieved through the major factor shaping priority setting rather than 
through a priori judgments; and that neither differences in labour intensity nor in research 
production functions should shape TAC’spriority setting - but perhaps centres’ priorities for 
projects - at this time.
TAC considered several further adjustments to congruence analysis based on 
current data. The consequences of the various adjustments are discussed in Chapter 5. 
Recognizing that the products of research initiated today will have their impact many years 
in the future, TAC favoured future income and future prices in the analysis. Reasonable 
measures of income growth provide the vehicle for projecting income towards the future. 
These projections, in turn, promise strong influence on the relative incomes across 
countries, for the poverty modifier and, finally, for the weights attached to various 
commodities and actives of the System. Given the reliability of the data and the potential 
importance of the results, TAC’s analysis includes estimated incomes to 2003, roughly mid­
way through the long term planning horizon.
As for prices, adequate data and analysis are simply not available. For its 1992 
effort, TAC used a price series based on 1987-89 and prepared by the Secretariat in close 
collaboration with the centres. Shortly after the conclusion of that work a new set of 
international prices was provided by Pradado Rao; this series purports to put all prices on 
the same footing and appears to represent an improvement on the earlier set used by TAC. 
No one, however, has built a similar data set for future prices. Those who make predictions 
do so for only a limited number of the commodities with which the CGIAR deals (6 to 8 
of 25). TAC recognizes that ,e.g., urbanization will lead to higher relative prices for wheat 
(because of its convenience), that income growth will boost relative prices of livestock and 
feedstocks, especially maize, as well as reduce the relative price of rice as, especially, 
Asians seek more varied diets. Even so, there is little analysis to give precision to these 
probable occurrences. . Fortunately, while real prices must continue to decline if the
CGIAR is to achieve its promise, TAC requires only relative prices, seemingly more stable, 
even in the face of the forces mentioned above, than are absolute prices. Given these 
considerations, TAC has used the R.Rao construction to estimate relative prices for CGIAR 
commodities.
Given the Group’s manifest concern for rural women, TAC made further 
adjustments in the value of production by giving added weight to females engaged in 
agricultural production. The argument made at ICW95 was that income gains to women 
would mean more for the well being of women and children than were the same income 
gains to fall to men. An added consideration was that, on the average, the women of 
developing countries are poorer than the men. This last point rests in some measure on the 
widely held view that some 60% of those below the poverty line are females. In a personal
30
communication, Hillary Feldstein reports that further study suggests that the number is 54%, 
rather than 60%. Some members of the Group opined that women had been given
relatively less attention in the past and heavier weights now were required to redress the 
balance. Beyond that, some argue for giving extra weight to women in the expectation that 
this will hasten the reduction in population growth rates - with its further implications for 
conservation of natural resources in the future. Offsetting these considerations are the 
results of studies that suggest that poor female farmers behave more like poor male farmers 
in their production activities than like better off female farmers.
Clearly it was not the intent of the Group that TAC set priorities exclusively in 
terms of women, simply that poor females or, more generally, females, have a higher 
weight than males. For its purposes, TAC framed one set of priorities in which females and 
males, whatever their location and whatever their degree of poverty, were weighted equally, 
one set in which females had 25% more weight than males. The implications for modified 
values of production are presented in Appendix 2, Tables 3a and 3b compared with Tables 
5a and 5b. The implications for priorities are discussed in Chapter V. TAC also notes that 
the Centres, setting priorities in terms of projects, will be better able to tailor activities to 
the needs of women than can TAC, setting priorities at the global level. And beyond that, 
it seems likely that national programs and NGOs, working as they do on specific, local 
problems will have even more to do with gender based concerns. TAC will, through its 
review processes, weigh the extent to which centres have taken the Group’s admonition into 
account in their priorities and will monitor the extent to which NARS are incorporating 
such considerations into their own work.
Again as a result of discussion at ICW95, TAC set out two scenarios based on 
differing weights for rural and urban dwellers (See Appendix 2 and compare Tables 3a and 
b with Tables 6a and b). Group members developed several arguments for favouring rural 
dwellers. One was the believe that, on the average, rural people are poorer than urban 
dwellers. A second was that urban dwellers have more access to socially provided services 
than do rural dwellers, hence that the farmers’ real incomes are understated vis-à-vis the 
latter. A third was that rural people are more numerous than urban people. A fourth was 
that the CGIAR with its emphasis on agriculture can do more for people involved with 
agriculture, hence rural people, than for urban people. Consider these impressions.
The first proposition is consistent with FAO data (see.... ). Given that TAC is
using national averages as its point of departure in framing the poverty modifier and that 
this implicitly assumes that incomes are the same in rural and urban areas, evidence that
those incomes differ suggests that an adjustment for rural and urban income levels will 
better reflect where poverty is found. The second is probably true to some unknown 
degree. Both considerations, however, are offset to the extent that the purchasing power 
parity of the average current income in urban areas is less than that in rural areas. On 
balance, the three elements taken together might suggest a modest adjustment, based on 
the proportion of rural and urban populations, to better reflect the inter-country comparison 
of poverty. The fourth theme has merit but perhaps less than meets the eye.
TAC has observed that increased productivity in agriculture affects poverty 
alleviation through two avenues: first, through the immediate, direct returns to those 
holding the resources (especially labour) favoured by whatever induced the productivity
31
change and later through induced rounds of spending adding further to the demand for 
factors (again especially labour) and, second, through the positive consequences for real 
income of declines in the price of foodstuffs. Both rural and urban populations benefit 
from these effects. Furthermore, and quite obviously, for most countries the full realization 
of poverty alleviation will require a shift of labour to non-agriculture pursuits. This 
suggests that those concerned with agricultural poverty must see developments outside of 
agriculture, e.g., in urban areas, as critically important to those concerns in the long run. 
What does differentiate the two populations and gives some credence to more emphasis on 
rural/agricultural than on urban poor is that the effects of productivity change are more 
quickly felt in rural than in urban areas. Time preference, then, would itself suggest some 
added weight to rural populations.
Appendix 2 has two versions of modified values of production - one with equal 
weights and the second with what would seem to be the outer limit for added weight to 
rural populations - which reflect the implications of this judgment for modified congruence 
allowance. It is noteworthy that there is little difference between the two. TAC notes, 
however, that centres, framing their priorities around projects, might well have more options 
for favouring rural over urban populations and TAC will take note of centre reactions as 
a part of the review process.
A portion of the Group expressed its concern for marginal environments and 
advocated that TAC give such environments added priority. Several considerations seemed 
to motivate this concern. One was that such areas contain either most of the poor or the 
poorest of the poor. The former seems unlikely and the latter calls for further research. 
A second was that such areas suffer most from degradation and that a further concentration 
there would be congruent with the System’s interest in conserving natural resources. 
Possibly a third was the perception that, as relatively more research has been done on 
favoured environments and relatively little on marginal environments, the effects of 
diminishing returns in the former would suggest moving more resources to the latter. 
TAC’s notes that some XX percent of the System’s resources are currently destined to 
marginal environments (see Table) and that any change in that level should emerge from 
the general considerations that are shaping priority setting, e.g., value of production, 
probabilities of success, and alternative sources of supply along with perceptions about the 
locus, level, and consequences for poverty. Via the newly emerging project framework, 
TAC will follow centre priorities on such work.
Attention was given to the possibility of incorporating labour intensity in priority 
setting. Some commodities, some sectors, utilize more labour than others per hectare or per 
unit value of production. TAC reviewed the available data and found misgivings about its 
accuracy and current relevance among its suppliers. Moreover, TAC noted that the 
measures required to ensure that proper choices were made, measures that involve the basic 
elements shaping the demand for labour, are not available. TAC concluded that, at its level 
of priority setting, there is little opportunity at this time to bring labour intensity directly 
and systematically into the analysis of priorities.
Finally, TAC reflected on the possibility of employing the production function for 
knowledge, the description of the relationship between inputs to research and outputs from 
the process, in its assessments. Conceptually of great importance, there is little information
32
about the function, hence little opportunity to use it in priority setting at TAC’s level. For 
this case, and for several of the others, TAC will join those engaged in developing the kind 
of information required to permit their future use in priority setting.
3.10. Conclusions
This Chapter has listed and described the major elements on which TAC’s priority 
setting was based. Each section opens with a short statement on TAC’s conclusions or 
findings. The remaining text expands on those points and others. From the overarching 
goals of the System, through the avenues by which the activities of the centres satisfy those 
goals, to the modus operandi and tools used by TAC in its deliberations, by the devices 
through which the locus and level of poverty are incorporated in the deliberations, and to 
a brief look at several other elements that might play a role at the level of TAC or at the 
level of the centres, the chapter provides insights into the elements that guided priority 
setting. Several important considerations were not treated, e.g., the view of the regional 
fora and national programs and the perceptions of the centres. Some of these are explicitly 
treated in Chapter 5, where the findings of this chapter guide the discussion of TAC’s views 
on where changes in resource allocations should occur, some in Chapter 7 where the 
implications for priorities and the resource allocations themselves are laid out. Some must 
await the further interaction between TAC and the centres as the latter develop, review, and 
amend their medium term plans with TAC’s support.
33
CHAPTER 4 - OVERVIEW OF OUTCOME OF RECENT STUDIES 
RELEVANT TO CGIAR PRIORITIES AND STRATEGIES
4.1. Background
The TAC priorities and strategies exercise took into account the outcome of a 
number of recent studies specially commissioned by TAC or by the Group through its Task 
Force or Standing Committee mechanisms. The TAC commissioned studies comprised the 
following: the inter-centre review of rice (1992); the stripe study of genetic resources in the 
CGIAR (1993); the global livestock research strategy (1994); the CGIAR commitments in 
West Africa (1994); the inter-centre review of root and tuber crops (1995); the study of 
public policy and public management research (1995); the study on institution strengthening 
research and service (1995); soil and water aspects of natural resources management research 
(1995); and the mid-term review of ICLARM (1995). The Group commissioned studies 
comprised the reviews by the Task Forces on Sustainable Agriculture and on Ecoregional 
Approaches; and the work of the genetic resources policy committee.
TAC also took into account the outcome of other major initiatives such as the Inter- 
Ministerial Meeting in Lucerne - the Lucerne Declaration and Action Programme; the LAO 
Agricultural Towards 2010 Study; the ILPRI 2020 Vision for Pood, Agriculture and the 
Environment; and the NARS Working Group Action Plan to Strengthen NARS-CGIAR 
Partnership.
The outcome of the above studies is discussed in the following sections in the 
context of future CGIAR priorities and strategies.
4.2. The Lucerne Declaration and Action Programme
The Lucerne Declaration and Action Programme focused its attention on the CGIAR 
Research Agenda and on the question of broader partnership within the global research 
system.
The CGIAR was urged to pay particular attention in its research agenda to the 
problems of the poor in less-endowed areas, in addition to continuing work in high potential 
areas; to take account of the fact that Sub-Saharan Africa and South Asia were facing the 
greatest challenges to eradicate poverty and malnutrition; that priorities need to be given to 
increasing productivity, protecting the environment, saving biodiversity, improving policies, 
and contributing to strengthen NARS, while addressing more forcefully international issues 
of water scarcity, soil and nutrient management and aquatic resources; and that actions 
already taken to protect genetic resources needed to be reinforced.
In terms of research strategies, the focus must remain on producing international 
public goods through problem-solving strategic and applied research using an interdisciplinary 
and collaborative approaches; that the research agenda should reflects the views and goals
34
of global and regional forums on agricultural research; and that the CGIAR System should 
work in closer partnership and collaboration with public and private research organizations 
and universities from both the developing and the developed world.
To achieve broader partnerships, priority must be given to include more developing 
countries as active members of the CGIAR; and to complete the transition from a 
donor/client to an equal partners approach that would systematize participation by NARS in 
setting and implementing the Group’s agenda. As a means of improving dialogue among 
CGIAR, private sector and civil society, the CGIAR should convene a committee of NGOs 
and a committee of the private sector organizations.
4.3. World Agriculture Towards 2010
This FAO study attempts to provides insights on two issues in food and agriculture 
of global importance that seem to dominate all others:
the persistence of under-nutrition and food insecurity for large parts of the 
developing countries’ population; and
the process of increasing scarcity and degradation of agricultural and other 
environmental resources as it relates, directly or indirectly, to the process 
of meeting the food and income needs of a growing world population.
From a development perspective, the study addresses two themes, one related to 
economic policies that have a bearing on agriculture, with particular reference to the 
significance for agriculture of the recent thrust of policy reforms aimed at structural 
adjustment; and the second concerns the proposition that in the low-income countries with 
food security problems and high dependence on agriculture, relative shift of emphasis 
towards agriculture and rural development holds more promise than other strategies of 
generating benefits both for food security and poverty alleviation as well as for putting the 
whole economy on to a higher growth path.
The study recognizes that technological opportunities are not sufficient in themselves 
to stimulate ecologically sound and sustainable development. A strategy for minimizing 
development and environment trade-offs must also be in place to provide stability and 
confidence for the producers to take up the technological opportunities and make the 
necessary investments. Access to proven technologies, production inputs and services, 
markets; secure rights to access to land, water and other resources provide the and an 
appropriate regulatory environment to safeguard and protect public goods must form an 
important part of the support measures that can ensure that the minimization of trade-offs and 
actions to shift agriculture on to a sustainable growth path take place within a consistent 
strategic framework.
The study points out some major uncertainties. For example, the question of whether 
the trade-offs between environmental resources and more production will conform to some 
notion of tolerable environmental cost cannot be answered in a definitive manner, given the
35
present state of knowledge and the uncertainties concerning future developments in 
technology. Also, at present there is no way of making any definitive statements about the 
prospect of whether future world agricultural output with enough per caput production to 
satisfy the needs of all people can be produced year after year in a sustainable manner. 
Further, because of the complexities in the relationships between the economy, agricultural 
development and the environment, and the uncertainties in the strengths of the trade-offs and 
their associated risks, the stress must be on minimizing the trade-offs, and adopting 
precautionary principle.
The study makes the following projections to 2010:
world agricultural production will grow at a rate of 1.8 percent p.a. to 
2010 (2.6 % in developing countries and 0.7 in developed), lower than 
the 2.3 % of the ’70s and the 2.0 % achieved between 1980-92, 
responding to a decreasing demand from people that do not need more 
food and those that cannot afford to buy it;
while per caput food supplies in most developing regions will be at or 
above the 3000 calories, under-nutrition could still affect 200 million 
people in South Asia, and 32 % of the population in Sub-Saharan Africa;
per caput production of cereals in developing countries will continue to 
grow from 216 kg in 1988/90 to 229 in 2010, but consumption for all 
uses will grow faster from 235 to 254 kg, requiring imports to grow from 
90 million tons in 1988/90 to 160 million in 2010;
in developing countries the use of cereals to feed an expanding livestock 
sector may more than double to some 340 million tons, or about 23 % of 
their total use;
increased production will be achieved through a 90 million ha expansion 
in land under cultivation, mostly in SS A and LAC, and an increase in 
cropping intensities in developing countries from 79 to 85 percent 
(additional 30 million ha)
if all additional cropland were to come from the forests, tentative 
estimates for 69 countries indicate that deforestation would go down to 
0.25 % from the 0.8 % in the ’80s (though the latter included 
deforestation from all causes);
as marine catches are not likely to exceed 100 million tons, there will be 
a severe shortage of supply and increase in the price of fish, while 
aquaculture production may go up to 15-20 million tons from the present 
level of 12 million.
The study identified the following development priorities, with many specific
aspects:
36
increasing sustainable productivity in the higher potential areas; and
reversing degradation and stabilizing production in marginal lands.
The study proposes the following research to meet the above future priority 
development needs. Increasing sustainable productivity in the higher potential areas 
would require:
the improvement of farm production/recycling of biological inputs through 
research on: nutrient recycling processes and techniques; NRM at village 
level; and integrated crop/livestock management systems;
the raising of yield ceilings through research on: resistance/tolerance to 
soil moisture and nutrient constraint; micronutrient deficiencies; and soil 
dynamics under low- and high-input crop production;
the improvement of irrigation management through research on: raising 
the efficiency of flood irrigation; adapting surge irrigation to conditions 
in developing countries; economical wastewater treatment methods; and 
institutional modalities of successful irrigation management systems;
the decrease in crop-weed competition through research on: weed 
management; biological control methods and biodegradable herbicides; 
ways to reduce herbicide applications;
the use of energy more efficiently through research on: energy-agriculture 
interrelationships for different ecosystems; integrated management of 
energy and other inputs (e.g., agrochemicals); and potential of biofuels for 
environmental and land use policy situations.
Reversing degradation and stabilizing production in marginal lands would require:
the conservation of local water through research on: minimum tillage 
systems for low-income farmers in the drylands; and pasture improvement 
in extensive range areas;
the lowering of the relative costs of external inputs through research on: 
cheap techniques to use low-grade phosphate rocks; and pigeon peas 
mechanism for releasing bound soil phosphate, to transfer it to other 
crops;
the integration of crop-livestock production through research on: reduction 
in labour and other constraints to adoption of silvopastoral systems; 
production of high-protein feeds and forages in legume-based systems; and 
ley farming systems for low fertility soils.
37
4.4. 2020 Vision for Food, Agriculture and the Environment
This IFPRI-led initiative highlighted the fact that over 1.1 billion people live on 
incomes of a dollar a day or less per person, and the gap between the rich and the poor is 
widening. The projections to 2020 included the following:
90 million people are likely to be added to the world’s population every 
year;
the number of malnourished children is likely to decline only slightly from 
185 to 156 million by 2020;
demand for food grains will increase by less than 3 percent, for livestock 
production by 17 percent, and for roots and tubers by 1 percent;
2 billion ha have been degraded in the past 50 years, 180 million ha of 
tropical forests were converted to other uses during the ’80s, marine 
fisheries are collapsing in parts of the world, and seasonal and regional 
water shortages afflict most developing countries.
Priorities for the 2020 Vision were seen to be:
the eradication of hunger and malnutrition, sustainable management of 
natural resources, and efficient, effective, and low cost agricultural 
systems.
the strengthening of the capacity of developing-country governments to 
perform appropriate functions;
the strengthening of the agricultural research and extension systems in and 
for developing countries.
Recommended strategies for action called for:
the enhancement of productivity, health, and nutrition of low-income 
people and increase in their access to employment and productive assets;
the promotion of sustainable agricultural intensification and sound 
management of natural resources, with increased emphasis on areas with 
agricultural potential, fragile soils, limited rainfall, and widespread 
poverty;
the development of efficient, effective and low-cost agricultural input and 
output markets; and
the expansion of international cooperation and assistance and improve its 
efficiency and effectiveness.
38
4.5. Genetic Resources Policy Committee
The Genetic Resources Policy Committee of the CGIAR in its report to ICW’95 
highlighted the following priority needs:
the CGIAR should expeditiously develop its guiding principles on 
Intellectual Property, to be used for policy development by centres and to 
set up material transfer arrangements;
the CGIAR’s interests and perspectives on key policy issues need to be 
brought to the attention of the various policy-making bodies;
to recognize the ethical concerns of farmers, indigenous communities and 
others about issues relating to the conservation and use of genetic 
resources;
that the gender dimension becomes an integral part of CGIAR’s research 
agenda relating to the conservation and sustainable utilization of genetic 
resources;
centres working in the area of recombinant DNA technology adhere to the 
regulations in force in the places where they work in the field testing of 
transgenic material;
that the CGIAR strengthen its capacity to undertake analytical, 
information, coordination and representational functions on policy issues 
through the establishment of a genetic resources policy unit at IPGRI;
to further pursue the organization of a workshop explore the ethical 
dimensions, while CGIAR centres assist interested NARS in the 
development of appropriate policies and procedures for the recognition of 
rights, and gather information on the origins of economically important 
genes and traits;
that the IARCs promote, in cooperation with NARS and appropriate 
NGOs, symbiotic social contracts between private and public sector seed 
industry and rural women, to enhance their income and livelihood 
security;
that the centres be proactive in providing assistance to NARS, when 
requested, in the development of a regulatory framework for the safe use 
of biotechnology.
39
4.6. Task Force on Sustainable Agriculture
The Task Force recommended the need to address the issues concerning: the 
degradation of both marginal and high productivity irrigated lands; the problems of sustained 
cultivation of steep and sloping lands; the problems of deforestation and loss of biodiversity; 
the threat posed by declining reserves and increasing competition for water resources; the 
economic and policy measures required to increase sustainable productivity, while preserving 
the resource base. It proposed that an increasing proportion of funding be earmarked for 
sustainability related research; that the proposed activities for the development of a better 
information base be strengthened to indicate where areas most at risk are located; that 
research on the socioeconomic basis of sustainability be strengthened, re-examining the 
association of research on income generating crops with CGIAR activities; and research into 
the environmental, institutional and social aspects of resource management and planning be 
strengthened, as they affect adoption of research products.
The Task Force further recommended that sustainability research should be funded 
in the form of incentives or seed money to encourage the development of consortia and 
networks; that the various on-going initiatives relating to soil, water and nutrient management 
in and out of the CGIAR be consolidated into a coherent, integrated programme; that the 
inter-centre programme on integrated pest management that incorporates AROs and IARCs 
be strengthened into a Consortium; that the ecoregional and Systemwide activities related to 
sustainability issues be advanced and refined by adopting an interdisciplinary approach that 
integrates productivity, environment and sustainability concerns; that consortia be established 
and present networks strengthened to forge closer links with NARS, NGOs, AROs and 
others; that the capability of the CGIAR System in the management of broadly based research 
consortia be strengthened.
4.7. Task Force on Ecoregional Approaches
The main conclusions of the Task Force were to:
use the US$ 10 million set aside for Systemwide and ecoregional 
initiatives exclusively for the orientation phase, following a "matching 
fund" approach to attract NARS and other local partners;
substantially increase the CGIAR contribution to ecoregional programmes 
in the coming years;
raise IARC’s contribution to ecoregional programmes to the 39% level 
recommended by TAC, which can be partially achieved by reorganizing 
existing activities in NRM and production systems research;
apply a holistic, systemic approach to R&D that promotes the 
simultaneous integration of research, policy and action programmes in an 
appropriate mix for particular agro-socioeconomic settings.
40
develop the ecoregional programme in two phases: i) an orientation phase 
to establish the programme, including a research agenda, a governing 
structure, research management, division of labour between partners, and 
monitoring and evaluation mechanisms, and ii) an implementation phase, 
in which the ecoregional programme is operational;
adopt 15 year as an initial time-horizon for ecoregional programmes.
4.8. Action Plan to Strengthen NARS-CGIAR Partnership
The NARS Working Group propose an Action Plan to:
develop improved mechanisms to ensure a stronger reflection of the 
collective views of NARS in CGIAR priority setting;
improve modalities for developing formal partnerships between IARCs 
and NARS;
enhance institutional capacity building at both the regional and national 
levels;
establish with donor assistance four viable regional fora supported by a 
small Executive Secretariat, as focal points that handle the views of 
broadened NARS in setting CGIAR priorities;
develop case studies of NARS/CGIAR interactions involving each of the 
regions, to identify successes and failures in collaborative research;
strengthen individual NARS by broadening their membership, developing 
priority setting mechanisms and improving research-extension linkages;
improve technologies to facilitate communications among NARS and 
between them and other members of the global research system.
4.9. TAC Commissioned Studies
4.9.1. Stripe Study of Genetic Resources in the CGIAR
The study’s main conclusions were that:
centres operate collectively as a System for genetic resources by: defining 
a unified direction; developing interactive genetic resources conservation 
and development activities, particularly with regional partners; integrating
41
existing efforts across centres; and exercising greater accountability to the 
major stakeholders.
all work concerned with the conservation of genetic resources should be 
integrated into a single Systemwide Programme, within which policies will 
be developed and coordinated.
IPGRI be reformulated as the International Agricultural Genetic Resources 
Institute (IAGRI), to be responsible for the Systemwide Programme
there should be a specific Genetic Resources Programme Fund in the 
CGIAR, which would provide funds necessary to operate the Systemwide 
Programme;
IARCs germplasm collections should be held in trust, and they should not 
seek to benefit financially from their commercialization;
a standardized information management system for the GRP be created, 
to integrate databases across the CGIAR and to simplify communications 
with NARS
Following a joint session with Centre Directors and Cosponsors, TAC recommended
that:
IPGRI should become the lead Centre for the Genetic Resources 
Programme, its DG being the GRP Director with overall responsibilities 
for the principles of the CGIAR’s involvements in genetic resources, 
including some policy and representational responsibilities for fisheries 
and livestock;
Genetic Resources Units in CGIAR Centres should have a programme or 
equivalent status, their day-to-day management remaining with centres in 
which they are located;
there should be a GRP Fund from which earmarked financial contributions 
will be allocated to all centres following TAC recommendations.
4.9.2. Inter-Centre Review of Rice
The review panel concluded that:
the overall level of core funding for rice research in the CGIAR be 
maintained, and that regional allocations should be commensurate with 
global needs;
resources should be shifted from West Africa to Asia, which will consume 
over 90% of the increased output in rice required by the year 2030;
42
productivity improvement be brought about through technological change, 
the most important way to achieve the levels of production required by a 
twofold increase in demand;
research should focus on (i) raising the yield potential for irrigated 
environments, (ii) managing pest and diseases to close yield gaps, and (iii) 
managing natural resources to sustain yields in irrigated, upland and 
inland valley environments;
efforts in the less favourable environments should be closely monitored, 
to ensure that the potential gains that must be eventually realized are 
commensurate with the present high level of funding (20%);
IITA and WARDA activities should be integrated, so that the reduced 
level of core support can be effectively applied.
TAC concluded that:
the global share of CGIAR funding for rice research should not be 
reduced;
in principle resources allocated to research in Asia should be increased, 
possibly by shifting some responsibilities from IRRI to Asian NARS;
given CGIAR’s policy decision to support rice improvement in WA, TAC 
finds no compelling reasons at this time to adjust the level of the resource 
envelope which it assigned to WARDA in the MTP for 1994-98;
over the ensuing 5 years (1999-2003) CGIAR core funding to WARDA 
should be at a level consistent with its contribution to CGIAR priorities;
that an IITA-WARDA integrated programme for resource management 
research is required, at least for the inland valley ecosystem of West 
Africa;
the current regional rice research mandates of CIAT for LAC and 
WARDA for West Africa should continue to be supported;
the CGIAR may be over-investing in the rainfed lowlands and uplands, 
although the issue warrants further consideration.
4.9.3. Livestock
TAC’s Livestock Strategy Working Group in April 1993 recommended that:
the relative overall allocation of funds to livestock research should remain 
as it is;
43
the current concentration of support in sub-Saharan Africa should be 
redistributed to both global and ecoregional programmes;
on a programmatic basis the systemwide balance of efforts between animal 
health/disease research and animal production should continue as present, 
i.e., 1/3 and 2/3 respectively;
research should focus on productivity of milk, meat and traction by cattle, 
sheep and goats, giving greater emphasis to a holistic approach in the 
context of crop-livestock/agroforestry systems;
the relative imbalance of resources between those allocated to sub-Saharan 
Africa and other regions should be addressed;
integrated crop-livestock research on a regional basis should have a high 
priority (i) in the subhumid tropics and highlands of sub-Saharan Africa, 
(ii) in the upland rice production systems of Asia, and (iii) in 
ecoregionally-oriented production systems research in the other regions.
there is a need to develop an inter-centre framework for the coordination 
of livestock research among the CGIAR Centres and between them and 
other research institutes; the coordination framework must foster 
complementarities and synergies between global and ecoregionally focused 
research;
there are strong arguments for the integration of parts of ILCA and 
ILRAD programmes;
The Strategic Planning Task Force on Livestock in August 1994 recommended that:
new approaches should be followed to develop Integrated Health 
Management systems that maximize productivity by reducing the impact 
of some key animal diseases;
the programme in animal genetics be broadened to adapt new 
developments in molecular biology to methodologies for characterization, 
utilization and conservation of animal genetic resources, and techniques 
to identify superior genetic characteristics in tropical livestock;
animal physiology work be reoriented to strengthen research in animal 
genetics, nutrition and health;
new emphasis be given to work on feed resources, by integrating System- 
wide research on forage genetics and feeds nutritional value with regional 
activities on the production of feed resources;
44
strengthen research on livestock-related systems in an ecoregional context 
in close interaction with NARS, including increased emphasis on 
economic and social research, as well as on natural resource management.
The Global Consultation in January 1995 on the agenda for livestock research 
identified the following common research themes arising from separate Group reports, which 
would qualify for international research efforts:
Feed resources: production and utilization;
Production Systems Research: integrated management, including NRM;
Biodiversity: characterization, conservation and improvement of forage 
and animal genetic resources;
Animal Health: integrated management technologies;
Livestock Policy;
NARS strengthening: research on delivery of technologies.
The above recommendations provided a strategic basis to develop and implement a 
Systemwide livestock research programme covering the following:
a global research agenda on priority themes identified through global 
consultations and dealing mainly with animal genetics, health and 
nutrition, which ILRI will develop in collaboration with NARS and ARIs 
(e.g., rumen ecology);
livestock-related work dealing with common themes across centres in the 
areas of feed resources/systems/NRM/policy, coordinated by an Inter- 
Centre Livestock Programme Group (LPG) consisting of representatives 
of all CGIAR Centres engaged in livestock research, that operates as the 
overall linking mechanism between centres; and
Systemwide Initiatives to support collaboration among CGIAR Centres and 
between them and non-CGIAR institutions around research projects 
supported by a Special Fund, administered by the LPG along clear 
responsibility lines and on an open and competitive basis that includes the 
use of external review panels and a set of criteria to evaluate the proposals 
(e.g., the SLI on Feed Resources).
4.9.4. Review of CGIAR Commitments in West Africa
The panel recommended that:
45
there should be a decrease in IITA’s and ICRISAT’s research investments 
in the development and management of production systems (activities 
1.2.1 and 1.2.2 in the present classification); there should be an increase 
IARCs efforts on upstream germplasm enhancement and breeding 
(category 1.1) and on process-oriented research in natural resource 
management (categories 1.2.3/5.a and 2.5); long-term joint programmes 
be developed in which IARCs are broadly responsible for upstream 
research in germplasm improvement and natural resource management, 
devolving to NARS research on production systems and management; 
(TAC does not believe that outright devolution will be appropriate under 
the existing circumstances, relying on strengthening partnerships as a way 
to shift centres’ research towards the more strategic end of the spectrum);
an in-depth review of the current scientific approach to crop improvement 
in dry areas should be carried out, to know why "improved" materials are 
not better than the locals under field conditions; this should include a 
high-level review of ICRISAT’s crop improvement program, including 
CIRAD’s sorghum research;
millet research efforts should focus on agroecological zones where it can 
be integrated into more complex cropping patterns; this should involve 
shifting millet improvement from the Niamey site to a less arid area;
concentrate efforts to fortify NARS on training and information by 
consolidate similar training activities in the region along the lines of the 
"Inter-Centre Training Programme for Sub-Saharan Africa";
IARC’s regional research efforts should be harmonized by creating a 
common Board of Trustees for WARDA and IITA, with ex-officio 
representation of ICRISAT, ICRAF, and IRRI.
4.9.5. Inter-Centre Review of Root and Tuber Crops
The review recommended that:
global investments in R&D research should continue at least at the 
existing levels; and that there should be an increase in relative terms 
research on cassava and potato, while decreasing that on sweet potato;
an Inter-Centre Committee be created to increase inter-centre 
collaboration, to develop a Systemwide strategy; to promote collaboration 
on vegetative propagation and conservation, biotechnology, post-harvest, 
policy, training and mechanization; the Committee should commission a 
task force to rationalize international phytosanitation regulations affecting 
the shipments of vegetatively-propagated materials; the Committee should 
operate as long as it adds value to inter-centre activities in the CGIAR;
46
the Committee should take the responsibility of reconciling discrepancies 
in production/consumption data available from different sources to 
reassess priorities;
joint collaborative programme be worked out to devolve research on sweet 
potato to strong NARS; reductions in CIP’s research on sweet potato 
should be linked to opportunities for strategic partnerships with the 
Chinese Agricultural Research System;
CIAT, CIP and IITA activities in technology transfer aimed at overcoming 
constraints to the dissemination of improved cultivars be supported, but 
avoid duplicating the effective technical assistance work of other 
development agencies;
partnership with NARS, AROs and the private sector on postharvest 
technologies be explored, specially on characterization of starch and flour, 
food processing technologies and market research;
duplications with AROs on advanced biotechnology should be avoided, 
focusing CGIAR activities on safe and viable technologies for vegetative 
propagation, especially of cassava.
4.9.6. Soil and Water Aspects of Natural Resources Management Research
The study’s main conclusions are that:
there is a need to improve the state of information on land and water 
degradation and its impacts on agricultural, forestry and fisheries 
production; that the CGIAR System should develop improved 
mechanism(s) by which centres, collectively, can be involved with other 
partners in generating and interpreting improved scientific evidence on the 
extent and magnitude of the impacts of agriculture, forestry and fisheries 
on the degradation and/or enhancement of natural resources and the 
impacts of such degradation and/or enhancement of agriculture, forestry 
and fisheries production and food supply;
the CGIAR System has need for an integrated natural resources 
management (INRM) framework for research, linking NRM with poverty 
alleviation and protection of the environment; such a framework would 
involve four sets of inter-related linkages:
• links between productivity-enhancing and resource conserving 
research (e.g., crop improvement and natural resources 
management);
• spatial or landscape level linkages (e.g., upstream-downstream 
linkages in a watershed management framework);
47
• temporal linkages (e.g., links between present and future, or 
sustainability considerations);
• linkages between research and the diffusion/adoption of results from 
such research.
within the INRM framework, there is need for additional focus on specific 
subject matter; the linkages covered within the INRM framework need to 
be introduced into the CGIAR System not only through centre activities, 
but also to a great extent through work in the Systemwide Programmes, 
essentially those that implement the ecoregional approach; these include 
the emerging Water Initiative and the existing Soil, Water, Nutrient 
Management Initiative; the water-related research is a priority area in 
which the System’s scope and intensity of work should be significantly 
expanded, particularly related to water scarcity and competition, and water 
quality, but also in terms of the broader watershed management activities 
needed to optimize water availability and use;
there is a need for increased research to look at why there is a lack of 
application of known technologies and ideas, and why there is a lack of 
widespread progress in natural resources management using much of the 
available research-generated information and technologies; an intensified 
and expanded collaborative mechanisms and activities be developed among 
CGIAR Centres and between CGIAR Centres and their non-CGIAR 
partners to help focus increased research and institution strengthening on 
issues related to adoption, adaptation, and utilization of existing NRM 
technologies and knowledge that so far have remained unused;
there is a need for uniform and consistent criteria forjudging the priorities 
for NRM activities/research in the CGIAR Centres and Programmes;
4.9.7. Public Policy and Public Management Research
The main conclusions of the study were that:
the current efforts in research on common property resources be pursued, 
particularly in terms of solid empirical research;
the CGIAR mandate on P&M research be broadened to also capture the 
i) public-private interface, ii) co-production of public goods by public and 
private sectors, and iii) role and management of NGOs and other non­
profit organizations;
the work on global food and natural resource use projections be advanced;
the possibilities of research on the political economy of P&M decisions 
be explored.
48
the current level of investment in socioeconomic, policy, and management 
research be maintained;
the budget reporting on "policy" and "protecting the environment" 
programmes be disaggregated, to identify the real extent of P&M research 
in the System;
a Systemwide initiative in policy research as an instrument to enhance 
collaboration and coordination should not be pursued;
six specific criteria for priority setting be adopted, dealing with relevance 
to goals and efficiency,nature of outputs useful to NARS, comparative 
advantages, quality and impact assessment mechanisms;
a modality be defined that would allow TAC to allocate resources on a 
competitive basis in response to requests by two or more centres to 
support project preparation and coordination and methodological backup 
activities;
the potential of "new" institutional economics in P&M research be 
explored, that brings behaviour to understand determinants of choice of 
efficient institutions;
the centres should have a minimum capacity to collaborate with IFPRI on 
P&M research;
IFPRI’s and ISNAR’s involvements in ecoregional initiatives be reviewed, 
questioning the opportunities to derive lessons of international relevance 
and to achieve greater synergy in inter-centre collaboration;
the responsibility for studies on global food and natural resource use 
projection be assigned to IFPRI.
4.9.8. Institution Strengthening Research and Services
The main conclusions of the study were that:
that CGIAR activities in institution strengthening should be demand- 
driven, more structured, and based on collaboration with other partners 
(internal and external);
resources be shifted from services to research on institutional development 
related to agricultural research in developing countries, including the 
assessment of NARS institutional requirements, organizational structures, 
management practices, leadership skills, and planning, monitoring and 
evaluation tools;
49
careful review of TAC’s system to classify activities in relation to the 
"Fortifying NARS" category be made, to clearly distinguish those 
providing direct services to NARS.
additional resources be made available for ISNAR to conduct a more 
comprehensive analysis on NARS strengths and weaknesses, as a basis to 
propose a plan for inter-institutional collaboration;
there should be a closer collaboration among centres in all facets of 
institution strengthening support to NARS, particularly between the other 
centres and ISNAR;
generic, methodological tools that can be used by other organizations and 
consultancy firms be developed and disseminated, freeing resources to 
devote a greater portion towards R&D activities;
studies on successful and unsuccessful examples of institution-building 
activities, be carried out by ISNAR in collaboration with major donor 
agencies and/or with selected NARS;
joint projects between other centres and ISNAR on areas such as setting 
research priorities, programme planning and evaluation, project 
preparation and budgeting, could be developed under the umbrella of a 
"Systemwide initiative" in institution strengthening;
ISNAR should assist in the strengthening of emerging regional groupings, 
which can facilitate the Centre’s efforts both in institutional strengthening 
and NARS participation in CGIAR priority setting;
the information services provided to NARS partners be distinguished from 
those given to the Centre’s own scientists and to donors/public, as well as 
between advice on research in mandated commodities from that on 
management of research in general.
4.9.9. ICLARM Mid-Term Review
The review concluded that:
the inadequate resources available to ICLARM for carrying out its mission 
be increased, to which TAC and the donor community should give the 
utmost priority; (TAC agrees with the members of the CGIAR that higher 
priority should be given to aquaculture research, and therefore 
recommends that additional US$2 million are allocated to ICLARM’s core 
funds for 1996, in addition to a transfer of US$2.5 million to its core 
programme. ICLARM agreed core research agenda for 1996 will then be 
twice as large as it was in 1994;
50
the additional US$ 1 million which TAC conditionally recommended for 
ICLARM be allocated;
existing funds be restructured, to move some from the short-term and 
project based restricted core to unrestricted core;
the technical issues surrounding the feasibility of the sites proposed by the 
Egyptian Government be considered, including implications on its 
strategic plan and research priorities; (TAC considered that the Egyptian 
offer will need to be studied carefully from scientific/technical, institution 
development and financial perspectives);
strategic alliances with capable NARS and internationally assertive 
advance science institutes be sought to enhance its scientific capacity.
51
CHAPTER 5 - ANALYTICAL PROCESS : ACTIVITIES
Introduction
As discussed in Chapter 3, since 1992 the CGIAR’s research and research-related 
work has been classified in five major undertakings: conservation and management of 
natural resources; germplasm enhancement and breeding; development and management of 
production systems; socioeconomic, public policy and public management research; and 
institution building. In 1995, in preparation for the Lucerne Consultation, these activities 
were regrouped as follows1:
increasing productivity which combined activities in germplasm enhancement and 
breeding and production systems development and management;
protecting the environment which included activities in natural resources 
management;
saving biodiversity which included activities on germplasm collection and 
conservation;
improving policies which included all the socioeconomic policy and management 
work; and
strengthening national research programmes which included all the activities under 
institution building.
TAC began this priority analysis by considering, in Systemwide terms, what should 
be the relative distribution of CGIAR efforts among these five undertakings or broad activity 
categories. It thereby took the 1996 allocation of resources to each of these categories as a 
starting point in the discussion. Consideration was then given as to whether, on the basis of 
internal and external developments, a shift in priorities was warranted. Emphasis was 
thereby given to the outcome of the important Systemwide studies that had been conducted 
either by TAC or other institutions or CGIAR Committees and which are reported upon in 
Chapter 4.
The Committee reiterated that there is no clear dividing line between these categories 
of activities, and that in many instances, activities may overlap into several categories. 
There was no basis for quantitative approaches to support an analysis of this kind. TAC 
therefore relied on a delphi approach to engage in a broad discussion as to the future balance 
of effort of the CGIAR in each of these undertakings. TAC Members responded to a 
questionnaire and subsequently met to discuss their responses. In the ensuing discussion, 
consensus was reached on the broad directions the CGIAR would take. TAC also took into
TAC/CGIAR, 1995. The 1996 CGIAR Research Agenda. TAC Secretariat, FAO, Rome. The 
definition of activities concerned is listed in Annex 1.
52
account the recommendations of each of the regional fora. In the following sections, a 
rationale is provided on the future directions of the priority of each of these activities. 
Concrete recommendations on target percentages are provided in Chapter 8.
5.1 Increasing Productivity
5.1.1. Germplasm Enhancement and Breeding
TAC is in favour of an increase in resources allocated to germplasm enhancement 
and breeding research relative to other activities. It was noted that TAC’s 1994-98 MTP 
recommendations would have had the category at 21-23% of CGIAR resources but that it is 
at about 20% of the proposed 1996 budget, without including support transferred from 
complementary to the agreed agenda, and at 19% if such transfers are included. TAC 
Members agreed that the allocation should reach at least the range recommended earlier and 
perhaps should go to the upper end of that range. The rationale for this judgement is laid 
out in the following paragraphs.
In particular TAC recommends an increase in pre-breeding activities (1.1.1) relative 
to other activities in (1.1). This work emphasizes the incorporation of novel genes from 
various sources, including from like species, from primitive materials and wild relatives, and 
from unrelated species. TAC notes that the pre-breeding, by its nature, has the broad 
generic applications which are desired characteristics of CGIAR activities and that 
opportunities are good in this arena. As well, TAC notes that there are still opportunities 
through the applications of molecular biology in each of the areas in which breeding is 
currently carried out. These are largely in the areas of marker assisted breeding, with its 
promise for more efficiency and more rapid progress. TAC does not recommend an 
expansion in work on transgenic transfers and observes that, with few exceptions, this field 
awaits advances in the underlying science, which itself is largely in the hands of advanced 
scientific institutions.
TAC recommends no increase in conventional breeding for crops (1.1.2), noting that 
several NARS are full partners in such activities while many more still rely heavily on 
CGIAR for finished or quasi-finished materials. TAC’s sense is that centre efforts need to 
be rebalanced, with emphasis on incorporating primitive and novel genes through pre­
breeding and on those crops of more importance to countries with less advanced NARS while 
reducing its efforts on those crops of most importance to countries with stronger NARS. The 
former will continue to reap large benefits from the CGIAR’s intraspecific work, while the 
latter will gain most from the prebreeding activities.
TAC recommends an increase in livestock improvement (1.1.3). It was noted that 
there is a considerable investment in improving species and breeds of importance to the 
CGIAR and that the System must stay abreast of such work. Overall, TAC recommends that 
ILRI consider more effort on molecular genetics related to disease resistance and is disposed 
to support such an increase, as part of the overall effort to enhance resistance to 
trypanosomiasis and theileriosis in livestock.
53
Discussion of tree breeding (1.1.4) concluded that a marginal increase is justified 
in this area. There are further opportunities through identifying and testing elite materials, 
the phenotypic selection and initial rounds of cloning of elite materials. It was thought that 
large scale cloning and distribution should be the province of NARS. Methods to improve 
cloning were discussed, but TAC converged around the idea that much of such work is 
underway in non-CGIAR centres, including some in developing countries. The importance 
of high-value trees yielding special chemicals for the pharmaceutical and other industries was 
considered. The possibility of extending the CGIAR’s efforts to fruits and other high-value 
trees was discussed. It was concluded that there are limited opportunities for generic work, 
that some location-specific breeding is underway in the private sector, and there were 
questions about the extent to which the poor would benefit as special conditions appear to be 
required for the adoption of improved materials. TAC should continue to encourage CIFOR 
and ICRAF for further clarification of their roles in this area.
TAC recommended an increase in the investment in fish breeding (1.1.5); in 
particular the opportunities for improving carp species was noted. ICLARM’s work on 
tilapia produced notable genetic gains, is widely known among specialists in fish breeding 
and has promise for the poor. Some specialists now see good opportunities for carp. Few 
others are working in this arena, largely because the possibilities for improving common fish 
strains were thought to be limited. TAC believes that the successes with tilapia, and the 
realization of the promise for carp, will bring increased non-CGIAR investment into play. 
Against this background, TAC recommends an increase in CGIAR funding to this activity.
5.1.2. Production Systems Development and Management
In considering activities under this category, TAC started from the following
presumptions:
that the over-riding significance of the CGIAR’s work is the improved production 
and supply of food for the poor, especially in rural areas;
this will lead to the general economic advance of those regions and thus create 
diversification in the opportunities for income generation;
all of this will be done in the context of environmental care so that land and aquatic 
resources of all kinds will be preserved, and if possible enhanced;
TAC’s concerns are that the research for which the CGIAR is responsible should 
have equal concern for the conservation of on-site resources as well as resources at 
locations away from the immediate places at which new production technologies are 
deployed;
although most attention in the CGIAR has been paid to the conservation of 
biodiversity, TAC believes that equal regard should be paid to the conservation of 
physical attributes of the environment, including the especially valuable attributes 
of soil, water and terrain;
54
TAC believes that there should be continuing investment in study of the 
effectiveness of the transfer of new technologies to their ultimate users especially 
of those technologies which introduce novel approaches to environmental 
management.
TAC has consistently emphasized that research on agricultural production systems 
must be carried out with a sustainability perspective. The purpose must be to enhance 
productivity with minimal degradation, or if possible improvement, of the natural resources 
at the site of production and locations external to that site. Otherwise the potential for food 
production by future generations is put at risk.
The CGIAR currently allocates approximately 24% of its core resources to research 
on production systems. This is a rapid reduction from 33% of resources that were allocated 
to this category during 1991. In 1992, TAC had recommended a reduction of efforts in this 
category from 33% to about 29% in view of the increasing national capacity to deal with 
these activities which mostly consist of applied and adaptive research. TAC also noted at 
that time that the mode of operation for this type of research should evolve towards greater 
use of networks and consortia. While the Committee is aware that a reduction in this type 
of work is already greater than TAC’s 1992 recommendations, it believes that a further slight 
decrease in resources allocated could be justified, primarily through more efficient 
organization of the work and the growing ability of NARS-based networks to assume 
responsibility for this type of work. In the future, this type of research should have both off­
site and on-site components, and should have a direct impact on sustainability. Production 
systems research, which has no direct bearing on sustainability, should be gradually phased 
out. Over the next three to five years, TAC will carefully keep track of ongoing activities 
to monitor developments.
Given the importance of productivity increasing resource conserving technologies 
TAC recommends that in the future, activities under "Production Systems Development and 
Management" should not be portrayed separately from those under category 2 "Protecting 
the Environment". As work on production systems should not be to the exclusion of 
conserving natural resources, nor should work on natural resources management be to the 
exclusion of productivity. This perception suggests combining both categories of activities. 
However, such a change must be made over time so that comparability with past allocations 
can be assured. More on this in Chapter 8 on recommendations. Current activities on 
production systems research are labelled under four headings: cropping systems; livestock 
systems; tree systems; and fish systems.
• Cropping Systems
Cropping systems research currently accounts for approximately 15% of CGIAR 
core resources and approximately two thirds of all production systems research. This 
predominance of work on cropping systems reflects the traditional focus of the Group on the 
major food crops of concern to the poor. With the expansion of the CGIAR in 1992, 
increasing emphasis has to be given to the newer production sectors, in particular trees and 
fish and also to livestock. Increased resources for these three production sectors will need 
to be derived from reduced investment in cropping systems research.
55
• Livestock Systems
Livestock production systems research currently accounts for 6% of the CGIAR’s 
core resources which is well below what was recommended by TAC during 1992. TAC in 
particular believes that much greater resources have to be allocated to livestock research, 
particularly in mixed farming systems because of its current inadequate coverage. TAC 
thereby notes that the contribution of livestock to the sustainability of production systems is 
vital, and that there is a strong increase in demand for livestock products over the next two 
decades and of the importance of intermediate products such as traction and manure to poor 
smallholder farmers.
• Tree Systems
Tree systems research currently account for only 2% of the CGIAR core resources 
and there is a need to expand the CGIAR’s efforts in both forestry and agroforestry systems. 
This will be further illustrated in Chapter 6 on production sectors.
• Fish Systems
Research on fish systems is currently allocated 1% of the CGIAR’s core resources. 
At Lucerne, it was recommended that the Group strengthen its efforts in aquatic resources 
research management. TAC, therefore, believes that increased resources can be allocated 
to research on this activity.
5.2. Protecting the Environment
This category of work accounts for approximately 16% of the CGIAR’s core 
resources and primarily deals with research on natural resources management. TAC has 
made a comprehensive review of the future of this type of work.2 TAC believes that the 
overall allocation of resources to this type of work is adequate and can be kept constant but 
that the CGIAR needs to give greater emphasis to strategic research and to its off-site 
components. There is also a need for additional focus on specific subject matters, in 
particular between productivity enhancing and resource conserving research, spatial or 
landscape linkages, temporal linkages and linkages between research and diffusion/adoption. 
As already indicated under Section 5.1.2. there is a need to integrate activities in this 
category with those of production systems into a joint category of "Sustainable Production 
Systems".
TAC Secretariat, 1995. Priorities and Strategies for Soil and Water Aspects of Natural Resources 
Management Research in the CGIAR. FAO, Rome. (SDR/TAC:IAR/95/14)
56
5.3. Saving Biodiversity
The CGIAR currently allocates approximately 10% of its core resources to activities 
on biodiversity, in particular, germplasm enhancement, conservation, characterization and 
evaluation. This activity accounted for 6% of resources during 1991 and TAC, at that time, 
recommended that it should be increased to about 8%. The rapid increase in resources 
allocated to saving biodiversity exceeding substantially the levels that were originally 
recommended by TAC reflects the growing importance that is being attached to this work 
as also illustrated at the Lucerne Conference.
The CGIAR is actually a major depository of large ex situ collections for the main 
food crops which were largely assembled before the beginning of the last decade. The 
Convention of Biological Diversity (CBD), which occurred in 1992, and the subsequent 
developments in this arena created a totally new environment in terms of germplasm value, 
germplasm conservation and exchange.
In such a new environment, the CGIAR should play a major role at an international 
(or transnational) level. The CGIAR/FAO Agreement that was signed at ICW’94 placed 
CGIAR collections under the trusteeship of FAO for the benefit of mankind. The 
international centres still have the responsibility of preserving them, characterizing them and 
boost their utilization as a way to promote sustainability of agricultural production and to 
alleviate poverty.
