Rationale and background. The People's Republic of Bangladesh has received 22 loans from the International Fund for Agricultural Development (IFAD) since 1979, more than any other country. It ranks third, after China and India, in terms of the total amount of the resources borrowed. A new Country Strategic Opportunities Paper (COSOP) for Bangladesh will be presented to the IFAD Executive Board in early 2006 in order to coincide with Bangladeshi government planning processes and its new National Strategy for Accelerated Poverty Reduction. Given the significance of the Bangladesh programme for IFAD and the timing of the upcoming new COSOP, this Country Programme Evaluation (CPE) in Bangladesh was a high priority for the Office of Evaluation (OE) in 2005.

The last CPE of Bangladesh conducted by the IFAD Office of Evaluation in 1994, covered the first 12 projects financed by IFAD between 1979 and 1992.  This 2005 CPE covers all projects that became effective from 1994 to 2004. Over the ten-year period, IFAD provided some USD 118 million of financing for a total of nine projects at a cost of about USD 393 million. The latter includes various co-financiers and the Government. As at 31 December 2004, five of the nine projects were on-going.

Evaluation approach. The overall goal set for the CPE was to learn, together with partners, from past experiences in Bangladesh and improve the future performance of this country programme. The specific objective was to assess results and impact of IFAD strategies and operations in Bangladesh between 1994-2004.

Current OE guidelines for CPEs were used as a framework and followed throughout the implementation of the evaluation. The CPE evaluated (i) the strategic dimension of the current IFAD country programme; (ii) the operational dimension of the IFAD country programme; and (iii) selected special issues identified in collaboration with the partners of the evaluation. The strategic dimension covers the design of the country strategy and the design of the programme put into place to realize that strategy. The criteria used were relevance and coherence.  The operational dimension covers the performance of the programme, including the performance of the projects, the performance of IFAD and the performance of partners.  The basic criteria used were effectiveness, efficiency and impact.  In addition, the overarching criteria of sustainability, innovation, replication and impact on gender were applied to the evaluation of the operational dimension. The programme is rated on selected criteria in order to summarise the analysis and standardise evaluation findings for future comparison with evaluations of other IFAD country programmes.

The evaluation process began in October 2004 with a review of documentation, followed by a preparatory mission to Bangladesh to consult with partners and establish a Core Learning Partnership (CLP) to guide the evaluation. An Approach Paper, circulated to partners, documented the evaluation process.  The methodology described in that Paper consisted of: (i) the OE review of all available reports and documentation; (ii) individual assessments of each project covered by the evaluation, including self-assessments by the on-going projects; and (iii) fielding of an Evaluation Team ¹ to verify preliminary conclusions and findings.  From these activities and sources, the Evaluation Team sought qualitative and quantitative evidence to indicate how well the programme met the selected evaluation criteria.

The Evaluation Report was provided in draft to partners for comment prior to its presentation, in collaboration with the Government of Bangladesh (GOB) and IFAD at a National Roundtable Workshop in Dhaka 24-25 July 2005 to discuss findings, recommendations and specific actions that could be taken to implement them.  Subsequently, the partners concluded an Agreement at Completion Point wherein the evaluation recommendations are recorded together with the follow-up actions that partners agree to take to implement them.

Country context. The Bangladesh country context is especially difficult due to very low income levels, exceedingly high population density, and environmental conditions prone to natural disaster.  Notwithstanding these formidable constraints, GOB has been successful during the period 1994-2004 in achieving several important improvements in country conditions. The country experienced strong and highly stable macro-economic growth. As external assistance fell, levels of trade increased. Remittances rose dramatically, increasing six-fold. Advances in human development included improved levels of health and education as well as improved social and economic conditions for women. The country also improved its capacity to manage natural disasters. However, these positive development achievements were contrasted by increased inequality in the distribution of income and slower rates of poverty reduction than had been achieved earlier.  Although GOB efforts at reform are underway, public sector bureaucracies still operate at relatively low efficiency levels and corruption continues to detract from the benefits of macro-level policies. 

