Completion evaluation

Country background

Physical and social features. Belize is a country of 22 965 km², located in Central America, bordering the Caribbean Sea, between Guatemala and Mexico. The population of the country stood at 291 800 inhabitants in 2005, up from 211 000 in 1994, with one of the lowest population densities (12 inhabitants/km2) in Latin America. The country is divided in six administrative districts: Belize, Corozal, Orange Walk, Cayo, Stann and Toledo. During last decade, the country has been struggling against the severe damages of several hurricanes: Mitch (1998), Keith (2000), Chantal and Iris (2001) causing considerable impact on socio-economic assets and development.

Rural poverty. In 2003, Belize ranked 91 on the global Human Development Index (HDI) scale, i.e., in the mid-level range of countries with medium human development. According to the 2002 Poverty Assessment Report, 24.5 per cent of households live under the poverty line.  Poverty has a much greater incidence in rural areas (33.7 per cent) than in urban areas (17.2 per cent). Children are more likely than any other population group to suffer from poverty (39.0 per cent), especially in rural areas (51 per cent) than urban areas (27 per cent), and for indigenous Mayan children (83 per cent). The Government of Belize (GoB) continues to put the primary focus of its strategies on the fight against poverty – foremost through the education, health and productive sectors and with the southern districts as a key focus.

Small farmers. The majority of small farmers¹ mainly produce maize, beans and rice. In remote rural areas, small farmers practice a traditional shifting (slash & burn) cultivation system called Milpa which contributes significantly to domestic food production. Milpa cultivation is predominantly found in indigenous communities in the southern districts. In response to increasing population pressure, the Milpa system has been evolving towards a more sedentary pattern in recent years, using a combination of perennial crops and food crops. At the time of project design, there was virtually no access to credit for small farmers in the project area. The public and private sector commercial banks were not interested in the risky agricultural productive sector or in costly small rural clients, whereas credit unions (CUs) were providing services in some rural areas.

The project

The Community-initiated Agriculture and Resource Management Project (CARD) was initially conceived as the second phase of another IFAD-funded project, the Toledo Small Farmer Development Project (TSDF). The performance of the TSDF was assessed as moderately unsatisfactory² but due to the consistently higher rates of poverty in the southern districts of the country, and the emerging benefits in agricultural productivity, a second phase was recommended. In view of the lessons drawn from the TSDF evaluation, a more participatory, bottom-up, community-led approach was designed to enable communities to decide on their own development alternatives and opportunities. During project appraisal, the new project was renamed as the CARD project rather TSDF Phase II, to reflect the substantial design changes.

CARD proposed a more integrated and participative approach for the rural poor and the indigenous communities to become involved in their own development. The overall goal of the project was to reduce rural poverty in southern Belize and raise production, employment and income of the rural poor. The purpose was to develop the productive potential of balanced sustainable land use systems and ensure support services to poor smallholder families in the southern region. CARD became effective in June 1999, under a co-financing agreement with the Caribbean Development Bank (CDB). CDB was also designated Cooperating Institution and assumed major responsibility for project supervision. The Ministry of Agriculture and Fisheries (MAF) was the Executing Agency (EA). The Development Finance Corporation (DFC) acted as the financial agency for the implementation of the Rural Financial Services (RFS) component. At the time of closing in 2006, the IFAD loan had only reached 58 per cent disbursement equal to the Special Drawing Rate (SDR)1 017 000 compared to the approved loan amount of SDR 1 750 000³ .

Project area and target group. At IFAD appraisal, the original target area was defined as rural communities in Toledo and Southern Stann Creek districts. A phased approach for targeting of specific communities was proposed, with an aim to commence a pilot process in eight communities and gradually increase by ten communities per year to reach 54 communities ⁴ selected among the poorest sections of the population. Targeted households were classified according to three groups:

1.  smallholders and artisanal fisheries households
2. subsistence Milpa farmers and
3. women from smallholder families. An additional emphasis in targeting was to seek to address the needs of the indigenous Mayan communities who comprise 11.9 per cent of the population in Stann Creek and 65 per cent of the population in Toledo.