The present difference between the current allocation and 1992 is due to a more 
positive public attitude towards conservation of genetic resources, the increasing involvement 
of the international centres on sustainability and agroecological research and the entry into 
the CGIAR of centres like ICLARM, ICRAF and CIFOR, which are giving increased 
emphasis to genetic resources conservation and utilization than before.
As in many agricultural research organizations, CGIAR genebanks have been centred 
on the mandate crops and their wild relatives. Forest and some animal resources are now 
within the CGIAR responsibilities. These new fields bring their own specific methodological 
problems, like in situ preservation technologies or genetic diversity characterization. The 
CGIAR cannot, however, rely on indoor research only to solve such problems and must 
make the best use of other actors’ activities in the domain. In situ conservation implies large 
scale country specific activities which cannot be taken up, nor should they be, solely by the 
CGIAR Centres. TAC is in favour of increased support to CGIAR activities related to the 
conservation of animal genetic resources, thereby focusing on genetic tools to help 
characterize populations and identify those populations which actually justify conservation.
Resulting from the stripe study on the situation of genetic resources in the CGIAR, 
and due to the new environment created by the signature of CBD, a more uniform treatment 
of biodiversity issues among the CGIAR Centres was considered necessary and a 
coordination mechanism for such was proposed and created centred at IPGRI and under the 
supervision of the Intercentre Working Group on Genetic Resources.
57
A modest increase in the priority for saving biodiversity should be specifically 
allocated to germplasm related activities in the different centres and to more research on 
conservation per se, such as on core collections, seed pathology and seed physiology. Such 
activities are mainly those of germplasm collection, characterization (be it molecular or not), 
evaluation, rejuvenation, the study of diversity levels sampled and/or maintained at ex situ, 
in situ and on-farm level work, methods of preservation for some species, etc.
Considering also that for some species (if not for all), molecular characterization 
may be the most efficient way of doing such tasks, and it is also relatively new in the 
CGIAR system, this is another area where an increase in activity is expected, even 
considering that advanced research institutions should take up a major part of the task 
(national programmes do not have a comparative advantage in this case).
A consideration here is that although it could be thought that some of the above 
mentioned tasks could be performed with funds that are now used for 
preservation/rejuvenation/characterization of the present collections and that such tasks will 
diminish as some of those tasks approach a steady state of relative completion, that is not the 
case at least for the next five to ten years. The present status of the CGIAR collections 
indicate that they will still demand a long time to reach the point where most of them will 
be characterized, safely duplicated and adequate samples placed under long-term storage. 
Since this is part of the agreement that was signed by the CGIAR when the collections were 
placed under the trusteeship of FAO, the Centres cannot decrease their activities in such 
tasks.
Taking all these considerations together, a small increase in allocation for such 
activity category is justified, mainly to cover the new research areas, focusing on generic 
activities and technologies that are specific of the CGIAR responsibilities and not already 
covered by others.
5.4. Improving Policies
TAC takes as its starting point the recommendation of the Stripe Study on Policy and 
Management Research in the CGIAR that the current level of investment in socioeconomic, 
policy and management research be maintained.
Within that broad category, however, some changes in emphasis are called for, in 
particular, the relative weight between economic and social analysis (activity 4.1) and policy 
analysis (activity 4.2). The line between the two categories is to some extent arbitrary, but 
within the continuum of issues covered under these two categories, TAC would recommend 
a shift in balance towards economic and social analysis.
TAC recognizes that policy work within the CGIAR has been valuable at a time 
when many developing countries were liberalizing their market and trade regimes in the 
1980s, but now that many of the reforms are in place, the urgent need that attended that 
transitional period has lessened. Further, liberalized regimes by definition would involve
58
relatively less sectoral intervention and therefore there will be in the long term less need for 
the kind of sectoral policy work that the CGIAR has been concentrating on.
On the other hand, as the CGIAR is moving toward more research on natural 
resources management, there will be increasing need for work that seeks to understand 
decision making by collective entities other than the state. The scope for social and 
economic analysis will correspondingly expand. The CGIAR with its close contact with 
some of the issues raised by natural resources management would be well placed to initiate 
research on them. Further, with the basic social science in this area being substantially 
incomplete, research in this area can be expected to yield results which are of international 
significance.
TAC recommends that research on the governance and management of public 
systems (activity 4.4) remain the same as currently, notwithstanding the fact that the research 
part of the work done on institution building and advice to NARS (former activity 5.3) will 
in the future be included in categories 4.3 and 4.4.
With respect to activity 4.4, research on organization and management of institutes, 
TAC recommends that, in line with its views expressed in the strategic study on institution 
strengthening research, more resources be allocated to this type of work.
5.5. Strengthening National Programmes
• Training
It is TAC’s view that current limitations on NARS are less in the area of human capital and 
more because of limitations on operating capital, on physical facilities, on appropriate 
incentive systems, and on effective management and administration. NARS do continue to 
request assistance from the CGIAR, especially for support for graduate training. TAC notes 
that such training is better financed through the loans and bilateral arrangements that are 
available, especially to national programmes, than through the limited funds available to the 
CGIAR. As for the other limitations, only that pertaining to management and administration 
is within the CGIAR’s purview (see discussion on institution building/advice to NARS). 
TAC recommends a continuing reduction of Investment in training until such time as the 
more evident limitations on NARS have been resolved.
• Documentation, publication and dissemination of information
TAC notes that this activity includes some activities which really pertain more to centres’ 
efforts to keep the world apprised of their work than to developing the information facilities 
of the NARS. TAC recommends that these expenditures be transferred to the administrative 
budgets of centres and treated as a part of overhead. This will result in an apparent 
reduction of resources allocated to activity 5.2. On the other hand, TAC notes the growing 
opportunities through modem communication technology and feels strongly that more 
investment is required in this area. While TAC feels that bilateral and loan arrangements
59
are the logical source of such funding, TAC believes that the CGIAR should play an 
advocacy role on behalf of the NARS.
• Institution building/advice to NARS
Since there is great need to increase applied research to underpin strengthening of NARS and 
to synthesize the experiences gained within the CGIAR, TAC recommends an increase in 
investment in this activity. This is consistent with TAC views on the future of institution 
strengthening in the CGIAR. Given its view that work in this area is much like work under 
4.3 and 4.4, TAC will recommend that, in the future, the two categories be combined and 
reported with undertaking 4. (TAC notes that to ensure comparability in budgets over time, 
there should be a period during which the two are reported as now and in the new format.)
• Networks
It is the view of TAC that the CGIAR should gradually phase out its investments in networks 
aimed at institution strengthening as this should be the responsibility of individual NARS and 
their respective subregional organizations, such as CORAF, INS AH, ASARECA and 
SACCAR in the case of Africa. This view is supported by deliberations that have taken 
place in the recently organized regional fora whereby great emphasis was placed on close 
linkages through networks within subregions due to the need to address common agricultural 
issues, TAC therefore recommends a further gradual reduction in investment in this
subcategory.
60
CHAPTER 6 - ANALYSIS BY PRODUCTION SECTOR AND
COMMODITY
6.1. Production Sectors
The 1992 review of CGIAR priorities and strategies did not try to address the 
relative balance of CGIAR efforts among agriculture (crops and livestock), forestry and 
fisheries. There were three major reasons for this. First, on the basis of valid analytical 
grounds, a different quantitative analysis was chosen for agriculture, forestry and 
fisheries, the results of which could not be added together. Second, CGIAR activities in 
forestry at that time were only just beginning while those in fisheries had still to be 
accepted. Third, TAC’s analysis did not provide grounds for an absolute decline in 
support for agricultural research. The Committee’s judgement, at that time, was that the 
proposed research programmes in agroforestry, forestry and fisheries should not be 
funded at the expense of critical research needs in crops and livestock.
With the rapid development of research programmes for these new production 
sectors since 1992, TAC now has the necessary information to proceed in making 
recommendations on the balance across different production sectors. Crops and livestock 
were thereby treated as sectors on their own. The Committee limited its quantitative 
analysis to CGIAR commodities only, or in the case of forestry and fisheries, to those 
commodities or aspects only of immediate relevance to the CGIAR. The analysis was 
also based on the likely production of these commodities in the year 2010 and the value 
was estimated on the basis of 1989-91 international commodity prices. A discussion of 
the methodology used is provided in Annex II. An overview of the importance of each of 
the production sectors as well as the need for research and specific role of the CGIAR 
therein is provided in Annex III. Please note that CGIAR commodities account for 
approximately 74% of the value of all crops, 52% of livestock, 79% of forestry and 72% 
of fisheries.
Table 1A provides an illustration of the relative importance of each of the 
production sectors both globally and by region based on their share of value of production 
by 2010. Crops account for 52.5%, livestock for 21.7%, forestry for 19.7%, and 
fisheries for 6.1% of the global value of commodities of relevance to the CGIAR. On a 
regional basis, Asia has the largest share of production value accounting for 58.3%, Latin 
America for 18.4%, Africa for 15.2% and the WAN A region for 8%.
Table IB presents an overview of the value of each of the commodities making up 
that of the production sector. It is shown that rice, fuelwood and charcoal, milk, beef 
and buffalo meat, and wheat are the five most important commodities globally, based on 
value of production. Regionally, however, there are strong differences, with fuelwood 
and charcoal being the most important commodity in sub-Saharan Africa, rice in Asia, 
beef and buffalo meat in Latin America, and wheat in the WANA region. TAC then 
modified these production values to take into account poverty by giving greater weight to 
those countries with more poverty than to other countries (see also Annex II for
61
methodology aspects). The relative importance of crops then declined from 52.5% to 
49.2% and of fisheries from 6.1% to 5.7%. The livestock and forestry sectors increase 
in importance if greater weight is given to poverty from 21.7% to 22.6% and from 
19.7% to 22.5% respectively. Further quantitative analysis shows only marginal 
differences if more strict poverty variables are incorporated therein. These data compare 
with the current allocation of CGIAR resources to production sectors of about 74% to 
crops, 18% to livestock, 5% to forestry, and 3% to fisheries. Note, however, that the 
estimate of relative share by production sector is based only on the CGIAR portfolio that 
can be allocated by commodity and therefore accounts for less than 50% of the overall 
effort.
These results would suggest that the CGIAR is relatively over-investing in crops 
and relatively under-investing in the livestock, forestry and fisheries sectors. The forestry 
sector, in particular, appears to receive, on the basis of its share in value of production, 
an inadequate share of the relative CGIAR effort. These results confirm the outcome of 
the analysis presented in Chapter 5.
6.2. Analysis by Commodity
The CGIAR commodity portfolio accounts for approximately 25 agricultural 
commodities but it is difficult to express the CGIAR’s interest in forestry/agroforestry and 
fisheries in specific commodities.
An overview of important factors to consider in assigning CGIAR priorities for 
each individual crop is provided in Annex III. As already referred to under Section 6.1, 
TAC undertook a quantitative analysis of the relative importance of commodities of 
interest to the CGIAR based on their estimated value of production in the year 2010.
Table 2b provides an estimate of value of production in the year 2010 by region and by 
commodity. Rice, fuelwood and charcoal, milk, beef and buffalo meat, wheat, maize and 
fish are the highest ranking commodities in terms of their global importance but strong 
regional differences occur when adjusted for poverty (Table 2b). Some commodities gain 
in their relative importance while others decline. Those commodities of particular 
importance to the poor, whose value increases after modification for poverty, are 
sorghum, millet, cassava, banana and plantain, pulses, groundnut, coconut, sheep and 
goats, and milk. The other commodities listed decline in relative importance. This 
relative importance further increases or decreases as the poverty variable is more 
narrowly targeted.
Table 2 compares, for crops, the current allocation of CGIAR funds with their 
share in the commodity value. While rice accounts for nearly a third of the total value, it 
receives only 24.6% of core funds. Wheat accounts for 13.5% of the value and receives 
11% of the funds, whereas maize, with 10.3% of the value and 14.4% of CGIAR funds 
appears to be relatively over-funded on the basis of value of production considerations. 
Other commodities that, on the basis of these demand considerations, appear to have a 
larger share of CGIAR funds than their share of production value are barley, sorghum, 
millet, cassava, potato and banana and plantain.
62
However, these figures must be treated with caution as they only reflect the value 
of future demand and do not take into account other considerations such as the 
opportunities for technical breakthroughs, the chances of success of research, alternative 
sources of supply, or the cost or relevance of the research. In considering future 
priorities for commodity research in the CGIAR, TAC started on the assumption that the 
current 1996 portfolio and its balance provided the basis for the future and that only 
marginal changes would be made. It was also noted that CGIAR commodities have been 
selected on the basis of their importance for the poor so there is no immediate candidate 
for elimination from the CGIAR portfolio.
TAC Members expressed their views on the future priority of commodities 
through a questionnaire that had been circulated by the Secretariat. The following 
changes in priority ranking have been suggested. Only when at least two TAC Members 
voted either positively or negatively, have these suggestions been included.
Candidates for increased priority
Rice is the most important commodity and is the major staple of most of the world’s 
poor, particularly in Asia. High returns from investment in rice have been obtained and 
according to the congruence analysis in Table 3, the CGIAR is under-investing in rice. 
The Inter-Centre Review of Rice recommended that the share of CGIAR resources 
allocated to rice research should not be reduced and also recommended a redistribution of 
its regional allocation.
Livestock - The CGIAR is currently under-investing in livestock research and the level of 
investment is well below that recommended by TAC in 1992. Furthermore, milk and 
sheep and goats are commodities of high importance to the poorest.
Forest - The CGIAR is greatly under-investing in forest commodity research.
Fish - The CGIAR is currently under-investing in fish research which is of great 
importance to the poor, particularly fishermen in poorer coastal areas.
Sweet potato - According to Table 2, the CGIAR is under-investing in sweet potato 
research. On the other hand, the Inter-Centre Review of Root and Tuber Crops Research 
noted that some responsibility currently assumed by the CGIAR could be transferred to 
national systems, particularly in Asia.
Cassava is a commodity which is particularly important for the poor, also for the urban 
poor. It is labour intensive and payoff to CGIAR investments has been high. However, 
on the basis of the congruence analysis, the CGIAR is already over-investing in this 
commodity.
Potato is a commodity of rapidly growing importance in developing countries. There is 
good potential for further productivity improvement. A major effort is needed to 
successfully combat the potato blight problem.
63
Soybean is a commodity which is rapidly gaining in importance in Africa and it is of 
importance to the poor. The IITA External Review Panel recommended that greater 
emphasis be given to soybean research. The payoff from additional research could be 
high.
Candidates for reduced priority
Pieeonpea is of importance primarily in India which accounts for 85 % of global 
production. TAC has asked ICRISAT to undertake a major review during 1997 of the 
appropriateness of further investments in pigeonpea research. The Committee should 
carefully look at the outcome of this review.
Cowpea - CGIAR investment in cowpea research is well above that which can be justified 
in the congruence analysis. The IITA External Review Panel also suggested that IITA 
carefully assess the possibility of increasing its investment in soybean research and 
correspondingly reducing its cowpea research.
Maize - The congruence analysis in Table 2 has shown that the CGIAR is relatively over- 
investing in maize research. Furthermore there are alternative sources of supply, 
particularly from the private sector.
Sweet Potato - As already indicated, the Roots and Tubers Review has suggested that 
given the overwhelming importance of sweet potato in China, where an increasing 
proportion is used for pig production, the CGIAR could gradually reduce its investment in 
Asia and hand over responsibilities to national research systems.
Wheat is a commodity of importance to the poor, also to the urban poor. A good return 
to CGIAR investment has been obtained. However, the growing strength of national 
research systems and the emergence of a strong private sector would suggest that a slight 
reduction in its priority could be considered.
Chickpea is of importance particularly in Asia and in the WANA region. The CGIAR is 
relatively over-investing in chickpea compared to its share in value of production. To 
date, little impact has been obtained from the CGIAR investments in chickpea research 
and progress in obtaining breakthroughs has been slow.
Table 1a: 2010 value of demand of CGIAR commodities by production 
sector; commodities valued in 1989/91 international $
Production sectors (% of total)
Region Crops Livestock Forestry Fisheries Total
Sub-Saharan Africa 6.3% 2.7% 5.6% 0.6% 15.2%
Asia 34.4% 9.3% 10.0% 4.7% 58.3%
C. & S. America 7.7% 6.5% 3.8% 0.5% 18.4%
WANA 4.2% 3.1% 0.4% 0.3% 8.0%
Total 52.5% 21.7% 19.7% 6.1% 100.0%
4>
Regional share (% of sector)
Region Crops Livestock Forestry Fisheries
Sub-Saharan Africa 12.0% 12.5% 28.3% 10.3%
Asia 65.5% 42.9% 50.5% 77.4%
C. & S. America 14.6% 30.1% 19.1% 8.0%
WANA 8.0% 14.5% 2.1% 4.2%
Total 100.0% 100.0% 100.0% 100.0%
{
Table 1b: 2010 value of demand of CGIAR commodities: values in 1000 int’l $ and shares by commodity
Regional values and shares of commodity
Sub-Saharan Asia Latin America and West Asia and Overall value Share of 
Commodity Africa Share Share Caribbean Share North Africa Share of commodity grand total
Rice 5536690 4.1% 119371469 88.8% 6371498 4.7% 3080386 2.3% 134360043 17.5%
Wheat 2097288 3.3% 40257850 63.7% 5744851 9.1% 15078283 23.9% 63178272 8.2%
Maize 5275084 11.0% 27767295 57.8% 11579343 24.1% 3438061 7.2% 48059784 6.3%
Barley 347939 5.2% 846017 12.7% 599640 9.0% 4872428 73.1% 6666025 0.9%
Sorghum 3302430 38.1% 2352206 27.2% 2681450 31.0% 326331 3.8% 8662417 1.1%
Millet 3112521 56.4% 2335319 42.3% 28788 0.5% 41791 0.8% 5518419 0.7%
Cassava 9094680 63.5% 2206144 15.4% 3028944 21.1% 88 0.0% 14329858 1.9%
Potato 525052 4.2% 7863944 63.1% 2101902 16.9% 1977514 15.9% 12468412 1.6% cr*
Sweet potato and yam 4628121 29.7% 10600972 68.0% 350252 2.2% 8716 0.1% 15588061 2.0% Ui
Sweet potato 547636 4.8% 10559409 93.3% 198672 1.8% 8716 0.1% 11314433 1.5%
Yam 4080486 95.5% 41563 1.0% 151580 3.5% 0 0.0% 4273628 0.6%
Banana 1608592 16.0% 5162804 51.4% 3031567 30.2% 234788 2.3% 10037751 1.3%
Plantain 3110477 78.7% 90394 2.3% 750298 19.0% 0 0.0% 3951168 0.5%
Pulses 5613884 19.1% 16898684 57.5% 4119553 14.0% 2756104 9.4% 29388226 3.8%
Beans 2784840 21.2% 6248441 47.6% 3732009 28.4% 357300 2.7% 13122590 1.7%
Chickpea 412569 5.0% 6167893 74.5% 244111 2.9% 1455887 17.6% 8280461 1.1%
Pigeon pea 397834 12.9% 2656594 85.8% 41375 1.3% 0 0.0% 3095803 0.4%
Lentil 55991 3.0% 980001 53.3% 45512 2.5% 757595 41.2% 1839100 0.2%
Broad beans 218072 17.1% 829347 65.2% 42754 3.4% 182357 14.3% 1272529 0.2%
Cowpea 1744577 98.1% 16408 0.9% 13792 0.8% 2965 0.2% 1777742 0.2%
Soyabean 235130 0.9% 7510748 30.1% 17000509 68.2% 184002 0.7% 24930389 3.3%
Groundnut 3344029 16.7% 15614744 78.0% 897539 4.5% 171848 0.9% 20028160 2.6%
Coc onut 326142 6.2% 4566089 86.2% 403239 7.6% 0 0.0% 5295470 0.7%
Beef and buffalo meat 10023010 13.9% 24384940 33.7% 30367315 42.0% 7565729 10.5% 72340993 9.4%
Sheep and goat meat 3887348 18.8% 8906057 43.1% 1492599 7.2% 6374033 30.9% 20660037 2.7%
Total milk 6926299 9.5% 38033997 51.9% 18112646 24.7% 10198311 13.9% 73271254 9.6%
Fuelwood and charcoal 39762311 33.7% 56381380 47.7% 19897240 16.8% 2103035 1.8% 118143966
Roundwood 15.4%2964668 9.0% 19879322 60.5% 8944718 27.2% 1067579 3.2% 32856287 4.3%
Fish food 4815590 10.3% 36080074 77.4% 3741864 8.0% 1967100 4.2% 46604627 6.1%
Grand total by region 116537286 447110448 141245755 61446128 766339617 100.0%
Regional share of grand total 15.2% 58.3% 18.4% 8.0% 100.0%
Table 2a: 2010 value of demand of CGIAR commodities by 
production sector, weighted by the modified welfare indicator
(income threshold value 9000 PPP $, exponent set as 2)
Production sectors (% of total)
Region Crops Livestock Forestry Fisheries Total
Sub-Saharan Africa 10.7% 4.6% 9.4% 1.1% 25.7%
Asia 27.6% 10.2% 8.8% 4.0% 50.6%
C. & S. America 7.5% 5.3% 4.0% 0.5% 17.2%
WANA 3.4% 2.5% 0.4% 0.2% 6.5%
Total 49.2% 22.6% 22.5% 5.7% 100.0%
Regional share (% of sector)
Region Crops Livestock Forestry Fisheries
Sub-Saharan Africa 21.8% 20.2% 41.6% 18.5%
Asia 56.1% 45.2% 39.2% 69.9%
C. & S. America 15.2% 23.3% 17.6% 7.9%
WANA 6.9% 11.2% 1.6% 3.7%
Total 100.0% 100.0% 100.0% 100.0%
Formula: VOD * (1 - ((1-Gini coefficient)*Income 2003)/9000) ~ 2
Table 2b: 2010 value of demand of CGlAR commodities weighted by the modified welfare indicator
(income threshold 9000 PPP $, exponent set as 2)
Regional values and shares of commodity
•Sub-Saharan Asia Latin America and West Asia and Overall value Share of 
Commodity Africa Share Share Caribbean Share North Africa Share of commodity grand total
Rice 4708715 7.6% 52468947 84.9% 3445657 5.6% 1172201 1.9% 61795520 
Wheat 15.8%1791056 6.9% 15272602 58.8% 2457037 9.5% 6437137 24.8% 25957832 6.6%
Maize 5245360 26.5% 6138786 31.0% 6614470 33.4% 1793637 9.1% 19792253 5.1%
Barley 309564 13.3% 270648 11.6% 258472 11.1% 1495443 64.1% 2334128 0.6%
Sorghum 2847467 52.8% 1472576 27.3% 886960 16.4% 189814 3.5% 5396817 1.4%
Millet 2662424 62.6% 1555708 36.6% 7545 0.2% 24975 0.6% 4250653 1.1%
Cassava 7773848 72.9% 1077705 10.1% 1805777 16.9% 49 0.0% 10657380 2.7%
Potato 458058 8.9% 2932794 57.3% 903435 17.6% 828327 16.2% 5122614 1.3%
Sweet potato and yam 3814821 69.0% 1502790 27.2% 203267 3.7% 4282 0.1% 5525160 1.4%
Sweet potato 451399 21.8% 1496898 72.4% 115298 5.6% 4282 0.2% 2067878 0.5%
Yam 3363421 97.3% 5892 0.2% 87968 2.5% 0 0.0% 3457282 0.9%
Banana 1390477 23.8% 2873076 49.1% 1496793 25.6% 87399 1.5% 5847746 1.5%
Plantain 2635696 86.4% 45602 1.5% 370949 12.2% 0 0.0% 3052248 0.8% O'
Pulses 4856807 26.3% 10334089 55.9% 2126655 11.5% 1177081 6.4% 18494632 4.7%
Beans 2409282 29.0% 3821123 46.0% 1926591 23.2% 152596.3 1.8% 8309592 2.1%
Chickpea 356931 7.3% 3771865 77.3% 126018.7 2.6% 621782.5 12.8% 4876597 1.2%
Pigeonpea 344183 17.3% 1624593 81.6% 21359.1 1.1% 0 0.0% 1990135 0.5%
Lentil 48441 4.9% 599302 60.2% 23495 2.4% 323555 32.5% 994793 0.3%
Broad beans 188663 23.7% 507172 63.7% 22071 2.8% 77881 9.8% 795788 0.2%
Cowpea 1509307 98.8% 10034 0.7% 7120 0.5% 1266 0.1% 1527727 0.4%
Soyabean 194470 1.9% 1841788 18.1% 8096180 79.4% 62910 0.6% 10195349 2.6%
Groundnut 2868138 26.8% 7389260 69.1% 363815 3.4% 70925 0.7% 10692138 2.7%
Coconut 274289 9.0% 2579933 84.8% 188983 6.2% 0 0.0% 3043205 0.8%
Beef and buffalo meat 8522888 24.9% 10429568 30.5% 12146542 35.5% 3146953 9.2% 34245951 8.8%
Sheep and goat meat 3345380 31.5% 4002707 37.7% 630936 5.9% 2642769 24.9% 10621792 2.7%
Total milk 5978934 13.8% 25530003 58.7% 7827801 18.0% 4135895 9.5% 43472633 11.1% 
Fuelwood and charcoal 34155303 44.8% 29603040 38.9% 11383137 14.9% 1045313 1.4% 76186793 19.5%
Roundwood 2435470 20.5% 4879431 41.1% 4157042 35.0% 400464 3.4% 11872407 3.0%
Fish food 4131776 18.5% 15658752 69.9% 1766160 7.9% 832083 3.7% 22388771 5.7%
Grand total by region 100400939 197859809 67137614 25547659 390946022 100.0%
Regional share of grand total 25.7% 50.6% 17.2% 6.5% 100.0%
Formula: VOD * (1 -((1 -Gini coefficient)* Income 2003)/9000) ~ 2
Table 3 : Allocation of research funds to CGIAR crops 
compared to modified values after quantitative analysis
Commodity Commodity- Share Expenditure Share
value, modified in 1996 
mill US $
Rice 61795520 32.2% 21.8 24.3% 
Wheat 25957832 13.5% 9.7 10.8% 
Maize 19792253 10.3% 12.7 14.1% 
Barley 2334128 1.2% 3.06 3.4% 
sorghum 5396817 2.8% 3.6 4.0% 
Millet 4250653 2.2% 2.6 2.9% 
Cassava 10657380 5.5% 7.6 8.5% 
Potato 5122614 2.7% 3.9 4.3% 
Sweet potato and yam 5525160 2.9% 2.5 2.8% 00
Banana and plantain 8899994 4.6% 7.2 8.0% 
Beans 8263559 4.3% 4.7 5.2% 
Chickpea 4722034 2.5% 2.2 2.4% 
Pigeonpea 1967514 1.0% 1.4 1.6% 
Lentil** 1180508 0.6% 0.8 0.9%
Broad (faba) bean 787006 0.4% 
Cowpea* 1574011 0.8% 1.0 1.1% 
Pulses, CGIAR crops * ★ ★ 18494632 9.6% 10.1 11.3% 
Soyabean* 10195349 5.3% 1.7 1.9% 
Groundnut 10692138 5.6% 3.4 3.8%
Coconut 3043205 1.6% n.a. n.a.
Grand total 192157675 100.0% 89.89 100.0%
Note: * IITA allocation for soyabean and cowpea split according 
to ratio in PS-report 1992, page 161 
** CGIAR expenditure for lentils and faba bean can't be split
★ ★ ★ Pulses split according to regional values 1992/44 production data
69
CHAPTER 7 - SYSTEMWIDE PROGRAMMES AND THE 
ECOREGIONAL APPROACH
7.1 Introduction
An important innovation in the plan for 1994-1998 was the recommendation that 
donors provide funds for programmes that transcend the interests of individual centres. 
This Chapter lists the Systemwide activities that are being developed using this new 
approach to the implementation of CGIAR priorities, and makes recommendations about 
their management by the System during the period 1998-2000.
7.2 Background
The previous review of CGIAR priorities and strategies (TAC Secretariat, 1994) 
proceeded in three stages: the determination of (a) what to do in relative terms 
(priorities), (b) the appropriate strategy to follow in implementing those priorities 
(strategies), and (c) who should do the research and how it should be organized 
(structure).
TAC’s proposals for strategies and structure were first discussed by the Group at 
MTM’92. In line with the vision presented in the expansion report (TAC Secretariat, 
1992), TAC proposed a future CGIAR System undertaking two separate but 
complementary research activities:
global, concentrating on strategic research on an agreed slate of 
commodities and subjects, and
ecoregional, covering strategic and applied research on natural resource 
conservation and management, production systems and location-specific 
aspects of commodity improvement.
While it was felt at MTM’92 that the Group had a long way to go before 
reaching finality on some of the issues, "the ecoregional concept was overwhelmingly 
endorsed, with both TAC and CGIAR Centres being encouraged to move from concepts 
to operations". (Extract from Summary of Proceedings and Decisions, CGIAR Mid-Term 
Meeting 1992, Istanbul, Turkey.)
TAC’s recommendations for medium-term resource allocations 1994-1998 
(Review and Approval of Centre Medium-Term Plans 1994-98, AGR/TACTAR/93/11) 
and its revision of Chapter 13 of the priorities and strategies document (TAC Revision of 
Chapter 13, AGR/TACTAR/93/18, Part II, Rev.l), were presented to the Group at 
ICW’93. In his covering letter to the Chair of the CGIAR, the TAC Chair noted that 
TAC had made two kinds of recommendations on funding in the CGIAR for the period 
1994-1998: centre-specific funding and programme funding for particular CGIAR 
Systemwide initiatives. He noted also the innovative but experimental nature of such
70
programme funding, though TAC believed that it would provide an attractive mechanism 
to promote partnerships among centres, national programmes and other actors in the 
global agricultural research system. TAC currently believes it is necessary to take stock 
of Systemwide activities before adding to the existing set of initiatives and programmes. 
The Committee’s position in this regard is outlined in the final section of this Chapter 
(pp. 12-13).
The justification for programme funding of Systemwide initiatives was further 
elaborated in the document on medium-term resource allocation. Under the heading 
"Systemwide Considerations", TAC recognized that the process of developing MTP 
proposals at centre level was limited in its ability to deal fully with concerns of particular 
importance at the System level transcending the centres’ own interests. There were also 
inconsistencies in the centres’ perceptions of the relative priority of inter-centre 
undertakings as a part of their normal activities.
In that document, TAC went on to recommend funding by 1998 of seven 
ecoregional programmes and a cross-ecoregional programme on alternatives to slash-and- 
burn agriculture. It also proposed inter-centre programmes on conservation of genetic 
resources, on livestock research (to facilitate the establishment of new livestock research 
programmes of both a global and ecoregional nature and specifically to develop, with 
plant-oriented centres, integrated programmes on livestock feed and production systems), 
and on water management research (to study the efficiency with which increasingly scarce 
irrigation water is used for crop production along with resource degradation issues such 
as salinization and waterlogging. The total recommended was US$ 10 million out of 
US$ 270 million, or 3.7%.
7.2.1. Terminology
The terms used above and in the next section were taken directly from the TAC 
documents that are quoted as references. However, the terminology has evolved 
considerably with time. At TAC 66 in March 1995, a note was prepared to help clarify 
use of the terms "Systemwide", "initiative" and "programme". This note ("TAC Note on 
Systemwide and Ecoregional Concepts," Report of the Sixty-Sixth Meeting of TAC, June 
1995) drew heavily on, but did not follow in all respects, the conclusions from a meeting 
on "The Management of Systemwide Programmes and Ecoregional Initiatives" convened 
in December 1994 by Centre Directors.
"Systemwide" was defined in the note as referring to inter-centre research or 
research-related activities, on a regional or global basis, or some combination thereof. It 
was recognized that such activities would generally involve organizations outside the 
CGIAR, for example, partnerships with NARS for the implementation of the ecoregional 
approach. "Initiative" was defined as the start-up or design phase of an activity, and 
"programme" as work that was under way.
Although the question has not been addressed by TAC previously, Systemwide 
activities (SWAs), initiatives or programmes, seem to differ from ongoing inter-centre 
collaborations in three main aspects:
71
the entering into by centres of formal commitments for SWAs, in 
contrast to the more flexible collaborative arrangements made by 
individual scientists and programme leaders;
the separate accountability to the Group required of SWAs; and
the practice that has been adopted of showing the costs of SWAs in the 
CGIAR’s annual financing plan, whereas those of regular inter-centre 
collaboration are usually not identified separately.
There has been some inconsistency in deciding what those figures for the costs of 
SWAs should cover. In some cases, funding has been recommended only for facilitation 
of the start-up phase. In other cases, it has also included seed money for new research 
and research-related activities, such as the setting up of databases. Guidelines need to be 
developed that clearly define the conditions for special support of such research and 
related activities in future SWAs. Such guidelines would best be premised on an 
assessment of experiences with SWAs to date. TAC’s proposed approach to this larger 
task is outlined in the penultimate section of this Chapter (pp. 10-12). In the interim, 
since TAC envisages that it would recommend for future approval only new initiatives for 
which there are strong arguments, this issue may be dealt with on a case by case basis.
The terminology from TAC 66 is adopted in the later sections of this Chapter for 
discussing the future management of Systemwide programmes and the ecoregional 
approach. However, ambiguity remains in relation to the boundaries of Systemwide 
programmes (SWPs). At one extreme, only the facilitation and any new research 
components are included. At the other extreme, all research related to the SWA both 
inside and outside the System is included. It is, therefore, important when discussing 
SWPs to specify in each case whether the SWP refers only to facilitation and new 
research, all related CGIAR work, or all related work inside and outside the CGIAR.
7.3. An Overview of CGIAR Experience in the Initiation and 
Implementation of Systemwide Programmes and the Ecoregional 
Approach
The adoption of an agreed agenda for the System’s research has been one of the 
most important changes that has occurred since TAC proposed programme funding in 
1993. The financing plan for the 1996 CGIAR Research Agenda provided for six SWPs: 
genetic resources, rice/wheat, Latin American ecoregion, alternatives to slash-and-burn, 
mountain agriculture, and on-farm water husbandry in West Asia and North Africa. 
Provision was also made for financing the start-up of a global livestock programme 
during 1996. In addition, funds were endorsed for the design phases of a global water 
management programme; an initiative on soil, water and nutrient management; an inter­
centre initiative on coastal zone management; an ecoregional initiative on desert margins; 
initiatives on research indicators and property rights/collective action; an initiative on 
integrated pest management; and ecoregional approaches to the humid and sub-humid 
tropics/inland valleys of sub-Saharan Africa and to the humid/sub-humid tropics and 
subtropics of Asia.
72
Thus, of the seven ecoregional programmes proposed by TAC at ICW’93, in 
1996 four were still entirely or significantly involved in their design phase, two 
(rice/wheat; an ecoregional approach to enhancing agricultural research in Latin 
America) were in the process of implementation as SWPs, and the last (an ecoregional 
programme for the east and central African highlands) had been subsumed in the cross- 
ecoregional SWP for sustainable mountain agricultural development. Of the other SWPs 
proposed by TAC in 1993, the cross-ecoregional programme on alternatives to slash-and- 
burn agriculture (ASB) was under way in 1996, as also were the inter-centre programme 
on genetic resources and the global livestock programme. However, the proposed global 
water management programme was still at the design stage, as already noted.
Meanwhile, the following initiatives had been added: a cross-ecoregional SWP 
for sustainable mountain agricultural development (SMAD) subsuming the east and central 
African highlands initiative, an inter-centre initiative on coastal environments, and 
initiatives on research indicators, property rights and collective action, integrated pest 
management (IPM), and soil, water and nutrient management (SWMN).
The last of these (SWNM) was the outcome of a succession of well-documented 
deliberations beginning with IBSRAM’s Position Paper on Soil, Water and Nutrient 
Management research - a New Agenda (the Greenland Report), and advancing through the 
Zschortau Plan and the Feldafing Consultation. Checking the list of eight topics proposed 
at MTM’94 by the CGIAR Task Force as follow-up to UNCED/Agenda 21, six were 
addressed specifically by SWPs and initiatives, the seventh (a global long-term forestry 
research network) was considered to be a central part of CIFOR’s core programme, while 
the eighth (global digital data sets for use in geographic information systems (GIS) was 
not yet covered though there was some use of GIS by ecoregional programmes, such as 
the one in Latin America.
Although no systematic evaluation has been done, the establishment of SWAs 
seems to impose heavy transaction costs on centres and adds significantly to TAC’s 
workload. Moreover, review processes have still to be implemented that will expose 
SWAs to the same rigorous scrutiny that is applied to centre-based activities.
The following notes summarize the status of the six SWPs that were endorsed for 
financing in 1996:
Genetic Resources: The proposal for a Systemwide genetic resources programme 
emanated from the Stripe Study of Genetic Resources in the CGIAR and was accepted at 
MTM’94. The budget for 1996 was US$ 1.6 million, of which US$ 0.95 million was for 
collaborative research activities, US$ 0.19 million was for public awareness and an 
information network, and the remainder was for coordination and administration.
Rice-Wheat: The rice/wheat consortium for the Indo-Gangetic Plain builds on the rice- 
wheat initiative established in 1990 with funding by ADB. It addresses the problem that 
the 12 million ha of rice-wheat cropping systems on the Plains, which were central to the 
Green Revolution there, are running into sustainability problems. The CGIAR budget for 
1996 of US$ 1 million covered US$ 0.2 million to ICRISAT as convening centre and the 
remainder for research activities by the three centres involved (CIMMYT, IRRI,
73
ICRISAT). The financial commitment by the four regional NARS significantly exceeded 
that of the CGIAR.
Latin America Ecoregion: Implementation of this SWP commenced during 1996, 
though some design work had still to be completed. The 1996-97 workplan (note the 
change of timebase) allocated most of the resources to implementation of the 
programme’s research and training activities; only 5-6% of the budget was for 
coordination. The 1996 Research Agenda budgeted US$ 5.7 million for this SWP, 
US$ 0.9 million to initiate the programme and US$ 4.8 million for centre activities by 
CIAT and CIMMYT, mainly CIAT (US$ 4 million).
Alternatives to Slash-and-Burn: This consortium was formed in February 1992, after a 
year of preparation, and was probably the earliest of the SWPs in which the CGIAR has 
had a significant involvement. The plan was for a concerted attack on a global scale to 
accelerate the development of sustainable land uses to replace forms of slash-and-burn 
agriculture that cause deforestation and environmental degradation. The programme was 
initiated by a consortium of 16 member agencies, including four CGIAR Centres and nine 
NARS. Seed money was provided by UNDP and UNEP in 1992 and 1993, then GEF 
funded two phases of the programme from March 1994 to May 1996. At the time of 
writing, a further phase of GEF funding was under discussion. The CGIAR budget for 
1996 provided US$ 6.8 million for this SWP, US$ 0.5 million to support coordination 
and US$ 6.3 million of centre-research activities, predominantly by ICRAF.
Mountain Agriculture: This SWP is a combination of CIP’s work in the Consortium for 
the Sustainable Development of the Andean Ecoregion (CONDESAN), the East African 
highlands initiative, and the related longer-standing work of ICIMOD in the Himalayas. 
CONDESAN was effectively convened in 1994, following a workshop in 1992. The East 
Africa highlands initiative, convened by ICRAF, also includes other CGIAR Centres and 
ASARECA, the regional NARS organization.
Global Livestock Programme: ILRI has developed an inter-centre initiative using an 
innovative process of competitive bidding for resources. From 1996, for a period of 
three years, three ecoregionally-based projects for the improvement of the production and 
utilization of livestock feeds will be supported in tropical Latin America, in WANA, and 
in the East African highlands, respectively. The budget for 1996 was US$ 2.8 million, 
with US$ 0.50 million for convening costs including a planning workshop, and the 
remainder as seed money for strategic and applied research.
The expenditures budgeted for programmes and programme design in 1996 can 
be compared with those recommended by TAC in 1993. For the comparable categories, 
expenditure in 1996 is estimated to be US$ 12.1 million, which is 4.0% of the CGIAR 
total of US$ 299.2 million. This is a slight increase on the 3.7% proposed in 1993, and 
entirely in design costs. The three SWPs proposed in 1993 - genetic resources, livestock, 
and water management - were projected to cost US$ 6 million in 1998 (in 1992 dollars). 
In the Financing Plan for 1996, genetic resources was allocated US$ 1.6 million, 
livestock US$ 2.8 million (expected to rise to US$ 4.0 million in 1997) water 
management (still in the design phase) US$ 1 million, and on-farm water husbandry 
US$ 0.6 million.
74
The US$ 12.1 million quoted above is only for the costs of facilitation and 
seeding new research. Figures were provided in the 1996 Agenda for the costs of related 
centre research in four SWPs - rice/wheat, ASB, Latin American Ecoregional, and 
SMAD. The costs of related centre research (US$ 15.75 million) far exceeded those of 
facilitation and seed money (US$ 1.75 million). Another SWP, the one for Hillsides in 
Latin America, does not appear at all in the 1996 agreed research agenda budget paper, 
because no special funding was requested or recommended. The costs of facilitation were 
being borne entirely by Cl AT and CIMMYT from centre-specific sources.
7.4. Analysis and Recommendations
7.4.1. A Classification of the CGIAR’s Systemwide Activities
The SWAs that have been undertaken up to and including the 1996 Research 
Agenda can be classified into two broad categories:
those undertaken to implement the ecoregional approach; and 
those undertaken to strengthen more specific aspects of the System’s 
work.
None of those in the first 
A CGIAR research activity may be group (Table 1) appear to address, in 
characterized as ecoregional if it meets the following the RAEZs in which they are 
general criteria, namely, if it: operating, all of the very 
(0 is research on the technical and human comprehensive range of objectives 
dimensions of problems in the sustainable described for the ecoregional 
improvement of productivity; approach (see box). Some of those 
SWAs in Table 1 are quite selective. 
(ii) addresses landscape units in the For instance the rice-wheat SWP 
agroecosystem of a priority agroecoregional addresses one farming system only, 
zone; albeit a very important one, in the 
(iii) has effective and clearly identifiable warm semi-arid tropics and 
partnership linkages with national research subtropics of Asia. Likewise, the 
systems and other research agencies of the alternatives to slash-and-bum SWP is 
region, and shows the complementarity of limited to the farming practices of 
function across the partners; (formerly) forested areas in the warm 
(iv) has close linkages with global strategic 
humid and sub-humid tropics of the 
commodity/subject matter research activities. world. Nevertheless, they all deal 
with the implementation of some part 
of the ecoregional approach. The 
(Gryseels & Kassam, 1994) initiative on coastal environments
_________________________________________  identifies a different type of
agroecological zone to those
described in TAC 1994, one that is relevant to fisheries research, as well as to studies of
other downstream effects of agriculture and forestry.
75
What distinguishes the second group (Table 2) is that they are much more 
targeted than the first and all are actually or potentially global. Their primary purpose is 
to fill gaps in the System’s work or to improve its technical quality and cost-effectiveness 
or both. The livestock SWP is somewhat different in that it arose out of a restructuring 
of livestock research in the System.
Further examination of plans for the second group reveals that, with the possible 
exception of the genetic resources SWP, they will be implemented predominantly in a 
regional context. This raises the question, which TAC intends to pursue, as to whether 
sufficient emphasis is being placed in this type of SWA on the CGIAR’s global role, that 
is on strategic research to develop new generic understanding and methodologies that can 
be used internationally.
The common feature of these diverse SWAs in Tables 1 and 2 is that they all 
appear to deal with the expediting of change in the System. Thus, adoption of the 
ecoregional approach represents a very different mode of operation for the CGIAR, and 
the SWAs in Table 2 are aimed at significant strengthening of research on important 
global issues (genetic resources, natural resource management) or of effectiveness in key 
subject matter areas (integrated pest management, property rights and collective action).
The conclusion that SWAs are concerned with the management of change in the 
CGIAR has a very important implication. These special mechanisms that have been set 
up to expedite change do not need to continue indefinitely. As soon as a change has been 
effected, it becomes part of the mainstream operations of the System. This is reflected in 
TAC’s proposals for the future management of SWAs.
7.4.2. TAC’s Proposals for the Management of Systemwide Activities
TAC proposes different forms of management for the two groups of SWAs 
during the period 1998-2000. In the case of implementing the ecoregional approach, 
there is an urgent need to monitor the progress of the ten SWAs listed in Table 1, and to 
document the lessons emerging from that experience. Before the end of this planning 
period, a good deal of experience will have been gained of the design phase (SWIs) of 
regional and cross-regional activities using the ecoregional approach, and for at least 
some programmes there should be clear indications of whether or not the approach is 
achieving its objectives in practice.
To oversee the evaluation and documentation of SWAs implementing the 
ecoregional approach, TAC will set up a Standing Committee, which will draw on the 
special expertise of CGIAR Centres, NARS, and the donor community. One of its 
specific charges will be to report to the Group before the end of the 1998-2000 period on 
its interim assessment of the value of the ecoregional approach.
This is necessary in order for the System to have early warning of any major 
deficiencies in the approach. The difficult history of efforts to organize large 
multidisciplinary, multi-institutional agricultural research programmes, for example in 
farming systems research which is simpler than the ecoregional approach, indicates a need
76
for a realistic and cautious attitude by the Group. A specific output of this exercise 
would be the development of guidelines to specify the criteria for recommending support 
to the initiative and programme phases of such activities.
The Standing Committee would draw on the valuable work done by the CGIAR 
Task Forces on Ecoregional Approaches to Research (noting that true partnership between 
the CGIAR and NARS goes far beyond the ecoregional approach) and on Sustainable 
Agriculture. Other relevant experience should be taken into account, such as that from 
farming systems research (Meindertsma, 1995) and from the experience of a number of 
industrialized countries that have set up multi-institutional centres for research in fields 
such as engineering.
The establishment of such a Standing Committee does not need to wait until 
1998. The report of the consultancy currently being commissioned by the centre directors 
to document and analyse experiences with the development and implementation of 
ecoregional SWPs would be a very valuable input for that Committee.
A different approach is proposed for the management of the second group of 
SWAs (Table 2) during the period 1998-2000. Because of their wide diversity, they each 
require separate treatment. TAC recommends that they be reviewed as part of the 
System’s normal review process. Each EPMR, and associated internally commissioned 
reviews, should examine the centre’s convening role, if it has one, and its components of 
SWAs. A separate decision will be needed on whether or not to carry out a stripe review 
or a strategic study of the research area in question. In some cases, the CGIAR may 
need to join with outside agencies in an even more comprehensive evaluation.
Two key questions need to be answered in any review of SWAs from either
group:
Has the value added by making the activity Systemwide, rather than 
leaving it as a series of centre-based components, outweighed the 
additional transaction and management costs? and
Has the SWA reached the point where its separate existence is no 
longer warranted? It may have done its job and no further special 
action is needed, or it may be possible to provide the collective action 
still needed through a centre, with appropriate incentives and 
accountabilities. Alternatively, the SWA may have failed, or be very 
likely to fail, and should be abandoned or substantially restructured.
The reviews of SWAs should also give particular attention to the effectiveness of 
their operational modes. In the Study of Priorities and Strategies for Soil and Water 
Aspects of Natural Resources Management Research in the CGIAR, TAC has proposed 
seven criteria for evaluating the operating modes used in SWAs. These should be tested 
in practice.
There is one area of CGIAR research that, in TAC’s view, requires further 
examination. It is the centre-based work in activity category 1.2, production systems
77
development and management, and category 2, protecting the environment, that is related 
to the SWI on soil, water and nutrient management. It is not easy in practice to 
distinguish between productivity-enhancing and resource-conserving research on natural 
resources such as soil and water. It needs a more detailed examination than was possible 
during TAC’s study on priorities and strategies for soil and water aspects of natural 
resources management research in the CGIAR.
7.4.3. Will New Systemwide Activities be Required 
During the Period 1998-2000?
TAC recommends that, in the absence of strong new arguments, no new SWAs 
for the implementation of the ecoregional approach be added until (a) the experience to 
date has been gathered, assessed and documented properly, and (b) the TAC Standing 
Committee has been able to present to the Group its interim assessment of the value of 
the ecoregional approach. While the ten existing initiatives/programmes (Table 1) are not 
comprehensive, they cover sufficient objectives of the approach to allow a full interim 
evaluation.
The situation with the second group of SWAs is different. The Group may wish 
to strengthen additional specific areas of work in 1998-2000, in order to remain flexible 
and responsive in the pursuit of its goals. For instance, research on postharvest 
processing, which has been one of the System’s objectives since the mid-1980s, may 
require strengthening in the light of the report of the current stripe review.
However, because experience indicates that the initiation and implementation of 
SWAs adds significantly to the managerial workload of centres and their partners, TAC 
counsels against adding many more SWAs of any kind during the 1998-2000 period. It 
would be better to proceed slowly until the implementation of the ecoregional approach is 
under control and some of the SWAs in the second group have achieved their main 
purposes.
Also this would enable the System to undertake a special priority-setting exercise 
for future SWAS. Many of the topics listed in Table 2 were first identified as priorities 
from outside the System. In future, the CGIAR with the assistance of TAC, needs to 
re-examine the agreed research agenda to see which areas require special action to raise 
them to the highest possible level of performance.
The Group may still wish to see separate reports on areas of research covered by 
former SWAs, even when they appear in centre plans and budgets rather than in TAC 
documents. There will probably be a desire also to make sure that investment in research 
and research-related activities are SWPs and the ecoregional approach is properly 
measured in activity category 5 (some redefinition is likely to be needed), and that the 
Systemwide overhead costs of convening and facilitating SWAs are known and assessed.
78
Table 1: A List of Systemwide Initiatives and Programmes 
Undertaken by the CGIAR to Implement the 
Ecoregional Approach
Regional Initiatives for:
Sustainable natural resource management options to arrest land 
degradation in the desert margins of sub-Saharan Africa
The warm humid and sub-humid tropics of sub-Saharan Africa
The humid and sub-humid tropics of Asia
On-farm water husbandry in West Africa and North Asia
Central American and Andean hillsides
Coastal environments
Regional Programmes for:
Sustainable rice-wheat based cropping systems in the Indo-Gangetic 
Plain
Enhancing agricultural research effectiveness in Tropical America
Cross-Regional Programmes for:
Alternatives to slash-and-burn agriculture
Sustainable mountain agricultural development
79
Table 2: A List of Systemwide Initiatives and Programmes 
Undertaken by the CGIAR to Strengthen Specific Areas 
of CGIAR Research
Initiatives in:
Water management
Agricultural indicators
Property rights and collective action
Soil, water and nutrient management
Integrated pest management
Programmes in:
Genetic resources
Livestock research
80
CHAPTER 8 - RECOMMENDATIONS ON FUTURE PRIORITIES
AND STRATEGIES
8.1. By Activity
(Only section drafted yet is from Sir Ralph Riley and Dr. Lucia Vaccaro on sustainable production systems)
The previous classification of some of the CGIAR research under "Protecting the 
Environment" served initially to emphasize the concern for sustainability and monitor the 
associated research. Now, however, the notion and objective of sustainability has become 
all pervasive in the production systems research of the CGIAR. Similarly, research 
concerned with protecting the environment must be involved with sustainable agricultural 
production. Consequently TAC considers that there is no longer any justification for 
maintaining the category "Protecting the Environment", since it seems to imply commitments 
to a kind of ecological research which is independent of agriculture and this conflicts with 
the purposes for which the CGIAR exists. It also leads to an underestimate of the true 
magnitude of the work that the CGIAR conducts on the sustainability aspects of production 
systems. These considerations have led TAC to propose to the Group that there should be 
research on sustainable production systems defined under four headings, namely crop-based, 
livestock-based, tree-based and aquatic biology-based systems. In drawing attention to these 
systems TAC nevertheless points out the prevalence of numerous mixed systems which have 
particular relevance to sustainability and in which multidisciplinary research is often 
necessary. The main categories of production systems categorized above can be 
disaggregated as follows:
Crop-based systems research should be considered under the following categories (which 
are not exhaustive). The estimated areas occupied by each system is as indicated (area to be
inserted). (14% of resource)
a. crop-based rainfed systems
b. mixed crop and livestock rainfed systems
c. arable irrigated systems
d. mixed arable crops and livestock irrigated systems
Livestock-based farming systems research must take into account several distinctive farming
systems: (12.2% of resource)
a. Dry range-land, principally with cattle
b. Peri-urban cattle, primarily for milk production
c. Mixed sheep, goats and cattle, especially on desert and steppe fringes
d. Village cattle, from which milk, meat and traction are all demanded.
