SETTING AGRICULTURAL RESEARCH PRIORITIES FOR
THE CENTRAL AND WEST ASIA AND NORTH AFRICA REGION
(CWANA)

Toward a new NARS/NARS and CGIAR/NARS collaboration spirit

III. RECOMMENDATIONS

The CWANA-wide priority setting exercise was concluded by a series of recommendations geared to facilitate the implementation of the identified regional research agenda and enhance the likelihood of impact within a relatively short-to-medium term horizon. These recommendations fall within five major headings outlined below: i) General; ii) Gap analysis; iii) Mechanisms to facilitate implementation; iv) Challenge programs of importance to the region; and v) Follow-up committee.

3.1 General recommendations

Building on the discussions and debate that took place during this regional consultation, two major recommendations emerged. They specifically addressed the CGIAR involvement in the region and the level of its resources allocated. The recommendations were stated as follows:

• In view of the state of poverty, natural resource degradation, and the level of food insecurity in the region, the participants strongly recommended that the CGIAR considers the inclusion of CWANA as another of its geographic priority region, in addition to Sub-Saharan Africa and South Asia.
• The participants considered that the resources allocated by the CGIAR to the CWANA region are insufficient, in view of the tremendous challenges faced by the region, and should therefore be significantly increased.

The analysis of the current status of regional cooperation/collaboration (both NARS/NARS and NARS/CGIAR) highlighted four major "gap areas" requiring immediate action in order to better address the identified research priorities of the region.

• · Various partnership arrangements and models are in place in the region. These involve NARSs, international CG and non CG-centers, regional centers, UN organizations, etc., with virtually no interaction among them. In view of the potentially high rate of duplication and waste of resources, it is suggested to make a thorough and critical assessment of the existing partnership arrangements to propose corrective measures in order to enable these partnerships to better address the identified research priorities of the region.
• Throughout the priority setting process it became very clear that the two SROs (AARINENA and CAC-NARS Forum) were not well equipped (in terms of resources, organization, and representativity) to adequately address the priority needs of the region. It is therefore strongly recommended to strengthen both organizations to enable them to fully play their role with regard to the implementation of the identified research priorities.
• The gap analysis has also identified the insufficient involvement of many CG Centers among which ILRI, ICRAF, CIFOR, ICLARM, IWMI, ISNAR, and IFPRI. Considering the mandates of these centers and the important contribution they could make with regard to improving agricultural research and development of the region, their involvement in the region should be significantly enhanced.
• Fruit trees and dryland oil crops are gaining in importance throughout the region. As these crops are not covered by the mandate of any specific CG-Center, it is suggested to expand the mandate of ICARDA to include these crops.

• To re-enforce regional cooperation and collaboration and facilitate implementation of the identified regional research priorities, three key mechanisms were identified, namely networks, coordination meeting, and traveling workshops(12).
• Networks: the existing networks need to be reviewed and consolidated. To undertake such a review, it is suggested that AARINENA and CAC-NARS Forum hold a joint meeting to be facilitated by ICARDA.
• Coordination meetings: to enhance regional collaboration and cooperation, two types of coordination meetings are suggested:
  - Sub-regional meetings, including all stakeholders involved in agricultural research/development activities in each sub-region (including CG and non-CG centers as well as regional centers active in the region). The sub-regional meetings should be organized under the leadership of AARINENA or CAC Forum, with participation from ICARDA and GFAR, and be held every two years on a rotational basis among countries of the sub-regions.
  - Regional (CWANA) meetings with participation from all relevant stakeholders (the latter including CG and non-CG centers as well as regional centers active in the region) under the leadership of AARINENA and CAC Forum and facilitated by ICARDA and GFAR. It was suggested that regional meetings be held every four years on a rotational basis among sub-regions.
• Scientist-to scientist interactions: to facilitate the implementation of the regional research agenda and the likelihood of its impact on the ground, it is suggested to foster interactions among regional scientists. Regional traveling workshops represent one of the mechanisms that would greatly facilitate and stimulate interaction. It is therefore suggested to encourage the organization of annual traveling workshops based on specific programs at the sub-regional level.
  

