Evaluation objectives and methodology

At the request of the Tunisian Government, IFAD undertook an evaluation of its cooperation programme with Tunisia. The overall objective of the evaluation was to draw lessons from over 20 years of project experiences (1980 to 2002) as a contribution to the definition of future IFAD strategy in Tunisia.

Evaluation objectives and methodology were defined during the first stage of the evaluation. During the second, three preliminary studies were undertaken covering target groups, institutional arrangements and the integrated development projects model. At the third stage, an evaluation mission visited Tunisia from 27 October to 21 November 2002. The mission worked closely with a representative cross section of programme partners, especially during the start-up workshop in Tunis, regional workshops in the three governorates – Kairouan, Siliana and Zaghouan – and a final meeting during which the mission’s tentative conclusions and recommendations were presented and discussed. The final stage of the evaluation was a Round Table meeting held to conclude the agreement at completion point, which reviewed recommendations by the partners and detailed responsibilities for the implementation of agreed follow-up.

The weakness of project monitoring-and-evaluation (M&E) systems proved a hindrance to the mission’s analyses and appraisals, in that the available data covered only material outputs, with considerable gaps in data on project impact.

IFAD cooperation in the context of tunisia’s policies and development

The economic context and reducing inequalities. Tunisia’s economic performance during the last 20 years has been outstanding. At the same time, substantial efforts to distribute the benefits of growth have led to a significant reduction of poverty. Nonetheless, marked disparities still persist between the urbanized coastal areas and the disadvantaged central areas of the country. The core population of poor Tunisians is flanked by a larger section of vulnerable people, composed primarily of the illiterate, women working in the informal sector and in agriculture, and smallholders working in rainfed agriculture. Boosting employment and closing the regional gaps are among the foremost challenges of the coming years, and feature prominently as priority objectives of the Tenth Plan (2002-2006).

Trends in the agricultural sector. The last 30 years have seen a marked increase in the number of farms (44 per cent) and a 12 per cent growth in smallholdings. This is clearly the outcome of land fragmentation. Smallholdings, which occupy the most marginal and least productive land, have to face three major constraints: highly irregular rainfall, problems with access to financial resources, land fragmentation and the lack of property title.

Up to the year 1986, the government pursued a policy that favoured strategic goods – through subsidies and total state responsibility for all support services – but yielded few benefits for smallholders. Additionally, imports of these goods weighed heavily on the balance of payments. In 1986 the Government began to implement a structural adjustment policy. Production and support services were reorganized and the bulk of the public companies were privatized. Occupational and trade associations were promoted and the Ministry of Agriculture was restructured. Retargeted state action included a process of regionalization to cater to the disadvantaged interior and the introduction in 1989 of regional councils, responsible for preparing regional development plans and land-use planning. Programmes to equip rural areas were strengthened to improve living conditions and reduce regional disparities. New, nationwide agricultural sector strategies – in such areas as resource conservation, development of marketing channels and consolidation of trade associations – took concrete form at the regional level with the Commissariats régionaux de développement agricole (CRDAs).

The IFAD–Tunisian Government cooperation programme. IFAD’s programme in Tunisia includes nine projects for a total value of USD 289.85 million, of which IFAD financed USD 114.7 million, or 39 per cent of the total, as shown in Table 1 below.

Projets	Effective date r	Total cost	IFAD contribution	IFAD share of total cost	Gov’t. share of total costl	Additional cofinancing
Completed projects Kef and Siliana Sidi Bouzid irrigated Sidi Bouzid rainfed Oued Mellegue Sidi M'Hadheb	1981 1984 1986 1989 1992	45.9 16.3 13.3 31.4 37.2	19.0 7.3 6.0 12.0 11.9	42% 45% 45% 41% 32%	56% 47% 48% 38% 29%	WFP - WFP IsDB IsDB
Ongoing projects Kairouan Siliana Zaghouan	1995 1996 1999	28.2 41.7 33.6	12.1 11.2 16.1	43% 27% 48%	20% 20% 26%	IsDB AFD -
Approved project Tataouine	-	44.3	18.7	42%	37%	OPEC

Programme development strategies. Two strategy options characterize the programme as a whole. The first concerns the choice of the least favoured regions and populations in the central and southern parts of the country. The second concerns the focus on the agricultural sector and the solid anchoring of projects within sectoral policy lines. The first strategy document of 1992 stressed resource conservation and boosting agricultural productivity, but also and equally the need to develop non-agricultural activities in order to create jobs and enhance the value of agricultural production. The second strategy (as set forth in the country strategic opportunities paper, or COSOP) in 1998 attributed the muted impact on technical progress to the lack of farmer involvement and weak development of rural associations. It recommended pursuing participatory approaches, seeking greater gender equity, reinforcing rural financial services and promoting sustainable natural-resource management.

