The World Bank
INTERNATIONAL BANK FOR RECONSTRUCTION AND DEVELOPMENT
INTERNATIONAL DEVELOPMENT ASSOCIATION
1818 H Street, N.W.
Washington, D.C. 20433
U.S.A.
(202) 473-3980
Cable Address: INTBAFRAD
FAX: (202) 676-0199
February 10, 1995
Dr Donald Winkelmann
Chairman
Technical Advisory Committee
Mr Chairman,
Attached is the final report of the Panel on CGIAR Commitments in West Africa. I again thank you and your predecessor for this interesting task.
The Panel benefitted initially from a Desk Study of Center commitments, done by the TAC Secretariat 1/. We visited West Africa in the last week of June 1994 and met TAC members and national program representatives during TAC 64 at Bouake, Côte d'Ivoire. The Panel traveled to Mali, Burkina Faso, Niger, Nigeria, Ghana, Benin, Cameroon, and the Gambia in August of 1994. It presented its draft report to TAC at the Committee's 65th meeting. The Panel met at the TAC Secretariat from February 7, 1994 through February 9, 1994 to consider the various comments on the draft report made during TAC 65 and afterwards.
1/
The Desk Study (June 1994) is not attached to the Panel's final report, but is available from the TAC Secretariat.
The report now incorporates the Panel's reponses to comments made by the Centers, the Center Directors' Committee, TAC itself, the NARS and others at TAC 65. I have replied in detail to the comments sent by the TAC, the CDC, ICRISAT, IITA, ILRI, ISNAR, and WARDA and discussed Dr Pinstrup-Anderson's remarks with him by phone; I have copied the replies to you and to the TAC Secretariat.
Let me reiterate some of our main points and highlight the changes we have made since the draft report, both in response to comments and upon further reflection about some of the issues.
Comments by the TAC
The final report responds fully to TAC comments on the draft. Here we highlight a few of the main points made by TAC. One of TAC's principal observations concerned issues in the report that went beyond the sub-region. TAC noted that it might be helpful if a section of the final report covered general systemic issues; this has been done in Annex 3 of the final report, which also discusses how this study might be done in other regions where the Centers are active.
Genetic exports. TAC noted that the draft report's comments on 'genetic exports' required further clarification because it considered that some of the Panel's suggestions exceeded the brief of the CGIAR. The Panel agrees with this view; the final report states that actions in the area of genetic exports might be taken jointly by the Special Project for African Agricultural Research (SPAAR) and regional institutions in West Africa, not by CGIAR institutions.
Nigeria. TAC commented that the draft report appeared to recommend an excessive concentration of Center efforts in Nigeria. We believe that the Center's overall allocation to Nigeria is reasonable, in view of that country's importance and the diverse opportunities it offers for research with implications outside its borders. We have recommended that a few additional staff of IFPRI and ICRISAT work on Nigeria, but this does not affect the aggregate very much.
Comments by the CDC
Several comments in the CDC summary are copied from the remarks sent by particular Centers. I have answered them in the individual replies and in the final report as appropriate. Reference to some of the CDC's observations follows.
The small NARS. These NARS basic problem is that they are too small to justify major investments by individual Centers in view of other demands on Center resources. Their basic need is more trained people; to fill that need, they must have academic training. Those are the reasons why a stable, long-term arrangement with academic institutions is both essential and preferable to a like arrangement with a Center. This does not exclude Center contribution to these countries through the usual mechanisms, nor need it interfere with the development of a pluralistic research system.
Administration costs. The draft report did not endorse the opinion that IARC administrative costs were higher in West Africa than in industrial countries; it only referred to such an assertion having been made in a TAC paper and then preceded to say why the Panel lacked the information to evaluate the assertion and to speculate about why such an assertion would be groundless anyway.
Water and irrigation issues. The final report argues that many of what appear to be researchable problems in West Africa on water and irrigation have solutions that are already well known and can be easily transferred from other regions where irrigation is much more important than West Africa. Examples are marginal cost water pricing, strong property rights, efficient water markets, and the transfer of public irrigation schemes to private farmers.
We assumed that most issues about IIMI, including its regional allocation to West Africa, were covered by its recent external review 1/. On one particular point, the CDC/IIMI noted that "the report also appears to discount the strong interest shown by governments in exploring fully the possibility of applying irrigation to their food security needs". Naturally we do see the importance of developing irrigation in the region where it is economically efficient. But some of the "possibilities of applying irrigation" in the region are good illustrations of the failure to develop a long-term perspective, as discussed in Annex 2 of the Report. The Centers should not provide research to support inefficient policies which lead to unsustainable production systems just because governments are interested; research on managing irrigation for wheat production in Nigeria comes to mind.