81
Tree-based systems research with particular attention to the following:
(3% of resource)
a. Agroforestry which deploys crops within the forest garden, with concern for the 
uses of wood for fuel, building materials and farm uses such as stakes and fencing 
b. Forestry, when trees and tree products constitute the harvest
Aquatic biological production systems with concentration on the following:
(3% of resource)
a. Aquaculture, sustainable fish farming
b. Inland water, capture fishing and its sustainability
c. Maritime ponds for farming marine fish, arthropods and molluscs
d. Capture marine fisheries
• Generalized components of production research
There are several components of sustainable agricultural production which are more 
effectively addressed by research in a generic manner rather than in relation to a specific 
production system. These are listed below together with brief comments.
Water - About 80% of the water used by the world’s people is deployed in agriculture. 
Clearly it is one of the most crucial of primary resources needed for agriculture. The 
management of water must be improved in its on-farm use, in the command area in irrigation 
systems and within major catchments. Moreover, TAC takes the view that the provision of 
potable domestic water is as important in alleviating poverty as is the pro0§fonfotfe&Mrixe)
Soils - In any crop production system a thorough understanding from research is necessary 
concerning soil physics, chemistry and biology. Only from such knowledge can the 
sustainability of any system be determined. Combined generic research on water (2.2.1) and 
soils (2.2.2) will be necessary especially where erosion by water is important. Knowledge 
from soil physics will help to alleviate the effects of wind erosion. (3% of resource)
Slopes - Much agriculture, in many kinds of cropping systems, is conducted on land at 
angles of slope which make soil erosion a serious risk. Soil preparation, cultivation and crop 
harvesting present serious problems. Generic research on agriculture on slopes may find
wide-scale application. (1% of resource)
Coastal research - About xxx million of people in developing countries live within xx miles 
of the sea shoreline and at less than xx feet above sea level. TAC considers that the 
problems that such people face can only be addressed in a specialized way. Research is 
consequently necessary on the interaction between agricultural production on the shore and 
hinterland and marine production from the adjacent sea into which the land water drains. 
This requires simultaneous multidisciplinary research on coastal production systems whether 
from agriculture or from fisheries. (0.5% of resource)
82
• Special multidisciplinary subjects related to sustainability
There are some aspects of research which do not neatly fit into the study of single production 
systems but which are perhaps better studied in a generic manner so that principles, once 
established, can be applied widely wherever needed in any aspect of the farming industry. 
Some of these are listed below.
Postharvest technology (PHT) - Any sustainable production systems must pay high regard 
to ensuring that the food produced is available for people and is not lost after harvest to 
microbial, insect, rodent or physical wastage. TAC believes that research on PHT should 
be increased. It will shortly receive the report of a Stripe Study on PHT which will make 
specific recommendations. In addition the recently received report of a Stripe Study on 
Roots and Tubers recommended that greater attention should be paid to PHT. Increased 
PHT research is necessary to ensure that producer/consumers gain the maximum benefit from 
their products and also to ensure that there is a market in the processing industries for
producers who have surpluses. (0.1% of resource)
Human health - Some forms of agriculture promote the development of organisms that cause 
human diseases (malaria, bilharzia, etc.). In addition the nutritional value of some products 
may not be well balanced, while still others may have toxic constituents (cassava, potato, 
etc.). TAC considers that there should be an opportunity for CGIAR research to address 
issues that are relevant to the health of the consumers and producers of its mandated 
products. (0.1% of resource)
Integrated pest management - There are some aspects of IPM in relation to which research 
on general principles may be necessary before they can be applied to specific problems. This 
category will allow for such research which will be additional to that automatically developed
for particular production systems. (0.1% of resource)
ANNEX I
ACTIVITIES AND THEIR DEFINITIONS
Category 1 : Increasing Productivity
1.1 Germplasm enhancement and breeding
- Pre-breeding
- Crops
- Livestock
- Trees
- Fish
1.2 Production systems development and management
- Cropping systems
- Livestock systems
- Tree systems
- Aquatic systems
Category 2: Protecting the Environment
Category 3: Saving Biodiversity
Category 4: Improving Policies
4.1 Economic and social analysis
4.2 Policy analysis
4.3 Governance and management of public systems
4.4 Research on organization and management of institutes
Category 5: Strengthening National Programmes
5.1 Training and conferences
5.2 Documentation, publication and dissemination of information
5.3 Institution building/advice to NARS
5.4 Networks
Annex I - Page 2
Category 1: Increasing Productivity 
1.1 Germplasm enhancement and breeding
1.1.1 Pre-breeding activities (including applications of techniques in molecular 
biology).
1.1.2 Crops: Crop germplasm enhancement and breeding: incorporating 
primitive and novel germplasm into useful material for breeding purposes, 
as well as germplasm evaluation and conventional breeding.
1.1.3 Livestock: Breed improvement.
1.1.4 Trees: Tree germplasm improvement: breeding of improved trees 
including multipurpose trees and shrubs.
1.1.5 Fish: Breed improvement.
1.2 Production systems development and management
1.2.1 Baseline studies of production systems (including constraint analysis and 
monitoring of sustainability)1: Characterization of the socioeconomic and 
agricultural aspects of farming systems including analysis of constraints to 
production and sustainability.
1.2.2 Development and management of farming systems, including 
socioeconomic evaluation of new technology or practices l: Design and 
testing of farming systems and components for more productive and 
sustainable systems.
1.2.3 Cropping systems
(a) Plant nutrition - crop and pasture nutrient requirements, the availability, 
cycling and uptake of nutrients (including the role of mycorrhiza and other 
symbionts), tillage and fertilizer management.
(b) Plant protection and pest management (diseases, insect pests and weeds) - 
the economic control of diseases, insect pests and weeds of crop, pasture 
and tree species including systems for integrated pest management.
(c) Seed production - increase of seed of elite materials, its certification and 
release.
(d) Postharvest technology - the development of ways of treating commodities 
to reduce losses in the storage and marketing system and improve the 
quality and value of foods through processing.
These are generic activities common to the crop, livestock, tree and fish production 
sectors.
Annex I - Page 3
1.2.4 Livestock systems
(a) Livestock nutrition including studies on feeds, pastures and fodder -
assessment of the nutritional status of livestock in relation to the availability 
of feed resources.
(b) Animal health - epidemiology, biology, immunology and genetics of animal 
pests.
(c) Livestock reproduction - reproductive biology of livestock and the
reduction of reproductive wastage from reproductive diseases and other 
causes.
1.2.5 Tree systems
(a) Silviculture and tree production - the management of trees in agroforestry, 
plantation and natural forest systems to enhance and sustain productivity.
(b) Tree nutrition - tree nutrient requirements, the availability, cycling and 
uptake of nutrients (including the role of mycorrhiya and other symbionts), 
and fertilizer management.
(c) Tree protection (diseases, insect pests and weeds) - the economic control of 
disease, insect pests and weeds of tree species including systems for 
integrated pest management.
1.2.6 Aquatic systems
(a) Fish reproduction - reproductive biology of fish and the reduction of 
reproductive wastage from reproductive diseases and other causes.
(b) Fish nutrition including studies on feeds - assessment of the nutritional 
status of fish in relation to the availability of feed resources.
Category 2: Protecting the Environment
2.1 Ecosystems analysis, ecological characterization and environmental concerns - 
the characterization, classification, mapping and analysis of aquatic and terrestrial 
ecosystems, especially in relation to the functioning and use of ecosystems 
including human use patterns and pressures, climate, hydrology, soil and 
landform.
2.2 Biology and ecology of useful organisms and pests - study of the distribution, 
production and dynamics of economically important plants, animals and fish and of 
the weeds, insect pests and diseases which affect them, and vectors related to 
hazards to human health.
Annex I - Page 4
2.3 Land resources conservation and management - research on the maintenance or 
improvement of the potential productivity of the land resource base and its 
components especially the edaphic, climatic, hydrological and biological resources.
(a) Soil and landform - research on monitoring, maintaining or improving the 
physical and biological characteristics as well as chemical fertility of soils.
(b) Water - research on the conservation and management of rainfall and/or 
irrigation water.
(c) Plants and animals - research on the factors affecting the productivity and 
conservation of natural vegetation including forests and rangelands, and 
research to monitor natural populations of wildlife.
2.4 Aquatic resources conservation and management - research on the maintenance 
or improvement of the potential productivity of the aquatic resource base, 
including research on the population dynamics of aquatic resources and their 
exploitation.
2.5 Processes and mechanisms of sustainable resource systems.
2.6 Modelling of landscape and watershed level phenomena.
Category 3: Saving Biodiversity
3.1 Germplasm collection, conservation, characterization and evaluation - 
collection and maintenance of in vitro (and in situ) germplasm collections and the 
distribution, characterization and documentation of collections.
Category 4: Socioeconomic, Public Policy and Public Management 
Research
4.1 Economic and social analysis
(a) Human nutrition - study of the relationship between such factors as 
nutritional composition of commodities, food quality, income, price, 
socioeconomic characteristics and the nutritional status of people.
(b) Gender, human health hazards and sociocultural organization - analysis of 
gender, health and sociocultural organization in agricultural communities.
(c) Microeconomic and social analysis - research to determine the economic 
and social effects and implications of technologies or policies as they affect 
people, by examining farm, household or village data.
Annex I - Page 5
(d) Market and trade analysis - research to determine the market level 
economic conditions that may result from various technologies, institutions 
or policies and to analyze the impact of trade and macroeconomic policy on 
markets.
(e) Impact assessment and priority setting - research to assess the impact of 
research including cost/benefit analysis and to improve the analytical basis 
on which research priorities are set.
4.2 Policy analysis - Research to determine the desirability of alternative policies from 
the viewpoint of society, taking into consideration productivity, equity, 
sustainability, and environmental concerns.
4.3 Governance and management of public systems (including irrigation systems) - 
Analysis of organizations for the management of public systems (including 
irrigation systems) and the development of innovations to improve their 
performance.
4.4 Research on organization and management of institutes - analysis of research 
and research management processes aimed at the development/enhancement of 
approaches, methodologies and tools for conducting these processes. The 
procedures generated relate to biological/technological research, i.e. technology 
generation efforts and organization and management of NARS.
Category 5: Strengthening National Programmes
5.1 Training and conferences
(a) Training - human resource enhancement including specialized training 
courses, postgraduate research, study tours, etc.
(b) Conferences and seminars - to foster the build-up of NARS capacities and 
the effective functioning of international research collaboration; fora for 
discussion of scientific cooperation among the partners in the global system 
(IARCs, NARS, specialized institutions); stimulating horizontal transfer of 
information and technology among national research systems.
5.2 Documentation, publication and dissemination of information - Efforts to use 
systematically the global knowledge base in areas and disciplines of relevance to 
centres’ research programmes and to make available to NARS relevant information 
on progress and output of centres’ research programmes, through newsletters, 
publications, electronic media, and abstracting services.
Annex I - Page 6
5.3 Institution building/advice to NARS - assisting NARS through the provision of 
advice and counsel. This covers a range of subjects/topics and includes the 
biological sciences (conduct of research) and the organization and management 
field (organization and management of NARS). Primary objective: build-up of 
NARS capacities (institution building).
5.4 Networks - Organizing, coordinating, managing or backstopping of collaborative 
research efforts among various partners in the global research systems with the 
primary objective of building up national capacities. This category does not 
include activities of research networks.
ANNEX H
METHODOLOGY
1. Baseline figures
For agriculture, the reference was data from the FAO AT 2010 study. This was 
available at country level as in the previous data base with the following exceptions to the 
TAC 1992 work:
pulses were given by AT 2010 as an overall aggregate value. Therefore, we 
calculated from the FAO Production Yearbook a regional average share by 
excluding peas from the pulses production values, for the year 1994. The 
importance of peas is rather low, from about 9% of pulses total for Asia to only 
2% for LAC and WANA. Furthermore, to divide pulses into the six CGIAR 
pulse commodities the regional importance of each measured in 1992/94 value 
terms was used to split the commodity group.
oilcrops were given in the AT 2010 study as global demand in 2010 expressed in 
oil-equivalents. A series of ratios had to be used to approximate country demand 
2010 from production in 1992/94. (However, this procedure can lead to zero 
demand 2010 where no production occurred in 1992/94, as the case for some 
oilcrops for WANA, ie coconut, which may not be realistic.)
sweet potato and yam were given as one item by the FAO AT 2010 study. As for 
pulses, the relationships of regional values in 1992/94 were used for splitting it to 
the single commodities.
For forestry, the data was taken from the FAO Forestry paper "Statistics today for 
tomorrow". Roundwood was taken only for the artisanal component, i.e. at 40% of 
demand, as done for this commodity for the analysis presented at TAC 68..
For fisheries, we used the data contained in Ms Lena Westlund’s report on 
"Apparent historical consumption and future demand for fish and fishery products - 
Exploratory calculations". The figures in the analysis were given aggregated to regional 
and sub-regional level. As in the previous work presented at TAC 68, the data is 
modified to represent the average amount of fish food which can be supplied by artisanal 
fishermen. According to the availability as regional figures, for the use in the analysis 
average, by population weighted, regional weights had to be calculated. These average 
weighted "poverty weights" (see below) found are 0.858 for SSA, 0.434 for Asia, 0.472 
for LAC, and 0.423 for WANA. (In the analysis focusing on the poorest countries, only 
the fish demand from China could be excluded from the regional demand figures. 
Therefore, one can assume that the results for fish food are a bit too high.)
Annex II - Page 2
2. Modification process
The expected demand 2010 was valued with the 1989/91 commodities’ 
international price adjusted for purchasing power parity of the countries (Prasada 
Rao/Zarqa/FAO, unpublished) which were used already for TAC 68. This value of 
demand (VOD) was further weighted with new modifiers taking into account issues of 
countries’ welfare/poverty depth, rural/urban population distribution and gender issues in 
the countries of the analysis.
The most important part of the modifiers is represented by a formula which takes 
into account the distribution of future income and attaches thereto a weight in function of 
the gap of this welfare indicator to a certain threshold.
The data necessary to derive country level value are:
1991 capita PPP income in the developing countries, taken from the UNDP 
human development report for 1994
country information on economic growth (World Bank data in WB Atlas 1996 or 
WB Trends in Developing Economies, 1994), the best figure between average 
1985/94 performance (WB Atlas) or the years 1991 to 1994 (WB Trends) was 
used)
overall economic growth was adjusted (subtraction) for population growth 
estimates (UNDP human development report estimates for period 1992-2000)
Gini coefficients for income distribution within the countries. (This unpublished 
data was provided by the TAC Chair and allowed the assessment of approximately 
50 out of 120 countries for this item, but which have about 85% of the total 
population of the countries in our database. For the countries with no Gini 
coefficients, we used weighted regional values. The figures found are 40% for 
SSA; 35.4% for Asia; 55.7% for LAC; and 35.8% for WANA.) 
the UNDP human development report provided the information on percentages 
population distribution (rural/urban).
FAO-Agrostat data is used for the numbers of males and females economically 
active in agriculture.
The steps applied with the above data to derive country weights are:
1. PPP incomes/caput in 2003 were estimated. The best figure for economic growth 
from the two WB sources was taken, and the population growth was deducted 
therefrom. Starting default values were set at +0.5%. The formula used is:
future value = present value (1+rate)term(12years)
These 2003 PPP estimates were then assessed as welfare indicators and converted 
into weights after transforming into a scale. Following formulae were suggested by Mr 
Martin Ravallion/World Bank (e-mail 5 March 1996) to TAC and used in the study:
2. Country welfare indicator W as influenced by the income distribution 
(negatively) by the Gini coefficients and the PPP income/caput (positively) at
Annex II - Page 3
country level. An even and smooth distribution of income with Gini coefficient 
values close to 1 would lead to negligible differences between welfare indicators 
and PPP income. Basically, the following formula will reduce the values of the 
PPP incomes adjusted for distribution of wealth:
W = (1 - GINI COEFFICIENT) * PPP INCOME
3. Therefrom, Mr Ravallion suggested to derive weights W* constructed to be 
highest for the lower values of the welfare indicator W. After a certain value Z, 
these should be zero. Mr Ravallion suggested that both the Z-values and the alpha 
exponent should be tested as the weight W* is depending on the values (see 
below). The applied is
W* = (1 - W/Z) aIpha
Three sets of weights W* were obtained for the analysis by using three 
combination of Z (upper PPP threshold) and alpha values (exponent):
Z value 6000 or 9000 PPP $, alpha value a 2 
Z value chosen as 6000 $, alpha value as 2 or 3.
Two further sets of country weights were obtained for W* values with z set as 
6000 and alpha as 2 by incorporating into the modifier weights already known since the 
TAC January 1996 meeting in Rome:
Inclusion of producers’ gender as additional factor in the modifier with the 
weights given as 1 (full) for the share of female and 0.8 (reduced) for the share of 
male agricultural producers. The formula is
Weight = W* times (F 4- 0.8 M), and
Adding the population distribution figures to the modifier, giving the share rural 
population full weight and the share of urban population a reduced weight.
Weight = W* times (R + 0.8 U).
Accordingly, the analysis used five modifiers (weights) and leads to six results for 
the final comparison when including the value of future demand of CG commodities 
expressed in present economic terms. It should be noted that for the analysis of the 
fisheries sector (see above, baseline figures) five sets of average, by population weighted, 
regional weights were calculated as depending upon the five modifiers used.
The utilization of different Z values leads to the exclusion of countries which 
obtain a distribution adjusted PPP income in 2003 above that value. These are for
Z set a 9000 PPP $ the countries Singapore, South Korea, Taiwan in Asia; and 
Cyprus, Qatar and United Arab Emirate for WANA; and
Annex II - Page 4
Z set as 6000 PPP $ additionally the countries of Mauritius in SSAfrica; China, 
Malaysia and Thailand in Asia; Chile in the region LAC; and for WANA, the 
countries of Bahrain, Kuwait, Oman, and Saudi Arabia drop out additionally.
Obviously, the exclusion of large countries as China after setting the z-value low 
to 6000 PPP $ are influencing the composition of the data set and the results of the 
quantitative analysis.
Annex II - Page 5
Table: Prices used in calculations
Commodity International Price Proportion 
Price, $/mt US$/mt int’l price to 
period 1989/91 (TAC, 1992) TAC, 1992
(1/2)
1 2 3
Rice 292 284 103%
Wheat 144 144 100%
Maize 124 104 120%
Barley 114 128 89%
Sorghum 124 93 133%
Millet 158 132 119%
Cassava 68 66 103%
Potato 110 180 61%
Sweet potato 76 82 93%
Yams 137 105 130%
Banana 154 150 103%
Plantain 96 144 67%
Chickpea 785 339 232%
Cowpea 266 591 45%
Pigeonpea 723 393 184%
Broad bean 229 591 39%
Lentil 547 489 112%
Beans 539 591 91%
Pulses* 555
Soybean 234 265 88%
Groundnut 491 585 84%
Coconut 106 143 74%
Tomato, fresh 192 195 99%
Onion, dry 202 235 86%
Cabbage 100 136 74%
Orange 176 512 34%
Lemon&lime 274 774 35%
Pineapple 151 293 51%
Grape 340 809 42%
Apple 307 378 81%
Sugar 141 218 64%
Coffee 959 2182 44%
Tea 1410 1862 76%
Cocoa 663 1670 40%
Tobacco 1498 2522 59%
Rubber 500 1050 48%
Cotton (lint) 1351 1244 109%
Cotton (seed) 175 149 117%
Seed cotton 554 * 504* 110%
Jute 270 329 82%
Hemp 559 291 192%
Sisal 222 419 53%
Palm oil 165 365 45%
Beef&Buffalo meat 2226* 1458 153%
Sheep&Goat meat 2099* 1652 127%
Pigmeat 1342 1026 131%
Poultry meat 1267* 727 174%
Milk, total 268* 306 88%
Eggs, hen 1005 840 120%
Inland catch 785** 763 103%
Inland culture 1517** 1474 103%
Marine catch 658 ** 639 103%
Marine culture 1606** 1561 103%
Fish food* 938
Charcoal 182** 182 100%
Fuelwood 55 ** 41 134%
Industrial roundwood 113** 131 86%
Note: * composite prices, ** imputed prices 
______ Pulses and fish food weighted average price
ANNEX m
1. CROPS
1.1. Background
Crops and their products provide about 52% of the total value of production of 
agriculture, forestry and fisheries in developing countries. In Asia this share amounts to 
59%, in sub-Saharan Africa 41%, in Latin America and the Caribbean 42% and in West 
Asia-North Africa about 51%.
Research to improve the productivity of the most important food crops in 
developing countries has been the central theme of the CGIAR since its inception. The 
CGIAR has a multidisciplinary research approach to increasing crop productivity. The 
research has four main objectives: to increase yield potential; to narrow gaps between 
potential and actual yields; to improve yield stability; and maintenance research to prevent 
the erosion of attained yield levels. Crop productivity research in the CGIAR consists of 
two sets of activities: germplasm enhancement and breeding, and cropping systems 
research. These two sets of activities fall into one of the five major "undertakings" called 
"increasing productivity". Germplasm enhancement and breeding include: pre-breeding 
activities; the incorporation of primitive and novel germplasm into useful material for 
breeding purposes; and germplasm evaluation and conventional breeding. Cropping 
systems include: plant nutrition; plant protection and pest management; seed 
production; and postharvest technology.
The pay off to crop productivity research in the CGIAR has been large, and the 
impact of research on rice and wheat has been particularly impressive (Anderson et al, 
1988). Significant farm-level impact has also been achieved through research on maize, 
millet (particularly in India), groundnut (in India), cassava and phaseolus beans. 
Although encouraging progress in the development of technology for the other crops has 
been achieved, evidence of impact is still largely anecdotal. Progress has been 
particularly slow for grain legumes.
This section discusses important factors for assigning CGIAR priorities to 
particular crops. For each crop under consideration, the importance of the commodity in 
the diet and the production system, research opportunities and history, the strategic 
breeding goals, and the role of the CGIAR hitherto are highlighted. Implications with 
respect to the future priority ranking of each commodity are discussed in Chapter 6.
TAC acknowledges the importance of mixed cropping systems and the difficulty 
of allocating priorities to the crops involved, which are often of minor importance 
globally but can play a significant role in particular farming systems. In addition, for 
many crops, particularly roots and tubers, food legumes and vegetables, the database is 
weak. Estimates on their value of production and yield levels are often crude guesses, 
and in a quantitative analysis these crops may, therefore, get a lower priority ranking than 
they merit.
Annex III - Page 2
1.2. Cereals
1.2.1. Rice
Globally, rice is the most important crop in terms of its contribution to diet and 
value of production. Of the 146 million ha harvested globally in 1994, about 142 million 
ha were in developing countries, producing 506 million tonnes of paddy. Asia is the 
primary producer, accounting for 93% of production in developing countries. Latin 
America and the Caribbean accounts for 4%, West Asia-North Africa for 2%, and 
sub-Saharan Africa for 1%. Rice provides between 35% and 80% of the calories 
consumed by 3.3 billion people in Asia, and 8% of food energy for 1 billion people in 
sub-Saharan Africa and Latin America and the Caribbean. Only about 4% of world rice 
production is traded on the international market; most countries rely almost entirely on 
domestic production to meet their demand. A striking exception to this is in sub-Saharan 
Africa where 35% of total rice consumption is derived from imports. Price formation on 
the international market is heavily influenced by subsidies and other protective measures. 
In West Africa and Latin America, rice is a relatively new staple in the diet. Per caput 
consumption in West Africa has doubled over the past two decades, and continues to 
grow by 2% annually, while in Latin America it has increased by about 20%.
Rice production increased in varying degrees in all developing regions by an 
average of 2.7% annually during the 1968-80 period, but since then, the growth has 
decelerated to 1.5% per year. West Africa saw the highest rates of increase during the 
1980s at 8.5% annually. About three-quarters of the increase can be attributed to higher 
yield levels in irrigated rice in Asia, attained through the widespread adoption of high- 
yielding varieties, fertilizer and irrigation. Production increases in Latin America and the 
Caribbean resulted largely from the spread of new varieties. However, in sub-Saharan 
Africa and West Asia-North Africa they resulted from an increase in the area cultivated 
rather than from increases in yields. In West Africa, although yields grew at 1.9% per 
annum during the 1980s, area increases still represented fully 75% of the overall 
production growth.
Over the last 25 years, the rice areas in Asia showed a mean yield increase of 
62% (from 2.26 to 3.73 t/ha). However, yields vary widely between countries in all 
developing regions. For example, in Asia, the average yield of rice in India, which has 
one-third of the region’s rice area, is only 2.82 t/ha, whereas it is 5.87 t/ha in China, 
which also has about one fourth of the region’s rice area.
Significant advances in rice production have been made over the past two decades 
in Latin America and the Caribbean (LAC). About 220 new rice varieties were released 
for flooded environments, with 40% coming from crosses made at CIAT, 13% at IRRI 
and almost all of the rest has parentage form IARCs’ progenitors. Modem semidwarf 
rice varieties now account for more than 90% of all flooded rice production, itself 
representing 70% of total rice production in the region. Average yields in flooded areas 
have risen from 3.0 tons per hectare in the mid 1960s to 4.5 t/ha in 1990; and total rice 
production doubled between 1967 and 1990, making the region largely self-sufficient in 
rice. With rice prices falling by 40% in real terms over the period, consumers have
Annex III - Page 3
benefited greatly. Rice is well established as a "wage good", and the crop has become 
the most important source of calories and proteins for that 20% of the region’s population 
with the lowest incomes. Rice is particularly important from the standpoints of growth 
and equity. Rice is preferred by the poor because it is cheap, nutritious, appealing, easy 
to prepare, and easy to store and transport.
If past trends in demand continue, world rice production will need to increase by 
21% by 2005, and by 65% by 2025 (1.7% annually). While Asia has achieved marginal 
self-sufficiency in rice for the present, further increases through higher yields or 
increased cropping intensity will be necessary to keep pace with demand. The leading 
rice-growing countries in Asia will need to increase their rice production by 100% by 
2025 (2.3% annually).
These escalating demand levels will require a concerted research effort to 
continue the development of improved technologies for production. To date, the pay off 
from CGIAR investments in rice research has been large. The internal rate of return 
from international rice research over the last 30 years can reach at least 80%, and during 
this period the new rice varieties allowed for an increase in rice production which was 
sufficient to feed about 600 million more people (IRRI, 1991). The impact of new 
technology has so far been confined primarily to irrigated areas - which make up some 
53% of the world’s harvested rice area - and to favourable rainfed areas. Further 
research must be conducted for these areas to protect and build on what has already been 
achieved. Recent work at IRRI has given strong indications that the high yield levels 
obtained on farms with favourable management conditions are not sustainable due to a 
variety of factors such as poor quality of irrigation water, the lack of micronutrients, and 
the vulnerability of improved varieties to pests and diseases. In order to meet the 
problem of yield erosion, further efforts in maintenance research, as well as in lifting the 
yield ceiling, will be required. However, if rising demands are to be met, other rice 
growing systems will also have to receive attention. These include: shallow rainfed rice, 
which accounts for almost 25 % of the harvested area in Asia; deep-water and floating 
rice, which accounts for about 13% of the harvested area in Asia; and dryland or upland 
rice, which accounts for 75 % of the harvested area in Latin America and the Caribbean 
and 50% of that in sub-Saharan Africa.
In 1983, 25% of total CGIAR allocations were spent on rice. In 1985, TAC 
recommended that the overall effort forbice be reduced and that the existing shift in 
research emphasis away from applied research on irrigated systems be reinforced. TAC 
considered that the CGIAR System’s future efforts on rice should concentrate more on 
non-irrigated systems, and in basic research on irrigated rice in collaboration with 
specialized institutions.
These recommendations reflected the successes that had already been achieved in 
rice research, especially in the more favourable environments. Today, more than two- 
thirds of the rice lands of developing countries are planted with high-yielding modem 
varieties. Furthermore, the CGIAR System’s collaboration in rice research has 
significantly strengthened many national research programmes, allowing them to assume 
an increasingly large share of the responsibility for research. This is particularly true of
Annex III - Page 4
some of the largest rice producing countries, e.g. India, Thailand, the Philippines, 
Bangladesh, China and Korea.
Non-irrigated wetland and dryland rice systems comprise almost half the global 
area under rice production. The production constraints of these systems are more 
complex than those of irrigated rice because of the lack of control in water management 
and the more limited knowledge base for research. In Latin America and the Caribbean 
and in sub-Saharan Africa, CGIAR emphasis has shifted to dryland and rainfed lowland 
rice research.
The 1986 recommendation for the movement towards basic research and to target 
more difficult environments was made in the belief that the exploitation of genetic 
diversity was fundamental to achieving higher and more stable yields, resistance to major 
pests and disease, and better drought tolerance. For both irrigated and non-irrigated rice, 
it will be necessary to develop new and better breeding techniques, to increase knowledge 
of the factors determining resistance and tolerance, to raise yield potential by using 
biotechnology and to exploit genes from species closely related to O. Saliva. The CGIAR 
System should therefore emphasize strategic research. In so doing it will catalyze and 
support basic research in other institutes, and play an active role in encouraging the 
application of new techniques to the rice production problems of developing countries.
In considering future priorities for rice research, TAC should also consider the 
substantial impact obtained from CGIAR efforts in rice research, in Asia, Latin America, 
the Caribbean and West Africa. In Latin America, a Latin American Irrigated Rice 
Fund, mainly supported by the private sector and IARCs, has been created, that should 
ensure continuity in rice research activities at the regional level. This process clearly 
shows that Latin American rice producers are aware of the value and innovation of new 
technologies.
In 1992, TAC recommended a continuation of current levels of CG investment in 
rice research, but with a shift in focus towards more strategic germplasm research 
necessary to lift the yield ceiling of the crop, and to sustain current yield levels.
The future of rice research holds exciting challenges and opportunities. Rice 
research aims at making significant contributions to environmental goals such as the 
protection of tropical forests and reduction in agrochemical use, as well as in feeding 
people through devoting its efforts to the development of improved rice gene pools and 
integrated crop management.
1.2.2. Wheat
After rice, wheat is the single most important food source in the developing 
world, contributing more calories to diets than all other cereals combined. It is higher in 
protein content than almost all other cereals. Within wheat, a distinction can be made 
between durum and bread wheats, and between bread wheats, between winter, facultative 
and spring wheats. Durum wheat accounts for 5 % of developing country wheat 
production, and 70% of it is grown in West Asia-North Africa.
Annex III - Page 5
In 1992/94, developing countries accounted for 45% of world wheat production 
(551 million t) and 46% of world wheat area (219 million ha). Half the total increase in 
production in the 1970s and 70% in the 1980s came from the developing world.
CIMMYT varieties now cover at least 50 million hectares and account for 70% of 
improved varieties. In 1992/94, Asia accounted for 71% of the developing world’s 
production, West Asia-North Africa for 20%, Latin America and the Caribbean for 7% 
and sub-Saharan Africa for 7%. In the West Asia-North Africa region, wheat is the most 
important food crop in terms of its calorie contribution.
Wheat production in the developing regions as a whole increased by 5 % annually 
in the 1970s and by 4.3% in the 1980s. The five largest producers - China, India, 
Turkey, Pakistan and Argentina - raised production at an average annual rate of 5.4% in 
the 1970s and 4.3% in the 1980s, largely through yield increases. In the remaining 
developing countries, the growth rate was only 1.5% during the 1970s, but increased to 
3 % during the 1980s. Trends in yield levels over the past two decades have varied 
considerably. China experienced an increase of 75% in the 1970s and 49% in the 1980s; 
India 25% in the 1970s and 45% in the 1980s; West Asia-North Africa 35% in the 1970s 
and 16% in the 1980s, sub-Saharan Africa 55% in the 1970s and 38% in the 1980s;
Latin America and the Caribbean 37% in the 1980s. Improved varieties and associated 
technologies have had a major impact on wheat production in the developing world, 
causing an absolute yield increase from 1.64 t/ha to 2.23 t/ha in the past decade. Today, 
some 60% of the wheat lands in developing regions are sown with modem varieties.
Wheat imports by developing countries doubled in the 1970s and further 
increased substantially in the 1980s. Many countries financed their purchases of wheat 
with limited foreign exchange, indicating the high priority assigned to wheat as a food. 
Even the countries that produce wheat have become more reliant on imports during the 
past two decades. Among countries consuming 100,000 tonnes or more annually, per 
caput wheat imports declined only in Turkey, India, Pakistan and Zimbabwe.
Growth rates in consumption are closely linked to rising incomes and 
urbanization. The correlation with rising incomes reflects not only greater overall food 
consumption, but also a switch to wheat in preference to other starchy staples, and the 
use of wheat as animal feed. In West Asia-North Africa, where wheat originated, 
consumption is high at all income levels and in both rural and urban areas. Other factors 
contributing to increased wheat consumption are the lagging production of many other 
staple foods; and food aid and pricing policies, which lower wheat prices and create a 
bias in favour of wheat products.
For the developing regions as a whole, the annual demand for wheat is projected 
to grow at 3 % over the coming decade. Demand will rise particularly rapidly in sub- 
Saharan Africa, at 5.1% per annum, and at 2.9% in other regions.
Expansion in wheat area has declined from 1.7% per year in the 1950s to under 
1% currently, and is projected at 0.8% in the future. Consequently, wheat yields will 
need to rise by 2.2% each year to meet the projected demand growth of 3%. Semidwarf 
wheat varieties are already sown in most of the wheat area and fertilizer applications are 
relatively high on much of the irrigated land. However, in most developing countries
Annex III - Page 6
absolute yields are still comparatively low, less than half the average yield in Europe. 
Even the current yield levels of the five largest producers cannot be considered high: 
China, 3.45 t/ha; India, 2.36 t/ha; Turkey, 1.99 t/ha; Pakistan, 1.94 t/ha; and 
Argentina, 2.14 t/ha.
Diseases, insect pests and environmental stresses, especially drought, are 
important constraints but they are not the only ones: crop and water management, 
socioeconomic factors and the policy environment are equally important for achieving 
further sustainable increases in yield. The impact of CGIAR investments in wheat 
research has been impressive. Varieties to which CIMMYT has contributed, now cover 
about 47 million ha, and between 50% and 70% of improved wheat varieties released 
during the last 30 years have been based on crosses made by CIMMYT.
In the West Asia-North Africa region, where most of the wheat is rainfed, winter 
rainfall is low and erratic and crop yields are limited by biological and environmental 
constraints as well as by management and socioeconomic factors. Except in Turkey, 
research in this region has not addressed the needs of high elevation areas, which require 
winter or facultative wheat varieties with tolerance to a range of environmental stresses, 
including cold. In the lowland areas of the West Asia-North Africa region, tolerance to 
heat and salinity, as well as to cold, are required. For the lowland irrigated areas of the 
semi-arid tropics and subtropics with summer rainfall, where the crop is grown during the 
cool season, varieties with better tolerance to relatively high temperatures are required. 
Aluminium toxicity is a constraint to bread wheat production in large areas of highly 
leached acidic soils in the subhumid and humid subtropical areas of southeast China. For 
the higher elevation areas of the cool tropics and subtropics with summer rainfall, spring 
wheat varieties with better adaptation to biotic and abiotic stresses are required.
According to the ACIAR analysis, the highest pay off from future investments in wheat 
research can be obtained in the warm and seasonally dry subtropics with summer rainfall, 
and in the cool subtropics.
In its 1986 assessment of priorities, TAC considered the importance of wheat as 
a food crop and the increasing reliance of developing countries on wheat imports. It also 
considered the strong research programmes on wheat in developed countries and the 
growing strength of national programmes in Latin America, Turkey, India and China, as 
evidenced by the remarkable yield and production increases achieved in those countries 
during the 1970s and 1980s. The well organized international wheat trade, the export 
capacity of some developing countries, and the increasing demand for wheat in countries 
with unfavourable environments for its production make the concept of self-sufficiency 
inappropriate for many areas.
In 1986 TAC considered that the trend of the centres concerned to transfer a 
number of research functions to national systems while continuing to provide them with 
technical support was reasonable and should be accelerated. This led to the 
recommendation that the System’s overall efforts in wheat research should be gradually 
reduced by 10% over the following five years. TAC also recommended that research 
should concentrate on increasing production on marginal lands, including those in tropical 
areas.
Annex III - Page 7
The pay off from investment in wheat research has been very high, but further 
efforts are required to sustain the increased yield levels achieved. In 1992, TAC 
recommended a continuation of CG efforts at current levels, but noted that in the long 
term the priority of wheat was likely to decline given the growing importance of 
alternative sources of supply.
1.2.3. Maize
Among the food crops, maize ranks third after rice and wheat both in terms of 
calorie contribution and in terms of value of production. For the 1992 to 1994 three-year 
average, developing countries produced an estimated 43 % of world production (522 
million t) from about 84 million ha (66% of total maize area). The crop is grown in all 
the developing regions. Of the total for all regions, China alone accounts for 44%, Latin 
America and the Caribbean for 30%, the rest of Asia for 14%, sub-Saharan Africa for 
8%, and West Asia-North Africa for 47%.
Where grown for human food, maize is an important source of calories for the 
poor. The crop is widely grown in mixed cropping systems by subsistence farmers. For 
all developing countries, annual per caput human consumption is only 20 kg, but in Latin 
America and the Caribbean (the homeland of maize) it is 80 kg, and in sub-Saharan 
Africa 60 kg; in some countries of both regions, per caput human consumption is as high 
as 100 kg per annum. Maize provides about one-third of the mean calorie intake in these 
two regions, but little more than 5 % in the other regions. Maize stover is an important 
byproduct in many countries.
The use of maize for livestock feed has become increasingly important and now 
accounts for about 54% of consumption in developing countries. In the subtropical areas 
of South America, it is the main use, and it is important for this purpose in the rest of 
Latin America and the Caribbean, and in the West Asia-North Africa region. In the 
1970s and the 1980s, the use of maize as feed in developing countries grew by 5.3% per 
annum, and in Asia and West Asia-North Africa it grew at three times the rate for direct 
human consumption. It grew at twice the rate for food use in sub-Saharan Africa, but 
from a low base, so that use of maize for feed is still relatively low in that region.
During the current decade, demand for food maize for the developing regions as 
a whole is expected to grow at 1.6% per year, for feed maize at 4.9%, and for food and 
feed maize combined at 3.5%. Total regional demand is projected to grow at 3.1% for 
sub-Saharan Africa, 3% for the West Asia-North Africa region, 3.8% for Asia and 3.3% 
for Latin America and the Caribbean.
During the past decade, developing countries achieved a 22% increase in yields. 
However, this average figure masks China’s considerable gain of 50% (associated largely 
with the adoption of improved varieties) at one end of the scale, and a decline of almost 
15 % in West Africa at the other. The variation in yields per ha is equally dramatic: 
these range from more than 3 t in subtropical South America and China, through just 
under 2 in West Asia-North Africa, to about 1.6 in Central and tropical South America, 
about 1.5 in South and South-East Asia, about 1.1 in East and Southern Africa and India,
Annex III - Page 8
and less than 1 in the other sub-Saharan African regions. Sub-Saharan Africa achieved 
some increase in production during the 1970s, but this was the result of an increase in the 
area harvested. During the 1980s, yield gains were 16% (from 1.96 to 2.28 t/ha) for the 
developing regions as a whole, 12% for sub-Saharan Africa, 7% for Latin America and 
the Caribbean, 38% for West Asia-North Africa and 27% for Asia.
In the long term, the global pattern of use will continue to change with rising 
incomes and urbanization. Although consumers in developing countries will tend to 
spend less on maize as they switch to other foods, maize consumption will increase 
because of its increasing use as feed. In the low-income countries, particularly in sub- 
Saharan Africa, this scenario is likely to develop more slowly, and in the medium term 
the problem will be one of increasing demand for maize for human consumption against a 
background of declining per caput production. For example, in East and Southern Africa, 
where maize is the staple food and is grown on about 30% of the cultivated crop area, 
production will need to double by the year 2000 if the region is not to face massive bills 
for food imports. Increased production in sub-Saharan Africa will need to come mainly 
from increased yields.
The demand for hybrid maize has increased rapidly in recent years, particularly 
in Asia and Latin America.
The potential for increasing yields is quite high and the pay off from CGIAR 
investments in maize research has been substantial, particularly in the lowland tropics.
The main constraints are environmental stresses (particularly drought), diseases and insect 
pests, and low levels of external inputs. Both improved open-pollinated varieties and 
hybrids are required, depending on local needs and the efficiency of national seed 
producers. In the lowland tropics, the development of better varieties and improved 
management practices relevant to farmers’ needs and constraints would contribute 
considerably to improved production. In sub-Saharan Africa, low fertilizer rates and poor 
management currently pose a greater constraint than does the availability of high-yielding 
varieties. In East and Southern Africa, where there are extensive lowland and highland 
areas ideally suited to maize production, the pay off from the development of appropriate 
technology for small-scale farmers is exceptionally high, as the case of Zimbabwe shows. 
In some environments with bimodal rainfall, short-cycle maize outperforms both sorghum 
and millet.
In 1986 TAC considered that the CGIAR System’s major effort in maize research 
was justified and should be maintained over the long term, and that some additional 
support should be given in the short to medium term to accelerate the promising results 
from work in progress. The recommendation took into consideration the crop’s value as 
food, feed and a source of income for low-income groups and small-scale farmers 
worldwide; the projected increase in demand; and the expectation that strategic research 
could successfully address the constraints to higher yields in many developing countries. 
TAC recognized the urgent and specific needs of sub-Saharan Africa and recommended a 
shift of effort to those areas where maize is the staple food.
In future, for sub-Saharan Africa, emphasis should be placed on development of 
maize-based cropping systems which utilize the available resources efficiently, maintain
Annex III - Page 9
long-term productivity of the lands, and minimize postharvest losses. Varieties with 
improved nitrogen and water-use efficiency, resistance to parasitic plants {Striga spp.), 
and resistance to storage pests (weevils and pathogens) are required for these systems. 
End-user requirements must be considered in breeding programmes in order to increase 
processing efficiency and the recovery of end products. This should enhance adoption of 
higher-yielding improved varieties.
In sub-Saharan Africa as a whole, perhaps the greatest constraints to increased 
maize production are Striga spp., particularly Striga hermonthica and Striga asiatica. 
These obligate root parasites frequently cause yield losses on cereals of more than 50%, 
and in many areas infestation has become so severe that cereal production has been 
abandoned. Maize is particularly susceptible to parasitism and yield loss. Control of 
Striga spp. under African farming conditions is a complex problem. Whilst host-plant 
resistance can contribute, sustainable control can be achieved only by integrating several 
control options into diverse cropping systems that involve not only maize, but also other 
cereals, legumes, roots, tubers, and vegetables. In this regard, legume rotations which 
have major effects on controlling the parasites, and additionally provide food and improve 
soil fertility, have tremendous potential. Maize resistance to the parasite can contribute to 
Striga spp. control. However, for this option to be realized, improved Sm'ga-resistant 
high-yielding maize cultivars which farmers prefer for their cropping systems, even in the 
absence of Striga pressure, and which have consumer acceptance, need to be available. 
Use of Sm'ga-resistant maize and any associated parasite control cannot be sustained 
without meeting this prerequisite.
In 1992, TAC recommended maintaining current efforts in maize research, but 
noted the rapidly growing involvement of the private sector in the maize industry, making 
a reduction of maize priority likely in the long term.
1.2.4. Barley
Barley is the fourth most important cereal crop. It is grown on about 70 million 
ha and global production is 160 million t. Developing countries account for about 18% 
(26 million t) of global production and 25% (18.5 million ha) of the harvested area. In 
most developing countries barley is a typical crop of poor farmers and of hostile 
environments. In Tibet, Ethiopia and the Andes, it is cultivated on the mountain slopes at 
elevations higher than other cereals. In many areas of North Africa, the Near East, 
Afghanistan, Pakistan, Eritrea, Yemen it is often the only possible rainfed crop, and 
therefore neither the area nor the production reflect the actual importance of the crop.
Compared with average yields of 3.7 t/ha in Europe and 3.0 t/ha in North America, 
yields average 0.8 t/ha in Africa and 1.7 t/ha in South America, 2.1 t/ha in Asia and 
1.3 t/ha for West Asia-North Africa. West Asia-North Africa accounts for 75% of the 
harvested area in the developing regions, Asia for 11%, sub-Saharan Africa for 5% and 
Latin America and the Caribbean for 5 %. Some two-thirds of the production is in West 
Asia-North Africa, and in no other developing region is the crop as important relative to 
other commodities. Asia accounts for another 21% of production, China for 13% and 
India for 6%.
Annex III - Page 10
Barley grain is mostly used as feed for animals, malt and human food. Barley straw 
is used as animal feed in the Near East, North Africa, Ethiopia, Eritrea, Yemen, in the 
Andean region and the Far East. Malt is the second largest use of barley, but the CGIAR 
System is not directly involved with improvement of malting quality. In many countries 
such as the highlands of Tibet, Nepal, Ethiopia, the Andean countries, North Africa, 
Afghanistan, India and Russia, barley is still used as human food either as bread (mixed 
with bread wheat) or for specific recipes.Developing hull-less barley with improved yield 
and resistance to common diseases offers great potential for these areas.
In the dry and cold areas of West Asia significant progress has been made using a 
breeding methodology developed at ICARDA and based on the use of locally adapted 
germplasm and selection for specific adaptation. This has led to the adoption of varieties 
in very dry areas where it was thought breeding can not have an impact. Consistent yield 
increases of about 20% have been reported by farmers who adopted the new cultivars.
Significant progress has been made in decentralizing breeding activities to North 
Africa. The same approach is now being gradually implemented in West Asia and in the 
Far East, but more needs to be done for the adoption of this breeding methodology by 
NARS.
The livestock industry accounts for almost one-third of the value of agricultural 
production in West Asia-North Africa, and the increasing demand for meat will mean an 
increased demand for barley as feed.
The main constraints to improved production are environmental stresses 
(especially drought), and insect pests and diseases.
In 1986 TAC recommended that the overall allocation to barley research be 
reduced slightly, but that the effort for West Asia-North Africa be strengthened by 
phasing out research for other regions. The recommendation took account of the 
relatively low importance of barley elsewhere (excluding barley grown for malt) and the 
strength of many national agricultural research programmes.
The future challenge is to consolidate on past achievements and to develop a new 
methodology for introducing farmers’ participation in breeding as a way to exploit 
specific adaptation and bypass problems to technology transfer.
In 1992, TAC recommended maintaining CG efforts, particularly in areas where 
poor farmers are heavily dependent on barley.
Annex III - Page 11
1.2.5. Sorghum
Some 70% of the world’s sorghum production (60.9 million t) and 90% of its 
sorghum area (43.5 million ha) are located in the developing regions. Sorghum is a major 
crop of the lowland semi-arid tropics with summer rainfall, where it has a special 
importance, together with millet, as a staple food for millions of very poor people in 
drought-prone, high-risk areas. In West Africa, sorghum is an important crop in the 
subhumid areas, where it is intercropped with millet, maize and cowpea. Sorghum is also 
an important crop in the medium-altitude areas of Ethiopia, and East and Southern Africa. 
Sorghum tends to have a negative elasticity of demand, and is usually substituted by other 
foods when income permits. In many areas, the stalks and foliage — used as fodder, fuel 
and construction materials — are as or more important than the grain.
Although the average contribution of sorghum to diets may be low in most 
developing regions, in semi-arid West Africa it contributes 13% of calorie intake and 
over 11% of protein, making it the second most important food commodity after millet.
In India, it accounts for almost 6% of calorie intake, but in some selected states, e.g. 
Maharashtra, accounts for a more significant share of calories. In Latin America and the 
Caribbean, most of the crop products are used for feed.
Of the area harvested in developing countries, the three-year average 1992 to 
1994 shows that Asia accounts for 38%, sub-Saharan Africa for 52%, West Asia-North 
Africa for 8% and Latin America and the Caribbean for 12%. India, the largest single 
producer, accounts for 33 % of the sorghum area in the developing regions and China for 
3.5%. In sub-Saharan Africa, some 60% of the sorghum area is located in West Africa, 
the rest being in East and Southern Africa. However, there is little correlation between 
area harvested and production share because of the considerable regional variation in 
yields: these range from 3.9 and 3.7 t/ha in Peru and China respectively to 0.95 t/ha in 
India and 0.9 t/ha in Western Africa, where many national averages are even lower.
In the 1970s, substantial yield increases were achieved in China, Latin America 
and the Caribbean, and also in India from a very low level. During the 1980s and early 
1990s, sorghum area in Asia declined by 25% and production by 5%. The decline in area 
was mainly in India and China and was offset by further yield increases. India’s area 
declined by 21% but production increased by 9% due to substantial increases in yields.
The decline in area was not uniform, some states showing stable or increasing sorghum 
cultivation. China’s area declined by 52% and production by 27%. In sub-Saharan Africa, 
sorghum area has increased by 48% during the same period and production by 23%. In 
Latin America and the Caribbean, there was a decrease in area of 30% and in production 
of 25 %. The pattern observed seems to indicate substitution of other crops for sorghum in 
favourable areas and consolidation or expansion of sorghum growing in rainfed areas that 
are not suitable for other crops.
The world’s most urgent food production problems lie in drought-prone areas 
such as those of India and the Sahelian zone of Africa, where sorghum and millet are the 
staple food crops. The events of recent years have demonstrated as never before the
Annex III - Page 12
extreme vulnerability of such areas, where the effects of a series of bad years have led to 
famine and dependence on food aid.
