CHAPTER 12 - IMPLICATIONS OF TAC'S ANALYSIS FOR CGIAR PRIORITIES

12.1. Introduction
12.2. The Analytical Framework
12.3. The Status of TAC's Priority Analysis
12.4. Conclusions
12.5. Final Observations

12.1. Introduction

As demonstrated in the preceding analysis, the task of priority setting for the CGIAR is multidimensional, complex and not easily reduced to a single analytical approach. As one element in the analysis, TAC attempted to use a framework allowing the formal consideration of a range of issues reflecting the nature of the CGIAR's mission and goals. The advantages of the spreadsheet approach used are many: it allows both sequential and simultaneous analysis of modifiers; it requires TAC at every stage to recognize that increasing some activities means decreasing others; and it is fully transparent since if, for example, TAC is asked how sustainability was taken into account and what were its impacts, the results are clear from the analysis.

But there are also disadvantages to the approach. Transparency brings with it the temptation to choose modifiers and weights to yield particular outcomes. The database on which TAC had to rely had many limitations. The many caveats associated with the estimation of value of production of commodities already gave strong indications of the utmost care that has to be taken in interpreting the results of the analysis. The spreadsheet does not maximize any goal function. It simply reallocates relative weights in proportion to previous levels when a modifier, change of base or weight is introduced. It is therefore a mechanical means of accounting. Finally there is always the danger that in an analysis producing numbers - i.e. relative allocations to commodities - those numbers will be given greater credence than they deserve.

Thus TAC presents its current views on future priorities for the CGIAR with a strong request that the whole analysis, not just the spreadsheet analysis, be taken into account.

Priority setting in the CGIAR has been and should continue to be an interactive process. Major stakeholders in the process have had the opportunity for reasoned input. TAC has consulted with a significant number of leaders of national programmes in Latin America and the Caribbean, in sub-Saharan Africa, in Asia and West Asia-North Africa. In each of these regions, meetings were held under the auspices of regional institutions (IICA, SACCAR, CORAF, AARINENA and APAARI) to discuss TAC's draft proposals on CGIAR priorities and strategies with representatives of national research systems. Board, management and staff of CGIAR Institutes have major stakes in the outcome and have made most valuable contributions. Members of the CGIAR and other interested parties commented on an earlier draft at ICW'91 and provided further comments in writing. TAC has carefully considered these inputs prior to finalizing its views. This process of consultation was very valuable but in no way does it imply an endorsement by its participants. This final document is the responsibility of TAC alone.

We begin this chapter by reviewing how the Committee proceeded. We then discuss TAC's views on: (a) priorities by category of research activity; (b) the implications of the agroecological and regional agroecological analysis for CGIAR priorities; (c) the implications of the analysis for the distribution of CGIAR resources across regions; (d) the implications of the analysis for the distribution of resources across production sectors; and (e) the implications of the analysis for the distribution of resources across commodities within production sectors. This sequential approach allows the Committee to put before the Group the multidimensional results of its analysis.

12.2. The Analytical Framework

The approach taken by TAC can be depicted in two diagrams (Figures 12.1. and 12.2). Figure 12.1 lists, on the left-hand side, the five major activity categories TAC has used. These are:

1. conservation and management of natural resources, including germplasm conservation (biodiversity);

2. germplasm enhancement and breeding;

3. development and management of production systems for agriculture, forestry and fisheries;

4. socioeconomic, public policy and public management research;

5. institution building (including training, information, organization/management counselling and networks to strengthen national research capacities).

TAC's analysis began by adopting an agroecological zone approach using nine agroecological zones defined in the main by moisture and temperature regimes. These agroecological zones were then rationalized into 21 regional agroecological zones - four for sub-Saharan Africa (SSA), one for West Asia-North Africa (WANA), seven for Asia and nine for Latin America and the Caribbean (LAC). These appear across the top of Figure 12.1. The same numbers refer to the same agroecological zones. For example SSA 1, Asia 1 and LAC 1 are all in agroecological zone 1, which is the warm arid and semi-arid tropics. The regional agroecological zone became the basic building block for both quantitative and qualitative analysis. The vertical lines indicate that the regional agroecological zones were most useful for developing TAC's analysis under activities 1, 2 and 3, whereas regions were used for all five activities. In this analysis, TAC could look across both regions and agroecological zones by activity, and down a regional agroecological zone or region across activities.

The next stages of the analysis are depicted in Figure 12.2. Again beginning with activities, TAC considered production sectors and commodities in the light of the analysis by agroecological zone and region. Under activity 1, certain resource management issues transcend production sectors, while others are specific to a single production sector and germplasm conservation is specific to commodities. Activity 2, germplasm enhancement and breeding, is necessarily commodity- and production-sector specific. Activity 3, development and management of production systems, can focus at several levels: (1) commodity-based cropping systems; (2) multiple cropping systems; (3) crop-livestock systems; (4) agroforestry systems, including crops and trees, livestock and trees, or crops, livestock and trees; and (5) integrated production systems involving crops, livestock, trees and fish. The diagram implies that production sectors and commodities are less important divisions for activities 4 and 5, socioeconomic, public policy and public management research, and institution building, though clearly some activities under this category are differentiated by production sector and commodity.

FIGURE 12.1. CGIAR PRIORITIES I: ACTIVITIES/REGIONAL AEZ, REGION, GLOBAL

ACTIVITIES	REGIONS AND REGIONAL AGROECOLOGICAL ZONES (RAEZ)
	SSA				WANA	ASIA							LAC
	1	2	3	4	9	1	2	3	5	6	7	8	1	2	3	4	5	6	7	8	9
1. Conservation and Management of Natural Resources -including Germplasm Conservation
2. Germplasm Enhancement and Breeding
3. Devt. and Mgt. of Production Systems for Agriculture, Forestry & Fisheries
4. Socioeconomic, Public Policy and Public Management Research
5. Institution Building a. Training b. Information c. Org/Mgt. Counsel. d. Networks