The evaluation. The objectives of the evaluation were to assess the project in terms of: (i) overall performance; (ii) rural poverty reduction impact; (iii) innovation and sustainability; and, (iv) the performance of IFAD, CDB, the GoB and other implementing partners. The evaluation was conducted as a collaborative process by IFAD and CDB⁵

As there had never been a functional monitoring system, the evaluation was constrained in generating quantitative assessment of project performance, despite benefiting from the comprehensive and realistic Project Completion Report (PCR)⁶. Consequently, the evaluation conducted assessments on three levels to cross-verify the minimal secondary data as follows:

1. assessment of a representative sample of community level activities⁷
2. organizational level assessment
3. and project level assessment, covering policies, partnerships, management and operations. Stakeholder meetings were conducted with key project partners such as Government of Belize (GoB) agencies, CDB and IFAD's Latin America and the Caribbean Division (PL), the Toledo Development Corporation (TDC), local and national NGOs, as well as with direct project participants in the targeted communities.

Project performance

Design features. There was an in-depth assessment with the communities and with potential development partners during the design process to identify a more relevant and appropriate project design. The design incorporated many of the lessons from the TSFD⁸ . CARD proposed a more participatory approach for enabling the communities to draw better benefits from development opportunities, while at the same time safeguarding cultural identity and increasing involvement in national development processes. The CARD design recognized that technical support for agriculture development was insufficient to address the complexity of poverty. An approach was required that acknowledged the interplay between resource management, production systems and culture. The project financed four components: (i) Community Promotion and Local Organizations Strengthening; (ii) Technical and Marketing Support Services and Small-scale Infrastructure Investments component; (iii) Rural Financial Services (RFS); and, (iv) Project Coordination to manage the overall programming and activities.

Changes during implementation. The essential elements of the design did not change during implementation but the implementation mechanisms were continually adjusted. The fundamental change in the project implementation was that the recommendations to appoint appropriately skilled staff to the Project Management Unit (PMU) did not eventuate.

As a result of delayed implementation and poor results, by the 2002 mid-term review (MTR), a major project restructure was required. Following continued slow progress, a decision was made by the Project Steering Committee (PSC) in early 2004 that the focus of the project should not be on attempting to implement the project as designed but rather to focus on limited activities that had been showing potential for success. This led to virtual cessation of work within the communities and a focus on activities with producer organizations and the two participating credit unions. A project extension⁹ was proposed by IFAD in 2005 and was considered feasible¹⁰ but unfortunately, due to the fiscal constraints of the GoB at that time, and the concerns over the poor performance of the project, the extension was not approved by GoB. In consequence, the IFAD financing was closed by the original closing date of June 2006.

Implementation and outputs

Community development and participatory planning. Activities were meant to strengthen local leadership and organizations and identify social and economic constraints to development of communities. Participatory project planning and management skills were introduced to improve the capacity of local organizations to plan and manage their own priority sub-projects.  Strong interest was showed by community members at first. Participatory planning processes were well received, but few plans led to implementation of the sub-projects identified. Financing was provided by CARD for only 59 sub-projects¹¹ across the project area, most of which were not identified by the communities and/or which were too small to achieve the desired results (community sub-projects provided tended to be standardized by the project, e.g. three pigs per community, three sewing machines for women's groups, etc). The evaluation found very few examples where the community sub-projects had been implemented in accordance with the wishes of the community. Even training was mainly conceived as one-off, top down, educational activity. Conversely, in the isolated cases where community priorities had been supported, the results were very positive, with groups demonstrating improved financial and organizational capacity, as well as recording increases in their enterprise activities.

