Document of The World Bank Report No: 31033 IMPLEMENTATION COMPLETION REPORT (TF-26659 TF-26655) ON A GRANT IN THE AMOUNT OF US$ 30 MILLION TO THE BRAZIL FOR A RF Demonstration Projects - PDA September 20, 2004 This ICR was prepared based on the findings of project analyzes prepared by the client and monitoring data available, as well as various contributions from the Implementation Agency and the project partners. The ICR was prepared by Werner Kornexl (LCSES) and Christoph Diewald (consultant). CURRENCY EQUIVALENTS (Exchange Rate Effective December 2003) Currency Unit = Real R$ at completion Cruzeiro (CR$) at appraisal Cr$306.69 = US$ 1.00 US$ 0.0036 = Cr$1.00 Rate at completion R$ 3.00 = US$ 1.00 FISCAL YEAR ABBREVIATIONS AND ACRONYMS ABC - Agência Brasileira de Cooperação, Ministério das Relações Exteriores Brazilian Agency of Cooperation, External Affairs Ministry AMA - Unidade de Apoio ao Monitoramento e Análises de Projetos (PPG7) Monitoring and Analysis (PPG7) APA - Área de Proteção Ambiental Environmental Protected Area BASA - Banco da Amazônia S/A Amazon Bank BB - Banco do Brasil Bank of Brazil BIRD - Banco Internacional de Reconstrução e Desenvolvimento International Bank for Reconstruction and Development CE - Comissões Executivas do PDA e do PDPI (conforme contexto) Executive Commissions of PDA DED - Deutscher Entwicklungsdienst (Serviço Alemão de Cooperação Técnica e Social) Germany Service of Technical and Social Cooperation FNMA - Fundo Nacional para o Meio Ambiente Environmental National Fund FUNAI - Fundação Nacional do Índio, Ministério da Justiça National Indian Foundation FUNBIO - Fundo Brasileiro para a Biodiversidade Brazilian Biodiversity Fund GAP - Grupo de Análise de Subprojetos (no âmbito do PDA e PDPI) Subprojects Analysis Group (PDA and PDPI) GAPEP - Grupo de Apoio à Preparação e Execução de Projetos (no âmbito do PDA e PDPI) Supporting Group to the Preparation and Execution of Projects PDA and PDPI) GEF - Fundo Mundial para o Meio Ambiente, Nações Unidas Global Environmental Facility GTA - Grupo de Trabalho da Amazônia (Rede de ONGs) Amazon Working Group (NGOs Network) GTZ - Agência Alemã de Cooperação Técnica (Gesellschaft für Technische Zusammenar-beit GmbH) German Technical Cooperation G7 - Grupo dos 7 (países mais ricos do mundo) Group of Seven IBAMA - Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis Brazilian Institute for the Environment and Renewable Natural Resources KfW - Agência Alemã de Cooperação Financeira (Kreditanstalt für Wiederaufbau) German Development Bank MMA - Ministério do Meio Ambiente Ministry of Environment MRE - Ministério das Relações Exteriores Ministry of External Affairs ONG - Organização Não-Governamental Non-governmental organization OSCIP - Organização da Sociedade Civil de Interesse Público Civil Society Organization of Public Interest PDA - Projetos Demonstrativos (subprograma do PPG7) Demonstration Projects (PPG7) PDPI - Projetos Demonstrativos dos Povos Indígenas (parte do PDA) Demonstration Projects of the Indigenous People (part of the PDA) PGAI - Programa de Gestão Ambiental Integrada (PPG7) Integrated Environmental Management Program PNUD - Programa das Nações Unidas para o Desenvolvimento (UNDP) United Nations Development Program PROAMBIENTE - Programa de Desenvolvimento Sustentável da Produção Familiar Rural da Amazonia Sustainable Development Program of the Amazon Rural Familiar Production PROMANEJO - Projeto de Apoio ao Manejo Florestal Sustentável na Amazônia Forest Resources Management Project PRONAF - Programa Nacional de Fortalecimento da Agricultura Familiar National Program to Strengthen the Familiar Agriculture PRORENDA - Programa de Viabilização de Espaços Econômicos para Populações de Baixa Renda Promotion of Smallholder Family Agriculture PROVÁRZEA - Projeto Manejo dos Recursos Naturais da Várzea (PPG7) Floodpain Management Project RESEX - Reserva Extrativista Extractive Reserves RFT - Fundo Fiduciário da Floresta Tropical, Banco Mundial Rain Forest Trust Fund, World Bank RMA - Rede Mata Atlántica Network of NGOs for the Atlantic Rain Forest SAF - Sistema Agro-Florestal Agro-Forest System SCA - Secretaria de Cooperação da Amazônia, MMA Amazon Cooperation Secretariat SPRN - Subprograma de Políticas de Recursos Naturais (PPG7) Natural Resources Policy Program - NRPP ST - Secretaria Técnica do PDA Technical Secretariat of PDA UC - Unidade de Conservação Conservation Unit UE - União Européia European Union UG - Unidade de Gerenciamento Management Unit Vice President: David de Ferranti Country Director Vinod Thomas Sector Manager John Redwood Task Team Leader/Task Manager: Werner Kornexl BRAZIL RF Demonstration Projects - PDA CONTENTS Page No. 1. Project Data 2. Principal Performance Ratings 3. Assessment of Development Objective and Design, and of Quality at Entry 4. Achievement of Objective and Outputs 5. Major Factors Affecting Implementation and Outcome 6. Sustainability 7. Bank and Borrower Performance 8. Lessons Learned 9. Partner Comments 10. Additional Information Annex 1. Key Performance Indicators/Log Frame Matrix Annex 2. Project Costs and Financing Annex 3. Economic Costs and Benefits Annex 4. Bank Inputs Annex 5. Ratings for Achievement of Objectives/Outputs of Components Annex 6. Ratings of Bank and Borrower Performance Annex 7. List of Supporting Documents Project ID: P006571 Project Name: RF Demonstration Projects - PDA Team Leader: Werner L. Kornexl TL Unit: LCSRF ICR Type: Core ICR Report Date: December 30, 2004 1. Project Data Name: RF Demonstration Projects - PDA L/C/TF Number: TF-26659; TF-26655 Country/Department: BRAZIL Region: Latin America and the Caribbean Region Sector/subsector: Forestry (100%) KEY DATES Original Revised/Actual PCD: 07/13/1993 Effective: 09/01/1994 04/15/1996 Appraisal: 11/21/1993 MTR: 09/01/1996 12/01/1999 Approval: 10/20/1994 Closing: 12/31/1998 12/31/2004 Borrower/Implementing Agency: Banco do Brasil (Recipient) / MMA/Secretaria da Amazônia (implementing agency) Other Partners: GTA, RMA, KfW, GTZ, FFEM STAFF Current At Appraisal Vice President: David de Ferranti Shavid Javed Burki Country Director: Vinod Thomas Rainer Steckhan Sector Director: John Redwood Constance Bernard Team Leader at ICR: Werner L. Kornexl Robert Kaplan ICR Primary Author: Werner L. Kornexl 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The Project in its Context: The project is part of the Pilot Program to Conserve the Brazilian Rain Forests (RFPP) and was one of the first projects appraised within that Program. The "Memorandum of the Director" (appraisal report) was approved and the Grant Agreement with Banco do Brasil (BB, the recipient) was signed in 1994. The latter became effective in early 1995. Sub-project implementation began in 1996. The RFPP was created in 1992 with the support of the G-7 countries, the European Union and the Netherlands, as a joint program with the Brazilian Government and civil society. Its long-term goals are to reduce the rate of deforestation and support sustainable development in the Amazon and Atlantic Rain Forests of Brazil. The RFPP was planned as a set of twelve (mostly pilot) projects with different lines of action. These included strengthening state institutions, environmental management, experimentation with and dissemination of new forms of natural resource use and conservation, "extractive reserves", indigenous land demarcation and protection, forest and floodplain resources management, biodiversity conservation, and science and technology generation and dissemination. It has been implemented by different public sector agencies with civil society participation under the leadership of the Ministry of Environment (MMA). The RFPP projects were designed largely independently of each other, but there have been efforts to integrate them during implementation. The "Demonstration Projects Type A" (PD/A1 There was to be also a "PD/B", a similar small grants project aimed at municipal governments rather than at communities and NGOs. This project was never prepared., hereinafter referred to as "the project") was specifically designed as the component of the RFPP that would contribute to the trial, adaptation and dissemination of innovative, promising and sustainable development approaches mainly by community associations and NGOs, but also by local governments. The rationale for promoting the sustainable use of natural resources was to reduce pressure on existing forests (i) through the rehabilitation and use of already-cleared areas, (ii) through the economical and low-impact use of existing forests, or (iii) by conserving biodiversity in protected areas and elsewhere. From the very beginning, the participation of civil society, local communities and stakeholders was a guiding principle of RFPP, although the government was initially reluctant to accept this. Moreover, most government institutions had little or no experience in working together with such groups at the time. This was in part due to the failure of earlier investments that didn't manage to garner sufficient social and political support. Community-based initiatives were considered to be important for overcoming destructive patterns of natural resource use (deforestation), and for duly recognizing local traditional forest-related knowledge, but also as a means of building constituencies for sustainable development. Strategically, the inclusion of largely excluded communities was thus an important element for supporting policy changes. In a sense, the project became the flagship of integrating civil society not only in RFPP, but also in environmental policy-making in Brazil. Accordingly, the project was prepared with maximum participation from emerging civil society organizations. Within the RFPP framework, the project was specifically directed towards the demands and expectations of civil society and its organizations. It had more participation from civil society in decision-making and implementation than other RFPP projects. The latter were geared more - 2 - towards governmental or academic "clients", although most of them also benefited from civil society's considerable participation or even co-management. Worldwide, there had been few if any experiences of the kind attempted within the project. The notion of community-driven projects was gaining ground, but mainly with the objective of poverty reduction or as an alternative means of providing public services. There were few models to learn from that dealt specifically with natural resource conservation and use2. Since the project was not designed for poverty reduction per se, poverty targeting was not an issue, although it was expected that sub-project proponents would represent mostly poorer rural communities3. While at the time of appraisal there were already donor funds available for such projects in Brazil, there was still a large demand that remained unmet. The project was meant to fill this gap with a new funding mechanism as an alternative to the National Environment Fund (FNMA, created in 1989 and still operating). The latter had been deemed too cumbersome in funding civil society initiatives, mainly because it depends on the federal budget and its funds must adhere to budgetary and other federal rules and regulations. The project was to fund initiatives by rural communities, NGOs and municipal governments in the Amazon and the Atlantic Forest regions. The two regions differ markedly both economically and socially, as well as in their level of development. The Atlantic Forest is a highly threatened biome along the Brazilian coast that also extends into the interior across many states. Most of Brazil's economic activity and population is concentrated in this region. Only some 8% of its original forest cover still remains. The Amazon (5 million km2 according to the legal Brazilian definition), which has suffered relatively less devastation so far (about 16% of forest cover), is among the poorest and most sparsely populated regions of the country. The name "Demonstration Projects" implies showing successful experiences to others, the intention being that they would adopt them. This was indeed a key objective of the project's second component, while the first component was about innovation and experimenting with novel ways of using and conserving natural resources from the rain forest prior to presenting and disseminating them. Forest conservation and management, community participation and the sustainable use of natural resources were not widely accepted concepts in much of Brazilian society and the public sector in the early 1990s. Rural communities received scant development assistance mainly for education, health care, technical support and extension (this still holds true to a great extent). The main pillars of economic development in the Amazon were large-scale undertakings, such as roads and hydro-energy, and particularly cattle ranching, without significant environmental and social impact mitigation. Sustainable resource use was still emerging as a concept. The capacity of communities to find their own solutions to sustainable resource use was largely untapped. Some knowledge and many untested hypotheses about adequate technical and environmental solutions in both biomes existed, but were rarely used. Moreover, community-based sub-projects were perceived by some to be running against the political and economic mainstream. Thus, the project concept was truly groundbreaking, also in a political sense. The situation has changed significantly since then, as the project has set positive - 3 - examples in both regions, consolidating knowledge and practices, building "social capital" and contributing to an increase in environmental and political awareness. Civil society organizations were still emerging in 1992, seven years after the end of the military government's time in power. The environmental NGO community was fragmented and not well organized during that period, particularly in the Amazon. Through the project, the RFPP attempted to strengthen civil society networks in both the Amazon and Atlantic Forest regions. Both networks, the Grupo de Trabalho Amazônico (GTA, Amazon Working Group) and the Rede Mata Atlântica (RMA, Atlantic Forest Network) were the political and, to a lesser extent, technical counterparts in the preparation and implementation of the project and the RFPP, in which they have been playing a very active role since the very beginning. GTA was first formed in 1991 in response to the RFPP's announcement. It represents a diverse mix of community-based associations, including "traditional populations" (such as fishermen, rubber tappers, Brazil nut gatherers, indigenous peoples), rural labor unions and other smallholder associations, as well as more conservation-oriented NGOs and larger advocacy groups. There is also a wide variety of objectives and stakeholder interests within GTA and most of the member organizations have few, if any resources. GTA's