IFAD assistance to Ethiopia. IFAD has provided loans totalling US$206 million to finance 13 projects in Ethiopia since 1980. A further US$288 million in cofinancing for these projects has been secured from the African Development Bank (AfDB), the Belgian Survival Fund (BSF), the Government of Ireland and the World Bank. Counterpart funding from the Government of Ethiopia amounted to US$98 million, for a total project portfolio equivalent to US$592 million. IFAD's assistance to the country has also included the provision of a few small country grants, as well as some larger regional grants in which Ethiopia's estimated share is about US$4 million. Appendix 1 provides some basic data about the projects funded by IFAD in the country.

Evaluation objectives, methodology and processes. The main objectives of the country programme evaluation (CPE) have been to: (i) assess the performance and impact of IFAD-funded operations in Ethiopia; and (ii) develop a series of findings and recommendations that can serve as building blocks for the preparation of the new country strategic opportunities programme (COSOP) for Ethiopia by IFAD and the Government of Ethiopia. The new COSOP will be submitted to the IFAD Executive Board for consideration during its ninety-fifth session in December 2008.

In line with the usual procedure for CPEs, this evaluation covers IFAD assistance to Ethiopia over a 10-year period (from 1997 to 2007). More specifically, it includes an assessment of 7 of the 13 loan-funded projects approved since 1980, a review of non-lending initiatives (policy dialogue, partnership-building and knowledge management) and an analysis of grant-financed activities.

The findings of the CPE are based on: (i) a comprehensive desk review of existing evaluative evidence and other documentation; (ii) self-assessments by the Eastern and Southern Africa Division (PF) and the authorities of three IFAD-assisted projects, namely the Agricultural Research and Training Project (ARTP), the Pastoral Community Development Project (PCDP) and the Rural Financial Intermediation Programme (RUFIP); (iii) surveys conducted by microfinance institutions participating in RUFIP on the results of the rural financial services that have been provided; (iv) five weeks of field work in Ethiopia by an Office of Evaluation (OE) multi-disciplinary evaluation team; (v) key informant and focus group discussions conducted during the evaluation mission; and (vi) information provided by project partners, including PF, the Government, donor organizations and others.

A preparatory mission for the CPE was conducted in May 2007, and the main evaluation mission took place in September-October 2007. Comments from the Government and PF concerning the CPE mission's aide-memoire were duly considered in preparing the CPE report, which has been enriched by a comprehensive internal OE peer review, as well as the written comments received from PF and the Government. The report has also been shared with the main cofinanciers for their comments. In addition, a senior independent adviser¹ was hired by OE to review the draft final report. The main issues² emerging from the CPE were discussed at the CPE national roundtable workshop held in Addis Ababa on 26-27 June 2008.

Economy and poverty. Despite recent impressive economic growth, Ethiopia remains among the poorest countries in the world. The population, numbering close to 80 million, have a per capita income of about US$200 (the same level reached in 1973 before the economic decline that occurred during the Derg regime of 1974-1991). Since 1992, Ethiopia has undergone a process of decentralization and market liberalization, but many economic activities continue to be managed by the State and by political parties. Moreover, the country has to deal with a number of significant macroeconomic imbalances. These disequilibria are largely attributable to low savings rates, which constrain efforts to develop Ethiopia's modest capital and technology base. About 80 per cent of the country's households obtain their livelihood from traditional low-productivity agricultural activities, which are subject to recurrent droughts. Some 39 per cent of all households are below the national poverty line (down from 46 per cent in 1996).

Poverty reduction strategies. In 2002, the Government introduced its first poverty reduction strategy paper (PRSP). This was followed by a second-generation PRSP in 2005, entitled "Plan for Accelerated and Sustained Development to End Poverty (PASDEP), 2005/2006 – 2009/2010". The PRSPs reflect the agricultural development-led industrialization policy which was introduced in the 1990s to give high priority to agricultural and rural development. The share of the total government budget allocated for agriculture and food security is unusually high in Ethiopia (over 10 per cent) compared to other African countries (where it is generally less than 5 per cent), and this provides a conducive environment for implementing and sustaining agriculture-related development projects.

The Quality Of The Country Strategic Opportunities Programme (COSOP)

The strategic directions for IFAD's cooperation with Ethiopia in the 1990s were set forth in the IFAD special programming mission document of 1989, which focused on promoting smallholders' incomes and farm production in low-income and food-deficit regions.

Together with the Government, IFAD formulated its first COSOP for Ethiopia in 1999. This paper outlined, among other issues, the main objectives for the country programme. There were few resources available for the preparation of the 1999 COSOP, yet its formulation was nonetheless carried out, broadly speaking, on the basis of a participatory process and in accordance with the guidelines of the time. As such, it did not include measurable objectives or indicators. Nor did it set out a detailed targeting strategy. Instead, it concentrated on: (i) a set of subsector priorities for project portfolio development (including rural finance, small-scale irrigation, agricultural diversification and marketing); (ii) portfolio management (sector development programmes, beneficiary participation, baseline and socio-economic surveys and integration of project management units in government structures); and (iii) policy dialogue directions (reorienting the role of regional agricultural bureaux, reducing the role of government in economic activities that can be performed more successfully by the private sector, and promoting the reform of land tenure systems).