There have been three generations of projects within this general context. The first-generation projects (1981-1985) centred on productive activities, and did not include farmer participation, the promotion of farmers’ associations or support for women’s activities. In the second generation (1988-1998), the focus was on the most degraded areas, and the approach that of integrated development in support of natural-resource conservation, higher agricultural productivity and enhanced community infrastructure. Resource-conservation activities and the promotion of production-oriented community infrastructure predominated. Support for women’s activities (1989) and participatory community development (1995) do appear at this point, but account for no more than four per cent of the budget. The Zaghouan project (1998) launched the third generation of projects, with the expansion of the participatory approach to more activities, but there was basically not much change with respect to the preceding generation. The Tataouine project of 2002 also falls into this category, proposing a partnership-based approach involving local stakeholders and the promotion of rural organizations; it is structured on three levels: the region, for structural investments; agropastoral lands, for participatory natural-resource management; and individual farms, for the development of non-agricultural microenterprises and marketing channels.

Review of principal outputs

The main achievements are found in the area of large-scale investments: soil and water conservation (SWC), rangeland development and various infrastructure works. Support in the form of credit, training and organizational advice was very modest, in line with budget forecasts. Overall, the activities undertaken by government services with no or limited participation by beneficiaries (SWC and community rangeland improvements) presented performance levels that exceeded 100%. On the other hand, agricultural support components that depended on credit encountered serious implementation problems, especially once credit was no longer financed by the project. Smallholders’ access to credit was restricted by a number of factors. Components for community development and promotion of women, which were weak to begin with, saw very limited implementation. The main outputs are listed in the following table.

Table 2. Principal Project Outputs

Type of activity	Unit	Total	Type of activity	Unit	Total
Resource conservation			Hydraulic infrastructure
Soil and water conservation	Ha	170188	Cisterns and wells	unit	4430
Consolidation	Ha	31753	Water supply system	km	87
Forest plantations Forest clearings	Ha Ha	3930 1540	Road infrastructure
Rangeland development			Rural tracks	km	285
Pasture improvement	Ha	16957	Credit		0
Fooder plantation	Ha	12408	Clients	unit	9092
Livestock production			Total amount	DT	15548742
Minor infrastructure	unit	781	Groups
Livestock purchasesl	unit	2528	All types		183
Agriculture			Training
Intensive cereal grain cultivation	Ha	47508	Girls		1431
Tree plantations	Ha	44486	All categories		2025
Irrigation	Ha	4007

A total of 452 275 people live in the area covered by the programme, with projects targeting 76 210 farms. Based on very approximate data, due to the lack of project impact data generated by the M&E systems used, an estimated 230 090 people were reached by IFAD-cofinanced projects, or some 70 per cent of the projected figure. Considering that some 40 per cent of this total are poor smallholders (92 000 people in all), IFAD-cofinanced projects have likely reached some 20 per cent of the rural poor as assessed in 1980 when the programme began – a fairly substantial achievement. The available data give no indication; however, how many rural poor who received programme support actually were able to rise out of poverty as a result.

Impact and sustainability

Did the approaches implemented actually reach the most vulnerable and facilitate their social and economic integration? Targeting projects along a central axis running from north to south focused interventions on disadvantaged and environmentally degraded regions with irregular rainfall. Within these regions, however, the criterion of choice for targeting the poorest was farm size. This is inappropriate, mainly because it fails to take into account the great diversity of smallholder activities and fluctuations in farm production in response to the vagaries of the weather. Moreover, the choice is not compatible with the geographic selection criteria used for such activities as resource conservation. Practically speaking, however, for those projects for which data are available, smallholders with under 10 ha were, on the whole and in terms of absolute figures, the main beneficiaries of project investments. Projects did help to reduce poverty, but were less successful in reducing inequalities. Farmers with over 20 ha benefited more from the projects than did other beneficiaries. The considerable number of smallholder beneficiaries with less than 10 ha is more a reflection of the way the agricultural sector is structured and the preponderance of smallholders in the target regions than a result of project targeting. Three projects that did not apply farm size as a selection criterion (the two Sidi Bouzid projects and the Mellegue project) all the same helped to launch a process of overall economic development in their target areas to the benefit of the entire population (though this did not help to reduce inequalities).

