2.1 Evolution of the CGIAR Resource Allocation Process
2.2. Assumptions on 1998 Core Funding for the CGIAR System
2.3. TAC's Recommendations on CGIAR Priorities
2.4. Linking CGIAR Priorities to Resource Allocation
2.5. Preparation and Analysis of Centres' Medium-Term Plans
2.6. Initial Review of the MTP Proposals
Initially, centres' programmes and budgets were prepared annually, complemented with multi-year projections of requirements, and reviewed by TAC and endorsed by the Group. The programmes and budgets were based on centres' long-term plans. Starting in 1987, centres prepared five-year, medium-term plans, based both on CGIAR priorities as proposed by TAC and endorsed by the Group, and on their own long-term strategic plans. At ICW'90, the Group reviewed a report ('Review of the Resource Allocation Process') which examined the experiences gained from the five-year allocation process. The Group endorsed the recommendations which, among others, called for a more transparent linkage between System priorities and centres' operational programmes, and the consideration of constrained resource supply in a so far largely demand-driven resource allocation process. The completion by TAC in 1992 of the Review of CGIAR Priorities and Strategies, including its implications for resource allocation, offered an opportunity to further improve the linkage between System priorities and resource allocation to the centres and programmes in the framework of the development of new MTPs by centres.
First Round of MTPs (1987-89). Using TAC's 1986 priorities paper as a reference, the development of centre MTPs became the mechanism for translating the System's priorities into operational programmes. Since the development of centre MTPs was, from a core resource point of view, open-ended, centres were able to present the full potential of their core programmes and activities in an environment of unconstrained supply of core funding. However, the initial round of MTPs had two major limitations. Firstly, the review and approval of centre MTPs was phased over a three-year period, causing the MTP time horizon of individual centres to vary from 1988-1992 to 1990-94. This reduced the scope for a comprehensive monitoring of the achievement of System priorities resulting from the implementation of individual centre MTPs. Secondly, the sum of centres' core resource requirements was substantially in excess of actual core funding during the implementation of the MTPs. Consequently, each year, centre core requirements needed to be adjusted downward mechanically, which generated increasing discrepancies between approved MTP programmes and centres' actual operational programme levels.
Second Round of MTPs (1992-93). In the framework of the current round of MTPs, the resource-allocation process has been modified so that it:
· provides a framework which explicitly links resource allocation decisions with recommendations on CGIAR priorities and strategies;· introduces ex ante a resource supply constraint on core programmes, consistent with reasonable projections of core resource supply:
· clearly defines the centre programme building blocks in terms which allow easy identification of their relation to the System priorities;
· improves the justification and review of centre proposals, thus enhancing transparency and equitable treatment of all centres' proposals; and
· provides, through consolidation of all centres' MTPs covering a common five-year period (1994-98), a System five-year plan - allowing global monitoring of the implementation of System priorities within realistic resource boundaries.
The specific objectives of the re-design of the resource-allocation process were:
· to improve the matching, in programme terms, of the System ('top down') priority setting by TAC and the CGIAR, and the ('bottom up') preparation by individual centres of MTP proposals; and· to improve the matching, in financial terms, of approved allocations of core resources to centres with the supply of core funds by donors.
The medium-term resource-allocation process is underpinned by the following principles:
· transparency, i.e., the rationale of the centre allocation recommendations can be related to the recommendations on System priorities as recommended by TAC and endorsed by the Group;· supply constraint, i.e., centres were asked to prepare MTP proposals on the basis of the indicative core resource envelopes recommended by TAC; however, centres had a reasonable margin of flexibility since they were also requested to present alternative proposals at 10% above and below the level of the core resource envelope. Centres could also present supplementary proposals in excess of this margin,
· interactive, i.e., starting from a centralized indicative planning allocation endorsed by the Group, centres constructed their MTP proposals, and presented these to both TAC and the CGIAR; subsequently, TAC reconciled the MTP proposals both in substance - i.e., the convergence of the sum total of centres' medium-term programme proposals with the System priorities - and financially -i.e., the compatibility between the sum total of centres' proposed funding requirements and the core funding assumption used for the System as a whole.