Technical services and marketing assistance. Based on the community plans, extension and technical assistance for improving technology on existing farms or diversification into new crops or non-agricultural activities were to be planned. Small-scale infrastructure was to be supported to improve farm access, storage and post-harvest processing. Results were mixed. The assistance for coconut growers and honey producers was initially successful, but eventually failed due to crop disease and other technical problems. CARD's most successful work was with the Toledo Cacao Growers Association (TCGA), a Mayan farmer's organization that produces organic cacao beans under a guaranteed rolling contract with Green and Black (UK-based customer)- which has also invested heavily in development of the Toledo cacao industry. As a result, 300 000 cacao grafted plants were transplanted, expanding the total acreage by about 800 acres. TCGA has expanded from 130 to 937 members. Late in the project, other similar initiatives were planned with the Livestock and Grain producers associations but these were not able to commence implementation before the end of the project. Infrastructure activities were largely confined to installation of culverts on a Ministry-led process of installation that involved minimal participation by the communities. The first installations were successful but the second batch has not yet been completed.

Rural financial services. The RFS component was designed to focus on small, short term loans, based on group guarantee, savings and other forms of innovative collateral. A credit fund was to be provided to the credit unions (CUs) as intermediary financial institutions (IFIs) for retail lending. An institutional fund was designed to strengthen the capacity of the IFIs to operate micro credit schemes. Several factors delayed the start-up of component activities including late hiring of the Rural Finance Expert (RFE) and the lack of eligible IFIs willing to operate in the project area. Negotiations with the two prospective CUs identified during appraisal failed. Eventually, two small local inexperienced credit unions: the Citrus Growers Credit Union (CGCU) and the Toledo Teachers credit union (TTCU) were accredited under relaxed criteria. There were also product design features that took a long time to resolve. Yet, the CUs managed to approve 1 056 loans amounting to BZ$1 478 193. The majority (62 per cent) of Toledo Teachers' loans went to agriculture, while nearly half (46 per cent) of Citrus Growers' loans were for commerce or retail trade. Loan delinquency is currently high in both IFIs with a portfolio at risk ratio of about 20 per cent, for Toledo Teachers and 33 per cent for Citrus Growers in December 2004, compared with the target ratio of = 12 per cent at appraisal. Portfolio at risk ratios are reported to be decreasing as the CUs consolidate their programs. Loan delinquency is mainly caused by external factors such as delays in payments of farmers by the Belize Marketing Board. However, both CUs acknowledge that they lent out loans too fast, too soon.

Project coordination. The design of CARD as a rural development project covered a broader scope of activities than was usually handled by MAF. For this reason, it was established a separate PMU that reported to MAF and which was largely autonomous in operations. In implementation, the PMU struggled to fulfil its role. The ‘innovativeness' of the project's approach made it difficult to find staff with the adequate experience. Staff members were not provided with basic, practical tools for implementation. The planned, gradual phased approach of the design was never implemented, nor was the plan to intensively support the communities on an on-going basis throughout the project period.

On the contrary, the project approach was fragmented, responding to isolated requests for support for vehicle use, or small non-viable sub-projects. The highly stratified administrative set up under the MAF, used by the project was at odds with the original design and made decision making slow. Procurement processes were a consistent issue with the project, in terms of overly bureaucratic documentary requirements, practical difficulties in complying with requirements due to low availability and quality of available suppliers and service providers. Poor communication processes between partners was also a continuing issue. The internal monitoring system was unreliable and incomplete for all stages of the project and the limited project data has constrained evaluation of project performance. Operations improved substantially late in the project when a Project Director (PD) was recruited who had project management experience and who was able to manage the coordination and communication mechanisms of the PMU more effectively.

Assessment of project performance
relevance

The design of the CARD project was found to be relevant to the target group and the Belize National Poverty Elimination Strategy focusing on the poverty "hotspot" of the Toledo District; supporting ecologically sustainable agricultural practices and addressing concerns with the traditional Milpa agricultural system. The focus on areas of highest poverty, on small farmers operating at subsistence level and indigenous people was also consistent with the policy directions of the GoB, the IFAD Strategic Framework and, retrospectively¹² , the IFAD Regional Strategy for Latin America and the Caribbean. The design project responded to the stated interests of the communities that had participated in the design phase. The community-initiated approach allowed the communities to have a say in their own development and was designed to build from the activities of previous projects. The issues with the results of the project design are more in relation to underestimation of the capacity for translating the design into implementation. Similarly, some sub-projects were supported by CARD that were not relevant to the communities, but this was due to operational deficiencies rather than a lack of relevance of the project per se.

