Interim Evaluation

In 1993, a General Identification Mission for Peru recommended that two projects be carried out: one whose objective would be management of productive natural resources in the Southern Highlands and another which would strengthen rural-urban linkages in the Puno-Cusco Corridor. That recommendation resulted in the formulation and implementation of both projects: (i) the Management of Natural Resources in the Southern Highlands Project (MARENASS) (loan 386-PE); and (ii) the Puno-Cusco Corridor Development Project (loan 467 PE).

The Interim Evaluation (IE) Mission visited the country from 3 to 27 April 2002. The Mission held meetings with an IFAD External Evaluation Mission that was visiting the country at the time and exchanged information, ideas and opinions. Meetings were also held, in Lima and Cusco, with an IFAD mission that was in Peru to formulate a new project in the Southern Highlands. In Lima, project-closing meetings were held with representatives of the Ministry of Agriculture (MINAG) and with the IFAD operations manager for Peru. The IE Mission focused on matters relating to: (i) analysis of the project's impact on the living conditions of beneficiary families; (ii) level of participation, ownership and empowerment; (iii) impact on social capital; (iv) impact on physical and financial assets; (v) environmental impact; (vi) sustainability of the results achieved by the project and possibilities for institutionalization after the project ends; (vii) innovative content of the project and replicability in other contexts; and (viii) cost-benefit ratio of the principal activities carried out. At the start of the Mission, a workshop was organized with the entire staff of the Project Executing Unit (PEU) in order to undertake a process of self-evaluation, four mini-workshops with the Area Coordination Offices, a conclusions workshop with the PEU and, finally, in Cusco, a project-closing workshop, attended by the principal public and private institutions associated with the project.

The IE consisted of the following phases: (a) revision of documents by each consultant prior to arrival in Peru; (b) field work in Peru; (c) analysis of the information and final drafting of the reports.

In keeping with the new IFAD evaluation process, a central learning consortium was formed, with the following members: (i) the IFAD operations manager in Peru; (ii) a representative of the Peruvian government; (iii) the MARENASS project director; (iv) the head of the cooperating institution, the Andean Development Corporation (CAF); (v) IFAD operations evaluator in Latin America and the Caribbean; and (vi) the head of the IE Mission. An Expanded Learning Consortium was also established.

MARENASS is a Special Project of MINAG, with technical, administrative, financial and managerial autonomy. The project headquarters is located in Abancay, in the department of Apurímac. The cooperating institution is the Office for Rural Development Operations of the CAF. The project was launched in February 1998 with an initial transfer of funds to 99 rural communities, chargeable to their technical assistance funds. In March of the same year, the first "talking map" competition was held.

MARENASS was initially proposed because it was recognized that in rural areas, and especially in the area targeted by the project, the natural resources situation had deteriorated to a critical point, particularly in the areas occupied by the beneficiary communities. This region is home to a significant portion of the country's rural poor population, and one of the causes of the high rates of poverty is directly related to the deterioration of productive natural resources. In these areas, management of productive natural resources should be seen as a crucial weapon in the battle against rural poverty, while also recognizing the human factor on which all efforts to analyse, reverse or curb the process of deterioration hinge.

The project area is between the south-eastern and south-central regions of the Republic of Peru. It encompasses all the provinces in the departments of Apurímac and some provinces in the departments of Ayacucho and Cusco. The total surface area covered by the project is 55 869 km². The project area, when MARENASS began, was the poorest in the country. The poverty index was 2.6%—considerably above the national average of 2.0%. Chronic malnutrition affected 83.7% of the population and the unemployment rate was 12.9%. Only 24.3% of the population had access to safe drinking water and 29.2% to electricity. The area had been severely affected by violence during the 1980s and 1990s (in 1992, there were 800 victims of violence in the department of Ayacucho alone). The target group for the project consists of the farm families living in the communities in the area.

The general objective of the project to increase the amount of arable land and enhance the commercial value of the productive natural resources of farmers in the Southern Highlands of Peru. The specific objectives were: (i) assessment of the extent and nature of natural resource degradation in the rural communities of the target area; (ii) ongoing identification of efficient ways (traditional and modern) of recovering, conserving and producing natural resources; (iii) implementation of participatory methods to facilitate transfer of the technologies identified; (iv) strengthening of community and functional structures needed to ensure expansion of the agricultural frontier and sustainability of natural resource management; (v) support for the process of clarifying land rights; and (vi) dissemination of the experience acquired at the local, regional and national levels.

The project has two fundamental components: (i) training and technological change, which constitutes the productive strategy and is aimed at training and social assimilation of traditional and modern technical knowledge as solution alternatives, as well as horizontal dissemination of this knowledge, with rural participation; the competitions occur under this component; and (ii) production support services, a component that is part of the overall support strategy for the project and complements the first component.

The appraisal report on MARENASS proposed a methodology for training, production support and natural resource management based on: (i) farmer-to-farmer training in the use of technological alternatives for natural resource conservation and recovery; (ii) transfer of resources to communities to enable them to contract for technical assistance services and thus develop a market for such services; (iii) the Pachamama Raymi (celebration or worship of Mother Earth) methodology, which entailed the organization of competitions for the dissemination-extension of technological alternatives; (iv) "seed" funds for marketing and funds (awarded through competitions) funds for developing new farmland (expanding the agricultural frontier). Through the Pachamama Raymi methodology, resources and responsibilities were transferred directly to the communities.