One issue that was discussed at length during the regional consultation relates to the Challenge Programs being developed by the CGIAR. At first, the participants endorsed three key CPs of direct relevance and interest to the region, namely the CP on Central Asia and the Caucasus (CAC), the CP on Desertification, and the CP on Global Genetic Resources and Genomics. The latter coincides perfectly with the region needs as germplasm improvement and biotechnology were identified as two research topics of high priority for the region. Second, to have the identified regional research priorities better reflected in the CPs put on the fast track by the CGIAR, the participants strongly recommended that: 1) the dry areas be included as a special focus in the CP dealing with water and agriculture and 2) the CAC region, the Atlas Mountains of North Africa, and the Anatolian Highlands be included in the CP dealing with mountain agriculture. Finally, it was recommended to encourage and stimulate the development of new ideas, derived from the identified regional priorities that could be presented as proposals for new CPs, with the understanding that these new CP proposals need not be restricted to CG CPs, but could be initiated within the region for direct presentation to potential donors(13).

3.5 Follow-up Committee

The final recommendation made was about the establishment of a committee to ensure that the region moves rapidly forward on the priorities identified and on the recommendations of the meeting. It was, therefore, suggested that AARINENA, the CAC Forum and ICARDA (as a facilitator) establish such a committee and be its key members.

(12) It must be pointed out that, with the exception of the suggested regional (CWANA) meeting, the other mechanisms have been widely used by ICARDA and the collaborative NARSs of the region. As shown in Annex 2 (Table A16), these mechanisms are regarded effective means to enhance regional collaboration and cooperation.

(13) With regard to these new CPs, it was suggested that in addition to the lead CG center(s), NARSs and other stakeholders, e.g., non-CG centers such as ICBA, ACSAD, etc., should also be included at the preparation stage of the proposals.

IV. OTHER ISSUES ADDRESSED BY THE REGIONAL MEETING

4.1 Role of the private sector

Although insufficiently represented in the various meetings held within the framework of the research priority setting process, the central role of the private sector was nevertheless emphasized. There was agreement that mechanisms need now to be developed to foster a public/private research partnership at the national and sub-regional levels and capitalize on its valuable capacity and experience as well as those of NGOs, increasingly active in the region. The few NGO and private sector's representatives who attended the meetings strongly insisted that in order to promote an effective multi-stakeholder dialogue and a bottom-up approach, there is a need to shift the focus from purely academic (research-driven) objectives and discussions towards more emphasis on elevating the debate to address concrete development issues and research-development linkages.

4.2 Natural resource management paradigm

Over the past decade there has been a clear additional research focus on integrated research programs, e.g., NRM(14). This additional focus on natural resource management is expected to significantly foster adoption of new technologies and will therefore scale up the reach of the products of commodity research. Given that the increasing focus on natural resource management is a common trend in the region, it is essential that it be clearly reflected in the regional research priorities and the associated modus operandi.

4.3 Time dimension of research priorities

In view of the rapid changes occurring throughout the world, a question was raised regarding the time dimension of the identified research priorities. The question raised was whether the aim of this regional exercise is about the identification of short, medium or long-term priorities? Although the question has no single and definite answer, the following issues must be considered when setting research priorities: i) in general, and especially in view of potentially high transaction costs, a regional approach is usually not the most appropriate approach to tackle short-term research issues; ii) a research priority setting exercise consists in re-visiting national agricultural research priorities in order to identify common regional concerns, i.e., it is not necessarily about the identification of new research activities but mostly about existing activities requiring better regional integration to achieve efficiency, synergy, and impact; iii) the time horizon is often determined by the research priorities identified. For example, research topics such as natural resource management represent necessarily a mid-to-long term endeavor; and iv) the setting of regional research priorities is necessarily a dynamic and "plastic" enough process to be reevaluated in the light of regional and/or global changes (Box 1).

(14) However, this by no means implies less emphasis on commodity research, which remains of vital importance in the region.

(15) Borrowed from Cosgrove, W.J. and F.R. Rijsberman, Editors. 1999. World Water Vision: Report. Worlds Water Commission's Staff Report. Version of 17 December 1999.

V. LIMITATIONS OF THE REGIONAL PRIORITY SETTING PROCESS

The key innovation of the CWANA regional priority setting exercise was to set the right conditions for a dialogue where "non-traditional stakeholders", i.e., farmers, NGOs, private sector, grassroots organizations, etc., would play a central role. The sub-regional brainstorming meetings, the regional survey, and the final regional consultation were all built-in mechanisms expected to ensure a bottom-up approach based on a broad multi-stakeholder dialogue that would enable the identification of regional research priorities that would truly reflect the agricultural priority problems faced in the region. However, while this regional process does indeed represent a major departure from past consultations on research priority setting, the derived decisional implications must be kept in perspective as, in spite of the efforts deployed, the process was nevertheless beset by a number of limitations key of which are highlighted below(16).