FIGURE 12.2. CGIAR PRIORITIES II: ACTIVITIES/RESOURCE MANAGEMENT. PRODUCTION SECTOR AND COMMODITY

ACTIVITIES	RESOURCE MANAGEMENT, PRODUCTION SECTOR, COMMODITY, ETC.
1. Conservation and Management of Natural Resources Including Germplasm Conservation	RESOURCE MANAGEMENT
	PRODUCTION SECTORS
	AGRICULTURE								FORESTRY	FISHERIES
	CROPS					LIVESTOCK
	COMMODITIES					SPECIES
	Rice	Wheat	Sorghum	Cassava	etc.	Cattle	Sheep	etc.
2. Germplasm Enhancement and Breeding
3. Development and Management of Production Systems for Agriculture, Forestry and Fisheries	(1) COMMODITY-BASED CROPPING SYSTEMS
	(2) MULTIPLE CROPPING SYSTEMS
	(3) CROP-LIVESTOCK SYSTEMS
	(4) AGROPORESTRY SYSTEMS
	(5) INTEGRATED PRODUCTION SYSTEMS ¬ ----------------------LAND USE CONTINUUM-------------------®
4. Socioeconomic Public Policy and Public Management
5. Institution Building - training - information - org./mgmt. Counselling - networks

12.3. The Status of TAC's Priority Analysis

12.3.1. By Activity Category
12.3.2. By Agroecological Zone and Regional Agroecological Zone
12.3.3. By Region
12.3.4. By Production Sector
12.3.5. By Commodity Within the Agricultural Sector

12.3.1. By Activity Category

TAC began its priority analysis by considering, in Systemwide terms, what should be the relative distribution of CGIAR efforts among the five broad activity categories. The Committee acknowledged that there is no clear dividing line between these categories and that in many instances activities may overlap into several categories. Furthermore, a number of CGIAR objectives cross-cut all these categories, such as the strengthening of national research systems, and improvement of the sustainability of production systems.

TAC was also mindful of the fact that the categorization of activities differed from that of the 1986 priority analysis in two major ways. First, in 1986, TAC had differentiated the activities relating to germplasm and production systems by production sector. Second, in 1986, separate categories had been used for product utilization and human linkages. For the reasons already discussed in Chapter 3, TAC felt that, given the addition of agroforestry, forestry and possibly fisheries to the CGIAR mandate, integrated categories were preferable with differentiation by production sector being a subsequent component of analysis. The little utilization work that is done by the System could be considered as one of the activities under the production systems category. Human linkages are considered as part of socioeconomic, public policy and public management activities.

However, the overlapping and the cross-cutting nature of CGIAR activities, and this redefinition of activity categories make a precise determination of current CGIAR efforts by activity category difficult. TAC's best estimate, based on analysis of the CGIAR Secretariat, is that in 1991, approximately 13% of CGIAR resources were in natural resource conservation and management (including germplasm conservation), 21% in germplasm enhancement and breeding, 33% in the development and management of production systems, 9% in socioeconomic, public policy and public management research and 24% in institution building. These estimates take account of the recent expansion of the CGIAR and include activities with respect to agroforestry, irrigation management and banana and plantain.

Setting current levels aside, an assessment was made of the optimal future balance of CGIAR core resources between activity categories, across production sectors by the year 2010. Throughout this report, considerable thought has been given to the challenges faced by the CGIAR during the next two decades, and their implications for research emphasis. In Chapter 4, an assessment was made of the international research needs by region and agroecological zone. Furthermore, in discussing important factors that determine CGIAR priorities by region and agroecological zones, attention was given to the implications of the weights attached to particular modifiers.

In all regions and agroecological zones, there was a perceived need for an expanded effort in research on natural resource conservation and management. These additional efforts refer both to activities in ecosystem conservation and management (currently receiving 7% of core resources) and germplasm collection and conservation (currently at 6% of core resources). This need emerged even more strongly when modifiers such as yield gap, soil degradation risk, and deforestation were considered. TAC therefore recommends a substantial increase in CGIAR efforts in this category, from 13% to between 17 and 19%. However, the analysis suggests that, relative to the new level, more attention should be paid to resource management in sub-Saharan Africa, West Asia-North Africa and Latin America and the Caribbean and less in Asia. Nevertheless, the investment in research in natural resource management should be increased in all areas.

Table 12.1. Priorities by activity category by region (core resources only)

Activity Category	1991 Base	2010 Rec.	Region1/
			SSA	WANA	Asia	LAC
1. Conservation and Management of Natural Resources Including Germplasm Conservation (Biodiversity)	13	17-19	+	+	-	+
2. Germplasm Enhancement and Breeding	21	21-23	-	-	+	0
3. Devpt. and Mgmt. of Production Systems	33	28-30	+	0	-	-
4. Socioeconomic, Public Policy and Public Management Research	9	10-12	-	-	+	+
5. Institution Building (incl. Training, Information, Org./Mgt. Counselling and Networks)3/	24	19-21	+	+	-	-
TOTAL	100	1002/

^1/

+ = more than the new System level priority but possibly lower than current allocation
0 = equal to new System level priority
- = less than new System level priority but possibly higher or lower than current allocation

^2/the mid points of the ranges add to 100

^3/ networks of category 5 refer to capacity building networks only. Research networks are included in the other categories.

Germplasm enhancement and breeding has been the central thrust of the CGIAR since its inception. Increased emphasis is warranted on this activity in Asia, where substantial efforts are now required to lift the yield ceiling, particularly of rice. In West Asia-North Africa and sub-Saharan Africa, the emphasis on germplasm enhancement and breeding could be somewhat reduced where a greater effort on resource management is required. Overall, the conclusion is that research on germplasm enhancement and breeding should continue at the marginally higher Systemwide level of between 21 and 23% of total activities.

Overall, in the long term, there should be a reduced need for CGIAR research on the development and management of production systems, because of increasing national capacity to deal with this category, which consists mostly of applied research. It is important to note, however, that the results obtained from applied research activities in this category feed into the planning of, for example, strategic resource management research. Continued emphasis is warranted in West Asia-North Africa, and more emphasis relative to the reduced System level is needed in sub-Saharan Africa, particularly to exploit the scope for growth. In Asia and in Latin America and the Caribbean, the resources devoted to this activity could be reduced below the new System level. As national research systems assume greater responsibility for research on the development of production systems, the modes of operation should evolve towards greater use of networks and consortia. Overall, TAC tentatively concludes that at the System level, efforts in this category could be reduced from 33% to between 28 and 30% of total activities.