The CPE has determined that the subsector priorities of the COSOP are relevant, although more analysis of the constraints on private-sector development would have been desirable, especially before a decision was made to embark on investment in agricultural marketing. Because so few resources were allocated for the preparation of the COSOP, very little analytical work was conducted as a basis for setting priorities (e.g. in terms of geographic focus, subsector engagement and so on). On the positive side, it should be noted that all subsector priorities for portfolio development have been adhered to. In addition, IFAD has financed a pastoral community development project as part of the United Nations response to the crisis situation caused by the drought in Ethiopia.

The portfolio management directions of the COSOP have generally been complied with, although limited progress has been made in conducting baseline surveys and strengthening project-level monitoring and evaluation (M&E) systems. The policy dialogue objectives outlined in the COSOP were rather ambitious, especially in light of the limited human and financial resources available for pursuing these objectives. The COSOP did not present a comprehensive targeting strategy for rural poverty reduction, leaving targeting priorities and modalities to be defined within the context of individual projects and programmes. Moreover, the COSOP did not provide a great deal of guidance on how to ensure linkages and synergies among the various projects and programmes funded by IFAD in the country.

Assessment of the 1999 Ethiopia COSOP

6=highly satisfactory; 5=satisfactory; 4=moderately satisfactory; 3=moderately unsatisfactory; 2=unsatisfactory; 1=highly unsatisfactory.

IFAD-funded projects in Ethiopia

As mentioned above, the CPE covered 7 out of 13 loan projects which have accounted for 73 per cent of all IFAD lending to Ethiopia since 1980. The first three projects included in the CPE were designed before the 1999 COSOP and correspond to the pre-COSOP period. These projects are the Southern Region Cooperatives Development and Credit Project (SOCODEP), Phase II of the Special Country Programme and the Agricultural Research and Training Project (ARTP). The other four projects, which correspond to the post-COSOP period, are the Pastoral Community Development Project (PCDP), the Rural Financial Intermediation Programme (RUFIP), the Agricultural Marketing Improvement Programme (AMIP) and the Participatory Small-scale Irrigation Development Programme (PASIDP).

In addition to these loan-supported projects, the CPE assesses two small country grants and five larger grants. The latter were regional and interregional grants for Ethiopia together with other countries in the PF region. These grants have primarily financed activities of international research institutions belonging to the Consultative Group of International Agricultural Research (CGIAR), but in a few cases they have also supported activities of NGOs. Support has been furnished for a wide range of activities, from the testing of stress-tolerant cereal varieties to livestock pest control and rural finance. It is estimated that, since 1993, Ethiopia has taken part in 15 large and small regional grants having a total value of US$12 million, of which Ethiopia's "share" is estimated at US$4 million.

Overview of the Loan-funded Operations Assessed by the CPE

	Appr.	Eff.	Clos.	Total cost	IFAD loan	Cofinanciers	CI.	Criteria evaluated	Eval. sources
US$ million
Southern Region Cooperatives Development and Credit Project (SOCODEP)	1993	1994	2005	21.9	17.45	BSF	UNOPS	All	Compl. eval., PCR
Phase II, Special Country Programme II: (Small-scale irrigation)	1996	1999	2007	31.9	22.6	Government of Ireland	UNOPS	All	Interim evaluation, PCR, field visit
Agricultural Research and Training Project (ARTP)	1998	1999	2007	90.6	18.2	World Bank	World Bank	All	Field visit, project information
Pastoral Community Development Project (PCDP)	2001	2004	2009	59.9	20	World Bank	World Bank	All	Field visit, project information
Rural Financial Intermediation Programme (RUFIP)	2001	2003	2010	88.7	25.7	African Dev. Bank	World Bank	All	Field visit, project information
Agricultural Marketing Improvement Programme (AMIP)	2004	2006	2013	35.1	27.2	None	UNOPS	Relevance	Project information
Participatory Small-scale Irrigation Development Programme (PASIDP)	2007	2008	2015	57.7	20 loan, 20 grant	None	Direct by IFAD	Relevance	Project information

Performance and impact

The overall performance of the project portfolio. The portfolio's performance (measured in terms of relevance, effectiveness and efficiency) is assessed as being satisfactory⁴ for the post-COSOP portfolio (rural finance and pastoral community development) and moderately satisfactory for the pre-COSOP portfolio, with the exception of small-scale irrigation, which has been supported since the 1990s through three different operations.

Relevance. The objectives of the loan-supported projects have been found to be highly relevant in the case of rural finance, pastoral community development and small-scale irrigation. These projects responded to the needs of the rural poor and were aligned with the main Government and IFAD policies and strategies related to rural poverty reduction. Moreover, the lessons from the OE interim evaluation of the second phase of the Special Country Programme were duly incorporated into the recently launched Participatory Small-scale Irrigation Development Programme (PASIDP).