5.1 Criteria, weights, and scores

By definition, a research priority is, literally, an activity one implements first. The choice of well defined common evaluation criteria (and their associated weights) logically linked to sub-regional and regional research objectives must, therefore, be crucial in a priority setting exercise. In other words, the evaluation (whether through the scoring method or some other relevant method) of the research themes under consideration must be considered as a necessity, not an option. The mere listing of important (priority) research themes is not sufficient. It needs to be backed up (validated) by an agreed upon evaluation method. This is even more so in the case of a multi-objective (multi-criteria) framework because of implicit tradeoffs, i.e., giving more weight to a given criterion, for example, is only possible at the expense of the other(s) as the sum of the weights must equal unity. However, if the concerned stakeholders use different evaluation criteria and/or assign different weights to common criteria, the identification of (shared) sub-regional research priorities becomes a difficult task, to say the least. Therefore, it is important to understand that although the priority setting process should not be analysis-driven (in the narrow sense of the term) it needs, however, to be supported by some analysis. In other words, the identification of research areas of convergence at the regional level does require the definition of a framework of analysis based on agreed-upon criteria. The latter are used to specify the contribution (through assigned weights) that the research alternatives under consideration are expected to make towards the stated objective(s) of the region. In the course of the regional priority setting process, five such criteria were suggested for discussion and further refinement:

• productivity (competitiveness): the likely impact of the research priority identified on yield, cost/risk reduction, and product quality;
• Poverty alleviation (equity): the extent to which resource-poor smallholders landless laborers in rural areas, especially marginalized areas, would benefit from the identified research priority;
• Resource conservation (sustainability): the contribution of the research priority to the protection of the resource base;
• household food security: the likelihood of the research priority to increase the stability of staple food availability; and
• Contribution to development: the extent to which the considered research priority would contribute to overall development.

Once the common criteria are agreed upon, the next step would consist in assigning scores to the identified research alternatives in order to estimate their expected contribution to the agreed upon criteria. To discriminate among the research alternatives considered, a global scoring is then computed(18). Yet the framework of analysis (based on the scoring method, widely used in priority setting) that was suggested from the outset (sub-regional brainstorming meetings) generated a great deal of debate. Presented as a tool to facilitate the ranking (prioritization) of identified research themes, such a framework was viewed instead as an unwarranted "sophistication" of the exercise. Research priorities were, therefore, mostly addressed in general terms with virtually no mention of the criteria and the weights used to discriminate among them. With a few exceptions, NARSs' contributions were mostly focused on the general orientations and strategies adopted at the national level to foster agricultural research and development. Obviously, a scoring exercise would have greatly facilitated and improved questionnaire analysis and ultimately the identification of relevant and comparable sub-regional research priorities(19).

However, these acknowledged flaws notwithstanding, the discussions and sharing of experience achieved throughout the process have definitely enabled all stakeholders to move closer to the identification of shared research priority areas.

5.2 Achieving regional consensus

One major problem that emerged during the regional consultation stems from the heterogeneous nature of the five sub-regions. The intrinsic sub-regional heterogeneity implies that some of the priority problems identified during the brainstorming meetings, for example, would necessarily be sub-regional specific. This sub-regional specificity generated a misunderstanding during the regional consultation as sub-regional representatives were under the impression that any priority specific to a given sub-region would not be included in the agricultural research portfolio of CG-centers and other partners. As a result, lengthy discussions took place on the need to include some specifically sub-regional priorities into the regional agenda. It was reiterated that the purpose of this priority setting exercise is to identify areas of coincidence among the 5 sub-regions with regard to priority problems to be addressed through a regional research agenda. Obviously, such an approach is quite different from a linear addition of sub-regional priorities. Therefore, some priorities identified by the sub-regional groups would be best handled at the sub-regional level and, perhaps, even at the national level in some cases. One example illustrating this situation is grape production which was identified as a priority research area only in the CAC region. Clearly, the inclusion of grape production into the regional research agenda would not be justified, regardless of its importance to the CAC region. However, this by no means implies that sub-region specific priorities would be automatically dropped from CG centers and other partners' research agendas. Such priority areas could be tackled within a sub-regional framework that could involve the concerned NARSs and CG and/or non-CG centers. Once this issue was clarified to the satisfaction of all concerned, the comparative analysis got underway leading to the identification of regional priorities shown in Annex 1.