GOB has actively pursued poverty reduction as the fundamental goal of its Fourth and Fifth Five-year Development Plans, spanning 1990-1995 and 1995-2000 respectively. It has continued to give primacy to this goal in the more recent interim Poverty Reduction Strategy Paper (iPRSP) of March 2003, an instrument that took the place of the five-year plan, and in the Poverty Reduction Strategy Paper (PRSP) now entitled "Unlocking the Potential: National Strategy for Accelerated Poverty Reduction", available in draft at the time of the CPE in February 2005. GOB has increasingly involved the NGO sector to complement its own investments and activities.  Bangladesh is known internationally for its successful development of the microfinance sector, including widespread outreach in rural areas and amongst women.

Evaluation of the strategic dimension of the programme. Considering the strategic dimension of the programme, the evaluation noted that from 1994-1999 IFAD had no explicit strategy for Bangladesh.  Operations were guided generally by the IFAD mandate to reduce rural poverty and the Fund's strong commitment to being responsive to the needs of its target group - the rural poor.

The evaluation found the implicit country strategy underlying the country programme from 1994-1999, based on the IFAD mandate, to be relevant to GOB policies and country conditions.  However, as the strategy was only an implicit one, its coherence could not be judged. Nor could the evaluation go beyond affirming that there was a general congruence between IFAD's broad aims and the needs and policies of the country.

In 1999 when IFAD prepared its first COSOP for Bangladesh, its central goal was to "promote self-managing grassroots community organizations that will create and sustain viable, cost-effective institutions and also empower the rural poor."  The COSOP also advocated targeting landless, women, small and marginal farmers, and the extremely poor, as well as vulnerable groups including indigenous people and charland dwellers. It called for support to production of higher value agricultural outputs in livestock and fisheries. The evaluation found that while the presentation of the goal and objectives in the COSOP was not very clearly put, it did have an underlying logical consistency. The main strategic thrust, the target group, the policy dialogue issues, the planned partnerships, the choice of sub-sectors, and the proposed pipeline of projects were well-aligned.   It was also relevant to conditions in country and to IFAD experiences, advocating a move to go beyond service delivery by GOB and NGOs, to the creation of sustainable community-based organizations (CBOs). 

However, the COSOP was fundamentally flawed because of the approach to CBOs that it advocated. The South Asia Poverty Alleviation Project (SAPAP) project funded by another donor that provided the model inspiring this COSOP goal was unsuccessful and eventually cancelled.  The one on-going IFAD project in Sunamganj that was subsequently designed using that same model is also now meeting with many institutional constraints. Although it is still too early to judge the Sunamganj project, the results to date are not promising.  Problems related to social cohesion and self-managed organizations were arising with the application of the model at the time the COSOP was drafted. IFAD was unrealistic about the associated risks and over-optimistic about the prospects of resolving them.

The programme put into place from 1994-2004 to implement IFAD's strategies had nine projects: seven were designed under the strategy that prevailed prior to the 1999 COSOP and two were designed subsequently.  Most of the nine projects were area-based and tended to include investments in agricultural production, infrastructure, and microfinance. Community development (including beneficiary training) and institution building (including project management) were also part of the activities financed in most projects. Within these general parameters, projects were distinguished by their own areas of special emphasis: fisheries, irrigation infrastructure, microfinance, community organizations, etc. The largest expenditures were for infrastructure (ranging from transport, to flood and drainage control, to aquaculture and irrigation structures, to buildings and markets). The next largest areas of expenditure were institution building and microfinance. Cofinanciers included the Asian Development Bank (ADB), the Swedish International Development Cooperation Agency (SIDA), the Danish International Development Agency (DANIDA), the World Food Programme (WFP), and the Government of Japan (GOJ).