On a related issue, the CPE recognizes the importance for the Government of investing in the development of the National Agricultural Research System, which is seen as a key feature in promoting food security in the country. However, it also calls attention to the type of contribution IFAD can make to the process, especially in view of the fact that research results have often not become available until long after a project's completion. This has limited the opportunities for transferring new technologies to the rural poor and for promoting their adoption. Furthermore, while the CPE supports the introduction of research into drought-prone under-served areas, some uncertainty remains about the selected approach, which would focus on the construction of large research centres with costly infrastructure based on the assumption that highly qualified researchers could be convinced to come to remote and marginal areas and to live and work there on a permanent basis. Similar comments were made during the internal formulation process in 1998 by IFAD's Technical Advisory Division. In fact, according to the CPE, IFAD's experience with the implementation of ARTP indicates that these comments are pertinent even today.

The design of the recent agricultural marketing project is broadly consistent with IFAD's Private-Sector Development and Partnership Strategy. However, while recognizing that the marketing project has been in place for just over two years, the CPE notes that the project needs to explore opportunities for greater public-private partnerships.

Effectiveness. The effectiveness of the interventions undertaken in the areas of rural finance, pastoral community development and irrigation is assessed as satisfactory. Coverage of beneficiaries or intervention areas has been expanded beyond the levels initially planned, and the overall quality of services has matched the needs of the beneficiaries. Effectiveness has been appraised as being moderately satisfactory in the case of agricultural research (benefits to the broader research system in Ethiopia, while still limited, are potentially transferable to extension efforts and to farmers) and as moderately unsatisfactory in the case of cooperative development (there has been only limited progress in terms of the quality of services, and problems of insolvency have arisen).

In the area of pastoral community development, effective and innovative models of local governance have been introduced for planning and implementing investments in community infrastructure, as well as in income-generating activities for the poorest, and this has provided stakeholders and beneficiaries with a sense of ownership. Communities are actively engaged in the planning and implementation of microprojects to which they contribute in kind or with cash. Substantial improvements in living standards are noted.

In the sphere of rural finance, outreach efforts have achieved impressive results through the development of microfinance institutions (MFIs), which in Ethiopia can be regarded as small banks, as most of them are allowed to mobilize savings. The scope of outreach activities has expanded at an average annual rate of 34 per cent, with a total of 1.72 million clients by the end of June 2007, which corresponds to about 20% of all rural households. The average annual growth rate for outstanding loans has been 105 per cent, and the rate of expansion for net savings has been 50 per cent. Most of RUFIP's end-of-programme targets (i.e. project year 7) had already been surpassed in its fourth year. Discussions with various stakeholders in the sector indicate that RUFIP has been the major catalyst for this growth, thanks, in particular, to its facilitation of linkages between MFIs and the banking industry. RUFIP has also created a new type of microfinance institution in Ethiopia from scratch, namely rural savings and credit cooperatives (RUSACCOs). While many RUSACCOs have been created, their membership is still limited. In addition, members have received insufficient training, mainly because of delays in the AfDB-funded component, and this raises concerns about the sustainability of these cooperatives.

Small-scale irrigation projects have placed priority on districts that are classified as highly or very highly vulnerable. About 70 per cent of the schemes included in Phase II of the Special Country Programme have been in such areas. Significant progress has been made in these areas towards the main objective of increasing yields and cropping rates by expanding irrigated agriculture, and the targets for beneficiaries and for the land area to be brought under irrigation have been surpassed. However, advances have been relatively modest in the case of objectives relating to water management and the settlement of water-rights issues, user organizations, soil conservation, crop husbandry and vegetable seed production, and the development of economic activities for women.

The support provided to cooperatives (through SOCODEP) did not achieve this activity's key objective of establishing a model for cooperative development in Ethiopia. Although quantitative targets for re-registering and restructuring cooperatives were surpassed, by the end of the SOCODEP intervention these cooperatives were weak, and some were on the verge of bankruptcy. The contribution made to the development of the cooperatives' commercial activities (e.g. flour mills and retail shops) was modest. Owing to a lack of business management skills, the activities concerned were performed poorly. Today, many of these ventures have closed their doors due to competition from the private sector. In most of the credit components, loan disbursement and loan recovery were unsatisfactory.

The CPE concludes that the combined support of the World Bank and IFAD in ARTP made a significant contribution to the strengthening of the National Agricultural Research System (NARS) of Ethiopia, in particular the large human resource development components funded by the World Bank. However, the effectiveness of IFAD's contribution to the transfer of technology from research institutions to farmers is assessed as moderately satisfactory. While the overall research system is generating an increasing number of agricultural technologies, the adoption of these technologies by the majority of farmers, which is expected to result in major improvements in national yields, will take many years. The reasons for this situation may be found both in the research system and in the constraints to which input supply and extension systems are subject.

Efficiency is assessed as satisfactory for rural finance, due to the favourable operating cost ratios when compared to regional standards in the industry. Efficiency is assessed as moderately satisfactory for pastoral development and small scale irrigation. While unit costs for construction are within the parameters of comparable interventions, because of the incomplete status of much of the infrastructure (pastoral development) and delays in implementation (irrigation), the benefits will accrue to the project much later than expected. Efficiency is assessed as being moderately unsatisfactory in the cases of both cooperative development and agricultural research. This is due to the fact that the level of project outputs is significantly lower than expected and to delays and high unit costs in construction and delivery. This situation is, however, attributable to many factors that are beyond the control of the project management teams, such as the focus on quantitative targets and the wide geographic coverage of activities which is called for in the project design.

