Agricultural Resources Management Project - Interim evaluation (2004) - IOE
Agricultural Resources Management Project - Interim evaluation (2004)
Interim Evaluation 1
The Agricultural Resources Management Project (ARMP) was designed in late 1995, and was the fourth IFAD supported intervention in the Hashemite Kingdom of Jordan. The project closed at the end of December 2003, and based on the successful implementation experience the Government of Jordan (GOJ) has requested a second phase of the project. The project was subject to a Mid-Term Evaluation (MTE) in 2000, hence the approach adopted for the Interim Evaluation (IE) was therefore to update the assessment of impacts and the changes that had occurred as a result of the MTE recommendations. In undertaking the evaluation, the field mission fully implemented the Methodological Framework for Project Evaluation (MFE) developed by the IFAD Office of Evaluation (OE) in 2002 and followed its rating system2 .
The overall goal of ARMP, within the GOJ strategies for the development of rainfed farming, was to improve the income stability of vulnerable, resource poor farmers in the two Governorates of Karak and Tafila by safeguarding and improving the productive potential of their natural resources and enhancing their returns to labour. This was to be achieved by arresting soil degradation and restoring soil fertility, promoting activities for women through credit and strengthening the capacity of the Ministry of Agriculture (MOA) both to implement these measures and to meet the technology needs of farmers. To realise these aims t he project interventions were arranged in five components: Resource Management; Agricultural Development; Institutional Strengthening; Women's Development; and Co-ordination and Management. At the time of design the estimated number of rural households living below the poverty line in the project area was estimated at about 7 000. The total project cost was estimated at about USD 18.6 million, of which the IFAD loan provided about USD 12.8 million. The project had a seven year implementation period.
In 2000, after three years of implementation, the project management proposed to revise the targets in the resource management and agricultural development components. These proposals were endorsed by the MTE, and the re-allocations were approved in 2001. The revised targets did not change the nature of the project itself (and hence the objectives), but simply increased and re-balanced the targets on a demand-driven basis. In addition, significant changes were made to the categories for which credit was available: funds were switched from support for supplementary on-farm soil and water conservation (SWC) activities and land consolidation and moved to women's Income Generating Activities (IGAs), where demand was proving to be very high.
Resource management. At the time of the interim evaluation 81% of the revised target for soil and water conservation activities (7 500 ha) had been achieved, covering 2 700 farms. On-farm, the project had constructed 2 871 cisterns of 30 m 3 capacity to conserve runoff and supply supplementary irrigation needs. The cisterns have contributed very significantly to the expansion of the area under tree crops, especially olives, and have greatly improved the productivity of the crop. As well as other erosion control measures, such as gabions and terraces, the project has constructed almost 900 km of stonewalls to protect some 5 880 ha. Stone walls have proved the most popular activity with beneficiary farmers probably because an unintended benefit has been significant increases in land values. Off-farm, 29 rock-faced earth mini-dams and hafiras3 were completed between 1998 and 2003 (58% of the revised target), benefiting some 18 population centres, mainly in respect of livestock watering. The project has undertaken a programme of spring rehabilitation, which involves permanent protection of the source to prevent damage and contamination, and improvements to the canals which distribute the water downstream, benefiting numerous farmers who mostly have very small traditional olive groves. So far works have improved the utilisation of 69 springs (66% of the target), helping some 1 121 farmers.
Agricultural development. New planting has taken place so far on about half the area which has been improved through SWC activities. The overwhelming majority of trees planted are olives (140 000) although about 5 000 pistachio seedlings have also been released from MOA nurseries. The programme of fruit tree rehabilitation has consisted of deep-pruning of old olive trees that had become unproductive, and providing one application of fertilizer. Initially this activity was unpopular with farmers until the benefits were demonstrated, at which point demand greatly increased. Olive rehabilitation normally takes place where springs and canals have been rehabilitated.
Credit4 . Loans have been provided through the Agricultural Credit Corporation (ACC) either for agricultural development or for women's IGAs. At the time of the IE a total of 1 152 loans had been dispersed, of which 1 034 loans have been for women5 (260% of the initial target), hence, over 90% of the total loans, and 80% in terms of absolute value have been provided to women's. Approximately 44% of loans taken by women for IGAs financed dairy processing (with sheep milk), whilst other activities have mostly centred on other forms of food processing. Overall, the loans appear to have been very profitable (more so than at the time of the MTE): the mission calculated that average annual profits were JOD 829 (USD 1 160) or 57% of average annual household incomes. The result of stringent lending conditions has been a high repayment rate (89%). By June 2002, 31% of resources allocated for the institutional strengthening for ACC had been utilized, but ACC remains reluctant to use these loan funds, arguing that they should be provided as a grant. This has meant that ACC's outreach in the project area has not changed.