Effectiveness

There was potential for the project to be effective but the potential was not realized for most initiatives within the project period. At the community level, based on the sample of project activities reviewed, the effectiveness was minimal. Feedback from the communities was that CARD wasted their time with unproductive meetings, did not respond to their articulated needs and provided small, inconsequential sub-projects that made no difference to their living conditions. This is confirmed by the sparse project information showing that CARD financed small-scale, fragmented activities over a large number of communities rather than following the recommended phased, intensive interventions in a gradually increasing number of communities. In contrast the interventions at the organizational level, by the end of the project, had proved more effective. Credit unions have grown and expanded their activities and are showing signs of combating their issues with high arrears. The support for the TCGA has been very effective and has been the main contributor to progress towards achieving CARD's development objectives. Hence, CARD is assessed as moderately ineffective.

Efficiency

As had been identified during the project design process, human resource issues were a major risk to efficiency. Experienced and capable staff is known to be difficult to attract and retain in projects in southern Belize. The risk became reality in implementation. Delayed recruitment of staff, poor installation of basic procedures and lengthy procurement procedures meant that the project started badly and was never able to recover. Most major operational issues were not resolved until late in the project period. After the MTR, substantial efforts were made by all partners to improve efficiency by provision of short-term training and technical consultancy. However, few results were evident from the support provided apart from the increased performance with the credit and cacao development activities. The long periods without key staff such as the PD, Accountant and Monitoring and Evaluation Officer meant that internal project processes were inadequately installed, new staff members were insufficiently oriented, vital management information was not generated, and decision-making was not strategic. Loan disbursement only reached 58 per cent by time of loan closing. Overall economic analysis for the project is not possible in relation to the community sub-projects, or the economic results of the rural finance as sufficient data is not available. However, analysis of the cacao production suggests a positive economic benefit of over 40 per cent return on investment (ROI) given the prevailing production system, levels and market. However, even though the cacao production strongly contributed to the success of cacao project, success cannot be solely attributed to CARD. Also, by the time of evaluation, this potential economic benefit has only been achieved by a limited number of farmers. The estimated average costs per beneficiary are in line with other IFAD projects (US$387) but the operating costs to total investment are high (43 per cent of loan amount). Therefore, the project is rated as moderately inefficient. It is likely, given the improvements later in the project, that if an extension had been granted, a greater level of efficiency would have been achieved.

Performance of partners

CARD had a complex stakeholder mix. All three financing partners must share a level of responsibility for the divergence of the project operations from the design and the resulting poor performance. Much valuable project time was lost at the project management level in delayed recruitment, in unclear procedures and in insufficient orientation and training. Unclear definitions of roles and responsibilities and lack of clarity in communication between the partners was a continuing issue.

Performance of IFAD

IFAD played the predominant role in the design of the project but did not sufficiently follow up the issues in translating the design into implementation early in the project. The elements of policy dialogue identified in the design were not progressed in implementation. Given the known lessons from the TSFD, and the re-emergence of the same issues, as well as the need for IFAD to take responsibility for the designs proposed, IFAD's performance is assessed as moderately unsatisfactory. IFAD responded well after mid-term when the need for more intensive support was identified and the technical assistance (TA) for the microfinance activities was helpful. However, the TA provided for the staff training and Monitoring and Evaluation (M&E) was of limited effectiveness due to lack of follow-up to identify whether the theory was being put into practice.

Performance of CDB

CDB acted as the IFAD cooperating institution¹³ for the whole project period. CDB financial processes were conducted in accordance with the agreed procedures and financing flowed effectively once the required documentation had been submitted by PMU. However, those procedures were found to be unduly complicated and bureaucratic for the small level of financing required for individual contracts. 

Insufficient orientation was provided to the incoming staff. As a result, procedures were not adequately understood or followed. This proved a major bottleneck in expediting implementation and hence performance is assessed as moderately unsatisfactory. After MTR, CDB also responded by providing a more intensive support with improved processes; and by reducing delays of payments when cashflow was constrained, through the provision of some direct payments to suppliers.