Rural poverty impact

The CPE assesses the overall impact on rural poverty as being satisfactory, with the exception of the early support for cooperatives provided through SOCODEP, for which the impact is assessed as moderately unsatisfactory. Given the nature of the project, in which emphasis is placed on the construction of large research centres, the impact of the support provided for agricultural research is not rated because it could not be gauged at the time of evaluation. Because implementation of AMIP and PASIDP has begun so recently, the CPE did not assess or rate the impact of these two new projects. The CPE ratings for impact and other evaluation criteria are shown for each project or programme in Appendix 2.

Impact on household income and assets. In terms of the number of households in which income levels and asset ownership have improved, the most significant contribution has been made by the support provided for rural finance, followed by the support furnished for pastoral communities and irrigation. Rural financial services and the MFI industry are making a significant contribution to poverty reduction in Ethiopia by reaching the poor, although they do not always reach the poorest. Impact studies consistently identify widespread and significant improvements in household income, consumption and asset-building among the vast majority of MFI clients, who are mainly the "economically active poor". As is common in microfinance, the available information suggests that although some changes begin to occur when the very first loan is disbursed, it is not until after the fifth loan (usually by the fifth year) that significant improvements in income and living standards can be seen.

There is evidence that pastoral community activities are having a broader impact in terms of poverty reduction. Through an income-generating scheme, interventions have directly contributed to improving the income levels and asset accumulation of some 10,000 of the poorest community members, of whom 78 per cent are women (including female heads of household). According to the evidence gathered through focus group discussions, it appears that activities are simple and affordable for very poor households (e.g. petty trading, breeding of poultry, fattening of goats and sometimes oxen). These results can also be attributed to the effective and participatory rural appraisal which was conducted when intervention plans were being prepared. In addition, many more households may have obtained indirect income benefits from the community microprojects, in particular the water supply schemes.

Some 31,000 households in densely populated drought-prone areas have been reached through the support provided for small-scale irrigation projects, and many of these households are gradually seeing an improvement in their incomes. Findings suggest that increases in crop yields over the traditional yields are in the range of from 25 to 40 per cent, and in cases where irrigation facilities have been built around springs, the increases have been between 75 and 100 per cent. Thanks to these irrigation projects, the targeted irrigation farmers' physical and financial assets have started to increase, although experience suggests that it may take six or more years for irrigation farmers to experience the full benefits.

Food security. The most direct and significant contribution to an improvement in food security for rural households has been made by the support furnished for small-scale irrigation projects. Information collected by the OE interim evaluation team on Phase II of the Special Country Programme shows that some farmers were experiencing a reduction in the number of "hungry months" from about six to two (July and August) thanks to larger and more reliable yields and higher income. It has also been reported that the range of dietary intake is widening due to crop diversification.

The support provided for rural finance and pastoral community development has also made important direct and indirect contributions. In the case of rural finance, the various impact studies show that the first impact for new clients is consumption smoothening, as these interventions enable households to meet their food requirements throughout the year. Generally speaking, for most rural clients, the first few loans are used to purchase oxen (usually for use in ploughing, but also for sale after the oxen are fattened). Multi-access loans have permitted the diversification of the income base, and this, combined with growing savings deposits, has improved clients' capacity for coping with drought and other external shocks. In pastoral communities, many microprojects have been undertaken to improve the water supply, which enables these communities to deal more successfully with recurrent droughts. The support furnished for cooperative development made much less of a contribution than expected to an improvement in food security.

Unfortunately, no baseline surveys have been conducted on the nutritional status of children, which is often considered to be one of the most reliable indicators of overall food security. Baseline studies, although recommended in the COSOP, have generally not been conducted in IFAD-supported projects and programmes. Data on food security and child malnutrition are particularly important because Ethiopia still suffers from high levels of child malnutrition (51 per cent according to the World Health Organization (WHO) and 48 per cent according to PASDEP) even though these levels did decline somewhat in the period 1995-2005.

Market access. Providing greater access to markets has not been among the key objectives of these interventions, with the notable exceptions of AMIP and SOCODEP. Therefore, not surprisingly, the programme's contribution in this sphere has been modest when compared to the contributions made in other areas. For example, some rural finance clients who have bought oxen or other transport animals have improved their access to markets. Similar effects may be seen for some beneficiaries of the income-generating scheme supported in pastoral communities. The support for cooperative development included the construction and rehabilitation of roads, which did improve market access for some households, though fewer than targeted. Limited achievements were made in promoting viable service cooperatives that provide efficient access for their members to markets and services.

Human capital. The main contributions to development have come from the support provided to pastoral communities and the BSF-financed water supply, health and basic sanitation component of SOCODEP, while the support furnished for rural finance has made more indirect contributions. More than 10,000 staff and community members in pastoral areas have been trained, and households are starting to benefit from the services provided by health posts and new schools (although no data are available on the quality of the teaching or of the learning process). The results of the BSF health component were - according to a BSF-financed impact study and the OE project completion evaluation of SOCODEP - positive and substantial. In rural finance, the impact has been more indirect. Some impact studies have reported that some clients have improved their income levels and are therefore in a position to send their children to school. However, progress in supporting skills development by MFI staff and RUSACCO members has been modest due to the lengthy procurement procedures employed by the cofinancing partner (AfDB).