The main constraints to sorghum production being addressed through research are 
drought and biotic stresses. The former targets drought escape through earliness and 
evaluation of the effects of specific traits that have been associated with drought 
resistance. The latter include Striga spp., which cause serious losses and prejudice 
sustainability of production where the land is planted to successive crops of sorghum, 
particularly in sub-Saharan Africa. Grain moulds, causing severe reduction in grain 
quantity and quality, are a problem wherever improved cultivars have been adopted in 
more favourable production environments, particularly in India. Insect pests cause 
substantial losses in grain yield in different regions - shoot fly in the post-rainy season 
crop in India, midge and head bug in Western Africa, and stem borer in all areas. Foliar 
diseases are important constraints in Western Africa, Latin America and parts of Asia, 
where they affect both grain yield and stover quality for animal feed.
A major objective of varietal improvement research is broadening the genetic 
base of breeding materials. This is achieved by deliberate introduction of new genetic 
materials into resistance breeding for the constraints mentioned above. It also includes 
development of broad based populations targeting improvement of specific traits that are 
important for sorghum in many areas. The main targets are dual purpose varieties and 
hybrids which combine high yields of both grain and stover. It includes forage sorghum 
hybrids, as forage uses of sorghum are increasing rapidly in areas of Asia and Latin 
America. Other research emphasizes development of management options to mitigate the 
same biotic and abiotic stresses and their integration into management packages suitable 
for small scale farmers of the semi-arid tropics.
In 1992, TAC recommended maintaining CG efforts, particularly in those areas 
where poor farmers were heavily dependent on sorghum.
1.2.6. Millet
In Asia and sub-Saharan Africa, pearl millet is the most important crop grown 
under dryland conditions in the lowland semi-arid tropics and subtropical areas with 
summer rainfall. There it is a staple food, together with sorghum (in sub-Saharan Africa) 
or wheat (in Asia). Pearl millet provides food for some of the world’s poorest countries 
and poorest people. It produces grain and fodder under conditions too hot, too dry, and 
on soils too poor for sorghum and maize. Its straw is a valuable livestock feed in those 
farming systems.
Because some countries combine their statistics for sorghum and millets, the data 
for millets tend to be unreliable, especially for sub-Saharan Africa. It appears that millets 
are harvested from about 34 million ha annually in developing countries, of which 26 
million ha are pearl millet and the remainder an array of other small-seeded grasses 
(finger millet, foxtail millet, proso millet, tef, and other species) that are harvested for 
grain and classified as millets. For pearl millet, India accounts for about 38% of the 
area, and West Africa for about 46%. Pearl millet is the staple cereal of the Sahelian
Annex III - Page 13
Zone of Africa. In semi-arid West Africa, it accounts for about half the daily calorie 
intake and one-third of the protein for local people.
Average millet grain yields are only 500 to 800 kg/ha. Yields increased 
moderately during the 1970s (12%), showed a more modest increase of 10% during the 
1980s, and have stagnated during the early 1990s. World production of millets increased 
modestly (8%) during the 1970s (with a slight increase in pearl millet in West Africa), 
declined by 3% during the 1980s as yield increases (10%) were counterbalanced by 
declines in production area (12%), and stabilized during the early 1990s. In Africa, the 
area under production has increased by 32% since the 1960s, and this was accompanied 
by an 8% increase in yield. In Asia, the area decreased by 37%, but yields rose 38%, 
leading to a net decrease of 13% in production. Clearly, unless millet yields can be 
further improved and stabilized, the future for dryland food production in the semi-arid 
tropics will continue to look bleak. In India, ICRISATs efforts in pearl millet 
improvement have met with substantial success: over 3 million ha or a third of the area 
is now sown to improved pearl millet hybrids and open-pollinated varieties based on 
parental materials of ICRISAT origin.
The main constraints of pearl millet production are the same as for sorghum: 
environmental stress (especially drought and soil fertility), crop establishment, birds, 
Striga, diseases, and insect pests.
Recent research developments in cytoplasmic diversification of hybrid parents, 
alternative cultivar types such as topcross hybrids that exploit heterosis but reduce genetic 
vulnerability compared to single-cross hybrids, identification of molecular markers for 
genes controlling downy mildew resistance and components of grain yield all augur well 
for making pearl millet an even more reliable component of agricultural systems in the 
harsh environments of the semi-arid tropics. These technologies, together with 
participatory approaches to identifying the real needs of farmers and their families in 
these environments, will help the people of these regions help themselves to achieve a 
more sustainable and self-sufficient food production system.
In 1992, TAC recommended maintaining current efforts in millet research.
1.3. Roots, Tubers, Banana and Plantain
1.3.1. Cassava
Cassava is an important food crop in Africa, particularly in the humid and 
subhumid tropics. It is also important in parts of Asia and Latin America and the 
Caribbean. Besides roots, in Africa, the leaves are eaten as a green vegetable in some 
parts of sub-Saharan Africa and provide a cheap and rich source of protein and Vitamins 
A and B. The crop is grown mostly by small-scale farmers, for whom it is a major 
source of cash income and food energy. It tolerates low-fertility soils, drought and can 
be left in the ground as a food reserve for long periods. Cassava ranks among the 15
Annex III - Page 14
most important agricultural commodities in developing countries with respect to value of 
production, and is the most important in sub-Saharan Africa.
World production in 1994, all from developing countries, was about 152.5 
million t from about 15.8 million ha and a production of 159.1 million is forecasted for 
1995 - some 42.7% of the total area in developing countries devoted to root crops. This 
represents an increase in production of 19.6% and in harvested area of 11.8% during the 
past decade. Sub-Saharan Africa accounts for approximately 47% of world production, 
Asia for about 31%, and Latin America and the Caribbean for 20% (77.4% of this from 
Brazil. Currently Brazil, Nigeria, Indonesia, Thailand and Zaire are the world’s largest 
producers. Thailand is the dominant world exporter. In contrast, there is very little 
export from Africa where production is almost entirely used as food.
Cassava is the most important root crop in Africa, where it accounts for 59.2% 
of the harvested root crop area. It has three main roles namely, - a major source of cash 
income for those households which produce and/or process the crop, a low cost 
carbohydrate staple for low income urban and rural consumers, and a food security crop 
in vulnerable areas. It is used mainly in processed forms (about 70% of total production) 
such as meal or flour, while the remainder is used in fresh form. It is a major source of 
dietary energy for over 200 million people contributing an average of more than 200 
calories per day per capita. Nigeria and Zaire are the largest African producers, 
accounting for 55.8% (40.6 million t) of production. In Asia, there are many more end 
uses and all the principal producing countries have starch industries. In Thailand, cassava 
is produced largely for export as pellets for animal feed. However, the share of native 
and modified starch has been increasing significantly since the early 1990s. In Latin 
America and the Caribbean, cassava’s principal use is as food, but an increasing amount 
of cassava is being processed into cassava chips and especially into starches. From the 
early 1980s to the early 1990s yield gains were about 5.9% (from 10.2 to 10.8 t/ha) for 
the developing nations as a whole, 11.6% for Africa (from 6.9 to 7.7 t/ha), 5.5% for 
Latin America and the Caribbean (from 10.9 to 11.5 t/ha), and 3.1% for Asia (from 12.7 
to 13.1 t/ha). Current average yields for Thailand, Indonesia, India, and China are 14.0, 
11.9, 22.5, and 14.8 t/ha respectively, compared with an average 7.7 t/ha in Africa.
With real incomes increasing slowly or not at all in most sub-Saharan African 
countries, there seems likely to be an increasing demand for cassava as a human food, at 
least until the year 2010. Also, the crop has a special significance as a food reserve. 
Enhancement of all of cassava’s roles (income generation, cheap food, and food reserve) 
requires breeding for high yield, early bulking and processing qualities. In addition, 
improvements in processing technologies to achieve better quality and diversity of 
products, and to reduce processing labour are required, all of which will impact 
favourably on women. Besides the need to diminish the cyanogen content for safety 
reasons and to reduce processing and food preparation time required by women, cassava 
research in Africa should address the following principal issues: improvement of the role 
of cassava as a subsistence and famine relief crop; utilization of the crop’s potential for 
income and employment generation; and the generation of marketable surpluses with 
significant added value to meet rising urban demand for new products. Overcoming 
significant biotic and abiotic constraints are the major challenge in Africa. The successful 
strategy applied during the 1980s for the biological control of mealy bug is currently 
being extended to some of the crop’s other major pests.
Annex III - Page 15
Cassava production in Asia has increased at an annual rate of about 1 % during 
the past 25 years. Thailand is the region’s largest producer, with about 39% of total 
production, and has become an important exporter of starch and cassava chips and pellets 
for animal feed. The market in Asia seems likely to remain healthy due to the demand- 
led diversification of cassava’s end uses i.e. modified starches. There exists a strong 
demand, especially from the industrial sector, for more high-yielding clones with superior 
starch contents. In addition, increasing attention is being paid to improve yields while 
decreasing the risk of endangering the natural resource base. The latter is being 
addressed through a multi-country project that aims to reduce soil erosion and improve 
soil fertility in upland cassava systems through an FPR approach.
Cassava production in Latin America and the Caribbean declined from 1975 to 
1984 at an annual rate of 1.4%, especially in Brazil where it declined at an annual rate of 
2.0%. After 1984, production remained more or less stable, decreasing slightly in 
NE-Brazil (due to major droughts) while increasing in S-Brazil and Colombia. Cassava 
remains a small-farm crop grown in marginal areas where soil fertility and moisture limit 
the production of other crops. Nevertheless, cassava farmers sell a high proportion of 
their production. Urbanization has led to decreasing per caput consumption of fresh 
cassava, but the crop is increasingly being used in animal feeds through the intervention 
of small-scale farmer associations producing low-cost dried cassava chips. In addition, in 
most countries in the region cassava starch processing has become increasingly important.
As a result of sustained training efforts in several disciplines and at a range of 
levels, the national programmes in sub-Saharan Africa have steadily strengthened. In 
addition, strong and effective collaborative links have come into place between national 
programmes and the CGIAR system for the realization of common research goals, 
especially in the fields of biological control, plant breeding and cellular biotechnology. 
However, national research capacity is still limited and a continued strong research input 
by the CGIAR system is justified. The scope of research that is required is similar to that 
which TAC recommended in 1986, with some alteration in emphasis. Findings from the 
Collaborative Study of Cassava in Africa (COSCA) showed that increased commercial 
opportunities for cassava, drive production increases, and generating income for rural 
people. Therefore, future research should emphasize postharvest technology, quality of 
roots for various end-uses, pest and disease control and to a lesser extent foliage 
production for use as vegetable. These issues have remained a high priority. In addition, 
cassava market assessment needs to indicate opportunities for improved or novel cassava- 
based products. This information will serve technological interventions, strengthening the 
gradual transformation of cassava from subsistence toward market orientation.
In Asia, demand was buoyant and national programmes, though relatively young, 
were strong. The main requirements from the CGIAR System seemed to be improved 
germplasm and consultation services on technical problems, especially production 
agronomy. These observations in general are still valid. However, a strong cassava 
market diversification, away from cassava pellets, and towards (modified) starches has 
taken place in most Asian countries. This, together with changing government policies, 
has put further pressure on improved cassava varieties, especially in terms of improved 
starch content. Furthermore, the continuing shift of cassava production towards more 
marginal areas, emphasizes the need for R&D to maintain the fragile resource base.
Annex III - Page 16
Since 1986, several studies carried out in Latin America and the Caribbean have 
shown that cassava is increasingly being used in animal feeds. The rapid increase in 
demand for feed, coupled with the cereals deficit, suggests considerable future demand 
for dried cassava. In the absence of price distortions, cassava is highly competitive with 
cereal grains. In addition, new technologies and new product demand have significantly 
strengthened demand for industrial, fermented and modified starches throughout the 
continent. Also, while urbanization has led to a decrease in the per capita consumption of 
fresh cassava, pilot studies indicate increased demand on the part of urban dwellers and 
new "convenience food" cassava products. Overall, the major areas for market expansion 
for cassava in Latin America and the Caribbean are seen to be animal feed, refined flours 
and starches. A series of cassava industry and market studies are being conducted to 
guide applied cassava research in the area of postharvest, processing and product 
development. Significant efforts continue to be needed in further linking small-scale 
cassava farmers to growth markets following the successful concept of Integrated Cassava 
R&D Projects. In addition, cassava productivity research aims for integrated approaches 
to overcome the major biotic and abiotic stresses, ensure the environmental soundness of 
technology options and maintain the competitiveness of the crop.
In 1992, TAC recommended a continuation of CG efforts in cassava research.
1.3.2. Potato
Approximately 30% (about 89 million t) of the world’s potato crop is currently 
produced in developing countries, mainly by small-scale farmers, compared to only 15% 
two decades ago. Potato is a labour intensive crop. The nutrient value (including 
Vitamin C) of potato is high, and the crop is particularly useful as a source of energy and 
protein and as an infant weaning food. High yields «are possible, demand is growing 
rapidly due to positive income elasticity of demand for the crop at low income levels, and 
potato has a high value as a cash crop. It ranks among the ten most important food crops 
in developing countries with respect to gross value of production.
In 1992/94, developing countries accounted for about 37% of the area harvested. 
China is the largest producer, accounting for 42% of the 89 million t of potatoes 
produced in developing countries in 1989, while the rest of Asia accounted for 18%, 
Latin America and the Caribbean for 14%, West Asia-North Africa for 15% and 
sub-Saharan Africa for 3%. Yields vary from about 6 t/ha in sub-Saharan Africa to 18 
t/ha in West Asia-North Africa, compared with an average of 20 t/ha in developed 
countries. During the 1980s, yields in the developing regions as a whole increased by 
13%, from 10.9 to 12.2 t/ha. A further 6% increase was obtained in the first years of 
the 1990s, as average yields rose to 13 t/ha.
Among the major constraints to increased production are the high costs of 
production, various diseases and pests, the perishability of the crop during storage, and 
the difficulty of developing varieties adapted to higher temperatures. As in the case of 
other roots and tubers, national research capacity in potato research was generally weak at 
the start of CGIAR activities with this commodity. Only 2% of the world’s potato 
production is traded on international markets because of the perishability of the crop, 
whose high water content makes its transport over long distances risky. Quarantine 
regulations also restrict international trade in potato.
Annex III - Page 17
Potato has responded well to research, and plant breeding has already brought 
about significant improvements in the crop in developing countries. Virology research in 
the potato has advanced greatly, and the safe movement of germplasm is now a reality. 
The adoption of improved potato varieties is often delayed by the absence of national seed 
or multiplication systems. There is also a need for greater attention to the integration of 
potato in sustainable cropping systems.
In 1986, TAC recommended that the level of support for potato research should 
continue in the short to medium term, given the short history of research for tropical and 
subtropical regions. TAC further recommended that, in view of the stronger national 
programmes then beginning to emerge and the spillovers from research in developed 
countries, CGIAR support be reduced in the medium to long term. In 1992, TAC 
recommended maintaining CG efforts at current levels.
To sustain production increases in the future, a coordinated effort to develop 
more durable host plant resistance to potato late blight is required. Greater emphasis is 
also required, through molecular virology, in building immunity to the main viruses 
which are second to late blight in terms of their importance as constraints to potato 
production.
1.3.3. Sweet potato
Sweet potato is now widely grown as a staple food in developing countries 
outside tropical America, where it originated. Although sweet potato statistics are 
dominated by the production level of China (the world’s largest sweet potato producer 
accounting for about 80% of production), the crop is also grown in many small countries 
with typically very low income levels. Sweet potato has very little research history, and 
outside the CGIAR only very little research is conducted on the crop. It is well adapted 
to warm tropical lowlands and produces relatively well under low-input conditions on 
good soils. Depending on variety, the crop can be harvested in three to six months. 
Sweet potato fits well into the multiple cropping systems of Asia. The protein content of 
the roots is marginally greater than that of cassava and about half that of potato and yam. 
Sweet potato provides large shares of calories, protein and Vitamin C, as well as 
Vitamin A in yellow cultivars to the diets of the poor. When eaten as a vegetable, the 
green leaves provide additional protein, vitamins and minerals. Production costs and 
labour inputs are low in terms of the yield and calories produced.
Per caput production of sweet potato has decreased during the past 20 years, and 
the area harvested has also diminished. As income levels have risen the consumption of 
sweet potato has fallen. There has also been diversification in the uses made of sweet 
potato. For example, in China, only about 26% of sweet potato production is now used 
for human consumption, as against 35% for livestock feed, 28% for industrial uses 
(starch and alcohol) and 11% for seed or processed snacks.
Of the 9.1 million ha of sweet potato harvested on average between 1992 and 
1994 in developing countries, Asia accounted for 82%, sub-Saharan Africa for 15%, and 
Latin America and the Caribbean for 3%. About 131 million t of sweet potato are 
produced altogether, of which 98% is from developing countries. China dominates world 
production, producing over 85 % of developing country output, and this masks the
Annex III - Page 18
importance of sweet potato in many small countries such as the Pacific Islands. In terms 
of gross value of production sweet potato ranks eighth among the major agricultural 
commodities in developing countries.
The demand for sweet potato is increasing in sub-Saharan Africa, where the 
harvested area is relatively small. Production is estimated to have increased by 25% in 
the 1970s and by 13% in the 1980s, and is now 2.6 times higher than that of Latin 
America and the Caribbean, where it declined during the 1970s but increased by 9% 
during the 1980s.
Although current yields in sub-Saharan Africa average only 6 t/ha, the crop’s 
high yield potential has been demonstrated by the CGIAR System’s research in that 
region, which has led to varieties that can produce more than 40 t/ha in four months 
when grown in the wet season. Similar results have been obtained from new Asian 
varieties. Current yields in the developing regions as a whole average around 14t/ha, 
with an average yield of about 18t/ha in China. Substantial potential exists for an 
expansion of the importance of sweet potato and its foliage as a livestock feed.
Pests and diseases, such as the sweet potato weevil, stemborer, viruses and 
mycoplasma-like organisms, are major production constraints. Integrated pest 
management, including the use of transparence resistance customs, appears to show 
promise for the future. Unlike cassava, the crop cannot be stored in the ground beyond 
maturity, as it sprouts easily and is subject to pest attacks. Nor does it store well once 
lifted, although slicing and drying alleviate this problem to some extent.
In 1986 TAC considered sweet potato to be a neglected crop and recommended 
that the research effort be increased substantially. It recognized a need for greater 
collaboration between the CGIAR Centres and other institutions involved in research on 
the crop, such as AVRDC. The role of sweet potato in the development of new foods 
and food processing technologies could make it a highly valuable cash crop and 
employment generator in the medium to long term. In 1987, sweet potato was added to 
CIP’s mandate, and in 1990, AVRDC decided to stop further work on sweet potato.
In 1992, TAC recommended maintaining current CG efforts in sweet potato
research.
1.3.4. Andean Root and Tuber Crops
In 1993, CIP began a programme to reduce wild and domesticated Andean root 
and tuber crop species threatened by genetic erosion. Collection, preservation, and 
utilization of nine of the lesser known Andean root and tuber crops are currently 
underway in collaboration with national programmes, universities and NGOs in Ecuador, 
Bolivia, Peru, and Brazil. The initiative also supports in situ conservation by identifying 
microcentres of diversity and by monitoring the handling of RAT by farmers at four sites 
in Peru and one in Bolivia. In addition, the project is conducting research aimed at 
identifying and managing major disease constraints.
Annex III - Page 19
1.3.5. Yam
Yams are cultivated throughout the tropics, and in parts of the sub-tropics and 
temperate zones. They are of major importance in sub-Saharan Africa, and in the Pacific 
and Caribbean islands. Estimated world production is 28.1 million tonnes of which 95% 
is grown in sub-Saharan Africa, mainly in the West and Central regions, and in small 
amounts elsewhere. Virtually all production is used for human food. It is the second 
most important root/tuber crops in Africa with production reaching just under one third 
the level of cassava. Nigeria is the largest producer (about 20 million tonnes), but the 
crop is important wherever it is grown. By virtue of its excellent palatability, it is high 
value crop and, in spite of rather limited research attention, the popularity of this food 
crop never wavers. More than 95% (2.8 million ha) of the current global area under yam 
cultivation is in sub-Saharan Africa, where the crop accounts for about 21% of the area 
cultivated with root crops in the continent’s root-crop belt. Nigeria alone accounts for 
about 70% (16 million t) of the world production of yam.
Yam is a preferred food and a food security crop in some sub-Saharan African 
countries. The most intensive area of production is in West Africa, in the southerly part 
of the lowland moist savanna zone. It is also grown in certain parts of the forest zone 
and, over the past decade or so, it has gradually extended into lower rainfall areas of the 
savanna, using alluvial soils of inland valleys. Because it is highly appreciated when 
prepared as a fresh starchy food, it is marketed into non-producing areas thus employment 
in transportation and sales at urban and rural markets. Unlike some other tropical 
root/tuber crops (cassava, sweet potato and aroids), yam tubers can be stored for periods 
of up to four or even six months at ambient temperatures. This characteristic contributes 
to the sustaining of food supply, especially in the difficult (food scarce) period at the start 
of the wet season. In West Africa, the white yam - Dioscorea rotondata - is the most 
highly prized type and the one that has received most attention from the CGIAR System.
Yam production is limited by various diseases and pests. Nematodes cause 
serious damage both in the field and in storage. Postharvest losses also result from 
fungal and bacterial rots and insects as well as from increased respiration and sprouting 
when tubers break dormancy. A further hindrance to yam cultivation is high costs, which 
are a consequence of the heavy labour requirements at planting and harvest and sizeable 
expenditures on planting material (‘seed’ yams, in the form of whole small tubers or tuber 
pieces). The need for staking is another cost, but this is not a major limitation in the 
prime production ecology (the savanna) because high solar radiation obviates the necessity 
for staking. Production is carried out mainly with hand tools, and labour demands are 
high for planting, weeding, staking and harvesting. The cost of planting material is high: 
20-30% of the previous harvest. In sub-Saharan Africa, mean gross yields are 10 t/ha (7- 
8 t/ha net, after allowing for next season’s planting material).
International research efforts on yam are fairly recent and small, but results are 
promising. Within the CGIAR System, non-stake lines capable of producing 20 t/ha have 
been produced and new techniques for the production of planting materials should reduce 
the drain on harvests. These techniques have already led to a small seed-yam production 
industry among yam growers in Nigeria. Research has also found ways of triggering 
flowering, thereby allowing plant breeding to begin. In recent years, yam breeding has 
made progress in the development of improved cultivars which achieve stable high yields
Annex III - Page 20
and can even perform well under conditions of natural soil fertility with no staking. 
Yield potential of the most recent elite germplasm is in the range of 25-30 t/ha.
In 1986, TAC recommended that the effort on yams be increased to a level 
sufficient to make a rapid impact on production and postharvest problems. TAC viewed 
the increased efforts devoted to yam as a short-term thrust to determine whether the 
apparent breakthroughs in seed propagation and the development of non-staking varieties 
could make the anticipated impact on production in farmers’ fields.
In view of advances made in the last five years, IITA is implementing extensive 
regional testing of improved germplasm. Within the next five years, with the present 
level of multidisciplinary research input, new pre- and postharvest technologies will be 
available for evaluation with farmers in selected countries in sub-Saharan Africa, 
including countries where yam is now a minor crop.
In 1992, TAC recommended maintaining CG efforts, but asked that the next 
external review of IITA consider the future role of the CGIAR in yam research.
1.3.6. Banana and Plantain
Banana and plantain are staple food crops for millions of people in developing 
countries. About 90% of production takes place on small farms and is consumed locally. 
Only 10%, mainly from commercial plantations in Latin America and the Caribbean, 
enters world trade. In terms of gross value of production, banana and plantain rank 
eighth after rice, milk, beef, wheat, maize, soybean and groundnut.
Banana and plantain production is threatened by pest and disease pressures, 
which have been increasing the past 15 years. These include black sigatoka leaf spot 
disease, Fusarium wilt (Panama disease), banana weevil, a complex of plant parasitic 
nematodes and several virus diseases (banana bunchy top, banana mosaic, banana streak, 
and others). Black sigatoka disease causes such severe leaf necrosis that fruit yield 
decreases by 30-5). Weevil and nematode damage generally reduce plant vigour and 
increase susceptibility to wind lodging.
Proper management of soil fertility, including provision of soil organic matter 
through regular mulching, is essential for maintaining the perennial productivity of banana 
and plantain. In the humid lowlands of sub-Saharan Africa, the so-called ’yield decline 
syndrome’ of plantain is observed after one or two cycles of cropping in large-scale field 
plantations. Whilst the reasons for this yield decline are complex, the pressure on land 
and the associated shortening of fallow periods and decline in soil fertility exacerbate the 
problem. Similarly in the banana production systems of midaltitude and highland areas, 
declining yields are undoubtedly related to reduced soil fertility and mineral deficiencies 
arising from use of poorer soils. Postharvest losses of plantain and banana are a serious 
deterrent to expanding production in some countries. Surplus production during the main 
cropping season is the primary cause, but losses can also be attributed to poor methods of 
harvest, transportation and storage of the fruit.
Plantain and banana are generally considered intractable to genetic improvement 
due to their triploid nature which results in almost complete sterility. Nevertheless, in
Annex III - Page 21
recent years the CGIAR system and other regional plantain and banana improvement 
programmes have made excellent progress in breeding hybrids with resistance to black 
sigatoka, improved yields and acceptable fruit quality. In addition, research in cellular 
biotechnology and virus diagnostics have provided ways to achieve delivery of improved 
germplasm on the scale necessary for achieving impact with small scale growers. With 
respect to future research, it is now possible for the focus to shift away from control of 
black sigatoka disease to other biotic stresses (nematodes, weevil, virus diseases) using an 
integrated approach. In addition to high and stable yields, increased tolerance of drier soil 
conditions, improved plant architecture, including root systems, suckering behaviour, and 
reduced plant height, and postharvest requirements (preferred quality together with better 
handling characteristics) should also be objectives for genetic improvement programmes.’
The main challenges to research include breeding for resistance to Black Sigatoka 
disease, Fusarium Wilt (Panama disease), Bunch Top Virus and banana weevil, and the 
development of improved production systems.
In 1990, the CGIAR decided to extend its support for banana and plantain 
research beyond the humid and subhumid tropics of sub-Saharan Africa, to include Asia 
and Latin America and the Caribbean.
The main challenges to research include breeding for resistance to Black Sigatoka 
disease, Fusarium Wilt (Panama disease), Bunch Top Virus and banana weevil, and the 
development of improved production systems.
In 1990, the CGIAR decided to extend its support for banana and plantain 
research beyond the humid and subhumid tropics of sub-Saharan Africa, to include Asia 
and Latin America and the Caribbean.
In 1994, CG efforts in banana and plantain research were streamlined through the 
integration of INIBAP into IPGRI.
1.4. Food Legumes
1.4.1. Chickpea
The oldest records of the cultivated chickpea are from Turkey, and it is assumed 
that the crop spread out globally from that area. Generally the crop is grown on small- 
scale farms as both food and cash crop. The seeds are used whole, dehulled or as flour. 
Immature shoots and seed may be used as vegetables. In 1994, world production was 7.9 
million t from 10.2 million ha, of which 97% was from developing countries. For the 
1992 to 1994 three-year production average, Asia accounted for 76% of production, 
Africa 3 %, and Latin America and the Caribbean each for 3 %. Yields have shown a 
steady increase of about 0.5% annually over the past two decades. During 1971-73 the 
average yield globally was 644 kg ha1; during 1991-93 it was 705 kg ha1, and during 
1992-94 720 kg ha'1.
The small-seeded desi types, which account for about 85% of world production, 
are grown on the Indian subcontinent, in Ethiopia, Australia and in parts of Mexico,
Annex III - Page 22
Afghanistan and Iran. The large-seeded kabuli types are grown in the Mediterranean 
region, parts of Mexico, and to some extent on the Indian subcontinent. In the tropics and 
subtropics with summer rainfall, chickpea is mostly grown on residual soil moisture or 
sometimes under irrigation. In the subtropics with winter rainfall, the crop is generally 
sown during the spring. It usually receives few inputs other than labour, insecticides and 
seed.
Chickpea is an important dietary item in South East Asia, India and the West 
Asia-North Africa region and Ethiopia. The protein content of the seed is about 20%. The 
average yield for all developing countries was about 720 kg/ha, but the Central American 
yield is almost twice as high, and experiments in India with limited irrigation have 
produced yields of over 5 t ha'1. Changes in yield and production reflect climatic factors 
and changes in agronomic practices and varieties. The area harvested globally has 
remained stable (around 10 million ha).
Consumption has followed production, and it is expected that demand may 
increase with population in India, though at slower rate, and in West Asia-North Africa, 
where chickpea is consumed by all income groups. In developed countries, consumption 
is increasing rapidly.
The major constraints to production include disease susceptibility of local 
varieties, environmental stresses, drought, diseases, pests and poor crop management. 
CGIAR efforts have already produced significant results, notably the combination of 
blight resistance and frost tolerance, which has enabled winter sowing and a potential 
doubling of production in the low-elevation areas of West Asia-North Africa region. This 
has led to potential yield increases of 50 to 100%. A break-through has been achieved by 
the breeding of wilt resistant, extra-short duration varieties that can grow under the 
harshest conditions in Eastern Africa and Southern Asia where soil-borne diseases can be 
devastating and drought escape adds a main stability factor to the crop. Analysis and 
resolution of the "wilt complex" has enabled more targeted improvement. Higher 
yielding, disease and pest-resistant lines have been made available by breeders. New, 
more effective strains of rhizobia have been identified, leading to increases in nodulation 
and biological nitrogen fixation. Sources of resistance to biotic and abiotic stresses have 
been identified from the annual wild Cicer species and efforts are nearing completion to 
transfer genes for resistance to cyst nematode and cold in chickpea.
There is active research collaboration with national programmes and advanced 
research institutions. Research in West Asia - North Africa has focused on increasing 
productivity and stability through varietal improvement for disease and cold resistance, 
and the development of better production technology. To upgrade the level of resistance 
in varieties for ascochyta blight, use of molecular marker technology has shown potential 
and will have to be increasingly used. Tissue culture techniques offer opportunities to 
enable crossing of chickpea with currently non-crossable Cicer species to unlock and use 
large variability for useful traits. Current research points to the possibility to increase 
seed yield through increased shoot biomass and manipulation of the crop phenology. 
These will have to be further explored. Emphasis will have to be laid on the developing 
cultivars for adaptation to specific niches in different cropping systems, for which 
increased decentralized breeding and participatory approach will have to be used. For 
Eastern and Southern Africa, research into drought and soil-borne diseases and increased
Annex III - Page 23
productivity has resulted in recent releases of high yielding, disease resistant varieties of 
both the desi and kabuli type in suitable agroecological areas.
1.4.2. Cowpea
Cowpea is widely grown in the warm semi-arid and subhumid regions of 
sub-Saharan Africa and locally important in the Caribbean Islands, Brazil, PDR Yemen, 
the Indian subcontinent and southeast Asia. About 80% of the production in Africa is in 
West Africa, with Nigerian production accounting for about 70% of the cowpea in West 
Africa.
Cowpea is usually grown by subsistence farmers and in mixtures with maize, 
sorghum, millet and cassava. It is a pivotal crop for enhancing sustainability of cropping 
systems because it fixes large amounts of nitrogen for its own growth with residues 
returning to the soil, is quick growing and produces an excellent ground cover to reduce 
soil erosion, and has the capacity, in the case of some cowpea varieties, to cause "suicidal 
seed germination" of the parasitic plant Striga hermonthica that attacks cereal crops, often 
with devastating effects. The dry seed is an important source of Vitamin B and protein 
(22% edible protein) and provides an estimated 6.5% of total protein consumed in semi- 
arid West Africa. Cowpea leaves are a preferred vegetable and an excellent source of 
protein in many areas of Africa. Cowpea haulm is also an important source of livestock 
feed.
Average yields in developing countries are about 240 kg/ha. Some countries 
have made progress in the release and adoption of improved varieties, e.g., Ghana, and 
the benefits of this are evident in national statistics (Ghana National Statistics: 1977-79, 
mean yields 300 kg/ha compared with 600 kg/ha in 1987-89, and a continuing upward 
trend is reported. However, the best short- to medium-duration varieties so far developed 
can yield 2,500-3,000 kg/ha in field conditions on research stations, and short-duration 
varieties can achieve over 2,000 kg/ha in 60-90 days. The major constraints to farm 
yields are three insect pests, flower thirps, Manica pod borer and pod-sucking bugs.
Only low levels of resistance have been found in the cowpea germplasm for each of these 
three insect pests.
In 1994-95, a very efficient method was developed to regenerate and produce 
transformed cowpea plants. This major research breakthrough has made it very probable 
that the CGIAR system will make significant progress in developing cowpea varieties with 
good levels of resistance to these three insect pests. In addition, progress in research on 
biological control of flower thrips, based on natural enemies, indicates that this 
technology may also be a feasible control measure.
The reduction in damage from insect pests is expected to increase average 
cowpea grain yields in West Africa and other cowpea growing regions by at least 100%. 
In marginal areas (the Sahel) where the rainfall is 350 mm per year or less and often soils 
are very poor, the damage caused by insect pests is much less than in more humid areas. 
Recent results have shown that relatively high grain yields are obtained from breeding 
lines which have been improved for tolerance to drought and heat. Future research will 
seek to continue to develop varieties for these ecologies, combining tolerance to drought 
and heat, with improved P use efficiency.
Annex III - Page 24
In 1986 TAC recommended that the resource allocation to cowpea be maintained 
for the medium term, but with an expansion of efforts in tropical America and Asia. The 
factors leading to this recommendation were: the importance of cowpea as a subsistence 
crop in sub-Saharan Africa; its qualities of genetic diversity, fast maturation, wide 
environmental adaptability, resistance to drought, ability to fix nitrogen, and easy 
placement in cropping systems which, if exploited, could make it the most valuable of the 
pulses in the semi-arid to subhumid tropics; the potential value in other regions of a short- 
duration legume; the rapid growth occurring in production and consumption; the already 
promising results emerging from cowpea’s short research history; and the limited capacity 
of national research programmes. TAC also recommended that research supported by the 
CGIAR System continue to concentrate on increasing yields and their stability and on 
improving management practices.
While all of the factors cited to support the TAC recommendations of 1986 
remain valid, the important role of cowpea (especially semi-determinate and spreading, 
dual purpose types) in the sustainability of crop-livestock production systems in marginal 
environments is now better understood and increasingly emphasised by national 
programme scientists (eg refer to the Second World Cowpea Research Conference, 1995 - 
Recommendations and Resolutions). Thus, in future research on this crop, the CGIAR 
system could make major contributions, not only in the needed area of insect pest 
management, but also in optimising the contribution that cowpea can make to resource 
management.
1.4.3. Broad (faba) bean
Faba Bean is a spring crop in temperate regions and a winter crop in subtropical 
regions with mild winters. It is grown at high elevations in tropical and subtropical 
regions. Two main groups exist: small-seeded types, found in Egypt, Sudan, Ethiopia, 
Eritrea and Afghanistan; and large-seeded types, found in other parts of West Asia-North 
Africa.
Developing countries account for approximately 90% of the global production of 
3.8 million t. (FAO 1994 Production Yearbook). Of the developing country share, China 
accounts for 62%, West Asia-North Africa 17.6, Africa 7.5%, and Latin America and the 
Caribbean 3.7%. The protein content is high (25% of edible portion, and faba bean is a 
popular food in West Asia-North Africa, though it provides 9.9% of the region’s protein. 
Faba bean is also a source of Vitamin B. Developing country yields of mature seed 
average 1,300 kg/ha, more than double that of many other pulses (FAO 1994 Production 
Yearbook). It is estimated that about 20% of the crop is consumed green and is not 
accounted for in production estimates. Demand is likely to increase as population rises: 
faba bean is a preferred pulse in North Africa and parts of West Asia, and provides 
variety to diets elsewhere. The crop is important in rotation in low monetary-input 
agriculture because of its high biological nitrogen fixation (120 by N/ha) and beneficial 
residual effect for subsequent cereal crops.
The constraints to production include: diseases, the parasitic weed, Orobanche, 
field and storage pests, poor crop management, and soil salinity in some areas.
Annex III - Page 25
In 1986 TAC recommended that CGIAR support for faba bean research be 
phased out for the following reasons: the crop is not important globally; China, the 
largest producer, has a strong national programme; and there are only 1 million ha under 
the crop in other developing countries, excluding China. The CGIAR was advised to only 
support the conservation and management of faba bean germplasm collections.
In accordance with these instructions, ICARDA relocated its faba bean 
programme to Morocco in August 1989 and had transferred the programme to the 
Moroccan national programme by the end of 1992, which was provided funding by a 
BMZ special project for a regional network on faba bean for Maghreb countries led by 
Morocco. There has been limited progress in this initiative. Consultation with the national 
programmes of the countries where faba bean is an important crop in the farming system 
has revealed the need for continued crop improvement efforts on this crop in the CGIAR 
system. The external programme review of ICARDA in 1993 emphasized the fact that the 
devolution of faba bean improvement work to NARS was premature and the NARS 
wanted a review.
According to ICARDA, it is necessary that CGIAR should continue to support 
faba bean improvement research for the following reasons:
1. The China national programme is not as strong as was asserted by the TAC 
commission on crop priorities. It’s strength was greatly overestimated. ICARDA 
has just become involved in an ACIAR supported trilateral project including 
NSW Agriculture, China and ICARDA. Die primary reason for instigation of 
this project was the request of the Chinese national programme of an urgent 
need to strengthen faba bean research in the country.
2. The Morocco national programme has asked ICARDA to become involved in 
strengthening its faba bean programme. The programme is not strong enough to 
work without backup support. Also, BMZ has sent feelers for ICARDA to 
become reinvolved in the BMZ network project for Maghreb in backstopping the 
North African Faba Bean Improvement Programme because of continued 
weakness of NARS in faba bean germplasm 
and crop improvement research.
3. National programmes both within and outside WANA have been asking for 
backup support in faba bean improvement and for providing nurseries of 
improved germplasm for their use.
4. There has been a realization also throughout Europe and Australia, that, with the 
largest germplasm collection of faba bean in the world, ICARDA is in a unique 
position to promote globally the genetic improvement of faba bean. The second 
International Conference on Cool Season Food Legumes held at Cairo, Egypt in 
1992 unanimously recommended this.
In the past four years the ICARDA Genetic Resources Unit has collected faba 
bean germplasm in Morocco, Tunisia, Baluchistan in Pakistan, Bangladesh and Nepal. 
The past year work was initiated on regeneration of the germplasm collections with 
support from GRDC, Australia. This year this Australian support has allowed disease
Annex III - Page 26
screening to be restarted along with the preliminary evaluation of the germplasm for 
agronomic traits. ICARDA is requesting TAC to consider reestablishment of a fully 
supported faba bean improvement programme at ICARDA to meet the requirements of the 
region and the global mandate of ICARDA for faba bean.
1.4.4. Lentil
Global lentil production is growing rapidly. It has risen by 110% from 1.3 
million t in the period 1979-81 to 2.8 million t in the period 1992-94, due to a fifty 
percent increase in area to 3.38 million ha and an increase in productivity of 38% from 
600 kg/ha to 820 kg/ha. Developing countries account for 87 % of the world lentil area. 
The major producing regions are Asia (58% of the area) and WANA (37% of the acreage 
of developing countries). Lentil is the most important pulse in Bangladesh and Nepal, 
where it makes a large contribution to the diet. The above expansion in production and 
productivity in Asia has come mainly from India, Iran, Nepal and Turkey. Other 
significant producers in the developing world include Argentina, China, PDR Ethiopia, 
Morocco, Pakistan and Syria. The expansion of production is fuelled by the rising 
demand of an increasing population and this trend will continue.
The crop is important for its use as a pulse and as a small ruminant feed. In the 
drier areas of West Asia and North Africa lentil is a key component of the traditional 
farming systems integrating barley, small ruminants and lentil.
For the Mediterranean Basin, research emphasis at ICARDA was initially on the 
development of harvest systems to reduce the high cost of production from a hand 
harvest. Such systems, involving cultivars with better standing ability and height, 
flattened seed beds and cutter bars, have and are being transferred to NARS in West 
Asia. Significant on-farm adoption of improved technology has occurred in Egypt, 
Ethiopia, Iraq, Lebanon, Sudan, Syria and Turkey. Vascular wilt is the most important 
disease of lentil and resistance is now being exploited for disease control. To control the 
damage of Sitona weevil to lentil nodules, genetic engineering will be employed to 
transfer a gene for toxin production into the lentil roots.
In the highlands of West Asia, lentil is usually spring sown, but sources of 
winter hardiness are under exploitation. As yield increases of above 50% have been 
realized from early winter sowing, the focus is now on exploiting this gain on-farm.
In South Asia, a network of researchers is targeting two production systems 
which offer scope for a major expansion in production: the relay sowing of lentil into rice 
paddy and the development of early lentil to sow after long season rice. Combined disease 
resistance is being incorporated into genetic material for both systems. The first lentil 
cultivars with combined disease resistance are poised to impact on-farm production in 
Bangladesh and Pakistan.
Lentil production, productivity and demand are rising rapidly in the developing 
world and ICARDA has made a positive assessment of the potential pay-off from planned 
research. Consequently, ICARDA will maintain and manage the genetic resources of 
lentil, continue to address the role of lentil in the farming system and focus improvement 
research to complete the above research agenda.
Annex III - Page 27
1.4.5. Phaseolus Bean
Phaseolus bean, or common bean, is the world’s most important food-legume. 
Common beans are grown in two forms, as dry beans and snap beans (the green pods are 
consumed as a vegetable). Global production of dry beans is estimated to be 18 million 
metric tons annually, with a market value of US$ 10.7 billion. Dry beans accounts for 
57% of the world’s food-legume production, having twice the production and market 
value of chick peas, the next leading food pulse. Another 3 million metric tons of snap 
beans are also produced annually. Nearly 80% of dry bean production occurs in the 
developing countries on small-scale farms.
Latin America, the centre of Phaseolus domestication, produces nearly half the 
world’s supply of dry beans. Brazil, Mexico, and Central America are the major 
producing regions in this continent. FAO production statistics for 1990-1995 show that 
dry bean production in Latin America is increasing by 3.4%. Most of this increase is due 
to higher yields (4.0%) as area under production shows a slight negative trend (-0.7%). 
The adoption of improved bean varieties appears to be a factor attributing to the higher 
yields and to the slowing down of bean expansion into marginal areas.
Africa is considered to be a secondary centre for bean genetic diversity. About 3 
million hectares of beans are planted annually in eastern, central and southern Africa, 
usually as mixtures of varieties. FAO statistics for 1980-1995 show that bean production 
is increasing at 1.2%, significantly below population growth rate. To meet future 
consumption demands by the year 2000, an average yield increase of 300 kg per hectare 
will be needed. Meanwhile, declining bean production and consumption is contributing to 
an overall deterioration of human nutritional trends in these regions.
Beans are nutritionally rich, especially in protein and iron, and are a good source 
of dietary fibre and complex carbohydrates. Given their nutritional quality and high 
consumption levels, beans make an important contribution to human nutrition, especially 
for poor consumers. In addition to high quality protein, a single serving (1 cup) of beans 
provides at least half the USDA-recommended daily allowance of folic acid (a B vitamin 
that is especially important for pregnant women) and 25-30% of the daily recommended 
iron levels. Similarly, the same serving of beans provide 25 % of the daily requirements 
of magnesium and copper, and 15% of potassium and zinc. Furthermore, the presence of 
beans in the diet significantly increases the utilization of maize and rice proteins due to 
complementarity in amino acids.
Beans play a very important role in human nutrition in the eastern Africa 
highlands, Mexico, Brazil, and Central America. In eastern and southern Africa, beans 
are the second most important source of protein after maize, and the third most important 
source of calories after maize and cassava. In Latin America beans are ranked forth as a 
protein source, similar in overall importance to milk and beef, and sixth as a source of 
calories, exceeded by such staples as cassava, potato and beef.
Consumption of beans is high in large part because beans are a relatively 
inexpensive food. In Brazil, the world’s largest consumer of beans, the cost of calories 
from beans is less than from rice or cassava, only maize is a cheaper source. Beans are
Annex III - Page 28
the cheapest source of calories and protein in Uganda and Rwanda, and the cheapest 
source of protein in Tanzania.
In Latin America and Africa, the demand for beans is tied to food markets, both 
rural and urban, and small-scale bean farmers are increasingly producing for the market. 
As women are the primary producers of beans in many regions of Africa, bean marketing 
represents an important source of income for the family.
Beans are not, however, simply a food for the poorest of the poor. Per capita 
bean consumption in the United States rose from 2.6 kg/year in the period 1976-1978 to 
3.4 kg/year in 1991 according to USD A data. Increased consumption reflects a nutritional 
move away from high-fat animal products to low-fat, high-fibre products like beans.
Major research options for improving bean productivity in Latin America and 
Africa have focused on public sector breeding efforts. There is little private sector 
interest in bean seed production outside Argentina, Brazil and the United States. 
International research on beans at CIAT has traditionally concentrated on improved 
resistance to diseases and pests, and more recently on tolerance to drought and low soil 
fertility, and improved yield potential.
Improved, CIAT-based varieties are grown over 800,000 ha, most of which is in 
Latin America where CIAT has been working longest. Nearly 40% of the bean 
production area in Central America is now planted to varieties resistant to the devastating 
bean golden mosaic virus. In Brazil more than 200,000 ha of CIAT-based new bean 
varieties are being grown in four states alone. Even in Africa, with more recent CIAT 
involvement, nearly 45 % of the Rwandan farmers (primarily women) were growing 
improved climbing bean prior to the civil war.
In recent years, international bean research led by CIAT began to seek non- 
genetic solutions to difficult production problems, in addition to the development of 
improved varieties. Pilot integrated pest management studies in the Andean region of 
South America has been successful in reducing pesticide applications by more than 50% 
in targeted regions. The use of climbing beans in Rwanda, and more recently in Uganda 
and Kenya is giving greater support to agroforestry efforts as stake production has 
become a new market business in these regions. In part of Uganda and Kenya, farmers 
are seeing advantages in use of green manure fertilizers to combat not only declining soil 
fertility, but associated problems with root rots and bean stem maggots. Additional work 
in the Andean region addresses problems of land degradation and soil nutrient depletion 
through building on existing farmer knowledge and using farmer participatory approaches.
In the area of building national programme capacity for bean research, CIAT 
pioneered a strategy for grouping countries into regional research networks to facilitate 
the development and transfer of new technologies in a more efficient and economical 
manner. The first regional network (PROFRUOL) was begun in 1978 in Central 
America, followed by three networks (RESAPAC, EABRN, and SADC-Beans) in Africa 
(during 1984-1986), and in the Andean region of South America (PROFRIZA) in 1987. 
The initial coordination of the networks was managed by CIAT, but as the networks grew 
stronger and regionally established, the coordination responsibilities were devolved to the
Annex III - Page 29
region. By mid-1996, all the regional networks will be locally managed. CIAT remains 
as a full research partner in the networks.
1.4.6. Pigeonpea
Pigeonpea is widely grown by subsistence farmers in the warm semi-arid and 
sub-humid tropics. It is often grown on poor soils and with few inputs. It is an important 
food in India, and is popular in parts of East Africa and Central America. The seeds are 
used whole, dehulled or as a flour; and in the Caribbean and South America, immature 
seeds and pods are used as a vegetable. The woody stem is valuable as firewood, thatch 
and fencing, and the leaves are an important source of nitrogen for the soil.
World production of dry seed is about 2.7 million t, most of which is grown in 
developing countries. The crop is an important source of protein (20% of mature seed) 
and Vitamin B. India accounts for about 91% of world production, followed by sub- 
Saharan Africa (6%). The remainder comes from Latin America and the Caribbean, and 
from Asia. There is limited international trade. Agro-industry has developed for canning 
green pigeonpea for export from Ecuador and Dominican Republic, and for dehulling and 
split pea (dhal) production for export from Malawi and Kenya.
Average developing country yields are about 700 kg/ha, but vary from 500-600 
kg/ha in central and southern India, sub-Saharan Africa and Asia to 1000-1, 200 kg/ha in 
northern India and Central America when the crop is grown as a sole crop. The main 
production constraints are variable yields associated with abiotic stress, diseases and pests 
and subsistence production conditions. The crop’s potential for wider use in semi-arid 
areas with high temperatures and poor soils is considerable, as a complement to phaseolus 
bean or cowpea in the drier and more marginal areas of Eastern and Southern and 
equatorial Africa, and Central America. Countries in Asia and in Eastern and Southern 
Africa have shown an active interest in exploiting pigeonpea’s multipurpose potential in 
farming systems where drought and heat tolerance are important considerations.
Globally, traditional pigeonpea farming systems have developed around medium- 
and long-duration cultivars (maturity in 180-280 days), often inter-cropped or mixed with 
cereals such as maize, sorghum and pearl millet. This cropping system has contributed to 
the sustainability of farming systems and to intensification of land and moisture use in 
rainfed areas. ICRISAT has collaborated with the national programmes in collection, 
characterization, and conservation of the biodiversity of landraces and wild related 
species. Source of disease resistance and tolerance to insect pests have been identified, 
and are being used by the NARS for genetic enhancement of medium- and long-duration 
pigeonpea. The potential of this material for agro-forestry and alley cropping on 
degraded land resources, and for crop-livestock systems, is clear.
Development of short-duration (100-150 days to maturity) and short-statured 
pigeonpea types by ICRISAT in collaboration with the NARS has greatly broadened the 
adaptation of pigeonpea into new production environments, and significantly increased 
productivity per unit area and time. Recent adoption of this material in the intensive rice- 
wheat cropping system areas, is contributing to the sustainability of the cereal 
productivity. There has been widespread adoption of disease resistant short-duration 
cultivars in drought prone area farmers in peninsular India, and there is significant
Annex III - Page 30
potential in other countries. Development by ICRISAT of the first hybrid pigeonpea 
variety and hybrid seed production technology has led to release of hybrid cultivars by the 
private and public sectors in India. Improved technology for hybrid seed production of 
cytoplasmic male sterility is in development by a collaborative research network with the 
NARS and private industry. Introgression of genes from medium and long duration 
landraces, and from the wild relatives using transformation, embryo rescue and tissue 
culture, is designed to broaden the genetic base and resolve the major biotic and abiotic 
constraints of short duration pigeonpea. Active research collaboration with the national 
systems is used to achieve spillover of technologies and materials internationally.
1.4.7. Soybean
Soybean was originally domesticated in China, and is now cultivated throughout 
East and South-East Asia, the Americas (particularly the USA and Brazil) and to a very 
limited extent in sub-Saharan Africa and West Asia. In the northern hemisphere, its 
cultivation now extends from the tropics to 52°N.
Soybean has high protein (38%) and fat (18%) contents. The crop’s main use is 
for oil and protein products in the food industry. The residue after oil extraction is used 
for flour, protein products and animal feed. Although soybean is an important food crop 
and an inexpensive source of protein and Vitamin B in East Asia, efforts to introduce it as 
a food crop elsewhere have met with limited success. However, it is gaining importance 
in many parts of sub-Saharan Africa. For example, in Nigeria, the use of soybean flour 
has become very popular. It is used to fortify traditional foods and in most dishes requires 
no additional labour or cooking time. The final products are highly acceptable and the 
women who have been exposed to the appropriate methods to incorporate soybean into 
traditional foods now use it almost every day. This has had a major impact on the 
nutrition and health of children in both rural and urban areas. While Nigeria currently 
serves as the best example in Africa of a country where soybean is commonly used as a 
food, it is also gaining in popularity in Uganda, Zambia, Cameroon, Benin, Ghana and 
Cote d’Ivoire. In these countries today, soybean is no longer considered to be a crop that 
is unpalatable and difficult to process.