Throughout this report, reference has been made to the need for an expanded effort in socioeconomic, public policy and public management research. In all regions and in most agroecological zones, there was a need for greater emphasis on policy research with respect to land use and sustainability issues, poverty alleviation and equity (particularly gender equity), irrigation management, and issues related to self-reliance. It is therefore proposed to increase CGIAR efforts in socioeconomic, public policy and public management research, from 9% to between 10 and 12% of System activities. Even more emphasis relative to the new System level may be needed in Asia and in Latin America and the Caribbean, because of pressing problems of sustainability, deforestation and equity.

TAC considered whether, in the long term, the resources allocated to institution building could be reduced. It is to be recalled that this category of institution building refers only to specific activities and that CGIAR Centres also contribute to strengthening national research systems through their work in the four previously discussed categories of activities in natural resources, germplasm enhancement, production systems and policy.

The study by Pardey and Roseboom (1991) has revealed the considerable increase in trained human resources in national programmes. For many national systems, the major constraint is now a shortage of operational funds rather than of trained manpower. As already noted in Chapter 10, TAC's view is that the allocation of CGIAR resources to institution building has reached a ceiling and that the CGIAR should focus on its complementary role of strengthening national research systems through technology development and partnership.

Although additional efforts in training will be required in the new and expanded areas of CGIAR activity (forestry, fisheries, natural resources and irrigation management), the overall emphasis on training can be reduced in the medium and long term from its present level of 9% to about 7%. However, in the medium term, a continued strong effort in training will be required in sub-Saharan Africa. In the area of information services, TAC considered that the current level of activity (8%) could only be reduced slightly since these are essential activities for partnerships with national research systems. In the area of organization and management counselling (now 2%), efforts should clearly increase. The responsibility for networks intended to strengthen national research capacity could increasingly be handed over to national programmes and the level of support to this type of activity could, therefore, be reduced from its present level of about 5% to 4%.

In TAC's view, therefore, the overall proportion of CGIAR activities in the category of institution building should be reduced from 24% to between 19 and 21%, but with less in training, information and capacity building networks, and more in organization/management counselling.

These judgements are displayed in Table 12.1, which shows the proposed shifts among activity categories, with an indication of differential efforts by region relative to the new proposed level of activity. These judgements were then used as indicators as TAC proceeded to evaluate other dimensions of the matrix.

12.3.2. By Agroecological Zone and Regional Agroecological Zone

TAC evaluated carefully the analysis of research priorities by agroecological zone developed in Chapter 4, along with the results of the analysis in Chapter 9, in making judgements as to whether or not the shifts in emphasis implied by the analysis were appropriate. It should be recalled that the analysis had suggested that in agriculture, the relative emphasis should increase for tropical agroecological zones (AEZs 1-4) and the cool subtropics with winter rainfall (AEZ 9), while in forestry the relative emphasis should increase for the tropical agroecological zone in general (AEZs 1-4), with greater increases in sub-Saharan African agroecological zones than in those of other regions. TAC supported these shifts in emphasis because the other areas benefit to a much greater extent from ongoing research in developed countries.

The current allocation of CGIAR resources by agroecological zone or regional agroecological zone is not available. The proposed new allocations cannot therefore be compared with existing allocations. However, based on its knowledge of current efforts, TAC felt that the shifts in emphasis implied by the analysis were already well under way in the CGIAR System. There is some evidence to support this perception in the regional analysis which follows.

12.3.3. By Region

In the 1986 analysis of priorities, TAC began at the global level, made recommendations with respect to priorities among activities and among commodities, and then evaluated the regional implications. The current exercise began with regional agroecological zones and, therefore, allowed a more comprehensive analysis of the regional distribution of CGIAR resources. The details of the regional analysis are given in Chapter 9. Here, TAC summarizes the changing pattern of regional allocations that the use of a modified base and the application of modifiers suggest.

Table 12.2. Impacts of Baseline and Modifiers on Regional Distribution of Values Relative to Current Allocation: Agriculture

	Baseline Components			Baseline	Modified Baseline			Current (1991) Allocation of Core Resources
	VOP%	Number of Poor %	Usable land %
Weight	0.33	0.33	0.33		0.25	0.5	1.0
Region
SSA	9.1	16.2	29.4	18.2	25.8	34.0	47.4	43.0
WANA	9.3	5.4	7.5	7.4	7.8	8.1	8.4	13.0
ASIA	59.0	72.1	27.9	53.0	46.4	39.5	29.1	29.0
LAC	22.6	6.3	35.2	21.4	20.0	18.4	15.1	15.0
TOTAL	100	100	100	100	100	100	100	100

Table 12.2 summarizes the previous analysis and adds a column containing TAC's estimate of the 1991 regional allocation of CGIAR resources. As pointed out in Chapter 9, the effect of using a three-part base (incorporating land use and the number of poor people, as well as the more usual starting point, value of production) already substantially shifted baseline values in favour of sub-Saharan Africa and away from West Asia-North Africa, Asia and Latin America and the Caribbean (value of production only, 9.1; three-part baseline value, 18.2). Applying the modifiers and increasing their weights further shifted resources towards sub-Saharan Africa, and also towards West Asia-North Africa. For example, with the modifiers weighted at 0.5 each, the sub-Saharan African modified baseline value is almost four times the original value of production. One might conclude that modification had gone too far.

However, when one considers 1991 estimates of CGIAR allocations by region, one discovers that the allocation to sub-Saharan Africa in fact lies between the analytical results obtained with modifiers weighted at 0.5 and 1.0. Applying the modifiers with a weighting of 0.5 reduced Asia's percentage share in value of production from 59.0% to 39.5%. A weighting of 1.0 suggests that Asia's share should be similar to the current allocation. A weighting between 0.5 and 1.0 would give a result resembling the current CGIAR allocation, except for the West Asia-North Africa region. As noted in Table 12.2, current regional allocation of CGIAR core resources is as follows: 43% to sub-Saharan Africa, 13% to West Asia-North Africa, 29% to Asia and 15% to Latin America. It should be noted here, however, that the benefits of research are not necessarily only accrued by the region where the investment occurs. The benefits of research investments often spillover into other regions and agroecological zones particularly with respect to investments in strategic research.