objectives have been (i) to build up a network for the exchange of information and opinions, (ii) to monitor and participate in the process of RFPP preparation and implementation, (iii) to organize civil society for the discussion on the social and environmental issues related to Amazon development, (iv) to monitor relevant government policies, (v) to hold a dialogue with and lobby government, and (vi) to promote mechanisms for the participation of local populations in development policies. At the time of appraisal, GTA had already grown to 300 member organizations; now there are over 500. The group has headquarters in Brasilia and offices in 15 regions of the Amazon. RMA was created in 1992 with the objective of promoting exchanges among member NGOs working on Atlantic Forest conservation issues, but also to represent civil society in that region in the RFPP. The number of member organizations grew from ten to eighty at the time of appraisal, much less than GTA. Today there are some 250 member entities. In RMA, most member NGOs focus on environmental conservation and rehabilitation, and less so on sustainable development, although the latter is gaining increasing attention. At the time of appraisal, RMA had no office infrastructure of its own. This ICR is based on official Bank and government records, on the Brazil Rain Forest Unit's institutional memory, experiences of the task manager and government staff, and a number of evaluation studies. The most important of these are two recent studies of twelve sub-projects in the Atlantic Forest and twenty sub-projects in the Amazon region (see Annex 7 for references). Although these studies are based on a non-random sample of sub-projects and are not rigorously scientific, independent evaluations, they contain a wealth of insights and lessons learned. Evaluation of Objectives The project had three stated objectives: - 4 - (i) to stimulate the development, adaptation and dissemination of environmentally, economically and socially sustainable systems of natural resource management and conservation by local communities through a small grants program; (ii) to inform other communities, groups that represent them, researchers, and government policy makers about the lessons learned from local sub-projects (dissemination); and (iii)to strengthen the capacity of local populations, and to plan and implement activities leading to conservation and sustainable development through support to the NGO networks, the GTA and RMA. As reflected in the latest CAS (2003), these objectives are still highly relevant to the Bank and Brazil today. In fact, the relevance of these objectives and the demand for community-driven development projects has increased over time. Community-based projects, environmentally and economically sustainable land use, and learning and dissemination of best practices have become part of the Government of Brazil's (GOB's) strategy to combat increasing deforestation rates. Today, both GOB and the Bank are seeking ways to mainstream successful pilot experiences on a larger scale. The objectives (which were formulated in different ways throughout the appraisal report) covered most elements of an innovative, first-of-a-kind pilot operation, such as testing, capacity building and dissemination. However, the treatment of evaluations and lessons learned in the objectives - a critical step before adapting models and mainstreaming for general use - was vague. While recognizing the importance of testing, adaptation, monitoring and evaluation, the objectives jump from experimentation almost directly to dissemination, without first requiring solid answers to the questions "what works where, and under what circumstances?". This had consequences on implementation. The true intent of the third objective, which aims at community-level capacity building, is not entirely consistent with the way it was originally stated. While it implies that the primary beneficiaries of capacity building would be the communities that carry out sub-projects, the intent was actually to provide support to the NGO networks as such, as a principal objective, not only as a means of capacity building in communities. Strengthening networks as an instrument to publicize the project and providing technical assistance to potential and actual proponents were no doubt legitimate efforts. However, the actual intent for Amazon NGOs "to act effectively and in a coordinated fashion on environmental matters" should have become a separate objective, or even a distinct project (as was done during recent years for both networks under the RFPP). Also, this third objective already defined the main channel of capacity building (the networks), thus virtually excluding other alternatives. The assumption that the networks could provide truly "technical" assistance was unrealistic, as they had little capacity to fulfill this task. Both networks contributed, however, to the preparation and implementation of the project, which had two other mechanisms for providing TA to communities (GAP and GAPEP, as discussed below). The five-year project implementation period was too short to achieve the intended objectives. The choice of five years is most likely due to both Bank and donor practice, which rarely permits longer project periods at the design stage, as well as the short time horizon of the RFPP itself (originally planned for just three years). Sub-projects were limited to three years at most (unless a - 5 - second grant was awarded). This is considered insufficient for testing and developing innovative practices with the complex social, entrepreneurial and agronomic processes that exist within communities. After adding the minimum amount of time for preparation, selection and approval, there was little left to analyze the results of the sub-project and lessons learned, and to disseminate them adequately for other sub-projects to use. Not surprisingly, the project was extended twice, in 1999 and 2002. 1 There was to be also a "PD/B", a similar small grants project aimed at municipal governments rather than at communities and NGOs. This project was never prepared. 2 The social investment funds (such as FONAMA in Bolivia, FONCODES in Peru, and the Rural Poverty Reduction Projects in the Brazilian Northeast) had no specific natural resource management dimensions, and had largely limited community choices from a selection of pre-designed investment models. 3 The word "community" is admittedly only vaguely defined. It may be taken here as group of usually poor families in a particular rural area, confronted with similar problems and with a minimum of common concerns, cohesion and willingness to participate in an organized way. Proposals were accepted only from legally established associations or from a formal NGO (which may represent a community). 3.2 Revised Objective: N/A 3.3 Original Components: Development, adaptation and dissemination of environmentally, economically and socially sustainable systems of natural resource management and conservation by local communities Well-designed for the most part, this component introduced a number of innovations that have contributed to the project's success. Particularly noteworthy is the acceptance of a "bottom-up" approach to natural resource management, i.e., the definition and adoption of models by communities, rather than the selection and execution of models that were pre-defined "at the top". Proponents were to receive grants of up to $210,000 for sub-projects of 1-3 years. The project introduced appropriate eligibility criteria for proposals and proponents, which were largely applied (witnessed by the fact that less than 20% of the proposals received were actually approved). For instance, the rules excluded associations formed only for the purpose of making a proposal ­ they had to exist for at least one year prior to that. Proposals also had to address a set of predefined "themes". In addition, there were detailed criteria for environmental sustainability. The project required counterpart contributions from the beneficiaries, ranging between 10% and 30%. Sub-project screening included a formal and a technical filter before a proposal was presented to the Executive Commission. The project's Technical Secretariat (ST) screened sub-projects for eligibility and requested clarifications and adjustments from the proponents. Thereafter, two parallel, independent technical reviews of the proposals were solicited for each sub-project from consultants selected from a predefined roster (GAP) before they were sent to the Executive - 6 - Commission for approval. In cases where a proposal was considered worthwhile but weak, consultants from a predefined technical assistance roster (GAPEP) were called to visit and assist the proponents in improving their proposals. A particular merit of the project design was the effective sub-project selection and funding mechanism: · The project has been coordinated by a Technical Secretariat (ST) in the Ministry of Environment (MMA), not by the Grant Recipient (Banco do Brasil, BB). · An Executive Commission, consisting of an equal number of civil society and government representatives, meets periodically to set overall guidelines for the project and to approve sub-projects for funding. · Two rosters of consultants, GAP and GAPEP provide assistance to the ST or to proponents in analyzing or reformulating proposals, and GAPEP also provides backstopping for projects under implementation. · The NGO networks help publicize the grants program among their member groups, provide training in proposal preparation and technical assistance to communities, and assist in disseminating lessons learned. · The Banco do Brasil (BB), as the recipient and manager of the funds, receives donor grants, enters into (civil law) contracts with associations (grantees) after approval of sub-projects, and makes disbursements to grantees when requested by the ST. Thus, grant funds did not go through the federal budget. This resulted in faster, more timely disbursements to beneficiaries and simpler procurement rules, and thus in better planning and implementation. Banco do Brasil, a large public commercial bank, has the widest network of agencies in Brazil and the capillarity needed to reach communities all over the country, even in remote areas (although it has never been truly geared towards serving communities). · UNDP, under an agreement with BB, contracted staff and consulting services for the ST, GAP and GAPEP, and was used also as a mechanism for training, evaluation and dissemination services. Civil society's participation in the project preparation and decision-making process was a main pillar of the design. "Sponsoring" NGOs often represented the implementing communities as sub-project proponents and were officially in charge of sub-project implementation and accounting for sub-grants. In some cases, community associations made proposals themselves, without an NGO intermediary. In other cases, there were NGOs proposing their own projects, without a community involved. Although municipal governments were eligible to propose projects, there was little demand from that side, mainly because the project was not really publicized to them. Many features of this funding and approval mechanism were replicated by other projects within the RFPP (PROVÁRZEA, PROMANEJO)4, while the mechanism was being considered by other government agencies, although FNMA continues to practice a more traditional, "governmental" approach. - 7 - There were, however, also some design shortcomings. For example, it did not include mandatory in loco appraisal of all sub-projects before approval, which would have increased costs but also provided better information than desk reviews. Approval was instead based only on the proposals submitted and on specific information from network or government representatives in the Executive Commission about a community's situation and capacity. Routine technical assistance (TA) for sub-project design and formulation (through GAPEP or otherwise) should have been the rule in a two-stage selection process, where pre-qualified proposals would have received TA for preparing the final proposal. Overall, we conclude that there were inadequate resources allocated to TA, particularly for the vast Amazon region with its remote sub-project locations. This was apparently due to the Bank's and the donors' desire to contain project costs and to channel a maximum of funds directly to final beneficiaries. The ceiling for sub-project grants (US$ 210,000) was too high to handle for groups that had never administered any such grants. The Bank and donors may have succumbed to NGO network pressure on that score. The denomination of the ceiling and of sub-grants in dollars went against GOB's macro-economic policy under the "Plano Real" of 1994 to eliminate indexing, and may have reduced incentives for the economic use of resources and sustainability. Dollar-denominated grants had been meant to shield against traditional hyperinflation. The "Plano Real" indeed introduced reasonable currency stability and could have offered the opportunity to set a limit in the national currency (possibly with some protection against inflationary price increases), but during negotiations there was still some skepticism as to the success of the Real Plan, after the failure of earlier such plans. The limit was reduced only in 2001 to R$300,000 (today about US$110,000), after a significant devaluation of the Real in 1999, and contracts were then drawn up in the local currency. Funding of working capital for sub-projects with productive and marketing objectives was not allowed. At the time of appraisal, it was assumed that this should be provided by the public credit system, but there was insufficient appreciation of the difficulties of first-of-a kind, high-risk community-based projects to access credit. Lack of working capital became a frequent bottleneck for processing and marketing by associations and cooperatives without access to credit sources when it came to paying suppliers (community members) for the raw materials delivered5. Since access to credit is an important element of sustainability, we conclude that grant financing of working capital should either have been allowed from the beginning (with rules for revolving funds and safeguards for the gradual capitalization of community enterprises), or that a mechanism for access to public-sector credit should have been built into the project design. Otherwise, the project should not have approved sub-project proposals that required working capital to be financially viable6. The project did not require a "business plan" or any kind of economic planning exercise from proponents of productive sub-projects, which would have been essential for obtaining credit, if not at the beginning, then at least after a startup phase. It is doubtful whether the kinds of initiatives proposed should have required full-fledged business plans, although TA could have used customary elements of a business plan to discuss a sub-project's feasibility with the proponents before approval. - 8 - 4 The general project idea was also adopted by the then ongoing, Bank-financed PLANAFLORO and PRODEAGRO projects, as part of their mid-term reformulation in 1995/96. 