What was the project impact on improving women’s living conditions and enhancing their participation in development? Activities targeted specifically at women reached only a small number of beneficiaries and had little impact on income-generation. Women did benefit, however, from investments to improve living conditions (e.g. drinking water supply and new tracks) or investments to boost farm income, but then so did the household as a whole. Projects failed to reflect the growing role of women in running farms. Other activities not specifically targeted at women failed to take their specific situation into consideration. While the Zaghouan project did afford the opportunity for women to participate in planning activities, thus acquiring a new right to make themselves heard, women are still either excluded or heavily underrepresented in formal rural associations. This limited project impact can be traced to the absence of clearly defined, gender-oriented objectives and strategies to promote women. Project emphasis on land-based activities was also a factor, inasmuch as women have little access to land ownership. Weak participation from the institutional side, the lack of training for project and CRDA staff, and the failure of project supervision and M&E to address these issues were further constraints.

Did participatory approaches lead to permanent rural associations and sustainable investments? Participatory approaches were implemented on a rather small scale and so their impact has been very limited. Moreover, projects paid little attention to support for rural organizational and capacity-building activities, focusing instead on material investments. The result was that farmers were not very motivated, and thus took limited responsibility for managing and maintaining project investments; this was also reflected in the low number of smallholder associations established. These gaps are due to inappropriate project design and an institutional climate unfavourable to such approaches. Another factor was that project implementation focused on government-defined sectoral strategies and material outputs. Even this limited experience, however, led to the introduction of new arrangements for direct farmer/government cooperation to the satisfaction of both.

Were the projects able to adjust to the constraints of low-rainfall areas? Projects helped to buffer the impact of drought on farmer output and income by adopting a preventive approach. They implemented new, drought-adapted measures such as the cultivation of hardy trees, hydro-agricultural schemes, and, massively as of 1989, soil and water conservation and rangeland improvement. Various factors combined to curtail the expected impact of investments, primarily the lack of attention to farmer strategies, weak farmer participation in the search for workable solutions, gaps in the research and development programmes, limited direct support to farms from the second generation of projects onward and limited access to bank loans.

Did the projects engender a sustainable dynamic for economic development? The injection of significant capital flows, mainly to productive sectors and individual farms, did help to make local economies somewhat more buoyant, up through the Mellegue project. Project areas also witnessed an increase in population stability, thanks not just to residents’ involvement in productive activities and increased revenue but also to infrastructure improvements (tracks, access to drinking water). In the more recent projects, the emergence of a sustainable economic dynamic is less visible, essentially because these areas were more seriously degraded and priority has been assigned to protecting resources rather than to productive activities. Overall, the projects have not pursued new areas of activity capable of tapping local potential and bringing greater added value; similarly, they have not provided support for non-farm or quasi-farm activities, even though these are key elements of farmers’ strategies. Support for producers in the form of access to credit or assistance in launching economic projects was also weak. The experience in Siliana, however, has shown the potential of an approach based on support for local initiatives.

Is the integrated agricultural development approach an appropriate model for attaining the objectives of IFAD-cofinanced projects? This approach, which targets better living conditions and incomes for rural people and – in the longer term – rational natural-resource use and conservation, is potentially suitable for the degraded and difficult environmental and living conditions of the IFAD target area. But in actual fact the approach could not be implemented, mainly because of the close focus on community infrastructure and resource conservation to the detriment of on-farm production support activities. This imbalance was further accentuated by limited access to the credit that was to have financed the bulk of productive activities. And while the planned interventions followed a strictly sectoral criteria, the projects failed to develop truly integrated strategies that could have meshed into a coherent framework, nor was any mechanism set up within the CRDAs to bring together the various sectoral departments for discussion or integration.

Organization and management: how was project performance affected by institutional arrangements and implementation modalities?