TAC considered the implications of the analysis for the balance of allocations between regions very carefully. TAC is aware of the strongly divided opinions held both within and outside the CGIAR on the special needs of agriculture in sub-Saharan Africa. Some argue that the CGIAR has gone much too far in reallocating resources towards sub-Saharan Africa, given the massive needs in Asia. Others argue that even more resources should be transferred to sub-Saharan Africa because of that region's rapid population growth rates, pervasive poverty, severe sustainability problems and lack of progress to date in improving the productivity of crops and livestock important to the poor.

The analysis presented here supports neither of these extreme views. TAC feels that the 0.5 weighting should be the maximum one used if undue influence by particular modifiers in favour of any given regional agroecological zone is to be avoided. Thus, using the analysis only as a rough guide, the Committee is of the view that there is no substantial case for a further shift of CGIAR resources towards sub-Saharan Africa. As the forestry analysis also suggests, there are pressing problems in Latin America and the Caribbean and Asia, as well as in sub-Saharan Africa. In considering the results of the analysis, TAC expressed a preference for the use of 0.5 weighting as a guide for allocating priorities by region by the year 2010. A case could then be made that the regional reallocation, which has taken place over the past 10 years may already have gone too far towards both sub-Saharan Africa and West Asia-North Africa.

As noted, rapid population growth rates, coupled with declining per caput food production in sub-Saharan Africa, make a compelling case for that region. The fragility of its tropical agroecological zones, the generally limited national research capacities and the slow rate of progress in productivity improvement to date add to the apparent urgency. Many of Africa's development problems are also political in nature however, and cannot be solved through research alone. On the other hand, the magnitude of population numbers, the extent of the poverty problem, the narrowing yield gap and the limited scope for land expansion all argue strongly for more long-term strategic and applied research in Asia. In addition, TAC is concerned about the generally low rate of progress in obtaining impact from international research efforts in sub-Saharan Africa. Although significant breakthroughs have been achieved (such as biological control of cassava mealy bug, and the adoption of improved wheat and maize varieties), in general the impact obtained in the region has been below expectations.

TAC's position is that in the medium term a modest reduction in resources allocated to sub-Saharan Africa and West Asia-North Africa, and an equivalent increase in resources allocated to particularly Asia and also Latin America should be scheduled, moving the CGIAR toward a distribution of resources by the year 2010 not unlike the priority index suggested by the modified baseline (weight 0.5). TAC recommends that, for the purposes of guiding the resource allocation process, by 1998 the target distribution by region should be 39% to sub-Saharan Africa, 11% to West Asia-North Africa, 33% to Asia and 17% to Latin America.

TAC recognizes that the analytical basis of this recommendation refers mainly to the agricultural sector, but on balance considered this proposed distribution appropriate for the System as a whole. Within the forestry and fisheries production sectors, the regional emphasis could be shifted more towards the distribution suggested in Tables 9.18 and 9.19, which calls for a much greater emphasis on the Asia region, particularly in the fisheries sector.

12.3.4. By Production Sector

The analysis undertaken to date gives much less insight into the question of what ought to be the relative balance of CGIAR efforts among agriculture (crops and livestock), forestry and fisheries. This is so for two main reasons.

First, for valid analytical reasons, different baselines were chosen for agriculture, forestry and fisheries. The results of the modification analysis cannot therefore be added together. As a result, one can compare relative distributions within sectors but not across them. There is no obvious way of obtaining cross-comparability unless one reverts to annual gross value of production as a basis. But TAC had rejected value of production as a direct congruence indicator for a variety of reasons, one of which was the fact that it includes a large number of commodities not covered by the CGIAR mandate. Non-CGIAR commodities account for approximately one-third of the value of crop production and for 45 % of the value of livestock production in developing countries. In addition, about 60% of the value of production of forestry is from sawlogs and veneer, which are not priority items for CGIAR research, while approximately 55% of fisheries production is from deep sea or ocean fishing which, according to TAC, should be outside the scope of the CGIAR.

Second, CGIAR activities in forestry are only just beginning and those in fisheries have yet to be finally accepted. The current allocation of CGIAR resources across sectors cannot therefore be used as the starting point for analysis.

What can be said is that TAC's detailed analysis of the research needs in agriculture, presented in previous chapters of this report, does not provide grounds for an absolute decline in support for agricultural research. On the contrary, it portends increasingly urgent needs to meet the rapidly rising demand for food. TAC's judgement at this juncture is that, whatever the level of support earned by the yet to be proposed long-term research programmes in agroforestry, forestry and fisheries, these should not be funded at the expense of critical research needs in crops and livestock. TAC will review the issue of balance between sectors as further information becomes available on research needs in agroforestry, forestry and fisheries.

12.3.5. By Commodity Within the Agricultural Sector

12.3.5.1. The commodity portfolio
12.3.5.2. The congruence between the modified values of production and current allocations

The final dimension of the analysis relates to the relative priority to be assigned to specific commodities. TAC carefully considered the information on particular issues related to each individual commodity as presented in Chapter 5 on crops and Chapter 6 on livestock, and the outcome of the in-depth quantitative analysis on modified value of production of commodities discussed in Chapter 9. It is very important to recall that through the process of modification of the baseline and the development of priority indices by RAEZ, several indicators related to efficiency, equity, sustainability, strength of national research systems, and food import gap were incorporated explicitly already in the quantitative analysis related to each individual commodity.

In developing its recommendations on assigning priorities by commodity, TAC also took into account, qualitatively, appropriate additional considerations such as:

· projected growth of demand for that commodity between 1990 and 2010;

· importance of the commodity for the poor;

· alternative sources of research supply, particularly the role of the private sector, universities and advanced institutes;

· regional distribution of production;

· concentration of production in one country;

· impact achieved through previous and ongoing research on the commodity;

· possibilities for technical breakthroughs;

· comparative advantage of the CGIAR;

· expected productivity gains as discussed in Section 9.11;

· size of spillover effects as discussed in Section 9.13.