The impact of irrigation interventions on human assets, in the form of the development of skills and knowledge, has been limited by the generally poor quality of extension work, an unimaginative use of trials and demonstrations, and the limited institutional support that has been provided. While a large training programme was made available in order to provide support for cooperatives, any lasting impact on the institutions involved was undermined by frequent government restructuring and re-deployment of personnel. Furthermore, capacity-building efforts largely ignored the importance of changing people's attitudes towards the cooperative model.

Social capital. The most significant contribution made in this domain has come from the support furnished for pastoral community development. Communities have been empowered through the effective use of participatory methods and the formation of the woreda (district) development committees and community development committees, which include members from government, the private sector and civil society. These committees may serve as a model for woreda and community planning throughout the country. Significant impacts have also been observed in the area of rural finance, where credit groups, local networks and RUSACCOs are helping to develop social capital at the grassroots level. While the support for cooperatives was expected to make a major contribution to social capital development, at the project completion point most of the cooperatives were still weak, both financially and with respect to management capacity and business skills.

In the sphere of irrigation, the impact on social capital, through the establishment and strengthening of local organizations for water management, has been more limited. The situation has been complicated by the presence of three different organizations within the same scheme: the traditional water user group, the "modern" water user association (WUA) and an irrigation cooperative. Traditional water-user groups have not been utilized effectively in the move to "modern"' organizational forms (WUAs and cooperatives). The cooperative promotion departments, which are mandated to strengthen WUAs, have focused on the promotion of irrigation cooperatives, even though the cooperative concept is unattractive to some (perhaps many) farmers because of the coercive application of cooperative schemes during the Derg administration. This has been taken into consideration in the design of the latest small-scale irrigation intervention (PASIDP).

Institutional impact. The support provided for pastoral areas has effectively contributed to new approaches and systems for planning and implementing public investments at the community level. The woreda development committees and community development committees are in operation and are contributing to a sense of local ownership. The support for rural finance has made a significant contribution towards building an inclusive financial system that can sustainably address the financial needs of the poor. Mechanisms for linking the MFI sector and the banking industry have been introduced, and a diversification process has been initiated in terms of the products offered and the range of institutions servicing the poor, including RUSACCOs. Finally, the capacity of the regulatory framework in respect of both MFIs and RUSACCOS has been strengthened, in particular by helping the National Bank of Ethiopia to upgrade its Microfinance Supervision Division and give it the status of a full department. Also, some steps have been taken to reinforce self-regulatory mechanisms in the microfinance industry by supporting the Association of Ethiopian Microfinance Institutions (AEMFI).

The overall National Agricultural Research System (NARS) of Ethiopia has been significantly strengthened through ARTP for human resource development and facilities. Through its involvement in this support effort, IFAD has helped to introduce competitive research grants and to establish the basis for improving linkages with the extension system. The chances that IFAD's support for six agricultural research centres in remote drought-prone areas will have a positive institutional impact will depend on how the current problems of these centres are solved. These problems include a failure to complete construction work, a lack of potable water, inadequate accommodation facilities, and difficulties in attracting and retaining high-quality staff. At the project's close, major efforts were reportedly being made to solve the water-supply problem.

Sustainability. It is likely that most of the benefits promoted through IFAD-supported activities will be sustained after the relevant projects come to an end. In fact, in Ethiopia, sustainability prospects are significantly better than they are, on average, for IFAD-funded projects across all regions (see table 3). In recent years, more than 10 per cent of the Ethiopian Government's budget has been allocated for agriculture and food security. Therefore, within the public domain, budgetary resources are usually available to support the continuation of activities in this field. Another positive element is that project management units are well embedded within the decentralized government structure (Phase II of the Special Country Programme, PCDP) or in permanent national organizations (RUFIP, ARTP).

High staff turnover and a thinly spread agricultural budget may, however, have a negative impact on operations. Outside the public domain, there are "sustainability threats" for some activities. The water user associations do not generate sufficient income to rehabilitate and maintain the main civil engineering structures involved in the small-scale irrigation schemes. Within the realm of rural finance, MFIs in Ethiopia have excellent portfolio quality and good operational efficiency. In spite of this, the return on assets and on equity are both negative owing to a combination of low interest rates (approximately 12-20 per cent, despite the fact that no official cap exists) and high inflation (estimated at around 40 per cent as of mid-2008), which results in negative real rates of interest. If inflation remains high, negative real lending rates may reduce people's motivation to save and may encourage borrowing for economic activities that may not be profitable in real terms. Finally, the significant shortage of inexpensive capital relative to demand may, as has been seen in other countries, result in the administrative and rent-seeking allocation of the limited supply of inexpensive capital that does exist.