Performance of GoB

The GoB was effective in complying with most of the provisions of the loan agreement except for effective recruitment processes. In terms of financial commitment, allocation of the agreed counterpart funding during the initial stages of the project was satisfactory. Yet, the performance of the PSC, under the Chair of MAF was patchy. At times, the PSC was seen by the PMU as attempting to micro-manage the project¹⁴ . At other times, MAF did not call a meeting for close to 12 months, between year 2003 and 2004 and again with a large gap in meetings towards the end of the project, with inconsistent attendance from representative agencies. The PSC was responsible for the unsatisfactory staffing performance of the project, including appointment of inexperienced PDs for the majority of the project period. The government improved the policy and regulatory environment for microfinance. As a result, the performance of GoB was also considered moderately unsatisfactory.

Performance of other partners

Other partners. The relationships with services providers in the early stages of the project was not adequately handled by CARD, nor adequately supported by the three financing partners. Service providers, in the main, fulfilled the requirements of the terms of reference but those requirements were rigid and not always in the best interests of the communities or groups. Yet, in the later stages of the project, the partnership between service providers and producer groups was considered to be instrumental in achieving substantial benefits. The communities themselves showed tenacity in endeavouring to become involved with the project but frustration and cynicism set in, when support for priorities was not forthcoming. Local organizations were largely by-passed in project activities.

Project impacts

Rural poverty restriction impact

In reviewing the impact of CARD, there are several over-riding factors that need to be highlighted. The overall impact at the community level was very poor. The expected impact in terms of rural poverty was not achieved, and in some cases, negative social impact such as failed community groups, disillusioned leadership, frustration and confusion were noted. The few community-based activities that have been successful are those that are closely linked to the culture and to income generation, particularly the deer dancers and the community-based natural resource management activities. The three impact indicators related to social, human, gender and institutional developments are rated as moderately unsatisfactory. Conversely, activities that were more intensively supported at the latter stage of the project through the rural finance activities, or producer organizations such as the cacao production and honey production generated more benefits. The four indicators related to economic development are rated as moderately satisfactory. Consequently, overall the project impact has been rated as moderately satisfactory.

Physical assets have been improved through the road infrastructure improvements but outcomes are limited. The project made minimal contribution to improve food security, despite interests of community to strengthen and diversify food sources. Social capital development through training and skills development was exceptional in the case of the organic cacao industry where traditional knowledge and technical skills combined to generate a quantum leap forward in industry development.

Other aspects of training and skills development were dismal. Training was inappropriate, inappropriately delivered and in the main, was not followed-up or assessed. Activities designed for institutional development only showed noticeable outcomes for the TTCU and the TCGA. Both organizations had long-term support from the project and this is clearly displayed in the robust growth of both organizations. Human capital improvements were again achieved mainly in the case of rural finance and the cacao industry.

Agriculture development impact in terms of basic crop production, has been low in comparison to expectations. Yet, the commercialization of the cacao production has had a very positive result in encouraging agricultural production on individual farms that is consistent with the culture and current production systems as well as being commercially viable. Financial assets have been strengthened through the increased access to rural finance and as a result of the income generated through cacao production. Access to markets has also been increased, particularly for cacao and livestock. For most other sub-projects, the marketing aspect has been neglected. Few of the sub-projects implemented directly targeted environmental protection. Yet, those targeted at ecotourism, organic production or tree planting have demonstrated positive outcomes.

Sustainability, innovation and replicability

The likelihood that the small initiatives supported at the community level will be sustained or replicated is varied. Few of the activities reviewed by the evaluation are considered as fully sustainable. Just over 50 per cent were considered moderately sustainable¹⁵ . The main thrust for sustainability will be realized through the cacao industry where the known market potential is considerable and the system is already fully operational. Few project initiatives were innovative, yet there are a few successful innovations that will be both sustained and have potential for replication resulting in a rating of moderately satisfactory. The cacao production and the complementary organic production system is an improved traditional process that is now being replicated by most of the TCGA members. There is potential for using the system for other organic crops. Some initiatives commenced by CARD have the potential for implementation/scaling up under the current European Union (EU)-financed Belize Rural Development Project (BRDP). The BRDP design has incorporated some lessons from CARD and yet is still experiencing similar implementation issues. The communities visited during the evaluation already perceive it as a continuation of previous failures – a project that has ostensibly been designed for them but to which they cannot access.