Women's Development Programme (WDP). Despite insufficient staff numbers or external support, the WDP has evolved into a much bigger programme than expected. Three NGOs and two universities have been contracted to provide training for the beneficiaries, and by project completion, 343 training workshops had been conducted (89% of target), reaching approximately 2 526 women. This is most impressive. However, the mission concluded, in agreement with the MTE findings, that the majority of WDP beneficiaries were not drawn from the poorest groups.
The WDP officers have found resistance to the concept of small enterprise groups, and none have been formed through ARMP. The mission concluded that the WDP staff neither had the appropriate skills nor did they have suitable training and support to introduce this concept.
Institutional strengthening. The project has arranged a very comprehensive training programme for staff, which has been very useful and will have long term benefits. However, there has been reluctance to use loan funds for more innovative aspects of the project, including hiring NGOs to provide support in specific skills, and to undertake studies. The mission considered that this had curtailed the extent of potential benefits and the development of new capacities and more participatory approaches.
Coordination and management. The IE considered that implementation arrangements for the project had generally been very satisfactory, with adequate coordination and management procedures. However, participatory approaches, aspects of community development and involvement and group formation have not been adopted, even after very strong recommendations made by the MTE and agreements reached in the ACP. This is unfortunate, as GOJ policies and MOA strategies now require exactly the sort of approaches described in the original design of ARMP.
Relevance. Overall, the mission considered that ARMP's SWC and agricultural objectives remained very relevant to the intended target groups. This conclusion was derived partly from considerations of the impacts on poverty presently being achieved, a summary appreciation of the poverty situation prevailing in Karak and Tafila, and GOJ's agricultural and rural development policies. If the credit activities had been implemented as designed, then their continuing relevance to the intended target groups would also be high. The success of the project has in many ways just highlighted the considerable potential which still exists in the two governorates. The relevance of objectives is rated as high (4).
Effectiveness. Overall, the mission concluded that the objectives of the project had been significantly achieved. Major improvements have been recorded in soil and water conservation, agriculture, and skills levels. The project completion report shows over 97% of physical targets achieved. By upgrading and protecting the natural resource base, through technical support to farmers and by providing credit and skill development programmes (especially for the women) the vulnerability of a section of the rural communities has been decreased. The re-allocation of targets and resources after the MTE improved the potential effectiveness of the project by better aligning targets and funds with beneficiary demands and project capacity. The mission concluded that the performance of the project in terms of its implementation effectiveness was high. The only detracting factor has been the lack of relative poverty of the groups that had been reached, which, it was further concluded, partly reflected the practical difficulties in targeting the interventions. The evaluation rated overall project effectiveness in achieving its objectives as substantial (3).
Efficiency. At appraisal, the incremental benefits were projected to be derived only from crop production, and the Economic Rate of Return (ERR) was estimated at 10.6%. The mission considered that this potentially underestimated the project benefits, as benefits are also derived from off-farm water storage structures, and the credit programme (through the promotion of IGAs). When all productive activities are combined, the Project's revised ERR was calculated by the mission to be 34.8%. This is a very good result, (which is not unduly sensitive to price/cost changes) and indicates that borrowing for ARMP development activities is well worthwhile and should lead to increasing economic returns for Jordan . This would also suggest that the project has been efficiently implemented. Improvements to physical resources, enhanced and rejuvenated agricultural production, and promotion of off-farm income generation have also proved positive for family incomes: the likely increases to farm household incomes indicated in the Appraisal (between 30% to 400%) are probably correct. The evaluation rated the efficiency of the project as high (4).
Physical and financial assets. Households either adopting SWC activities or benefiting from spring rehabilitation have seen their physical assets improved, probably best illustrated by the increased values now placed on their holdings. Impacts on household and financial assets have been reported by beneficiaries as a result of the IGA programme, and mission calculations indicate that these impacts are significant. Approximately 1 800 households have been given improved access to financial services through the credit programme. These improvements are impressive and are considered sustainable. Access to markets - an important factor in poverty reduction - has improved since the beginning of the project, as a result of beneficiary (and other) demand. However, the mission concluded that beneficiaries of the SWC, farm development and credit programmes were mostly in the middle-income bracket (in the context of rural incomes), therefore, although the project has had an impact on assets, it had less impact on poverty than it might have had, if it had been able to target the poorest sector of the rural population. Nevertheless, the overall rating for this sub- criterion is high (4).