Innovation, replication and scaling up. The IFAD portfolio has contributed to the introduction of a number of systems and approaches that are innovative in the Ethiopian context. For example, in agricultural research, innovations have included: (i) a system of competitive research grants; (ii) Farmer Research Groups, through which farmers are involved in research activities on an ongoing basis (this approach will be continued and scaled up with the help of funding from the Japan International Cooperation Agency (JICA)); and (iii) a system of research extension advisory councils which is supported by public policy and the government budget. In the area of pastoral community development, a community-driven development approach has been introduced, and the planning and management of community investments are now being conducted by the woreda (district) development committees and the community development committees. The potential exists for scaling up this approach and system, not only in pastoral areas (further funding will soon be coming from the World Bank), but nationally as well. In the sphere of rural finance, RUFIP has helped to link MFIs with the banking industry, and large MFIs are now accessing funds from commercial banks.

The regional and interregional technical assistance grants provided to Ethiopia and other countries have mostly been used to fund agricultural research on the part of international research institutions. While useful research results have been produced in areas that are relevant to Ethiopia, there have been limited linkages with the loan portfolio, and the mechanisms for replicating and scaling up these interventions have been weak. This has raised questions as to the rationale for IFAD to provide this type of support. By contrast, two small country grants for RUFIP have been used to fund preparatory studies which have proved useful for project design purposes, thus providing an example of grant use which is directly connected to the lending programme.

In general, while the Ethiopia CPE notes that projects and programmes have introduced innovations in technology or in social areas, the replication and scaling up of tried and proven innovations have not been systematic. Although recently greater efforts are being deployed in this area, in the past insufficient attention and resources have been devoted to policy dialogue, knowledge management and partnership-building, all of which are essential ingredients for replication and scaling up (see the following section). Direct supervision and the provision of implementation support, together with the maintenance of an IFAD country presence in Ethiopia since 2005, are steps in the right direction which can contribute to more effective innovation scouting and promotion.

Non-project activities

The implementation of non-lending activities (knowledge management, policy dialogue and partnership-building) has been limited, mainly because of a lack of resources and the fact that in the past a high priority has not been placed on such activities. The situation in this respect is improving, however. First, the country programme manager now has more resources available than s/he did a decade ago, which allows the manager to engage more effectively in non-lending activities. Second, the CPE found that the country presence has contributed to improvements in donor coordination, an exchange of experiences and policy dialogue.

Policy dialogue. IFAD's main contribution to policy dialogue has been made during the project design phase. Furthermore, in some cases, a policy dialogue component has been included in the project design (e.g. in the pastoral community development and rural finance projects). However, supervision reports note that policy dialogue components are lagging behind "operational" components and that engagement by government agencies has not always been as expected. The current CPE concurs with that judgement. Moreover, IFAD's contribution and capability to engage in policy dialogue at the national level has been challenged by various factors, including the limited resources available for conducting analytic work, the lack of a country presence until 2005, and the definition of an ambitious policy dialogue agenda. Finally, the Government and development partners have established a high-level forum and a technical working group on agriculture. IFAD takes an active part in these groups, which have, among other issues, discussed important policy matters relating to the sector.

Knowledge management. This area of activity was identified as a high priority in the COSOP, but limited progress has been made in this respect. As mentioned earlier, project-level M&E systems, which are at the foundation of a vibrant knowledge management system, have generally performed unsatisfactorily. Under the civil service reform programme, public institutions are improving their management information systems and are conducting planning, budgeting and reporting functions based on output targets and deliverables. However, relatively little attention continues to be paid to impact issues, and baseline and repeat surveys focusing on changes in household livelihoods are therefore generally not done. As a means of stimulating knowledge management, in 2007 IFAD launched the Country Programme Forum to facilitate contacts and meetings among IFAD project stakeholders (the Government, IFAD, other donors) with a view to exploring synergies between projects and different actors and exchanging experiences and lessons.

Partnerships. At the federal level, there is a solid partnership with the Government, especially the Ministry of Finance and Economic Development, Ministry of Agriculture and Rural Development, Ethiopian Institute of Agricultural Research and others. These agencies regard IFAD as a flexible and valuable organization that is working to reduce rural poverty by promoting innovations in remote areas, and they realize that this area of endeavour is not usually considered to be a priority by other development organizations.

In recent years, cofinancing partnerships have declined in importance as several major development partners adopted the budget-support modality. Furthermore, some of IFAD's traditional partners, such as AfDB and the World Bank, did not place priority on small grassroots-type agricultural and rural development interventions in Ethiopia in the period assessed by the CPE.

Non-governmental and civil society organizations (NGOs and CSOs) have traditionally played a less important role in development cooperation in Ethiopia than in some other African countries and have therefore not been as widely involved in IFAD-supported projects and programmes in the past. However, the capacity of NGOs and CSOs is improving and, as suggested by the experience gained in support activities for pastoral communities, NGOs and CSOs can play an important role in supporting communities and grassroots organizations. The capacity of private-sector service providers is also expanding, albeit from a low base level, and this remains an area in which further inroads can be promoted within the context of IFAD operations.