TAC began its analysis by comparing the modified values of production of the top 45 agricultural commodities of importance in developing countries with the current list of CGIAR commodities (Table 12.3). First, TAC addressed the portfolio issue. The question asked was: Are there strong candidates, with high modified values of production, that should be considered for inclusion in CGIAR activities? TAC then looked carefully at current CGIAR commodities which have relatively low modified values of production but which are important for limited subsets of regions and/or countries. This led to a discussion of whether any current activity should be discontinued. Next, TAC looked at the congruence between the modified values of production and current allocations to determine if there was a need to consider altering the relative distribution of resources among agricultural commodities. The following sections present the information TAC considered and the recommendations of the Committee.

12.3.5.1. The commodity portfolio

Of the 45 commodities listed in Table 12.3, 22 are already CGIAR commodities, and tomato is on the list of vegetables recommended by TAC for inclusion in 1988. Three major vegetables (tomato, onion and cabbage) were explicitly considered in the analysis discussed in Chapter 9. The modified value ranking (2.1%) for the three vegetables combined (Table 9.17) would place vegetables within the top 16 commodities, thus further supporting TAC's earlier recommendations that they be included in the CGIAR portfolio (TAC/CGIAR, 1990).

Table 12.3. Ranking of commodities based on value of production, weighted baseline and modified baseline

VOP (%)		Weighted Baseline (%)		Modified Baseline (%) (W = 0.5)
1. Rice*	17.7	1. Rice*	15.6	1. Rice*	13.2
2. Milk*	8.9	2. Milk*	9.5	2. Milk*	9.7
3. Wheat*	6.4	3. Wheat*	5.7	3. Beef & Buffalo*	5.9
4. Beef & Buffalo*	5.0	4. Beef & Buffalo*	5.3	4. Cassava*	4.5
5. Pigmeat	4.8	5. Pigmeat	4.7	5. Maize*	4.2
6. Maize*	4.1	6. Maize*	4.4	6. Wheat*	4.0
7. Orange	3.5	7. Orange	3.3	7. Coffee	3.8
8. Sweet Potato*	2.9	8. Coffee	3.1	8. Groundnut*	3.7
9. Cotton	2.8	9. Groundnut*	3.0	9. Banana*	3.6
10. Eggs	2.8	10. Cassava*	2.9	10. Pigmeat	3.3
11. Potato*	2.8	11. Eggs	2.8	11. Orange	3.0
12. Coffee	2.7	12. Sugar	2.8	12. Sugar	2.9
13. Sugar	2.7	13. Cotton	2.7	13. Cotton	2.6
14. Groundnut*	2.6	14. Potato*	2.7	14. Eggs	2.4
15. Tobacco	2.6	15. Banana*	2.6	15. Sheep & Goats*	2.3
16. Grape	2.5	16. Sweet Potato*	2.3	16. Potato*	2.1
17. Soybean*	2.5	17. Tobacco	2.2	17. Cocoa	2.0
18. Banana*	2.1	18. Grape	2.1	18. Poultry	2.0
19. Cassava*	2.0	19. Soybean*	2.1	19. Grape	1.9
20. Poultry	1.9	20. Poultry	1.9	20. Yam*	1.9
21. Sheep & Goats*	1.7	21. Sheep & Goats*	1.9	21. Tobacco	1.8
22. Tomato	1.2	22. Phas. Bean*	1.3	22. Phas. Bean*	1.6
23. Phas. Bean*	1.1	23. Coconut*	1.3	23. Millet*	1.5
24. Coconut*	1.1	24. Cocoa	1.2	24. Sorghum*	1.5
25. Apple	1.0	25. Sorghum*	1.1	25. Soybean*	1.5
26. Rubber	1.0	26. Tomato	1.1	26. Coconut*	1.4
27. Cocoa	0.8	27. Yam*	1.1	27. Sweet Potato*	1.4
28. Onion	0.8	28. Millet*	1.0	28. Rubber	1.3
29. Sorghum*	0.8	29. Rubber	1.0	29. Palm Oil	1.1
30. Tea	0.8	30. Apple	0.9	30. Tomato	1.1
31. Lemon & Lime	0.7	31. Palm Oil	0.8	31. Cowpea*	0.9
32. Millet*	0.7	32. Tea	0.8	32. Tea	0.9
33. Palm Oil	0.7	33. Lemon & Lime	0.7	33. Apple	0.7
34. Barley*	0.6	34. Onion	0.7	34. Onion	0.7
35. Yam*	0.6	35. Barley*	0.6	35. Pineapple	0.7
36. Chickpea*	0.5	36. Pineapple	0.6	36. Barley*	0.6
37. Pineapple	0.5	37. Chickpea*	0.5	37. Lemon & Lime	0.6
38. Broad (Faba) Bean*	0.4	38. Cowpea*	0.5	38. Broad (Faba) Bean*	0.4
39. Cabbage	0.4	39. Broad (Faba) Bean*	0.4	39. Chickpea*	0.4
40. Cowpea*	0.2	40. Cabbage	0.4	40. Cabbage	0.3
41. Jute	0.2	41. Lentil*	0.2	41. Lentil*	0.2
42. Lentil*	0.2	42. Pigeonpea*	0.2	42. Pigeonpea*	0.2
43. Pigeonpea*	0.2	43. Jute	0.1	43. Jute	0.1
44. Hemp	0.0	44. Hemp	0.0	44. Sisal	0.1
45. Sisal	0.0	45. Sisal	0.0	45. Hemp	0.0
99.5		100.1		100.0

* CGIAR Commodities

Of the remaining commodities, coffee, pigmeat and cotton rank within the top 15 commodities. However, as noted in Sections 5.7 and 6.2, TAC's view on these and the other non-CGIAR commodities (eggs, pigmeat, cocoa, poultry, sugar, tobacco, rubber, tea and fruit of different types) is that for a variety of reasons (such as the existence of private sector research or other sources of research supply), there is no compelling case for considering major commodity improvement efforts in these commodities at this time. TAC recognizes the importance of these commodities for smallholder farming systems and for incomes of landless labourers, but would encourage centres to undertake research on these commodities within the framework of production systems research. Furthermore, the CGIAR is now in the process of giving much greater attention to resource management issues which implies a shift away from the original focus on plant breeding that characterized the CGIAR. TAC considers that the CGIAR should consolidate its commodity portfolio before embarking on new initiatives.