Human assets. Access to potable water has improved either directly as irrigation facilities have been developed, or indirectly as a result of increased incomes. The project activities, by their nature, have resulted in increased workloads, both for women and children, as activity levels have been raised. All beneficiaries have either received skills training, or have profited from increased access to information via the project staff. Overall, the rating for this sub-criterion is substantial (3).
Social capital and empowerment. Although the design called for participatory approaches, in practice these have not been adopted. Because of the lack of participation, social cohesion and local self- help has not been an outcome of ARMP, neither has the project made any contributions to building local capacities for community empowerment, nor in fostering other self-help initiatives. Overall the rating for this sub-criterion is negligible (1).
Food security. The project has changed farming technology and practices by increasing farmers' use of fertilizer for tree crops, increasing the amount of irrigation water, more effective pruning, increased use of contour ploughing, and more rational cropping patterns for land of varying gradients. As a result land areas and productivity have increased, and the combined effect will be to improve the robustness and resilience of the farming systems, thereby improving food security. The overall rating for this sub-criteria is therefore high (4).
Environment and communal resource base. Changes in the natural resource base have largely been positive as water storage facilities constructed by the project have resulted in improved pasture, and the area of protected and stabilised soil on private farms has increased as a result of the SWC measures and the expansion of tree cropping. However, the project-sponsored cisterns for supplementary irrigation requirements do not adequately serve the new areas planted to olives and unintentionally the result is the depletion of what is already a severely depleted aquifer. On balance, however, the project has reduced short-term environmental risk and thereby contributed positively to livelihoods. The rating is substantial (3).
Institutions, policies and the regulatory framework. No significant changes in public institutions policies or service provision have resulted from the project's activities. The major change required is modernisation of the inadequate extension services, and ARMP has served to highlight this need. As a result of the credit experience ACC has been willing to modify its lending terms and conditions somewhat, however the changes made are small, and still do not mean that the poor will be eligible for credit. No substantial contribution has been made in promoting gender sensitive policies. Overall, the evaluation concluded that the impact of the project on institutions and policies had been limited, and the rating is therefore modest (2).
Gender equality. The project design included not only the women's programme, but specific references for inclusion of women-headed households in the SWC and agricultural components to enhance women's gender equality. As a result of demand, the credit allocations for women's IGAs have been significantly increased, which has also had the effect of increasing the skills training offered to women. Overall, therefore, women's access to project benefits has been good and more than half of project beneficiaries are women. Women beneficiaries confirmed that they felt more empowered and played a stronger role in decision making in the household as a result of being income earners, particularly those that had no previous employment. However, the project has not contributed to or facilitated progress in gender sensitive policies (paragraph 18). On balance, the impact on gender equality is rated as high (4).
Sustainability. The IE concluded that the project SWC interventions would be largely sustainable after project closure, as the on-farm structures require little maintenance. By comparison, there are significant maintenance requirements for the mini-dams, hafiras, and check dams on State land, and it is probable that the long term maintenance of these structures will remain a government responsibility, with little possibility of cost recovery from users. As benefits increase from the planting or rehabilitation of tree crops, then annual inputs from the farmers are likely to continue and will maintain yield levels. The sustainability of the increased area under olives may be threatened by low prices for olive oil or high prices for groundwater.
Although ACC has not established a specific revolving fund it has confirmed its commitment to continue to provide loans by going beyond the initial allocations for credit. The credit line is profitable for ACC and there is a demand for it, hence there is no reason why it should not be continued as part of ACC normal/ongoing operations. Mission estimates are that despite some incongruence between loan duration and actual funding requirements, the IGAs undertaken are in general profitable for the borrowers, and hence will be continued. On balance, therefore, the evaluation concluded that the major project activities would be largely sustainable, and that the overall rating should be substantial (3).