1/
We further recognize our failure to say nothing about ICLARM or about fisheries in general, but this error can perhaps be rectified by the current ICLARM external review.
Organization and Process
We saw no reason to change our view that the current organization of the CGIAR commitments in the region is reasonably efficient and cost-effective. (You will not be surprised to learn that the Centers had no disagreement with this conclusion). While administration costs need to be investigated in individual Center reviews, we doubt that savings from various reforms are very consequential.
A key process issue is how to incorporate NARS opinions into the formulation of IARC programs. Though we have received comments on the draft report only from Togo, the Gambia, and Ghana (from Professor Haizel, who is a regional representative to the CGIAR) on the draft report, the report is based partly on discussions with national programs in seven countries (Mali, Burkina, Ghana, Niger, Nigeria, Benin, and the Cameroon) and Dr Ouayogode is himself a NARS representative. Annex 1 of the report reviews WARDA's experience with this question. We adhere to our conclusion that current mechanisms of collaboration between the IARCs and the NARS are efficient and do not require systemic changes.
The Panel found four other structural or process reforms that would, in its view, significantly improve the impact of the IARCs in West Africa. The first is to lessen the concentration of ICRISAT research at Sadore. Given that ICRISAT has already moved to diversify its work away from Niger, one may consider this a moot point, but it is one that requires continuing re-evaluation in light of ICRISAT's overall resource allocation in the region and in light of the universal difficulties in raising agricultural productivity in the semiarid tropics. A second concerns the ILRI programs in Nigeria, where we have recommended a consolidation of two small programs and a firmer commitment to a stable ILRI presence in the humid and subhumid zones. The third is to make IFPRI the strong convenor of social science research across the Centers in West Africa. The fourth is to deemphasize and reorient production systems and management research in the IARCs. We return to these points below.
Production Systems and Management Research (Category 3)
This issue provoked voluminous comments from individual centers and from the CDC. There was, I think, some misunderstanding. We did not say that the Centers should "forego research" on Category 3; we did say that they should concentrate on basic process problems, as ICRAF is doing, and develop explicit partnerships with the national programs to devolve the leading role in applied production systems research to the NARS. Nor did we say that no one should do this work; the NARS should do most of it The growth of national research capacity in the region, and the many failures in this category worldwide, impose a smaller role for the IARCs.
We therefore maintain our recommendation that the Centers should devolve that part of their work in Category 3 that is not consistent with their main responsibility for strategic research to the national programs. Resulting savings should be invested in Category 2 (Germplasm Enhancement and Breeding) because most productivity gains have come from that category. I expect that implementing this recommendation will require some attention from the external reviews of each Center.
The Role of IFPRI
The draft report of the Panel recommended that IFPRI be designated as the convenor of all Center social science research in the region, including that of ISNAR, to provide stronger leadership in that field. Despite the general unpopularity of this recommendation, we have retained it. We believe that social scientists based in West Africa and those working on it from IARCs outside the region, are too isolated, lack a coherent strategy and unified approach to common problems, and have failed to exploit important opportunities in this field. The recommendation is definitely not that IFPRI should do all this research, but that it should lead it.
Dr Ryan and Dr Pinstrup-Andersen have alike objected that such an arrangement will weaken microeconomic research now done in the commodity Centers in close contact with natural scientists. While we take this objection seriously, it is not an insurmountable problem. The point of the arrangement would be to ensure that such work fits into the policy analysis, much of which is impossible without a good knowledge of the technical relations.
Institution Building, Training and Information
We have kept the draft's recommendation that the Centers, with the exception of ISNAR, should limit their activities in Category 5 (institution building) to training and information and should abandon organization and management counseling because it is not in their comparative advantage. The overall size of training and information activities should also be reduced and the savings transferred to Category 2 in view of the many sources of training and information services outside the Centers. We maintain the opinion that ISNAR's future in the region is not well-defined and believe that designating IFPRI as the regional convenor of IARC social science research would define that role more clearly.
The Impact of ICRISAT and ILCA
In a letter to Dr McCalla, I said that ICRISAT and ILCA, in contrast to IITA, have "really had no impact in West Africa and it is very hard to see what the impact of some of their current work will be." It might have been better initially to put that conclusion in a broader context; it has been difficult to achieve research impact in dry areas everywhere - in the United States, Latin America, Australia, the Middle East, as well as Africa - and this is the challenge for everyone, not only for the Centers.
My earlier conclusion provoked a long reply from ICRISAT and a briefer one from ILRI. With those replies, and with other additional information, we have again looked carefully at impact.