12.3.5.2. The congruence between the modified values of production and current allocations

TAC considered the relative distribution of resources among existing CGIAR commodities. It looked at the results of TAC's modified congruence analysis and of the analysis by ACIAR. The modified values for CGIAR commodities and estimates of current (1991) actual allocations of core resources in the expanded CGIAR are presented in Table 12.4. The regional breakdown of the modified values with weights of 0.5 is given in Table 9.17 (Section 9.9.1.2) and with weights of 0.25, 0.5 and 1.0 for selected commodities in Table 9.22 (Section 9.10.3).

For all the developing regions combined, the relative modified values of commodities are: cereals 36.3%; roots and tubers 14.5%; food legumes 13.0%; oil crops 2.0%; vegetables 3.2%; banana and plantain 5.2%; and livestock 25.8 %. These figures compare with 1991 core resource allocation of cereals 44.3%, roots and tubers 11%, food legumes 16.3%, banana and plantain 2.3% and livestock 26.1%.

In the strictest sense of congruence analysis, divergences between allocations and modified values could be used to raise the question of whether CGIAR resources should be reallocated from cereals and food legumes towards roots and tubers and bananas. In so doing, TAC also recalled the caveats associated with the estimation of value of production of commodities, and prior to making recommendations the outcome of each analysis was therefore considered with great care.

Cereals

Actual allocation to cereals is approximately 2% higher than indicated because investment in rice breeding networks has been considered within the category of institution building. TAC notes that in 1986, it had recommended that the relative emphasis given to rice should be reduced. TAC had based its recommendation both on the over-emphasis on rice relative to the commodity's importance in global food supplies and on the strength of national programmes in Asia. For a variety of reasons well known in the CGIAR, the relative funding for rice has not declined as much as recommended by TAC. This reflects increases in nominal expenditures in both Asia and sub-Saharan Africa.

The Committee debated reaffirming the earlier recommendation to reduce the relative allocation to rice, particularly in Asia, primarily because of now even stronger national programmes. The West African question is more complex. Rice is rising rapidly as a component of diets, substituting for traditional staple cereals and roots and tubers, especially in urban areas. Further, weaker national programmes and higher research costs make African research in general more expensive. Finally, the CGIAR has already decided to have a major upland rice improvement effort in sub-Saharan Africa. TAC had, therefore, recommended a minimum effort that has a reasonable chance of success.

TAC recognizes the importance of rice as the staple food of a large number of people in developing countries and the substantial pay off from CGIAR efforts on rice research in Asia and Latin America. TAC has also considered the large benefits from rice research that accrue to consumers and producers. TAC therefore recommends a continuation of current levels of CGIAR investment in rice research, but a shift in focus of rice research towards more strategic germplasm research necessary to lift the yield ceiling of the crop, and to sustain current yield levels.

Table 12.4. Modified values of production and 1991 estimated allocation of core resources among CGIAR agricultural commodities^1/

Commodity		Modified VOP Baseline (w == 0.5) (%)	1991 Allocation (%)
Cereals		36.3	44.3
	Rice	19.2	18.4
	Wheat & Barley	6.7	10.8
	Maize	6.1	9.7
	Sorghum	2.2	3.2
	Millet	2.1	2.2
Roots & Tubers		14.5	11.0
	Cassava	6.6	5.4
	Potato	3.1	4.3
	Sweet Potato	2.0	1.1
	Yam	2.8	0.2
Food Legumes		13.0	16.3
	Beans	2.3	5.4
	Broad Beans & Lentil	0.9	2.2
	Chickpea	0.6	2.2
	Cowpea	1.3	3.2
	Groundnut	5.4	2.2
	Pigeonpea	0.3	1.1
	Soybean	2.2	N/A
Oil Crops		2.0
	Coconut	2.0	-
Vegetables		3.2	-
Banana & Plantain		5.2	2.3
Livestock		25.8	26.1
	Beef & Buffalo Meat	8.5	N/A
	Sheep & Goat Meat	3.3	N/A
	Milk	14.0	N/A

N/A not available
^1/commodity research only

TAC is mounting an inter-centre review of rice in the CGIAR in conjunction with external reviews of IRRI and WARDA. This inter-centre review will also consider ongoing efforts on rice research at CIAT, IITA, IBPGR, IIMI and at the stronger national programmes. Thus TAC will continue its discussion of relative rice priorities in its March 1993 meeting and provide the Group with updated views at that time.¹

¹TAC/CGIAR, 1993. Investment in Rice Research in the CGIAR: A Global Perspective -Report of the Inter-Centre Review of Rice.

With respect to wheat, TAC has been impressed by the continued large pay off and impact in the farmers' fields of CGIAR investments in this commodity. While a comparison of modified value of production with current CGIAR resource allocation would suggest an over-investment in wheat research, TAC recognizes the many distortions in the marketing and trade of wheat. It noted the existing of wheat research programmes in developed countries but reaffirmed the special role of the CGIAR in catering for the needs of developing countries. TAC also took into account the higher costs associated with strategic research and the need to maintain efforts in maintenance breeding. TAC recommends that CGIAR efforts on wheat should give primary attention to strategic germplasm research. Overall, TAC recommends a continuation of efforts in CGIAR investment in wheat research in the medium term, recognizing the potential for further gains to be made in increasing the productivity of wheat. In the long term, the priority of wheat is likely to decline given the growing importance of alternative sources of research supply.

TAC recommends, in the short to medium term, maintaining current efforts in maize research, given the importance of the crop in mixed cropping systems of resource poor farmers. TAC has noted that the private sector is rapidly becoming involved in the maize seed industry, particularly with respect to hybrid maize varieties. In the long term, a reduction in the priority of maize research is therefore likely.

Pearl millet is an important crop in sub-Saharan Africa, particularly in Sahelian West Africa, where approximately half the world's production is found. Millet is generally an important crop for the poor and is grown in farming systems in less endowed regions. In Asia, millet is mainly produced in India. In 1986, TAC recommended a greater concentration of effort in pearl milled on sub-Saharan Africa over the short term. This was because of the short research history and the weakness of national programmes in the millet producing countries of sub-Saharan Africa, compared with the programme of India. ICRISAT responded positively to this recommendation by transferring some of the millet responsibilities from the ICRISAT Centre to West Africa and the SADCC region.