Innovation and scaling up. The project involved only limited innovation, rather success has been achieved by making known improved practices available to more people. This is replicable – hence the request for a second phase. The deep pruning of old olives can be described as a new technique, which can be replicated through the extension services wherever it is appropriate. To an extent, the introduction of check dams was also innovative. The project has provided training to many women who could not take loans because they could not meet ACC's conditions, if loan conditions could be modified then this programme could easily be expanded. Overall, the rating for innovations and scaling up is modest (2).
Other poverty impact. The emphasis for development within GOJ has shifted more towards ways of achieving poverty alleviation and creating employment – a strategy to which all Ministries are expected to contribute. For MOA, the successful experience with ARMP is the most appropriate initiative on which to build to help realize this strategy. Whilst this affect may not yet have been seen, it promises to be a significant factor in future planning. The rating for this sub-criterion is modest (2).
Overall, rural poverty impact as a result of the project was rated as substantial (3).
IFAD and the Cooperating Institution (CI). In assessing IFAD's performance, the mission concluded that whilst the intentions and relevance of the interventions could be rated very highly, there were some aspects of the design which were poor – especially the targeting criteria and the practical aspects of the implementation approach for community involvement. The arrangements for the involvement of ACC also did not offer any progress in terms of enhancing the outreach of credit to the poor. Overall, IFAD facilitated a design which "did the right things", but did not systematically ensure that "the right things were done right".
Overall, the mission concluded that the CI had performed well in supporting project implementation, and decided to rate the performance of both IFAD and the CI as substantial (3).
Government and its agencies. Overall, the arrangements made by MOA for project implementation have proved very satisfactory and commitment to the project is high. The only difficulties experienced during implementation relating to the MOA have been the reluctance to change Ministry operational arrangements to meet the implementation mechanisms as specified in the loan agreement and the reluctance to use loan funds for Technical Assistance (TA) or for contracting skills through NGOs etc. The Project Management Unit (PMU) has proved a very effective mechanism for project implementation, which has been demonstrated not only by the physical achievements of the project, but by the very pragmatic way that problems have been overcome, the decisive management involved in the modification of project targets at mid-term and the very good M&E arrangements and output. The ACC has been an effective partner in terms of disseminating the credit funds, but cannot be said to have contributed much to the achievement of the project's poverty objectives. Overall, the performance of the government agencies is rated as substantial (3).
Overall assessment and conclusions
Participation and targeting. In practice there has been little beneficiary participation in project planning or implementation. In its recommendations the MTE provided detailed suggestions as to how progress might be made with addressing participation in the second half of the project, and after this a community-based approach was developed in the spring and canal rehabilitation activities. However, no such approach has even been tried in the WDP, where it could have encouraged the participation of poorer women. The IE came to the conclusion that the targeting mechanisms in ARMP had been weak. Inclusion or exclusion as a project beneficiary had become both confusing and rather arbitrary, and there did not seem to be many factors which tilted the benefits towards the poor.
SWC and agriculture. The evaluation concluded that there had been substantial achievements in most of the major objectives of both the SWC and the agricultural development components. Firstly, the degradation of land and water resources has been reduced or arrested on both public and private land where project interventions have taken place. Secondly, there has been a shift from cereal to tree crop production on sloping land, with the consequence of improved conservation and the potential for improved farmer incomes, and thirdly, rangeland grazing and water access has been improved. However, the objective of optimising the long-term productive use of the land and water resources still needs to be addressed in the light of the continuing dependency of most project farmers on abstracted groundwater. In addition, the sustainability of the improvements achieved so far may be affected by reduced prices of olive oil, the reliance on purchased groundwater, and the lack of formally structured Water User Associations on communal irrigation schemes.
Credit. In terms of the disbursement of credit funds, ARMP had been very successful and had had a significant impact on household incomes for over 1 000 beneficiaries. The high repayment rate on these loans has meant that the exercise has been profitable for ACC and is a good indicator of sustainability of the programme. However, the mission concluded that one reason for the high repayments had been the stringent lending conditions, which meant that borrowers predominantly came from non-poor groups. The intended institutional strengthening and other support for ACC have largely not been realised, with the result that ACC's has not extended its outreach as planned.
Women's programme. The evaluation concluded that the WDP had achieved its objective of increasing family incomes through IGAs. The outreach has been very creditable, especially given the under-staffing of the Unit, and has demonstrated the popularity of and need for the credit which was made available. In addition, the provision of skills training pre-credit has enhanced the capacities of and served to empower a very large number of women, much more than was ever planned for in the design, and this will bring its own longer term benefits, even if credit is not made available to all.