5 In sub-projects funded by Germany, some working capital was made available from grants as an "initial investment in a stock of merchandise". 6 Other programs funded by MMA (such as the Agro-Extractive Program) provided, at least initially, for funding of working capital, but they seem to show that this can also be problematic, if working capital is not appropriately managed and recovered for the following year's purchase of raw material. MMA ceased to allow funding of working capital, but the issue persists. Informing other communities, groups that represent them, researchers, and government policy-makers of the lessons learned from the local-level projects This component, which was also to include the evaluation of sub-project results, was not adequately designed. There was no provision for a mechanism that would produce quality material for dissemination-proven models and lessons learned. The evaluation of sub-project experiences was foreseen, to some extent, in the Monitoring and Evaluation (M&E) Plan attached to the appraisal report, but its objective was more to monitor implementation and project performance than to analyze what worked, what didn't, where and for what reason. The M&E Plan contained the elements and indicators of traditional project monitoring and supervision, financially and technically, and provided for the hiring of consultants to conduct evaluation studies. It required that all sub-projects define their own indicators, collect data and report on progress. It did not define a minimum set of common indicators to be measured by all sub-projects internally and did not require the collection of baseline data. Such indicators and data, while certainly increasing costs, would have been necessary for conducting more quantitative analyses and evaluations. Moreover, resources allocated to the evaluation and learning tasks were inadequate (but were later provided, in part, by GTZ). The GAPEP consultants were apparently expected to carry out such learning-related work. The project called, however, for an early study to identify and assess experiences from other projects, which were either completed or ongoing at the time, to have material ready for early dissemination. The lack of an early definition of the evaluation and learning process led to different expectations. Donors and GOB expected lessons for the mainstreaming of programs regarding which procedures, rules, interventions, technologies and projects proved successful. Associations and NGOs, on the other hand, saw project implementation more as an exercise of community-level learning-by-doing (which also requires its own indicators). That kind of learning experience was indeed very rich, judging from the evaluation reports. Both kinds of learning and evaluation are needed, but should have been foreseen from the beginning. Dissemination as such was also less than satisfactory. The appraisal report only mentions the general audiences and vehicles for dissemination, and allocates one staff in the ST to lead this task (to be carried out by consultants). It did allow, however, for the funding of sub-projects just for - 9 - dissemination purposes. Strengthening the capacity of local populations to plan and implement activities leading to conservation and sustainable development through support to the NGO networks This component did not do enough for strengthening the capacity of communities to plan and implement their projects ­ the stated objective ­ and did not provide the necessary tools. TA should have already been available for preparing proposals and during preparation, not only for technical or economic matters, but also for the communities' mobilization, discussion of proposals and internal organization. The component's actual content was the strengthening of GTA (and to a lesser degree RMA), so that it could "act effectively and in a coordinated fashion on environmental matters". The networks were expected to publicize the small grants program and provide some technical assistance to grantees, especially for proposal writing. There had been an earlier strategic planning exercise carried out by GTA with project preparation funds. Support to GTA then declined over a three-year period, after which it was assumed to have its own funds to carry on. The component had the clear goal of "creating a strong, independent association adept at articulating and pursuing the interests of its members at the end of the funding period". In the case of RMA, which was to receive a fraction of the funds allocated for GTA, only services geared towards supporting the project (publicity, technical assistance to sub-projects and dissemination of results) were expected. 3.4 Revised Components: N/A 3.5 Quality at Entry: Before project start-up, a well-prepared Project Manual, including forms for sub-project proposals, had already been approved by the Bank. This manual set down the rules and procedures, although it was changed several times later on, and established a benchmark to be emulated by similar projects in the RFPP. Quality at entry was substantially enhanced by the very active participation of organized civil society, leading to full ownership on the beneficiary side. This was an unusual feature at the time, certainly among most Bank-funded projects in Brazil. There were, however, some operational gaps in project readiness. At start-up, Banco do Brasil, ST, the Bank and KfW still needed time to agree on the standard contracts with the beneficiary associations. The project partners (BB financially and the ST technically) were not fully prepared to take on the project. BB was not entirely willing to take over its role in the project and gradually passed on the responsibility of scrutinizing sub-project expenditures and other tasks to - 10 - the ST. While BB first played an active role in disbursement and accounting, it gradually shrunk to a mere financial intermediary with few responsibilities7. This attitude is related to readiness for implementation, but not to quality at entry.This transfer of responsibilities increased the workload of ST even more. 7 Surprisingly, BB questioned, after grant effectiveness, a standard provision of the grant agreement that allowed the Bank to ask BB for recovery of funds used for ineligible purposes. BB argued that, considering the agreed fee it was to receive on disbursement transactions (1.65%), it would not be in a position to cover such liabilities. The issue was settled in 1996 by means of a letter of comfort from the Bank assuring BB that it would first consult with it in case of alleged misuse before taking any action. This attitude is related to readiness for implementation, but not to quality at entry. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: Component 1: Stimulating the development and adaptation of the dissemination of environmentally, economically and socially sustainable systems of natural resource management and conservation by local communities The project did not pursue a stated target in terms of the number of sub-projects funded. Out of the 1,047 proposals submitted to the ST, a total of 147 projects were actually implemented in the Amazon, and 47 in the Atlantic Forest ecosystem. (Only 19% of all proposals submitted received financing). Over half (659) of the proposals received were rejected as they did not meet the necessary criteria; 58 others were not recommended for funding by the GAP; 130 were not approved by the Executive Commission or the ST; and six grants were cancelled after approval. About 14% of the proposals sent to the Executive Commission by ST between 1995 and 2001 were evaluated as interesting but technically weak. In these cases, consultants from the GAPEP roster were asked to assist the proponents in revising their proposal. Out of 40 re-evaluated projects, only 19 were eventually approved by the Commission. By April 2004, 170 sub-projects had been concluded, and 24 are still being implemented (financed by KfW and the French Fund for the Global Environment, FFEM). All projects financed by the European Commission were fully implemented by the closing date (Dec. 31, 2003)8. The large percentage of rejections (81%) may be a reflection of the lack of technical assistance to proposing communities and NGOs and of the hasty preparation of proposals in order to respond swiftly to the opportunity of receiving a grant. This warrants further study. In terms of size, 89% of grant resources were assigned to large sub-grants (mostly up to the ceiling of $210,000), 8% to medium-sized ones and 3% to small ones. The high sub-grant ceiling reduced the number of grants that could be funded. Staggering ceilings in accordance with demonstrated prior experience with grants may have led to better size distribution and avoided the temptation "to go for the maximum". Smaller sub-projects appear to have been able to improve their social organization at least as well as the large ones, and thus may have been more efficient. - 11 - Almost half (47%) of the sub-projects focused on agroforestry systems (SAFs), and many of the others had SAF sub-components. These had become somewhat of a trend in forest conservation and rehabilitation of degraded areas in the 1990s. In many cases, they may have been imposed on communities by outside proponents without due regard for local conditions and farmers' interests. Their rate of success was not overwhelming. Almost a third (30%) of the sub-projects concentrated on forest management for the extraction of non-timber products (although there were also a few cases of management for timber), with the intent of increasing the economic value of standing forests through the collection of seeds, rubber, Brazil nuts, oils, medicinal plants and fruits being more frequent than management for timber. In some cases, these products were also processed by communities for value-added. The majority of grant beneficiaries were community associations and cooperatives, followed by NGOs and rural labor unions. In the Atlantic Forest, most of the beneficiaries were NGOs, whereas in the Amazon most were smallholder organizations. Sub-projects in the Amazon had more "productive" objectives, while in the Atlantic Forest environmental and conservation objectives dominated, at least initially. However, environmental NGOs in that biome became increasingly aware of the need to implement alternative sources of income generation to strengthen community support for conservation activities. Economic efficiency and sustainability were initially evaluated as the sub-projects' weakest aspects. Nevertheless, according to the latest evaluations, beneficiary producers did effectively benefit economically from adopting new land-use practices, by diversifying the production system, reducing inputs, promoting a shift to perennial crops, reducing or controlling the use of fire and making gainful use of the standing forest. Impacts on cash income and increased consumption of subsistence crops were rather modest (5-15%). Many projects were able to increase benefits through economies of scale and increased professionalism. Although not always quantified in the producers' self-evaluations and in reports, the reduction of expenditures on agrochemicals and other inputs, the diversification of farm production, the increase in crop yields, more rational use of natural resources and the reduction of work time were seen as the most important gains from agroforestry. Wherever raw materials were processed (e.g., fruit pulp, Brazil nuts), economic success was often hampered by inadequate planning and management capacity, compounded by a lack of experience in marketing and knowledge of markets. The complexity of such economic processes was underestimated by the communities that had very few entrepreneurial abilities. Nevertheless, with continued support from other donors and regional institutions, most cases should achieve economic sustainability over the coming years. The expectations of project proponents in terms of economic gains over a three-year period were unrealistic for the most part. This is not surprising, given the lack of prior experience and prudent technical assistance. For example, it normally takes well over three years for agroforestry systems to provide substantive gains. An evaluation right after the closing of a sub-project may often suggest poor sustainability, but later field visits frequently indicate a better picture as a result of internal learning and adaptation processes. In environmental terms, the Atlantic Forest sub-projects demonstrated especially positive impacts, not only by restoring native forests or implementing SAFs and other improved natural resource - 12 - management systems, but also by ensuring a broader debate on NRM issues in the sub-project regions and a change of attitudes. The integration of environmental-awareness campaigns, capacity building and environmental education, linked to the economic and social needs of the beneficiaries, were seen as most effective, while isolated activities rarely led to a change of habits. There were very few cases where there was a complete loss of investments and a failure to reach development objectives. In the technical evaluations carried out thus far, only about 6% of the sub-projects were classified as total losses, where neither technical nor organizational strengthening goals were reached. Although there are many more sub-projects needing follow up, this ratio is surprisingly low for a pilot project. Cases of misuse of funds for ineligible purposes were registered in less than 2% of the sub-projects. Where irregularities were noted, disbursement to the grantee was suspended, and in some cases restructuring of the sub-project was attempted. Where problems persisted, sub-projects were cancelled and the occurrences were registered officially, as required by the agreement with BB. MMA declared that the amounts in question will be included in the "Active Debt of the Union" (i.e., without onus to BB), to be recovered and returned to the respective donor (KfW or EC). 