About 56% of the global area harvested is in developing countries. Tropical and 
subtropical South America produces 63% of the developing country share (61% of this 
from Brazil, which has a large export trade), China 24%, temperate South America 14%, 
and southeast Asia 4 %. The crop provides nearly 5 % of protein consumption in China 
and southeast Asia. Its fat contribution to diet is 20% in Brazil, 6-7% in China, India 
and Thailand, and 4-5% in Indonesia. Latin America and the Caribbean produces 
36 million t of soybean annually. In the past decade the region’s area under soybean 
increased by about 1.4% per year while yields increased at 2%, reaching 2.1 t/ha (more 
than the world average of 2.0 t/ha). Among the major constraints limiting production in 
Latin America and the Caribbean are acid soils, aluminium toxicity, photoperiodism, and 
pests and diseases. Yields vary considerably, from 1.1 t/ha in sub-Saharan Africa to 
2.1 t/ha in Latin America and the Caribbean.
Demand for oilseeds in developing countries is expected to grow at 3.7% 
annually until the year 2010, and production will need to increase accordingly. In
Annex ni - Page 31
sub-Saharan Africa, vegetable oil is already in short supply, and several countries of the 
region imported substantial quantities of both soybean cake and soybean oil during the 
1980s. Most countries in Latin America and the Caribbean also have a deficit in 
vegetable oil. Furthermore, soybean has substantial potential as a source of livestock 
feed, particularly for poultry. In sub-Saharan Africa locally produced soybean is used as 
a raw material in large and medium scale oil mills especially in Zimbabwe, Zambia, 
Uganda, Nigeria, Ghana and Còte d’Ivoire. Most of these mills began using soybean 
during the past six years.
Soybean research has been under way for some time outside the CGIAR System, 
with AVRDC working on both vegetable and grain type soybean breeding and production 
with a focus on Asia and INTSOY working on processing and utilization. The System’s 
own work is based in sub-Saharan Africa, and has progressed well in both crop 
improvement and postharvest processing and utilization. Breeding lines have been 
developed, tested by NARS, released and are now being grown by farmers in Nigeria, 
Ghana, Zaire and Uganda. These varieties have the ability to nodulate with naturally 
occurring rhizobia, improved seed longevity and improved levels of resistance to pod 
shattering and the major pests and diseases.
The main objectives of future research related to crop production is to develop 
soybean varieties that give a maximum contribution to the productivity and sustainability 
of the cereal-based cropping systems of the moist savannas of Africa. Major traits which 
are under improvement are the ability to cause "suicidal germination" of seed of Striga 
hermonthica, nitrogen fixation and phosphorus use efficiency. Cropping systems research 
will be conducted to develop appropriate technologies to increase productivity and 
sustainability including resistance to pests, diseases and pod shattering. Another objective 
of future research is to develop a better understanding of the relatively new disease, Red 
Leaf Blotch, that is found only in Africa and to the develop appropriate disease 
management strategies.
The major thrust of the processing and utilization research in the future is to 
work with more African NARS using the successful research and technology transfer 
methods that were developed with NARS in Nigeria.
In 1986 TAC recommended that research support for soybean be increased, with 
efforts continuing to focus on sub-Saharan Africa, and this is still justifiable. The needs 
of Asia and Latin America and the Caribbean were being successfully met by strong 
national programmes. This recommendation was based on: the crop’s importance, given 
increasing oilseed demand in sub-Saharan Africa and globally; the high level of interest in 
and apparent potential for the crop in sub-Saharan Africa; the high pay off from the 
modest research effort to date; and the excellent potential for developing solutions to 
some of the more important production problems in the tropics. In 1992, TAC again 
recommended that higher priority and more resources be allocated to soybean research in 
the CGIAR.
Annex III - Page 32
1.5 Oil Crops
1.5.1. Coconut
The coconut palm is a pan-tropical crop, grown on approximately 9.3 million ha 
in 82 countries. Many of the producing countries are small islands in the Pacific and 
Indian Oceans and also the Caribbean. Coconut is both their primary subsistence crop 
and their only significant source of export earnings. There are few, if any, alternative 
crops which can substitute for coconut in these countries. Coconut is the major tree-crop 
component in several agroforestry systems throughout the world, although its wide use in 
home gardens is probably not reflected in official production statistics.
At least 96% of the total world production of coconut comes from smallholdings 
with about 70% of the crop consumed in the producing countries. Coconut can be grown 
in harsh environments such as atolls and tolerates high salinity, drought and poor soils. It 
plays an important role in sustaining often fragile ecosystems in island and coastal 
communities and is used as a source of food, drink, fuel, animal feed and shelter. It is 
also a cash crop, used to produce many items for sale at either the local, national or 
international level. The main internationally traded products are copra, coconut oil, copra 
meal and desiccated coconut.
In 1986 TAC identified coconut as a priority commodity for support through 
international research. Following that, the CGIAR requested TAC to explore the 
desirability of establishing an international research initiative on coconut, and the form 
such an initiative might take. Subsequent studies undertaken on behalf of TAC identified 
several constraints and opportunities in coconut production which could be addressed 
through this research effort.
There are four major constraints to increased coconut production in developing 
countries: the low productivity of many coconut trees, due to old age and poor nutrition; 
the failure of many replanting programmes; fluctuating productivity due to variable 
environmental conditions; and inefficient handling and processing, with low farm-gate 
prices to smallholders. The productivity of the crop can be increased by the use of 
locally adapted, high-yielding, pest- and disease-tolerant varieties in replanting or new 
planting schemes. To increase the productivity of existing plantations it would be 
necessary to apply better agronomic practices, including the control of diseases, insects 
and weeds and also the use of fertilizers, plus identify and promote profitable and 
sustainable intercropping systems. Furthermore, there is a need to develop improved 
methods of handling and processing coconut, and to further diversify the coconut products 
traded.
Coconut breeding in several countries over the past 30 years has demonstrated 
that several improved varieties and hybrids are capable of yielding up to 3-6 t 
copra/ha/year under favourable conditions and this could potentially be tapped to increase 
the average world yields of 0.5 t/ha/year. Progress has also been made in identifying the 
causal agents of diseases of previously unknown etiology, such as cadang-cadang disease 
in the Philippines and lethal yellowing disease in the Caribbean and further nutritional 
studies have shown that coconut responds well to fertilizer application, particularly
Annex III - Page 33
potassium and chloride. Intercropping and the grazing of cattle under trees have shown 
that the total productivity of coconut lands can be improved, without threatening the long­
term sustainability of the system.
These findings suggest that a well organized and adequately funded international 
research effort could yield a high pay off. The long-term nature of coconut research and 
the likely benefits to smallholder producers, make coconut particularly suitable for an 
international research initiative. The TAC/CGIAR identified priority research areas for 
such an initiative to be: germplasm conservation and improvement; disease and pest 
control; sustainability of coconut-based farming systems; postharvest handling and 
processing; and the socioeconomics of coconut production. It is to be noted, however, 
that in a number of countries, research on coconut is funded by the private sector through 
levies on producers.
To enhance the sustainability of coconut production, there is an urgent need to 
promote the effective conservation of coconut genetic resources and their efficient 
utilization in breeding programmes. Thus among the five priority areas for international 
collaboration, germplasm conservation and improvement was identified as the research 
activity to be initially supported.
Upon the recommendation of 15 coconut-producing countries, and the support of 
the TAC/CGIAR and its donors, in 1993 IPGRI established the International Coconut 
Genetic Resources Network (COGENT) as part of its programme. The main objective of 
COGENT is to strengthen national programmes in the conservation and utilization of 
coconut genetic resources and establish the foundation for collaboration on the broader 
aspects of coconut research and development. The networking approach was selected 
because it reduces duplication of work, encourages sharing of limited resources and 
promotes complementation and synergy of research activities in national programmes and 
advanced research institutions.
The current research priorities of COGENT include: the development of an 
international coconut genetic resources database to enhance dissemination of genetic 
resources data to breeders worldwide; collecting to secure germplasm in areas which are 
threatened by genetic erosion and to fill up gaps in national collections; conservation in 
national and regional field genebanks; germplasm evaluation to identify suitable varieties 
for farmers; development of other complementary conservation methods and molecular 
methods for assessing genetic diversity and promoting safe germplasm movement. To 
complement these efforts, a strong training programme is being pursued to increase the 
number and upgrade the skills of researchers in national programmes.
The future research priorities include: application of research results to promote 
efficient genetic diversity assessment, safe germplasm movement, and effective 
conservation and exchange; plus strategic utilization of coconut genetic resources to 
support the development of varieties and hybrids with high productivity and adaptation to 
biotic and abiotic stresses. In the near future, research projects will also address users’ 
perspective to promote multi-purpose uses and increased competitiveness of the coconut 
and the gender issues related to sustainable germplasm conservation and utilization.
Annex III - Page 34
1.5.2. Groundnut
About 21.7 million ha are cultivated to groundnut in the world, of which 13.8 
million ha are in Asia (India 8.5 million ha; China 3.6 million ha), 6.8 million ha in sub- 
Saharan Africa, and 0.5 million ha in North and Central America. The production area 
in WANA is less than 100,000 ha. Groundnut is grown under a wide range of 
environmental conditions in areas between 40°S and 40°N of the equator. Most of the crop 
is produced where average rainfall is 600 to 1,200 mm and mean daily temperatures are 
more than 20°C. The main use of seed is as a source of edible oil, but the high oil (45- 
50%) and protein (26%) contents also make it an important food crop. Since the mid- 
1970s, edible groundnuts have increased in importance in both domestic consumption and 
export trade. Large quantities are consumed in the areas of production.
As a combined oilseed and food crop, groundnut ranks second only to soybean.
It is a valuable source of B Vitamins (particularly niacin which is low in cereals), and the 
cake after extraction of the oil is a high protein animal feed. With proper processing, the 
cake also is utilized for making products such as biscuits and baby or invalid foods. The 
green haulms provide good quality fodder and can be made into hay. In drier parts of the 
semi-arid tropics, groundnut fodder is valued as highly as pod yield. Groundnut is a 
valuable cash crop for millions of small-scale farmers in the semi-arid tropics. It 
generates employment on the farm and in marketing, transportation, and processing. It is 
a valuable source of foreign exchange when exported. It is an important component of the 
fat content of diets in India, Myanmar, and China, and of the protein content of diets 
throughout sub-Saharan Africa.
Asia contributes 71.6% to the annual world groundnut production (FAO, 1994). 
China is the largest producer (34.1%), and India the second largest producer (29.5%). 
Sub-Saharan Africa produces 18.6%, Central America and South America 2.6% of the 
world production. WANA accounts for 0.5% of production and Europe and Oceania 
contribute less than 0.2%. About two-thirds of the world groundnut production is used for 
oil extraction. China, the USA, and Argentina are the leading exporters of the crop. In 
India, most of the crop is processed for oil, with the oilseed cake used mainly for animal 
feed. In sub-Saharan Africa groundnut is a major food crop and only part of the produce 
is marketed.
Average productivity is highest in the region of South America (1.83 t ha'1), 
followed by Asia (1.48 t ha'1), and Africa (0.78 t ha1). In most countries of Africa and 
many countries of Asia, average productivity remains below the world average of 1.31 t 
ha'1. China has shown a dramatic increase in productivity (2.69 t ha'1), while the average 
yield in India remains below 1.0 t ha'1 in spite of marginal improvements in recent years.
The main constraints to productivity in Asia and Africa are diseases and insect 
pests, unpredictable and unreliable rainfall, low soil fertility, lack of improved agronomic 
practices and production technology, lack of technology-responsive cultivars adapted to 
local conditions, low financial inputs and lack of suitable small-scale farm implements and 
of the infrastructure to supply quality seed of the currently available improved cultivars. 
Aflatoxin contamination in the field and during storage reduces the marketability of the 
produce.
Annex III - Page 35
Foliar diseases (rust, early leafspot, and late leafspot; all world wide), virus 
diseases (groundnut rosette virus and peanut clump virus in Africa; peanut stripe virus 
and peanut bud necrosis virus in Asia), aflatoxin contamination of the produce 
(particularly in the semi-arid tropics), foliar and soil pests (leaf miner, Spodoptera and 
white grub in Asia; millipedes, termites and aphids in Africa), nematodes (in Asia and 
Africa), drought, and low soil fertility are priority research targets within the target 
production systems of the semi-arid tropics.
There is close research collaboration in Asia and Africa between ICRISAT and 
the national programmes and advanced research institutions. Impact at the farm level is 
reflected by release of improved cultivars - 26 by 16 national programmes outside India; 
19 other varieties undergoing on-farm testing in 8 countries; 17 cultivars in India, 
including 4 developed from segregating material supplied by ICRISAT. Improved 
packages of cultivation practices have been developed in collaboration with national 
programmes in India, Indonesia, Vietnam, Nepal, and Sri Lanka. In Malawi, a large scale 
on-farm demonstration programme has been launched in collaboration with an NGO and 
the national programme to popularize CG 7 cultivar among the farmers.
A strategic research commitment by the CGIAR to groundnut improvement is 
justified by the crop’s important dietary contribution, its importance as a cash crop and 
income generator, its potential in meeting part of the global demand for vegetable oils, its 
secondary value as animal feed and fodder, its contribution to the sustainability of mixed 
cropping systems; and the evidence that major production constraints can be resolved 
through research. In 1992, TAC recommended that the priority given to groundnut 
research be increased moderately.
1.5.6. Vegetables
Many vegetables are grown in developing countries, and the kinds vary 
considerably from place to place, with strong social preferences dictating the choice of 
species used. Vegetables provide a valuable source of income to producers near large 
urban areas. As a group, they are high-yielding and are well adapted to small-scale 
operations if markets are close, and to large-scale operations as infrastructure improves 
and transportation and cold storage become available. All income groups need and prefer 
them as supplementary foods, and demand in developing countries is expected to increase 
by 3.4% a year throughout the 1990s.
Of the current production of 333 million t in the developing regions, Asia 
accounts for 72 %, West Asia-North Africa for 17 %, Latin America and the Caribbean for 
7 % and sub-Saharan Africa for 5 %. Production during the past two decades has been 
growing at 3.2%. The four most important vegetables in terms of area harvested in the 
developing regions are tomato (1.6 million ha), onion (1.3 million ha), peppers 
(0.9 million ha) and cabbage (0.8 million ha).
Inclusion of a vegetable initiative in the CGIAR System would complete the 
commodity portfolio from a nutritional point of view. The major constraints are diseases 
and insect pests, and there is much scope for varietal improvement. Poor marketing 
facilities are also a constraint given the perishability of many vegetables. Modest
Annex III - Page 36
increases in production can lead to temporary gluts, and a major research need in many 
areas is to extend the production period.
In 1986 TAC indicated that highest priority among new ventures within the 
CGIAR System should be assigned to research on vegetables. Research should be directed 
at the potential for increased vegetable production in both tropical and subtropical areas, 
with special emphasis on indigenous tropical vegetables. In 1988 TAC recommended that 
the CGIAR create and support an international entity which would help establish and 
coordinate regional collaborative vegetable research networks in Asia, sub-Saharan Africa 
and Latin America and the Caribbean. CGIAR support was to be limited initially to 
tomato, pepper, onion, and leafy green vegetables. However, TAC also recommended 
that studies and consultations with relevant institutions be carried out to determine the 
importance of other commodities such as okra and eggplant, and to identify the major 
constraints to production increases and marketing, as well as their research ability. The 
new entity would then have the flexibility to phase new research topics into its 
programme as necessary.
TAC further recommended that the highest priority be assigned to supporting 
research for tropical environments, with activities for subtropical environments to be 
initiated once those for tropical environments had become operational.
Two important operational considerations in TAC’s deliberations were the 
integration of this new initiative with the System’s current efforts on commodities which 
either are vegetables (green bean, vegetable cowpea, potato, sweet potato and soybean) or 
produce vegetables as byproducts (bean leaves and cassava leaves); and the 
complementarity of a CGIAR initiative with the work of AVRDC.
In 1990 TAC recommended that vegetables were an appropriate subject matter 
for inclusion in the expanded CGIAR effort, and that collaborative vegetable research 
networks in sub-Saharan Africa and Latin America and the Caribbean be implemented.
1.5.7. Tropical Forages
Grasses and legumes are an intermediate product that contribute directly to 
livestock production and indirectly to more sustainable land use.
Among the major agricultural commodities, the gross value of production of meat 
and milk from ruminants ranks 1st in Latin America and WANA regions, 2nd in 
Sub-Saharan Africa and 4th in Asia. The major constraint to production is the quantity 
and quality of feed. Forages are the main feed source in Latin America, WANA and 
Sub-Saharan Africa, while in Asia, forages are used to supplement crop residues.
Tropical grasses and legumes are also used widely for uses other than feed.
There is widespread use of legumes as ground covers in tree and fruit plantations, as 
green manure crops and in natural fallows for soil improvement and weed control. 
Grasses and shrub legumes are used as barriers for controlling soil erosion. Pastures are 
used in rotations with crops with the aim of improving soil biological, chemical and 
physical properties in addition to providing feed for livestock.
Annex III - Page 37
Forage improvement programmes based on collection and evaluation of wild 
grass and legume species for direct use as animal feed or for soil improvement have 
identified a number of genera of commercial importance for the subhumid and humid 
tropics. These include, among the grasses - Andropogon, Brachiaria, Cenchrus, 
Panicum, Paspalum and Urochloa, among the herbaceous legumes, perennial Arachis, 
Centrosema, Desmodium, Pueraria and Stylosanthes, and among the shrub legumes - 
Calliandra, Cratylia, Gliricidia, Leucaena and Sesbania.
The introduction of selected wild accessions of Andropogon gayanus and 
Brachiaria spp. into the native savannas of Latin America has resulted in a 12-16 fold 
increase in livestock productivity with improved grasses being used over 20 to 80 percent 
of the farm area. Monitoring of well managed grass and grass-legume pastures has 
demonstrated beneficial effects on soil properties. Nevertheless, legumes such as Arachis 
pintoi, Centrosema spp. and Stylosanthes spp. have found wider adoption as cover crops 
than for improved legume.
Limitations have emerged as some of these natural species have become widely 
adopted and enhancement programmes based on traditional recombination and selection 
have successfully been carried out to overcome them. They have focused on overcoming 
specific limitations such as susceptibility to insect (spittlebug in Brachiaria spp.) or 
disease tolerance (anthracnose in Stylosanthes pp.) attack while maintaining adaptation to 
infertile soils and high feed value.
The present focus of research is to develop forage components for specific 
agro-ecosystem niches in Latin America and South East Asia where a demand has been 
identified, e.g. ground covers for tree crops, legumes for fallow improvement on 
hillsides, short-term pastures for crop-livestock systems, dry season fodders for 
dual-purpose cattle and multi-purpose grasses and legumes for intensive farming systems.
2. LIVESTOCK (Still to be updated to take into account TAC and CG
deliberations in 1993-94)
2.1. Background
Livestock and their products contribute about 29% to the total value of 
production of agriculture, forestry and fisheries in developing countries. In sub-Saharan 
Africa their share amounts to 19%, in Asia to 28%, in West Asia-North Africa to 35% 
and in Latin America and the Caribbean to 38%. However, these figures under-estimate 
the substantial contribution that livestock frequently make to crop production through 
draught power and manure.
Livestock products provide 6% of calorie intake and 19% of dietary protein 
consumed in developing countries. Animal products are the only reliable sources of 
Vitamin B12, zinc and iron. Meat and milk are highly income-elastic products. Their 
consumption increases with incomes and urbanization. Given economic growth and 
technological improvements in developing countries, livestock’s contribution to 
agricultural production can therefore be expected to increase.
Annex III - Page 38
Domestic animals enhance the economic viability and sustainability of farming 
systems. They diversify production and management options, increase total farm 
production and income, provide year-round employment, and provide insurance in times 
of need. Sales of livestock products provide funds for purchasing critically needed crop 
inputs and for financing farm investments. Livestock often form the major capital reserve 
of farming households.
Among domestic livestock species, ruminants have special importance because 
they convert into edible products crop residues, byproducts, weeds and other biomass that 
cannot be directly consumed as food by humans. Ruminants provide the only practical 
means for using vast areas of natural grasslands in regions where low, unreliable or 
seasonally limited rainfall combined with poor, acid soils make crop production 
impractical. In crop producing regions, traction raises crop productivity while manure 
enriches the soil. In addition, ruminants provide farmers with the economic incentive 
required to plant nitrogen-fixing forage crops and maintain pastures in crop rotations, 
which reduce erosion, conserve soil moisture and enhance soil fertility. The key to 
enhancing these positive aspects of livestock production is good management. It should 
also be noted that poor management, and especially overstocking, can cause degradation. 
Population growth in semi-arid rangeland areas is exacerbating these problems.
Expansion of grasslands is the major factor that leads to deforestation in Latin America 
and the Caribbean.
In general, the returns from smallholder livestock development projects in 
developing countries have been low (World Bank, 1985). This further highlighted the 
need for research to expand the knowledge base for more effective livestock development 
planning in the future. Although valuable progress has been made, to date pay offs from 
CGIAR investments in livestock research have been slow to materialize.
2.2. Regional Importance
Cattle are especially important in Latin America and the Caribbean, and in the 
warm semi-arid tropics and cool tropics of sub-Saharan Africa and India (for milk).
Sheep and/or goats are important in West Asia-North Africa, East and Southern Africa, 
semi-arid West Africa and temperate South America. Although small ruminants provide 
only a small proportion of the global production of meat and milk, the aggregate data 
mask their importance in some regions. It is estimated that they provide 30% of the meat 
consumed in West Asia-North Africa and 20% of that consumed in sub-Saharan Africa. 
Small ruminants are also important generators of cash income.
The first major thrust of the CGIAR System’s research programme on improving 
ruminant production has been to enhance nutrition through improved management 
practices and the development of better pastures, forages, and other feed sources. 
Inadequate year-round feed supply is the major constraint to ruminant production in many 
areas of Latin America a the Caribbean, West Asia-North Africa, and sub-Saharan 
Africa.
Annex III - Page 39
The second major research thrust is to control ruminant diseases, particularly 
tsetse-transmitted trypanosomiasis, which is a major constraint in large parts of sub- 
Saharan Africa, and a form of theileriosis, East Coast fever, a major constraint in East 
and Southern Africa. This research is of a basic and strategic nature. Progress being 
made in understanding the biology of these diseases, the nature of host defence 
mechanisms and novel means of vaccination provides a basis for developing improved 
methods of control for other economically important livestock diseases worldwide.
The CGIAR has focused its research on the most important ruminants in 
developing countries, i.e. cattle, sheep and goats and on sub-Saharan Africa, which 
accounts for more than two-thirds of the CGIAR’s resources for livestock research. 
However, although located in sub-Saharan Africa, ILRAD has a worldwide mandate for 
animal disease research. About 21 % of CGIAR investment in livestock research is 
allocated to CIAT in Latin America and the Caribbean, for pasture improvement 
research, and 11% to ICARDA in West Asia-North Africa, for the improvement of 
forage production systems. In general, CGIAR-supported livestock research has not yet 
led to significant farm-level productivity increases, but CIAT’s technologies for pasture 
improvement on the acid soils of Latin America and the Caribbean are gradually being 
adopted, and ILRAD may be on the verge of a breakthrough with recent progress in the 
development of new vaccines to provide immunity to theileriosis.
TAC has recognized the importance of the domesticated buffalo in areas to which 
it is climatically adapted. However, since 85% of buffalo are found in only five countries 
of Asia, TAC’s position to date has been that the research needs for this species could 
best be met through regional efforts. Similarly, TAC has recognized the importance of 
the camel in arid and semi-arid environments. Again, TAC feels that the research needs 
for these species could best be met through network activities or by regional institutions. 
TAC considers that the CGIAR has no comparative advantage to initiate activities on 
buffalo or camel research.
Poultry and swine account for almost half the monetary and nutritive value of 
livestock in developing countries. However, TAC has not considered their research needs 
to be of sufficiently high priority to justify their inclusion in the form of commodity 
improvement programmes in CGIAR activities. Evidence from Asia and from Latin 
America and the Caribbean indicates that, as the demand for chicken and pigmeat 
increases, more intensive production systems are adopted, and technology from developed 
and other developing countries is rapidly and effectively applied in these systems. Both 
the poultry and pig sectors also benefit substantially from private sector research. 
However, consideration needs to be given to the production of feed crops to meet the 
rapid growth of demand caused by the expansion of poultry and swine production as 
population and urbanization increase in the next 20 to 30 years.
Demand for livestock products is rising rapidly in response to urbanization, 
population growth and income gains, while yields of both meat and milk are low 
compared with those of developed countries. Low productivity is associated with a 
number of interacting factors: poor nutrition and acute seasonal feed deficiencies, limited 
availability of water in arid and semi-arid areas, poor management, disease, and low 
genetic potential as feed and health constraints are removed. In some regions, sub-
Annex III - Page 40
Saharan Africa in particular, the low productivity of cattle may also result from producers 
placing greater value on the number of animals owned than on their output of meat or 
milk, since animals are a means of storing wealth for future expenditures and as an 
insurance against drought. Overall, there is a significant need for research to increase 
ruminant production in developing countries.
2.3. Livestock Research in sub-Saharan Africa
As already noted, CGIAR efforts in livestock research have focused on sub- 
Saharan Africa. At present, livestock production systems in this region are predominantly 
subsistence oriented, and concentrated in areas that are tsetse free or only lightly infested. 
The majority of livestock is found on mixed smallholder farms, which account for 
approximately 60% of the region’s ruminant animal units. The productivity of livestock 
in terms of milk and meat in sub-Saharan Africa is the lowest of any world region. 
Nonetheless, it has been amply demonstrated that sub-Saharan Africa can produce meat 
and milk at prices that are competitive with imports, provided that markets are not 
severely disrupted by dumping, artificial currency exchange rates, and other 
macroeconomic distortions.
Milk accounts for 26% of the value of sub-Saharan livestock production, beef for 
37%, sheep and goat meat for 14%, pigmeat for 5%, and poultry for 8%. During the 
past two decades, increases in production have resulted largely from the expansion of 
herds and flocks, rather than from improved animal productivity.
The major constraints to improved productivity are natural resource limitations; 
technical barriers, such as inadequate feed supply, diseases, poor genotypes and 
inadequate management; and socioeconomic factors, such as inadequate government 
policies and marketing opportunities, and the lack of infrastructure.
Despite these constraints, there appear to be substantial opportunities for 
increasing livestock productivity throughout the region. It is felt that sub-Saharan African 
products can successfully compete with other foods in local markets and with imports.
Among the region’s highly diverse agroecologies, the subhumid tropics (AEZ 2) 
the higher rainfall areas of the semi-arid tropics (AEZ 1), and the cool tropics (AEZ 4) 
appear to have the greatest development potential for livestock production, particularly 
through integrated crop-livestock systems. In these systems, advances in dairy 
production, animal traction, poultry and pig production offer particularly promising 
opportunities for productivity and income gains. In the drier parts of the arid and semi- 
arid areas, pastoral systems continue to support a large number of ruminant livestock. 
The opportunities for technical intervention in such systems appear limited, but there is 
some potential for encouraging offtake through more favourable government policies, 
especially with regard to prices.
In the humid tropics (AEZ 3), the major stresses are pests and diseases, 
especially trypanosomiasis and dermatophilosis. Where pastures are established, the
Annex III - Page 41
control of weeds presents a special difficulty. These problems are likely to constrain 
production in this zone, at least in the short term.
Cattle will continue as the predominant source of meat, milk and traction in sub- 
Saharan Africa. The importance of small ruminants, poultry and pigs is expected to 
increase rapidly during the next two decades. Camels will remain important in arid 
areas, donkeys and horses in the cool tropics, and wildlife for tourism and game/bush 
meat in arid and semi-arid zones.
CGIAR efforts should focus on four principal research areas: feed supply, 
animal health, genetics, and sustainable production systems, particularly for the warm 
subhumid, warm semi-arid and highland (cool tropics) zones.
2.3.1. Feed Supply
Highest priority should be given to improving the quality of animal feed and its 
year-round availability. To support the intensification and spread of smallholder mixed 
crop-livestock farming systems, research on the production and feed value of forages, 
multipurpose trees, crop residues, cereals, root crops, grain legumes and less 
conventional feeds is needed. Greater efforts are required to enhance crop-livestock 
interactions. For example, as shortage of feed is the major constraint to increasing 
livestock productivity, the latter could benefit substantially from an increase in crop 
productivity.
2.3.2. Animal Health
The vector-borne diseases, trypanosomiasis, theileriosis, cowdriosis, babesiosis, 
anaplasmosis, and dermatophilosis, as a group constitute the most serious constraint to 
increased animal production in sub-Saharan Africa. Existing chemical control methods 
are often too costly for widespread adoption, and are also unsustainable due to the 
development of resistance to drugs and pesticides. Cost considerations deter the 
development, testing, licensing and introduction of new chemical products for use in sub- 
Saharan Africa. Continued strategic research on host-parasite relationships and control 
methods is necessary to provide strategies for overcoming the reduced productivity caused 
by this complex of diseases.
As animal agriculture systems intensify, strategic and applied research will be 
needed to resolve animal health problems related to intensification.
2.3.3. Genetics
Sub-Saharan Africa has valuable indigenous livestock germplasm that needs to be 
identified, characterized, preserved and utilized to enhance the productivity of animals 
while retaining their adaptation to the region’s environments. Strategic research to 
identify and manipulate genes that confer disease resistance, physiological adaptation to 
the environment, and productivity traits will grow in importance. Advances here will 
have global application.
Annex III - Page 42
2.3.4. Sustainable Production Systems
Research is needed to optimize the contribution of livestock to sustainable 
farming systems in each agroecological zone, particularly in the warm subhumid zone 
where production potential is high and the tsetse fly is retreating. Farming systems 
research must take into account agronomic, animal production, epidemiological and 
ecological factors. Policy research should give particular attention to macroeconomic 
issues related to natural resource use and the infrastructure needed to support the efficient 
development of animal agriculture.
The ability of governments to establish policies that foster the development of 
sustainable land use systems is hampered by serious deficiencies in livestock data. 
Databases on livestock, organized by agroecological zone, must be improved and 
expanded. National governments, the development community and the CGIAR should 
evaluate needs and develop improved methods for surveying and analyzing the current 
and potential role of livestock by agroecological zone in sub-Saharan African farming 
systems, especially by applying the new technologies of modelling and analysis provided 
by geographical information systems.
3. FORESTRY AND AGROFORESTRY
3.1. Background
Tropical forests cover only one-seventh of the earth’s land area, yet their 
importance is greater than this implies. In addition to wood, forests supply many non- 
timber products including foods and beverages, fibres, resins, building materials, fodder, 
ornamentals, medicines and fuel. More importantly, they provide environmental services, 
notably watershed protection, climate regulation, protection and improvement of soils, 
and provision of habitat for wild plants and animals. A wide range of cultural, spiritual 
and recreational benefits are also derived from tropical forests. They are an important 
component of the earth’s global carbon budget and the repository of perhaps half of all 
species of living things.
Thus the potential of forests to contribute to rural and urban welfare, economic 
growth, sustainable agricultural development and global environmental functions is vast. 
Yet it is constrained by accelerating deforestation and degradation of forest lands. About 
15.4 million hectares of tropical forests and woodlands were converted to other uses or 
destroyed each year between 1980 and 1990 (0.8% p.a.); 4.6 million hectares of this was 
tropical rain forests. By 1990, 1,756 million hectares of tropical forests remained; 52% 
in Latin America, 30% in Sub-Saharan Africa and 18% in Asia-Pacific.
The pressures on the remaining forests to provide ever-increasing volumes of 
timber, fuelwood, non-timber products and land for agriculture and other uses continue to 
grow - seemingly exponentially. Simultaneously, the importance of tropical forests to the 
well-being of the rural poor, in regulating the global climate, and as a reservoir of 
biodiversity is increasingly recognised. The need to do more with less, while at the same
Annex III - Page 43
time conserving a substantial area of tropical forests as a heritage for all people, is a 
fundamental problem which must be faced if these forests are to be successfully managed 
for the sustainable production of multiple goods and services.
Two decades ago, the Food and Agricultural Organization of the United Nations 
(FAO) estimated that slash-and-bum agriculture was practised on 30% of the arable soils 
of the world and provided sustenance for 250 million of the world’s poorest people. It has 
been suggested that the 200 to 500 million slash-and-bum farmers account for about thirds 
of global forest clearance annually (Myers 1994). Slash-and-bum agriculture remains the 
dominant land use at the margins of the humid tropical forests, and empirical evidence 
suggests that the numbers of people engaged in slash-and-bum agriculture may have 
doubled.
Agroforestry technologies have the potential to provide sustainable alternatives to 
the practice of shifting cultivation and to ameliorate the secondary forest fallows and 
grasslands that follow in its wake. In sub-humid savannas and woodlands and the semi- 
arid tropics, they can also play a decisive role in increasing the agricultural productivity 
and sustainability of small-scale farming systems. In these agroecological zones, 
agroforestry technologies can prevent soil erosion, bring a halt to deteriorating soil 
fertility, and provide food, fuelwood, building material, fodder and numerous other 
valuable products that have the potential for generating additional income.
3.2. Current Status of Forestry and Agroforestry Research
3.2.1. Forestry Research
Investment in forestry research and the human resources available to conduct 
research in developing countries are both low in comparison to the agricultural sector and 
in comparison to the value of goods and services derived from forests. Pardey et al 
(1988), however, assert that “comparison...between the share of crop research in 
agricultural research and crop production’s share of value-added in agriculture 
(AgGDP)....[indicates] that forestry research absorbs a larger share of research capacity 
than agriculture” However, this analysis ignores the fact that a large proportion of 
research effort in forestry is not directed towards outputs or outcomes that can be readily 
captured by crude aggregated indicators such as AgGDP, nor does AgGDP capture the 
many and varied ‘non-tradeable’ goods and services provided by forests. In addition, the 
number of forestry researchers active in research areas that have a direct ‘commodity 
orientation’ in developing countries would be less than 50% of the research cadre. 
Reliable current estimates for regional / global investment in forestry research are not 
available and therefore the summary presented by TAC (1994) remains, in the absence of 
more recent information, the most reliable source:
“Expenditure on forestry research in developing countries in 
1981 amounted to US$ 186 million, of which 60% was allocated to 
Asia, 21% to sub-Saharan Africa and 19% to Latin America and the 
Caribbean (Mergen et al, 1988). More recent data are not available, but
Annex III - Page 44
if past trends are an indication, current annual expenditures may be in 
excess of US$ 200 million.
Developing countries account for only 12% of total investment in 
forestry research world-wide. Forestry research intensity in developing countries 
is considerable less than one-tenth of agricultural research intensity. Forest 
research expenditures as a percentage of the value of production have been 
estimated at 0.019 for low-income developing countries, 0.059 for middle-income 
developing countries and 0.070 for semi-industrialized countries. The 
corresponding ratios for agricultural research expenditures were estimated at 
0.451, 0.863 and 0.816 respectively (Mergen et al, 1988). (TAC 1994)
Current statistics that assess the number of forestry research institutes and 
qualified researchers within them are available. FAO (1995) surveyed 764 forestry 
research organisation in 112 countries, 7% of which were located in Sub-Saharan Africa, 
3.8% in West Asia North Africa, 13.4% in Asia Pacific and 14.7% in Latin America and 
the Caribbean. Manpower, in terms of graduate staff in these institutions, does not 
follow the same pattern: more researchers (25% of the global total) are now located in the 
Asia Pacific developing countries. In Sub-Saharan Africa, Universities represent only 
14% of the forestry related research Institutes, the figures for WANA, AP and LAC are 
10.3%, 20.5% and 31.3% respectively. The following table provides summary data.
Table 1: Summary of Forestry Research Capacity in Developing Countries
Forestry N° of Graduate N°of
Region Research research researchers researchers
Institutes (%) Institutes (1985
values)*
Sub Saharan Africa 7a) 54 4.0 (7.3) 1106
West Asia-North Africa 3.8 29 2.9 (5.7) 815
Asia-Pacific (excl. China) 13.4 102 25.8 (9.4) 7177
Latin America - 14.7 112 9.7 (5.4) 2687
Caribbean
Global Totals 100 764 100 11785
adapted from FAO 1995, *1985 Pardey et al, figures for Asia Pacific exclude China.
3.2.2. Agroforestry Research
Agroforestry research has been widely taken up over the last two or three 
decades by small teams in agricultural and forestry research institutes and university 
departments throughout the world. Until recently, however, greater inputs have been 
made by development projects and NGO’s practising social and community forestry. 
ICRAF was established in 1977 to conduct and support agroforestry research. At that 
time, agroforestry was still lacking the theoretical basis and accumulated data of 
traditional areas such as forestry and agriculture. Early work therefore concentrated on 
developing the conceptual and methodological basis for agroforestry research, bringing 
together relevant information from disparate sources to make it available to a wide
Annex III - Page 45
audience, and supporting agroforestry training and education with a view to developing 
greater national capacity in agroforestry research and development. Building on this 
foundation, ICRAF initiated a collaborative research programme in 1985. The goal then 
was to strengthen national research capacity and ultimately to generate agroforestry 
technologies suitable for farmers.
In 1991, ICRAF became a member of the CGIAR. At present, ICRAF’s work is 
concentrated in four ecological regions of Africa: the humid lowlands and the semi-arid 
lowlands of West Africa, the highlands of East and Central Africa, and the plateau lands 
of Southern Africa, as well as in the humid tropics of Southeast Asia and Latin America.
Worldwide agroforestry research is expanding (Sanchez, 1995), with a number of 
other institutions (e.g. CATIE, EMBRAPA, ICAR) playing a significant role.
3.3. Global Issues and Research Needs
3.3.1. Forestry
People have cleared and modified forests for millennia, as human needs, 
perceptions and numbers have changed: forests have been modified to increase the flow of 
benefits to the users-managers. But some changes have had unintended or “perverse” 
effects, especially in recent decades. As pressures on land and competition for access to 
it have increased, inequities have developed in the distribution of the costs and benefits of 
forest use. These problems have occurred at the levels of forest communities, nations, 
regions and the entire world. They have affected the poor and the rich, foresters, farmers 
and corporations, local and distant users, and will affect future generations. This 
situation will inevitably be exacerbated by growth in population. In fifty years, the world 
will need at least three times as much food as at present. This will increase pressures to 
convert forest lands to agriculture. The principal consequences and causes for concern
are:
local livelihoods will be lost or impaired for many millions of people who derive 
much of their income or subsistence from foods, fibres, medicines or other 
products and services from tropical forests; and
the environmental services from forests which support and sustain agricultural 
productivity could be seriously damaged, thus exacerbating agricultural and food- 
security concerns, as well as global environmental externalities.
In Asia and Latin America, much of the increase in global food production over 
the past thirty years has resulted from intensification (partly through scientific 
breakthroughs developed by CGIAR Centres). Rates of yield increase for many food 
crops are now slowing. Part of the increase in global food production has been through 
creating new farmlands by clearing forests, particularly in Africa, but this cannot continue 
without serious environmental and social costs.
Annex III - Page 46
Currently, around 3,400 million m3 of roundwood are removed annually from the 
world’s forests for human use. A little less than half of this is used for industrial 
purposes, with the remainder being used principally as fuel for cooking and heating. The 
annual contribution of forest products to the world economy currently approaches US$400 
billion. About one-third of this is generated in developing countries, where forest 
products contribute 2.7% of their combined gross domestic product. [The remaining two- 
thirds is generated in industrialised countries, contributing 2% of their total GDP.] In 
developing countries slightly more than half of the economic value of forest products 
comes from the use of wood in energy, [compared to less than 10% in the industrialised 
countries].
The global value of trade in forest products was about US$ 98 billion in 1991, 
equal to 3.3% of world mercantile trade and almost one-fourth of the total world trade in 
agricultural, fishery and forest products. In the three decades since 1961, world trade in 
forest products has more than tripled (measured in 1990 US dollars). Exports of forest 
products from developing countries have increased by a factor of six; the share of forest 
products exports which originate in developing countries has risen from 8% in 1961 to 
13% in 1990 (still a very low share, given that these countries represent 75% of the 
world’s population). The structure of forest products trade has also changed significantly. 
In 1961, 60% of developing country exports were unprocessed roundwood; by 1990 the 
value of roundwood exports had more than doubled in constant dollars but the share had 
dropped to 20% of the total.
Numerous studies have shown that official statistics consistently undervalue 
forests as sources of products and employment. The diversity of uses and products is 
enormous and most are processed or traded in the informal sector. Preliminary CIFOR 
analyses of the true contribution of forests to developing countries’ economies suggest 
that it is probably at least double that shown in national accounts. Forests may contribute 
as much as 6% of economic product and up to 10% of total employment. However just as 
the true extent of forest values are not adequately documented, so the costs to poor forest- 
dependent people of forest degradation and loss are not widely recognised. Deforestation 
is generally perceived as a global environmental problem and the magnitude of the losses 
incurred by developing country peoples has been largely ignored in the international 
debate.
New technologies and policies are essential to help reduce the destruction and 
degradation of tropical forests, to avoid enormous social and economic losses. These will 
not just be limited to the forestry sector as conventionally defined. Individuals and 
countries who generate their incomes from forest products will suffer, and consumers will 
face shortages and higher prices.
Yet as social recognition of the non-consumptive and amenity values of forests 
increases, the predominant demand on forestry is likely to change from timber to 
environmental services. Tropical forests in future will provide a much wider range of 
social, economic and environmental benefits, and for a much wider range of beneficiaries. 
Many forests will be of more value for environmental and watershed protection, or for 
provision of non-timber products to local communities, than for their capacity to produce 
industrial cellulose. Although the aggregate demand for finished products is projected to
Annex III - Page 47
continue to increase, the amount of industrial raw materials from forests may rise more 
slowly, or possibly even decline, due to expanding technological abilities to find 
alternative raw materials (for most uses, from construction to paper), to increase 
processing efficiency, to increase functional product life and to recycle. In contrast, our 
reliance on environmental services from forests will not decline, and may increase. This 
transition is further complicated by equity considerations, because the beneficiaries from 
industrial use of forests may be quite different from those who benefit from non-industrial 
usage, or from the provision of environmental services from forests.
As prices increase for timber and other products from managed natural forests, 
and as plantation technology and "tree-husbandry" techniques improve and become more 
widespread, forest products will increasingly come from "cultivated" rather than from 
"natural" sources of supply. Indeed, rapid expansion of high-yielding plantations in the 
tropics could provide an alternative, commercially attractive resource that may reduce 
harvesting pressures on natural forests. The average annual timber productivity of natural 
forests, world-wide, is currently lm3/ha/year, thus 4 billion hectares presently yield 
approximately 4 billion m3 of logs. At the other extreme, only 100 million hectares of 
well-managed, well-located tropical plantations could, hypothetically, generate the same 
volume of timber harvest. By increasing the productivity of forest management, the area 
required to produce a given timber yield could be significantly reduced. Intensification of 
forest management - of both natural forests and plantation forests - is inevitable. Similar 
trends can be demonstrated for a variety of non-timber forest products under intensified 
management and domestication.
World-wide, there are already about 100 million hectares of forest plantations. 
They comprise 2.6% of the world’s forest areas and met 15% of the world’s wood 
requirements in 1995. There are clear trends of increasing areas of both industrial and 
non-industrial plantations in the tropics. In 1990 there were 43 million hectares of tropical 
plantations - up from 21 million hectares in 1980, so most of them are very young still - 
and almost 75% of this is in Asia (Table 2). Most of the 28 million hectares of non­
industrial tropical plantations are in farm strips or woodlots in Asia (82%), with 12% in 
Latin America and the Caribbean and 6% in Africa.
Table 2: Area of plantations including woodlots (’000 ha) in the tropics
Region 1965 1980 1990
Africa TT7S 2 724 3773 W%)
Asia incl. China 4 421 13 046 29 245 (68%)
Australia and 70 269 420 (1%)
Pacific
Central America 219 486 786 (2%)
and Caribbean 
South America 597 4 448 8 470 (20%)
Total 6 667 20 973 42 694
Source: Evans (1992) Plantation Forestry in the Tropics. Clarendon Press, Oxford, p35
Although they provide comparable commodities, plantations generally provide 
fewer environmental benefits than natural forests, but nevertheless remain a better 
environmental option than many other land uses. Almost two billion hectares of land in
Annex III - Page 48
the tropics have suffered moderate to severe degradation in the past fifty years (Table 3). 
Physical and chemical degradation is manifested as soil compaction, sheet and gully 
erosion, increased soil acidity and reduced availability of nutrients. Biological 
degradation includes declining soil organic matter, losses in beneficial soil micro­
organisms and weed encroachment. In the humid tropics alone, there are 250 million 
hectares of degraded forest fallows, Imperata cilindrica grasslands and degraded pastures. 
In densely populated South and Southeast Asia, nearly 20% of farmland is unproductive 
due to salinity, waterlogging and loss of top soil. The restoration of forest cover on this 
land could do much to improve the welfare of these extremely poor rural people.
Table 3: Human-induced land degradation, 1945-1992
Total Degraded Area as % of Total 
Region Degraded Area Vegetated Land (1992)
(m ha)
Asia TZ5 20
Africa 494 22
South America 244 14
North & Central 158 8
America 
Oceania 103 13
Sources: Swaminathan (1995) from World Resources Institute (1992).
Tropical plantations will have only a small beneficial impact on tropical 
deforestation, as long as the major cause of deforestation is agricultural expansion. 
However, plantations can help relieve some of the other pressures leading to deforestation 
in specific socioeconomic conditions. Plantation forestry should be considered as 
complementary to the management of natural forests and not as a substitute. The real 
issue is the potential of plantations to optimise benefits from low-potential sites.
Forest plantations, at their best, foster local socioeconomic development, and 
provide employment, raw materials, infrastructure and environmental and recreational 
services for local people. At their worst, plantations can take land out of food production 
in areas of acute food shortage, increase landlessness or destroy culturally important 
species, habitats and landscapes. Plantations and societies interact, and ultimately many 
plantations have suffered financially when managed in a way that led to social problems. 
Where plantations form a major part of the land-use system, and where competition for 
land is strong, the social and cultural implications and feedback can be significant.
The notion of forestry providing a broad range of products and services 
dominates many countries’ forest policies as they attempt to ensure social, environmental 
and rural stability, as well as timber supplies. Thus, the area of plantations designed 
specifically for non-industrial products and services has greatly increased. Much still 
needs to be done to ensure that plantation forestry is sustainable, through incorporating 
appropriate responses to ecological, economic, social and cultural needs. Today the areas 
of land suitable and available for forestry are decreasing and are often part of an intricate 
mosaic of land use and ownership. The main benefits from integrating trees and other 
land uses appear to be in the less-productive marginal uplands where ecological benefits
Annex III - Page 49
translate into economic advantages, i.e. higher productivity on a sustained basis. 
Commercial incentive is often the strongest stimulant for tree growing but frequently local 
markets for forest products are not well established and/or there is poor market 
information, and consequently little private, small-scale investment in commercial 
forestry. Much can be learned by studying institutions and markets in those few countries 
where small-scale forestry is flourishing.
The applications of social forestry to increasing agricultural productivity, to soil 
conservation and to the provision of wood products has two elements, i.e. local 
participation and the sustainable increase of productivity on a fixed area of land. Local 
participation will occur only if farmers are able to take up new technologies - new 
incentives (or the removal of disincentives) may be necessary. Government commitment 
to promoting new technologies through legislation, technical support, market development 
and finances may be a key factor. Overall there must be economic gain; conservation 
without discernible economic benefits is difficult to promote.
The perceived ecological benefits are nutrient and soil conservation in fragile and 
hilly lands and restoration of the productive capacity of degraded lands, through soil 
biological processes controlling the decomposition of plant residues. Loss of biodiversity 
- both flora and fauna - from tropical forests, and release of vast carbon stores with the 
simultaneous reduction in the capacity for new carbon sequestration, have potentially 
devastating consequences. These global externalities provide compelling reasons for the 
world community to invest in reforestation and forest conservation.
We are currently witnessing the transition from harvesting naturally occurring 
forest products to domestication and cultivation of trees for specific purposes, analogous 
to the transitions millennia ago in the domestication of crops and livestock. Much of the 
world’s future supplies of timber and non-timber forest products are likely to come from 
domesticated sources such as mixed-species plantations. Simultaneously, timber and non- 
timber products from natural forests are likely to come increasingly from sources which 
are retained primarily for their environmental functions and which are under management 
which is “certified” as socially and environmentally benign.
3.3.2. Agroforestry
As each year passes, more and more people in the tropics face crippling poverty 
and famine. Understandably, their priority is firmly set on food production rather than on 
pursuing long-term conservation goals. And, tragically, as they struggle to meet then- 
subsistence needs using resources at unsuitable levels, their future becomes more bleak 
and environmental stability more tenuous. A few daunting statistics highlight the current 
land-use problems.
About 15 million hectares of tropical forest are destroyed each year by human 
activities such as unsustainable commercial logging, shifting cultivation, expanding 
settlements, land speculation and ranching.
Annex III - Page 50
On current trends, by the end of the decade, 2.4 billion people will be either 
unable to obtain their minimum energy needs, or will be forced to consume 
fuel wood faster than it is being replenished.
Between 1970 and 1990, available arable land fell from a world average of 0.38 
ha per capital to 0.28 ha.
Each year, 25 billion tonnes of productive top soil are lost to agriculture - mostly 
a direct result of erosion in the wake of poor land use
Vast areas of agricultural land are being depleted of nutrients - largely an outcome 
of fallow systems in their struggle to produce more food from small areas.
Creating replacement fields for depleted agricultural soils is the cause of more than 
half of the world’s annual deforestation.
Without significant change, these problems can only get worse as, during the 
decade, the world’s population surges by a further one billion. Part of the solution to this 
crisis lies with a new ‘green revolution’ - approaches to land use that will simultaneously 
feed people and conserve natural resources.
In the 1980’s and 90’s, external factors changed rapidly. Worldwide concerns 
about global warming, high rates of deforestation, accelerating land depletion and the 
need for sustainable land-use practices were brought to the forefront of the political 
agenda by the Brundtland Commission on Environment and Development and the Bellagio 
Strategy Meetings on Tropical Forests which helped set the stage for the 1992 United 
Nations Conference on Environment and Development (UNCED). It became clear that 
acceptable long-term solutions also require research on ecology and conservation, along 
with ways to decrease rural poverty.
Through such avenues as UNCED and the subsequent recommendations of Agenda 
21, the world community has recognized that these approaches must not only address the 
link between improved productivity and environmental protection, but must combine 
poverty alleviation with sustainable resources management based on sound scientific 
principles.