5.3 Participatory bottom-up approach

The effective participation of all concerned stakeholders in setting the regional research agenda was one of the basic tenets of this regional exercise. The valuable lessons learned from previous donor/research-driven programs have triggered the need to develop mechanisms that would expand collaboration and dialogue through sustainable links and strategic partnerships with "non-traditional" stakeholders, especially NGOs, farmers, grassroots organizations, the private sector, etc., the ultimate clients of agricultural research products. Through this regional priority setting process there was a genuine attempt to remedy this lack of collaboration and catalyze the research and development capacity of these "non-traditional" stakeholders. A lot remains to be done, however. At the NARIs' level in particular, the collaborative relationships with universities, NGOs, the private sector, farmers, and farmers' organizations are, by and large, at an embryonic stage and need therefore to be significantly consolidated.

5.4 Inadequate regional capacity to address the emerging research issues

To a large extent, the success (or failure) of sub-regional (regional) integration will hinge on the resource capacity of the concerned partners. Historically, human capacity building at the NARSs level has largely emphasized biological disciplines such as crop improvement, in particular breeding. The acute lack of capacity in other key disciplines such as social sciences, combined to the shift in research focus towards relatively "new" issues such as poverty alleviation, natural resource management, etc., raises serious concern as to the capacity of NARSs to adequately implement the sub-regional (regional) research agenda. Moreover, it was pointed out that many NARSs of the region are in need to improve their capacity to undertake research priority setting at the national level. This lack of capacity points to: i) the need to enhance training in such disciplines; and ii) the importance of involving "non-traditional" stakeholders such as universities, NGOs, and the private sector, which may have a clear comparative advantage with respect to such disciplines. Probably as a reflection of participants' background and interests, fisheries and forestry were not adequately addressed over the course of process.

5.5 Relevance of survey-derived priorities

By design the results of the questionnaire are expected to reflect priorities as perceived by individuals (the questionnaire recipients) and not necessarily research priorities as identified at the national level (Ministry of Agriculture, for example)(20). Reconciling individual priorities, as derived from survey analysis, and national priorities, as presented at sub-regional and regional meetings by NARSs' representatives, proved to be an arduous exercise. For one thing, individual priorities are necessarily biased as they usually reflect the recipient's background. For example, a crop breeder will tend to focus on crop breeding while a livestock scientist will focus on livestock, etc. In the course of the survey, some have suggested that instead of filling the questionnaire on individual basis, it would be more appropriate to fill it on a collective basis at the NARS level, for example. However, doing so runs the risk of limiting participation to NARIs' scientists and thus would not ensure participation of "non-traditional" stakeholders, which was precisely one of the objectives pursued through the survey. Moreover, farmers and other recipients not familiar with NARS/CGIAR collaboration, i.e., "non-traditional stakeholders" have received little assistance, either from NARIs or from the SROs, in filling the questionnaire, when they did. For one thing, the reported survey results (Annex 2) provide further confirmation of the difficulty in reaching a regional consensus on agricultural research priorities in view of the heterogeneity of the CWANA region.

(14) However, this by no means implies less emphasis on commodity research, which remains of vital importance in the region.

(15) Borrowed from Cosgrove, W.J. and F.R. Rijsberman, Editors. 1999. World Water Vision: Report. Worlds Water Commission's Staff Report. Version of 17 December 1999.

(16) Although important, these limitations do not, however, imply that the process will be of no contribution. It in fact represents a milestone in effective regional collaboration and cooperation. Moreover, using Ghandi's talisman (Box 2), it could be considered as a major contribution to the region in its quest to alleviate poverty and improve food security.

(17) Borrowed from Speth, J.G. 1993. Towards Sustainable Food Security. CGIAR. Sir John Crawford Memorial Lectures. International Centers Week. October 25, 1993. Washington, D.C.

(18) In its simplest form the total score of each research alternative could be computed as follows: Yi = S(wj * yij)
where Yi is the total score of research alternative i; wj the weight of criterion j (the sum of the weights is 1); yij the score of research alternative i with respect to criterion j; The global score can be further weighted by the probability of success (feasibility) of research alternative i (Ps), the probability of adoption of research alternative i (Pa) and the expected cost of research alternative i (Ci), usually approximated by researcher time. The equation then becomes:
Yi = [S(wj * yij) * Ps * Pa]/Ci

(19) Table A13 (Annex 2) shows the range of weights assigned by some NARSs of the region to the suggested criteria.

(20) Not to mention sample bias and its likely implications on survey results.