The pay off from CGIAR investments in research on pearl millet has been substantial, particularly in the more endowed farming areas of India. ICRISAT is now shifting its focus towards the drier areas of more limited potential, and towards more strategic issues. TAC endorses this view and recommends to maintain current CGIAR efforts on millet research.

With respect to barley and sorghum, TAC saw on balance no reason to alter ongoing CGIAR efforts and recommended that the level of investment in these commodities should continue. Research on these commodities should continue to focus on those areas where poor farmers are heavily dependent on these crops.

Roots and tubers

With respect to investment in roots and tubers, TAC noted the divergence between modified value of production and CGIAR allocations. Nevertheless, TAC recommends maintaining current efforts in cassava and the other root and tuber crops.

The Committee recognized the rapid rate of progress in cassava research, and the importance of cassava for low income consumers and producers. It also considered the negative income elasticity of cassava in most regions, and the need for the development of post-harvest technology. On balance, TAC recommended a continuation of efforts in cassava research.

TAC noted that over 80% of the global production of sweet potato is in China (which has a relatively strong national research system), that there has been a steady decline in the importance of sweet potato as a staple food, and that there have been shifts in product utilization towards livestock feed in Asia. The Committee recognizes that very little research on sweet potato is conducted outside the CGIAR, that ongoing CGIAR efforts are of very recent origin, and that outside China sweet potato is a very important crop in a large number of small countries typically with very low income levels. For the medium term, TAC therefore recommends that CGIAR efforts be maintained at their current level.

The relative importance of another root and tuber, potato, fell in the modified ranking, mainly because the crop is predominantly grown in the subtropics and cold tropics. The production and consumption of potato is growing rapidly in developing countries, and good rates of progress have been obtained from CGIAR investments in potato research. Furthermore, potato is an important crop for low-income farmers and consumers. TAC recommends that, in the medium term, current efforts in potato research be maintained.

TAC noted that yam is produced mainly in West Africa, with Nigeria accounting for 70% of world production. In its 1986 review of priorities, TAC had recommended the continuation of efforts on yam in the short term, followed by a performance review in five years. The assessment of the Third External Review of IITA was that the comparative advantage in crop management research related to yam lay with the Nigerian national research system. IITA should concentrate on germplasm conservation and focus its research on the critical constraints to germplasm improvement. TAC endorses this view, and suggests that a review of the effectiveness of CGIAR research on yam should be part of the next external review of IITA, which has the global mandate for this commodity.

Food legumes

In 1986, TAC had recommended that investments in research on faba bean (broad bean) and lentil be phased out. A CGIAR involvement was to be limited to the maintenance of genetic resource collections. The responsibility for faba bean research is in the process of being transferred to a national programmes. While ICARDA's efforts in this regard have been successful, it will take more time than expected before its responsibilities can be fully discharged. ICARDA has also been requested to undertake an in-depth assessment of the potential pay off from further research on the improvement of lentil. The outcome of the assessment will be considered by TAC when ICARDA's next medium-term plan is presented. TAC notes that lentil is an important crop in farming systems of resource poor farmers of the West Asia-North Africa region, and that outside the CGIAR very little research is carried out on lentil. In the short term, therefore, current efforts in lentil could continue, while reaffirming TAC's view that in the long term the role of the CGIAR in faba bean and lentil research should be primarily in maintaining genetic resource collections.

In view of the progress made by the centres concerned on phaseolus beans in Latin America and the Caribbean, and sub-Saharan Africa, and on pigeonpea in India, TAC proposes to reduce the emphasis on these two commodities. CIAT has successfully developed several improved varieties of phaseolus beans that are being widely adopted by farmers. The pay off from CGIAR investment in research on this crop has been substantial. TAC also recognizes that phaseolus beans are an important crop for poor farmers and of substantial significance in the diets of low income consumers. On the basis of congruence criteria, the CGIAR is overinvesting in phaseolus beans however (Table 12.4). CIAT is already proposing a substantial reduction in the scale of its research programme on phaseolus beans. TAC endorses this trend, and recommends a modest reduction in the priority of this crop.

Pigeonpea is an important crop in one country only (India) which in addition has a strong national research programme. The rapid progress achieved by ICRISAT in developing a hybrid pigeonpea variety, adds further weight to the argument that responsibilities for pigeonpea research, particularly in India, can now gradually be transferred to the national research programme. TAC recommends that CGIAR efforts in pigeonpea research be progressively reduced in the long term to limit its activities to maintaining genetic resource collections only, and be diminished significantly in the medium term.

Chickpea is an important dietary item for poor consumers in southeast Asia, India and the West Asia-North Africa region. In the short to medium term, the priority of this crop could be maintained.

TAC recalled that cowpea was largely produced in Nigeria, but that the crop could be an important commodity for resource poor farmers throughout different agroecological zones and cropping systems of West Africa, where national systems are generally still weak. TAC concluded that it would therefore be appropriate to continue CGIAR support for this commodity in the short to medium term.

TAC noted that soybean ranked above several other legumes and the rapid progress of research on soybean conducted in Asia, particularly with respect to multipurpose varieties. TAC recognizes that soybean has substantial potential in developing countries and that the demand for livestock feed is growing rapidly. The Committee recommends an increased resource allocation to this commodity in sub-Saharan Africa, in view of the likelihood of rapid progress, particularly with respect to its potential as a nutritious food, cash crop, and protein-rich livestock feed.

Among food legumes, the only commodity that appears underfunded is groundnut. TAC has noted the substantial pay off from CGIAR investments in groundnut research in Asia as reported by the Third External Review of ICRISAT. In sub-Saharan Africa, progress has been slow. TAC recommends increasing current efforts in groundnut research modestly given the opportunities for further gains to be made through research on this commodity.

Oilcrops

The modified ranking of coconut is only 2.0% (Table 9.17). However, TAC observed that its earlier recommendation to include coconut in the CGIAR portfolio was based on its importance as a smallholder multipurpose tree crop in several farming systems throughout the tropics. TAC hereby notes the prospects of high returns from research investments, the benefits to low-income producers, and the lack of continuity in historical research efforts. Coconut is also an important crop for the sustainability of agricultural production in coastal ecosystems.

The priority ranking of another oilcrop, groundnut, has been considered within the category of food legumes.