The performance of IFAD and its partners

The performance of IFAD. In the majority of cases, IFAD has contributed to good project design. This is especially the case in IFAD-initiated projects and programmes in such areas as rural finance and small-scale irrigation. As a consequence of the operating model used in the past, under which supervision was delegated to cooperating institutions, IFAD was perceived as a flexible but distant partner in project execution. With the adoption of the direct supervision and implementation support policy and the establishment of a country presence in Ethiopia, this perception is rapidly changing. The CPE found that IFAD's country presence is an importance feature of the operating model which can help to further strengthen its development effectiveness, even though the present country presence arrangements (e.g. limited resources and delegation of authority) may act as a constraint on its opportunities in the future.

Cooperating institutions and cofinanciers. The World Bank has served as the cooperating institution (CI) in three of the projects reviewed in the CPE. The best performance has been observed in PCDP, where the Bank's Ethiopia country office is in charge of the provision of support and where all items have truly been cofinanced by IFAD (40 per cent) and the World Bank (60 per cent). The least satisfactory performance was seen in ARTP, where support was provided through brief missions from Washington and where IFAD was fully financing three separate components of the project. The United Nations Office for Project Services (UNOPS) served as CI on two of the reviewed projects and provided a moderately satisfactory level of service, but did not focus enough on correcting problems faced by these projects.

The CI and cofinancing partnerships with the World Bank and AfDB have been hampered by the cumbersome procurement procedures and regulations of these organizations. This has resulted in delays in implementation, particularly in the case of research (ARTP, World Bank) but also in the area of rural finance (RUFIP), where AfDB regulations have hindered progress on the capacity-building components; this, in turn, has had a negative impact on the IFAD-financed credit component.

The performance of the Government and its agencies. The Government's overall performance is assessed as satisfactory. IFAD has been engaging in an increasingly constructive and useful dialogue with key government ministries and agencies. A useful dialogue has been maintained with the Development Bank of Ethiopia in connection with the credit component for which it is fully responsible. In the instances in which performance has not been fully satisfactory, the major problem has been a lack of clarity regarding the assignment of responsibilities. For example, in the support for cooperatives provided under the SOCODEP project, too many agencies were involved and major institutional changes took place which hurt performance.

Overall Rating

* The rating scale adopted by the Office of Evaluation is as follows: 6 = highly satisfactory; 5 = satisfactory; 4 = moderately satisfactory; 3 = moderately unsatisfactory; 2 = unsatisfactory; 1 = highly Unsatisfactory.
** Ratings considered here are those of SOCODEP, SCP II, ARTP, RUFIP, PCDP.
*** The ratings shown in the 2007 edition of the Annual Report on Results and Impact of IFAD Operations (ARRI)
refer to the evaluations conducted by the Office of Evaluation in 2006.

Conclusions and recommendations

Conclusions

Clear portfolio development directions, but limited analysis of resource needs. The 1999 COSOP was prepared at very little cost but provided concise and clear directions for portfolio development and non-lending activities. Given the limited resources available for its preparation, the COSOP's analytical underpinnings were, understandably, inadequate. Among other shortcomings in this regard, different typologies of rural poverty in the country were not well captured. The COSOP also implicitly assumed that policy dialogue and knowledge management would be taken care of through IFAD-financed projects, without any accompanying activity or a specific budget allocation. Finally, the COSOP did not clarify how the different subsector programmes would reinforce each other (for example, how to provide financial services, irrigation and marketing services to the same clients and communities).

Satisfactory portfolio performance. In terms of many of the key evaluation criteria used by the Office of Evaluation, the performance of IFAD-supported projects in Ethiopia has been better than the average for IFAD operations globally (see table 3). This is an achievement that warrants acknowledgement. In particular, performance has been good in areas such as small-scale irrigation, rural finance and pastoral community development, where IFAD operations have had an impact in terms of reducing rural poverty. Progress has also been made in the critical area of local governance. Performance and results have so far been more limited in the area of cooperative development and in the sphere of agricultural research, where any large-scale impact on farmers' livelihoods may not become apparent until after the relevant interventions have come to an end. This has also bee true with regard to the engagement of the private sector. Overall, sustainability prospects are good, partly thanks to the fact that the Government has allocated a sizeable share of its budget to agriculture and rural development.

Valuable innovations. Innovations have been introduced in a number of IFAD operations. Community-driven approaches have fostered local partnerships among the public sector, private enterprises and civil society. In the area of agricultural research, ARTP introduced competitive funding facilities which can also be accessed by private entities and non-governmental organizations. Participatory research activities with farmers and a system for linking up research, extension and farmers have been established. The Government of Ethiopia and some donors (the World Bank, the Japanese International Cooperation Agency) are extending further financial support for these innovations. In the area of rural finance, linkages between MFIs and banks have been facilitated, and rural savings and credit cooperatives (RUSACCOs) have been introduced. The replication and scaling up of tried and proven innovations have not received systematic or sufficient attention, however.

Opportunities for further improvements. Opportunities exist, for example, for bringing in computerized management information systems for use by MFIs and for introducing business development services for rural finance clients. In addition, IFAD has not taken full advantage of its grant programme in Ethiopia. The majority of the grant funds have gone to research projects that are not closely enough linked to the lending portfolio, while small grants that are tied into a given project have proved their validity for generating useful knowledge and piloting innovations.