To guide the MTP process, the Group had indicated at ICW'91 that prospects for growth in core funding for the next few years were extremely limited, and any growth was likely to be limited to the CGIAR's new initiatives, mainly in the forestry-related activities. Subsequently, at MTM'92, the CGIAR endorsed a financial planning assumption of US$ 270 million (in 1992 dollars) for 1998 which projected no growth, in real terms, in core resources from the 1992 estimated level of US$ 255 million except for an increment of US$ 15 million for forestry and fisheries: this excluded the possibility of reallocation of existing resources towards these new activities. Since then, the global development assistance environment has worsened significantly, at least in the short term. The level of CGIAR core resources actually provided in 1992 was US$ 8 million lower than the projection at MTM'92. By TAC 61, 1993 funding was projected to contract further to US$ 231 million (US$ 221 million in 1992 dollars), which compared to the original planning assumption (which assumed no growth) now implies a 15% real growth in core resources by 1998, in addition to the US$ 15 million increment for forestry and fisheries.
The constraint on funding currently being experienced raised the issue of whether the planning assumptions should be revised to reflect the lower funding level prevailing in 1993. However, centre medium-term planning had proceeded on the assumption of US$ 270 million core funding for the System in 1998. Reducing the medium-term planning levels would have required the centres to revise their MTP proposals, which would have delayed the reconciliation between centre plans and System requirements well beyond ICW'93. Furthermore, it is unclear whether the present low levels of core funding will persist until 1998. It was therefore proposed and accepted at the May 1993 MTP Workshop (see Section 2.4) that TAC proceed with recommending the allocation of 1998 core resources at the level of US$ 270 million (in 1992 dollars). It was also agreed that TAC would outline a contingency programme for allocating core resources at a funding level of US$ 240 million, or about 90% of the original planning level for 1998. It was understood that TAC's recommendations to that effect could go well beyond proposals for across-the-board reductions of all centres' programmes and requirements, or a proposal to simply postpone the initiation of new activities. At the MTP Workshop, it was made clear that TAC's proposals should be based on considerations of structural adjustments of the CGIAR through the restructuring, streamlining or phasing our of entire programmes at the level of subject matter, regions, commodities or institutions, based on criteria such as relative priority ranking of commodities or regions, management capability, institutional delivery capacity, potential alternative sources of supply, etc.
Consequently, TAC formulated recommendations for the allocation of core resources for 1998 at levels of US$ 270 and 280 million fin 1992 values), from which it derived recommendations for core funding in 1994 and each intermediary year of the MTP period, for the System as a whole as well as for each centre. TAC also presented a strategy, process and timetable on how to adjust the System to enable it to cope with funding at a US$ 240 million level (in 1992 values) in 1998 while assuring the integrity of the System.
During 1992, TAC finalized its review of CGIAR priorities and presented its report to the Group. Compared to TAC's previous reviews, the approach had been modified to take into account the current goals of the CGIAR and its expanded mandate, emerging trends in world agriculture, including the greater emphasis on sustainability and resource management issues, and the evolution of scientific capacity in developing countries. TAC also strove to achieve meaningful interactions with stakeholders, transparency in decision making, the development of mechanisms which would facilitate CGIAR priority setting as a continuing activity, and enable the implementation of agreed CGIAR priorities to be monitored.
In assessing priorities and strategies for the System, TAC used an analytical framework with three dimensions: an activity dimension, including the five categories of CGIAR research and research-related activities; a spatial dimension with nine agroecological zones, and four geographic regions; and a product dimension, with four main production sectors and their respective commodities. Quantitative tools, such as congruence, scoring and economic surplus models were used in addition to informed judgement to assist in the formulation of recommendations.