In light of ICRISAT's comments about the scientific impact of research, including publications, apart from their eventual production impact, we have modified the report in one place to give due credit to scientific impact. We note here that ICRISAT has produced more than twice as many advanced lines and varieties of millet, sorghum, and groundnut in West Africa in 1993-94 as it did in 1985-92; this is not production impact, but it is a hopeful sign.
ICRISAT commented extensively about the reasons for its lack of production impact in West Africa, most notably for the failure to have measured adoption of its varieties at even the local level. In making the criticism, we recognized fully the environmental difficulties involved, as well as the debility of the national systems of technology generation and transfer. While it is clear that ICRISAT is doing the right thing in moving resources away from Sadore to other sites in the region and in changing its aggregate structure to achieve better synergies between African and non-African research, we have not changed two major recommendations about ICRISAT. The first is to devolve much of Category 3 research to the national programs and to concentrate more on basic process questions.
The second is to commission a fundamental scientific review of ICRISAT's crop improvement work in West Africa, including that of CIRAD for sorghum. The Panel did not have the competence to do this, but my strong belief is that such a review is needed because the plausible solutions have been tried for a long time - introducing exotic materials, exploiting the characteristics of local materials, intensive screening for sources of biotic stress resistance, and increasing the harvest indices - with little noticeable field impact. Furthermore, the contrast between ICRISAT Asian successes with crop improvement and its failures with resource management suggests that we will observe a similar contrast in West Africa - as clearly seen in ICRISAT's section of "Current CGIAR Research Efforts and Their Expected Impact on Food, Agriculture, and National Development" (CGIAR Secretariat, March 1994) - this expectation logically shifts more emphasis to crop improvement. The failure to find durable solutions is not exclusive to ICRISAT, but has occurred with colonial research, the independent NARS, bilateral programs, and regional programs like SAFGRAD.
ICRISAT has replied to this second recommendation to suggest in its place a study of the constraints to adoption of improved materials of millet and sorghum. The Panel could not accept this suggestion. The basic constraint to adoption has long been evident: the introduced materials are not better than the locals under field conditions, even in farmer-managed trials when there are no problems with extension, input supply, risk, or marketing. We need to know why the "improved" materials are not really improved.
There is one other question about ICRISAT. The draft report recommended that the question of separating ICRISAT's African activities from its non-African work be considered again in the near future. The final report omits this recommendation to avoid giving the false impression that the Panel has endorsed the notion.
ILRI contended that the Panel "has not added any significant evidence on which to base its of assessment" beyond what was in the Impact Study of Africa. I assure you that we have. I have again reviewed ILCA's own submissions about its impact in West Africa, including Dr Fitzhugh's presentation at the World Bank last fall (which summarized 20 years of ILCA research), the ILCA material in a CGIAR paper ("Current CGIAR Research Efforts and Their Expected Impact on Food, Agriculture, and National Development", March 1994), and an ILCA study ("Potential for Impact: ILCA Looks to the Future", April 1992) published soon after ILCA's past external review. I have also interviewed Dr Boubacar Hassane, the president of the national livestock owners' association of Niger and who wrote his PhD on fodder production in northern Nigeria, to discuss the impact of ILCA's work. I see no convincing reason to change the Panel's report on ILCA's current impact. While, as ILRI's reply to the draft report says, it may be true that "the jury is still out" on the future impact of both fodder banks and alley farming for livestock production, the benefits of research to the present have been less than the costs. Given the paucity of ILRI's expected allocation to West African research, I remain very doubtful about the future impact.
A Common IITA and WARDA Board
The draft report erroneously justified its proposal for a common IITA and WARDA Board in terms of an insignificant benefit, i.e., cost savings. The real benefit of a common Board is to harmonize research. We have, therefore, retained the recommendation and have further incorporated Dr Lampe's suggestion that IRRI be represented on the proposed common Board.
Let me say that the Panel strongly supports the Centers' efforts in West Africa, admires the significant successes that have been achieved and expects those successes to multiply in the future. We hope that nothing in this report will be used to diminish the overall level of resources available for Center activities in the region. While our report perhaps emphasizes criticisms of the Centers by the national programs, we found that the NARS are generally very appreciative of the Centers' research; the best proof of this is their growing interest in closer scientific collaboration.
I would like to close by thanking you and Dr McCalla again for this assignment, and by expressing my appreciation to Dr Bakary Ouayogode, the other Panel member, as well as to Dr Philip Kio of the TAC Secretariat, for their work on the report.
Sincerely,
John McIntire
Panel Chairman