Institutional strengthening. The very comprehensive training programme provided for project staff had been instrumental both in increasing capabilities and fostering good team spirit. Similarly, considerable training has been provided for beneficiaries. However, whilst skill levels may have been deepened, they have hardly been broadened, as little use has been made of the funds included for NGO support, research and studies. This has acted to limit the implementation of the project in the ways envisaged in the appraisal, especially in terms of developing participatory approaches and widening the scope of IGAs.
Coordination and management. The evaluation concluded that the project implementation mechanisms had largely been very effective, and had contributed both to the extent of the project achievements and the dynamic nature of the processes created. However, there is a need to arrange for a handover mechanism, so that completed areas and activities are on-passed to District based officers for routine follow up operations. In terms of coordination, the evaluation foresees a need to strengthen the role of the ACC as an implementing partner. At present credit activities are divorced from the central soil and water conservation objectives and activities and the PMU and ACC manage their separate aspects of the project on parallel tracks, without clear common objectives.
Participation. In the second phase ARMP could make a significant contribution by assisting MOA to apply participatory approaches in its poverty alleviation and agricultural development efforts. What is needed is the adoption of a strategic approach to participation and its mainstreaming at all levels within the Ministry. The mission has prepared a series of steps to guide this process (see recommendations, Chapter VIII). In order for the poverty focus of the project to be improved it is necessary to sharpen and clarify the targeting mechanisms. In the design of the new phase consideration should therefore be given to directing resources specifically towards the poorer groups, learning from the experience of phase I. A lost opportunity in Phase I was the lack of reaching the target group with credit for IGAs, and phase II needs to consider alternative strategies.
Water. In phase I the efficiency of water use in agriculture was not given the highest priority, but t he limiting factor in GOJ's development strategy for agriculture through conservation is water, not land. Firstly, this means that the project must stop relying on ground water for irrigation and gear the extent of the agricultural area to the scarcest resource, i.e. water. Part of the reason for the present groundwater abstraction is that the present capacity of the cisterns (30 m 3 ) is almost invariably too small for the supplementary irrigation requirements. However, the evaluation found that the present project proposals to address this problem may actually serve to exacerbate it, and are not based on adequate examination of the irrigation needs. Secondly, there was considerable variation between farmers in the frequency of irrigation and in the amounts of water applied, suggesting that both over-irrigation and under-irrigation may occur. In a low-rainfall area farmers, irrespective of their level of experience (and it should be borne in mind that some of the project's beneficiaries come from non-agricultural backgrounds), need to be given clear guidelines and recommendations regarding efficient water-use practices. Thirdly, in order to protect the investments already made and ensure a mechanism for management and maintenance on community schemes, and to settle any disputes which may arise, it is vital to ensure that there is an established local organisation which will take responsibility for the completed works. The present systems were found by the mission to be largely inadequate, and in phase II this needs to be given high priority. To address these three issues in water, the evaluation made a number of recommendations for consideration in phase II (Chapter VIII).
Agricultural extension. Jordan's agricultural extension service is seriously under-staffed and its effectiveness is consequently compromised. This presents a problem in terms of the continuing support required by farmers in order to fully realise the development potential from improved land and water resources. At the moment the extension agents working with the project appear to have to continue their activities with an ever-growing portfolio of clients. This is addressed under the decentralisation proposals of the evaluation (Chapter VIII).
Credit for development. In considering a second phase of ARMP, credit needs to be given a more central role and to be considered in the wider view of rural financial services, following IFAD's policy guidelines (which had not been developed a the time of the original design). ACC should continue to play a key role in any MOA development strategy, as it remains the only formal organisation able to provide financial services to producers in the rural areas. The need, however, is to identify the mechanisms through which ACC can grow into a much more broadly based lender, able to reach out to those individuals and groups who need ACC services in order to climb out of their poverty. Despite the achievements and successes in phase I, the limitations demonstrated show that there is still some way to go.
The evaluation has identified three issues on which progress needs to be made. These are, firstly, arranging for better integration of development objectives and coordination of activities between the PMU and ACC branches, as well as providing a mechanism for monitoring credit utilization. To achieve this it is recommended that a Credit Unit is included in the PMU. Secondly, the possibility of lending through groups must become a reality, at least for a small proportion of the credit funds provided. And thirdly, further institutional strengthening should be provided for ACC, to share with the Corporation the burden and risk in accepting the challenge inherent in making changes to their credit mechanisms. ACC needs external expertise and stimuli to establish the new mechanisms suggested, and needs support from an organisation that has positive experience in anticipating and addressing the difficulties that will arise. Detailed recommendations are made by the evaluation covering these aspects (Chapter VIII).
