Target groups and targeting mechanisms

Targeting. Over the years the ability to identify the poor during design has greatly improved in IFAD projects in Ghana. The importance of targeting in design documents grew from being non- existent in VORADEP to being one of the main thrusts of project design in UWADEP. The categories of poor targeted in IFAD project designs can be classified as follows:

(i) smallholders and near-landless in the densely populated areas of the Upper East and Upper West Regions, where land is limited, fertility low, local off-farm employment is lacking and water is at a premium;

(ii) women, particularly those in women-headed households in the north, who often lack access to sufficient labour;

(iii) smallholders in the Northern Region, where low rainfall, poor roads and deteriorating soils have led to seasonal food insecurity and vulnerability;

(iv) smallholder food crop producers (without cocoa) in the central regions, where social/kinship support systems have weakened and dependency on seasonal employment and small-scale enterprise is high.

Most of the projects, however, based their targeting on cut-off or exclusionary indicators which determine who is within the target group and who is outside. These are quantitative criteria referring mostly to a maximum landholding size for eligibility for project assistance. A consistent weakness in the design of all but the last project is the absence of targeting mechanisms which define how those eligible will be reached effectively. Little attention is paid to the technical feasibility (or to the costs) of applying such quantitative criteria which proved impractical during implementation and/or resource intensive. The result is that criteria-based targeting has almost always been applied poorly, half-heartedly or not at all. In a few instances, too much time and effort are spent by project staff pursuing the objective of targeting the poor without commensurate results.

In the context of traditional communities in Ghana any targeting mechanism which requires precise measurement of landholding size is impractical, since even honest estimates by the farmers themselves are often based on traditional means of approximation which vary from one locality to the next. During implementation none of the projects in the Northern and Upper Regions have targeted project services using the defined quantitative criteria even in the more targetable activities such as input supply and credit (see paragraph 63). This, however, did not seem to have affected the poverty coverage of the projects. More important, little consideration has been given to the socio-political consequences of attempting to apply these criteria. Undertaking this type of selection in close-knit and relatively homogeneous communities can easily create tensions and conflicts.

The CPE mission's fieldwork amply confirmed the view that much of rural society in Ghana is characterized by strong social cohesiveness and complex interdependency webs. While these networks have weakened somewhat in the south and central regions due to the long-term impact of cocoa farming and commercialization, they are still very strong in the three northern regions. The social network in these communities is a key dimension of well-being and livelihood security. Poverty is defined by the people themselves as a communal rather than as an individual phenomenon. This implies that when targeting is based on a system in which outsiders divide the community into "haves" and "have-nots", it is likely to be neither implementable nor sustainable. In this particular social context, targeting should be aimed at the community as a whole rather than attempting to forcefully break up or bypass the local power structure. In addition, poverty in the Northern and Upper Regions is so pervasive as to justify limiting targeting to area targeting. The percentage of better off households is so low that it is sufficient, in cost-benefit terms, to select poorer areas for project participation rather than attempting to target individuals or households.

The one case in which it appears that eligibility criteria are being applied in IFAD projects in Northern Ghana is in the distribution of incremental irrigated land at rehabilitated dams under LACOSREP. Distribution was the responsibility of the WUAs and the new lands were given to the poorest (though women did not benefit proportionately) excluding chiefs, large landowners, those with another plot in the scheme, wage-earners, etc. The poorest have been identified jointly with the communities involved.

SCIMP has done more than the other projects in attempting to implement eligibility-based targeting criteria. This is because the Staff Appraisal Report (SAR) for this project was clearer than others on the importance IFAD attaches to targeting as social differentiation is more pronounced in this project area. Nevertheless, the results of applying these criteria are not encouraging. The average holding size for SCIMP group members was 3.6, 2.4 and 1.9 ha in Brong-Ahafo, Volta and Ashanti Regions, respectively, compared to the SAR cut-off of 1.6 ha. Moreover, data available to the CPE missions indicate that just over one third of beneficiaries (37%) of production credit (the largest category in credit disbursement) fall within IFAD's target group as defined at design. While this may partly reflect the unsatisfactory efforts of the NGOs in ensuring the right composition of groups, it is more indicative of the constraints involved in implementing quantitative targeting.

The UWADEP project is attempting to implement targeting methods which are designed to address these constraints. These mechanisms appear to be less costly than criteria-based targeting in terms of the human and financial resources needed to evaluate potential beneficiaries, as well as in terms of the credibility of the project in the eyes of participating beneficiaries. Verification, however, will be needed during implementation.

Gender Targeting. There has been progress in targeting poor rural women beginning with VORADEP through to UWADEP. Government institutions with experience in mobilizing women and coordinating women related activities, as well as local NGOs, were identified and mandated in appraisal documents to support women's group promotion and development. However, the use of many agencies to support the formation and development of women's groups has resulted in a considerable overlap at the expense of effective targeting. In addition, many of the participating agencies including banks, NGOs and the extension services, still have a majority of male field staff which reduces the overall effectiveness of extension services in reaching women.