The appraisal report estimated the project cost at USD 19.1 millions. The expected contribution from the OPEC Fund did not materialize, however, and total financing for the project has therefore been USD 15 142 800, of which 79% was contributed by IFAD, 19% by the Government of Peru and the remainder by the beneficiaries themselves. The direct beneficiaries of the project are 360 communities formed by 52 800 families. The project goals included reconstruction of approximately 2 000 hectares of terraces, the extension of irrigation to another 3 000 or 4 000 ha, doubling of the carrying capacity of pasturelands and planting of 500 trees and/or shrubs per family.

Initially, project operations were managed by a project coordinating unit (PCU) under the International Contracts Administration Unit (UGECI) of MINAG. In September 2000, the Project Executing Unit (PEU) was created. This unit, though also affiliated with MINAG, operates with technical and administrative autonomy. During the year 2000, the situation in the country was particularly unstable owing to the fall of the third Fujimori government, the institution of a provisional government, political and institutional uncertainty, and the occurrence of a new electoral process that concluded in mid-2001 with the establishment of the present government.

The PCU detected a certain lack of consistency between the objectives set out in the appraisal report and the methodology proposed under the project. It therefore reformulated the general development objective more precisely as follows: to increase the capacity of communities and families to carry out their own development activities in a sustained manner, exercising their civic rights and duties, in a framework of gender equity. This "redefinition" made it possible to focus the project activities on improving the living conditions and overcoming the poverty of community members by developing their capacity for natural resource management.

During 2002, MARENASS has begun withdrawing, as planned, from the first 99 communities that were incorporated in 1998; 261 agreements with the communities that entered the project after that date remain in effect. In the communities that continue to have a contractual relationship with the project, a greater effort will be made to incorporate issues in the competitions that were not envisaged in the original project design, such as improvement of housing and rural businesses or microcredit activities, using the production and marketing funds entrusted to the organized groups of women (OGW). Specifically, this work will focus on the following areas:

• Training and Technological Change Component. Training of technology suppliers (yachaqs) and community leaders (yachachiqs) will be strengthened as a means of optimizing private technical assistance services through them, and technical assistance fund transfers will continue.
• Pachamana Raymi Competitions. The five basic thematic focuses of MARENASS will be maintained: (i) livestock management; (ii) crop and water management; (iii) soil management and conservation and forestation; (iv) management of range- and pasturelands; and (v) improvement of housing, gardens, community beautification, refuse collection and other activities that will help improve quality of life. The experience and comments of the communities suggest that it would be best to hold competitions involving entire families or communities.
• Production and Marketing Funds. Support for the organized groups of women will be stepped up in order to boost their participation.

The policy of establishing partnerships with municipal governments will be continued because, of the communities included in the project since 1998, most of those that were awarded funds for the development of new farmland have received significant support from municipal institutions. Municipal participation has been especially important in less-urban municipalities. The Environmental Education Program in schools will be carried out through an institution specializing in this area and will seek to encourage greater participation in training for community leaders.

Principal outcomes of project implementation

As of December 2001, 20 015 families in 360 communities (an average of 55 families per community) located in 69 different districts had participated directly in project activities. Hence, the total goal of 360 communities had been reached. One of the basic project activities has been the "talking maps," an activity which was also the subject and content of the first competition. The project and the community use the talking maps to establish goals and a plan of action that begins with training and dissemination activities within the community. The talking maps portray the community graphically at three levels: the past (30 years before the project), the current situation (as of the project start-up date) and the future (in 20 or 30 years). Based on these talking maps, each year communities develop a community plan of action. This is an instrument that enjoys wide social acceptance (bolstered by the competitions between communities) and that forms the true basis for "real and participatory " planning in the community.

The main tool of the project is mass dissemination and application of known and proven technological alternatives for improving natural resource management. The project has achieved the full introduction of at least ten technologies (organic fertilizer, terracing, irrigation by means of a system of channels, gardens, pasture management, slow-formation terraces, biological insecticides, gathering of native seeds, management of bovine, ovine and camelid livestock, etc.), which have led to substantial progress in natural resource management and recovery and have boosted production substantially. The family greenhouse gardens have been particularly successful and have enhanced diets and food security.

Notable progress has been made in the physical execution of the project, except for construction of terraces. This is because families have preferred to focus on plots of land located closer to their homes and devote their efforts to building terraces of smaller size but better quality. As for irrigation, in most cases improvements have been made to existing plot irrigation (and the corresponding infrastructure), rather than constructing new infrastructure to irrigate other areas. Several irrigation projects of differing scopes are still under way.

Although complete and appropriate management of community land is not yet a reality, the process has begun. The training and dissemination methodology used under MARENASS, Pachamama Raymi, uses the modality of competitions in which rural families participate voluntarily, competing among themselves, first within individual communities and then between communities. Families' participation is motivated by the project-financed monetary awards given to the winning families in each community and the winning communities organized at the supra-community or microbasin level. The three keys to the methodology's success are the innate "competitiveness" of the farmers, the possibility of wining cash prizes, and the fact that Pachamama Raymi is basically managed by the beneficiaries themselves, reducing the project's presence to a minimum.