Overall assessment

CARD has been largely perceived in Belize as an unsuccessful project by the Government and by the beneficiaries. The investments in the communities did not generate the expected performance in terms of outputs. Overall, the magnitude of project benefits for the communities has been below expectations. Yet, there were several initiatives that were rated as highly satisfactory and which have far reaching potential for southern Belize.

Aggregations of the impact, sustainability and innovation ratings show that where outputs were generated, results were positive. The overall project achievement for CARD is assessed as moderately satisfactory.

The table below shows the rating of the CARD project.

Performance ratings of the CARD project¹⁶

Source: IFAD evaluation mission 2007
*/   This is an average of relevance, effectiveness and efficiency.
**/ Determined based on the ratings for relevance, effectiveness, efficiency, impact,
sustainability, innovation, replication and up-scaling. 

Conclusions

Project success was seen where suitably qualified staff members with practical experience in field-based rural development were appointed, where community interests were supported, and where ready markets were available. Although the design was assessed as valid, there were several major gaps in project implementation that were instrumental in the poor performance including: (i) poor human resource management; (ii) a fragmented, externally driven approach, rather than the recommended strategic, phased approach; (iii) lack of timely and sustained follow-up on project interventions; and (iv) ineffective monitoring and evaluation.

There were other weaknesses in the project, but if these four aspects had been adequately addressed, most of the other challenges could have been more easily overcome.

Lost opportunities. Cases of lost opportunities abound throughout CARD implementation. Some sub-projects stimulated activity but when support was not sustained, momentum declined. Provision of sub-projects that had not been identified or prioritised by the communities resulted in loss of ownership. The model proposed for the community development aspect was never adequately implemented and therefore at this stage cannot be correctly judged as successful or not.

Access to markets and market potential. Marketing support was seen as a major need by both groups and individuals; by men and women. Once the markets have been opened, producers were ready to actively invest their own time and resources in production. Poor to no results were seen in all supply-driven initiatives where there was no specific market and/or marketing process identified. One of the reasons for the exemplary success of the cacao production was because there was a confirmed buyer for the production that was not only willing to buy, but who was also an active partner and financier in the cultivation process. There are other products in the south, particularly organic products that have substantial market potential, and which are appropriate to the local culture and environment. More could have been done to harness that potential.

Local institutional development. CARD has left an institutional legacy amongst the IFIs and TCGA that is likely to be sustained. Long term inputs to producers' organizations, particularly the TCGA, yielded multiple benefits in social capital generation as well as in economic and institutional development. The credit unions were an effective mechanism for developing rural credit for small farmers in the south and have potential for further growth. Yet they are still at a vulnerable stage of development. The development of the local community institutions is still well below that envisaged at appraisal.

Community group expectations were unduly raised due to the repeated failure of sub-projects to deliver on the expectations that had been raised. The process of planning in the community was taken as a "promise". Training and skills development did not meet the needs of the groups.

Complexity and accountability. The CARD structure was too complex and was unable to function effectively. Decision-making was too centralized and remote from the operations of the project. The role of the PSC should have been to support and enable, rather than control and manage. Lack of orientation and accountability processes installed in the project decision-making committees, and between stakeholders meant that each did not adequately fulfil their role, or require that the project fulfilled its role effectively.

Similarly, a project must maintain supervision over the delivery of services, must check with the project participants to ensure satisfaction with the service and follow up to ensure that the service yields the required results. The partnership arrangements between GoB, CDB and IFAD were appropriate but communication and lack of clarity in roles affected the responsiveness of the partnership.