Project design. Project design has generally been of good quality. However, in the case of agricultural research (ARTP), concerns expressed by the Technical Advisory Division about the project design were not fully responded to. The CPE finds that these concerns are still pertinent.

Supervision and implementation support. Under IFAD's traditional operating model, these functions have generally been outsourced to cooperating institutions. This is now changing, however, with the implementation of IFAD's new supervision policy. The evaluation considers this to be a good policy that is likely to enhance IFAD's development effectiveness in the country. Cases of complex and "heavily procedural" approaches to procurement have been observed in the context of ARTP and RUFIP which have caused delays and hurt performance.

Since its establishment of a country presence, IFAD has been becoming a more active partner. IFAD's field support officer is now participating regularly in supervision and implementation support missions, which contributes to better communication and knowledge management. IFAD's country presence is still limited, however, especially in terms of delegation of authority and the resources available.

Recommendations

Target food-deficit areas. Poverty rates are higher and development challenges are more significant in Ethiopia's drought-prone food-deficit areas. For IFAD, this constitutes a rationale for focusing on food-deficit districts and for supporting dynamic economic changes in the rural economy through trickle-down effects (e.g. through microfinance and support for the development of small businesses and microenterprises in rural areas), as well as for supporting the development of agriculture (irrigation) and livestock assets with the view to improving food security. In planning and implementing such support, attention should be devoted to identifying measures that will promote linkages between different subsector-specific programmes (for example, linkages between rural finance, on the one hand, and small-scale irrigation and agricultural marketing, on the other).

Build on successes. For the coming 10 years, priority should be given to areas in which IFAD has developed a lead position, such as small-scale irrigation and rural finance. However, in the case of rural finance, a second phase will depend on the findings of the interim evaluation to be conducted in 2009 and, in particular, on progress in addressing the current problem of negative interest rates. As the inflation rate is heavily influenced by macroeconomic aggregates which are beyond IFAD's control, IFAD needs to raise the issue. It is also important to discuss the realignment of the interest rates charged by MFIs in order to address the problem posed by negative returns on assets and equity. Support for pastoral community development, which was initiated by the World Bank, has been a success for which continued IFAD involvement seems justified subject to a continued commitment on the part of the Government of Ethiopia to the involvement of NGOs, communities and CSOs in local development planning. Currently, IFAD is participating in the formulation of an operation to support sustainable land-use management around Lake Tana, which, if approved, will open up a new strategic focus area for IFAD. Natural resource degradation is an issue that clearly warrants attention, but the strategy for dealing with this problem needs to be carefully worked out.

Use grants as a "smart" tool for knowledge management and the promotion of innovations. IFAD should increase the use of grants for preparatory studies, baseline surveys and impact studies, which could be outsourced to independent third parties (agencies other than the implementing institutions). This would be a more effective use of grant resources than free-standing research programmes that do not have direct linkages to the loan portfolio and lack mechanisms for dissemination to rural households. Grants should also be used for scouting, testing and promoting successful innovations.

Anchor the policy dialogue in IFAD operations. While project design and implementation offer opportunities for policy dialogue, supplementary activities (e.g. analytical work, workshops, attending donors' groups) at the national level may also be needed, and grant resources should be allocated for this purpose. In addition, well-targeted study tours for government officials should be considered an effective policy dialogue tool.

Continue current partnerships and intensify efforts to partner with NGOs, the private sector and bilateral donors. Opportunities exist for building partnerships among the public sector, civil society and the private sector at the regional and subregional level (as tested in pastoral community development operations). Such partnerships are particularly helpful in supporting water user associations in irrigation projects. Based on an early review of AMIP, the CPE recommends that ways and means be found to strengthen public-private partnerships. IFAD should also more actively explore partnership opportunities with some bilateral donors that are working in the areas of agriculture and rural development. Finally, lessons learned from experiences in rural finance indicate that efforts should be made to avoid complex cofinancing arrangements under which each organization applies its own specific regulations, especially in the area of procurement.

Strengthen the IFAD country office. Given the scale of IFAD operations in the country, the CPE recommends that the current country presence arrangements be strengthened. In this connection, consideration could be given to outposting the country programme manager.

---

1/ Mr Seydou Traoré, former Minister of Agriculture of the Republic of Mali.

2/ The three main themes discussed at the workshop were as follows: (i) Improving the IFAD Country Programme Approach; (ii) The development of a results-based management framework; and (iii) Towards better implementation capacity.

3/ Figures derived from the following sources: World Bank, World Development Indicators 2006; UNDP, Human Development Report 2005; Economist Intelligence Unit, Ethiopia Country Profile, 2006.

4/"Relevance" is defined as the extent to which the objectives of a development intervention are consistent with beneficiaries' requirements, country needs, institutional priorities and partners' and donors policies. An assessment of a project's coherence in achieving its objectives is also entailed. "Effectiveness" refers to the extent to which the development intervention's objectives were achieved, or are expected to be achieved, taking into account their relative importance. "Efficiency" is a measurement of how economically resources/inputs (funds, expertise, time and so on) are converted into results.

5/IFAD uses a six-point rating scale in which 1 represents the lowest score and 6 the highest. A 5-point rating is considered to be satisfactory.