The main outcome, presented in Chapter 12 of the report, was TAC's recommendations on CGIAR priorities in Systemwide terms by: (a) category of activity; (b) region; (c) agroecological zone; (d) commodity; and (e) production sector.
Overall, these recommendations represented the following changes in the focus of CGIAR activities compared to 1991, expressed in terms of relative core resource allocations:
By category of activity· a substantial increase in priority for research on the conservation and management of natural resources, including germplasm conservation (from 13 to 18%); and for socioeconomic public policy and public management research (from 9 to 11%);· a slight increase in priority for research on germplasm enhancement and breeding (from 21 to 22%);
· a reduction in the priority assigned to research on the development and management of production systems (from 33 to 29%), and on institution building (from 24 to 20%);
By region
· an increase in priority for Asia (from 29 to 33%) and Latin America and Caribbean (LAC) (from 15 to 17%), and a reduction in priority for sub-Saharan Africa (from 43 to 39%) and that of West Asia and North Africa (WANA) (from 13 to 11%);By commodity
· an increase in priority of groundnut and soybean, and a reduction in priority of phaseolus bean and pigeonpea. The priority of cereals and root and tuber crops, other food legumes, and banana and plantain is to be maintained. The overall priority of livestock should also be maintained but TAC considered that the CGIAR was over-investing in livestock research in sub-Saharan Africa;By ecoregion
· TAC identified six priority ecoregions: (i) the warm arid and semi-arid tropics, and (ii) the warm humid and sub-humid tropics in sub-Saharan Africa; (iii) the cool subtropics with winter rainfall in WANA; (iv) the warm arid and semi-arid tropics and subtropics and (v) the warm humid and sub-humid tropics and subtropics in Asia; and (vi) the warm humid and sub-humid tropics and subtropics in LAC.
TAC's analysis of CGIAR priorities was largely based on 'needs' considerations, which were to be complemented subsequently, through the MTP process, by centres' proposals on how these needs could be addressed effectively, taking into account the role of other non-CGIAR institutions. For this purpose, TAC assigned each centre a tentative, priority-based resource envelope as the basis for the development of each centre's MTP proposal. The methodology for developing the resource envelopes and the outcome of this part of the process are described in Chapter 14 of the Review of Priorities and Strategies. In recommending the tentative distribution of resources, TAC attempted to ensure consistency with its recommendations on CGIAR priorities. This was achieved:
· at the System level, by allocating 1998 core resources in relative and actual dollar terms for a total of US$ 270 million (in 1992 values) among the five categories of activity and 14 sub-activities, as well as among the four geographical regions;· at the centre level,
- first, by considering the impact of the priority recommendations on the distribution of core resources by centres, particularly in terms of directional changes in the distribution among centres and within centres, among activities and commodities;- subsequently, the System resource allocations and the directional changes in centre allocations were translated into centre-specific dollar amounts (expressed in 1992 values); the latter representing the 1998 indicative resource envelope for each centre, which constituted the starting point for the preparation of centre 1994-1998 MTP proposals.
Finally, in order to assure a certain degree of flexibility in the final allocations, out of the total US$ 270 million TAC allocated only US$ 255 million to centres' indicative resource envelopes; this provided TAC with an allocatable reserve of US$ 15 million (in 1992 values), to which another US$ 5 million was originally added on account of a livestock reserve fund. This fund of US$ 5 million was composed of US$ 3 million from ILCA and US$ 2 million from ILRAD in view of the uncertainty at that time surrounding the future evolution and direction of CGIAR livestock research. At TAC 59, TAC presented its preliminary views on CGIAR priorities for livestock research which concluded that support for livestock research should not be reduced from current levels. TAC subsequently returned these funds to ILCA and ILRAD for planning purposes to supplement their resource envelopes.
In order to allow appropriate flexibility in the development of MTP proposals, centres were urged to develop alternative proposals at levels of core funding above and below their planning envelopes.