Work in regard to resource management and conservation is organized and executed by the communities themselves, using their own means: families or communities make investments beforehand (mainly in labour, but also in materials) and although they may later win an award, it will never equal the value of the investment.

The competitions between communities are the instrument that has made it possible to achieve two objectives: first, community cohesion and, second, mass dissemination of resource management techniques and their subsequent application. Although the level of participation in the competitions between families is quite variable (averaging 40% of the families in each community), by decision of the assembly, the competitions between communities necessarily involve all the families in each community. The competition and the award provide thea strong initial impetus. Later, concrete results become the incentive to continue with the practices introduced: production improvements that translate into higher earnings for the farmers, thanks to more effective use of their productive natural resources and the consequent appreciation in the value those resources, which constitute their main asset.

The farmers believe that natural resources begin with their own dwelling and extend from there to encompass the garden, animal corrals, farmland, irrigation, organic production and pastures. They are reluctant to upgrade their corrals or farmland before "putting in order" or improving their own houses. They have a logical conception of an undivided whole that is the space within which their lives and their productive and reproductive activities take place. In this context, improvement of housing, with the introduction of stoves, terracing, painting, pens for guinea pigs, latrines (even showers!), shelves, tables, construction of room dividers and refuse collection have all had a very strong impact on the living conditions of community members. The methodology of comprehensive interfamily competitions involves the entire family in a reappraisal of the roles of heads of household, women and young people.

The system of transferring resources directly to the communities necessitates the opening of bank accounts by each community that participates in the project. This methodology, which was tested and validated under the Promotion of Technology Transfer Project to Peasant Communities in the Highlands (FEAS), is a powerful instrument for involving communities in civil society and in the formal economy. In addition, it helps empower communities by making them directly responsible for managing resources and by strengthening community organization.

The funds for organized groups of women(OGW) have helped finance microbusinesses managed by the women's groups. In several communities mixed groups of men, women and young people have been formed. Most of the groups organized have their own bank accounts, and those that do not use the community account. The MARENASS funds are transferred into these accounts, as are the revenues of the microbusinesses. The businesses managed by the women run the gamut from agricultural production and livestock breeding and fattening to micromarketing and microcredit operations, which extend loans directly to users under agreements established by the group itself, stipulating the form of repayment. Some groups are also working to preserve biodiversity through the recovery of seeds of native species and the development of small nurseries. This fund has achieved remarkable success: the average of the capitalization process for the project as a whole (all 360 communities have one or more women's groups) is around 50%.

A computerized monitoring and evaluation (M&E) system was established only in July 2001 (three years after project start-up). However, there is a sizeable information base deriving from participatory assessments used to develop the talking maps (past, present and future) that form the basis for the community plans (the participants' objectives for the project). For all 360 communities, there is also graphic documentation of the situation prior to the project, as well as a pre-project statistical sample of 25 of the 101 communities served since 2001.

Impact of MARENASS

The project has had the following impacts on physical and financial assets: (i) family assets: physical assets, including the improvement of housing, corrals and facilities for animals, terraces, plot irrigation infrastructure, etc.; financial assets have been generated by reducing families' expenditures on external inputs (fertilizers and agrochemicals) and boosting their incomes through increased production and sale of surpluses; the monetary awards have also become, first, financial assets and, then, in almost all cases, physical assets; and (ii) community assets: meeting halls for community use, community lands, pastures, collection and distribution of water for irrigation, etc. In addition, some communities have won awards, increasing their financial assets, which in general have rapidly been converted to physical assets. Communities have also accumulated financial resources through transfers from MARENASS, fees paid by community members, and financing negotiated with other institutions, including municipal government institutions.

The increase in the area under cultivation, which was one of the objectives of MARENASS, has been relatively small, but it has occurred, thanks to the construction of terraces, especially in community areas, and to the rehabilitation or construction of irrigation infrastructure. Much more important than the increase in cultivated area are the improvements on family lands already under cultivation: terracing, sprinkle or channel irrigation, fertilizing with compost and humus, crop diversification and intercropping. These improvements have led to large and stable increases in production on family farms. The higher yields have not been due to the increase in cultivated area but to the improvements on areas already being farmed (in 260 of the 360 communities, 440 hectares of new terraces and 1 740 kilometres of irrigation channels were constructed and irrigation was improved on 2 475 hectares).

For the families, the impact has been significant in terms of the benefits obtained: production has doubled or tripled on a large part of the land farmed by each family participating in the competitions between families (20 000 families, approximately 46% of the 43 000 registered families in the area). Nearly 69 000 hectares of pastureland were fenced to facilitate rotational grazing; 262 kilometres of infiltration trenches and dugouts or pits were excavated to store close to 20 000 m3 of water, in addition to channels for temporary irrigation or drainage of wetlands (bofedales), etc. Construction and improvement of corrals and stables (more than 10 000), together with production of forage crops on irrigated terraces, have increased both milk production and the number of live offspring.