8 There was no funding from the Rain Forest Trust Fund (administered by the Bank) for sub-projects. EC funds were held in trust by the Bank and were used exclusively for sub-projects. Component 2: Informing other communities, groups that represent them, researchers, and government policy-makers of the lessons learned from the local-level projects As mentioned earlier, the design of this component jumped too quickly from experimental practice to dissemination. Good results were achieved nevertheless, even in the absence of a clear strategy on what to disseminate, to whom and how, and given the lack of a more systematic project-level learning mechanism. An early evaluation of experiences from other projects, as mentioned in the appraisal report and scheduled for the first year, was not carried out. A first in-depth evaluation of six sub-projects was conducted by FAO/CP in 1999, and a more comprehensive evaluation of the first 28 sub-project experiences was done in 2000. After 2000, with no new proposals to screen, the ST had more human resources free to focus on analytical work and learning lessons. Accordingly, more intensive monitoring and systematic learning took off only in 2001, after available funds for sub-grants had been committed. The Bank insisted strongly on this aspect of the project and valuable assistance was made available from GTZ. The final evaluation studies in the Atlantic Forest and the Amazon were conducted in 2002 and 2003, respectively, with a shared and consistent methodology, including common indicators permitting some quantitative analyses. These recent evaluation studies eventually produced extraordinarily rich material in terms of lessons learned. They are based on the interaction between evaluators and communities and their - 13 - leaders in 32 of the 194 projects, which also brought out the lessons drawn by the beneficiaries-actors themselves. They are also based on the conclusions that resulted from the team's comparison of approaches and outcomes from these (and other) sub-projects. The material produced is highly useful, even if quantitative data suitable for more rigorous analysis were not available. It should therefore be the basis for additional, in-depth studies, which might then also cover a broader selection of sub-projects. This result came somewhat late in the project process and still needs to be more widely disseminated and absorbed by both project management and proponents of new sub-projects, but it represents a satisfactory output of the component. Learning at the sub-project level could have been complemented by more analytical work based on a minimum of comparable baseline data and common monitoring indicators for most if not all sub-projects. This would strengthen the overall credibility of the lessons and provide an additional, well-founded basis in support of mainstreaming. The methodological approaches used, interactive evaluations in the field and the analysis of a common set of data, are all valid and should be applied in parallel. In addition, many visits to participating and non-participating communities were organized during project implementation. Such visits were highly effective as eye-openers and for practical demonstrations. Through them, substantial learning and dissemination took place locally, in the sub-project area, in neighboring areas and even farther away, leading other communities to take note of and adopt what was being demonstrated. The dissemination of experiences and adoption by surrounding communities (which needs to be more thoroughly corroborated) are positive indicators of the impact of project demonstration. Several regional and local seminars were organized by the ST, but many more were conducted by the sub-project proponents themselves with project funds. Numerous other interactions took place without ST intervention. Several local, regional and even national policy initiatives can be directly or indirectly attributed to the project and to active civil society participation. One of these is the PROAMBIENTE proposal. This is an innovative program to provide credit to farmers implementing sustainable production models, under which there is a rebate of interest payments to recognize "environmental services" provided by them, and the difference is funded from grant sources. The policies chosen by MMA for promoting conservation in the Atlantic Forests also reflect this initiative. Many ideas, innovations, concepts and approaches were adopted and also followed up by research bodies in the Amazon. In several cases, municipal governments and local institutions were influenced by novel projects in their jurisdiction. However, state governments have been less supportive. Partnerships with municipal governments were rare at the beginning of the project, but have increased over time. Sub-project performance was greatly enhanced wherever these partnerships have been maintained. (In some cases, however, reliance on partnerships with local or state governments led to disappointments, especially following changes in power in a municipality or state after an election.) In a successful experiment implemented in 2003, communities were trained to self-evaluate their experiences, problems and achievements. The results of such evaluations sometimes turned out to - 14 - be quite different from external ones, and yielded rich input for the discussion on how to best work with community-based projects. Self-evaluation turned out to be an essential instrument substituting for reliable baseline and monitoring data, and has now been adopted as a standard methodology for the second phase. The project produced various materials of satisfactory technical quality for dissemination: a newsletter, brochures and videos on sub-projects. While these were useful for awareness-raising by providing examples of actual experiences, they were rarely based on thorough analyses. Improved communication and dissemination strategies are part of the project's second phase (2004-2006). Continuous, thorough monitoring of the sub-projects would have been essential, but the M&E Plan defined at appraisal was only very partially implemented. During the first years, the ST had to concentrate most of its resources on the screening and approval of sub-projects. The project did not implement a management information system with financial and technical information on all sub-projects. The enormous diversity, geographical dispersion of sub-projects and often lack of accessibility made it hard to establish a working M&E system. Nevertheless, the ST made some 250 monitoring field visits to ongoing sub-projects between 1996 and 2003, initially concentrating more on financial and administrative capacity building. Later, field visits also included staff for technical monitoring. Most sub-projects were visited only once, due to insufficient staff and financial resources. Given the low ST staff rotation, many data were still available for final evaluations even in the absence of a structured system and on an informal basis. In 2001/02, designs began for a "second phase" (in which the Bank would not participate anymore as trustee). At that time, a more systematic learning process was still going on. Thus, not all potential lessons have yet been integrated into the design of the second phase. Component 3: Strengthening the capacity of local populations to plan and implement activities leading to conservation and sustainable development, through support to the NGO networks 9 As in the case of Component 2, the design of this component was not sufficient to achieve what was intended: making communities, associations and cooperatives stronger in terms of organization, production, marketing and representation of their interests and demands. Somewhat surprisingly, the organizational strengthening of communities was in most cases one of the most important and visible impacts of the project, as was increased environmental awareness and technical capacity. Significant improvements in management, social organization, the role of women and political representation were found in more than 90% of the sub-projects evaluated. Communities generally advanced in terms of their ability to deal with the outside world, their bargaining power, self esteem and technical, administrative and even political skills, even where external help was absent or inadequate. The project organized several early administrative capacity-building events involving almost 2,000 - 15 - producers. This was a key factor for successful financial management. Although financial reporting and procurement were a great challenge for the beneficiary associations, surprisingly few mistakes were made. Some 400 other capacity-building seminars were organized by the project, involving beneficiaries and other stakeholders. Almost half of these events focused on the implementation of agroforestry systems and a quarter on economic issues such as processing, marketing and management. The participatory approach was clearly taken seriously under the project. Without strong community involvement in planning and especially implementation, and without improvements in social organization, the sustainability of activities suffered after a sub-project grant was "closed". On the other hand, with ownership and stronger organization, sub-projects developed further, often even without further grant support. Despite the encouraging outcome, technical assistance to sub-projects, beyond training in financial and procurement management, could have been planned and organized better, and many of the communities' errors and omissions could have been avoided. Leaving capacity building to the NGO networks was not an adequate solution. While the networks themselves got stronger (see below), they were not sufficiently effective as such to provide or organize the technical support for individual sub-projects under this component. Few NGOs in the networks had the technical capacity and knowledge to provide such assistance. This was not properly evaluated at the time, although an objective assessment may have been difficult where the capacity of the major actors involved in the preparation was concerned. Indeed, few institutions in the country were equipped at the project's start-up to help with innovative uses of natural resources from the rain forests (particularly in the Amazon). These were mainly research institutes with too few staff to work with too many communities. The comparative advantage of organized civil society in mobilizing communities and assisting with their organization and project formulation was, however, clearly confirmed. Government teams simply could not have achieved as much as the NGOs did. In the case of sub-projects in the Atlantic Forest, more technical support was provided by the networks and their associate NGOs there than in the Amazon region. NGOs in the South and Southeast of Brazil have better technical capacity to assist with sub-project implementation and have more direct contact with beneficiaries than those in the Amazon. The lack of qualified TA was one of the project's most serious problems, and it is surprising that overall satisfactory results were still achieved under these constraints. Sub-projects managed to build up their own expertise and professional knowledge over time to some extent, without relying on competent TA. Today, some of the leaders and staff of sub-projects are considered and sought after as technical references. The true focus of the third component was, however, the strengthening of the regional NGO networks. This support lasted only a few years (1996-99), and was later followed by other grants from the Rain Forest Trust Fund directly to the networks, outside the PD/A project. Both GTA and RMA benefited from the grants they received under the project and managed to insert themselves more effectively into the political processes. They produced newsletters and other publications, held periodic assemblies and worked to channel information and opinions down to their member organizations and, to a lesser degree, from those organizations up to network - 16 - leaders. The Atlantic Forest Network opened an office in Brasilia with project funds to monitor and intervene in political processes related to Atlantic Forest conservation in the Brazilian Congress and Government. Funds for the Amazon Working Group (GTA) were mainly used for maintaining regional offices and for periodic meetings of its now more than 500 affiliated entities. GTA appears to have been less effective as a network and a unified voice on general development issues in the Amazon in the political process than RMA. This may be due to the fact that the stronger, well organized, more professional (and "urban") of its member organizations (such as ISA, Friends of the Earth, WWF, IPAM and IMAZON), perform analytical, advocacy and lobbying functions with considerable skill and impact on policies, with or without GTA. The latter had, however, substantial influence on the MMA, to such an extent that several individuals belonging to the network took over functions in the Ministry for a while. Civil society's influence, participation and decision-making power within the RFPP have clearly increased over time. This is largely, but not only, due to this project. At the same time, the government's receptivity towards civil society as a partner has also increased, leading to a much different governmental "culture" today than in the early nineties, at least as far as environmental policies are concerned. There were some particular challenges for the networks in assuming their role in the RFPP and in the project: w GTA had to reconcile its dual roles as advocacy group and support organization in the RFPP. The tension between autonomy and collaboration has marked GTA's existence and has occasionally led to charges of "disloyalty" on the part of government and donors, or of being too close to government and donors on the part of member organizations. Overall, however, GTA was not shy about speaking out and criticizing both government and donors. The need to play a mediating role, to cajole member organizations to perform under sub-projects, and to shoulder the pressures and frustration "from below" were novel experiences for the group. These had previously only been undergone by government. w Working with government, and more so with the Bank and donors, revealed a "culture gap", particularly with regard to contractual obligations related to accounting for funds imposed on the networks and other grantees, and observance of certain rules. The Bank was often perceived and publicly scolded as being overly bureaucratic, even controlling the networks at times. These challenges were compounded in 1998, when the management of GTA passed from an experienced professional to less experienced leaders from rural Amazonian NGOs, and when GTA agreed to plan and carry out a project for the mobilization and training of community leaders to prevent and control forest fires. These challenges, and a perception of little empathy on the part of the Bank and donors, seemed to have overwhelmed GTA's structure to some extent. w A third challenge was maintaining inclusiveness and effective representation within GTA, with its highly diverse mix of participating groups scattered over the vast Amazon area, dealing with different views, expectations, interests, levels of information and distances from political centers. This diversity has been both an asset and a strain on GTA. While the project - 17 - helped to interconnect the member organizations, democratic governance and "equitable participation among unequal partners" remain a serious challenge. There have been power struggles and allegations of concentration of power in the hands of a few regional leaders. An external evaluation found that with the growth of GTA since 1992, there was also a "gradual reduction in the level of participation in the network", leaving "most of the elected positions at the highest levels ... in the hands of a dozen persons", and leading to the "domination of regional meetings by a small group of core activists" (Berno de Almeida, op.cit.). w Lastly, there were problems with the flow of information, communication and consulting between the top of the network and the grassroots level. Meetings between the member organizations and the positions to be taken by the network have rarely been prepared from the "bottom up" and usually reflect the opinions and positions of the organizations' leaders. While similar challenges also affected RMA, a much smaller group, they have been mitigated by less diversity and inequality among its members and by more direct communication. The networks continue to rely financially on substantial RFPP grants. Expecting self-sufficiency after only a few years was unrealistic in the Brazilian context, where few NGOs have the capacity to raise funds from citizens and corporations, and government incentives for tax-deductible donations are fundamentally lacking. 