The adaptation of project institutional arrangements to institutional changes in the agricultural sector and the growing incorporation of project components into the CRDAs led to a very positive performance for activities familiar to the technical districts, such as resource conservation and rangeland development. Institutional support measures also made a direct contribution to strengthening the CRDAs and expanding extension facilities. Training, and to some extent the introduction of new approaches such as participatory planning, undoubtedly heightened the skills and interest of the staff involved. Nonetheless, difficulties in both the design and implementation of the institutional arrangements hindered project performance, especially for the more innovative activities and their acceptance by the CRDAs. These problems were linked to the lack of mechanisms for capitalizing on innovations within the projects and within the CRDAs, but also to short-staffing, attributable mainly to inadequate institutional review at the design stage. And whereas project objectives lie increasingly outside the narrow framework of agricultural development, institutional arrangements were only weakly integrated into the regional mechanisms. NGOs, which were to have implemented some of the participatory development aspects, were hobbled by their unclear legal status and role, and a certain reserve on the part of the government. Lastly, methodological support was lacking, particularly to backstop the introduction of new approaches by the projects.

M&E systems focused solely on material outputs and budgetary implementation, and failed to involve the stakeholders to whom the participatory approaches and the bulk of project outputs were addressed. Supervision, entrusted to FADES in seven of the nine projects, performed well for loan administration, but paid little attention to monitoring project outputs and none to their impact or to promoting new approaches. Monitoring by IFAD was extremely discreet and was not adapted to the innovations introduced by each project, which would have required specific monitoring.

Project contribution to national plans and policies

The co-financed projects made a direct contribution to the implementation of national development plans and policies, with which they were squarely aligned. They also contributed innovations by introducing wholly new approaches (such as targeting the underprivileged, microcredit delivery by NGOs, and M&E) and by acknowledging new approaches that they did not directly originate (such as the participatory approach and community development, activities specifically designed to meet women’s needs, supervised credit, and the link with research and development). However, these approaches had only limited budgets and aroused only limited interest on the part of the CRDAs. They remain very much circumscribed within the domains of project interventions, mostly because their development was so limited. Although the IFAD-cofinanced projects were not adopted into the national programmes, they can be seen as the source of an initial institutional learning process that allowed CRDA teams to at least gain some familiarity with the new approaches.

Recommendations

General strategy framework

Recommendations for a general strategy framework.

The following four recommendations set forth the basic orientations for a future IFAD strategy in Tunisia.

Reaffirm IFAD’s special focus on aid to disadvantaged areas and vulnerable populations.

In accordance with its mandate, IFAD should reaffirm its special focus on:

• Support to disadvantaged areas. The zoning which determined the concentration of IFAD interventions within governorates in the central, southern and north-western interior of the country remains valid. The Government has invested considerably in these regions, easing constraints, but some zones still face natural-resource degradation, low productive potential, insufficient public amenities and services, weak development of grass-roots organizations, weak rural capacity for taking the initiative and poor economic integration. The Government attaches great weight to reducing interregional disparities and developing disadvantaged areas. This will continue to call for sustained efforts.
• Support to vulnerable populations. Approaches to be implemented in the target areas, while favouring the participation of all local stakeholders, should be rounded out by measures specifically adapted to the characteristics of the most vulnerable sectors of the population, i.e. smallholders, women working in agriculture, landless farmers and unemployed youth. Such measures should seek to include these sectors in the development process. It would be preferable to target the poorest of the poor by developing a supply of products adapted to their special constraints, rather than mechanically applying rigid selection criteria.

Stress the introduction of innovations

IFAD’s role as an innovator is clearly acknowledged by the Tunisian Government, which sees its partner as a strong proponent of new proposals. In the past, however, such innovations were often confined within the project intervention sphere. IFAD should consolidate its stance as a bringer of innovative approaches by taking steps to ensure that these innovations, once tested and refined within the project context, can be more widely replicated and contribute to the development of national policies and strategies. Given the limited total amount of IFAD resources, testing new approaches should take priority over financing standard public-investment programmes.

Enhance the role of projects as policy dialogue laboratories

IFAD-cofinanced projects offer enormous potential for learning about and testing new approaches that deserve a wider role for a more effective contribution to national policy dialogue and policy development. In the absence of an IFAD representation in Tunisia, projects comprise the foremost instrument for policy dialogue. Project formulation and implementation should take this feature into account at every stage, and at the same time try to find ways of more closely coordinating the operational side of projects with policy definition at the central level. M&E systems to learn from these experimental processes, vehicles for disseminating the acquired experience (such as publications, handbooks and audio-visual aids), plus workshops to present and discuss the new approaches and close contacts with the government institutions involved, are just some possible avenues to explore.