Once TAC reached consensus on CGIAR priorities and strategies and had assigned the centres' tentative resource envelopes, guidelines were issued to assist centres in the preparation of their MTP proposals.2 The centres were invited to present their MTP proposals at the level of their resource envelope and at 10% above and below that range, and explain the programmatic impact of funding at each level. Centres could also present proposals that implied core requirements beyond the 10% range.
2 CGIAR Secretariat, 1992. Guidelines for the CGIAR Medium-Term Resource Allocation Process 1994-98. World Bank, Washington. D.C.
Prior to the formal presentation and discussion of the MTP proposal by TAC and the CGIAR, a working party, consisting of a TAC member and staff of the TAC and CGIAR Secretariats, visited each centre so as to familiarize themselves with the proposals. Subsequently, each centre presented its MTP proposal to TAC first and then to the CGIAR: seven at TAC 59 and ICW'92, and ten at TAC 60 and MTM'93; CIFOR's MTP proposal was presented at MTM'93 and at TAC 61. In each case, TAC, in collaboration with the CGIAR Secretariat, prepared an interim commentary on centres' proposed plans raising programmatic and financial issues and, in most cases, requesting additional information from centres. These commentaries were shared with members of the CGIAR prior to the presentation and discussion of the MTPs by the Group. Centres then had the opportunity to respond to the TAC interim commentary and to questions raised by the CGIAR before all of the MTPs were considered again, individually and collectively at TAC 61 in June 1993. It was understood that no final recommendations by TAC, nor decisions by the CGIAR would be made before all centres' MTPs had been considered individually by both TAC and the Group: this would ensure that TAC would be able to formulate consistent recommendations for all centres, and reconcile its centre -and System-level recommendations.
A summary of each MTP proposal, TAC's interim commentary and programmatic issues, the centre's response, and TAC's evaluation, assessment and final funding recommendations is given for each centre in Chapter 4 of this report.
Immediately after MTM'93 and before TAC 61, the progress made by centres and TAC in the MTP process was presented and discussed at an MTP Workshop held in San Juan. Puerto Rico. The Workshop was attended by representatives of most of the CGIAR members. Centre Directors and senior staff, the TAC Chair and some TAC members, and staff of the two Secretariats.
At the Workshop, a progress report on the MTP process - jointly prepared by TAC, and the TAC and CGIAR Secretariats - was presented and discussed.3 It presented an initial assessment of the experiences gained. The report also presented a preliminary analysis of the implications of aggregating all centre proposals (at the level of their base resource envelopes) and compared the outcome with TAC's recommendations on CGIAR priorities and strategies. A first assessment was made of the degree of convergence of centres' collective proposals with the CGIAR priorities by category of activity, region, production sector and commodity. It was concluded that, despite large discrepancies at the individual centre level, in the aggregate the congruence was high: the directional changes were consistent in all cases, though the pace of implementation of the changes clearly accelerated under the centres' collective proposals. The report also expanded on how TAC intended to complete the resource allocation process and presented the format in which the recommendations to the Group would be formulated.
3 CGIAR Secretariat, 1993. Progress Report on the CGIAR Medium-Term Resource Allocation Process. Paper presented at the 'Workshop on the Medium-Term Plans'. San Juan, Puerto Rico, 30 May 1993.
2.6.1. Overview of Proposals
2.6.2. Considerations with Respect to CGIAR Priorities
2.6.3. Institutional Considerations
2.6.4. Systemwide Considerations
In the aggregate, centres' MTP proposals submitted to TAC and the Group, up to MTM'93, called for core funding which exceeded the planning estimate of 1998 core funding of US$ 270 million by some 20%. Centres' complementary funding projections called for considerable increases also. In several cases, centres asserted that the constraint imposed by the resource envelope they had been assigned made planning quite irrelevant by limiting centres' ability to demonstrate to the full extent their capacity to develop more activities and absorb more resources effectively. Also centres argued that core funding should not necessarily stagnate at levels currently provided to the CGIAR, For these reasons, most centres presented a primary MTP proposal at a level equivalent to (or in excess of) 110% of the base resource envelope. A number of centres also noted that providing alternative scenarios at lower levels of core funding was not sensible in view of the high demands for their services, or inappropriate because of unwanted negative connotation such proposals may have within and outside the institutions.