9 This analysis draws, in part, on (a) Lisansky, J. and Sprissler, L. "Lessons Learned from the Rain Forest Participation in the First Decade of the Pilot Program to Conserve the Brazilian Rain Forest, World Bank, 2001, and (b) on Wagner B. de Almeida, A., "The Growing Pains of an Unprecedented Civil Society ­ Government Partnership in the Brazilian Amazon: The Case of the Amazon Working Group (GTA)", in Thinking Out Loud, Innovative Case Studies on Participatory Instruments, LCR Civil Society Team (PREM/ESSD), World Bank, 2001. 4.2 Outputs by components: N/A 4.3 Net Present Value/Economic rate of return: N/A 4.4 Financial rate of return: N/A 4.5 Institutional development impact: N/A 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: Given that Banco do Brasil is an agency controlled by the Federal Government, and that the actual implementing agency (ST) is part of a federal ministry, there should have been few, if any, factors affecting project implementation that were outside government control. The one factor that had an impact, the devaluation of the Real, substantially raised the amount of available - 18 - resources for the project. This helped to increase the number of sub-projects funded as well as the funds available to sub-projects with dollar-denominated contracts, and thus probably delayed the project's learning and evaluation effort. Local government changes sometimes caused severe difficulties for the implementing communities. This was manifested in the abrupt withdrawal of support or even outright opposition. 5.2 Factors generally subject to government control: Pilot Program projects were implemented quite independently from other government programs during the first years. After 2000, the interaction between the PD/A project and other governmental programs increased, and several other government staff members contributed to the project analysis, which led to a fruitful dialogue. Federal "ownership" of the RFPP and of this project has increased over time, even if counterpart contributions remained extremely low until the end of 2003, when the federal government decided to fund project staff and the operation of the ST. The rural community's access to credit, particularly for non-traditional agricultural purposes, is largely also a federal matter. This was not really addressed during the project, with the exception of a line of financing for agroforestry communities, which was, however, minimally disbursed. A factor outside federal, but within state government control, was the absence of a suitable extension service to assist rural communities in their endeavors. State extension agencies are largely ineffective, limited in transferable knowledge (traditional crops such as rice and beans, cattle), and certainly not geared towards innovative forms of natural resource use. New rules were established by the government over the recent years, making the contracting of consultants through UNDP (e.g., for the GAPEP, a non-bureaucratic and efficient instrument to provide quick and efficient technical assistance) more difficult. As a result, the burden of technical assistance fell to ST staff, who were ill-equipped and had too little time to take on this duty. 5.3 Factors generally subject to implementing agency control: The extraordinary motivation, dedication and effort on the part of project management and its staff in MMA, their identification with the project objectives and low staff turnover have greatly contributed to the results, as has the cooperation between the ST and civil society organizations and the networks. Factors that had a positive impact at sub-project level include: · Good local leadership; · Involvement of beneficiaries in sub-project preparation and implementation leading to ownership (although stakeholders other than beneficiaries could have been more involved, including local governments); · Preparedness to learn from other experiences; - 19 - · Capacity to plan in a participatory manner, considering local factors, taking advantage of opportunities, and making adjustments and corrections where needed; · Ability to take advantage of market opportunities, work with business partners and integrate economic and social development objectives; and · Work in partnership with local public and private organizations, seeking synergies. However, the deficiencies of the monitoring and evaluation system (including the lack of a management information system10) led to less than optimal analysis and learning. Also, regular technical feedback between the ST and beneficiary associations was mostly weak. Many of the (mandatory) sub-project reports referred to serious technical problems in the early phase of the project. Beneficiary associations had the impression that their reports were shelved by the ST (which they were for the most part). Too little effort was spent on addressing their problems and registering achievements, with ST having inadequate staff resources to help resolve the technical problems alluded to in the reports. An early warning system, through which associations could voice their problems more effectively, and prompt re-design of sub-projects with technical advice would have led to better results. The interaction between the ST and sub-projects improved over time, however, and hopefully will be stronger in the second phase, not repeating the same mistakes or approving projects with similar problems. 10 Project management failed to perceive, despite repeated Bank alerts, that 422,000 were still available in the final years of the EC grant and not committed to sub-projects. This amount was later cancelled, which could have been avoided by a better financial management system. 5.4 Costs and financing: The project was estimated to cost US$22.1 million at the time of appraisal. It was financed by grants made available from the European Commission, in the amount of 4.0 million ($4.5 million equivalent at time of appraisal), which were administered by the Bank; from the Rain Forest Trust Fund (RFT) in the amount of $3.0 million, from Germany through KfW, with 10.2 million (20 million Deutsche Mark, or $11.7 million equivalent at appraisal) and from France, through FFEM (US$ 0.9 million). Germany's technical cooperation contribution (GTZ) was not counted in estimated project costs. Total project cost also included an estimate of beneficiary contributions of $3.0 million expected from grantee associations, including GTA. No counterpart funds were required from the Federal Government or from BB, the grant recipient. Physical or financial contingencies were not included, as this was a financial intermediation project. The bulk of project cost was allocated to sub-grants (76%), another part to support NGO networks (3%), and the remainder to the various overhead tasks performed by the ST (sub-project selection, supervision, training, technical assistance, monitoring, evaluation, and management, 18%). About two percent was reserved for audit costs and fees for the grant recipient (BB). During the course of the project, Germany increased its contribution from 10.2 to 17.9 million - 20 - ($11.7 to $17.9 million at the original exchange rate). Most, but not all of the German grant had been disbursed by Dec. 31, 2003. Germany has, however, committed new resources to the project, making a second phase possible. The RFT contribution was increased from $3.0 million to $3.94 million, for additional funding for the work of the Technical Secretariat and for the institutional strengthening of the civil society networks, RMA and GTA. On the other hand, an (accidentally) unused amount of 422,230 of the EC contribution was cancelled in 2004. An additional contribution from the RFT of $2.25 million for a new Environmental Education component was negotiated in 1997, but never became effective, as BB did not countersign the amendment to the grant agreement. France made available an amount of 1.52 million (FF 10 million at the time) for sub-project grants, which had only been partially disbursed by Dec. 31, 2003. Appraisal estimates and actual disbursements under the project by the final closing date of Dec. 31, 2003 are shown in the following table. Disbursements are taken as equivalent to project costs (100% financing). For the beneficiary contributions, only the initial commitments in the approved sub-project proposals and in the plans of GTA and RMA are known, but no reliable records are available on actual counterpart contributions, most of which were in kind, not cash. Appraisal Actual Component US$ million % US$ million % Sub-grants 14.59 76 21.04 78 Support to GTA 0.50 3 0.93 4 Support to RMA 0.08 <1 0.45 1 Overhead tasks 3.56 18 4.02 15 Fees 0.36 2 0.52 2 Total 19.09 100% 26.96 100% Counterpart contributions 3.0 N.A. Due to an increase in contributions, the overall project cost rose 40%, and that of sub-project grants rose 44%. Support to the NGO networks rose 130% (and their counterpart contribution was not enforced). Spending for overhead tasks, which were of crucial importance under this pilot project (for monitoring, learning and dissemination), increased less than proportionally by only 13%, despite a higher workload. As the bilateral donors do not fund overhead costs, the RFT should have increased its contribution to the work of the Technical Secretariat even more than it actually did. Macroeconomic factors had more positive than negative impacts on fund availability. As the bulk of donor contributions was made through grants in Euro, the decline of the Euro-dollar exchange - 21 - rate from about $1.35 in early 1995 to 0$0.85 in late 2000, substantially reduced the dollar amount available for sub-grants and other purposes. On the other hand, the devaluation of the Brazilian real since early 1999 more than compensated for this decline, despite inflation. Increased foreign contributions combined with a real devaluation of the Brazilian currency made for a much larger amount available for sub-project investments. This should have resulted in a larger number of sub-projects than without these favorable factors, but due to the persistence of the ceiling of $210,000 and of the denomination of sub-grants in dollars until 2001, this resource-spreading effect was much less than it could have been. There have been five cases of irregular or unaccounted use of funds by sub-grant recipients (three funded by Germany and two by the EC, with overall disbursements of $252,435, about 1.26 % of overall sub-grant disbursements). BB cancelled all five contracts, notified the recipients of the irregularity and requested the return of the disbursed funds. BB refused, however, to take further judicial action to recoup the funds. Government (MMA) has since stated that the amounts in question will be included in the list of active debts to government, to be recovered from the recipients, which would relieve BB of its obligations. This course of action has not yet been finally and formally agreed upon among the parties (Bank, EC, MMA and BB in the case of EC-funded projects). An agreement will also have to be reached on the disposition of these funds (returned to the donor, or used in another (sub)-project) once recouped. The entries for sub-grants in the tables above and below include the amounts disbursed to the cancelled sub-projects. The funding of the project by sources, as of Dec. 31, 2003, was as follows: Component RFT EC Germany France US$ million Sub-grants - 3.40 16.76 0.88 Support to GTA 0.50 0.50 - - Support to RMA 0.25 0.09 Overhead tasks 3.12 - 0.90 - Fees 0.06 0.06 0.30 0.01 Total 3.94 4.05 17.96 0.90 Share 15% 15% 67% 3% (totals made not add up due to rounding) The German and French contributions are not yet fully disbursed, and Germany has made available further substantial contributions for the continuation of the project, including, but not limited to an Indigenous Peoples Development Project and a special line of funding for the Atlantic Forest region. Given the limited remaining funds in the Rain Forest Trust Fund, and since the German partners felt that there was no need for continued Bank involvement, the project is continuing without the Bank and another RFT grant. 