It is rather difficult to quantify the project's impact on the financial resources of families and communities. The most direct and visible impact is the increase in the amount of currency in circulation. Higher output, coupled with the sale of surpluses on the market after family needs are met, has translated into higher incomes, both in the short term (weekly, monthly earnings) and in the long term (income from the sale of animals). Fund transfers for awards have also had an impact. Close to USD 3 000 is being distributed annually through competitions within communities (between families, yachachiqs, and schoolchildren), and these amounts may be supplemented by awards received in competitions between communities (around USD 3 000 for an average of 5 communities).

The project has been remarkably successful in fostering widespread use of technologies that form part of the shared cultural heritage of the farmers but had been abandoned because, as the farmers put it, they had been "forgotten" or, more probably, they have been supplanted by technological paradigms that are suitable only for capital-intensive farming and lands with high potential. The most positive results are seen in the practices employed in the environment nearest to and worked most intensively by families, which have yielded improvements in the quantity and quality of the production for self-consumption.

MARENASS has achieved one important success: It has enabled at least 20 000 families of community members to move from a situation of subsistence and food insecurity to one in which they are rural producers with greater financial and physical assets, increased food security and production surpluses.

MARENASS's real impact of on physical and financial assets has been limited by its initial focus, which overemphasized productive and agricultural aspects and lacked a clear vision of the economic realities and characteristics of the beneficiary families and communities and of their own strategies for dealing with their economic situation and the contexts in which they operate. MARENASS continues to lack an approach and clear-cut proposals for addressing economic issues. Up to now it has supported the organization of local fairs and other one-time initiatives, but there is no ongoing strategy for this key area. The economic dimension has become the main obstacle to further progress. The most significant weakness of MARENASS is, without a doubt, the absence of a strategy and activities to promote linkage with the market for the families and communities (and the women's groups) that are beginning consistently to generate production surpluses.

Most of the families that work with MARENASS have placed less priority on expanding the arable area than on boosting the economic profitability of nearby lands already under cultivation, improving those lands in order to obtain higher yields. Their vision of space and land is based on overall living conditions, not simply on productive potential. Accordingly, they have quickly begun to channel a good part of their investments into improving their houses and immediate environment. In short, rather than simply pushing back the agricultural frontier, what the families and communities want is to expand their economic frontier.

The continued availability of a corps of technical support personnel within the communities (yachachiqs and, in some cases, yachaqs) has led to more effective training for families and technical support better suited to the conditions and systems of production in the community. The training system for farmers, which relies on yachachiqs and community leaders (men, women and young people) is making it possible to create a broad and growing human capital base. The formation of human and social capital is not complete, however; frequent changes of yachachiqs and community authorities necessitate an ongoing training effort which is far from finished.

The impact of MARENASS on human capital is directly related to the improvement in living conditions as a result of: (i) lightening of the burden of everyday tasks for the family, especially women; (ii) greater, more varied and more stable production throughout year (with a consequent reduction in vulnerability); (iii) refurbishment of physical assets and improvement of homes; and (iv) acquisition of new goods (increased family economic activity). These improvements are behind the optimism expressed by all those interviewed. Security about their potential for growth utilizing the resources at their disposal forms the basis for their claims.

The ideas about social and family equity disseminated through the gender and citizenship training, combined with the empowerment of participants and groups under MARENASS, have led to increased attention to and better—i.e., more equitable, effective and representative—distribution of benefits among the poorest sectors of the community. The improvement in women's status within the family and community has been due to better training for women, their capacity to manage funds and affirmative action. The greater visibility and prestige of women, and of their productive and reproductive role and contribution to the family, have led to a more equitable distribution of benefits and responsibilities within the family and have enhanced their status and position and the amount of respect they receive.

The most immediate impact of the improved practices has been a reduction of women's workload, since it is women who have traditionally been responsible for feeding and herding animals and for small-scale sales of small species and agricultural products. Improvements in the quantity, quality and diversity of family production are making it possible for women not only to cover the basic needs of their families, but also to contribute financially (sometimes for the first time) to family income through retail sales of small surpluses. Feeling more secure about their families' well-being has given the women interviewed by the IE Mission a new sense of self-assuredness.

The impact of the strategy of improving land and productive areas promoted by MARENASS has been felt first inside the home. In all the families visited, the house has become an edifice with several rooms devoted to different uses (parents' room, children's room, kitchen and, sometimes, a parlour or living room). Animals have been put outside the house in corrals and stables, thanks to the practices promoted by MARENASS. Guinea pigs no longer live in the kitchen but in pens. Manure is now picked up daily and placed in compost/humus piles.

Most of the communities in the project area have been battered by the economic crises and social conflicts that have occurred in the country in recent years and by macroeconomic policies that marginalized them. Thanks to their work with MARENASS, community councils have been re-established or strengthened through the official designation and recognition of the communities, training of community leaders, selection and planning of joint activities (community plans) for the capitalization of social goods or community assets or the regulation of their use. Community leadership has been strengthened as a result of greater recognition and appreciation of the functions of the community councils.