Group formation has been almost always used as a mechanism to facilitate reaching rural women. Membership of many women's groups is composed of more old than young women and the executive was largely dominated by the elderly which prevented more innovative ideas from being tested. The most effective gender targeting was in credit. Targeting women was not as effective in other activities, especially irrigation activities. Though identified, the issue of women's access to land was not adequately addressed in design or implementation. It is doubtful whether the project can alter this situation during its lifetime, which makes a strong case for providing support to women's non- farming activities, such as marketing and processing.

Women's groups participating in the projects have developed strong mutual support networks and have gained experience in the use of credit and in working on diverse income-generating activities. They have increased their exposure to agencies external to their communities and are now better able to articulate their needs and interest. Training has been provided to women's groups in various areas. However, training has focused more on providing information than on teaching skills. In addition, training is most often targeted at group executives who are usually the elderly, and the younger members often do not benefit proportionately. It also appears that in general, groups' executive members, who are trained do not impart the skills they acquire to other members of the group, and mixed groups with male secretaries appear to be less effective in terms of cohesiveness and repaying loans than all-women groups and those groups with literate/numerate female secretaries.

Beneficiaries Groups. Project designers often view groups as a targeting instrument, whereby the selection of group members is a method for screening out non-eligible individuals. In implementation, however, the process of group formation in itself rarely functions as an effective targeting mechanism. As interested community members were asked to form groups, they were not able to take the responsibility for excluding other members of their own community. The staff of NGOs/IAs were unable to successfully apply any screening mechanism. Group formation, nevertheless, remains a main feature of all IFAD-initiated projects in Ghana.

Beneficiaries' groups in IFAD projects varied in quality, but overall they lacked cohesion and sustainability. Two factors emerged. First, women-dominated groups were distinctly stronger than men's groups. Second, groups which already had a history of self-help activity, as part of community tradition, were less dependent on the offer of credit for their existence. Such groups demonstrated higher loan recovery, better record-keeping and much more cohesion.

The Water Users' Associations (WUAs). While still at mid-term LACOSREP seems to have produced one particularly successful experience in group activity: the WUAs associated with the project irrigation activities. The groups are different from project credit groups in that they are built around very clear and concrete economic assets (plots or irrigated land) with a valuable contribution to household food security in the dry season. The value of the asset and its continuing contribution to household income is closely linked to the effective functioning of the group. The development of these WUAs under LACOSREP should serve as an example for other IFAD projects in Ghana, and possibly in other countries.

The use of NGOs for group formation and animation has been included in all IFAD-initiated projects in Ghana. The experience has varied greatly with the experience of SRDP being the most positive . Under SCIMP the NGOs selected had little capacity or relevant experience to bring to the project, in spite of the long selection process. This is partly the result of the much greater demands which this project placed on its collaborating agencies. Instead, the project had itself to finance new staff, training, transport and other project-related needs of the NGOs.

Participation

Overall, because of the difficulties inherent in undertaking genuine participatory development, there have been only modest achievements on this account in IFAD projects in Ghana. While reference to beneficiaries' participation in design increased over time, none of the projects include effective mechanisms for this in project design. Projects are designed with little input from "beneficiary assessments" but mainly by technical experts who use their own judgment and experience to decide what is best for the target population. During implementation, beneficiaries are treated (and react) as passive recipients of project goods and services. This, however, has not precluded project benefits from reaching the target group; impact could have been higher with genuine participation. UWADEP's design advocates a stronger participatory element and a different approach to participatory implementation. The experience has to be monitored carefully from the start of implementation.

The inclusion of beneficiary representatives on project coordination committees, often indicated as the major vehicle for beneficiary participation, has been only irregularly applied, and even when applied had serious weaknesses as a participatory methodology. It is unlikely that one man and one woman would be able to represent the interests of tens of thousands of beneficiaries spread over a wide geographical area. No mechanisms are indicated for the selection of these representatives, so that it is not clear if they are elected democratically or appointed. In addition, they are always a small minority on the committees (two out of ten or more), so that their influence would be minimal even if they could represent all the beneficiaries.

The use of local NGOs for implementation of the credit component has allowed for some participation in implementation (e.g., SRDP self-initiated activities promoted by NGOs). However, this initiative has been taken by the NGO itself which had previous relevant experience in this field. The most positive experience in participation so far, is perhaps the LACOSREP's WUAs which are allowing a greater sense of ownership of project activities by the participants.