The design of the programme, taking the set of all nine projects, was consistent with IFAD's implicit strategy during the 1994-1999 period, i.e., IFAD's basic mandate with its special  target group focus. It was less consistent with the strategy outlined in the 1999 COSOP, oriented toward the promotion of CBOs. To IFAD's credit, the design of the lending programme after the COSOP reflects a de facto adjustment in the approved strategy. The evolution and development in the design of the projects shows evidence of learning, albeit delayed, over time.  However, the programme was wide-ranging. There was no indication that projects were conceived to contribute to larger programmatic goals or objectives beyond those of the individual projects.

Overall, the relevance of the strategy and the actual programme put into place to implement that strategy was moderately successful. 

Evaluation of the operational dimension of the Programme. Judging from Project Completion Reports submitted by the Borrower, Self-Assessments done by the Project Coordination Units (PCU), and reports on project performance by the IFAD Programme Management Department, the projects generally performed well in reaching, and sometimes exceeding, their targets in terms of physical and financial outputs. For the most part, they achieved their immediate objectives.  They have reached about 131% of the number of beneficiaries originally targeted in project designs.  Counting just six of the nine projects, more than 3.6 million persons or about 7% of the rural poor have been reached in some way by IFAD-financed investments.  A few examples of the physical outputs providing benefits that reached them include: 200 000 hectares of polders protected; 1 200 km of improved roads; 71 new landing stages; 60 000 on-farm demonstrations; and, at least one loan for each of some 300 000 borrowers.  Overall, the programme was successful in terms of effectiveness.

Using available indicators of efficiency, including average cost per beneficiary, average time overrun, and time to effectiveness, the performance of this Bangladesh programme was somewhat less impressive.  Although average time overruns have gone down since the last CPE in 1994, most projects were inefficient in project start up, including the selection of financial institutions, conclusion of subsidiary loan agreements, selection of technical advisors and selection of NGO implementing agencies. This resulted from a combination of slow-moving government bureaucratic agencies and inadequate support from IFAD, including design documents with insufficient detail. Notwithstanding delays in project start up, projects still reached their physical targets.  In some cases, maintaining extensive beneficiary numbers meant a reduction in the intensity or in the level of support provided to each.  In microfinance, for instance, access to finance was achieved, but opportunities for repeated borrowing were reduced. The programme was moderately successful in terms of efficiency.

Using the IFAD evaluation framework, six domains of impact were considered: physical and financial assets, human assets (including access to health and education); social capital (empowerment, including gender issues), food security, environment, and institutions and policies. While it has not been possible to measure or quantify impact, the evaluation found indications of the kinds of impact made by the programme, the degree of impact and the extent or reach of impact, in terms of numbers of beneficiaries. On the whole, the programme had a successful impact.

Significant household-level impact on income and asset accumulation was reported for the majority of beneficiaries, including women and the poor. One study found a 39% increase in household income for NGO credit participants in the Netrakona Integrated Agricultural Production and Water Management Project (NIAPWMP). Around 90% of borrowers reported an increase in their income as a consequence of borrowing from Employment Generation Project for the Rural Poor (EGPRP). Among the more than 1 700 respondents in a survey of AqDP, the "vast majority of participants reported a moderate or better increase in family income so far". Most Small-scale Water Resources Development Project (SSWRDP) beneficiaries have bought land since participating in the project. Respondents to project-level surveys also reported road improvement and water management infrastructure as highly beneficial, with a positive impact on product marketing and other types of travel.

Projects have not explicitly sought to improve access to potable water, basic health and education, however, ADIP reported that 40% of all beneficiaries had increased access to drinking water by acquiring tube wells with the increased income they had as a result of the project. SSWRDP and AqDP respondents to the Independent External Evaluation of IFAD (IEE) surveys indicated that access to health and education services improved as the result of improved roads and transport services.  Changes in nutritional status were not monitored. 