In addition, the members of the community councils are now more motivated, thanks to the availability of (i) financial resources; (ii) knowledge; (iii) larger supplies of inputs and tools of better quality for carrying out community works; and (iv) the presence of support personal (yachaqs and yachachiqs). The communities' capacity to negotiate with local entities and other programs has been strengthened enormously. Community organizations have become stronger as a result of the development and transfer of responsibilities under the project, including planning for the community's future, managing funds and overseeing family and community natural resource management processes; regulating livestock grazing in communal areas; and participating in competitions between communities. Community organizations have thus been legitimated and strengthened in a period in which, owing to the erroneous association of the idea of community with "collective," they tended to be dismissed as development agents because it was assumed that they would hinder the "private" initiatives of families.

The most prominent impact is the spectacular dynamics of planning and action ("ordenamiento") that characterize families and communities, encompassing everything from family homes to family parcels of land to communal areas. There has been a true mobilization that is involving a growing number of families in the three or four annual cycles of competitions between families.

MARENASS project activities have emphasized "capacity-building," recognizing that local stakeholders are pivotal to its facilitating interventions. The project has thus had a very great impact on the proficiency of these stakeholders in regard to all three dimensions of the concept of "capacity": knowledge, know-how and ability to take action.

The impact that managing funds has had on the capacity of members of women's groups to engage in commerce (bartering and sale of small surpluses) has enhanced the prestige and empowerment of those who have done the managing (the women). In most cases, the women's groups have mastered the concept and practice of teamwork (pooling and joint marketing of goods; mutual support among participating families) and avoided letting the individualistic attitudes that result from the break-up of communities prevail.

A true empowerment of local stakeholders has occurred, and this empowerment continues to grow. MARENASS has demonstrated that social capital is the key to improving the local ecosystem. Now it is necessary to cultivate (capitalize on) the lessons derived from experience, to share them among the people already involved and also with other, external actors who might become involved in the dynamic processes triggered by the project or become commercial, institutional or academic partners in the processes under way. The aim is to facilitate opportunities for coming together, negotiating, and encouraging mutual respect and understanding beyond the community environment—i.e., in the microbasin, the province, etc.

The families and communities participating in MARENASS have taken ownership of the project and, with it, of something that they feel was theirs already: the terraces, the houses, the water, the pastures, a technology with a high labour content that produces high returns with little or no external input. But, above all, they have taken ownership of a "friendly" project that has offered technologies that are within their reach and rooted in their culture and ancestral practices. The project's sustainability depends largely on this concept of "regaining ownership" and on acceptance of the idea, oft-repeated by the community members: "We are MARENASS."

Use of the various agricultural production technologies and strategies proposed under the project immediately boosted the productivity of the land and doubled family production. Moreover, the use of organic inputs (made from ingredients available in the community) has improved the quality of the products and helped spread knowledge and practices for the control and prevention of pests and blights, thereby reducing losses.

Improvements in the care of stables and corrals and in feeding of animals have led to a increases in meat, milk and cheese production and gradual improvement in the quality and breeds of animals (in addition to the savings or availability of funds that they bring to the family). Although the diversification and intensification of livestock production has emphasized animals that develop rapidly (smaller species), an increase and improvement of the cattle available in the communities is also discernible.

Through training, families and communities have (re)assumed responsibility for the management of local ecosystems. This is evident in the community plans for the future, in the intense dynamics of working with natural resources and in the development of capacity (greater knowledge, use of techniques and skills, negotiating agreement on and authority to implement common measures) for managing the process. Other activities have also been undertaken to reduce pollution, both by adopting organic farming practices that utilize fewer agrochemicals in general and by applying various environmental sanitation techniques, such as sanitary landfills, latrines, etc.

"Ordenamiento" (putting in order, upgrading, improving) is the term used to describe the set of actions undertaken by local stakeholders, beginning with the house—the family base and main element within the ecosystem—and encompassing all of the land and the entire community. The idea of "ordenamiento of the home, the farm, the community" has become an important organizing principle for mobilizing action. Local stakeholders have progressively taken over the management of their natural resources, starting with those belonging to family groups and then encompassing communal resources.

At the national level, MINAG and the Ministry of Economy and Finance (MEF), particular since 2001, have shown a keen interest in the achievements and advances of MARENASS. The results attained have suggested to both ministries that they should conduct a careful analysis of the project methodology with a view to replicating it in similar areas of the Highlands and other areas in which rural poverty is prevalent. The National Compensation and Social Development Fund (FONCODES), which is the institution responsible for implementing the Puno-Cusco Corridor Development Project financed by the IFAD, is also very interested in the experience of MARENASS.

At the local level, especially in its interaction with municipal institutions, MARENASS has had a significant impact. In 1998, municipal governments served as the entry point for establishing the initial contacts with rural communities. Now, they are evolving from communication and procedural channels to true cofinancers and stakeholders in the project strategies. MARENASS has made great strides in collaboration with local governments; at present, some 20 districts located in the provinces of Chumbivilcas, Espinar, Grau, Aimaraes and Puquio have signed on to the project.

Two fundamental approaches applied by the project stand out as most important: (i) MARENASS does not ask farmers what their problems are but what they want to do and how, without attempting to "lead" them or impose anything on them and working within the framework of the MARENASS proposal; and (ii) flexibility of proposals and in the transfer of responsibilities and financial resources to communities.