Move away from an exclusive focus on agriculture

For a closer focus on the specific problems of the target population, income-generation and better living conditions should become the core IFAD strategy. This means getting away from the exclusive focus on agriculture. Agriculture is only infrequently the main source of income for the poorest households – it should not constitute the sole source of development in an area where the potential for agriculture is quite low. Considering the problems poor people have finding work and marketing their farm produce, one future project goal might even be rural, peri-urban and urban integration.

Recommended thrusts for interventions

Anchor projects more solidly in local institutions and the local economy.

Future IFAD interventions in disadvantaged areas with vulnerable populations should seek to launch local development more solidly rooted in local economies and institutions and based on the sustainable development of available potential: natural resources, agriculture, the cultural heritage and local products. They should also make the most of the available skills, resources and know-how of local stakeholders. Interventions could take a variety of forms depending on local specifics and what stakeholders want. Approaches of potential interest might build upon the EPIDELs initiated with FERT support in the context of AFD cofinancing for the Siliana project, support for the promotion of marketing channels with strong added value, support for the development of suitable financial services, and local development funds for decentralized financing of public investments, etc.

Favour participation by local stakeholders and partnerships.

The suggested approach for promoting local development is based on the concept of giving local stakeholders more initiative, a greater role in decision-making, and more responsibility. Fostering partnerships with local stakeholders in all their diversity should make it possible to mobilize all available resources and skills. Support should be made available to allow these stakeholders, especially smallholder farmers and their communities, to exercise new responsibilities.

Assign more weight to development support measures.

While the most recent projects have emphasized major public investments, it would be desirable in future to achieve a more balanced approach, stressing support measures. Appropriate measures concerning rural financial services, support and advice for producers and research and development should be implemented. Above all, IFAD should resume its commitment, assumed within the framework of the 1998 COSOP, of working in the financial services field, offering Tunisia its vast expertise in the development of rural financial services and policies for low-income rural households.

Enhance women’s access to production inputs and services.

Future projects should accord priority to women, especially women already working in economically productive activities. They should consider men and women’s specific needs and constraints at all stages of project design and execution, tailoring intervention approaches and mechanisms to suit. This particularly implies favouring women’s access to extension, training, trade and occupational organizations, technology and financing.

Recommended modes of implementation

Take action at two levels.

For projects that target a particular geographical area and have sufficient critical mass to drive a process of sustainable development, two levels of project intervention are suggested. The first would correspond to the délégation, or expanded local development council, and would see to the preparation of sound development strategies and help to mobilize a broad swath of partners. The second and geographically smaller level could correspond to the imadat (a socio-territorial unit), and this would be appropriate for efforts such as natural-resource management demanding closer proximity, as well as facilitating the organization of local residents and their representation at the higher level.

Extend institutional arrangements to the various local stakeholders.

Participation in future projects should be open to the various local stakeholders. The novelty and experimental nature of the proposed approaches also demand close links between project staff working in the field and those operating at the central policy-formulation level. Future institutional arrangements should be tailored accordingly, based on a careful review of the institutional capabilities and organizations of the various institutions involved. Functions having to do with project implementation could be organized in three tiers. For geographically based projects, the first tier of local operations might be the délégation, in the form of a discussion and decision-making structure made up of local stakeholders. The second tier would be project management, possibly at the level of the governorate, CRDA or some regional institution. The third tier, at the central level, would be responsible for steering and capitalization. The technical and management support needed at each tier should tap existing local and national capacity in both the government and the non-government sectors. NGOs, in particular, could be called upon for specific functions, as could social and trade associations.

Adopt more open project mechanisms.

Project mechanisms should continue to remain open for greater flexibility. This would involve defining a general framework for intervention as well as the relevant mechanisms and methods. Project formulation should consider the time needed to get these methods and mechanisms up and running, regarding it as a preparatory and learning phase to be backed by appropriate technical assistance and training.

Provide better support for project implementation.

Project implementation modalities should stress training for stakeholders, possible recourse to external technical assistance, the implementation of M&E systems for guidance, and regular supervisory missions with clearly defined objectives covering all aspects of implementation.