The programme had a significant impact on gender, creating one form of social capital. At least half of all recorded beneficiaries in most projects were women, who acquired skills, self-esteem, income and improved social status. Other positive impacts accrued in the creation of groups of borrowers associated with microfinance activities in most projects.  However, not all projects were able to create the changes in social organization required at the village level to sustain benefits in post-project periods.  This issue, the weak point of the COSOP, continues to pose a challenge for sustainable local-level development.

Increases in income, together with more intensive production practices, especially poultry production and home gardening, have had a positive impact on food security. In surveys conducted for three projects (Agricultural Diversification and Intensification Project (ADIP), SSWRDP, NIAPWMP) most beneficiaries reported increases in the quantity and quality of food they consume.

Numerous public works sub-projects for flood control and drainage implemented with IFAD-financing had a positive impact on the environment. They helped to mitigate the environmental destruction, loss of resource productivity and human tragedy at times of natural disaster.  However, flood and drainage control schemes had some negative impact on the floodplain environment in terms of fish resources and biological diversity. Depletion of such common property resources particularly affects the rural poor.

The Country Programme impact on institutions and policies was achieved through the normal working processes of design and implementation of loans and grants. Through collaboration with PKSF, the IFAD programme made an impact on rules for lending to marginal and small farmers and on requirements for weekly repayments. IFAD likewise made an impact on nationalized commercial bank (NCB) relationships with microfinance NGOs. Through association with the Third Rural Infrastructure Development Project (TRIDP), IFAD made an impact on Local Government Engineering Department (LGED) policies and procedures with respect to facilities for women and guidelines for local government operation and maintenance (O&M) of infrastructure.  By linking NGOs and Department of Agricultural Extension (DAE) in the delivery of extension services, IFAD-financed projects had a positive impact on outreach and effectiveness.

 

The changes described as impact on policies and institutions above were innovative, and were replicated to some extent, and made sustainable. Those in infrastructure and agricultural extension were also picked up and replicated by other donors.  Some innovations in technology were achieved through agricultural research grants. However, given the importance that IFAD corporate policy attaches to innovation and replication, efforts to promote and replicate successful outcomes were limited.  Both grant and loan-funded projects missed opportunities to share information on results, foregoing potential benefits that could have accrued to those not directly associated with the projects. The Programme was only moderately successful in terms of innovation and replicability.

In terms of sustainability, the evaluation found reason for optimism where changes in practices have been accompanied by changes in institutions or policies. There is also reason for optimism with respect to the sustainability of benefits derived from the adoption of new technologies by farmers in a number of projects. However, the evaluation also found examples of cases where the probability of sustainability is low, especially where: (i) responsibility for infrastructure maintenance has been transferred by LGED to local stakeholders; (ii) financial services are being provided by NCBs and small NGOs; and (iii) access to water bodies has been granted to the poor in the context of a project. Overall, the programme was moderately successful in terms of sustainability.

Evaluation of IFAD's performance.  IFAD's performance was mixed and only moderately successful.  Initially, IFAD did not set to establish a strategic framework for providing its assistance, and when a strategy in the form of a COSOP was developed, results were disappointing.  IFAD sometimes failed to make realistic provisions to cope with recurring difficulties such as rigid government procedures in establishing and revising project implementation plans, relationships between projects and banks, and selection of technical advisors.  Performance in partnership building, resource mobilization, use of non-loan instruments and influence on policy was also weak.  However, performance did improve in recent years, especially in adjusting strategy and coping with corruption.  There was also some improvement in portfolio performance management, implementation follow-up, and project design towards the end of the period under review. 

Evaluation of the Government's performance. The evaluation found Government performance was also mixed and similarly rated it as moderately successful.  At the macro-level, the Government can be credited with high performance levels and orientation of expenditures to create favourable economic and social conditions for poverty reduction. As a result, the positive trends in production and income observed at the household level amongst project beneficiaries were certainly facilitated. However, excessive bureaucratic delays at the project level were very common. Central government sometimes showed limited willingness to fulfil commitments as, for example, in issues related to financial services and access to common property resources in EGPRP, ADIP, AqDP and the Sunamganj Community Based Resource Management Project (SCBRMP).