Outstanding policy issues. There were larger national issues which impacted upon the project and were not addressed. In particular, land tenure issues were identified in the logical framework as an important factor that would influence the eventual benefits of the project. Land speculation is increasing across the country. Those with more capability to access and pay for lands are leading the processes of surveying land with a view to obtaining land titles. The indigenous communities in particular fear loss of traditional lands and elements of conflict over land. Observations during the evaluation highlight that this is an escalating issue¹ .

Similarly, recognition of ethnic diversity, and its relation to poverty was an issue raised by both the Government and the community. Some activities of CARD demonstrated effectively how cultural practices can be harnessed for economic good, and how ethnic diversity can be seen as a potential rather than a challenge. Lack of decentralized decision-making in both governance structures and support organizations creates a barrier to development for the south.

Recommendation 1

Investment in the southern districts of Belize. The advances made through CARD show that southern Belize has potential that is firmly in line with the GoB strategic objectives towards economic development through niche export marketing and ecotourism as well as in poverty reduction. The unrealized potential that was present in the south at the commencement of CARD remains unrealized but still realizable.

There are a number of social and economic assets developed through the project which still require support to ensure that the full benefits from project investments can be realized. There is a need for detailed negotiations with the BRDP to definitively assess whether or not these priorities can be accommodated. At present, there is a general statement that this will occur but no mechanism for transfer of the knowledge from one project to the next.

Niche marketing in organic products has substantial economic potential for the south and requires further investment. This requires an enabling environment by the Government including: careful consideration of developments that may compromise the organic certification of the area; continued support to the TCGA, possibly in extending their activities to market other organic production; and/or establishment of similar producers organizations for other organic products; facilitation of export approvals and other documentary requirements for export of organic products.

Recommendation 2

The reflows from the Rural Credit Fund need to be quarantined for continued support to the two participating credit unions. The Belize Credit Union League (BCUL), with the Central Bank, is likely to play a major role in future nurturing of the credit unions in the southern districts.

Policy dialogue is required to support the professional development of the credit unions' needs. There is still additional credit funds required. Given the positive performance of the credit unions and the identified issues concerning future development, the GoB can assist in further development in the southern districts of Belize by ensuring that the funds are available to the south in perpetuity.

Recommendation 3

Realistic implementation. There are a number of recommendations in relation to improved support to development projects in Southern Belize. These include: (i) intensive support at the start-up of a project to ensure that the design is effectively translated into implementation; (ii) more emphasis at the start-up of the project to install and operationalise an appropriate monitoring system that will respond to the management information needs of the implementers, partners, and assist in detecting when a project deviates from its planned approach; (iii) there is a need for future project initiatives to clearly state intentions and gauge and manage the level of expectations generated; (iv) the groups to be financed need to either be in existence prior to the project and/or have specific self-generated objectives and a common cause. The methodology that requires communities to form small groups specifically to access funds is not viable; and (v) long-term investment, similar to the one given to the TCGA, should be provided to develop local institutions - both local governance structures and specific interest groups - which is likely to yield better results than short-term fragmented sub-projects.

Recommendation 4

Investment in people. Improved communication mechanisms and opportunities for cultural exchange need to be encouraged so that there is an increased mutual understanding of the different cultures that comprise the national identity of Belize. Supporting cultural initiatives like the deer dancers and the community-based national park management initiatives contribute to economic benefits as well as transforming cultural differences into an advantage rather than a challenge and there are other opportunities that can be similarly developed. Future initiatives must not ignore the traditional processes but can learn from and integrate the traditional processes, melding them with available technologies to provide lasting changes in agricultural systems. It is not a matter of stimulating these processes, but rather of supporting the evolution that is already occurring.

Policy level review in decentralization is needed to assess the representation of different ethnic groups in decision-making for their own development and the relationship between the roles and responsibilities of different government and non-government institutions needs more definition. At the same time, there is a need for community leaders, both traditional mayors and local councils to receive more orientation, training and basic tools for leadership. Lessons learned from other countries are that the connection of indigenous people to land and other forms of self-determination is a sensitive yet highly important issue. It is not an issue that was addressed by CARD, yet it is an issue that continues to face many of the communities that it aimed to assist.