However, subsequent to the interactions between centres, TAC and the Group, by TAC 61 all centres had provided adequate information on the programme and institutional implications of alternative core funding levels. This avoided TAC having to make judgements about centres' internal priorities and how these had to be re-arranged under different funding assumptions, and which is properly the prerogative of the centres themselves.
From the review of centres' proposals, a number of general observations could be made:
(a) There was a wide variation in the methodology, transparency and subjectivity of priority-setting analysis by the centres. The approach to internal priority setting ranged from being highly quantitative to highly qualitative. There was also a wide variation in the conciseness of documentation. This observation does not suggest that there was a "correct way" for centres to prepare their proposals, it rather points out the rich diversity of approaches that were encountered:(b) Centres were responding to the call for greater collaboration with other CGIAR centres as well as with other partners to the extent of giving the impression that each centre is proposing collaboration with 'everyone else'. Caution had to be expressed against proliferation of bilateral collaborative efforts leading to the dilution of many, currently-positive examples of collaboration, or the development of unsustainable institutional relations, outside a comprehensive framework:
(c) As indicated above, despite having been encouraged 10 limit themselves to constrained resources in their MTP proposals, as indicated by donors, centres still emphasized the need to increase their assigned level of resources. Only six centres presented their primary MTP proposals at the base resource level. On TAC's insistence, however, all centres provided information - in many cases as addenda to their MTP document - on the programmatic implications of core funding at lower levels;
(d) During the last three years, the System has gone through a process of rapid downscaling; this is reflected in the fact that several centres are currently operating at levels substantially below 1990-91 funding;
(e) When aggregated, centre proposals appeared to be broadly in line with TAC's recommendations on CGIAR priorities by region, by category of activity and by commodity. While differences were to be expected because of the very nature of the process at this stage (i.e., bottom up), it was not always clear at the time whether the proposed changes in centre's priorities reflected a true shift in emphasis or resulted from a re-categorization of ongoing activities. Table 3.3 on page 24 shows the distribution by category of activity, activity, region, production sector and commodity as this resulted from aggregating all centres' core proposals at the base level and puts it in context of prior distributions and recommendations.
TAC's comprehensive review of all the centres' MTP proposals occurred at TAC 61 in June 1993. The first step in the process was to consider centres' responses to issues raised by TAC and the CGIAR - in October 1992 and May 1993 respectively - and to assess whether developments within and outside the CGIAR System since early 1992 (when TAC's original proposals were endorsed) necessitated adjustments to the centres' indicative resource envelopes. Among these developments were: (a) the changes brought about by UNCED which had stressed the need for much greater consideration of sustainability and environmental issues; (b) a number of external reviews of centres.4 (c) sectoral reviews of rice, livestock, banana and plantain research; (d) the joint TAC/Centre Directors' review of the ecoregional approach to research; and (e) the decisions taken by the CGIAR related to these issues, especially those taken at MTM'93 which had major structural implications. TAC also took into account the centres' own estimates of the distribution of efforts by activity categories, sub-activities, regions and commodities. As it is to be expected from an iterative process (i.e., a top down followed by bottom up approach), in many cases the centres' estimates were quite different from the distribution used by TAC in its March 1992 deliberations on resource allocation. Sometimes, different interpretations in the classification of activities and categories of activities have caused such differences. In other cases, differences were caused by evolving interpretations in the classification of centres' operations in TAC's new classification of activities by category. A number of centres had challenged TAC's estimates and considered therefore that their indicative resource envelopes had been negatively affected by inaccurate estimates. The Committee carefully considered the centres' arguments in each case.