Evaluation of the CIs' performance. The United Nations Office for Project Services (UNOPS) and the ADB were successful in the performance of their responsibilities.  In addition, relationships with cofinanciers were productive. In the case of both WFP and ADB, collaboration added value to the IFAD programme in terms of approaches and funds.  However, differences in programming approaches remain a constraint to collaboration with some agencies, such as WFP. 

Evaluation of NGOs' performance. The performance levels of the NGOs involved in the programme tended to be lower than that of the larger NGOs, including Bangladesh Rural Advancement Committee (BRAC) and the Grameen Bank, with whom IFAD has collaborated in the past.  Those who lacked their own development programmes or strong links to the region or sector where they were operating did not often do well.  The performance of National Commercial Banks that participated in IFAD projects was also poor in many instances.

Selected special issues. On the special issue of the role of the private sector in agriculture, the evaluation sought to answer two questions posed by partners: Which activities have been most successful in promoting private sector development? and What has been learned about the roles of the private and public sector?. During the period under review, policy reforms were the public sector initiatives that were most powerful in promoting private sector development in agriculture. Government stimulated private sector growth by abolishing restrictions on sales of shallow-tube wells, reducing tariffs on diesel engines, and liberalising the retail fertilizer market. 

The most successful IFAD-supported public sector activities to promote the private sector were  the investments that supplied capital in the form of micro-credit lending by NGOs, combined with technical advice through training to the self-employed. These fostered increased production of poultry, homestead garden output and small income-generating activities such as rice trading. Follow-on effects of these public sector investments in rural finance and technical training were increased purchases of inputs and sales of outputs by private producers. These, in turn, increased the number of businesses trading and marketing inputs and outputs.

Through the study of the IFAD-financed projects during the period 1994-2004, the evaluation found the role of the public sector in poverty reduction to be a very important one, but one that has not been well defined or articulated. In particular, its role in supporting agricultural production must be re-examined. At the same time, the role of the private sector is expanding and potential for collaboration between the public and private sector is good.  This potential could be tapped to the benefit of the rural poor.

Considering the special issue of microfinance and employment generation, the key question was: What has been learned with respect to the promotion of self-employment, employment and entrepreneurship for the poor and poorest? The evaluation confirmed that in IFAD-financed investment projects, as elsewhere in rural Bangladesh, microfinance was an excellent means of promoting self-employment for the rural poor.  In particular, as conceived and delivered largely through NGOs, it was generally well suited for the poor who were landless: those with less than 0.2 ha of land, who represent a total of about 10 million households or 52% of the rural population.  However, microfinance has been less suited to the poorest amongst the landless, those with less than 0.02 ha of land, who are estimated to be some 6.4 million, an important component of the landless group.

The IFAD programme was most successful in reaching the poorest when it provided them with direct employment.  It often did so as the means to other ends that were being pursued, as with road construction. Project-built roads lowered transport costs and increased access to markets for the poor in more remote areas. Hence, there was a dual benefit. The employment had a poverty reduction effect, and so did the output resulting from the employment.

The EGPRP project was also designed to generate employment for the poorest by financing microenterprises of self-employed people and small entrepreneurs. The project was successful in delivering financial services to small-scale entrepreneurs – a new borrower group.  Partner lending institutions acquired new skills and new customers, while borrowers acquired access to formal financial institutions. However, the employment-generating effects of this lending were not as high as expected. In particular, employment generation effects were lower when loans were made to existing enterprises, rather than new ones.  Greater care in borrower selection and more attention to the design of the financial services would be required if employment generation for the poorest is to be the most important project outcome.