With regard to the technologies proposed by MARENASS, in view of the results seen by the farmers and their expectations for the future, some beneficiaries have declared: "They have returned our history to us and they have returned us to history."

Fully 80% of the funds have been transferred (in monetary form) to the beneficiary communities and families. The total cost (operating costs plus transfers) per family participating in the activities of MARENASS is around USD 350 (calculated on the basis of the 260 communities participating in the project at the end of the year 2000). That cost would be even lower if all 360 communities, consisting of some 50 000 registered families at the end of 2001, were taken into account. In that case, the cost per family would be no more than USD 150. At this time, it is impossible to estimate the added value attributable to the investments, capitalization or increased present and future incomes that families might earn. However, it is possible to conjecture based on case studies, specialist consultancies and the sampling work of the IE Mission. As a very cautious estimate, for every United States dollar spent under MARENASS, families and communities have probably made investments in physical assets worth between three and five dollars.

MARENASS has one truly innovative aspect: it has succeeded in synthesizing the lessons learned from prior isolated, fragmentary and incomplete experiences and putting them into practice. The project has also awakened the creativity of local actors, families and communities, which first experimented with and then adopted and replicated the techniques proposed by the project, all of which were based on local resources.

MARENASS has also been highly innovative in its working methodology. Working from a design based on previous innovative experiences, such as the Pachamama Raymi training system or the lessons from the FEAS project regarding transfer of funds and responsibilities for technical assistance, and applying a approach of dialogue and interaction with communities, the project has refined and integrated those experiences. MARENASS has viewed its mission as "working with people" more than "working with natural resources" through the application of techniques specifically for that purpose. This orientation has been manifested in the objective of "capacity-building."

With regard to the MARENASS methodology, it would be foolhardy to reduce it to a certain instrument or group of instruments. The approach and criteria are essential. Unquestionably, the talking maps, the system of competitions, the fund transfers, etc. had many merits. But without an approach oriented towards "facilitation" and without the commitment to "work with people," those mechanisms would have been sterile, becoming mere means to achieve the end of reaching goals or clients. The MARENASS methodology is the outgrowth of a series of methods and instruments (proposed during the project appraisal) and effective approaches (some prompted by the Pachamama Raymi methods and others arising out of the interaction between the project and the communities). Its key elements are: (i) building trust in communities; (ii) cultivating relationships; (iii) financing/rewarding results; (iv) developing talking maps; (v) farmer-to-farmer training; and (vi) transferring funds to the communities.

Conclusions

After four years of work under MARENASS, it can be concluded that to achieve success in the fight against poverty under the conditions that prevail in the highland region, especially the Southern Highlands, it is essential to reclaim and increase the productive value of natural resources—the most important asset available to families and communities. It can also be concluded, however, that a mere proposal for improving land through the management of natural resources will not suffice if it is not recognized that the people who live on these lands have their own ideas and skills and that those directly concerned—families and communities—should be at the heart of any proposal put forward.

The MARENASS methodology, because it emphasizes capacity-building and development of relationships, and does not focus only on "objects" (technology transfer as such), has made it possible to take full advantage of the virtues of the Pachamama Raymi method and the funds provided by the project, whose most visible result is that communities are managing their own program of experimentation and learning and are helping to multiply existing potential at the local and regional levels.

MARENASS shows that the demands, vision and expectations of the poor families and communities in the Highlands are not oriented solely towards solving problems related to the degradation of their natural resources. Rather, their efforts are aimed at other objectives, such as starting rural businesses, improving their homes and maintaining local roads. In short, they are seeking to improve their quality of life as "part and parcel" of managing their natural resources.

It can be stated that the levels of food security attained are sustainable, given the full incorporation of the technology and productive methodologies proposed under MARENASS. The technologies introduced, which require very few external inputs and involve no recurring costs, also seem sustainable. Likewise, the achievements with regard to level of organization, institutional empowerment of communities and management of funds seem solid and probably sustainable (at least in part).

MARENASS has yielded a long list of lessons learned: (i) the natural resource management practices proposed under the project have great potential for rapidly improving household economy and achieving food security, the top priority of local actors; (ii) there is a potential for investment by families and communities themselves in the Southern Highlands; (iii) a project such as MARENASS cannot be limited simply to a "productive" or "agricultural" approach; (iv) organic farming holds tremendous economic promise; (v) the system of yachachiqs demonstrates the potential of community organizations to oversee an "internal educational system"; (vi) the social impact of MARENASS has been greater than expected because no attempt has been made to impose organizational models; (vii) as an instrument of planning and monitoring, the talking maps are extremely useful in facilitating the development and execution of projects by the beneficiaries themselves; (viii) the transfer of funds to communities fosters empowerment; and (ix) capacity-building and development of relationships are two key aspects of MARENASS's intervention.