Recommendation 5

Professionalizing project processes.Project procedures were too complex and cumbersome. The allocation for appropriate procurement procedures needs to be in line with the expected type of sub-projects. Staff and communities must be guided by an overall strategy that is "user and culturally oriented". Introducing a skills transfer clause into the contract for a service provider can greatly increase the long term benefits to the project participants. There has been a recurring theme in project evaluations for Belize that poor recruitment processes and human resource management compromises project effectiveness. Experienced rural development implementers at the management and field level are needed to serve as mentors and provide on-the-job training to local project staff. Good practices for project management, procurement, field implementation and monitoring and evaluation could be transferred within the country. When staff turn-over does occur, the project partners need to have a commitment to proper orientation of newly appointed staff.

Recommendation 6

Productive partnerships. The partnership between CDB, IFAD and GoB was an appropriate arrangement that brought together expertise and resources towards a common objective. Some amendments in the implementing arrangements may have been of assistance to make the partnership more productive, for instance: (i) clearer distinction in the roles and responsibilities, particularly for CDB fulfilling a dual role as a financing partner of CARD, and as a cooperating institution for IFAD supervision; (ii) greater investment in establishing the PMU at the commencement of the project; (iii) more review of the specific procurement processes for CARD to fit with the small-scale procurement required; (iv) better communication mechanisms and practices to expedite critical correspondence, exchange views more regularly, increase mutual understanding of the challenges faced by the project and lead to more rapid identification of solutions.

---

1/ Defined as those having less than 20 acres of land.

2/ IFAD Office of Evaluation (1994), Belize Toledo Small Farmers Development Project Evaluation. The assessment of performance for TSDF at that time was "relatively poor". This has been translated in "moderately unsuccessful" (3) in accordance with current OE rating guidelines.

3/ CARD (June 2006), Audited Accounts.

4/ There were slight differences between the IFAD and CDB appraisal documents. The CDB appraisal proposed that CARD would work with the 48 poorest of the 72 targeted communities; not 54 according to the phased approach proposed by IFAD.

5/ Field work was conducted collaboratively in Belize from 7 to 25 May 2007, after a preparatory IFAD Mission in April 2007.

6/ CARD (2006), Project Completion Report.

7/ See Annex 5 for details of sample selection, assessment process and results.

8/ It also responded to most of the recommendations of the IFAD Technical Review Committee (TRC) and Operational Strategy Committee (OSC).

9/ IFAD (2005), CARD Project, Assessment of the Viability of a Project Extension.

10/ The evaluation reviewed the Feasibility Assessment for the Extension and confirmed that the case for extension was sound.

11/ CARD (2006), Project Completion Report (PCR).

12/ The Regional Strategy for Latin America and the Caribbean was produced in 2002 and highlighted strategies for empowerment of the rural poor, taking advantage of market opportunities and partnerships; these were already incorporated in the CARD design.

13/ CDB as a cooperating institution is contracted by IFAD to handle the fiduciary aspects and conduct the required supervision missions for the project. IFAD also undertook supervision missions sometimes with CDB, at other times without.

14/ Such as the case of distribution of coconut seedlings which was approved by the PSC in 2001, yet had not even been requested by the communities targeted. (See Monkey River Case Study in Annex IV).

15/ Of the 17 communities visited by the evaluation mission, nine communities were able to confirm that the activities that had been supported by the project had been partly sustained. The range of activities was varied, ranging from infrastructure (culverts installation) to livestock, sewing, and ecotourism. The definition of sustainability was merely that the impact of the specific activity could be discerned as continuing, and likely to continue. This needs to be considered in the context that the benefits to be sustained were at a low level, apart from the cacao production.

16/ As per OE's project evaluation methodology, a six-point scale has been used to attribute ratings to each of the evaluation criteria. Six-point rating scale reads as follows: 6-Highly satisfactory, 5-Satisfactory, 4-Moderately satisfactory, 3-Moderately unsatisfactory, 2-Unsatisfactory, 1-Highly unsatisfactory.

17/ Such as credit being accessed for survey of land, only to have more powerful people sequester the surveyors and submit alternative claims for land; new owners preventing villagers access to traditional hunting grounds, amongst others.