VI. PERCEIVED RISKS AND THREATS

In addition to the above limitations, a number of risks and threats are associated with this CWANA research priority setting process. These risks and threats - which could significantly affect the implementation and expected outputs of the regional approach as it is being developed in CWANA - arise from the complexity of the task at hand resulting from the interplay of several factors and circumstances, key of which are outlined below:

• § a regional approach to research involves various stakeholders very often with different backgrounds and interest. Consensus is therefore not straightforward, not to mention the lack of experience relative to a bottom-up and multi-stakeholder dialogue. A regional approach to research is about expanding collaboration and dialogue by forging strong and sustainable links and strategic partnerships with various traditional as well as "non-traditional" stakeholders, such as NGOs, the private sector, etc. It is also about empowering farmers' and grassroots organizations. Clearly, this abrupt paradigm shift in collaboration is a relatively novel experience in the region and will thus require time to be fully consolidated and institutionalized;
• by definition a regional approach to research is information (data) intensive. Its effectiveness is therefore strongly dependent on the availability and easy access to information by those in charge of the implementation of the regional research agenda. By and large, information available in the region is highly imperfect and is available only locally, i.e., it is not systematically made available electronically for large access through the internet, for example. It is therefore necessary to establish partnerships with local stakeholders to mobilize and tap this information when it exists. The lack of readily available information and relevant databases, often exacerbated by very frustrating bureaucratic procedures still in use in many countries of the region, may greatly affect the outcomes of the approach as well as the sustainability of the whole process(21);
• to ensure the effectiveness and sustainability of the regional process underway and of its expected products, ownership must be fully shared by all the concerned actors, i.e., NARSs, SROs, CGIAR and other international/regional centers, farmers, private sector, etc. Ensuring equitable ownership and credit within a multi-stakeholder framework is not an easy task. Given serious capacity discrepancies among the different stakeholders involved, there exists a serious risk of domination by the better endowed and/or better experienced partners. In such a case, "weak" partners may loose interest and confidence, hence affecting the process as a whole. In other words, unless true ownership of activities is ensured, participation of stakeholders, especially "non-traditional" stakeholders, will likely tend to be perfunctory, at best. Therefore, the issue of ownership, very often overlooked in sub-regional and regional debates, may prove a serious stumbling block to the process of regionalization of research;
• regional leadership is another important issues that, so far, has been totally ignored throughout the process. From the outset, AARINENA (for WANA) and the CAC-NARS Forum (for CAC) were supposed to lead the process throughout. However, as the process got underway, it became clear that the two organizations have yet to acquire a status of "a recognized regional authority" in their respective mandate region. To ensure regional ownership of the process, it is strongly suggested that CGIAR centers, e.g., ICARDA, play a catalytic role rather than assuming the lead of the priority setting exercise. Although there are many circumstances that may force some CGIAR centers, e.g., ICARDA, to take the lead of the process, such a strategy carries a very high risk and would likely lead to a lack of commitment on the part of regional partners, e.g., SROs and NARSs, which would fully rely on the CGIAR to ensure the success of the process with little or no input of their own. The sustainability of the whole process rests definitely on regional commitment. If the latter is lacking, a situation of "free-riding", i.e., benefiting without contributing, will most likely prevail;
• often than not, there is a tendency to express agricultural research priorities in terms of broad "development problems" and challenges rather than in terms of focused "researchable issues" that could contribute to solving key development problems. Obviously, agricultural research is a key component of the research-development continuum and hence can play a key role in achieving development objectives. It should be pointed out, however, that in addition to agricultural research, other key factors, e.g., infrastructure, education, political and social factors, etc., also play an important role in solving (or worsening) development problems. Therefore, as the implementation and effectiveness of the regional research agenda greatly depends on the degree of focus of the identified research priority areas, it is crucial to differentiate between the latter and the definition of development problems, no matter how important they are(22). The contribution of agricultural research is to improve the well-being of those who depend the most on agriculture for their livelihoods. This must be the ultimate aim of the regional research agenda;
• in addition to the above risks and threats it is worth highlighting some key elements of the environment within which the process of regional research and integration is to be implemented and which could significantly affect its expected outcomes. Among such key elements the following stand out:
  
  o Rampant degradation of natural resources, especially a sustained decline of supply and quality of water;
  o Alarming genetic erosion and loss of biodiversity;
  o Global warming and climate change;
  o Declining crop productivity due to recurrent droughts and soil nutrient depletion;
  o Relatively high cost of agricultural inputs.
  o Difficulty in harnessing globalization and new research tools such as biotechnology;
  o Insufficient national investment in agricultural research in spite of the very important, and acknowledged, role of agriculture in the overall economy;
  o Lack of national commitment to regional research exacerbated by donors' fatigue;
  o Insufficient efforts deployed for the institutionalization of a true bottom-up and participatory approach to research;
  o Very few attempts to genuinely empower farmers, especially the resource-poor farmers and women

(21) In some countries scientists may have to go through lengthy and frustrating procedures before they can obtain administrative permission, when they do, to undertake socioeconomic surveys, for example.