The desire to promote sustainable agriculture and rural development as a step 
towards improved natural resource conservation has had a direct bearing on agroforestry. 
While there is a continued focus on productivity and poverty alleviation, the goals and 
objectives have been modified to give greater emphasis to research related to the 
management of natural resources.
This is seen in the recent re-definition of agroforestry as a dynamic, ecologically 
based, natural resources management system that through the integration of trees in farm- 
and rangeland, diversifies and sustains smallholder production for increased social, 
economic and environmental benefits (Leakey, 1996).
The research needs in agroforestry are numerous and diverse, since agroforestry 
research requires a multidisciplinary approach linking social and biophysical sciences 
throughout the research-to-development continuum (ie strategic to adaptive research). 
Regarding the needs of the Intergovernmental Panel on Forests of the United Nations 
Commission on Sustainable Development (UNCSD) and the World Commission on
Annex III - Page 51
Forests and Sustainable Developments (WCFSD), ICRAF addressed the research and 
capacity building issues with three major agroforestry questions:
why aren’t farmers planting more trees?
how can agroforestry research contribute to the development of sustainable 
and profitable land-use systems?
how can agro forestry research, training and education promote the 
institutionalization and enhancement of sustainable land use and capacity 
building in research institutions?
Why aren’t farmers planting or retaining more trees.
Research is needed to examine and overcome the policy and germplasm constraints 
that prevent farmers from planting trees.
The policy constraints include: lack of land on tree tenure; the promotion of 
governments of non-sustainable or inappropriate land uses, and the economic disincentives 
of low prices, lack of marketing opportunities, poor infrastructure for trading and lack of 
credit, etc.
Policy changes conferring formal land or tree tenure and the removal of 
disincentives to plant trees will promote the availability of land to farmers, thereby 
increasing the land under productive cropping and promoting social well-being in rural 
communities.
The germplasm constraints are those of: the non-availability of seeds of many tree 
species, and where seeds are available a lack of information and knowledge about how to 
collect and use them. Farmers would like and would respond to access to improved 
germplasm, but frequently such material is in small quantities and without pathways for 
dissemination.
How can agroforestry research contribute to the development of sustainable and 
profitable land use systems?
One approach requiring research is to develop agroforestry practices that stabilize 
the agricultural side of the forest margin by the integration of trees in farmland for the 
production of timber and non-wood forest products. Multi-strata agroforests, as developed 
in Southeast Asia, are the ultimate example of such systems, but little research has been 
done to develop such systems elsewhere in the tropics. So far, there is little understanding 
on how these multi-strata agroforests - which are an attractive alternative to slash-and- 
bum agriculture - diversify agroecosystems and enhance biodiversity, sequester carbon, 
affect green house gas emission or sinks. Similarly, little is known of the economic and 
social benefits of these systems in terms of cash generation, food security and their ability 
to alleviate poverty and buffer and diversity economic returns.
Research is also needed for methods to improve agroforestry systems by 
integrating trees into farming systems in a way that develops a mosaic of land-uses on
Annex III - Page 52
both the farm and landscape scales. In this way, the productivity of the systems can be 
achieved, while also improving their ecological and economic stability.
The benefits of agroforestry systems will be further enhanced if the trees are 
domesticated so that they are both more productive, of higher quality, and superior in 
other desirable traits. The genetic improvement of agroforestry tree species for an 
increased range of timber and non-wood forest products should increase the overall 
economic returns from a unit of land in the longer term, so triggering improvements in 
marketing infrastructure and in the incentives to grow trees. Potential gains in this area 
are high, since little improvement work has been done to date on agroforestry species.
Although much agroforestry research has been focused on the restoration of soil 
fertility though biological nitrogen fixation, more is needed to ensure the rehabilitation of 
degraded farmland through the development of increasingly intensive and diverse farming 
systems. Current policies and price structures for inorganic fertilizers make them 
inaccessible to many resource-poor farmers. Research into methods to recapitalize 
depleted soils using trees to recycle nutrients from strata below the rooting zone of crops 
or applied fertilizers is particularly necessary. Such rehabilitation of abandoned non-forest 
land could increase the area of land under cultivation and increase its carrying capacity.
Large scale applications of rock phosphate as a slow-release fertilizer could also 
have a critical role in replenishing soils. However, much needs to be learnt about policy, 
environmental, social and economic factors of such an approach. Research needs also 
include the intra- and inter-specific variation among trees abilities to solubilize, uptake, 
use and recycle phosphorus, and on the soil process and land-use systems involved in 
maximizing the benefit from a capital investment in rock phosphate.
The maintenance of soil fertility is essential for the sustainability of production 
from this land, but sustainability will also depend on the control of potential weeds, pests 
and diseases. The risks of these should be minimized by integrated land-use management, 
but intricate mixtures of trees and crops also pose problems arising from their competition 
for resources (light, water and nutrients). Strategic research is needed to understand the 
processes so that trees and crops can be grown together to minimize competition for 
resources and reduce the risks of pest and disease outbreaks.
The capital value of fertile soils is also lost through erosion by wind and water and 
through leaching. Much is known about the ability of contour hedgerows and windbreaks 
to reduce erosion, but less is known for other agro forestry practices.
How can agroforestry research, training and education promote the 
institutionalization and enhancement of sustainable landuse and capacity building in 
research institutions?
The institutionalization of agroforestry through the development of national and 
regional research teams has occurred in many countries. This can be greatly enhanced by 
curriculum development in colleges and universities, postgraduate training for 
researchers; administrators and educationalists; workshops and courses; in-service
Annex III - Page 53
training; information dissemination to NGOs, farmers and others in rural development 
sector, and - importantly - by informing and working with farmers.
3.4. Priorities and Current Research Activities for Forestry and 
Agroforestry in the CGIAR
3.4.1. Forestry
From consideration of these factors and the potential beneficiaries of research 
findings, the following major research priorities are being addressed.
1. Underlying Causes of Deforestation, Forest Degradation and Poverty in 
Forest Margins
Policy and social science analyses (through country case studies) demonstrate 
how more effective co-ordination between forestry and non-forest sectors, and 
macro-economic policy reform, can improve land-use choices; enrich our 
understanding of the causes of deforestation and forest degradation and therefore 
provide preventive measures on a wider regional scale, and help assure that 
various stakeholders receive appropriate and lasting benefits from those 
resources.
2. Landscape-scale Conservation and Management: Forest Ecosystem 
Management
Appropriate forestry policies and technologies (including planning models and 
assessment methodologies) are needed at the landscape scale. The development of 
these policies and technologies will require both the acquisition and use of 
comparative base data among the ecoregional foci. Generic methods will be 
developed for (a) characterising vegetation for natural resource surveys; (b) 
indicators to be used in assessing biodiversity; (c) predictive modelling to map 
the distribution of species; (d) assessing plant functional attributes and site 
physical environment attributes; (e) undertaking natural resource surveys which 
integrate biophysical information with socioeconomic information.
3. Multiple Resource Management of Natural Forests
Guidelines and technologies which reduce environmental impacts associated with 
timber harvesting, while retaining economic efficiency, and policies and 
incentives which encourage timber concessionaires and loggers to utilise such 
techniques, should improve the efficiency of planning harvesting operations and 
reduce impacts of logging in natural forests of the tropics. Reduced-impact 
techniques in harvesting operations in tropical forests could result in a 50% 
increase in carbon sequestration by residual vegetation and a 25 % reduction in 
soil impacts caused by logging equipment.
Annex III - Page 54
Improved understanding of silvicultural interventions in natural forests and 
improved design of silvicultural research can increase the biological productivity 
and commercial value of logged-over forests, through the application of 
ecologically based silvicultural practices.
Increased productivity of secondary forests and identification of socioeconomic 
and policy factors that influence the adoption of management in secondary 
forests, will improve the contribution of such forests to the well-being of the 
rural poor and enhance the sustainable production of multiple goods and services. 
Substantial reduction in the conversion of primary forests to agriculture and 
pasture could result from increased use of secondary forests to generate incomes.
4. Assessing the Sustainability of Forest Management: Testing Criteria and 
Indicators
Certification of “sustainably managed forests” requires the identification of 
criteria and indicators which are objective, cost-effective, and relevant, in both 
industrial forests and in forests managed by local communities. Decision support 
methodologies for evaluation of the sustainability of forest management and 
generalised methods and technologies for testing criteria and indicators, including 
the social impacts of forest management, are sought by policy makers, 
government agencies and other organisations involved in assessing, controlling or 
certifying the sustainability of forest management.
The most important primary gain is the ability to discern clearly between 
sustainable and unsustainable forest management practices. This in itself will 
neither better the well being of forest dependent people nor enhance the 
conservation of forest areas unless it is adopted for use by:
Policy makers: to improve the policy framework to reward sustainable 
management practices and discourage unsustainable utilisation and conversion of 
forest lands.
Trade: Preferential treatment or better prices for forest products especially timber 
from sustainable sources through certification. This in turn is expected to be an 
incentive for the adoption of sustainable management practices.
Management: to reduce environmental impacts and degradation through an 
improved ability to diagnose the sustainability of management practices, and 
reduce social tensions and increase opportunities for income generation and 
maintenance of lifestyles.
5. Plantation Forestry on Degraded or Low-potential Sites.
Improved understanding of key factors determining growth rates on low potential sites 
can lead to guidelines and technologies which improve production and sustainability of 
fast-growing plantations on degraded and low potential sites, through better management 
practices, especially for smallholders involved in plantation forestry.
Where improved management practices are identified and adopted in fast-growing 
plantations, increased growth rates of 10-40% in the second and subsequent rotations can 
be achieved. Improved economics of small-scale plantations compared to other land use- 
alternatives, should reduce pressures on natural forests and increase opportunities for 
households and communities to generate income from forest products.
Annex III - Page 55
6. Conservation of Biodiversity and Genetic Resources 
This project will determine the impacts of human disturbance, logging and 
fragmentation on in situ conservation of biodiversity, and develop tools for 
measuring and monitoring genetic diversity. Outputs will be the establishment of 
a relationship between the intensity of disturbance and genetic diversity for 
species with different life history characteristics which will allow the formulation 
of management prescriptions for protected areas. The tools will include 
molecular marker techniques; and GIS and remote sensing applications. The 
impact will be through more effective conservation of biodiversity of forests and 
its effect on sustainable forest utilisation and efficient use of natural resources.
7. Local Livelihoods, Community-based Management and Devolution
Based on a better understanding of the livelihoods, property rights and decision-making 
systems of people living in forest areas, this project will identify policy options for 
improving the compatibility of conservation objectives with local people’s needs. It will 
identify institutional arrangements for successful integrated conservation and 
development. Tools include simple methodologies for the rapid measurement of forest 
incomes and assessment of income opportunities compatible with conservation and 
systems for the quantitative modelling of the people-forest interface. The application of 
the research results should have significant impacts on the well-being of communities in 
forest areas. The research should also result in improved performance of integrated 
conservation and development projects and extractive reserves.
8. Sustainable Use and Development of Non-Timber Forest Products 
Theories and models are required that explain trends in NTFP utilisation and 
assess the potential for NTFP based development. The project will estimate 
impacts as a result of NTFP based development on forests and biodiversity, as 
well as generating more reliable estimates of the number of people who depend 
on NTFP, and the nature of this dependency. Procedures for assessing the 
potential of NTFP development will become important tools in projects where 
NTFP based development is being considered.
3.4.2. Agroforestry
Priorities for agroforestry research span two major areas of activity: the 
development of alternatives to slash-and-bum agriculture and the mitigation of land 
degradation through soil depletion. Both of these research areas are important as means of 
counterbalancing the effects of deforestation on farmers livelihoods. Cross-cutting themes 
of this research agenda are soil amelioration, environmental rehabilitation, cash 
generation for poverty alleviation, tree domestication, agroecosystem function, impact at a 
landscape scale and capacity building.
ICRAF is the convening centre for two of the CGIAR’s systemwide programmes: 
Alternatives to Slash-and-Bum Programme which spans the 3 humid lowland ecoregions, 
and the African Highlands Initiative which is part of the Global Mountains Initiative and 
is based in East Africa.
Annex III - Page 56
Agroforestry research priorities are organized under the following four
programmes.
1. Natural Resource Strategies and Policy:
This programme focuses on farmers’ management of the resource base of 
agriculture, as well as on the ecological, social and economic interactions among these 
resources across different spatial scales. Its research agenda is driven by farmers’ needs 
and goals, and by ecological principles rather than being led by technological 
development.
It aims to;
develop typologies of land-use systems and farmers on the basis of ecological and 
socioeconomic parameters and at various spatial scales (household to continental), 
develop (and validate) ecological-economic models which predict the adoption 
potential and sustainability of improved agroforestry practices at the farming 
system and regional scales.
analyse policy constraints to the adoption of these improved practices and to 
design alternative policy instruments for addressing them, 
evaluate the actual ecological, social, cultural and economic impacts of adoption 
of agroforestry practices.
2. Domestication of Agroforestry Trees
Through its Germplasm Resources Unit (GRU), the Programme collects, 
characterizes and disseminates information and germplasm of priority agroforestry trees. 
This germplasm is then evaluated for its genetic diversity and domesticated through 
genetic selection, vegetative propagation and breeding, for use in various agroforestry 
technologies. The objective of this research is to improve the genetic quality of 
agro forestry trees species by the collection, evaluation and selection of germplasm for the 
compatible production of food, fodder, fuel wood, timber and other products with 
companion crops, and the provision of environmental services such as soil conservation 
and amelioration of soils.
3. Tree-Crop-Environment Interactions
Process measurements are made in priority agroforestry systems identified to 
solve specific land-use or socioeconomic problems in each ecoregions, so as to understand 
the function as well as improve the performance of the systems.
The objective is to contribute to the development of improved and sustainable 
agroforestry systems by:
gaining an understanding of biophysical interactions between the components of 
agroforestry systems
synthesizing results from a wide range of environments through models and 
obtaining predictive capability for evaluating alternative systems and management 
options
Annex III - Page 57
developing research methods for facilitating the conduct and improving quality of 
process-oriented research in agroforestry
4. Agroforestry Systems Improvement
Research activities are undertaken on the potential of agroforestry to address the 
constraints and opportunities that farmers experience.
This systems improvement research focuses on the monitoring and evaluation of 
the long-term biophysical, ecological and economic impact of agroforestry technologies 
used as alternatives to current systems. This work is undertaken on research stations and 
on farms by multidisciplinary teams of scientists at a range of locations in many 
countries.
The priority systems currently being investigated address the prospects and 
constraints of alternatives to slash-and-bum agriculture, the reclamation of abandoned and 
depleted lands in the humid tropics, and land depletion in the savanna woodlands and 
agrosilvopastoral systems of the subhumid and semi-arid tropics.
The objective is to develop, through a participatory research approach, 
biophysically-sustainable and economically-viable agroforestry systems which address the 
priority concerns of farmers and major environmental issues in the six ecoregions in 
which ICRAF works.
4. FISHERIES
4.1. Background
Fisheries play an important role in food production, income generation and the 
provision of employment in developing countries. The number of full-time fishermen in 
developing countries has been estimated at 12.9 million, of whom 80% live in Asia, 12% 
in sub-Saharan Africa, 6% in Latin America and the Caribbean and 2% in West 
Asia-North Africa (ICLARM, 1991).
In addition, there are many millions of part-time fishermen. Water covers 70% 
of the earth’s surface, and the total production of aquatic commodities amounts to 95 
million tonnes annually, of which 79% is in the form of finfish, 5% crustaceans, 9% 
molluscs, and 7% seaweeds. Fish and fish products provide 20% of animal protein and 
4% of dietary protein in developing countries, but these averages mask the fact that in 
several countries this share is at least twice as high. The total gross value of world 
fisheries production is almost US$ 25 billion per year, of which 52 % originates from 
marine capture fish, 18% from inland capture fish, 16% from inland culture fish and 14% 
from marine culture fish. Fish account for 5.6% of the total value of production of 
agriculture, forestry and fisheries.
Annex III - Page 58
Approximately 72% of the gross value of fish production originates in Asia, 17% 
in Latin America and the Caribbean, 8.1% in sub-Saharan Africa and 3% in West 
Asia-North Africa (TAC/CGIAR, 1990).
Of the global aquatic production, only 12% originates from aquaculture, but in 
value terms this share amounts to 29 %. Aquaculture differs from capture fisheries just as 
agriculture does from hunting and gathering. Aquaculture, as in the case of agriculture, 
and even more than in capture fisheries, requires ownership or control over the aquatic 
resources (commodity) and space. It implies action to direct energy flows in the 
ecosystem towards the commodity produced. During 1993, total world aquaculture 
production amounted to 16.3 million tonnes, of which 50% consisted of finfish, 4% 
crustaceans, 18% molluscs and 28% seaweeds. About 86% of world aquaculture 
production originates in Asia, but in terms of value this share amounts to 78 %. The 
value of culture fisheries exceeds that of many ‘traditional’ CGIAR commodities such as 
beans, sorghum, and groundnuts.
4.2. Global Fisheries Research Needs
Issues and their relation to CGIAR goals: Previous TAC reports, as well
as a recent study undertaken on behalf of several donor agencies, have stressed the need 
for more research on fisheries (see, for instance CEC/FAO/UNDP/Worid Bank 1991). 
Humans capture some 4,000 species of fish, molluscs, crustaceans and other aquatic 
organisms and culture nearly 200; capture and culture contribute to food security both 
through the provision of fish and through employment and income generation. Aquatic 
resource research contributes directly to several CGIAR goals, including the preservation 
of biodiversity, better management and conservation of natural resources, the 
improvement of the policy environment and the strengthening of national research 
systems.
Fisheries in a State of Transition: For more than four decades, from
about 1950, the per capita supply of fish rose continually despite rapid population 
growth. About 50% of the demand increase since 1970 is attributed to increased 
population demand, and the remainder to economic factors such as rising disposable 
incomes in many countries. However, in 1990, the rise in supply was halted, and a per 
capita decline in supply began to occur. The price of fish had begun to rise during the 
1980s and this rise was far greater than for any other food commodity groups, the prices 
for most of which continued to fall over the period. Trade in fish has also increased, 
making fish one of the most highly traded of agricultural commodities — nearly 40% of 
production by value is traded internationally.
World fisheries and aquaculture are now in a state of transition, learning to cope 
with increasing stock scarcity price increases. Key current issues include the need for 
better management of natural fish stocks, increased production through aquaculture and 
better stewardship of the aquatic environment to prevent a large gap between demand and 
supply (Williams in press).
Annex III - Page 59
Global Concern for the Future: Recent predictions are that the world will need 
between 110 and 120 million tonnes of fish by the year 2010, compared to the present 
supply of about 71 million tonnes of food fish (FAO 1995). The implications for food 
security in the developing world are serious as was noted at the December 1995 
Conference on the Sustainable Contribution of Fisheries to Food Security Fisheries. The 
major output of this conference was the Kyoto Declaration and Plan of Action, which was 
signed by 95 countries. Several parts of the document relate to research and its 
contribution to food security through fisheries and aquaculture, as shown in Table 1, 
below.
Table 1: Extracts from the Kyoto Declaration and Plan of Action on the Sustainable 
Contribution of Fisheries to Food Security (Kyoto Declaration 1995).
We, the 95 States which met in Kyoto from 4 to 9 December 1995 on the occasion of the 
International Conference on the Sustainable Contribution of Fisheries to Food 
Security...............
DECLARE that we should, without prejudice to the rights and obligations of States under 
international law:
1. Recognise and appreciate the significant role which marine fisheries, inland fisheries 
and aquaculture play in providing food security for the world, both through food supplies 
and through economic and social well-being;
7. Undertake in-depth studies to assess the social, economic and cultural importance of 
fisheries and fishery products;
8. Promote and strengthen scientific research as the fundamental basis for sustainable 
development of fisheries and aquaculture activities to ensure food security, as well as 
provide scientific and technical co-operation and support for those countries with lesser 
research capabilities.
9. Base policies, strategies and resource management and utilisation for sustainable 
development of the fisheries sector on the following: (I)maintenance of ecological systems; 
(ii) use of the best scientific evidence available; (iii) improvement in economic and social 
well-being; and (iv) inter- and intra-generational equity.
11. Assess the stock productivity in the waters under national jurisdiction, both inland and 
marine, adjust the fishing capacity in these waters to a level commensurate with long-term 
stock productivity etc....
12. Conserve and sustainable use biological diversity and its components in the aquatic 
environment and, in particular, prevent practices leading to irreversible changes, such as 
extinction of genes and species, genetic erosion and/or large scale destruction of habitats;
13. Study the effectiveness of multi-species management;
18. Promote the use of sustainable and environmentally sound aquaculture and ranching in 
coastal marine and inland waters through, inter alia: (I) establishment of appropriate 
institutional and legal frameworks; (ii) coordination of the use of lands an waters with 
other activities; (iii) use of the best and most appropriate genetic material in conformity 
with the conservation and sustainable use of the environment and conservation of 
biological diversity; and (iv) application of social and environmental impact assessments;
22. Provide, either directly or through regional, sub-regional or international 
organizations, technical and financial assistance to developing countries, in particular low 
income food-deficit developing countries and small island developing States, in order to 
assist them to realize the sustainable contribution of fisheries to food security and social 
and economic development;
Annex III - Page 60
Research Topics: Despite its obvious fragility, the resource base in the 
developing world is still poorly understood. Research for most tropical fisheries is still 
rudimentary, although national research capacity is developing to meet the challenge. 
Several research needs are priorities, including cost-effective data acquisition, especially 
as it relates to fisheries resources, aquaculture development and biodiversity, developing a 
holistic approach, integrating biological and social science research, exploring the impact 
of protected areas, studying the potential of aquaculture and its relationship to the 
environment, extending the genetic improvement of aquaculture species, overcoming 
aquatic environment degradation and analyzing the impact of research. Several other 
topics, including improving post-harvest handling of fish and diagnosis and management 
of disease in aquaculture are also important, but of less relevance to present CGIAR 
objectives and capabilities.
Cost-Effective Data Acquisition: In most fisheries, the several decades of 
rapid expansion of fishing have not allowed time to build up an understanding of the 
resource before it becomes fully exploited or overexploited. Cost effective systems for 
the acquisition of fisheries resource and fishing data (social and economic as well as 
biological) are still lacking. Scientists have developed some useful methods for the 
analysis of the resource base and its productive capacity but much more needs to be done 
before these methods can be effectively applied to the range of resource systems and the 
range of resource states (i.e. from newly exploited to degraded and over-exploited states). 
The role of biodiversity in the productivity of different types of resource systems is 
poorly understood and aquatic biodiversity is one of the least documented of any 
biological system. Available information on species is widely scattered and only now are 
systems such as FishBase starting to draw this information together.
Holistic Approach: Scientists are now realizing that there is a need to understand 
the functioning of whole aquatic systems in order to understand the impacts of fishing, 
environmental degradation and climate change. This realization has changed the scale at 
which resource systems work is targeted and the types of partnerships required in 
research.
Integrating Social and Biological Science: Managing the resource base is not
simply a question of obtaining sufficient biological knowledge. A host of social, economic 
and cultural factors surround the institutions and the decision-making processes. Research 
by the social science disciplines, preferably integrated with or in close collaboration with 
biophysical studies, can contribute significantly to policy and institutional insights, advice and 
development. For example, the condition of free and open access is often blamed for 
overexploitation and rent dissipation in fisheries. Research has suggested that policy 
measures which are likely to be most effective in resource management are those which 
remove the open-access condition, i.e., establishment of user rights. In certain natural 
resource contexts, co-management and community-based management are approaches in 
which user rights can be instituted appropriately but we still do not understand sufficiently 
well the contexts most suited to these forms of management. Successes and failures of these 
approaches should be critically assessed and refined.
Protected Areas: Protected areas are gaining popularity as an option for improved 
natural resources management; fisheries are one type of resource who which this approach
Annex III - Page 61
might be useful. However, far more research is required to establish the most effective form 
of protected areas in different cultural and resource systems. The ad hoc implementation of 
protected area regimes may lead to unwarranted disappointment in the method. Research on 
the benefits and management approaches to protected areas such as marine parks should be 
enhanced, especially through joint research with national systems, and with the participation 
of neighbouring residents.
Aquaculture Potential and Environmental Effects: Annual average production 
from aquaculture in 1990-93 was 15 million t. Potential growth of the sector has been 
estimated to reach 27 million t, under pessimistic conditions, and 39 million t, under the 
most optimistic. Non-traditional products, e.g. giant clams and sea cucumbers, that are seen 
to benefit coastal village farmers should be developed and the markets for such projects 
explored through economic studies.
Some forms of aquaculture have already had deleterious effects on the environment 
and vice-versa. Carrying capacities in both freshwater and marine environments should be 
assessed particularly in sites with multiple and conflicting use of water resources. Technical 
inputs to regulations and aquaculture development policy are needed.
Much of the current aquaculture development is on a fully commercial basis and 
producing products which have little impact on food security for the low income. Specific 
intervention to develop small-scale technologies for the adoption of fish farming for resource- 
poor farmers should be pursued further and market barriers minimized. Many of these 
technologies will need to be developed through farmer-participatory research tailored to 
specific situations. International public goods can be drawn from the research in terms of 
methodologies and on-farm results.
Species Improvement: Genetic improvement of aquaculture species should be 
expanded beyond Nile tilapia and beyond the first simple breeding goals (growth and 
survival) to include other marketable species such as carp and goals such as disease 
resistance, desirable maturation goals but with great caution so as not to result in erosion of 
biodiversity. Working through national partners, the CGIAR should support the development 
of new approaches to national fish breeding programmes which underpin the multiplication 
and dissemination of new breeds. Since selective breeding in fish is still so new, there is no 
ready dissemination pathway for such breeds. Emphasis should be given to domestication 
and improvement of native species. There is also need to develop policy guidelines and codes 
of practice to safeguard the environment and maintain biodiversity.
Aquatic Environment Degradation: The degradation of aquatic environments, in 
both coastal and catchment areas, and increasing competition between the fisheries sector and 
other sectors (tourism, shipping, agriculture) emphasize the need for a comprehensive and 
integrated approach to management. Tools that fully utilize and integrate the inputs of various 
disciplines should be developed and made accessible to policy-makers. Information databases 
should be maintained and made accessible to a wide group of users.
Impact Analysis: The impacts of market globalization on poor fishers and consumers 
of fish should be analyzed. The impact of adoption of aquaculture on production and
Annex III - Page 62
consumption of fish and on the environment should be assessed. Methods for assessing the 
impact of fisheries research on target beneficiaries should be developed and applied.
Importance of Linkages: At best, the CGIAR can only produce a tiny fraction of 
the fisheries research that is needed to overcome the food security threats described above. 
To maximize the effect of its research, strong emphasis needs to be put on developing and 
sustaining an increasing number of linkages with others in the field. In keeping with the 
state of the art in fisheries research approaches, CGIAR work should involve the indigenous 
knowledge of local people to the maximum extent possible. This in turn will be facilitated 
by collaboration with NGOs who work with such people. Linkages with NARS and 
advanced research institutions will ensure that CGIAR results are disseminated, and outside 
ideas are incorporated into system work. The importance of such linkages suggests the need 
for a special programme in this area.
ANNEX IV
SELECTED TABLES OF QUANTITATIVE ANALYSIS
Table 1a: 2010 value of demand of CGIAR commodities by production 
sector; commodities valued in 1989/91 international $
Production sectors (% of total)
Region Crops Livestock Forestry Fisheries Total
Sub - Saharan Africa 6.3% 2.7% 5.6% 0.6% 15.2%
Asia 34.4% 9.3% 10.0% 4.7% 58.3%
C. & S. America 7.7% 6.5% 3.8% 0.5% 18.4%
WANA 4.2% 3.1% 0.4% 0.3% 8.0%
Total 52.5% 21.7% 19.7% 6.1% 100.0%
Regional share (% of sector)
Region Crops Livestock Forestry Fisheries
Sub - Saharan Africa 12.0% 12.5% 28.3% 10.3%
Asia 65.5% 42.9% 50.5% 77.4%
C. & S. America 14.6% 30.1% 19.1% 8.0%
WANA 8.0% 14.5% 2.1% 4.2%
x
Total 100.0% 100.0% 100.0% 100.0% 3
Table 1 b: 2010 value of demand of CGIAR commodities: values in 1000 int’l $ and shares by commodity
Regional values and shares of commodity
Sub-Saharan Asia Latin America and West Asia and Overall value Share of 
Commodity Africa Share Share Caribbean Share North Africa Share of commodity grand total
Rice 5536690 4.1% 119371469 88.8% 6371498 4.7% 3080386 2.3% 134360043 17.5%
Wheat 2097288 3.3% 40257850 63.7% 5744851 9.1% 15078283 23.9% 63178272 8.2%
Maize 5275084 11.0% 27767295 57.8% 11579343 24.1% 3438061 7.2% 48059784 6.3%
Barley 347939 5.2% 846017 12.7% 599640 9.0% 4872428 73.1% 6666025 0.9%
Sorghum 3302430 38.1% 2352206 27.2% 2681450 31.0% 326331 3.8% 8662417 1.1%
Millet 3112521 56.4% 2335319 42.3% 28788 0.5% 41791 0.8% 5518419 0.7%
Cassava 9094680 63.5% 2206144 15.4% 3028944 21.1% 88 0.0% 14329858 1.9%
Potato 525052 4.2% 7863944 63.1% 2101902 16.9% 1977514 15.9% 12468412 1.6%
Sweet potato and yam 4628121 29.7% 10600972 68.0% 350252 2.2% 8716 0.1% 15588061 2.0%
Sweet potato 547636 4.8% 10559409 93.3% 198672 1.8% 8716 0.1% 11314433 1.5%
Yam 4080486 95.5% 41563 1.0% 151580 3.5% 0 0.0% 4273628 0.6%
Banana 1608592 16.0% 5162804 51.4% 3031567 30.2% 234788 2.3% 10037751 1.3%
Plantain 3110477 78.7% 90394 2.3% 750298 19.0% 0 0.0% 3951168 0.5%
Pulses 5613884 19.1% 16898684 57.5% 4119553 14.0% 2756104 9.4% 29388226 3.8%
Beans 2784840 21.2% 6248441 47.6% 3732009 28.4% 357300 2.7% 13122590 1.7%
Chickpea 412569 5.0% 6167893 74.5% 244111 2.9% 1455887 17.6% 8280461 1.1%
Pigeonpea 397834 12.9% 2656594 85.8% 41375 1.3% 0 0.0% 3095803 0.4%
Lentil 55991 3.0% 980001 53.3% 45512 2.5% 757595 41.2% 1839100 0.2%
Broad beans 218072 17.1% 829347 65.2% 42754 3.4% 182357 14.3% 1272529 0.2%
Cowpea 1744577 98.1% 16408 0.9% 13792 0.8% 2965 0.2% 1777742 0.2%
Soyabean 235130 0.9% 7510748 30.1% 17000509 68.2% 184002 0.7% 24930389 3.3%
Groundnut 3344029 16.7% 15614744 78.0% 897539 4.5% 171848 0.9% 20028160 2.6% > 
Coconut 326142 6.2% 4566089 86.2% 403239 7.6% 0 0.0% 5295470 0.7% §
Beef and buffalo meat 10023010 13.9% 24384940 33.7% 30367315 42.0% 7565729 10.5% 72340993 9.4% x 
Sheep and goat meat 3887348 18.8% 8906057 43.1% 1492599 7.2% 6374033 30.9% 20660037 2.7% ^
Total milk 6926299 9.5% 38033997 51.9% 18112646 24.7% 10198311 13.9% 73271254 9.6%
Fuelwood and charcoal 39762311 33.7% 56381380 47.7% 19897240 16.8% 2103035 1.8% 118143966
Roundwood 15.4% £2964668 9.0% 19879322 60.5% 8944718 27.2% 1067579 3.2%
Fish food 32856287 4.3% to 4815590 10.3% 36080074 77.4% 3741864 8.0% 1967100 4.2% 46604627 6.1% to
Grand total by region 116537286 447110448 141245755 61446128 766339617 100.0%
Regional share of grand total 15.2% 58.3% 18.4% 8.0% 100.0%
Table 1 c: 2010 value of demand of CGIAR agricultural commodities: values in 1000 int’l$ and shares by commodity
Region
Sub-Saharan Asia Latin America and West Asia and Overall value Share of 
Commodity Africa Share Share Caribbean Share North Africa Share of commodity total
Rice 5536690 4.1% 119371469 88.8% 6371498 4.7% 3080386 2.3% 134360043 23.6%
Wheat 2097288 3.3% 40257850 63.7% 5744851 9.1% 15078283 23.9% 63178272 11.1%
Maize 5275084 11.0% 27767295 57.8% 11579343 24.1% 3438061 7.2% 48059784 8.5%
Barley 347939 5.2% 846017 12.7% 599640 9.0% 4872428 73.1% 6666025 1.2%
Sorghum 3302430 38.1 % 2352206 27.2% 2681450 31.0% 326331 3.8% 8662417 1.5%
Millet 3112521 56.4% 2335319 42.3% 28788 0.5% 41791 0.8% 5518419 1.0%
Cassava 9094680 63.5% 2206144 15.4% 3028944 21.1% 88 0.0% 14329858 2.5%
Potato 525052 4.2% 7863944 63.1% 2101902 16.9% 1977514 15.9% 12468412 2.2%
Sweet potato and yam 4628121 29.7% 10600972 68.0% 350252 2.2% 8716 0.1% 15588061 2.7%
Banana 1608592 16.0% 5162804 51.4% 3031567 30.2% 234788 2.3% 10037751 1.8%
Plantain 3110477 78.7% 90394 2.3% 750298 19.0% 0 0.0% 3951168 0.7%
Pulses, overall 5613884 19.1% 16898684 57.5% 4119553 14.0% 2756104 9.4% 29388226 5.2%
Soyabean 235130 0.9% 7510748 30.1% 17000509 68.2% 184002 0.7% 24930389 4.4%
Groundnut 3344029 16.7% 15614744 78.0% 897539 4.5% 171848 0.9% 20028160 3.5%
Coconut 326142 6.2% 4566089 86.2% 403239 7.6% 0 0.0% 5295470 0.9%
Beef and buffalo meat 10023010 13.9% 24384940 33.7% 30367315 42.0% 7565729 10.5% 72340993 12.7%
Sheep and goat meat 3887348 18.8% 8906057 43.1% 1492599 7.2% 6374033 30.9% 20660037 3.6%
Total milk 6926299 9.5% 38033997 51.9% 18112646 24.7% 10198311 13.9% 73271254 12.9%
Overall agricultural commodities 68994717 334769672 108661933 56308415 568734737 100.0%
by region 12.1% 58.9% 19.1 % 9.9% 100.0% >
3
n3x>
HH
<
id
3
09
CO
Table 2a: 2010 value of demand of CGIAR commodities by 
production sector, weighted by the modified welfare indicator
(income threshold value 9000 PPP $, exponent set as 2)
Production sectors (% of total)
Region Crops Livestock Forestry Fisheries Total
Sub-Saharan Africa 10.7% 4.6% 9.4% 1.1% 25.7%
Asia 27.6% 10.2% 8.8% 4.0% 50.6%
C. & S. America 7.5% 5.3% 4.0% 0.5% 17.2%
WANA 3.4% 2.5% 0.4% 0.2% 6.5%
Total 49.2% 22.6% 22.5% 5.7% 100.0%
Regional share (% of sector)
Region Crops Livestock Forestry Fisheries
Sub - Saharan Africa 21.8% 20.2% 41.6% 18.5%
Asia 56.1% 45.2% 39.2% 69.9% >3
C. & S. America 15.2% 23.3% 17.6% 7.9% a3>x
WANA 6.9% 11.2% 1.6% 3.7% <
►3
Total 100.0% 100.0% P100.0% 100.0% eara>
Formula: VOD * (1 - ((1 -Gini coefficient)*lncome 2003)/9000) ~ 2
Table 2b: 2010 value of demand of CGIAR commodities weighted by the modified welfare indicator
(income threshold 9000 PPP $, exponent set as 2)
Regional values and shares of commodity
Sub-Saharan Asia Latin America and West Asia and Overall value Share of 
Commodity Africa Share Share Caribbean Share North AfricaShare of commodity grand total
Rice 4708715 7.6% 52468947 84.9% 3445657 5.6% 1172201 1.9% 61795520 15.8%
Wheat 1791056 6.9% 15272602 58.8% 2457037 9.5% 6437137 24.8% 25957832 6.6%
Maize 5245360 26.5% 6138786 31.0% 6614470 33.4% 1793637 9.1% 19792253 5.1%
Barley 309564 13.3% 270648 11.6% 258472 11.1% 1495443 64.1% 2334128 0.6%
Sorghum 2847467 52.8% 1472576 27.3% 886960 16.4% 189814 3.5% 5396817 1.4%
Millet 2662424 62.6% 1555708 36.6% 7545 0.2% 24975 0.6% 4250653 1.1%
Cassava 7773848 72.9% 1077705 10.1% 1805777 16.9% 49 0.0% 10657380 2.7%
Potato 458058 8.9% 2932794 57.3% 903435 17.6% 828327 16.2% 5122614 1.3%
Sweet potato and yam 3814821 69.0% 1502790 27.2% 203267 3.7% 4282 0.1% 5525160 1.4%
Sweet potato 451399 21.8% 1496898 72.4% 115298 5.6% 4282 0.2% 2067878 0.5%
Yam 3363421 97.3% 5892 0.2% 87968 2.5% 0 0.0% 3457282 0.9%
Banana 1390477 23.8% 2873076 49.1% 1496793 25.6% 87399 1.5% 5847746 1.5%
Plantain 2635696 86.4% 45602 1.5% 370949 12.2% 0 0.0% 3052248 0.8%
Pulses 4856807 26.3% 10334089 55.9% 2126655 11.5% 1177081 6.4% 18494632 4.7%
Beans 2409282 29.0% 3821123 46.0% 1926591 23.2% 152596.3 1.8% 8309592 2.1%
Chickpea 356931 7.3% 3771865 77.3% 126018.7 2.6% 621782.5 12.8% 4876597 1.2%
Pigeonpea 344183 17.3% 1624593 81.6% 21359.1 1.1% 0 0.0% 1990135 0.5%
Lentil 48441 4.9% 599302 60.2% 23495 2.4% 323555 32.5% 994793 0.3%
Broad beans 188663 23.7% 507172 63.7% 22071 2.8% 77881 9.8% 795788 0.2%
Cowpea 1509307 98.8% 10034 0.7% 7120 0.5% 1266 0.1% 1527727 0.4%
Soyabean 194470 1.9% 1841788 18.1% 8096180 79.4% 62910 0.6% 10195349 2.6%
Groundnut 2868138 26.8% 7389260 69.1% 363815 3.4% 70925 0.7% 10692138 2.7%
Coconut 274289 9.0% 2579933 84.8% 188983 6.2% 0 0.0% 3043205 0.8%
Beef and buffalo meat 8522888 24.9% 10429568 r 5: 30.5% 12146542 35.5% 3146953 9.2% 34245951 8.8% - S3
Sheep and goat meat 3345380 a>31.5% 4002707 37.7% 630936 5.9% 2642769 24.9% 10621792 2.7% 
Total milk 5978934 13.8% 25530003 58.7% 7827801 18.0% 4135895 9.5% 43472633 11.1% 
Fuelwood and charcoal 34155303 44.8% 29603040 38.9% 11383137 14.9% 1045313 1.4% 76186793 
Roundwood 19.5%2435470 20.5% 4879431 41.1% 4157042 35.0% 400464 3.4%
Fish food 11872407 3.0%4131776 ■ fo 18.5% 15658752 69.9% 1766160 7.9% 832083 3.7% 22388771 5.7% :-  (JnQ>
U\
Grand total by region 100400939 197859809 67137614 25547659 390946022 100.0%
Regional share of grand total 25.7% 50.6% 17.2% 6.5% 100.0%
Formula: VOD * (1 -((1 -Gini coefficient)* Income 2003)/9000)~ 2
Table 2c: 2010 value of demand of CGIAR agricultural commodities weighted by the modified welfare indicator
Income threshold 9000 ppp $, exponent set as 2 Region
Sub-Saharan Asia Latin America and West Asia and Overall value Commoditiy
Com m odity Africa Share Share Caribbean Share North AfricaShare ofcommodity share
Rice 4708715 7.6% 52468947 84.9% 3445657 5.6% 1172201 1.9% 61795520 22.0%
Wheat 1791056 6.9% 15272602 58.8% 2457037 9.5% 6437137 24.8% 25957832 9.3%
Maize 5245360 26.5% 6138786 31.0% 6614470 33.4% 1793637 9.1% 19792253 7.1%
Barley 309564 13.3% 270648 1 1.6% 258472 11.1% 1495443 64.1% 2334128 0.8%
sorghum 2847467 52.8% 1472576 27.3% 886960 16.4% 189814 3.5% 5396817 1.9%
Millet 2662424 62.6% 1555708 36.6% 7545 0.2% 24975 0.6% 4250653 1.5%
Cassava 7773848 72.9% 1077705 10.1% 1805777 16.9% 49 0.0% 10657380 3.8%
Potato 458058 8.9% 2932794 57.3% 903435 17.6% 828327 16.2% 5122614 1.8%
Sweet potato and yam 3814821 69.0% 1502790 27.2% 203267 3.7% 4282 0.1% 5525160 2.0%
Banana 1390477 23.8% 2873076 49.1% 1496793 25.6% 87399 1.5% 5847746 2.1%
Plantain 2635696 86.4% 45602 1.5% 370949 12.2% 0 0.0% 3052248 1.1%
Pulses, overall 4856807 26.3% 10334089 55.9% 2126655 1 1.5% 1177081 6.4% 18494632 6.6%
Soyabean 194470 1.9% 1841788 18.1% 8096180 79.4% 62910 0.6% 10195349 3.6%
Groundnut 2868138 26.8% 7389260 69.1% 363815 3.4% 70925 0.7% 10692138 3.8%
Coconut 274289 9.0% 2579933 84.8% 188983 6.2% 0 0.0% 3043205 1.1%
Beef and buffalo m eat 8522888 24.9% 10429568 30.5% 12146542 35.5% 3146953 9.2% 34245951 12.2%
Sheep and goat meat 3345380 31.5% 4002707 37.7% 630936 5.9% 2642769 24.9% 10621792 3.8%
Total milk 5978934 13.8% 25530003 58.7% 7827801 18.0% 4135895 9.5% 43472633 15.5%
Overall agriculture per region 59678391 147718586 49831275 23269798 280498051 100.0%
Share of region after modification 21.3% 52.7% 17.8% 8.3% 100.0% >3
3
3
<
-d
3
CTQ
3
CT\
Table 2d: Regional and overall shares of CGIAR crops after modification
Region
Sub-Saharan Asia Latin America and West Asia and Overall value Commoditiy
Com m odity Africa Share Share Caribbean Share North AfricaShare of com modify share
Rice 4708715 7.6% 52468947 84.9% 3445657 5.6% 1172201 1.9% 61795520 32.2%
Wheat 1791056 6.9% 15272602 58.8% 2457037 9.5% 6437137 24.8% 25957832 13.5%
Maize 5245360 26.5% 6138786 31.0% 6614470 33.4% 1793637 9.1% 19792253 10.3%
Barley 309564 13.3% 270648 1 1.6% 258472 1 1.1% 1495443 64.1 % 2334128 1.2%
Sorghum 2847467 52.8% 1472576 27.3% 886960 16.4% 189814 3.5% 5396817 2.8%
Millet 2662424 62.6% 1555708 36.6% 7545 0.2% 24975 0.6% 4250653 2.2%
Cassava 7773848 72.9% 1077705 10.1% 1805777 16.9% 49 0.0% 10657380 5.5%
Potato 458058 8.9% 2932794 57.3% 903435 17.6% 828327 16.2% 5122614 2.7%
Sweet potato and yam 3814821 69.0% 1502790 27.2% 203267 3.7% 4282 0.1% 5525160 2.9%
Banana 1390477 23.8% 2873076 49.1 % 1496793 25.6% 87399 1.5% 5847746 3.0%
Plantain 2635696 86.4% 45602 1.5% 370949 12.2% 0 0.0% 3052248 1.6%
Pulses 4856807 26.3% 10334089 55.9% 2126655 1 1.5% 1177081 6.4% 18494632 9.6%
Soyabean 194470 1.9% 1841788 18.1% 8096180 79.4% 62910 0.6% 10195349 5.3%
Groundnut 2868138 26.8% 7389260 69.1% 363815 3.4% 70925 0.7% 10692138 5.6%
Coconut 274289 9.0% 2579933 84.8% 188983 6.2% 0 0.0% 3043205 1.6%
Overall agriculture per region 41831190 107756308 29225996 13344181 192157675 100.0%
Share of region after modification 21.8% 56.1% 15.2% 6.9% 100.0%
>
3
n3x
<
3
CTQ
CO
-J
Table 3a: 2010 value of demand of CGIAR commodities by 
production sector, weighted by the modified welfare indicator
(income threshold value 6000 PPP $, exponent set as 2)
Production sectors (% of total)
Region Crops Livestock Forestry Fisheries Total