The IE Mission has tried to pinpoint the main reasons for the project's success. In the Mission's view, the following four key elements, among a considerable number of other factors, explain why MARENASS has proved so successful:

• Project design. The design combines creativity, scientific rigor and knowledge of the area and of the local reality. The extensive discussion and input from different sources and experiences that went into the design of the proposal have been crucial in developing a clear and sufficiently flexible strategic basis for effective implementation: (i) the incorporation of previous positive experiences in the direct transfer of resources to organizations; (ii) the understanding that productive natural resources are the most important form of capital available to community members and that capital has, for the community members, cultural and social significance, not just economic and conservation value; (iii) outsourcing of service delivery and the farmer-to-farmer system; (iv) the role and effectiveness of social organization in the Andean community and the approach to empowerment; (v) proven technologies requiring few external inputs; (vi) "friendly rivalry" as an instrument of motivation, (vii) a participatory methodology and genuine transfer of responsibility for decision-making to families and communities, and (viii) a very modest PCU/PEU headquartered in the project area.
• The project's form of implementation and the system of management. Marenass adopted project start-up procedures already tested under other projects, which yielded positive results. The methodology included direct support from the Corporation for Regional Rural Development Training (PROCASUR) and IFAD in the selection of personnel through open competitions, based on strict criteria. Particular care was taken in selecting the project manager. The human resources were trained prior to assuming their responsibilities, and for the most part the team has remained the same throughout the project. The project's management has been a crucial factor in its success, and the management methodology should be analysed and studied with a view to applying it in other projects. The management of human resources has been very innovative for a project of this type, as have the communications and decision-making systems. MARENASS has worked essentially to establish and develop "relationships" and to foster trust with regard to the agreements and commitments made with families and communities. This meant building TRUST and SOLID RELATIONSHIPS.
• The capacity of the PCU/PEU to learn, listen to and understand the views and culture of the communities and families. MARENASS offers a system of participation and true empowerment. The project transfers resources and, most important, responsibility for the management of those resources, based on precise agreements/contracts, to which MARENASS always adheres. The resources are awarded through competitions, and results, not intentions, are rewarded. One of the keys to MARENASS's success is having understood the role, the environments and the functions of the families and communities—roles and interests that are different but interdependent. MARENASS works at both levels.
• Support and oversight; reflection, transfer of experience, impetus. The IE Mission considers this another of the essential elements. MARENASS was open to and sought out—with support from the IFAD office in Lima—conceptual and methodological, and even tactical, support and contributions from a considerable number of specialists who have spent long periods working in the field. Doing so required a positive attitude on the part of both the project and the consultants and, above all, a capacity for conceptual and methodological supervision and oversight, a function which in this case was performed successfully by the IFAD office in Lima.

Conclusions

The success of MARENASS is prompting various public and private institutions to try to incorporate the strategy and methodology applied under the project into their operations. It is recommended that, in these desirable replication processes, it be borne in mind that the MARENASS methodology comprises a group of mechanisms, tools and approaches and cannot be reduced to a single instrument or method. The essential element is the focus and approaches that have enabled the Pachamama Raymi strategy to go beyond simple technology transfer and generate an intense and promising dynamic in the hands of the beneficiary families and communities.

The major weakness in the project, and the main obstacle to the sustainability of effective natural resource management, is its lack of an economic approach to the management of productive natural resources. This deficiency has two main facets: failure to calculate the cost of the recommended practices and the ambiguity of the proposals for achieving greater linkage with the market.

It is necessary to develop, with the communities, an economic approach to the sustainability of natural resource management that reflects the economic strategy and viewpoint of the families and communities. The word "economy" should be understood to mean the possibility of expanding options and selecting those that are most desirable in keeping with the vision and expectations of community members.

Thus far, MARENASS has not been able to adequately support families and communities in meeting these new challenges, but it would not be advisable for the project's efforts to be dissipated on this new task. It is essential and urgent to address this issue, but it will be necessary to look for complementary funds and resources outside of MARENASS in order to do so effectively. MARENASS might "drive" and guide this process, which could be entrusted to external service providers under contract. After having achieved good results with regard to natural resource management and "agricultural" and "productive" matters, it is imperative to devise approaches and activities for dealing with economic issues and increasing access to and ties with the market.

Specific recommendations:

• Production and Marketing Fund - Funds for Organized Groups of Women (OGW): It is recommended, in order to consolidate the women's groups, that activities be planned to: (i) train their members (men and women) to manage rural businesses; (ii) support the shift, within the women's groups, from a farmer mentality to a business mentality; (iii) identify business modalities that are not limited by legal constraints; (iv) seek similar experiences in microcredit and microenterprise to capitalize on experiences; (v) provide training in all the foregoing areas in the Quechua language; (vi) integrate the need to monitor businesses into the thinking of the community and, especially, into negotiations with women.
• Fund for contracting community facilitators: While the role of the facilitators has been basically positive, there are some risks and weaknesses at the moment. It is recommended that the MARENASS strategy be revised, with greater emphasis on training for community facilitators, clarifying their relationship with the community and the role they are to play in it.
• Award fund: It is recommended that the possibility of strengthening interinstitutional relationships be explored, in particular with the municipal governments, in order to maintain, at least partially, the availability of funds for awards in the communities from which MARENASS has already withdrawn or will soon withdraw.
• Fund for expansion of the agricultural frontier: It is recommended that the use of these funds be extended to competitions between communities that have been participating in the project for all four years.
• Withdrawal process: In 2002, the first 99 communities that were incorporated in 1998 concluded their participation in the project. A lower level of activity has been perceived in these communities relative to those that began participating in 1999. It is recommended that the strategy and proposal of the PEU for this withdrawal process be defined taking into account the fact that the 1998 communities probably deserve some "compensation" for their contribution to the learning curve of the MARENASS personnel.
• Agreements in force: It is necessary to ponder, with the communities, the future of the "agents" that have emerged as a result of MARENASS (yachachiqs and community facilitators).
• Institutional relations: This is one of the weak points of MARENASS. It is recommended that a greater effort be made to establish alliances with municipal institutions and with other programs and projects.
• Environmental education in schools: It is recommended that the performance of the Environmental Education Program be reviewed, involving the communities.
• Gender: It is recommended that gender training be continued and strengthened, based on the current strategy of focusing on the family and the issue of the rights of persons and organizations.
• Monitoring and evaluation: It would be advisable to strengthen and integrate the current systems so that they can perform the function for which (intentionally or not) they were created, establishing clear objectives within the systems; linking each one to a computerized information base containing not just numbers but also narrative text; equipping the personnel responsible for collecting information in the community with digital cameras and secretarial support to transcribe the information collected; training project technical personnel to read maps and interpret photographs; and corroborating graphic and verbal information with numeric data on concrete results of proven validity and reliability.
• Dissemination of information: This should be a shared responsibility of the entire PEU, but it should be coordinated and systematized based on activities undertaken by the Monitoring and Evaluation (M&E) Unit.

Recommendation for IFAD and the Borrower

The success of MARENASS is prompting several public and private institutions to try to incorporate the strategy and methodology applied under the project into their operations. It is recommended that, in these desirable processes of replicating the methodology, the following considerations be borne in mind.

The MARENASS methodology entails a process of analysing and systemtizing experience and an effort to delineate (perhaps even in the form of manuals) the processes and steps, duly analysed and discussed, so that this experience can be applied in similar areas. This systematization should describe all the main elements that have led to the achievement of positive results: (i) project design, (ii) staff selection and profile, (iii) staff training, (iv) implementation and presentation-dissemination, (v) ways and methods of relating with the beneficiaries, (vi) administration of a project such as MARENASS, in particular capacity for management; (vii) detailed processes and steps for applying the methodology; (viii) modalities and negotiations for establishing local and institutional relations.

It is recommended that national authorities, the institutions concerned, MARENASS and IFAD initiate a policy dialogue and an analysis/debate in order to begin "translating" the MARENASS experience into policy lines and methodologies for rural development, especially in the following areas: (i) transfer and management of funds by organizations; (ii) outsourcing and market for services; (iii) natural resource management and organic farming; (iv) technology transfer and farmer-to-farmer training; and (v) the role and relationships of communities and families.

It is suggested that additional resources—both financial and professional—be provided to support MARENASS in formulating viable proposals for developing and ensuring the sustainability of organized groups of women (microcredit, microbusiness). MARENASS should be supported through proposals and studies aimed at consolidating the process and identifying specific activities to "even up" the development of the communities that began participating in 1998.

It is recommended that resources be allocated for the development of proposals that will facilitate the application, in the short term, of the experience of MARENASS in projects already under way, such as the Puno-Cusco Corridor project and projects of other institutions.

MARENASS includes a series of innovative features and elements that create opportunities and point up the need to undertake studies that go beyond functional analyses for the purpose of implementing other projects or designing rural development policies. It is recommended that opportunities be sought to carry out studies of an academic nature that will yield a better understanding of anthropological and cultural mechanisms and aspects and propose scientific interpretations and explanations of the cultural and social factors behind what was observed empirically under MARENASS.

It is not considered advisable to proceed with a second phase of MARENASS, but an expansion/replication initiative is considered appropriate, given the high demand among communities wishing to participate that cannot currently be served by the project. At present, MARENASS is serving around 7% of the registered Andean communities. If the project's success is recognized, it might be desirable to carry out a dissemination and replication initiative. One fact that illustrates this success: the overall cost of MARENASS is around USD 40 000 per community, or an average of USD 350 per family, and it has been demonstrated that capital formation in the communities very quickly surpasses that amount.

Reallocation: An agreement should be reached among the PEU, MINAG, the Borrower, IFAD and the cooperating institution to review the amounts allocated in the budget to each category of expenditure. The total lack of financing for the Fund for Expansion of the Agricultural Frontier (because funding from the expected source did not materialize) should be remedied, since this is a strategic fund, especially for the consolidation stage in communities that are completing their cycle of participation in the project.

Some of the funds in the "unallocated funds" category could be shifted into "monitoring and evaluation," and the categories of operating costs, salaries and wages could be adjusted. A proposal to that effect should be prepared by the PEU, bearing in mind the operational plans for the next few years and the incorporation of the IE Mission's recommendations.

Expansion/replication: MARENASS is currently serving around 7% of registered Andean communities. If the success of the project is recognized, it might be desirable to carry out a dissemination and replication initiative. One fact that illustrates this success: the overall cost of MARENASS is around USD 40 000 per community, or an average of USD 350 per family, and it has been demonstrated that capital formation in the communities very quickly surpasses that amount.