6. Sustainability 6.1 Rationale for sustainability rating: - 22 - Sustainability in the case of this pilot project may refer to several distinct aspects: (a) the sustainability of the sub-projects financed, (b) the institutional and political sustainability of the project/program itself, (c) the financial sustainability of the NGO networks, and (d) the project's lasting impact on public policy. Sub-projects: The sustainability of sub-projects has social, economic and environmental aspects. In most cases, the final evaluations predicted good chances for sustainability in social and organizational terms, as communities did not abandon their undertakings after project support had ended. As a rule, the better the communities were initially organized and involved in the discussion, implementation and evaluation of their sub-project, the better the chances for sustainability appeared to be. Social organization was strengthened in almost all community sub-projects and is a key factor for economic and environmental sustainability. Even if sub-projects appear successful at the time of closing, this success must be sustained. There are not yet enough data available to assess whether this will be the case. Key long-term economic factors will be reasonable infrastructure (roads, communication, etc.) to link producers to markets, to ensure stable or growing markets for the products, access to credit at affordable rates and with compatible maturities, access to quality TA, and internal organization, rules and management apt to maintain the assets created, including working capital. It was largely beyond the scope of the project to create or guarantee these necessary conditions, except for internal organization. The resilience of successful undertakings in the face of inadequate access to infrastructure, technical assistance and credit, adverse market conditions or incompetent internal management, will be low if these factors should indeed be unfavorable. Access to infrastructure, and even to institutional credit, depends on the communities' ability to connect to and deal with the local and state "powers that be", apart from macroeconomic conditions and federal policies. It is difficult to assess whether all these factors will be favorable in the future. Environmental sustainability is a term with multiple meanings. It can refer to the long-term chances of the environment to be protected as a consequence of a sub-project (sustained positive environmental impact), or to the absence of negative environmental impacts that may eventually threaten the very basis of production and livelihood of communities. Given that most sub-projects made use of already "degraded" areas or of standing forests for low-impact extraction of timber and other products (rather than converting them to pasture or cropland), environmental impacts are likely positive, i.e., reduced deforestation, or improved habitats for biodiversity in agroforestry systems. There is the possibility of over-extraction of products from forests beyond a "safe yield", but the communities' increased awareness as to the fragility of local environments that has been observed should help prevent this. Where sub-projects were initially "development islands" in the midst of neighbors continuing to apply slash-and-burn practices, the project appears to have helped to disseminate the idea of using natural resources sustainably, even if others do not always follow suit. Larger-scale sub-projects, especially those implemented in the Atlantic Forest, have contributed a great deal to environmental awareness-raising around the sub-project by promoting seminars and workshops in partnership with local institutions. Most local and state governments, communities, and public and private institutions are gradually recognizing the concept and importance of "sustainable - 23 - development", even though their understanding of "sustainability" may vary widely. Institutional and political sustainability: The project mechanism has been broadly recognized by civil society, government and donors as an efficient and effective instrument. The prospects for a second phase of the project with donor support, the expectation that the Federal Government will continue to carry the administrative costs itself, and the fact that PD/A is now part of GOB's Pluri-Annual Plan are aspects that show a chance for continuity. It is unlikely, however, that Brazil would sustain such a program itself without support from donors or agencies like the Bank. The most promising factor for the political sustainability of the project as a program is the strong support of organized civil society, which will continue as a force even throughout changes of government. PD/A will likely remain as a government program beyond the piloting phase (although there may be two such programs, one for the Amazon and one for the Atlantic Forest). 6.2 Transition arrangement to regular operations: The second project phase will focus on the consolidation of promising sub-projects already funded in the first phase, strengthening their demonstration functions and disseminating best practices to other communities. A new line of sub-projects (PADEQ) will fund community-based projects on the Amazonian frontier ("arc of deforestation"), to support sustainable land- use practices and control and reduce the use of fire. A specific line of sub-projects only for indigenous populations (PDPI) already started in 2001, and includes lessons learned from this project. One of the objectives of the second phase of PDA will be the dissemination of good practices from projects funded during the first phase. The coming years will therefore be crucial for reaching out to neighboring communities, local institutions and other stakeholders. It remains to be seen whether the project truly gained knowledge about monitoring, dissemination and learning based on the lessons from the first phase. Since there will be a second phase without Bank/RFT participation, the short and medium-term continuation of the project has been secured. The real issue at hand, however, is "mainstreaming" - the adoption of proven models and best practices by more communities and smallholder farmers, and public support to such communities on a broad scale, beyond the limits of a Pilot Program. Such mainstreaming could involve lending by the Bank or IDB, GEF grants, or larger federal budget commitments. There are no such plans at this time, although proposed loans to the states of Amapá, Pará and Amazonas include or will include elements of the PD/A approach and of the lessons developed under this project. PROAMBIENTE, a proposal developed by civil society in the Amazon and embraced by MMA, is a smallholder credit program that also includes TA. It provides for recognition of producers' "environmental services", through the partial reduction of their interest or repayment obligations. The individual debt forgiven would be compensated from other sources (e.g., donors). The proposal is being discussed and carried by smallholder farmer organizations. PROAMBIENTE has been a priority program of the Ministries of the Environment and Agrarian Reform and might represent the most important mainstreaming action based on the PD/A project. However, the source of funds to cover debt or interest still needs to be secured. - 24 - A severe bottleneck to effective mainstreaming is the scarcity of human and organizational resources available for working with a multitude of communities on a professional basis, mobilizing, coaching, providing technical and organizational assistance, and monitoring performance and results. In the absence of a professional (public, private or mixed) support service with well-trained staff and the presence of active populations in many parts of the Amazon and Atlantic Forest, it is likely that there will only be as many community projects as there are able and willing civil or private sector organizations. Both donors and government urgently need to address this issue. 7. Bank and Borrower Performance Bank 7.1 Lending: The Bank showed flexibility and innovativeness in the preparation of this project. It supported novel institutional arrangements and took a stand for the active participation of organized civil society. Although lessons learned elsewhere were screened and largely incorporated into project design, the Bank, as an institution, could have done more to integrate practical (technical and economic) experiences from elsewhere. Despite including a monitoring and evaluation plan at appraisal, it should have insisted on the more detailed design and implementation of a solid monitoring, evaluation, learning and dissemination system in the project, if not at appraisal, then at the beginning and during implementation. The Bank played its role well as a partner to both government and civil society, helping the latter to get a foothold in project (and RFPP) preparation in a historical context of low government trust of NGOs. 7.2 Supervision: On the supervision side, the Bank should have insisted more on early baseline work (probably as part of each sub-project) and the design and implementation of a monitoring, evaluation and management information system. The Bank provided otherwise adequate technical assistance and maintained a healthy partnership with the donors KfW and the EC. The project greatly benefited from the technical cooperation between MMA and the German agency, GTZ. The latter provided technical assistance directly to the ST, helped effectively with its internal organization, with the hiring of consultants and in its support to technical evaluations, workshops and courses. This was crucial to the implementation of the project arrangements during the first years and also in the final evaluations. GTZ was probably less successful in establishing the spirit, system and methodology for monitoring, evaluation and learning, which were so important in a pilot and demonstration project. 7.3 Overall Bank performance: Satisfactory Borrower 7.4 Preparation: The Bank's main client in the preparation of the project was really MMA and its ST, rather than - 25 - BB, the recipient of grant funds. The MMA's team had little experience working with the Bank, particularly with Bank procedures, policies and rules, which were thus often perceived as rigid, even unreasonable. (The only other MMA experiences with the Bank were the National Environment Project and the National Biodiversity Project and Fund (GEF). There was no interaction between the Demonstration Projects team and the teams of the other projects.) The assigned MMA team worked well with civil society organizations, but other branches of government had, at least initially, difficulties in accepting new players in what had previously been an exclusive Government-to-Bank relationship. 7.5 Government implementation performance: The main client, GOB, fielded a team in MMA to manage the project, which was probably too small and under-funded for this type of pilot project (due largely to the insistence of the German donor who favored channeling a maximum of funds to communities), most of the time without any counterpart funds. There was an average of ten staff members, only five of whom were "technical". This is far less than what is required to evaluate, monitor, disseminate, build capacity, provide technical assistance and extract lessons. Although the staff was experienced and had diverse skills, the objectives of the project and the diversity of sub-projects required more professional experience in economics, conflict resolution and monitoring and evaluation. With effective support from the German agency, GTZ, the team performed quite well given the constraints, especially since many of BB's duties were transferred to the ST. Staff members were highly motivated and there was little turnover during the nine years of implementation. A public staff selection process in late 2003 finally fielded the government's own staff. The financial agent and grantee, BB, played a more ambiguous role. As a federal commercial bank, BB did not place high priority on the administration of grants (as against credit), especially to communities. It took its role under the grant agreements seriously, but with considerable reneging and grumbling, somewhat in contrast to its assigned "social role". BB's reluctance was also due to its perception that the fees charged on disbursements were low in view of the risks of a possible misuse of funds for which is might be held responsible. In the event, the Federal Treasury itself is assuming this risk. More had been expected from BB as grant recipient, at least in terms of financial management. 7.6 Implementing Agency: See comments above 7.7 Overall Borrower performance: Satisfactory 8. Lessons Learned - 26 - 1. Pilot projects should produce knowledge, first and foremost. Pilot projects, such as the Demonstration Projects, are meant to provide knowledge in "uncharted waters", to find out what works and what doesn't, in an effort to achieve given objectives. What proves successful may then be "mainstreamed" on a large scale. A pilot project must be designed to produce reliable knowledge, proven models and lessons from the start. We understand today that this implies that most of the following elements should be present in the design: w Pilot design should begin with brainstorming and defining the gaps in knowledge, uncertainties and possible pitfalls, formulating specific questions or hypotheses as to what conditions success or failure. w The "experiment" should be discussed with the actors (beneficiary community), so that everybody involved is clear about what needs to be found out and known. This does not mean that knowledge is the only or most important objective of a (sub-) project. Obviously, people want better income, employment, resource conservation, etc. in their projects. Managers of pilot projects and community participants should also not lose sight of the learning character. w Knowledge needs determined indicators ­ usually these are related to the objectives and expected results (goals), but also to the conditions under which these are achieved or not. Indicators may not always be quantitative, but they should be measurable. w Indicators and the methods used to measure them should be agreed upon with the actors of a sub-project. Preferably, little outside involvement should be required to do the measuring (observation, monitoring) and recording, and indicators should be measurable by the beneficiaries. The project as a whole will most likely have additional indicators and a methodology to facilitate the cross-sectional analyses of sub-projects. w Evaluation of results should involve the actors themselves (self-evaluation), but can be catalyzed and enriched by observations, lessons and information offered from the outside. w Lessons, successful models, and the conditions for success should be documented, together with the evidence on which they are based, to be accessible and understandable to the public or target audience. This may even be in the form of a "show and tell", i.e., the actors' demonstration of a successful venture to others, with the necessary explanations. Innovative projects "grope in the dark" to some extent, but should do this within a framework that allows a systematic analysis at both the local (sub-project) and the macro (project) level. Such systematic piloting has higher costs and requires more specialized staff than general, "mainstreaming" projects. These costs and human resources should be included in the project. The rule of thumb that "overhead" should not exceed 15-20% of project cost may not apply to pilot projects where learning is of the essence and is also the principal output. 