Sub - Saharan Africa 12.7% 5.5% 11.2% 1.3% 30.6%
Asia 26.7% 10.8% 8.6% 3.9% 50.0%
C. & S. America 6.3% 4.0% 3.6% 0.4% 14.3%
WANA 2.6% 1.9% 0.3% 0.2% 5.0%
Total 48.3% 22.2% 23.7% 5.7% 100.0%
Regional share (% of sector)
Region Crops Livestock Forestry Fisheries
Sub - Saharan Africa 26.4% 24.6% 47.1% 22.1%
Asia 55.2% 48.7% 36.4% 68.0% >3
C. 3 & S. America 13.1% 17.9% 15.3% 6.8% CPX
WANA 5.3% 8.8% 1.3% 3.1% <
3
Total 100.0% 100.0% 100.0% 100.0% 00CP
00
Formula: VOD * (1 - ((1 -Gini coefficient)*lncome 2003)/6000) ~ 2
Table 3b: 2010 value of demand of CGIAR commodities weighted by the modified welfare indicator
(income threshold 6000 PPP $, exponent set as 2)
Regional values and shares of commodity
Sub-Saharan Asia Latin America and West Asia and Overall value Share of 
Commodity Africa Share Share Caribbean Share North AfrcaShare of commodity total
Rice 4338815 9.2% 39916717 84.4% 2398641 5.1% 655832 1.4% 47310004 15.6%
Wheat 1658751 9.0% 11457808 62.2% 1511090 8.2% 3806032 20.6% 18433681 6.1%
Maize 4842020 36.3% 3021556 22.7% 4350613 32.7% 1107606 8.3% 13321795 4.4%
Barley 292350 21.0% 195439 14.1% 155668 11.2% 747252 53.7% 1390709 0.5%
Sorghum 2636540 59.2% 1272267 28.6% 397537 8.9% 143984 3.2% 4450328 1.5%
Millet 2453098 64.2% 1346668 35.2% 2298 0.1% 19763 0.5% 3821827 1.3%
Cassava 7164660 77.5% 757702 8.2% 1324598 14.3% 33 0.0% 9246993 3.1%
Potato 428274 11.8% 2185266 60.1% 550165 15.1% 472072 13.0% 3635777 1.2%
Sweet potato and yam 3440241 83.7% 513780 12.5% 153534 3.7% 2714 0.1% 4110270 1.4%
Sweet potato 407076 40.4% 511766 50.7% 87089 8.6% 2714 0.3% 1008645 0.3%
Yam 3033165 97.8% 2014 0.1% 66446 2.1% 0 0.0% 3101625 1.0%
Banana 1291156 27.6% 2332026 49.9% 998153 21.3% 54225 1.2% 4675560 1.5%
Plantain 2416397 89.8% 29190 1.1% 245296 9.1% 0 0.0% 2690883 0.9%
Pulses 4508986 29.4% 8719108 56.8% 1441726 9.4% 679190 4.4% 15349010 5.1%
Beans 2236741 32.6% 3223969 47.0% 1306097 19.1% 88050 1.3% 6854857 2.3%
Chickpea 331369 8.4% 3182409 80.4% 85432 2.2% 358776 9.1% 3957986 1.3%
Piqeonpea 319534 18.7% 1370706 80.4% 14480 0.8% 0 0.0% 1704721 0.6%
Lentil 44972 6.0% 505645 67.1% 15928 2.1% 186695 24.8% 753240 0.2%
Broad beans 175152 26.4% 427913 64.5% 14963 2.3% 44938 6.8% 662966 0.2%
Cowpea 1401218 99.0% 8466 0.6% 4827 0.3% 731 0.1% 1415241 0.5%
Soyabean 175726 2.7% 1103695 16.7% 5303084 80.2% 27811 0.4% 6610316 2.2%
Groundnut 2647693 30.2% 5872862 66.9% 215778 2.5% 38936 0.4% 8775269 2.9%
Coc onut 250387 10.9% 1927169 83.9% 118811 5.2% 0 0.0% 2296367 0.8% >
Beef and buffalo meat 7852640 32.0% 7965449 32.4% 6910054 28.1% 1839815 7.5% 24567958 8.1% 33
Sheep and goat meat 3100505 38.2% 3026374 37.2% 382254 4.7% 1617011 19.9% 8126144 2.7% c
Total milk 5563485 16.1% 21751704 63.1% 4735356 13.7% 2439120 7.1% 34489665 11.4% 
Fuelwood and charcoal 31568262 49.6% 23116086 36.3% 8228609 12.9% 698826 1.1% 63611784 21.0% <
Roundwood 2198113 27.1% 2960783 36.4% 2759346 34.0% 206898 2.5% 8125141 2.7%
Fish food 3823578 22.1% 11797125 68.0% 1178687 6.8% 540952 3.1% 17340342 5.7% 3CTQ
Grand total by region 92651677 151268773 43361300 VO15098073 302379823 100.0%
Regional share of grand total 30.6% 50.0% 14.3% 5.0% 100.0%
Formula: VOD * (1 -(1 -Gini)*PPP in 2003/6000)) ~ 2
Table 3c: 2010 value of demand of CGIAR agriculturalcommodities weighted by the modified welfare indicator
(income threshold 6000 PPP $, exponent set as 2) Region
Sub-Saharan Asia Latin America and West Asia and Overall value Commoditiy 
Com m odity Africa Share Share Caribbean Share North AfricaShare of commodity share
Rice 4338815 9.2% 39916717 84.4% 2398641 5.1% 655832 1.4% 47310004 22.2% 
Wheat 1658751 9.0% 11457808 62.2% 1511090 8.2% 3806032 20.6% 18433681 8.6% 
Maize 4842020 36.3% 3021556 22.7% 4350613 32.7% 1107606 8.3% 13321795 6.2% 
Barley 292350 21.0% 195439 14.1 % 155668 1 1.2% 747252 53.7% 1390709 0.7% 
sorghum 2636540 59.2% 1272267 28.6% 397537 8.9% 143984 3.2% 4450328 2.1% 
Millet 2453098 64.2% 1346668 35.2% 2298 0.1% 19763 0.5% 3821827 1.8% 
Cassava 7164660 77.5% 757702 8.2% 1324598 14.3% 33 0.0% 9246993 4.3% 
Potato 428274 11.8% 2185266 60.1% 550165 15.1% 472072 13.0% 3635777 1.7% 
Sweet potato and yam 3440241 83.7% 513780 12.5% 153534 3.7% 2714 0.1% 4110270 1.9% 
Banana 1291156 27.6% 2332026 49.9% 998153 21.3% 54225 1.2% 4675560 2.2% 
Plantain 2416397 89.8% 29190 1.1% 245296 9.1% 0 0.0% 2690883 1.3% 
Pulses, overall 4508986 29.4% 8719108 56.8% 1441726 9.4% 679190 4.4% 15349010 7.2% 
Soyabean 175726 2.7% 1103695 16.7% 5303084 80.2% 2781 1 0.4% 6610316 3.1% 
Groundnut 2647693 30.2% 5872862 66.9% 215778 2.5% 38936 0.4% 8775269 4.1% 
Coconut 250387 10.9% 1927169 83.9% 118811 5.2% 0 0.0% 2296367 1.1% 
Beef and buffalo meat 7852640 32.0% 7965449 32.4% 6910054 28.1% 1839815 7.5% 24567958 1 1.5% 
Sheep and goat meat 3100505 38.2% 3026374 37.2% 382254 4.7% 1617011 19.9% 8126144 3.8% 
Total milk 5563485 16.1% 21751704 63.1% 4735356 13.7% 2439120 7.1% 34489665 16.2%
Overall agriculture per region 55061723 113394780 31194657 13651397 213302556 1 0 QpO %
Share of region after modification 25.8% 53.2% 14.6% 6.4% 100.0% 3
n3x>
Formula: VOD * (1 -(1 -Gini)* PPP 2003)/6000)) ~ 2
<
OQ
<D
o
Table 4a: 2010 value of demand of CGIAR commodities by 
production sector, weighted by the modified welfare indicator
(income threshold 6000 PPP $, exponent set as 3)
Production sectors (% of total)
Region Crops Livestock Forestry Fisheries Total
Sub-Saharan Africa 14.2% 6.2% 12.5% 1.4% 34.3%
Asia 26.9% 11.0% 8.5% 4.0% 50.4%
C. & S. America 5.1% 3.0% 3.1% 0.3% 11.5%
WANA 1.9% 1.5% 0.2% 0.1% 3.8%
Total 48.2% 21.7% 24.3% 5.8% 100.0%
Regional share (% of sector)
Region Crops Livestock Forestry Fisheries
Sub-Saharan Africa 29.5% 28.4% 51.5% 24.3%
Asia 55.9% 50.8% 34.9% >67.9% 33
C. & S. America CD 10.7% 14.0% 12.6% 5.4% X
WANA 3.9% 6.9% 1.0% 2.4% <
< 3k>
Total 100.0% 100.0% 100.0% 100.0% CD
Formula: VOD * (1 - ((1-Gini coefficient)*Income 2003)/6000) ^ 3
Table 4b: 2010 value of demand of CGIAR commodities weighted by the modified welfare indicator
(income threshold 6000 PPP $, exponent set as 3)
Regional values and shares of commodities
Sub-Saharan Asia Latin America and West Asia and Overall value Share of 
Com m odity Africa Share Share Caribbean Share North AfricaShare of commodity total
Rice 3872990 10.2% 31978429 84.6% 1574528 4.2% 385639 1.0% 37811586 15.6%
Wheat 1497258 10.6% 9398100 66.4% 963511 6.8% 2287157 16.2% 14146026 5.9%
M aize 4336175 43.3% 2232154 22.3% 2792608 27.9% 661649 6.6% 10022586 4.1%
Barley 271034 29.7% 159359 17.5% 95455 10.5% 386866 42.4% 912713 0.4%
Sorghum 2368878 63.1% 1082910 28.8% 199283 5.3% 105481 2.8% 3756552 1.6%
Millet 2187063 65.4% 1141858 34.1% 819 0.0% 15789 0.5% 3345529 1.4%
Cassava 6402648 81.7% 537813 6.9% 895138 11.4% 20 0.0% 7835619 3.2%
Potato 390039 13.8% 1812404 64.3% 342822 12.2% 272524 9.7% 2817789 1.2%
Sweet potato and yam 2975172 85.4% 391447 1 1.2% 115982 3.3% 1539 0.0% 3484140 1.4%
Banana 1167976 31.2% 1880480 50.3% 656921 17.6% 33353 0.9% 3738729 1.5%
Plantain 2141182 92.3% 17152 0.7% 161760 7.0% 0 0.0% 2320094 1.0%
Pulses 4067167 32.0% 7288611 57.4% 948074 7.5% 386612 3.0% 12690463 5.3%
Soyabean 152402 3.4% 881620 19.6% 3454748 76.7% 12590 0.3% 4501359 1.9%
Groundnut 2368806 32.0% 4872469 65.8% 142526 1.9% 20187 0.3% 7403988 3.1%
Coconut 220320 12.9% 1417482 82.8% 73957 4.3% 0 0.0% 1711758 0.7%
Beef and buffalo meat 7033895 37.8% 6348658 34.1% 4157306 22.3% 1068251 5.7% 18608110 7.7%
Sheep and goat meat 2794179 43.3% 2359874 36.6% 244574 3.8% 1051311 16.3% 6449939 2.7%
Total milk 5042621 18.4% 17901246 65.5% 2919130 10.7% 1469529 5.4% 27332526 1 1.3%
Fuelwood and charcoal 28318173 53.7% 18298859 34.7% 5586479 10.6% 483461 0.9% 52686972 21.8%^
Roundwood 1916938 32.0% 2158102 36.1% 1791170 29.9% 119863 2.0% 5986074 2.5% g 
Fish food 3433515 24.3% 9604567 67.9% 763340 5.4% 334407 2.4% 14135830 5.8% x „ 
Grand total by region 82958432 121763594 27880130 9096228 241698384 1 00.0%
Regional share of grand total 34.3% 50.4% 11.5% 3.8% 1 00.0% p(TO
CD
Formula: VOD * (1 — ((1 — Gini)* PPP 2003)/6000) ~ 3 K)
Table 4c: Regional and overall shares of agriculturalcommodities after modification
Income threshold set as 6000, exponent as 3 Region
Sub-Saharan Asia Latin America and West Asia and Overall value Commoditiy
Com m odity Africa Share Share Caribbean Share North AfricaShare ofcommodity share
Rice 3872990 10.2% 31978429 84.6% 1574528 4.2% 385639 1.0% 37811586 22.4% 
Wheat 1497258 10.6% 9398100 66.4% 963511 6.8% 2287157 16.2% 14146026 8.4% 
Maize 4336175 43.3% 2232154 22.3% 2792608 27.9% 661649 6.6% 10022586 5.9% 
Barley 271034 29.7% 159359 17.5% 95455 10.5% 386866 42.4% 912713 0.5% 
sorghum 2368878 63.1% 1082910 28.8% 199283 5.3% 105481 2.8% 3756552 2.2% 
Millet 2187063 65.4% 1141858 34.1% 819 0.0% 15789 0.5% 3345529 2.0% 
Cassava 6402648 81.7% 537813 6.9% 895138 11.4% 20 0.0% 7835619 4.6% 
Potato 390039 13.8% 1812404 64.3% 342822 12.2% 272524 9.7% 2817789 1.7% 
Sweet potato and yam 2975172 85.4% 391447 1 1.2% 115982 3.3% 1539 0.0% 3484140 2.1% 
Banana 1167976 31.2% 1880480 50.3% 656921 17.6% 33353 0.9% 3738729 2.2% 
Plantain 2141182 92.3% 17152 0.7% 161760 7.0% 0 0.0% 2320094 1.4% 
Pulses, overall 4067167 32.0% 7288611 57.4% 948074 7.5% 386612 3.0% 12690463 7.5% 
Soyabean 152402 3.4% 881620 19.6% 3454748 76.7% 12590 0.3% 4501359 2.7% 
Groundnut 2368806 32.0% 4872469 65.8% 142526 1.9% 20187 0.3% 7403988 4.4% 
Coconut 220320 12.9% 1417482 82.8% 73957 4.3% 0 0.0% 1711758 1.0% 
Beef and buffalo meat 7033895 37.8% 6348658 34.1% 4157306 22.3% 1068251 5.7% 18608110 1 1.0% 
Sheep and goat meat 2794179 43.3% 2359874 36.6% 244574 3.8% 1051311 1 6.3% 6449939 3.8% 
Total milk 5042621 18.4% 17901246 65.5% 2919130 10.7% 1469529 5.4% 27332526 16.2%
Overall agriculture per region 49289804 91702065 19739141 8158497 168889508 100.0%
Share of region after modification 29.2% 54.3% 11.7% 4.8% 100.0%
Formula: VOD * (1 -((1 -Gini)*PPP 2003)/6000) ~3
N>
CT
Table 5a: 2010 value of demand of CGIAR commodities by 
production sector, weighted by the modified welfare indicator and 
by the agricultural producers’ gender
(income threshold set as 6000 PPP $, exponent as 2)
Production sectors (% of total)
Region Crops Livestock Forestry Fisheries Total
Sub-Saharan Africa 13.2% 5.6% 11.5% 1.3% 31.6%
Asia 26.5% 10.7% 8.6% 3.9% 49.7%
C. & S. America 6.1% 3.8% 3.5% 0.4% 13.8%
WANA 2.5% 1.9% 0.3% 0.2% 4.9%
Total 48.3% 22.1% 23.9% 5.7% 100.0%
Regional share (% of sector)
Region Crops Livestock Forestry Fisheries
Sub-Saharan Africa 27.2% 25.4% 48.2% 22.8% >a
Asia 55.0% 48.6% 36.0% 67.6% an>x
C. & S. America 12.6% 17.3% 14.6% 6.6% <
WANA 5.2% 8.6% 1.2% 3.0% ►d
a
crta
n>
Total 100.0% 100.0% 100.0% 100.0% u>
Formula: VOD * (1 - ((1 -Gini coefficient)*lncome 2003)/6000) ~ 3 * (F + 0.8 M)
Table 5b: 2010 values of demand of CGlAR commodities weighted by the modified welfare indicator and by producers’ gender
(income threshold set at 6000 PPP $, exponent at 2)
Regional values and shares of commodity
Sub - Saharan Asia Latin America and West Asia and Overall value Share of 
Com m odity Africa Share Share Caribbean Share North AfricaShare of commodity total
Rice 3830137 9.5% 34008556 84.3% 1977254 4.9% 543916 1.3% 40359863 15.6%
Wheat 1459633 9.4% 9709760 62.2% 1246468 8.0% 3185397 20.4% 15601258 6.0%
Maize 4292543 37.7% 2586841 22.7% 3583770 31.5% 911704 8.0% 11374858 4.4%
Barley 257596 21.6% 166658 13.9% 128292 10.7% 642230 53.8% 1194775 0.5%
Sorghum 2302849 60.1% 1086864 28.3% 326909 8.5% 117624 3.1% 3834245 1.5%
Millet 2146542 64.8% 1150349 34.7% 1893 0.1% 16188 0.5% 3314972 1.3%
Cassava 6371374 78.5% 652869 8.0% 1090812 13.4% 27 0.0% 8115081 3.1%
Potato 381503 12.3% 1865918 60.2% 453329 14.6% 396270 12.8% 3097020 1.2% 
Sweet potato and yam 3020294 84.0% 443801 12.3% 128780 3.6% 2190 0.1% 3595066 1.4% 
Banana 1155856 28.8% 1990788 49.6% 822238 20.5% 44638 1.1% 4013521 1.6% 
Plantain 2145288 90.5% 2491 1 1.1% 201043 8.5% 0 0.0% 2371242 0.9% 
Pulses 3990830 30.3% 7433390 56.4% 1189054 9.0% 573197 4.3% 13186470 5.1% 
Soyabean 154429 2.8% 945991 17.2% 4372456 79.5% 23635 0.4% 5496512 2.1% 
Groundnut 2327808 30.8% 5023512 66.4% 178192 2.4% 33279 0.4% 7562792 2.9% 
Coconut 223741 11.4% 1643114 83.6% 98056 5.0% 0 0.0% 1964910 0.8% 
Beef and buffalo meat 6919428 33.1 % 6773758 32.4% 5691049 27.2% 1528464 7.3% 20912698 8.1% 
Sheep and goat meat 2719878 39.3% 2540329 36.7% 316517 4.6% 1342625 19.4% 6919348 2.7% 
Total milk 4889712 16.7% 18448177 63.0% 3896559 13.3% 2035217 7.0% 29269666 11.3%> 
Fuelwood and charcoal 27895409 50.7% 19721075 35.9% 6776091 12.3% 576546 1.0% 54969122 21.2%g 
Roundwood 1940052 28.0% 2537488 36.6% 2274817 32.8% 175961 2.5% 6928318 2.7%>< 
Fish food 3375728 22.8% 10031437 67.6% 972885 6.6% 450466 3.0% 14830515 5.7%3
Grand total by region 81800631 128785584 35726464 12599575 258912254 100.0%$
Regional share of grand total 31.6% 49.7% 13.8% 4.9% 100.0% ffqCD
Formula: VOD * (1 -((1 - Gini)* PPP 2003)/6000) ~ 2 * (F + 0.8 M)
Table 5c: Shares of agriculturalcommoditiesafter modificationfor poverty and producers’gender
Income threshold set at 6000, exponent at 2 Region
Sub-Saharan Asia Latin America and West Asia and Overall value Commoditiy
Com m odity Africa Share Share Caribbean Share North AfricaShare ofcommodity share
Rice 3830137 9.5% 34008556 84.3% 1977254 4.9% 543916 1.3% 40359863 22.2% 
Wheat 1459633 9.4% 9709760 62.2% 1246468 8.0% 3185397 20.4% 15601258 8.6% 
Maize 4292543 37.7% 2586841 22.7% 3583770 31.5% 911704 8.0% 11374858 6.2% 
Barley 257596 21.6% 166658 13.9% 128292 10.7% 642230 53.8% 1194775 0.7% 
sorghum 2302849 60.1% 1086864 28.3% 326909 8.5% 117624 3.1% 3834245 2.1% 
Millet 2146542 64.8% 1150349 34.7% 1893 0.1% 16188 0.5% 3314972 1.8% 
Cassava 6371374 78.5% 652869 8.0% 1090812 13.4% 27 0.0% 8115081 4.5% 
Potato 381503 12.3% 1865918 60.2% 453329 14.6% 396270 12.8% 3097020 1.7% 
Sweet potato and yam 3020294 84.0% 443801 12.3% 128780 3.6% 2190 0.1% 3595066 2.0% 
Banana 1155856 28.8% 1990788 49.6% 822238 20.5% 44638 1.1% 4013521 2.2% 
Plantain 2145288 90.5% 2491 1 1.1% 201043 8.5% 0 0.0% 2371242 1.3% 
Pulses, overall 3990830 30.3% 7433390 56.4% 1189054 9.0% 573197 4.3% 13186470 7.2% 
Soyabean 154429 2.8% 945991 17.2% 4372456 79.5% 23635 0.4% 5496512 3.0% 
Groundnut 2327808 30.8% 5023512 66.4% 178192 2.4% 33279 0.4% 7562792 4.2% 
Coconut 223741 1 1.4% 1643114 83.6% 98056 5.0% 0 0.0% 1964910 1.1% 
Beef and buffalo meat 6919428 33.1% 6773758 32.4% 5691049 27.2% 1528464 7.3% 20912698 11.5% 
Sheep and goat meat 2719878 39.3% 2540329 36.7% 316517 4.6% 1342625 19.4% 6919348 3.8% 
Total milk 4889712 16.7% 18448177 63.0% 3896559 13.3% 2035217 7.0% 29269666 16.1%
Overall agriculture per region 48589442 96495585 25702671 11396601 182184298 100.0%
Share of region after modification 26.7% 53.0% 14.1% 6.3% 100.0%
Formula: VOD * (1 - ((1 - Gini)* PPP 2003)/6000) ~ 2 * (F + 0.8 M)
CP
Table 6a: 2010 value of demand of CGIAR commodities by 
production sector, weighted by the modified welfare indicator 
and by the rural/urban population distribution
(income threshold set at 6000 PPP$, exponent at 2)
Production sectors (% of total)
Region Crops Livestock Forestry Fisheries Total
Sub - Saharan Africa 12.9% 5.6% 11.3% 1.3% 31.1%
Asia 27.1% 11.0% 8.8% 4.0% 50.8%
C. & S. America 5.8% 3.7% 3.4% 0.4% 13.2%
WANA 2.5% 1.9% 0.3% 0.2% 4.8%
Total 48.4% 22.1% 23.7% 5.8% 100.0%
Regional share (% of sector)
Region Crops Livestock Forestry Fisheries
Sub - Saharan Africa 26.7% 25.1% 47.7% 22.2% >
Asia a56.1% 49.7% 36.9% 68.5% aCxD
C. & S. America 12.1% 16.6% 14.1% 6.3% <
WANA 5.1% 8.6% 1.2% 3.0%
aTJOP
CD
Total 100.0% 100.0% 100.0% 100.0%
Formula: VOD*(1 — ((1 — Gini coefficient)*lncome 2003)/6000) /V2*(R+0.8U)
Table 6b: 201 0 value of demand of CGlAR commodities weighted by the modified welfare indicator and by the population distribution
(income threshold set at 6000 PPP $, exponent at 2)
Regional values and shares of commodity 
Sub-Saharan Asia Latin America and West Asia and Overall value Share of 
Com m odity Africa Share Share Caribbean Share North AfricaShare of commodity total
Rice 4063377 9.1% 37844741 84.9% 2055608 4.6% 594879 1.3% 44558604 15.8%
Wheat 1573928 9.2% 10835138 63.2% 1299980 7.6% 3433049 20.0% 17142096 6.1%
Maize 4583760 37.8% 2816934 23.2% 3732853 30.8% 1002716 8.3% 12136262 4.3%
Barley 281233 22.3% 183553 14.6% 133572 10.6% 661593 52.5% 1259951 0.4%
Sorghum 2490003 59.7% 1205148 28.9% 344040 8.2% 132616 3.2% 4171807 1.5%
Millet 2312313 64.1 % 1276534 35.4% 1942 0.1% 1 8564 0.5% 3609352 1.3%
Cassava 6728556 78.4% 711869 8.3% 1137497 13.3% 29 0.0% 8577951 3.0%
Potato 412381 12.2% 2070468 61.3% 472773 14.0% 423772 12.5% 3379395 1.2% 
Sweet potato and yam 3215793 83.7% 485680 12.6% 136259 3.5% 2474 0.1% 3840206 1.4% 
Banana 1238442 28.5% 2192502 50.5% 864558 19.9% 49221 1.1% 4344723 1.5% 
Plantain 2315532 90.5% 2781 6 1.1% 216609 8.5% 0 0.0% 2559958 0.9% 
Pulses 4303193 29.8% 8261710 57.3% 1243140 8.6% 608961 4.2% 14417004 5.1% 
Soyabean 165729 2.9% 1042870 18.1% 4512701 78.5% 24842 0.4% 5746142 2.0% 
Groundnut 2484050 30.0% 5562381 67.3% 187622 2.3% 34567 0.4% 8268620 2.9% 
Coconut 234827 1 1.0% 1794139 84.2% 102824 4.8% 0 0.0% 2131790 0.8% 
Beef and buffalo meat 7427034 33.0% 7529282 33.4% 5917240 26.3% 1659495 7.4% 22533051 8.0% 
Sheep and goat meat 2938589 38.7% 2851899 37.5% 330531 4.4% 1475761 19.4% 7596780 2.7% 
Total milk 5281033 16.5% 20561858 64.1% 4058818 12.6% 2193457 6.8% 32095166 11.4% 
Fuelwood and charcoal 29803601 50.2% 21870954 36.8% 7096244 1 1.9% 646922 1.1% 59417721 2^^%> 
Roundwood 2062651 27.9% 2787984 37.7% 2351117 31 .8% 188039 2.5% 7389791 2.6%g 
Fish food 3616508 22.2% 11156820 68.5% 1017787 6.3% 491775 3.0% 16282890 5.8%S
Grand total by region 87532532 143070280 37213717 13642731 281459261 100.0%
Regional share of grand total 31.1% 50.8% 13.2% 4.8% 1 00.0% p
CnT>Q
Formula: VOD * (1 - ((1 - Gini)*PPP 2003)/6000) ~ 2 * (R + 0.8 U) CT\
Table 6c: Shares of agriculturalcommodities after modification by poverty and population distribution
(income threshold set at 6000 PPP, exponent at 2) Region
Sub-Saharan Asia Latin America and West Asia and Overall value Commoditiy
Com m odity Africa Share Share Caribbean Share North AfricaShare of commodity share
Rice 4063377 9.1% 37844741 84.9% 2055608 4.6% 594879 1.3% 44558604 22.5% 
Wheat 1573928 9.2% 10835138 63.2% 1299980 7.6% 3433049 20.0% 17142096 8.6% 
Maize 4583760 37.8% 2816934 23.2% 3732853 30.8% 1002716 8.3% 12136262 6.1% 
Barley 281233 22.3% 183553 14.6% 133572 10.6% 661593 52.5% 1259951 0.6% 
sorghum 2490003 59.7% 1205148 28.9% 344040 8.2% 132616 3.2% 4171807 2.1% 
Millet 2312313 64.1 % 1276534 35.4% 1942 0.1% 18564 0.5% 3609352 1.8% 
Cassava 6728556 78.4% 711869 8.3% 1137497 13.3% 29 0.0% 8577951 4.3% 
Potato 412381 12.2% 2070468 61.3% 472773 14.0% 423772 12.5% 3379395 1.7% 
Sweet potato and yam 3215793 83.7% 485680 12.6% 136259 3.5% 2474 0.1%
Banana 3840206 1.9% 1238442 28.5% 2192502 50.5% 864558 19.9% 49221 1.1% 4344723 2.2% 
Plantain 2315532 90.5% 27816 1.1% 216609 8.5% 0 0.0% 2559958 1.3% 
Pulses, overall 4303193 29.8% 8261710 57.3% 1243140 8.6% 608961 4.2% 14417004 7.3% 
Soyabean 165729 2.9% 1042870 18.1% 4512701 78.5% 24842 0.4% 5746142 2.9% 
Groundnut 2484050 30.0% 5562381 67.3% 187622 2.3% 34567 0.4% 8268620 4.2% 
Coconut 234827 11.0% 1794139 84.2% 102824 4.8% 0 0.0% 2131790 1.1% 
Beef and buffalo meat 7427034 33.0% 7529282 33.4% 5917240 26.3% 1659495 7.4% 22533051 1 1.4% 
Sheep and goat meat 2938589 38.7% 2851899 37.5% 330531 4.4% 1475761 19.4% 7596780 3.8% 
Total milk 5281033 16.5% 20561858 64.1% 4058818 12.6% 2193457 6.8% 32095166 16.2%
Overall agriculture per region 52049773 107254521 26748570 12315995 198368859 100.0%
Share of region after modification 26.2% 54.1% 13.5% 6.2% 100.0%
Formula: VOD * (1 -((1 - Gini) *P P P 2003)/6000) ~2 * (R + 0.8 U)
CP
C5~
Annex IV - Page 17
Table 7: CGIAR commodities’share of value of demand (VOD) and its modifications
Share of Share of Share of Trend for 
Commodity VOD modified VOD modified VOD the
(z=9000) (z=6000) commodity
Rice 17.5% 15.8% 15.6% 1
Wheat 8.2% 6.6% 6.1% i
Maize 6.3% 5.1% 4.4% 1
Barley 0.9% 0.6% 0.5% 1
Sorghum 1.1% 1.4% 1.5% t
Millet 0.7% 1.1% 1.3% t
Cassava 1.9% 2.7% 3.1% f
Potato 1.6% 1.3% 1.2% i
Sweet potato and yam 2.0% 1.4% 1.4% i
Sweet potato 1.5% 0.5% 0.3% 1
Yam 0.6% 0.9% 1.0% t
Banana 1.3% 1.5% 1.5%
Plantain 0.5% 0.8% 0.9% t
Pulses 3.8% 4.7% 5.1% t
Beans 1.7% 2.1% 2.3%
Chickpea 1.1% 1.2% 1.3%
Pigeon pea 0.4% 0.5% 0.6%
Lentil 0.2% 0.3% 0.2%
Broad beans 0.2% 0.2% 0.2%
Cowpea 0.2% 0.4% 0.5%
Soyabean 3.3% 2.6% 2.2% i
Groundnut 2.6% 2.7% 2.9%
Coconut 0.7% 0.8% 0.8%
Beef and buffalo meat 9.4% 8.8% 8.1% 1
Sheep and goat meat 2.7% 2.7% 2.7%
Total milk 9.6% 11.1% 11.4% t
Fuelwood and charcoal 15.4% 19.5% 21.0% t
Roundwood 4.3% 3.0% 2.7% I
Fish food 6.1% 5.7% 5.7%
Total 100.0% 100.0% 100.0%
Table 8: Allocation of research funds to CGIAR crops 
compared to modified values after quantitative analysis
Commodity Commodity Share Expenditure Share
value, modified in 1996 
mill US $
Rice 61795520 32.2% 21.8 24.3% 
Wheat 25957832 13.5% 9.7 10.8% 
Maize 19792253 10.3% 12.7 14.1% 
Barley 2334128 1.2% 3.06 3.4% 
sorghum 5396817 2.8% 3.6 4.0% 
Millet 4250653 2.2% 2.6 2.9% 
Cassava 10657380 5.5% 7.6 8.5% 
Potato 5122614 2.7% 3.9 4.3% 
Sweet potato and yam 5525160 2.9% 2.5 2.8% 
Banana and plantain 8899994 4.6% 7.2 8.0% 
Beans 8263559 4.3% 4.7 5.2% 
Chickpea 4722034 2.5% 2.2 2.4% 
Pigeonpea 1967514 1.0% 1.4 1.6% 
Lentil** 1180508 0.6% 0.8 0.9%
Broad (faba) bean 787006 0.4% 
Cowpea* 1574011 0.8% 1.0 1.1% 
Pulses, CGIAR crops ■*■*■*■ 18494632 9.6% 10.1 11.3% 
Soyabean* 10195349 5.3% 1.7 1.9% 
Groundnut 10692138 5.6% 3.4 3.8% 
Coconut 3043205 1.6% n. a. n. a.
>
3
Grand total 192157675 100.0% 89.89 100.0% n3>
Note: * IITA allocation for soyabean and cowpea split according <
to ratio in PS-report 1992, page 161 73
** CGIAR expenditure for lentils and faba bean can't be split 3
★ ★ * 0QPulses split according to regional values 1992/44 production data n
00
Table 9: Selected key indicators for population, land area and use, and agriculture by region
Landarea Population Population Population Food Food Production Production Rainfed Irrigated Total
10 ~6 ha 1993 2010* Growth Demand Demand Food 1993 Cash 1993 Arable Arable A T3 b 16
(10~6) (10~6) per annum 1993 2010* (10 ^6 ha) (10 ^6 ha) (10 ^6 ha)
(10A6 tGE) (10~6 tGE) (10~6 tGE) (10~6 tGE)
Sub-Saharan Africa 2241 520 874 3.10% 119 267 112 32 127 6 132
West Asia/North Africa 1253 346 513 2.35% 114 170 100 20 63 26 89
Asia 2003 2927 3687 1.37% 781 1089 888 185 226 134 360
Latin America and 2015 464 593 1.45% 136 186 218 50 107 17 124
Caribbean
Overall 7512 4256 5667 1.70% 1150 1712 1317 286 522 183 705
Notes: * FAO AT 2010 study
GE: Grain equivalents based on wheat energy (food)/price (cash)
>
n§x>
>—I 
<
TJ
P
OP
CD
VO
Annex IV - Page 20
Table 10: Selected socioeconomic indicators by region
Indicator SSA Asia LAC WANA Absolute value 
million
Population (% of LDC total) 12.2 68.8 10.9 8.1 4256
Number of poor (% of LDC total) 20.8 59.1 13.5 6.7 1332
Share of urban population 28 29 73 55 1515
Daily calorie supply (Kcal/cap, 1988-90) 2096 2441 2686 3011
Calorie supply to requirement (%, 1988-90) 90 109 114 125
Food import dependency ratio (%, 1988-90) 10.5 6 14 40
Income/caput (US $, 1991) 393 525 2461 2081
Income/caput (Int’l $, 1 991) 1049 2448 5371 4471
Arable land (%) 19 51 18 13 705
Irrigated land (%) 3 73 9 14 183
Use of fertilizer (kg/ha) 7.2 82.8 35.1 49.1
Agr. GDP/agr. labourer (US $) 278 277 2340 1710
Agr. GDP/total GDP (%) 32 21 11 15
Agr. land/labour ratio (ha/worker) 5.4 1.0 16.0 10.3
Arable land/labour ratio (ha/worker) 0.8 0.4 2.7 2.3
Deforestation (1988-90, % p.a.) 1.7 0.9 1.4 1.0 16.8 m. ha
Total wooded area (1991 -93, m. ha) 739 510 921 52 2222
(closed, open and forest fallow)
ANNEX V
SYNTHESIS TABLES ON SYSTEMWIDE INITIATIVES/PROGRAMMES
1. Attached are draft tables on systemwide initiatives/programmes prepared in support 
of the review of CGIAR priorities and strategies. The format follows that agreed to in our 
meeting with Don Winkelmann, Ted Henzell, and Hans Gregersen on January 26, 1996. 
Table 1 presents quantitative data on SWI/P budgets and collaboration. Table 2 presents a 
qualitative synthesis of SWI/P content, management, and expected results, This
memorandum summarizes and provides a preliminary analysis of the data presented and 
recommends steps to be taken to develop a more complete database on Systemwide activities.
2. The data for both tables were drawn from the following sources: Review of Proposals 
for Systemwide and Ecoregional Initiatives, September 8, 1994 (AGR/TAC:IAR/94/ll); TAC 
Commentaries on Systemwide Initiatives contained in the Reports of TAC 66, TAC 67, and 
TAC 68; and a review of the latest versions of SWI proposals (most of these were annexed 
to Centres’ 1996 Funding Requirements submissions).
Table 1.
Comments on the table:
3. As noted in Ted Henzell’s Chapter 6 of the draft report on priorities and strategies, 
the distinction between "initiative" and "programme" costs is ambiguous. This presented 
problems in apportioning the TAC recommended funding for 1995/96 between these two 
categories. We have consulted the TAC Commentaries for each recommended allocation 
and, on the basis of the intent expressed, made an estimate as to how much of the 
recommended allocation was considered by TAC to be for "initiative" (i.e., design, 
facilitation) and "programme" (i.e., implementation of research and research-related 
activities) activities. These estimates will need to be verified or, in cases where an estimate 
has not been possible, supplied by the centres.
4. The columns for non-CGIAR funds and other IARC costs are gaps in our in-house 
data. The above cited documentation contains, in some cases, references to actual or
anticipated funding from non-CGIAR sources. Given the selective, sometimes tentative, and 
probably dated nature of these figures, inserting them in the table could result in error. 
Henzell’s Chapter 6 does make reference to both CGIAR and, in some cases, non-CGIAR 
resources, apparently based on CG Secretariat data. We are checking these numbers against 
available in-house sources. However, for the table, it would probably be best to check with 
the CG Secretariat and/or canvass the centres directly. In that case, we would need to decide 
whether we want data for individual years, the MTP period, and/or the duration of the 
initiative/program me.
Annex V - Page 2
5. The data on collaborating CGIAR and non-CGIAR organizations is partially 
incomplete because, in some cases, the initiative had not reached the stage where specific 
collaborators had been identified/committed. Blank cells in the table generally mean the data 
are not available for these items. (See, for example, 2.3.1. Tropical America; 4.1. 
Integrated Pest Management.)
6. Comparing initial MTP allocations with actuals for 1995/1996, TAC has exceeded the 
initial target in the following cases:
1.1. Genetic Resources 
2.1.1. Desert Margins 
2.1.2. Africa Humid/Sub-Humid 
2.2.1. Rice/Wheat
2.2.2. Asia Humid/Sub-Humid Tropics 
2.3.1. Tropical America 
2.4.1. Water Husbandry 
3.2. Slash and Bum
7. TAC’s initial MTP allocations for these initiatives totalled $4.75 million; actual 
allocations were $9.44 million, or $4.70 million beyond the plan. The largest divergence 
was Genetic Resources whose actual allocation was more than three times the MTP target 
of $1.0 million. In most of the other cases, the target was exceeded on average by $0.30 
million.
8. There were two initiatives for which allocations thus far had not reached the MTP
level:
1.2. Livestock
1.3. Water Management
The MTP initial allocations for these initiatives totalled $5.00 million; actual allocations 
totalled $3.55 million, or $1.45 million under allocation.
9. There were 6 cases where either no initial MTP allocation was made or the initiative 
had not been identified at the time MTP funding recommendations were made:
1.4 Property Rights 
1.5. Agricultural Indicators 
3.1 Mountain Agriculture 
3.3 Coastal Environments 
4.1 IPM 
4.2 SWNM
TAC’s subsequent allocations to these initiatives totalled some $3.0 million. Nearly half of 
this was allocated to the inter-centre initiatives on IPM and SWNM; the balance was divided 
somewhat evenly between the other four initiatives.
Annex V - Page 3
10. Overall comparison of planned vs. actual allocations for all initiatives are summarized 
below by the four categories of initiatives/programmes used in the table:
Category Initial MTP 1995/1996 Over
Allocation Allocation Allocation
US$ Millions US$ Millions
US$ Millions
1. Global 6.00 7.96 + 1.96 
(+32%)
2. Ecoregional/Regional 3.25 5.10 + 1.85 
(+56%)
3. Global/Cross-Ecoregional 0.75 1.65 + 0.90 
(+120%)
4. Other NA 1.30 + 1.30
(NA)
Totals 10.00 16.01 + 6.01 
(+60%)
It is difficult to discern whether these changes in allocation represent shifts in TAC’s 
prioritization of SWIs, particularly in the absence of any rationale for the initial pattern of 
allocation. Although category 1 (Global) at nearly $8.0 million represents the largest actual 
allocation, the two closely related categories 2 (Ecoregional/Regional) and 3 (Global/Cross 
Ecoregional) together accounted for nearly $7.0 million. This amount would rise to $10.0 
million if the Livestock initiative were classified under category 3, where logic would seem 
to suggest it belongs, rather than category 1 which would correspondingly shrink to $5.0 
million. Furthermore, categories 2 and 3 account for about half of the $6.00 million total 
over allocation. Another way of characterizing this shift is that the combined actual allocation 
for categories 2 and 3 ($6.75 million) is two-thirds higher than their combined initial MTP 
targets ($4.0 million).
11. With respect to the data on collaborators, one of TAC’s objectives in encouraging 
Systemwide activities was to broaden partnerships with non-CGIAR organizations in order 
to stimulate multi-disciplinary approaches and mobilize additional research resources to 
address the CG’s new research agenda, particularly in the areas of sustainability and natural 
resources management. In quantitative terms, it is difficult to verify what the numbers 
represent. Most Systemwide proposals do not specify in detail the functional roles of the 
various partners, even when the organizations are identified by name. Thus, terms like 
"partner" and "collaborator" may connote a broad range of cooperative activities from 
participation in a workshop to formal agreement to collaborate in joint research. For 
example, CIP’s plan to implement the Andean component of the Mountain Agriculture 
initiative through CONDESAN implies, in the absence of any detail, extraordinarily high 
numbers of collaborators in all of the categories shown. Given the stage of the initiative, it 
is doubtful whether the actual numbers of collaborators would correspond to these figures. 
In contrast, the SWNM initiative shows equally high numbers of collaborators, but these have
Annex V - Page 4
been identified in the context of theme-specific research consortia and can be taken more 
seriously. To some extent, then, the numbers may not be strictly comparable.
12. With this caveat, a few patterns in the data are observable. The aggregate number 
of IARC collaborations (88) on the one hand, and NARS/RARS (129) on the other, is 
roughly balanced and even slightly leveraged with a ratio of 1:1.5. The aggregate number 
of ARI collaborations at 56 equates to an IARC:ARI ratio of 1.5:1. But the number of 
NGOs is negligible if one considers that all but 2 of the total 25 are identified with 
CONDESAN. The only real anomaly in the ratios for individual initiatives would appear to 
be Water Husbandry which, at this stage, seems to involve no other identified IARCs. To 
reach any firm conclusions about the significance of any of these numbers, one would need 
to compare the baseline ratios of centres-to-partners for centre-based research to determine 
whether the Systemwide approach has added significantly to the CGIAR’s external "reach."
13. In terms of multi-disciplinarity, the quantitative data are not informative. This would 
require a breakdown by areas of institutional disciplinary specialization to determine whether 
centres have been able to involve non-traditional partners in such areas as natural resource 
management, etc. Existing in-house information is not adequate to make such an analysis 
and would require canvassing the centres.
Table 2.
14. This table attempts to synthesize a great deal of qualitative information into the 6 
narrative columns explicating objectives, governance, priority setting, resource allocation & 
accountability, monitoring & evaluation, and outputs & impacts. In some cases, Systemwide 
proposals lent themselves well to this exercise; in others, it was necessary to reconstruct the 
information from the proposals since they were not structured into these categories. Thus, 
in some instances, the information in the table is more explicit than that contained in the 
proposal. We have also attempted to classify which CGIAR activities are covered by the 
initiative. All initiatives cover more than one category; however, not all categories are 
covered with equal weight. It might be useful to have the centres weight the categories to 
see where the centre of gravity of each initiative really lies, recognizing that there may be 
spillovers to other activities. Inasmuch as we do not yet have information on what has been 
accomplished beyond the design stage of SWI/Ps, the outputs & impacts column specifies 
planned rather than actual products/services to be produced and their anticipated effects.
Comments on the table.
• Objectives
15. In evaluating SWIs for purposes of resource allocation, TAC initially established 
criteria emphasizing the global relevance of their ecoregionally-based results/methodologies, 
their degree of multi-disciplinarity, and their extent of multi-institutional collaboration. It 
later added to these criteria complementarities, avoiding duplication, encouraging 
specialization, economies of scale, and internalizing externalities. A perusal of the objectives 
column permits only a few inferences to be drawn:
Annex V - Page 5
• As might be expected, the globalization of results and/or methodologies is 
mentioned as an explicit objective primarily in the global and cross- 
ecoregional categories (initiatives 1.1, 1.4, 1.5, 2.4.1, 3.1., 3.2, and 4.2).
The ecoregionally-focused initiatives tend to emphasize extrapolation of results 
at the regional level.
• Multi-disciplinarity is mentioned or evident in the objectifies of virtually all 
initiatives.
• The collaboration criterion has already been discussed in the context of 
Table 1. Here it may be noted that strengthening collaborative mechanisms 
or building the capacity of partners is mentioned as an explicit objective in 
virtually all initiatives.
Beyond these inferences, the information synthesized in the table cannot shed light on the 
more operational questions of research efficiency implicit in TAC’s later criteria. Answers 
to such questions ought logically to be posed in the programme phase when observable 
results begin to materialize.
• Governance.
16. TAC’s guidelines for SWIs have little to say on the subject of governance beyond 
emphasizing the need for "thorough consultation" among partners to ensure durable long-term 
collaboration. The Committee underscored the importance of consultation by providing seed 
money to help cover the transaction costs involved.
17. At least two of the initiatives, SGRI and IPM, may be considered as having "in- 
house" governance in the sense that they are overseen in each case by an Inter-Centre 
Working Group. Non-CGIAR organizations are not involved in these mechanisms as they 
function primarily to coordinate inter-centre activities.
18. The Livestock initiative also seems to approximate above model. The IC-LPG’s 
membership is limited to the lead centres for each consortium, even though non-CGIAR 
collaborators are involved in the programme of work. Collaborative links are thus at the 
working rather than the governance level where decisions are taken on priorities and resource 
allocation. A similar arrangement seems to be envisaged in the Property Rights initiative 
where the Steering Committee would be composed only of participating CGIAR centres and 
"outside experts," and the Mountain Agriculture initiative which is formally led by IARCs 
with invited participation by other participating institutions.
19. Most other initiatives — particularly the large multi-consortia types — adopt a broader 
approach, involving in their governing mechanisms a mix of IARCs, NARS/RARS, and 
donor representatives. In the cases of the Desert Margins and Inland Valleys initiatives, 
chairmanship of the governing body rotates among NARS representatives. The African 
Highlands initiative is overseen by the governing body of ASARECA.
20. It should be emphasized, however, that even when membership in governing 
mechanisms is broadly shared, steering committees typically meet only annually and day-to-
Annex V - Page 6
day decision-making and management of the initiative are usually lodged in a coordinating 
unit backstopped by the convening centre.
21. It is worth observing that in several cases, fairly elaborate multi-tiered mechanisms 
involving committees from the global to the national, regional, and local levels have been set 
up (e.g., Desert Margins, Africa Humid-Sub-Humid, Rice/Wheat, African Highlands, Slash 
and Bum). This approach has evidently been necessitated by the geographic spread of these 
initiatives and by a participatory approach to research planning and implementation. These 
initiatives would be prime candidates for assessing the costs and benefits of transacting the 
business of ecoregional research in a highly consultative mode.
22. There is some variation in the functions of governing mechanisms. Some would 
appear to have primarily an advisory and coordinating role (e.g., SGRI, IPM, Agricultural 
Indicators, Asia Humid Sub-Humid) while others in varying degrees define the programme 
of work, set priorities, allocate resources, and monitor activities and results (e.g., Livestock, 
Property Rights, Desert Margins, Africa Humid Sub-Humid, Inland Valleys, Rice/Wheat, 
Tropical America, Water Husbandry, Mountain Agriculture, African Highlands, Slash and 
Bum, SWNM). Within these two general approaches to governance there appears to be 
variation in the degree to which decision-making with respect to research planning, 
budgeting, and implementation is decentralized.
Priority Setting
23. There is also variation in approaches to priority setting. Formalized approaches 
appear to be the exception rather than the mie. Thus, SLI’s use of external peer review of 
proposals coupled with a scoring procedure based on predetermined criteria to select and 
allocate resources to specific projects does not appear to be replicated in any of the other 
initiatives, although scoring is also used in the Rice/Wheat initiative. In most cases, 
priorities are set through consensus achieved in planning/programme design workshops 
involving the convening centre and partners, either before or after an initiative proposal has 
been approved by TAC. In a few cases, the criteria for priority setting are specified (e.g., 
IPM, Mountain Agriculture); in quite a few others we have no information on the specifics 
of the priority setting process. In at least one case (Asia Humid Sub-Humid), priority setting 
for ecoregional research is itself an objective of the initiative and a systematic approach is 
to be employed for prioritizing issues for further R&D. There may be scope for 
extrapolating from this kind of exercise to other present or future initiatives. In general, this 
is an area where it might be useful to canvass the centres for more information.
Resource Allocation & Accountability
24. TAC’s guidelines for SWI’s stipulate that convening and participating centres’ annual 
programme and budget submissions include progress reports on implementation of initiatives. 
As is evident from the table, this is an area in which information is generally deficient. The 
resource allocation process is usually not described in detail beyond indicating that the 
steering committee has responsibility for this function. The convening centre normally 
assumes responsibility for financial reporting. However, programme and budget reports 
usually do little more than indicate the amount allocated or expended against a particular 
activity (e.g., "facilitation"). Particularly lacking are narrative descriptions of the receipt and
Annex V - Page 7
disposition of non-CGIAR resources. This, again, is an area where more information is 
required.
Monitoring & Evaluation
25. TAC has indicated that it has the responsibility for monitoring and reviewing the 
outputs of Systemwide and ecoregional initiatives. Yet, it is evident from the table that we 
have virtually no information on monitoring and evaluation procedures being used for 
initiatives beyond the identification, normally, of the steering committee or the convening 
centre as the entity responsible for this function. That it is usually restricted to an annual 
meeting suggests that evaluation is not envisaged as a routine process integrally related to 
research planning and implementation, but rather as a means of reporting results to 
stakeholders. Much more information is needed on the methodologies, if any, that are being 
employed.
Projected Outputs & Impacts
26. From the standpoint of accountability TAC has a reasonably good record in this, the 
final column of the table, of what the Systemwide initiatives promise to deliver. As indicated 
earlier, in some cases it has been necessary to state the outputs more explicitly than the 
proposals have. It would be good to have the centres verify them. As a general observation, 
it is evident that centres tend to carefully limit the deliverables to outputs and do not go on 
to define causal linkages to the impacts which are implicit in their statements of purpose in 
column 1. It may well be that TAC’s review of Systemwide activities would want to define 
accountability strictly in terms of outputs. If there is an inclination to push the review 
further in the direction of impact assessment, it might be useful to link it to the work of the 
IAEG and the CG Secretariat’s new project reporting system, both of which will, in any 
case, require centres to develop measurable indicators of project success, and document 
wherever possible the impacts of outputs on intended beneficiaries.
Recommended Next Steps
27. This synthesis has identified a number of ambiguities or gaps in our data on SWIs. 
The tables as currently drafted should be submitted to centres to verify/correct the 
information recorded. To complete the picture and as an aid to TAC’s planned review of 
Systemwide activities in 1998-2000, We would suggest simultaneously canvassing the centres 
for the following information:
• Estimates of "initiative" and "programme" costs met by the TAC allocation.
• Non-CGIAR funds and other IARC costs for individual years, MTP period, and/or 
duration of the initiative.
• Numbers of collaborator s/partners for initiatives based on a functional definition of these 
entities.
• Estimated baseline ratios of centres-to-partners of various kinds for centre-based research.
Annex V - Page 8
• Disciplinary specializations of partner institutions (quantified).
• Weighting of the CGIAR activity codes covered by each initiative.
• Detailed information on centres’ priority setting resource allocation, and monitoring & 
evaluation processes.
Annex V - Page 9
SYSTEMWIDE AND ECOREGIONAL/REGIONAL INITIATIVES/PROGRAMMES 1996
(Funding Allocations/Recommendations in US$ Millions)
TABLE 1
Systemwide Initiatives/Programmes Initial TAC TAC Init­ Pro­ Non- Other No.
Convening MTP Ree Ree iative gramme CGIAR I ARCs ’ No. NARS No. No.
Centre Alloc. 1995 1996 Cost* Costs Funds Costs IARCs RARS ARU NGOs
1. Global
1.1. Genetic Resources IPGRI 1.00 1.74 1.60
1.2. Livestock URI 4.00 0.50 2.80 0.50 *2.80 5 17 4
1.3. Water Management UMI 1.00 0.20 0.00 0.20
(Proposal Pending) (IFPRI) 0.05
1.4. Property Rights IFPRI NA 0.20 0.20 0.40
1.5. Agricultural Indicators ISNAR NA 0.00 0.67 2
IFPRI
Subtotal 6.00 2.69 5.27
2. Ecoregional/Regional
2.1. Africa
2.1.1. Desert Margins ICRISAT 0.50 0.05 0.50 0.55 6 10 5
2.1.2. Humid/Sub-Humid Tropics IITA 0.50
2.1.2a Moist Savanna IITA 0.35 0.35 0.35 4 5
2.1.2b Humid Forest
2.1.2c Inland Valleys WARDA 0.15 0.35 2 7 3
2.2. Asia
2.2.1. Indo-Gangetic Plains ICRISAT 0.40 0.40 0.20 0.30 3-5 4 4-6
Rice/Wheat
2.2.2. Humid/Sub-Humid Tropics IRRI 0.70 0.30 0.70 1.0 10 3
2.3. Latin America/Caribbean
Annex V - Page 10
Systemwide luitiatives/Programmes Initial TAC TAC Init­ Pro­ Non- Other No.
Convening MTP Ree Ree iative gramme CGIAR IARCs’ No. NARS No. No.
Centre Alloc. 1995 1996 Costs Costs Funds Costs IARCs RARS ARIs NGOs
2.3.1. Tropical America CIAT 0.75 0.15 0.90 0.15 0.90
2.4. WANA
2.4.1. Water Husbandry ICARDA 0.40 0.10 0.60 0.10 1 11 3
Subtotal 3.25 1.50 3.60
3. Global/C ross-Eco regional
3.1. Mountain Agric. CLP NA 0.20 0.15 0.35
3.1.1. L.A. Andean CIP 13 34 6 23
3.1.2. E. African Highlands ICRAF 0.25 0.25 10 10 2 1
3.1.3. Asian Highlands ICIMOD NA
(Proposal Pending)
3.2. Slash and Bum ICRAF 0.50 0.40 0.40 0.40 7 9 1 1
(CIFOR) NA 0.10 0.10 0.20
3.3. Coastal Environments ICLARM NA ■0.30 0.05
(Proposal Pending)
Subtotal 0.75 0.70 0.95
4. Others
4.1. Integrated Pest Mgt. ICWG NA 0.20 0.20 0.40 10
4.2. Soil, Water, Nutrient CIAT NA 0.90 14 22 24
Mgt. IBSRAM
Subtotal 0.20 1.10
TOTAL 10.0 5.09 10.92 b b c88 c129 c56 '25
* ILRI: Includes 0.2Ò tor design of 1997 projects. lCLARM:Tentative allocation only; no more than 0.05 to be expended in 1996 for proposal development. 
k Estimated partitioning of initiative and programme costs is incomplete and, therefore, not totalled.
' Institutions may collaborate in more than one initiative; totals, therefore, represent numbers of collaborations.
NA: Not Applicable.
Annex V - Page 11
SYSTEMWIDE AND ECOREGIONAL/REGIONAL INITIATIVES/PROGRAMMES 1996
TABLE 2
Sy» temwlde Inllbttl ve*/Frogntrnm«** Convening CGI Alt Activity Governance Priority Sotting Resource Allocation A Centre No. Accountability Monitoring it Evaluation Projected Output* St Impact*
1. Global
1.1. Genetic Resources IPG RI 1.0 The Systemwide Genetic Resources The independently manage 
     Resources for centre-managed Evaluation activities are SGRP outputs cut across 15 2.0 Initiative (SGRI) is an integral components of SGRP activities are components of SGRP are allocated considered outside of the 
Purpose: CGIAR centres and hence are too 3.0 component of SGRP. Both are planned by the individual
  by the centres individually within scope of SGRP. numerous to list here. See IPGRI, 4.0 convened by IPGRI.        participating centres. their TAC-approved budgets. Participating centres receive Programme Plans A Funding Improve coordination of the CGIAR's genetic resources research, 5.0
           Expenditures for genetic resources oversight from their conservation activities, strategies and technologies, documentation, Requirements for 1996, AppendixSGRP is governed by a steering SGRI*» IPGRI-managod activities activities (conservation, research, respective boards on the n.
information and training committee, the Inter-Centre are formulated on the basis of the 
    training, information A public management of those Working Group on Genetic recommendations of TAC’s Stripe awareness, physical plant components of the 
Objectives:  SGRI outputs are summarized in Resources (ICWG-GR), which Study of Genetic Resources in the operations) by all centres are 
     programme for which they     advises on policy, strategy, and CGIAR and those of the ICWG- reported     
the above document and include:
in tabular form in are responsible.