Women and poverty. Whilst the present WDP provides a building block for phase II, it should not just be copied. Rather, the need is to improve a number of aspects. In particular, in order to reach poorer women it will be necessary to encourage group formation, before moving to suggesting IGAs and providing credit. To facilitate group formation will require the assistance of a skilled and experienced local NGO, which will need to emphasise capacity building for beneficiaries (such as the organization of collective working, administration and marketing skills). The evaluation also suggests the sponsoring of women's centres, as it would be more cost efficient for poor women to work collectively and would assist in institutionalising women's groups. Finally, the range of IGAs needs to be expanded after feasibility and marketing studies which seek to identify products that have market demand and are financially viable.
Institutional strengthening. In phase I the Ministry was not convinced that the participatory approaches being suggested were either cost effective or appropriate to the development being sought. However, it is now timely to help the MOA with the sort of institutional strengthening which will let them develop new approaches to development. The key to this is for the Ministry to accept the need for assistance with the development of skills which it does not have. An appropriate package of institutional strengthening should therefore be included in phase II, covering both the technical training requirements of beneficiaries and Ministry staff, and sufficient to allow for the adoption and introduction of the suggested participatory methods. In addition, assurances should be sought that identified staff needs and planned training activities will be fully funded during implementation of phase II.
The evaluation considered that in phase II wider use should be made of the assistance and support available in other Departments in the MOA, both to address strategic issues (e.g. the marketing of olive oil) and to obtain technical support (e.g. with arranging training). In addition, care should be taken to link with the Ministry's own M&E Department, in order to harmonise reporting on the overall progress of projects. Consideration should also be given to the adoption of the M&E system devised for ARMP in a wider context: there is the potential to extend the methods used to many projects in the Region. With appropriate support ARMP could provide a local "centre of excellence" for training in M&E.
Decentralisation. The differentiation between project activities and the regular roles of the staff of the FDAs have been blurred. Whilst this may have come about because the project has more funding and is able to promise more to potential beneficiaries, it is not sustainable. Essentially the strategy should be for the project activities to temporarily complement the activities of the FDAs. This may require additional support for the extension services as well as a formalised handover process by which developed areas become the responsibility of the FDAs. In addition, the ongoing involvement of the Regional Agriculture Coordinating Committee (RACC) needs to be strengthened, so that relevant decisions about the project's day-to-day work programme can be taken by the agencies most closely involved. Both these aspects would serve to reinforce the decentralised nature of the project's activities. As such, the future purpose of the PMU might then be better understood as demonstrating the new approaches and techniques and training the local staff, before moving on.
Recommendations to this affect have been provided.
1/ The Interim Evaluation Mission was composed of Mr Frank Butcher , Mission Leader and Institutions Specialist; Dr Mouna Hashem, Sociologist and Gender Specialist; Mr Swithun Goodbody , Natural Resources Management Specialist; Mr Omar Imady, Credit Specialist; Mr Sarath Mananwatte , Economist and Financial Analyst. Dr Mona Bishay , Deputy Director of the Office of Evaluation undertook the preparatory field mission, supervised the evaluation and led the process of presentation of the Aide Memoire in the Wrap-up Meeting.
2/ Methodological Framework for Project Evaluation (EC 2003/34/W.P.3)
3/ Hafiras are reservoirs used for livestock.
4/ Credit was not a separate component of ARMP, but was the major means of achieving project objectives. Credit activities were implemented by the Agricultural Credit Corporation (ACC), which is part of the MOA, through a subsidiary loan agreement.
5/ This was the figure for end-June 2003.
6/ The evaluation methodology followed (see Appendix 2) uses three composite evaluation criteria: project performance (composed of relevance, effectiveness and efficiency) rural poverty impact (composed of six impact domains, sustainability, innovations and gender equality) and performance of partners (including IFAD, implementing agencies, and the cooperating institution). The ratings used to assess performance using these criteria are high (4), substantial (3), modest (2) and negligible (1). For the sustainability criterion the rating used is highly likely (4), likely (3), unlikely (2) and highly unlikely (1).