4 CIMMYT, ICARDA, ICLARM, ICRAF, IFPRI, ILCA, ILRAD, INIBAP, IRRI and WARDA
Following a collective discussion of the priority issues as these emerged from the considerations described above, at TAC 61 each of the TAC members was asked to indicate by ballot whether centres' resource envelopes should qualify, on priority considerations alone, for an upward or downward adjustment of, or for the maintenance at, the base resource level as indicatively set in March 1992. This evaluation was based only on priority considerations which, in line with the priority and strategy exercise, were essentially demand driven and of a qualitative nature. Therefore TAC members were asked to omit institutional considerations at this stage. The objective was to complete TAC's discussion on the CGIAR priorities and on how these were to be translated into centre resource envelopes.
After TAC members had indicated their assessment by ballot, the ballots were compiled and the outcome was shared among all TAC members. TAC members were invited to share their vote and explain the reasoning behind it. This exchange of decision-making information fostered an in-depth discussion involving all TAC members and led gradually to a consensus on the priority considerations for each centre. The outcome of this discussion was that five centres tentatively qualified for an upward adjustment of their base resource envelope, ten centres for maintenance and three centres for a downward adjustment of their envelopes as an expression of TAC's modified priorities.
Once consensus had been reached on the priority considerations. TAC members were asked to evaluate each centre. This step was introduced by an examination by all TAC members of centres' MTP proposals and their quality, the adequacy of centres response to TAC commentaries and CGIAR questions, recent centre performance, outstanding programme or management issues, centre financial performance and condition, etc. Following such a detailed discussion of each centre, TAC members were asked to evaluate each centre according to a common set of largely supply-oriented criteria. TAC discussed and agreed on the following, equally-weighted criteria, which were considered to be relevant and pertinent to the assessment of all centres:
· the strategic character of the centre's research programme, i.e. whether the centre had developed a coherent, forward-looking programme to address critical future strategic issues;· the potential for breakthroughs, i.e. an assessment of the chances that research would be successful and that the results would be converted to usable technology which would be widely adopted;
· past performance and likelihood of continuance (or improvement), i.e. whether the centre had delivered results and products in the past and whether there were firm indications that it could do so in the future;
· the external environment, institutional health and quality of management, i.e. whether the location of the centre and its relationships with host country/countries suggested a conducive environment for successful research; and whether the institution was appropriately endowed with human and physical resources and had a management team that would lead it towards achieving important results;
· collaboration with NARS, other CGIAR centres and advanced institutions, i.e. whether the centre was an active partner with NARS and had an appropriate set of links with other institutions to maximize the effectiveness of its resources.
Once the criteria had been discussed and agreed upon, each TAC member was asked 10 assign a score of between 0 and 20 for each criterion and each centre. The total score for each centre per TAC member could not exceed 100 points. The results of this second ballot were collated and shared among all TAC members. Information on centres' total and average scores, their comparison with System averages, centres' mean (excluding the highest and the lowest score), standard deviation and median, all led to a relative ranking of centres which reflected TAC members' evaluation on the basis of the criteria described above. This second input provided TAC with a basis to progress in the resource allocation process by reconciling the outcome of the priority considerations (first step and ballot) with institutional considerations (second step and ballot). A relatively high degree of convergence between the two evaluations generally emerged. However, in some cases, there was a divergence indicating that TAC members considered that the subject matter areas were of high priority, but that there were concerns about the centres' present capacity to deliver an effective research outcome.
Finally, TAC recognized the limitations of the MTPs having been developed at the centre level only. On the one hand, the centre-specific proposals did not, and could not deal comprehensively with matters of interest to more than one centre, and on the other, tended to ignore matters which transcend centres' own interests but are of interest to the System as a whole. With regard to the former, TAC considered it necessary to promote effective inter-centre collaboration as well as collaboration with national programmes and other institutions. With regards to the latter, TAC decided to foster the concept of Systemwide programme initiatives. In so doing, TAC referred to recent CGIAR decisions on global issues such as livestock and plant genetic resources, and to other CGIAR undertakings such as that on ecoregional research, and TAC's own working groups on specific subject matters.