(22) "Research initiatives are defined by the problems they seek to solve, not the product they hope to identify" (Collinson, M.P., and E. Tollens. 1994. The Impact of the International Agricultural Research Centers: Measurement, Quantification, and Interpretation. Issues in Agriculture, No. 6. Consultative Group on International Agricultural Research).

VII. MEETING THE CHALLENGE

In view of the complexity of the challenges facing the region, it is unlikely that the NARSs or other regional institutions would be able to satisfactorily address them on their own. This in turn highlights the urgency of establishing strategic partnerships to tackle the agricultural problems of the region. Within the framework of the CGIAR plank 4, ICARDA, in collaboration with the two SROs (AARINENA and the CAC-NARS Forum), has initiated and facilitated the piloting of an experimental regional bottom-up priority setting approach for the CWANA region, as outlined in this report. It must be clear, however, that the identification of regional priorities is just half the battle. The other half will consist in translating such priorities into concrete research proposals with the appropriate funding strategies. The proposals to be developed should seek to link the research initiatives to the development goals of poverty alleviation and food security improvement. The objective would consist in generating win-win options that would contribute to economic growth while preserving the natural resource base.

It is worth pointing out that the CGIAR's mandate is about producing international public goods for which it has a comparative advantage. Obviously, the international public goods produced by the CGIAR benefit most CWANA countries. NARSs of the region, on the other hand, have a more global mandate that goes well beyond the CGIAR mandate. Therefore, the next step is for the CGIAR to identify a niche within the regional research agenda, i.e., identify research opportunities that are common to both the NARSs and the CGIAR (Figure 2)(23).

Confronted to the mounting problem of poverty and food insecurity fueled by persisting trends of low crop productivity, degradation of the natural resource base, declining rates of investment in agricultural research, and inappropriate agricultural policies, NARSs of the region can only benefit from the implementation of the regional approach being developed. Clearly, the faith of this process will greatly depend on commitment, capacity, and good governance of the key actors involved, especially the SROs.

Through discussions and interactions, the priority setting exercise provided NARIs with the necessary information and tools for re-visiting/updating, in collaboration with (relevant) "non-traditional" national stakeholders, agricultural research priorities first at the sub-regional level before moving to the regional level. In the process, lessons were learned. These must now be internalized to make the necessary corrections and ensure a better implementation of the derived regional research agenda. Moreover, it should be pointed out that the effective (and efficient) implementation of the derived regional research agenda calls for new forms of partnerships and new roles and responsibilities of the key actors involved. Within the new partnership framework, the NARSs will undoubtedly have to play a key role. This points to the need for follow up if the upbeat spirit that characterized the process throughout is to be maintained and, why not, re-enforced. In the meantime, however, the leading role of the two concerned SROs cannot be overstated.

In view of the novelty of this regional process, relying on a bottom-up approach and a multi-stakeholder dialogue, the implementation of the derived research priorities will necessarily require time, resources, and probably further consultations at the level of individual NARSs as well as at the sub-regional level. To foster the process and ensure adequate implementation of the derived recommendations, a regional task force needs to be established under the leadership of the two SROs and participation of ICARDA (as recommended by the regional consultation). It must be clear, however, that although it is important that ownership of the process remains with the NARSs and the SROs, CG centers, ICARDA in particular, will have to play, at least in the short-to-mid term horizon, a true catalytic role to ensure the appropriate and timely implementation of the agreed upon recommendations.

It is never too early to start exploring avenues to secure the necessary human and other resources (financial, equipment, etc.) for the implementation of the agreed upon regional research agenda. Therefore, in collaboration with the two SROs and GFAR, the task force should also address the issue related to securing the necessary funding in order to implement the agreed upon research priorities with the understanding that NARSs' contribution has to be substantial in view of donors' fatigue. Finally, in order to ensure the sustainability of the regional process, it is crucial that the major threats outlined above, in particular lack of institutional commitment, lack of human capacity, and lack of funding opportunities must be clearly spelled out at the outset and dealt with accordingly.

Figure 2. NARSs/CGIAR research congruence