On the special issue of Government and NGOs, the key question posed was: What were the best arrangements for working relationships between NGOs and Government? Relationships with the Government worked best in IFAD-financed projects where NGOs functioned as intermediaries between government extension services at the lowest tier and the farming community.  This was especially true where the NGOs were real advocates for the communities they linked to Government and not just providing services for a fee paid to them by the Government. Yet, overall, the relationship prevailed was one where the Government was the contractor and the NGO was the service provider.  Limiting the role of the NGO to the subordinate one of service provider deterred at least one larger and stronger NGO from seeking to participate in IFAD projects.

Criteria and procedures for selection of NGOs were not well defined during much of the decade, resulting in the inclusion of some sub-standard NGOs.  This has changed with respect to the selection of NGOs for microfinance services, now done through PKSF.  However, there is still a problem with respect to partnership relationships with NGOs charged with forming groups and providing technical support to beneficiaries. These could be improved if the parameters of the partnership relationships, the selection criteria and the procedures were better defined.

The last special issue was related to infrastructure, the investment area that absorbs more financial resources than any other. The question was: What has been the importance of investments in rural infrastructure for poverty reduction? The evaluation noted that infrastructure investments in this programme were about USD 112 million, or 28.5% of total investment. The infrastructure development resulting from the use of these funds made a very substantial contribution to poverty reduction.  It stimulated increases in farm outputs and decreases in farm input costs, primarily as a result of improved transport facilities. It improved mobility, which broadened job opportunities and enhanced access to more remote communities by government, NGOs and the private sector. It improved food security with additional irrigation capacity facilitating higher, more diversified and intensified crop production. It also provided several million person-days of employment to poor people.

As most infrastructure investments were in public goods, such as roads and markets, they did reach even the poorest.  Although those investments also benefited better-off households, the poor tended to benefit more, in proportion to their income, from cost savings and value generated by access to new services and facilities. The poor could directly access, and benefit from, those public goods without depending on the skills or good will of staff responsible for administering projects.  Another advantage of these investments in public goods, in terms of targeting, was that even when benefits did accrue to affluent households that did not preclude the poorest from benefiting. Nor did it diminish the value of benefits available to the poor, whereas, when projects provided benefits that took the form of private goods like microfinance, irrigated land or training, the less poor were better able to capture those benefits and the poorest tended to be left out.  

Conclusions.  The Programme as a whole achieved the strategic objectives prevailing prior to the COSOP as follows.  The objective of expanded food production was met in the areas where it was an IFAD objective. Although expansion can be attributed in part to key policy reforms and positive social and economic conditions, IFAD-financed training combined with microfinance and infrastructure, including irrigation, flood control, and drainage also contributed to this outcome.

Only two of the nine projects had objectives that specifically targeted improved nutritional status and none monitored it as such.  Therefore, it is not possible to determine whether nutritional status actually improved and whether IFAD achieved its strategic objective in this area.  However, food security measured in terms of the frequency and quantity of food consumption improved due to increases in income or in agricultural outputs.

IFAD did influence some working policies and institutions in terms of procedures and approaches. However, overall IFAD-financing had a greater impact on delivering benefits directly to target group households than on the development of policies and institutions that could favour sustained future benefits to the poor.

IFAD experienced difficulties in improving the conditions of the poorest. A number of efforts were made especially to reach out to these "hardcore poor". Projects sought to employ the extremely poor directly, they supported businesses that would hire them, they enabled farmers to expand employment, and they made efforts to provide microfinance services so that the poorest of the poor could start their own economic activities and become self-employed. The most widespread benefits to this group came from infrastructure investments, through direct employment, the income effects of lower cost transport, and better access to markets and social services.

With respect to the strategic objectives of COSOP, there was little evidence of achievement in the promotion of self-managing grass-roots organizations, even though this was the central goal of the COSOP and an important objective of a number of projects. Fundamental questions of unequal social and economic relations in local communities still pose constraints that overshadow many community development efforts.