• Coordinate and oversee the development of the CGIAR’s Systemwide 
  management of genetic resources GR. The latter also identifies IPGRI*» annual Programme Plans Genetic Resources Programme (SGRP)             • Development of a Systemwide work in the CGIAR. IPGRI specific collaborative activities and and Funding Requirements
         Information is not available   Information Network for Genetic        provides its secretariat and the lead centres for those submission to TAC. on the monitoring and • Establish a systemwide information network on genetic resources and a       Resources (SINGER) to ensure  programme leader in the person of activities. evaluation of SGRI permanent secretariat at IPGRI to service the Inter-Centro Working Group on  compatibility among centre’s   its director-general. The Group Resource allocation for IPGRI-     activities.Generic Resources (KJWG-GR).   genetic resources data bases and meets an an annual basis. managod SGRI activities is international access thereto.
 proposed in the context of the • Undertake collaborative activities, including capacity buikiing ani SGRP programme components are 
training,              Centre’s annual programme and • Information and public in the areas of ex situ gcnebank/conscrvalion research, In situ independently managed by
conservation        budget request to TAC.        awareness products including research, information management, and policy research, and participating centres, with oversight Allocations for collaborativesocioeconomic analysis.            brochures, reports, displays at by their respective boards. An activities under SGRI are made by international meetings, a series of 
inventory of current activities is 
    the ICWG-GR. videos, and an annual report.• Undertake public awareness activities including production of an annual contained in IPGRI, Programme 
generic resources programme report Plans A Funding Requirements for (NB: A statement of priority
     • Conduct of an external review of 1996, Appendix H. setting and resource allocation
     technical, scientific, and financial modalities for SGRP and SGRI is
 aspects of centre genebanks; SGRI functions to ensure adequate 
     not available to the TAC identification of constraints to and coordination of SGRP and enable Secretariat.)
certain high priority  opportunities for increasing the collaborative quality of services offered.
activities to be initiated. It is also 
the primary interface between the 
    • Development of standards, CGIAR and collaborating partner
    methodologies, initiatives, and institutions on generic resources other instruments in support of 
globally. genetic resources work of CGIAR
• Collaborativo research outputs in 
the areas of ex situ genebank/ 
conservation research, In situ 
conservation research, information 
management, policy issues, and 
socioeconomic analysis.
£ •I 
Ì
8 UÌ a
<* f 4 S *
a '5 ■«
f 3
O
1
? Ufa a
1 4-8
4
WS
iI
2 4
4
I1f3 1I
lI l* §3 111llllllilllfa
?
I ijldi Ifij
£> i*il
I
£
a, H ill111
- ^ 3*C/5 s :
1M il•5? f: jlj 12 1!
2>
3*
3 3®
u 2 9
3fi3 I
4 ll
i j!I Uh
■U
a? ■s •
I g
1
51
1 if!
*
<N 1 I I
> li! II1Is
8 H 111I • • •
Annex V - Page 13
System wide Initiative*/Program me»* Convening: CGIAK Activity
Cmtf No, Governance Priority Setting
Resource Allocation St
Accountability Monitoring A Evaluation Projected Output* A impact*
1.3. Water Management UMI NA NA NA NA NA NA
(Proposal Pending) (IFPRD
Annex V - 14
Systemwide InltUUv**/Pr«*nsn»ne** Convening : CG1AR ActivityCentre No. Governance Priority Setting
Resource Allocai km St 
Accountability Monitoring St Rvaluadon Prelected Output* St Impact*
1.4. Property Rights IFPRI 1.2 (a) A Secretariat hosted by IFPRI The Steering Committee would The Steering Committee would The Steering Committee Planned outputs from the 
<b) would facilitate Initiative planning, advise on conceptual frameworks make available seed money and 
  would review Initiative implementation phase:Purpose(s): (c) research, and outputs; liaise with and methodologies, conduct supplementary grants to Initiative 
    progress and outputs at an 2.0 CGIAR centres and other
    literature reviews and, based on partners. annual meeting. • Tested conceptual models and Develop locally relevant policy recommendations and internationally 3.0 participating institutions; and the identification of research gaps methodological guidelines for 
geocralizable lessons on institutional issue*-specifically, those concerning 4.0 coordinate information exchange emerging from these activities, ling the role of local 
externalities, property rights, and community action-affecting agricultural 5.1 between all collaborating partners. identify priorities and action plans. collective action and property 
productivity and natural resource management around the world. 5.2 The Secretariat would be mamtairwl institutions in natural resource
5.4 in the subsequent implementation 
Objective^): phase. A Steering Committee 
comprised of representatives of
    • Cross-comparable field studies • Increase knowledge of the emergence and performance of voluntary, self- participating CGIAR centres and illustrating the conditions in which 
governing/adapting community organizations and different property outside experts and meeting annually different collective action and
institutions in natural resource igement, and the trade-offs associated would direct the initiative and advise 
         property institutions can bewith different type* of institutions in different resource/socioeconomic on methodologies. effective.
conditions.
         * Field studies that combino to • Identify policy instruments that facilitate the formation, improved
        illustrate key factors affecting the functioning, and spontaneous evolution of organizations and property
     emergence and evolution of institutions that assure optimal resource use. particular collective action and 
  property institutions.• Strengthen the capacity of national and CGIAR research centres, non­
governmental organizations, universities and local organization to do 
research on         • Locally relevant policy property rights and collective action issues and forge/strengthen
  recommendations to facilitato the linkages between them. development of efficient and 
effective collective action and 
property institutions for natural
• Enhanced ability of national and 
CGIAR centres to conduct 
collaborative, interdisciplinary, 
empirical research on property 
rights and collective action issues.
• Dissemination of lessons 
learned from the initiative to 
researchers, development 
practitioners, policymakers, and 
resource user organizations.
Annex V - Page 15
Systemwide Inltiatlves/Profm/nmes* Convening: CGIAR Activity Resource Allocation A 
C«lre No, Governance Priority Setting Accountability Monitoring A Evaluation Projected Output* A Impact*
1.5. Agricultural Indicators ISNAR 4.0 A 6-member advisory committee Mechanism and process not Mechanism and process not An external evaluation of • An on-going series of published 
EFPRI 5.1 representing IFPRI, ISNAR, NARS, specified. specified. the initiative would talas 
Purposcfs): country statistical briefs.5.2 regional agencies, TAC, World 
      place in 2000.5.4 Bank, and donors, wculd be created
     • A rolling series of regional Provide national, regional, and international decision-makers with basic 
  to provide oversight to ensure research investment indicators and related policy analyses to inform and  policy reviews summarizing major quality control. It would meet once 
thereby improve public policies regarding investments in national, regional,     agricultural RAD investment trends  a year with the initiative’sand international agricultural research.    and policy assessments of these management team and report to the trends.
directors of the convening centres.
Objectives): The management team, comprised 
      • On-line information products  of staff of the convening centres,• Accelerate the development and maintenances of a database of globally      including downloadable country would manage the initiative and
  statistical briefs and regional comparable agricultural research investment indicators with the CGIAR. present a formal, joint report to tie 
    research policy reports and primary             directors and the advisory committee• Synthesize, analyze, and report on the important policy implicalioc» of data files.annually
these data from national but, primarily, regional and global perspectives. • A training manual with 
      accompanying survey instruments.• Disseminate these data and policy analyses to relevant decision-makers 
and analysts.
• Enhanced CGIAR and NARS 
• Develop networks and working relationships    institutional capacity to maintain « with counterparts in
national, regional and global agencies to strengthen their capacity to compile, future global data base.
these agricultural RAD indicators and perform the necessary policy analyses 
so that policy formulation and decision-making are improved and costs of 
future data collection are reduced.
Annex V - 16
Sy»tomwide Initiative*/Programme** Convening CGIAR Activity
Centre No. Governance Priority Setting
Resource Al locall oo A 
Accountability Monitoring A Evaluation Prelected Output# A Impact#
2. Ecoregkmal/Regional
2.1. Africa
2.1.1. Desert Margin# ICR1SAT 1.2 (a) National convener* would allocate The RSC would endorse priorities The RSC would endorse budgets 
      The Coordinator would      • Availability of improved 1.2(b) research tasks and resources across based on the recommendations of
   based on the recommendations of review the research, ensure          conservation-effective production Purposc(a): 1.2(c) institution* and coordinate the RTCC. The priority setting the RTCC. The Coordinating
2.0 information sharing and that results are effectively technologies.
         process is not specified. Unit would be administered by synthesized, and report to • Arrest land degradation by promoting improved and innovative 3.0 multidisciplinary linkages/activities. 
        ICR1SAT. the RTCC and RSC.    • Improved methods of sustaining technologies that integrate effective nutrient management strategics with 4.0  A Regional Steering Committee long-term soil fertility.improved soil and water conservation techniques that are ecologically sound, 5.0 (RSC), serviced by a Coordinating 
economically viable and socially acceptable to fanners in the dryland areas of 
 Unit backstopped by ICR1SAT and a    • Improved soil and watersub-Saharan Africa. Regional Technical Coordinating
Committee (RTCC), would provide 
Objective^): policy guidance and overall initiative 
   * Improved understanding of            direction; and cadono research      crop/livcslock inter-actions and soil • Develop better understanding of the extent of land degradation through priorities, workplans A budgets and
              erosion and the economic and water/wind erosion and mining of soil nutrients in traditional crop/livestock monitor effectiveness. RTCC would
               policy measures needed for production systems in desert margins and the impact of and distinction           be responsible for program improved NRM.between causal factors, natural and human, with a view to combatting
degradation and achieving    planning/development of improved productivity and sustainable collaborative research, training, 
       Minimum data sets of climate, soil use/managemeot of soil resources. networking; and prepare workplans
      A water, land use and           A budgets for RSC decision. RSC    socioeconomic variables in support • Evaluate with the participation of farmers, NGOs, and NARS, membership would comprise NARS,
past/curreot indigenous and improved soil management program for I ARC  of dryland development s, donors, and the Initiative programmes.
arresting land degradation; design strategies to test improved option to 
enhance soil   Coordinator (1C), with chairmanship resilience in desert margins. rotating annually among NARS 
   * Dissemination of improved representatives. RTCC membership
               methodologies of climate • Develop integrated systems of nutrient management based on locally would comprise NARS and IARC
             monitoring, data capture/analysis, available agrominerals, wastes, fanner- acceptable agroforestry system, ani      representatives and the IC. RTCC      and crop monitoring among major crop rotations to improve biological nitrogen fixation. would meet twice a year under
  scientific institutes in SSA.  rotating NARS chairmanship.• Evaluate the role of livestock in ecological/economic linkages between • Better awareness among policy­
arid/semi-arid zones to control land degradation and loss of vegetation 
 makers/scientists of the need for biodiversity in the arid zone.
NRM to main-tain productivity and 
    sustainability of fragile lands.• Identify, evaluate, and assist in the design of policies that will «nharvY-. 
the adoption of improved soil management options for arresting land • Creation of a cadre of NARS 
degradation.
scientists A mechanism to conduct 
  production system research.• To enhance the institutional capacity of participating countries in Ural 
degradation research and extension of improved technologies with particular 
      • Policy guidel inea to pro-mote attention to multidisciplinary and participatory socioeconomic research. ecologically sound, economically 
viable production technologic*.
D
ag,?
I
j•I»>
s ot
I
«
Z
I III,, s ì liE
K
?
5 |S 3 -3 4
4
s ?l1
1 il
z li
il mi
■8 n! g
II 8l-B
M 8
I
ie 1
£
■e 9
I 1 j■a 8 | SsI
ÌJISÌ miiiJiiJi
Lfi E |
8 I
ls?i5. <
È u<
i
!
c H E E
■8 8 8
£ I 1 I «
s
H 2«:3 ri 1 a I
Annex V - 18
Convening: CG1AR Activity
Systemwide Inltb»tlv*s/l‘rogn»mm*s*
Centre No, Governance Priority Setting
Resource AI local ion St 
Accountability Monitoring St Evaluation Projected Outputs it Impacts
2.I.2.C. Inland Valleys WARDA 1.2 (a) Governance rests on three A system of bottom-up priority The Steering Committee selects Internal evaluation of the 
       • Creation of a regional critical 2.0 principles: equality, inter­ setting based on the research research proposals developed by consortium programme is a mass of competent researchers; 
Purpose^): 3.0 disciplinarity, and mutual benefit. priorities of individual countries, National Coordinating Units to be continuous process: the enhanced capacity of national 
5.1 Oversight is provided by an Annual weighted and aggregated to the funded by consortium research 
       Annual Workshop evaluates scientists; strengthened Imh Assist national and international research institutes to collaborate in the 5.2 Workshop (AW) comprised of regional level, is employed. The grants. The grants are cn-going research and 
          between farmers, extension development and testing of integrated soil, water, and crop management representatives of all countries and Annual Workshop assigns specific administered by the CCU.
  coordinating activities; the services (including NGOs), and technologies • adapted to the needs and resources of rice-based smallholder international partners involved. research responsibilities to 
      Steering Committee researchers.farmers - to improve the sustainable agricultural productivity of inland Management functions are handled participating institutes based an 
   critically reviews the valleys in Sub-Saharan Africa by a Steering Committee elected by analyses of their respective scientific quality of the 
     • A multi-level/year master plan the AW comprised of represent­ comparative advantage. research carried out. for inland valley research 
Objectives): atives of NARS and lead External reviews are 
     specifying priorities/strategics for institutions, and chaired by an planned after 3 years (1996) 
     complementary regional re-search • Develop a structure for cooperation between NARS, IARCs, and other elected NARS representative. It and at the end of the 5-year by NARS and IARCs.
institutions to define common research priorities and to conduct develops long-term research/training 
    programme (1998). The multidisciplinary research on inland valleys. plans based on AW priority setting, programme’s objectives, • Increased understanding of 
cvahiatcs/sclects research proposals mode of implementation, 
  agronomic, lie, and• Develop common methodologies for research, data storage and for consortium support, and and finances (transparency institutional factors that affect more 
interpretation, and the extrapolation and transfer of results. supervises implementation of and accountability) will be intensive and sustainable use of 
      activities. A Consortium examined. inland valleys.* Implement complementary tmilrìdìmriplinaiy aginecologi cal Coordinating Unit has research 
characterization with subsequent classification of inland valleys and land use. responsibilities and management, • A multi-scale agroecological 
finance, information and liaison 
• Develop zona lion for Wcst/Cenlral Africa, and lest improved low cost water management technologies. functions. A Consortium including a comprehensive inland 
Coordinator, an agroocologist based 
           valley classification system, based • Selectively transfer and test agronomic technologies at farmers’ field level at the convening centre, provides 
  on bio-physical and socioeconomic in inland valleys. scientific and administrative factors.
leadership. National Coordinating 
Uni ts/Coordina tors have • A description of idcotypes of 
information, proposal, fundraising, crops, cropping systems, and 
and meeting functions. National farming systems; integrated seni, 
Coordinators participate in the 
Annual water, and crop management Workshop. technologies for various 
agroecological zones; and 
improved biophysical and 
socioeconomic models.
• Idcotypes of integrated 
management strategies for inland 
valleys.
• A multi-level master plan for 
inland valley development, 
specifying prioritics/slrategies, to 
improve planning & resource 
allocation.
• Use of multi-scale agro­
ecological zona ilon for W/C Africa 
for improved development
planning.
• Increased efficiency in use of 
human/financial resources.
Annex V - Page 19
Sy$t*mwld« Inlttadvee/Programmee* Convening CGI AR Activity Governane* Priority Setting Resource Allocation & Centre No. Accountability Monitoring A Evaluation Projected Outputs St Impact*
2.2. Asia
2.1. Indo-Gangctic Plains 1CRISAT 1.0 National conveners would allocate Projects of differing substantive The convening centre would 
  The RSC would monitor the • Characterization of rice-wheat Rioe/Wbeal 1.2 (a) research tasks and resources across focus (e.g., productivity trends, 
 provide on a cost recovery basis effectiveness of the initiative 1.2(b) institutions, coordinate information soil fertility, water management, administrative, office  systems at different research sites.and in achieving its goals. The 
Purposefs): 1.2 (c) sharing, establish linkages between ecological inpacts, policy options) logistical support for the Facilitator would also • Development of appropriate and 
2.0 rice and wheat sectors, encourage 
          would be prioritized by country Facilitator and Facilitation Unit,    review initiative research improved crop establishment • Increase food production in the Indo-Gangetic region by increasing 4.0 multidisciplinary research. A
          based on comparative advantage establish and provide the financial and report to the RSC and methods for rice and wheat.productivity of rice-wheat cropping systems through the introduction of new 5.1
       Regional Steering Committee using a scoring model in which accounting of a multidonor RTCC.technologies suitable for the fanners of the region. 5.2 (RSC), serviced by a Facilitation countries rate their capabilities and account in a bank acceptable to the 
5.4 Unit backstopped by IRR1 and 
           resources for addressing research RSC  
• Improved sustainability of long­
for deposit of donor funds, 
         term soil fertility for rice-wheat • Guide policy and the strategic research framework to address emerging CIMMYT and administered by issues previously identified by a and authorize the Facilitator to systems.
sustainability and productivity issues and facilitate collaboration between 
      ICRISAT and a Regional Technical NARS/IARC regional workshop. draw on the account in accordance agricultural researchers and technicians in the region. Coordinating Committee (RTCC), with the directions of the RSC, 
  • Improved water management for would provide policy guidance and and provide the RSC and donors 
      intensive rice-wheat systems.Objective^): overall initiative direction; and an annual independently audited 
endorse research priorities, 
      report of the use of funds under • Improved understanding of the • Locate areas most seriously threatened by productivity and sustainability workplans A budgets and monitor 
   the mullidooor account. ecological consequences of issues; diagnose the biological and physical causes of these problem; effectiveness. RTCC would work prolonged and intensive cultivation 
develop, test and promote implementation of more attainable, high 
  with NARS to develop collaborative of rice-wheat.productivity cropping systems; research programmes and prepare 
   workplans A budgets for RSC • Identification of policy options • Establish a collaborative research manageaeot nrahanism to foster 
          decision. RSC membership would     to enhance sustainable resourceregional cooperation among rice-wheal specialists and experts in othsr social,
biological and physical disciplines to   comprise NARS, IARCs, donors, management in rice-wheat systems.address high priority, post-Grccn and the Initiative Facilitator, with 
chairmanship rotating amually 
   • Formation of an alliance of among NARS representatives.
 scientific and technical experts to • Train and strengthen national programme staff in new skills and RTCC membership would comprise 
            address issues of sustainable methodologies; develop linkages between national, regional, and intcmatkxml NARS and 1ARC representatives productivity in rice-wheat fuming 
programmes toward this end. and the Facilitator. RTCC would 
      systems in the Indo-Gangetic Plain.meet twice a year under rotating
NARS chairmanship.
Annex V - 20
Sy*l®mwide Inltbitlv*s/Programmes* Converting CGIAK Activity
Centre No. Governance Priority Setting
Resource Allocailoa St 
Accountability Monitoring St Evaluation projected Output* St Impact*
2.2.2. Humid/Sub-Humid Tropics IRRI 1.1 Governance of the initiative would Priority setting for ccoregional Information not specified in The Council would have Outputs:
1.2 (a) rest with a Council for Collaborative R&D is the second of the four proposal. general oversight of the 
Puipoac(s): 1.2(b) Rice Research in Asia. Its functions pn*x»cd initiative objectives (See 
   initiative. The Eoorcgional     * Knowlodgp/data bases «nd GIS 2.0 would include promoting   Purposcs/Objectives). It would be Systems Coordinator would maps presenting alternative • Develop an Ecoregional Action Plan using s system»-based operational 3.0 collaborative research to serve the 
   accomplished via a workshop be responsible to the scenarios of land uso under model for ecoregional research that addresses the link between agriculture 4.0 needs of the Asian region; guiding 
        involving major stakeholders in the Council for monitoring different constraints and their and other sectors and stresses the role of all the major stakeholders in ths 5.4 and strengthening established region and prospective IARC and progress of the EAP and potential for resolution by R&D;
warm humid/sub-bumid tropics and subtropics of Asia. consortia and networks; developing NARS collaborators. A systems ensuring the relevance of 
       litizatkn of stallholders to the concepts for an ccoregional analysis of the ecorcgioo- outputs. Regular monitoring 
      ccoregional concept; strengthened Objective^): approach to research in rice and employing simulation modelling of 
    workshops would be held to    collaborative mechanisms/new rice-based fanning systems; and relevant crops, geographic evaluate progress of the partnerships between LARCs/NARS • Conduct cx ante analysis of ecoregional issues and knowledge gaps and harmonizing the views of NARS and 
strengthen     information systems, and multiple research conducted by the collaborative mechanisms for ecoregional activities. providing feedback to lARCs  st key sites.and goal optimization-would be Working Groups to meet the 
the CGIAR on ccoregional 
• Specify R&D  conducted to identify research priorities identified in the • Prioritization of ecoregional priorities and develop an ecoregional action plan to respond 
  agricultural research needs. The issues that are ccoregional in       EAP. An IRRI cote project issues by the major regional to ecoregional issues. Council would meet annually and be •cope; these issues would then bo would provido technical and 
          stakeholders; a well-endorsed R&D comprised of NARS leaders having            prioritized for further R&D into logistical support for the Ecoregional Action Plan (EAP); • Implement action plan in a participatory mode to meet research priorities; responsibility for rice and all other 
  sustainable agricultural strategy analysis, prioritization, strengthened research sites for synthesize R&D results; create a supportive policy environment for agricultural research in their 
 and technology. knowledge synthesizing, and ecoregional work.ecoregional research. respective countries. monitoring aspects of the 
             initiative. It would also • Research-monitoring mechanism • Develop and implement monitoring mechanisms to evaluate improvements  Working Groups would be serve stakeholders, liaise       in place to address prioritized in the ecorcgioo. organized for each of the research
 with the Council, and ecoregional issues; knowledge themes identified by the EAP. organize issue-focused generated by research teams to 
workshops at intervals to 
  address priorities; knowledge to An Ecoregional Systems assess progress and endorse 
     support policy environments Coordinator would be recruited with
     synthesis of results. favouring NRM for sustainable terms of reference to be determined 
 agriculture.by initiative partners. The 
Coordinator would be responsible to 
the Council      • Mechanism to ensure utility of but be based at the
convening centre to facilitate linkage utility of ccoregional research 
    outputs; improved systems of with its systems analysis group. technology for resolving specific 
ecoregional issues.
Impacts:
• Improved total productivity of 
•elected ecosystems.
• More focused and rationalized 
research on the foundations of 
sustainable production systems.
• Strengthened cooperation 
between and within national and 
international partnerships for 
research on ecoregional issues.
Annex V - Page 21
Systemwide InltbUvee/Prognunme** Convening CGIAK Activity
Centre No» Governance Priority Setting
Resource Allocation A 
Accountability Monitoring A Evaluation Protected Output* * Impact*
2.3. Latin America/Caribbean
2.3.1. Tropical America CIAT 1.2 (a) The convening centre would be A common research agenda would The convening centre would be 
    The convening centre would Outputs:2.0 responsible for overall initiative be developed through a scries of responsible for financial reporting. be responsible for technical 
Purpose^): 3.0 execution and coordination of collective and bilateral The "indicative* initiative budget reporting of initiative 
4.0 consortia activities. An Ecoregicnal consultations between  • Fnhanood capacity for cross- the specifies consultation costs and 
           activities. The ESC would     country research prioritization, • Improve the management of natural resources devoted to agriculture in 5.1 Steering CommiIter. (ESQ
           convening centre and resources associated with projected      have the authority to review targeting and extrapolation through Tropical America (TA) in order to permanently reduce poverty and hunger, 5.2 comprised of national, regional, and
     representatives of existing outputs. (Allocation process not consortia activities to ensure development of comprehensive maintain resourco quality, and h agricultural productivity. 5.4 IARC representatives would oversee agroocosystem consortia specified.) their consistency with environmental and agricultural land 
initiative execution and throughout the region. Workplans initiative objectives and use data bases/modcls.
Objectives): management. It would meet and proposals for future activities 
      strategies.onoc/ycar and elect its own chair.
       would also be developed on the • Improved methods for research • Enhance the effectiveness of agricultural and natural resources Project activities would be executed basis of these consultations. 
management research in Tropical America by improving the capacity      on technology development and to by agroecosystem consortia overseen
      (Details of the priority setting     policy at agro-ecosystems level define and understand productivity problems in agriculture. by their respective steering criteria and process arc not 
    through better targeting and committees. Decentralized contained in the proposal.)
• Develop and adapt extrapolation.suitable solutions to those productivity problems and management of the initiative would
extrapolate results among agroecosystems through the development of be coupled with provision of • More effective methods for 
georeferenced information systems and analysis. centralized services by the 
   identifying and prioritizing convening centre including
   agricultural and natural resource information, facilitation, and data problems in priority watersheds.
services.
• Strengthened national capacity to 
manage/use georeferenced models 
and data for prioritizing and 
conducting resource 
and agricultural research.
Inpacts:
• More effective integration of 
IARC’. technology & NRM 
research with
problems/opportunities in Tropical 
America.
• Enhanced NARS capacity to 
identifying priorities in 
national/regional agricultural 
research planning/inplcmcntation.
• Availability to farmers of 
technology Sc policy options to 
increase productivity, incomes, and 
sustainability of resources.
Annex V - 22
Systemwide InltUdvee/Programrae** Convening CGIAK Activity
Centre No, Governance Priority Setting Resource Allocation A Accountability Monitoring A Evaluation Projected Output# A Impact#
2.4. WANA
2.4.1. Water Husbandry ICARDA 1.2 (a) The convening centre would play a Research priorities, defined as four 
 The initiative budget would be          The convening centre would      • Appraisals of indigenous water 1.2(b) major research role, manage and "strategic research themes" (water allocated in three parts: (a)     have overall responsibility Purposed): husbandry (WH) systems and 1.2 (c) coordinate the initiative, and provide in present land-use systems; water support to research activities 
   for project reporting. improved information/I real 2.0         liaison with participating resources and potential for water according to research themes NRCMs involving Promote sustainable and profitable agricultural production in dry raided 5.1
            institutions. NARS would designate capture; options for water   
expertise om user perceptions of 
specified in the proposal; (b) 
 participating institutions WH techniques, socio-economic area* of the WANA region through the efficient and sustainable utilization of 5.2 National Coordinators to lead 
         utilization; and dissemination, project-wide activities such as      would focus on specific     aspects of construction A available supplica of water from rainfall, groundwater, ani surface sources. 5.4 national teams participating in the development and inpact), were workshops and training; and (c) 
    themes/activities, synthesize requirements, project. Coherence across themes identified through a consultative facilitation by the convening 
    results, and report findings Objectives(s): feasibility of water husbandry and regions would be achieved workshop with regional NARS and centre. Research funds would be 
          to the Steering Committee modifications to production through an annual meeting of a international experts and a follow­ allocated on a pre-set percentage annually systems, and risk management.• Conduct strategic research in collaboration with national, regional and Steering Committee chaired by 
            up literature review to identify basis to the four strategic themes. international centres to increase at the regional and global levels the technical
         ICARDA'a DDG/Rcsearch and gaps in existing knowledge.     Allocations could be adjusted by          • Improved methodologies for knowledge base (methodologies) for improved and sustainable efficiencies in comprised erf the National Research on these themes would the Steering Committee annually. identification of WH sites of high the capture and utilization of water for agricultural purposes in dry Coordinator* and ICARDA'a Project be formulated into annual The project-wide activities budget 
Coordinator. The Committee would 
         workplans via National and would  
potential; sites for further 
be administered by the experimentation identified by such 
• Conduct applied, multidisciplinary, and participatory research to adapt
available methodologies to the    discuss and approve workplans for Regional Coordination Mootings Project Coordinator on the advice methodologies.realities of nalional/local circumstances - the following year and advise the 
            (NRCM) involving ICARDA and of the Steering Committee. The social, economic, and political - in the WANA region to facilitate their
         Project Coordinator on the its NARS partners. The plans facilitation budget would be used • Provisional recommendations Aprofitable and sustainable application by farmers and other resource users. utilization of the budget. would, in turn, be approved by the by the convening centre to support 
      training extension materials for Steering Committee. a project scientist. The centre
would also administer and account WH lechniques/systems, crop types, and cropping system
for initiative funds. management for water-use
efficiency.
• Improved information/local 
expertise on crop/shruh/lree 
suitability, agronomy, and 
production needs with respect to 
water management in dry areas.
• Ex-ante appraisals of likely 
adoption and socioeconomic impact 
of new WH techniques A systems. 
Ex-ante/ox-post appraisal* of 
biophysical impacts of water
husbandry innovations.
• Greater public awareness erf WH 
teefaniques/systems among land 
users, development A extension 
agencies, WANA scientific 
community, and globally.
* Strengthened inter-institutional 
and multidisciplinary national, 
regional, and international 
cooperation on WH research.
Annex V - Page 23
Syetomwlde Inltlallvea/Programmee* Convening CGIAK ActivityCentre No. Governance Priority Setting
Resource Allocation A 
Accountability Monitoring A Evaluation Projected Output* A Impact*
3. Gtobal/Crota-Ecoreglonal
3.1. Mountain Agriculture CIP
(1CR AF) 
(ICIMOD)
Annex V - 24
$y*ten)wide Ini! hi lives/Programme** Convening CGIAK ActivityCentre No. Governane* Priority Setting
Resource Allocation & 
Accountability Monitoring A Evaluation Projected Output* A Impact*
3.1.1. L.A. Andean* CIP 1.1 Global coordination for the Global Priorities in the form of broad Resource allocation among the 
       Centre boards of trustees of       • Geoieferenced data bases on 1.2 (a) Mountain Initiative (GMI) would be research themes were set three ecoregional activities would 
 the concerned centres would           natural resources and their uso to Purpose(s): 1.2(b) provided by a steering committee collaboralively via a "project be a shared function between the 
 provide the institutional        permit agroecological 1.2 (c) composed of the heads of the three design" meeting convened by CIP steering committee and the auditing for the monitoring characterization of the mountain 
• Provide a forum for promoting and supporting research in natural 2.0 ecoregional initiatives convened 
         and involving participating director-level staff of the of overall research activities regions involved.resource management and sustainable agricultural production in the mountain 3.0 respectively by CIP (L.A. Andean), 
   initiative institutions. Themes were participating centres. and GMl’s global regions of the world. 4.0 ICRAF (E. African Highlands), and not formally ranked or coordination. GMI would • Diagnostic studies of land 
5.1 1C1MOD (Asian Highlands). Its quantitatively evaluated, but The four research themes would 
 also be audited and degradation, its processes snd • Develop independent strategic natural resource management research 5.2 functions would include planning 
     criteria such as expected impact, be assigned the following reviewed through normal determinants, contributing a programmes in three mountain regions (L.A. Andean, E. African Highlands, 5.4 and monitoring of global 
           feasibility, comparative advantage budgetary allocations across CGIAR processes and mountain perspective to knowledge Asian Highlands) based on common concerns and issues cutting across those coordination, harmonization of of the implementing institution, regions for 1997: land use, 28%; reviews.
regions. research activities across the three of the environmental consequences value-added from a global biological diversity, 26%; of population growth for 
regions, timely information 
   approach, specificity to mountain characterizing mowlain agricultural development.• Provide a framework for increased exchanges of experience, germplasm, 
     exchange, fundraising, and a focal agriculture, and multi-disciplinarity ecosystems, 20%; policy research, and technologies among the three mountain regions. point for the mountain research figured prominently in the 13%; information 
 • Improved knowledge of agenda within and outside of the 
            exchangc/coordination, 13%.    integrated nutrient management • Advocate increased support for research to reduce the degradation of the CGIAR System. When deemed Themes were grouped into four 
     yielding cost-effective strategies tonatural resource base and promote sustainable agriculture in mountain necessary, project leaders of areas (see "Objectives") and maintain long-term soil
regions. participating national institutions assigned budgetary allocations. 
       productivity in the three mountain (NARS, universities, NGOs) would Specific projects were then derived     regions. Training of human Objective^): be invited to share their views and from these broad subject matter 
  resources in computer simulation contribute to the implementation of 
         areas. for a systems approach to • Characterize agroecosystems to permit comparison of lard use patterns, GMI. integrated nutrient/soil
extrapolation of technological options, and exchange of experiences with the Within each mountain area, the 
other tropical mountain ecoiegioos. convening centre would develop a 
 mechanism for collaboration with • Methods for assessing tbs • Empirically document in benchmark sites the extent, processes, and national institutions; functions 
           impacts of agricultural determinants of land degradation and develop a critical knowledge base for would include participatory 
       intensification on biodiversity in integrated nutrient management to accommodate agricultural intemification programme planning to Identify mounts in regions and for 
without damaging the resource base. research strategies and measuring 
 measuring levels of biodiversity programme success.
 over time.• Assess the impact of agricultural intensification on biodiversity, the 
potential for inter-mountain transfer of promising agricultural specks, and The Consortium for the • Identification of potentials for 
the risk to biodiversity of adverse consequences of such transfers. Sustainable Development of the transfer of crops between mountain 
         Andean Ecoregion (CONDESAN), regions; improved biosafety/risk• Identify policies and public sector investments that stimulate productivity convened by CIP and comprising 
 assessment of field testing ofand enhance resource conservation in mountain agriculture. some 35 institutions, would be the 
    transgenic cultivar# in mountaincollaborative mechanism for the regions.
Andean component of the 
initiative. • Improved understanding of the
economic, policy, and investment
constraints affecting development 
of the dairy sector and terraced 
agriculture in mountain regions.
• Creation of an inter-regional 
electronic communications system 
to provide access to integrated 
scientific knowledge on 
environmental/agricultural issues in 
mountain regions.
a
Annex V - Page 25
Systemwide Inltkatlvee/Programmee* Convening: CGIAR Activity
Outre No. Governance Priority Setting
Resource Allocation St 
Accountability Monitoring St Evaluation Pitted Outputs St Impacts
3.1.2. E. African Highlands ICRAF 1.2(a) AHI is planned to operate at three AHI'a research themes ate decided Until ASARECA’s governing role The Directors’ Committee 
(c) levels: national activities, technical upon through a lengthy is fully operational, AHI resources 
   approves workplans and Purpose(s): 2.0 coordination, and overall consultative process involving both are managed by ICRAF. CGIAR monitors annual progress 
3.0 governance.
            senior scientists and research funds allocated to ICRAF as the toward achieving them.• Improve and enhance land productivity in the East African highlarvU by 5.1
    managers within NARS and convening centre are expended for working with farmers to evolve policies and technologies that increase 5.2 Oversight of AHI would be lodged 
       IARCs. Research methods and coordination costs; they are agricultural production while safeguarding the natural resource base. 5.4 with the governing body of approaches are determined by included in the Centre’s core 
ASARECA (Association for TAPs using participatory rural budget and accounted for in its 
Objective^): Strengthening Agricultural Research appraisals involving dialogue annual programme and budget 
in Eastern and Central Africa) and a 
            among fanners, researchers and report. ICRAF also receives and • Develop a regional programme of research on the management of natural Directors' Committee comprised of 
  other change agents. Emphasis is coordinates transfer of resources resources, particularly soil, to promote the sustainability of agricultural and the directors of participating on identifying and prioritizing to initiative partners for specific 
livestock production through improved technologies and unierstaixiing of the institutions and donor 
  problems and documenting activities; these funds are not natural and socioeconomic environments. representatives. Most management farmers’ perceptions of incorporated into the Centre’s 
decisions would be entrusted to an 
 environmental, institutional, policy budgetary presentation.* Strengthen the capacity of NARS to deal with problems related to natural AHI Task Force and Coordinating and other constraints and deriving 
Unit which would report regularly priorities for research and/or 
to the Director’s Committee.
           interventions accordingly.• Increase cooperation between NARS in the region and among NARS,
IARCs and other regional research and extension programmes dealing with Technical coordination at the 
natural resources research. regional level would bo carried cut 
by a lead institution with guidance 
of a Technical Advisory Panel 
(TAP) drawn, where possible, from 
existing advisory
commi Uces/commodity networks in 
the region.
National teams based at selected 
zonal stations and working on 
priority themes within national 
programmes would operate in 
collaboration with fanner groups, 
development agencies, NGOs, under 
the guidance of team leaders and 
technical advisory panels identified 
by national authorities.
Annex V - 26
Systemwide Initiative*/Programmes* Convening CG1AK Activity
Cmtre No. Governance Priority Setting
Resource Ai location St 
Accountability Monitoring St Evaluation Projected Output* St Impact*
3.1.3. Asian Highlands ICIMOD The convening centre has 
 organized a regional conference on Proposal pending. sustainable development of the
mountain areas of Asia to set
priorities for research and 
development and to establish a 
regional coordination structure.
Annex V - Page 27
Systemwide I n I dadves/Program raw* Convening CGIAK ActivityCantre No. Governane* Priority Setting
Resource Allocation St 
Accountability Monitoring St Evaluation Projected Output* St Impact*
3.2. Slash and Bum 1CRAF 1.2 (a) A Global Steering Group cotnprisod See Governance and Resource GSG members arrive at a 
     Information not available. • Identification of gaps in (CIFOR) (c) of representatives of the ASB Allocation columns. consensus of research priorities biophysical and/or sociooconomio- 
Purpoee(s): 2.0 consortium, chaired by the director- and funding needs. The convening policy knowledge and 
3.0 general of ICRAF, and serviced by 
          centre coordinates proposal/budgct corresponding future research • Reduce global warming, conserve biodiversity, and alleviate poverty by 4.0 s global coordinator, sets general 
   preparation and funding by the agendas;promoting development of ecologically sound, economically viable, and 5.1 policy guidelines. Regional Steering core donor (Global Environmental 
culturally acceptable alternatives to slaah-and-bum agriculture. 5.2 Groups for Latin America, Asia and Facility). Identification of co­ • Standardized 
Weal Africa led by IARCs ensure financing sources is a 
    procedures/metbods for Objective(s): regional coordination, setting of responsibility shared by monitoring/asscssing key 
priorities and sharing of institutional 
          participating institutions. Block environmental and socioeconomic • Identify, evaluate and, where necessary, modify and develop land-use responsibilities among the region’s 
       grants are received by ICRAF indicators; s geo-referenced systems and technologies that lead to sustainable alternatives to slash-and- benchmark sites. National Steering which is accountable to GEF for 
bum agriculture and the reclamation   database with rcgicnal/global of degraded lands. Groups, chaired by the NARS programmatic and financial 
    applications;director-general for each study
• Identify, evaluate, and design policies and methods of reporting. The Centre coordinates implementation to country, ensure government support 
            resource allocation and reporting     • Methods for policy research on protect the environment by reducing the area deforested by the practice of and the participation of other requirements for participating key soci ÙC and nalifutinnal
slasb-and-bum and promote the establishment of sustainable syslezr». institutions.
institutions, local NGOs, and issues;
• Enhance human resource capacity for informed policy-making and the universities. Local Steering Groups, 
   • Dissemination to policymakers, dissemination and application of research results. chaired by the NARS-ASB 
 farmers and scientists of representative, comprise farmer- 
Area Focus: producer organizations, NGOs, information an policies, investments, and technologies to 
community leaders, stale 
  reduce deforestation caused by • Latin America governments and others. LSGs 
     slaah-and-bum agriculture and • Asia implement the identified and
  reverse concomitant land • West Africa prioritized project goals at the local 
     degradation;level and are responsible for
maintaining quality research and 
 • Strengthened human resourcesexecution of training and 
  and information systems.dissemination activities. Thematic 
working groups further coordinate 
specific activities of participating 
institutions.
Annex V - 28
Syttemwld® 1 Dittati ve* /Program me** Convening CGIAK Activity
Outre : No, Governance Priority Sotting
Reeource Allocation & 
Accountability Monitoring & Evaluation Projected Output* A Impact*
3.3. Coastal Environments 1CLARM
(Proposal Pending)
Annex V - Page 29
Systemwide Inltfa»tlve*/Progn»imw«* Convening CGIAK Activity
C«tre No, Governane* Priority S«Ulng
K«source Allocation St 
Accountability Monitoring St Evaluation Projected Output* St Impact*
4. Other*
4.1. Integrated Poet Management ITT A 1.1 The Inter-Centre Working Group- SIPMI would solicit and prioritize The ICWG-DPM would consider 
    No information is provided     Proposed prqjects/imtiative* are 1.2(a) Integrated Poet Management centre*’ proposal* for inter-centre projects formulated by two or in the proposal.
     explicated in terms of rationale, Purposc(s): 1.2(b) (ICWG-IPM) would constitute the IPM projects/initiatives based on more centre* on the basis of 
 objectives, activities, and expected 1.2 (c) Steering Committee of the      the following criteria: prioritizing criteria and endorse outputs. Outputs are generally • Articulate a coherent CGIAR policy on integrated pe*t management 2.0 Systemwide Programme cn 
       them to donors. characterized in terms of:(IPM) and serve as a focal point for IPM matters within and outside of tbs 3.0 Integrated Peet Management 
 • Economic/scicntific contribution CGIAR System. 5.1 (SPIPM). The SC would further to development of improved plant 
    Allocations for proposed initiatives • IPM technologies, strategies, 5.2 SP1PM objective* which are
             and animal health systems in and projects are ranked in the systems, and practices developed• Provide a forum for discussion of the System’s IPM policy, strategy, and 5.4 identical to those of the Systemwide LDCs SIPMI three-year budget estimate 
activities, particularly as they relate to project definition. Integrated Peet Management resourced accordingly.
Initiative • IPM-improvcd cultivar* (SIPMI). • Demand-driven and concerned 
 developed• Coordinate and improve the comrranical ion/exchange of information on 
 primarily with constraints limiting No additional information on IPM within/outsidc of the CGIAR System. ICWG-IPM membership would IPM implementation identified resource allocation/accountability • h in productivity of
consist of one representative of each through farmer-participatory is provided in the proposal.
Objcctive(s): CGIAR centre nominated by  specific cropsits methods.
Director-General. The Committee 
 • reductions in pesticide use• Strengthen inter-centre IPM communication/collaboration and with would elect a Chairperson and 
 • Improves understanding of external partners. Secretary from among its biodiversity and ecological
    * methods for measuring adoption membership; the Secretary would
    mechanisms as a basis for the and impact of IPM• Identify BPM joint project opportunità* and support implementation of initially also act as Coordinator of 
   devekpment of more effective IPM research and training. SIPMI and its activities. IPM programme*. • training of researchers,
 extension agents, fermerà in IPM• Promote public of CGIAR centres’ IPM activities. Since no CGIAR centre has a • Multidisciplinary, undertaken on
mandate corresponding to the focus 
     an inter-centre collaborative basis, • information, databases, and • Enhance the impact of centre*' IPM activities at the farm level by of SIPMI, its overall leadership 
 and focused on problems of global decision support systems producedencouraging farmer participation and effective collaboration with remains to be confirmed. A rotating or regional significance.
organizations primarily concerned with IPM implementation (NARS, IGOs, Chairperson with a permanent 
  • increased IPM-based collabora­NGOs). secretariat is envisaged. tion with and capacity of NARS
• Focus IPM on sustainability objective* (including preservation of For an initial period the secretariat 
biodiversity and environmental health) and on human well-being (health and for SIPMI would be provided by the 
e*juity). convening centre, ETTA.
Annex V - 30
System wide inlth*tlvr*/Prwgnunm«* Convening CG1AK Activity
Centre No. Governance Priority Setting Reeource Allocai Urn it Accountability Monitoring it Evaluation Prelected Output# St Impact#
4.2. Soil, Water, Nutrient Management CIAT 1.2(a) 
 The initiative would be comprised of Fcur priority themes were      Proposed resource allocations    Information on monitoring Outputs:*IBSRAM 1.2(b) four thematic consortia and identified through an
 recommended by the workshop Purpose(s): and evaluation is not1.2 (c) implemented through an IBSRAM/CLAT/DSE-spoosored were approximately equal for the provided in the proposal. * Economically viable, socially 
2.0
          ecoregional/mnlli-disciplinary workshop involving 20 national four initiative themes.* Promote application by land users of technologies and sustainable   acceptable and environmentally 4.0         approach involving collaboration programmes, 7 CGIAR centres, Information an accountability for management systems to increase long-term productivity, reduce poverty, and scamd technologies for SWNM.5.1
conserve and enhance the quality of land between IARCs and partners and 29 other agricultural research resources is not provided in the and water resources both on-site 
 5.2 within/outside of the CGIAR and development agencies: proposal.and off-site.  • Improved methodologies and 5.4 including NARS, AROs, and combating nutrient depiction,
         diagnostic tools for SWNM NGOs. Consortia would be jointly managing acid soils, managing soil
Objective^): research.convened by IARCs other erosion, and optimizing soil water 
  institutions. use. Criteria for priority setting • Develop effective, efficient and environmentally sound technologies • Impact assessment 
systems for land management and conservation and     are not specified in the proposalmake them available to     methodologies including indicators    Programme coordination would befarmers and other users. for sustainable and unsustainable provided by the convening centre (See also Governance.)
 assisted by a Global 
land use systems to monitor 
Steering 
* Develop new mechanisms which encourage the adoption of sustainable environmental, social and 
      Committee composed of consortia soil, water, and nutrient management (SWNM) technologies.  economic impacts.and ecoregional program leaders, 
          donor representatives, and special * Promote stronger partnership and capacity of all stakeholders in       • Easily accessible decision advisors from NARS and AROs; it
ecoregional programmes (NARS, NGOs, IARCs, and AROs) in order to       support systems (models, expert   would be implemented by a Globalplan and implement research and dissemination programs for sustainable land systems, GIS, global data bases) Coordinator located at the convening 
   for generating, testing, and centre. The coordination function 
    extrapolating SWNM option.  would provide the following• Design workable policy options and advice, including issues concerning 
  services with a minimal structure: equity (gender, resource access, tenure) to enhance adoption of the   • Critical mass of trained human       continuity of activities across technologies and systems generated by the initiative. resources capable erf' implementing different consortia and ocorcgioos 
   SWNM programmes and policies.through cross-theme Working
Groups, common activities on • Framework for full cooperation 
training and information, and 
  and partnership between provision of a focal point on SWNM stakeholder groins in the initiative 
issues within and outside of the coalition.
CGIAR.
   • Policy options that promote Integration of SWNM activities with 
 sustainable SWNM practices.ecoregional programmes is 
envisaged, including joint definition 
of priorities   • Mechanisms for training and through collaborative
surveys, shared data-bases and information exchange.
benchmark sites, joint working Benefits:
groups, and participation at 
ecoregional or SWNM meetings. • Provide linkage for focused 
Decisions on SWNM research 
     research on SWNM at benchmark programmes would as far as
    sites within ecoregional possible be reached through
 programmes and across ccorcgions.consensus between ecoregional 
groups and SWNM consortia via 
  • Address priority SWNM joint ecoregional-SWNM working 
    problems in selected ecoregiom, groups. These groups would
     the results of which would have a determine joint workplans based on
priorities assessed   significant potential for by the
    extrapolation and adoption in ecoregional programme and the similar ccorcgions.
consortia, prepare joint proposals 
for multi- and bilateral funding, and 
  • Increase the efficiency of ths ensure that implementation of the research system through effective 
activities is collaborative and based collaboration betwocn NARS,
on the comparative advantage of the 
different institutions.
b
Annex V - Page 31
* Full Titles:
1.1. Systemwide Genetic Resources Inilia live/Programme
1.2. Systemwide Livestock Initiative: Feed Resources - Production ani Utilization
1.3 Proposal pending.
1.4. Property Rights and Collective Action
13. Agricultural Research Indicators Initiative
2.1.1. Sustainable Natural Resource Management Options to Arrest Land Degradations in the Desert Margins of Sub-Saharan Africa
2.1.2.a/b. Ecoregional Programme for the Humid and Sub-Humid Tropics of Sub-Saharan Africa: Moist Savanna/Humid Forest
2.1.2.c Inland Valley Consortium An Ecoregional Programme for the Warm Humid and Sub-Humid Tropics of Sub-Saharan Africa
2.2.1. Sustainability of Rice-Wheat Based Cropping Systems in the IndoGangclic Plain
2.2.2. An Ecoregional Approach to Research and Development in the Humid/Sub-humid Tropics arai Sub-Tropics of Asia
23.1. An Ecoregional Approach to Enhancing Agricultural Research in Tropical America
2.4.1. On Farm Water Husbandry in West Asia and North Africa
3.1.1. Global Ecoregional Initiative an Sustainable Mountain Agricultural Dcvclopcmot: L.S. Andes
3.1.2 African Highlands Initiative
3.1.3. Proposal pending.
3.2. Alternatives to Slash and Bum Initiative
33 Proposal pending.
4.1 Systemwide Integrated Pest Management Initiative/Systcmwidc Programme on Integrated Pest Management
4.2 Soil, Water and Nutrient Management Initiative
Notes:
The purposes and objectives specified apply to the three regional components of this global initiative but have only been displayed under the Andean initiative for reason of space.
The outputs listed are generic to the initiative aa a whole. Each of the initiative's four research themes would generate outputs discrete to those themes; they are too nurarout to display in the table.
Annex V - Page
* Full Titles:
1.1. Systemwide Genetic Resources Initiative/Programn»
1.2. Systemwide Livestock Initiative: Feed Resources - Production and Utilization
1.3 Proposal pending.
1.4. Property Rights and Collective Action
1.5. Agricultural Research Indicators Initiative
2.1.1. Sustainable Natural Resource Management Options to Arrest Land Degradation in the Desert Margim of Sub-Saharan Africa 
2.1.2.a/b. Ecoregional Programme for the Humid and Sub-Humid Tropics of Sub-Saharan Africa: Moist Savanna/Humid Forest 
2.1.2.C Inland Valley Consortium: An Ecoregional Programme for the Warm Humid and Sub-Humid Tropics of Sub-Saharan Africa 
2.2.1. Sustainability of Rice-Wheat Based Cropping Systems in the Indo-Gangelic Plain
2.2.2. An Ecoregional Approach to Research and Development in the Humid/Sub-humid Tropics and Sub-Tropics of Asia 
2.3.1. An Ecoregional Approach to Enhancing Agricultural Research in Tropical America
2.4.1. On Farm Water Husbandry in West Asia and North Africa
3.1.1. Global Ecoregional Initiative on Sustainable Mountain Agricultural Development: L.S. Andes
3.1.2 African Highlands Initiative
3.13. Proposal pending.
3.2. Alternatives to Slash and Bum Initiative
33 Proposal pending.
4.1 Systemwide Integrated Pest Management Initialivc/Systcmwidc Programme on Integrated Pest Managems*
4.2 Soil, Water and Nutrient Management Initiative
Notes:
i The purposes and objectives specified apply to the three regional components of this global initiative but have only been displayed mlcr the Andean initiative for reasom of space.
2 The oUpUi listed arc generic to the initiative as a whole. Each of the initiative's four research themes would generate outputs discrete to those themes; they are too ramerei» to display in the table.