 Overall impact

As only two projects in the Ghana portfolio have closed - two are at mid-term and two are in the early stages of implementation - the programme's impacts cannot be comprehensively assessed and can only be subject to partial analysis. Notwithstanding this constraint, the CPE could still distinguish three sets of activities: those with clear positive impact, those with mixed impact and those without a significant impact

(a) Activities with Positive Impact

New Technology. The importation, multiplication and release of cassava pest predators have virtually eliminated Cassava Mealy Bug throughout Ghana. Similarly, releases of Cassava Green Spider Mite predators, through the NRTCIP, are now making rapid inroads on this pest. While improved cassava varieties through VORADEP in the 1980s have had some unassessed impact in the Volta Region, the release of three improved cassava varieties under NRTCIP has set the stage for a potentially massive expansion in cassava production nationwide. Similarly, evidence suggests that the improvement of sheep breeds under IFAD projects is having a positive effect on farmers' income so far.

Irrigation. In the Upper Regions in Ghana, having access to an irrigated dry-season plot makes the difference between food-secure and food-insecure households. LACOSREP irrigation activities have achieved thus far about 50% of the appraisal target of the dry-season cultivation of the Upper East Region. Apart from assuring a wet season rice crop, irrigated areas are producing high value cash crops such as onion, tomato and pepper in the dry season. While it is not clear whether the full appraisal target of irrigated land can be achieved, and hence full incremental output, there is little doubt that any addition to dry-season irrigation is a direct contribution to household food security.

Dam rehabilitation has allowed villagers to obtain additional irrigation by digging shallow wells which take advantage of overflow from the command area. Beneficiaries indicated that with these plots they are able to make it through the year without depending on food aid during the "starvation" season. On the institutional side, the promotion of WUAs with strong by-laws and eventual legal status, has initiated a process of institutionalizing effective beneficiary participation in irrigation schemes.

Potable Water. The provision of good quality water near communities reduced the incidence of water-borne diseases and is making significant contributions by reducing the time spent in fetching water, an onerous task usually left to women and children. VORADEP and SRDP built 34% and 53% of all borewells in their respective project areas. The beneficial effects of the borewells, in terms of health and productivity, are difficult to quantify. The wells in SRDP provide 35 000 people with water routinely and a considerably larger number from nearby communities during periods of critical water shortage.

Rural Roads. IFAD's support to rural roads has undoubtedly had a major impact. Quantitatively the greatest effects occurred through VORADEP. The road achievements of VORADEP are still very much in evidence where the largest market in one of the project districts is a direct consequence of a VORADEP-built road. The same positive impact of roads occurred in SRDP and SCIMP. The most striking impact is the dramatic increase in the population of the communities along the roads, amounting to an average 55% with many having their populations double due to migration from outlying communities.

Institutional Development. Project assistance to extension, in at least three projects has been substantial and provided a very solid base for the UAE to work from. The two initiatives under SRDP, Women Extension Volunteers and Demonstration Homes, have had a significant impact on women's access to information and training. Assistance, under NRTCIP, has resulted in root and tuber crop research being institutionalized within the national research system and allowed the root and tuber crop units and the Biological Control Unit to establish international linkages. Efforts in promoting the MOFA's goal of better research/extension/farmer linkages (specifically in LACOSREP) have had considerable success, and the model can be replicated elsewhere

(b) Activities with Mixed Impact: Credit

The credit components of the projects have shown that issues associated with smallholder credit are complex and not easy to handle particularly in the context of a changing economic environment. The picture that emerges so far is that of a mixed impact. At institutional level, development inputs directed towards the rural financial sector through IFAD projects have had some positive influence on the behaviour of the participating financial institutions. However, the loans portfolios created by the projects are showing signs of increasing non-performing assets of PBs caused by lower-than-expected repayment performance. This will affect the operations' sustainability. Groups in general and women in particular have benefited in various degrees from the credit programmes, but their repayment rates still remain low. Their ability to generate savings to build equity and self- finance operations is equally low.

Creation and institutionalizing group discipline with respect to credit has not yet materialized. Payment of group dues were several months in arrears and some groups stopped payment of group dues just after loans were granted. Credit management knowledge and discipline are found to be quite low overall. Whereas credit has been initially an essential ingredient in the establishment of groups in IFAD-supported projects its contribution to the continuity and sustainability of such groups is at best doubtful.

(c) Activities without Positive Impact

Two pilot activities carried out under SCIMP fall into this category: the Valley Bottom Rice (VBR) Pilot Project, and the Draught Animal Power (DAP) Pilot Project. In the first case, the beneficiaries were poorly selected, the works inadequately designed and the whole scheme badly managed. In one of the rice schemes, women were selected from a village 30 km away from the project site, were already cultivating rice near their homes and had little time to dedicate to the VBR activity. Overall, the role of government IAs in managing the schemes has been so dominant that it led to an attitude of dependency on the part of the beneficiaries. Yields have been low, and interest on the part of the beneficiaries weak leading to deterioration of project infrastructure. In the case of the DAP project, animals originally intended for ploughing had been diverted to carting. In general, while farmers were interested in the ownership of the oxen, the impact of their introduction on local farming and the household economies was minimal. A comprehensive specific assessment for these pilot projects and a decision on their reformulation or closing needs to be taken.