IFAD was partially successful in reaching target groups it identified in the COSOP, i.e., women, indigenous people and charland dwellers, landless people, small and marginal farmers and the extremely poor.  It succeeded in reaching women through microfinance, training, infrastructure and access to markets.  It also reached some of the landless, the mainstream poor, through microfinance. But it only partly succeeded in reaching further down to the "extremely poor", as noted above, or further up to the small and marginal farmers. Only one project made efforts to reach indigenous and charland people, with very limited success so far.

On the whole, the Programme achieved its objective to focus on fisheries and livestock, areas that are very relevant to the landless. Together with homestead fruit and vegetable gardening activities also included in many projects, they have provided viable income-generating opportunities to the mainstream and extremely poor.

Further conclusions were drawn on two key strategic directions for IFAD. The first concerns the delivery of microfinance. IFAD broke new ground working with rural microenterprises in EGPRP and it has a comparative advantage in future efforts to test and improve the models it launched to support rural micro-entrepreneurs that are not currently targeted by other programmes.  Small and marginal farmers with land holdings of 0.20 to 1.0 ha, many of whom are below the poverty line, represent another important group in terms of microfinance services.  They have largely been left out of the microfinance revolution in rural Bangladesh, partly because of the special nature of agricultural credit. Yet they require working capital to improve productivity and funds for investments to adopt even the most basic new technologies.

The second concerns rural infrastructure investments that bring advantages on several counts.  They proved to offer an excellent way to reach the poorest. Using labour-intensive approaches, they generated employment. Increasing "connectivity" they contribute to social and economic integration. Future investments in this area are likely to yield high benefits provided that weaknesses observed in the selection of sites, in beneficiary involvement in design, and in post-project O&M issues are addressed.

Other important conclusions concern those of the private sector and the use of IFAD loans. Funds have been used to directly assist private sector producers, especially the self-employed, through microfinance. They have also been used indirectly to stimulate growth in the private sector through investment in the institutional capacity of the public sector to provide support services and facilities. However, IFAD has not gone beyond these traditional approaches to look for other ways to access the potential of dynamic private sector operators for rural poverty reduction. IFAD has the potential to assist GOB to do this.

The NGO ‘sector' is very important in Bangladesh in terms of numbers, field presence, resources, and influence. Given that the new PRSP advocates collaboration with NGOs, they are likely to be an important participant in poverty reduction efforts for some time to come. But IFAD still lacks a clear vision of what kinds of partnerships, and for what purposes, it would like to create with NGOs, using them most often as suppliers of services.  Potential benefits may have been lost as a result of inattention to developing relationships with these important actors.

IFAD is a relatively small player that has tended to work somewhat in isolation, dedicating more attention to achieving impact than to sharing results with others, creating the conditions for them to be sustained, or promoting their replication.  It has not developed strong partnerships or mobilised significant levels of resources. GOB returns to investments made with IFAD funds could be higher if more effort was made to analyse, document and communicate acquired knowledge and experiences with others.

Development assistance projects can offer opportunities for misappropriation of funds and rent-seeking behaviour on the side of public sector and NGO staff involved in the delivery of benefits and services. Over-budgeting, for example, is one way that externally-financed projects unwittingly encourage this. Similarly, projects have offered the more powerful members of rural communities' opportunities to capture benefits improperly. Projects designed with unrealistic assessments of community relations feed these negative tendencies.  IFAD has a responsibility to contribute to GOB efforts to improve governance and combat corruption by improving its own practices.

Most partners, including government, donors, and NGOs prefaced their remarks about IFAD with observations on the constraints they face in the relationship due to the lack of an IFAD office in the country.  Development agency and government staff who has had some contact with the IFAD consultant who serves as a liaison officer in Dhaka indicated that his presence was very helpful.  However, this presence has not been used to support on-going projects, implementation of the programme, or formulation of new investments and activities. All these areas could benefit from a more constant backstopping and on-going dialogue with IFAD.

RecommendationsBased on its evaluation of the 1994-2004 Country Programme in Bangladesh, the CPE made eight recommendations. One covers the quality of the future strategy document.