Banana and plantain

The CGIAR has recently expanded its activities in banana and plantain research through the incorporation of INIBAP. Most of the investments of this Centre are not incorporated in the allocation figure given in Table 12.4 because they have been classified as network expenditures. Actual CGIAR efforts in banana and plantain are thus substantially higher than indicated. TAC recommends maintaining current efforts in CGIAR research on these commodities.

Vegetables

The analysis undertaken for assessing a possible expansion of the CGIAR and for the current review of CGIAR priorities, confirms TAC's 1988 recommendation that research on vegetables is a high priority for the CGIAR.

Livestock

TAC has carefully considered the relative priority assigned to livestock research. In particular, TAC noted that while congruence on livestock research appears close, in practice the modified value of production figure has to be treated with considerable caution. The figure includes the value of all cattle meat and milk, and small ruminants across all regions. The CGIAR does not conduct any livestock research in Asia however, nor does it conduct research on small ruminants in Latin America or on large ruminants in the West Asia-North Africa region. When adjusting the modified production value for these factors, the figure would be reduced from 25.8% to 16.5%, with a distribution of 40% to sub-Saharan Africa, 9% to West Asia-North Africa, 27% to Asia and 24% to Latin America. Actual resource allocation to livestock research amounts to more than 26%, distributed as 73% to sub-Saharan Africa, 9% to West Asia-North Africa, 16% to Latin America and only 1% to Asia. This would, in turn, suggest that the CGIAR is substantially overinvesting in livestock research, even when considering the important role of intermediate livestock products such as traction and manure. A disproportionate share of CGIAR resources for livestock research is allocated to sub-Saharan Africa. In addition to the regional emphasis, major questions remain about the distribution of species emphasis and between animal production and health research. Furthermore, it is now increasingly clear that in the future, much greater emphasis will have to be given to enhancing crop-livestock interactions. The major constraint to increasing livestock productivity is a shortage of feed, and this could, to a greater extent, effectively be addressed through more adequately focused crop productivity research. TAC is also concerned about the generally slow rate of progress in obtaining impact from CGIAR investment in livestock research. TAC, therefore, recommends a modest reduction in the priority the CGIAR currently allocates to livestock research. In the medium term, the allocation to livestock research in sub-Saharan Africa could be reduced substantially. When the livestock study, currently being undertaken by Winrock International, and the external reviews of ILRAD and ILCA are completed, TAC will revisit livestock research priorities.¹

¹TAC/CGIAR, 1993. Priorities and Strategies for Livestock Research in the CGIAR.

12.4. Conclusions

With respect to the assessment of priorities, in Systemwide terms, by activity category, TAC recommended a substantial increase in the priority allocated to conservation and management of natural resources including germplasm conservation, and of socioeconomic, public policy and public management research. The Committee recommended a reduction in the priority allocated to development and management of production systems and of institution building, while the current priority ranking of germplasm enhancement and breeding is to be maintained.

TAC considered that in the medium term the share of resources allocated to sub-Saharan Africa and to West Asia-North Africa should be reduced modestly, while the share allocated to particularly Asia and Latin America should be increased. The Committee did not make a recommendation on the level of priority by production sector, but notes that the proposed new programmes in forestry and fisheries should not be funded at the expense of critical research needs in crops and livestock.

TAC reaffirmed the priority it is currently allocating to the cereal and root and tuber crops. Among food legumes, it proposes to reduce the level of priority of phaseolus bean modestly, and that of pigeonpea significantly.

The priority ranking of groundnut and soybean should be increased, while that of faba bean and lentil, chickpea and cowpea should be maintained. The current level of effort on banana and plantain will also be maintained, and TAC reaffirmed its views on the priority level for research on vegetables and coconut. Finally, TAC considered the priority ranking of livestock and noted that the CGIAR is currently overinvesting therein, particularly in sub-Saharan Africa.

With respect to recommendations on forestry and fisheries, TAC reaffirms the outcome of the analysis made in Chapters 7 and 8. Furthermore, TAC is pleased with the good progress that is being made in the implementation of a new forestry research initiative in the CGIAR through the establishment of the Centre for International Forestry Research (CIFOR) and in ICLARM's efforts to develop a strategic plan for international fisheries research in the CGIAR.

12.5. Final Observations

The approach taken by TAC to this priority analysis is more comprehensive and a great deal more quantitative than previous TAC efforts. We have attempted to bring into the analysis quantitative indicators of the most important dimensions of the CGIAR mission and goals. We have also carefully reviewed the outputs of similar efforts such as those of ACIAR. A major conclusion arising from TAC's analysis is that the current constellation of activities in an expanded CGIAR is largely congruent with present and future research and research-related activity needs but that much greater emphasis still needs to be given to natural resource conservation and management. Although in the medium term further changes will be required in the overall balance of effort, these will not require a dramatic departure from current activities. Stated another way, TAC's review of activity balance, regional distribution of resources and commodity congruence suggests that the "founding fathers" of the CGIAR, and its changing membership since, have charted a course that allows for evolution and change and continues to address high priority issues.

This should not be surprising, because the challenges that faced the CGIAR at its birth have, if anything, become more serious since 1970. Population growth continues at high rates, particularly in sub-Saharan Africa, poverty and malnutrition remain pervasive, the need for increased productivity grows more acute as the opportunities for area expansion diminish, and long-term issues of sustainability have become both more prominent and severe. Thus TAC finds that the focus on applied and strategic research of international importance focused on the twin needs of productivity improvement and sustainable resource management for agriculture, forestry and fisheries is indeed more important now than it was in 1970. Despite a broadening commodity portfolio and additional interests in natural resource management (particularly in agroforestry and forestry), the CGIAR remains a highly focused organization. The System still devotes critically necessary levels of resources to selected commodities of major importance, and still focuses on a set of research activities that are most efficiently and effectively conducted internationally.

The ability of the CGIAR to continuously adapt to changing circumstances should be seen as a strength of the CGIAR. The Group has always been forward looking. The CGIAR in 1991 is both similar to, and different from, the innovative model created in 1971. It is similar in its commitment to improving the lot of the poor in developing countries by increasing their access to an affordable and sustainable food supply. It is different in its scale, breadth of activities, and emerging partnerships with developing countries. The challenge ahead remains enormous.