2. Novel experiences may not always be so new ­ look first what has already have been learned elsewhere The project was seen as a first-time experiment with novel ways of using natural resources. It - 27 - may have been innovative in Brazil and its rain forests, but it was not the world's first small grants community project, nor was it the first time that communities tried some form of cooperative production, processing, or marketing. Lessons from failures and successes of community projects, especially cooperatives, and some "dos and don'ts" are usually available elsewhere, even if they may not be directly applicable to the context at hand. A lot can go wrong in community enterprises, from lack of working capital, wrong selection of species for production or extraction, over-extraction, inept cooperative management, neglect of market conditions, to the ignorance of more subtle environmental impacts. Learning is certainly most effective when people go through experiences themselves, but needless frustration, loss or damage can be avoided by applying previously acquired knowledge. 3. Integrate local knowledge and needs in project design The objectives and design of a sub-project should be based on the community's needs and demands, so as to guarantee ownership and identification. This lesson is not new, but it has been confirmed yet again. It applies not only to overall development objectives, technologies and planned investments, but also to simple things like the selection of species to be used in an agroforestry system. It is difficult to select the "best" technical options in a truly first-of-a kind project, where neither technical assistance nor the producer can rely on much solid prior experience. Even then decisions should be discussed and made by the beneficiaries-actors. Professional agronomic advice is necessary and should be given (see lesson above), but often decisions made by the producers themselves make a community project successful. In this project, sometimes choices were made by technical assistance agents or sponsoring NGOs that did not meet community needs. Sometimes traditional local knowledge or women's concerns were ignored, jeopardizing the results. Sub-projects were generally more successful, were maintained over a longer period of time and had fewer problems wherever the creativity and needs of producers were integrated into their own design (particularly of agroforestry systems). Technical assistance should not only impart available and proven knowledge; it should respect the clients and help people learn from their own experience. It should thus concentrate more on giving producers choices as to how to achieve objectives and refrain from deciding for them. This type of attitude and assistance requires skills in participatory methodologies, in addition to broad technical knowledge. Technical assistance naturally tends to teach what it knows instead of listening to what producers may know or want to know. This may be one of the reasons why some research institutions coped badly with producer demands, focusing instead on proving specific, preconceived hypotheses. 4. Technical assistance, of the right kind, should be on call at the beginning Technical assistance has its place, even in "uncharted waters", to help communities face their - 28 - social, organizational, commercial and technical challenges. Yet it is needed not only during sub-project implementation, but already when communities are preparing their proposals. Even if it is more difficult to provide in remote areas, it must be planned for and made accessible early on. In the face of the scarcity of qualified technical assistance and huge distances, some sub-projects found adequate solutions that may help others: w Try to establish regional or local partnerships for TA that are closer to the sub-projects and can provide lessons and ideas taken from similar projects, while building bridges among them, disseminating lessons. w Technical assistance can be offered by those who have already undergone the learning process. An efficient way to disseminate knowledge and experience is to bring producers together. w Professionals can receive on-the-job training within the sub-projects. An important result of many projects was the build-up of professional knowledge, turning producers into qualified references or resource persons on specific technical issues. This also helps with sustainability and replication. 5. Participatory planning Process and the need for an appraisal procedure Technical assistance most likely cannot be provided to all those preparing applications, but it should be made accessible to those with favorable chances of eventually submitting good projects. A two-stage selection/approval process may be more appropriate. In a first round, all proposals would be screened for compliance with basic criteria and a selection would be pre-approved. First applications would be easy to fill, without much training. Pre-approved proposals would then have the right to technical assistance paid for by the project, although they would still have to go through a final appraisal and approval, to ensure the quality and transparency of the process. Before and after approval, a good practice was to have technical assistance agents hold participatory planning sessions with the community before the first funds were disbursed, going once more through the project and its phases and offering further advice for implementation. This helped to set expectations right, formulate realistic goals and implementation plans, and correct technical errors, while ensuring that community needs were heard and heeded. Such meetings can also serve to collect and establish baseline data and discuss monitoring indicators. 6. Smaller-scale projects are more efficient and let more people participate Grants to sub-projects ranged initially between US$20,000 and $210,00011 The thresholds were later changed to values in Brazilian Reais for new contracts.. Both large and small-scale projects were able to improve their social organization and technical and financial capacity. On average, larger-scale projects implemented more activities, but the smaller ones had equal social-organizational impacts. Wherever poor organization and lack of managerial and technical capacity were a problem, small-scale projects seemed to be as or even more efficient in filling this - 29 - gap. With smaller sub-grants, a given grant budget can benefit more communities, and a universal ceiling of $210,000 is too much to start with for many of them. (In the Northeast Rural Poverty Reduction Projects, the maximum grant is much less). On the other hand, larger-scale projects seemed to have had more impact on strengthening environmental awareness and technical capacity on a broader scale, with the promotion of special events and seminars, and including neighboring communities and stakeholders. Small grant programs should indeed make smaller grants, at least in a first community experience. 11 The thresholds were later changed to values in Brazilian Reais for new contracts 7. Integrated natural resource management on the farm tends to show better results than isolated activities Sub-projects that focus on improved practices for the entire holding and included farmers directly had more long-term measurable benefits. The cost-benefit ratio (grant amount vs. impact on incomes, environmental protection and community organization) of sub-projects focusing only on partial farm processes and products (e.g. fish tanks or processing units) was apparently less favorable. Isolated production activities may show good technical results, but are possibly less effective for environmental sustainability. The emphasis should be on entire holdings and their farming systems, rather than on investments in only a part of each system, and the aggregate impact of many individual systems on the environment should be considered. 8. Stay close to the actors, listen to them and be open for adjustments Pilot projects are experiments that can potentially fail, but a lot can be learned even from failures. Still, they can have high social costs in terms of disappointment, frustration, loss of confidence and a tendency for the implementing communities to return to their "old ways", especially where expectations had been high. For sub-project approval, an appraisal should not only evaluate written proposals; it should visit the sites and communities to get insights that written applications will not convey. Thereafter, implementation must be closely monitored, first and foremost by the actors themselves, and communications between them and project management must be open and effective. Staff or consultants must be able to take time to read and react to reports, visit communities often and help with solutions to problems that may arise. This adds to the costs, but increases the chances for success and contributes to overall learning. The capacity and flexibility to adjust a sub-project if needed ("adaptive planning") are also important, both for the authorizing project staff as well as the sub-project participants. Adjustments are made more easily and quickly with good sub-project ownership. 9. Sustainability will emerge only over the longer term, and should be reached in small steps The expectation that communities and their projects would reach a stage of likely sustainability - 30 - within three years may not have been truly harbored by anybody, but was nevertheless codified in the project rules. Three years are unrealistic, especially with regard to economic and environmental aspects. Where little organization existed before, sub-projects can contribute to satisfying some basic needs and strengthening community organization and capacity. However, it will have difficulty reaching economic sustainability and significant environmental impacts in a short period of time. Sub-projects with a longer organizational history tend to approach sustainability faster. Consolidation and maturity should probably not be expected before about five to ten years, as suggested by communities that had implemented other projects before receiving a grant under this project. It appears preferable to initially fund small-scale projects with shorter implementation periods within a sequence of grants over a longer time period and with longer commitments from donors or the Bank ("APL"-style, with trigger conditions related to organization, financial management and marketing). A gradual, step-by-step approach, with realistic expectations for each phase, is likely to be more effective, particularly for the design of projects in isolated regions with less experienced and organized communities. This will reduce operational risks, guarantee long-term involvement and make adjustments easier than in single large grants. 12 This confirms one of the findings of a World Bank research study by Mansuri, G. and Rao, V. "Community-Based and - Driven Development: A Critical Review", WB Research Paper 3209, 2004. 10. Partnership with local institutions contributes to sustainability Sub-projects eventually benefited from establishing partnerships with local governments and institutions. Cooperation with state agencies was more difficult, mainly because the project was a federal one with no links to state programs and therefore practically bypassed the state governments. A lack of cooperation with local government and institutions during preparation hampered many of the sub-projects because they were perceived as "falling out of the sky" by municipal governments and the local "powers that be", which were often not informed about what was planned, much less consulted. It helps sustainability to be on good terms with local government13. "Paper partnerships", however, did not contribute to the sub-projects' success. To be effective, partnerships may have to be reestablished over and over again, especially in remote areas with weak institutions. The involvement of the right person in an institution may be more effective than a formal partnership. 13 The political situation during the early nineties in Brazil may not have been ripe for community-driven projects, as state and local political elites in the Amazon did not much appreciate them. Some sub-projects may simply not have happened, had they depended on the initial cooperation of municipal authorities. - 31 - 11. Partnerships with the Private Sector ­ both sides have to learn and to adjust their way of thinking Communities entered into partnerships with the private sector in several cases, and such partnerships have become a magic word for solving community development in the Amazon. But the interest of a major company in working with a community for raw material collection or production is not enough. In fact, in some cases, it may create more problems for the communities. Successful partnerships (e.g., with Daimler-Chrysler, Cognis/Natura and the Pão de Açúcar supermarket chain) produced valuable lessons on what to look for in such partnerships: w The partnership should be built on a long-term commitment, where both parties understand that sustainable economic results will take time. Expectations of immediate results will only lead to disappointment. Businesses should not seek partnerships merely for the sake of "marketing benefits" (although they will always be an important factor). w Negotiation of long-term partnerships takes time. Adequate intermediation by an "honest broker" is needed to adjust expectations on both sides and to enable everyone to see both the opportunities and limitations (project cycle, product quality, procurement problems, communication difficulties, regularity of supply, etc.). The "honest broker" should not be responsible for implementing a project. Communities have to learn to negotiate by themselves. A wise division of labor between the partners is needed. Each should concentrate on what he can do best. Communities should process a product up to the point after which its comparative advantage ends, and where the private sector can follow up more efficiently. For example, creating high value products in the forest for export to New York consumers does not work without a savvy intermediator. 