Set clear strategic goals and specific attainable objectives

IFAD's next strategy for Bangladesh should clearly describe desired results and impact.  It should identify a limited number of specific objectives that can realistically be achieved with the available resources and within the time period foreseen by the COSOP.

To cover major substantive areas where it has already acquired experience to then build upon and deepen the related policy dialogue, IFAD should build partnerships and disseminate knowledge.  This would enhance results and impact while developing sustainable approaches for replication and up-scaling by the GOB. These would also serve to define IFAD's strategic niche and be main thrusts in the future strategy.

Development of financial services to microenterprises and small and marginal farmers

IFAD should continue its important new work in developing financial service providers and products for rural microenterprises and agricultural production.  It should lead the way in developing services in these areas, new frontiers for microfinance in rural Bangladesh.

Investment in infrastructure to provide economic benefits to the rural poor and employment to the poorest

Investment in infrastructure for transport, especially union-level roads, water control structures and social and administrative infrastructure should be continued as a means of providing benefits to the poor, especially better access to markets, to goods and services and to natural resources. Direct employment programmes should be used wherever possible to benefit the poorest.

Five further recommendations cover ways and means that operations could be improved to enhance the overall quality of the results and impact of investments financed by IFAD in Bangladesh. These are:

Build partnerships to tap private sector know-how, networks and resources

IFAD should work with its public sector counterparts to help stimulate the development of the private sector, particularly the participation of poor small-scale producers that make up the IFAD target group in that sector. IFAD should also help to forge partnerships with selected private sector operators to tap their know-how, networks and resources. Means to attain the latter could include: briefing of private sector suppliers of inputs and services on planned project activities to make them aware of input and output marketing opportunities; contracting larger private sector operators to furnish technical assistance and training in areas like seed supply; undertaking joint applied research projects on constraints faced by the target group that might otherwise be neglected by the private sector.

Set principles and procedures for NGO partnerships

IFAD should determine what roles can best be played by NGOs in association with the IFAD programme in Bangladesh and what results and impact that can be obtained through partnership with them.  IFAD should study the options and establish principles concerning what type of partnerships would be most beneficial. Adoption of clear procedures for NGO selection is essential.

Promote development of knowledge and communications in all projects

IFAD should devote more resources to developing and communicating the knowledge it acquires in its programme. There is wide scope for disseminating knowledge at all levels, from farmer to minister, and across several domains, from government to donor agencies to NGOs and the business community.  An explicit strategy for doing so in a targeted way so as to contribute to overall programme goals and objectives must be drawn up. Specific activities and investments should be designated to implement that strategy.  

Reduce opportunities for corruption in relation to projects

IFAD should do more to reduce opportunities for corruption. Two steps that will help are recommended above: better procedures and criteria for selecting NGO partners, and the establishment of communication components to disseminate information to the public. In addition there should be: (i) revisiting of cost estimates in design and budgetary allocations in implementation; (ii) more vigilance in obtaining compliance with existing reporting requirements; and (iii) full transparency in all financial, procedural, administrative and technical information related to project design and implementation.

Establish a permanent field presence in Bangladesh

IFAD should establish a formal presence in Bangladesh to improve efficiency in communications and dialogue, build closer partnerships, create stronger strategic alliances, improve policy dialogue, project design, supervision and implementation follow-up.  IFAD should take into consideration both strategic and practical factors to determine the type of presence it requires, the level, and the resources needed.

---

 1/ The evaluation was conducted under the overall responsibility of Ms Chase Palmeri, then Senior Evaluation Officer, OE. The CPE team comprised:  Mr Elliott Hurwitz, Team Leader; Mr Dewan Alamgir, Microfinance Specialist; Mr Charles Bevan, Agriculture and Rural Development Specialist; Mr Salah Rouchiche, Infrastructure Specialist; Mr Seraj Uddowla, Civil Engineer; Mr Sajjad Zohir, Institutions Specialist.