12. Agroforestry Systems ­ not only an agronomic model The project funded a large number of agroforestry systems all over the two biomes. There is, of course, no single ideal agroforestry model, as everything depends on local circumstances - soils, climate, demand, distance to markets and local resources. It is difficult to evaluate if (and which) agroforestry systems were successful after a three-year implementation period, as such systems need a longer time to mature. Nevertheless, some lessons can already be drawn: w Local producers must be free to pursue their own ideas and decisions as to what kind of species mix should be adopted. It is essential to involve women, because they are the ones who think of the need to feed the family, whereas men tend to favor commercial crops. Technical assistance should not define the choice of species but should assist the producers in applying best agronomic practices. w Producers should be stimulated to think of subsistence needs with more immediate economic returns first, before aiming at sales at distant markets. Women are important for keeping the focus on this need. Many agroforestry systems failed because of false market - 32 - expectations, so that producers lost interest in maintaining them. Even the best systems with cash crops will have difficulties overcoming long distances, lack of infrastructure or insufficient local market demand. w Systems should be simple, and keep inputs and labor demands within the limits of a smallholder family, especially at the beginning. Complex systems with long-term maturation may fail because they neglect fast-growing crops mixed with slower-maturing ones and trees, but also because the producer must invest considerable time in maintaining the system, which may break family labor constraints. w Agroforestry systems should take advantage of deforested land under regeneration (secondary growth) rather than using recently cleared land. Enriching re-growing vegetation with perennial crops showed reasonably good results, with lower seedling mortality and labor costs. Local market demand and transport infrastructure are essential for commercial crops. This may be why agroforestry in the Atlantic Forest tended to have better commercial success than in the Amazon. Local markets are better developed, distances are less and producers are generally better equipped than those in the Amazon. For long-distance markets, only non-perishable products should be considered. 13. Extractivism, Forest Management and NTFP The second largest number of initiatives financed by the project focused on the collection and processing of "non-timber forest products" (NTFP) and on managing forests for those products. Although many of the following lessons are not new, they had to be learned the hard way by many communities under this project: w Withveryfewexceptions,thecollectionofNTFPrepresentsaperiodic,oftenunstableand low income for the farmer, and is not his sole income. It also does not satisfy subsistence needs, as NTFPs have mostly commercial value. They must be complemented by subsistence crops. Extractive activities alone are not likely to keep the producers in the forest. The introduction of new NFTP products, which producers did not know how to collect, stock and sell, was rarely successful. w Products vary in size and quality, and are often subject to fluctuating world market prices, changes in consumer preferences and quality restrictions. Many markets are dominated by buyers who care little for the producers' fate. Concentrating on only one crop is risky. w Manyproducercooperativesorassociationshavetheambitionofrunningprocessingunits to add value to raw products. Few of these achieved their objectives. Planning rarely took into account limited management capacity, infrastructure and marketing knowledge, political interference, working capital needs and supply shortages. Processing projects should be kept as simple as possible to ensure success. Despite the ambition of community leaders to turn into instant entrepreneurs, a step-by-step approach should be followed, gathering experiences based on a realistic plan and broad community involvement. w Local markets are important. Except for a few private-sector partnerships where markets can be conquered outside the region, products should be developed for markets with local demand for modest quantities and where producers can gather experience in selling this kind - 33 - of product. The quality of raw material or semi-processed products should be improved to fetch higher prices locally (dried fruits, oils, improved manioc flour, etc.). The "vertical integration" of production processes has had limited success. A horizontal approach (diversification of raw materials) produces better economic results. 14. Community-driven development must be complemented by other policies supporting regional sustainable development Almost ten years of experience working with communities within the RFPP have led to a considerable change in the sustainable management of natural resources by communities in the region. Without a more comprehensive regional and local development strategy, however, which would also address the need for basic education, health, credit, communication, transport and energy, community-driven projects alone cannot be expected to turn the tide and become truly sustainable. Isolated communities are particularly vulnerable to adverse impacts. In many cases, this project was the first and only investment supported by the outside. While many communities made good use of the sub-grants, one should not expect them to be "living happily ever after". A longer-term and more comprehensive support strategy is needed to secure these first results. 9. Partner Comments (a) Borrower/implementing agency: not received from the implementation agency (b) Cofinanciers: Comments from GTZ, German Technical Cooperation The PDA Demonstrative Project was one of the first projects within the Pilot Program with technical assistance provided by GTZ, working together with the technical secretariat of PDA. GTZ shares the Bank's opinion about the Performance Ratings of PDA. PDA was very successful in establishing working relations in a multilateral program with numerous and different partners. The conception and implementation of a professional and participatory mechanism of project selection, within a political context where civil society participation in government decisions was rather negligent stands out as a model for other PPG7 projects. PDA pioneered the creation of an innovative, transparent and low-cost funding mechanism. This was to implement sustainable development projects together with local communities in remote regions of the Amazon and Atlantic rainforests, respecting the particular conditions of each partner. The internal organization and qualification of the project personnel to carry out the great variety of daily tasks was one of the principle tasks of GTZ cooperation. - 34 - The monitoring system remained a challenge throughout PDA´s implementation. The enormous diversity of subprojects in terms of size, area of work, target groups and thematic variability, difficult access and enormous geographic distances made defining needs extremely difficult. Cost-benefit relationships were difficult to establish. Over time these became clearer through the management of accumulated knowledge and shared experience within and between the subprojects. PDA played a pioneering role within Pilot Project introducing, testing and adapting innovative monitoring approaches such as participatory monitoring. An important contribution of GTZ was the methodological support of studies analyzing the experiences of 20 projects in the Amazon and 12 projects in the Atlantic rainforest. This highlighted important results and lessons learnt from the demonstrative projects Estudos da Amazônia: Avaliação de vinte projetos, PDA. Brasília: Ministério de Meio Ambiente, 2004. Estudos da Mata Atlântica: Avaliação de doze projetos PDA. Brasília: Ministério de Meio Ambiente, 2004. Implementation of a communication strategy and the dissemination of experiences and lessons learnt are important within the present project phase (2004-2006). The overall challenge is to create mechanisms allowing the internalization of this type of Fund within the government. GTZ supported the creation of an innovative mechanism with the OSCIP Coalizão Florestas to act as a catalyst for national and international, public and private funds within an institutional model which allows partnerships between the State and Civil Society organizations allowing administration of public funding. The agility and flexibility of this mechanism guarantees efficient and effective resource management without loosing necessary control. The most important results of PDA have been the formation of social capital and the strengthening of civil society in the Brazilian Amazon and the acknowledgement and support for the preservation of the remaining areas of Atlantic forest. This has contributed to the formation of an innovative public policy ­ PROAMBIENTE and has also impacted on the Brazilian Ministry of the Environment's Atlantic Forest Program and PDA Atlantic Forest. (c) Other partners (NGOs/private sector): Comments from the KfW, Kreditanstalt fuer Wiederaufbau We thank you very much for the preliminary version of the PDA Report. The Report presents a good and broad analyses of the project. We consider it excellent the fact that the report does not only show the positive results of the project but also provides a critical analyses of the existing problematic aspects. In this sense we do not have any additional comments as we fully agree with the report. We appreciate the idea of implementing an inter-institutional group to discuss a specific dissemination plan. An example of this could be the consolidation projects in phase III of PDA. 10. Additional Information - 35 - - 36 - Annex 1. Key Performance Indicators/Log Frame Matrix Outcome / Impact Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate N/A N/A Output Indicators: 1 Indicator/Matrix Projected in last PSR Actual/Latest Estimate % of proposals received which are approved 18,5% 18,5% No. and amount of funds by size category of large projects: 89% large projects: 89% executed sub-projects medium size projects: 8% medium size projects: 8% small size projects: 3% small size projects: 3% No./amount of funds by subject area Agroforestry Systems: 47% Agroforestry Systems: 47% Forest Manaement: 30% Forest Manaement: 30% No./amount of funds by type of organization ONGs - U$ 8.221.628 ONGs - U$ 8.221.628 Associações - U$ 6.546.247 Associações - U$ 6.546.247 Cooperativas - U$ 1.871.075 Cooperativas - U$ 1.871.075 Sindicatos - U$ 1.852.072 Sindicatos - U$ 1.852.072 ONGs Indigenistas 1.488.873 ONGs Indigenistas 1.488.873 Federações - U$ 916.282 Federações - U$ 916.282 Organizações Indígenas - U$ 807.059 Organizações Indígenas - U$ 807.059 Entidades Quilombolas - U$ 218.670 Entidades Quilombolas - U$ 218.670 Prefeitura - U$ 127.738 Prefeitura - U$ 127.738 No./amount of funds by region Amazon 147 projects, U$ 16.042.035,70 Amazon 147 projects, U$ 16.042.035,70 (Amazon/Atlantic Forest) Atlantic 47 projects, U$ 6.007.608,62 Atlantic 47 projects, U$ 6.007.608,62 % of acceptable proposals from 40 projects received training during proposal organizations which received training during not monitored preparation by GAPEP proposal preparation % of compliance with administrative and 90% 98% accouting norms % of organizations which receive training on not monitored not monitored organizational structure that experienced improvement % of proposals processed within established 15% 12% time limits % of sub-projects which receive funds within nm not monitored established time limits % of reformulated proposals recommended nm 47,5% by GAP after reformulation % of problematic sub-projects assited by TS nm 6,2% or GAPEP that reached their stated objectives % of successful sub-projects documented by nm 75% successful - 37 - subject area and region 1End of project - 38 - Annex 2. Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal Actual/Latest Percentage of Estimate Estimate Appraisal Component US$ million US$ million Sub-grants to communities 14.59 21.04 Support to GTA 0.50 1.00 Support to RMA 0.08 0.34 Overhead tasks 3.56 4.02 Fees 0.36 0.44 Counterpart contributions 3.00 3.00 Total Baseline Cost 22.09 29.84 Total Project Costs 22.09 29.84 Total Financing Required 22.09 29.84 Note1: Community counterpart contributions were required but not quantified and verified. Note 2: Table include contributions from the European Community, Germany (KfW) France (FFEM) and the RainForest Trust Fund Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) 1 Procurement Method Expenditure Category ICB NCB 2 N.B.F. Total Cost Other 1. Works 0.00 0.00 1.54 0.00 1.54 (0.00) (0.00) (0.00) (0.00) (0.00) 2. Goods 0.00 0.00 0.12 0.00 0.12 (0.00) (0.00) (0.00) (0.00) (0.00) 3. Services 0.00 0.00 1.18 0.00 1.18 (0.00) (0.00) (0.00) (0.00) (0.00) 4. Subprojects 0.00 0.00 17.58 0.00 17.58 (0.00) (0.00) (0.00) (0.00) (0.00) 5. Fees 0.00 0.00 0.46 0.00 0.46 (0.00) (0.00) (0.00) (0.00) (0.00) 6. Consultants 0.00 0.00 1.23 0.00 1.23 (0.00) (0.00) (0.00) (0.00) (0.00) Total 0.00 0.00 22.11 0.00 22.11 (0.00) (0.00) (0.00) (0.00) (0.00) Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equivalent) 1 Procurement Method Expenditure Category ICB NCB 2 N.B.F. Total Cost Other 1. Works 0.00 0.00 0.11 0.00 0.11 (0.00) (0.00) (0.00) (0.00) (0.00) 2. Goods 0.00 0.00 0.98 0.44 1.42 (0.00) (0.00) (0.00) (0.00) (0.00) - 39 - 3. Services 0.00 0.00 1.10 0.00 1.10 (0.00) (0.00) (0.00) (0.00) (0.00) 4. Subprojects 0.00 0.00 24.04 0.00 24.04 (0.00) (0.00) (0.00) (0.00) (0.00) 5. Fees 0.00 0.00 0.44 0.00 0.44 (0.00) (0.00) (0.00) (0.00) (0.00) 6. Consultants 0.00 0.00 2.73 0.00 2.73 (0.00) (0.00) (0.00) (0.00) (0.00) Total 0.00 0.00 29.40 0.44 29.84 (0.00) (0.00) (0.00) (0.00) (0.00) 1/Figures in parenthesis are the amounts to be financed by the Bank Loan. All costs include contingencies. 2/Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units. Project Financing by Component (in US$ million equivalent) Percentage of Appraisal Component Appraisal Estimate Actual/Latest Estimate Bank Govt. CoF. Bank Govt. CoF. Bank Govt. CoF. Sub-grants 0.00 3.44 3.40 0.0 98.8 Support to GTA 0.00 0.50 0.50 0.50 0.0 100.0 Support to RMA 0.00 0.09 0.35 0.09 0.0 100.0 Overhead tasks 2.65 3.12 117.7 fees 0.05 0.06 0.00 0.06 0.0 100.0 unallocated 0.30 0.35 0.0 0.0 Note 1: Table only includes the contributions from the European Community (CoF.) and the Rain Forest Trust Fund (Bank) , as projects financed by KfW and FFEM are still ongoing - 40 - Annex 3. Economic Costs and Benefits - 41 - Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle No. of Persons and Specialty Performance Rating (e.g. 2 Economists, 1 FMS, etc.) Implementation Development Month/Year Count Specialty Progress Objective Identification/Preparation 1994 10 Task Manager (1) S HS Community Development Specialists (2) Economist (1) Environmental Lawyer (1) Operations Specialists (2) Consultants (3) Appraisal/Negotiation 1996 6 Mission Leader (1) HS S Community Development Specialist (1) Environmental Lawyer (1) Financial Analyst (1) Operations Specialist (1) Consultants (1) Supervision 12/15/1996 2 Team Leader (1); S S 12/20/1996 Operations Specialist (1) S S 08/25/1997 8 Task Team/Mission Leader (1) 09/05/1997 Head of PPG7 (1); Anthropologist (1); Civil Society Liaison (1); Operations Analyst (1); Consultants (3) S S 03/30/1998 4 Task Team/Mission Leader (1) 04/02/1998 Anthropologist (1); Civil Society Liaison (1); Regional Planner (1); S S 09/02/1998 2 Task Team/Mission Leader (1) 09/16/1998 Anthropologist / Task Manager (1) S S 09/11/2000 4 Anthropologist/Task Mngr (1) 09/15/2000 Procurement Analyst (1) Biodiversity Specialist (1) Private Sector /Forest Spec (1) S S 08/27/2001 2 Private Sector /Forest Spec (1) 08/30/2001 Operations Analyst (1); - 42 - S S 09/23/2002 2 Private Sector /Forest Spec (1) 10/03/2002 Operations Analyst (1); S S 09/01/2003 2 Private Sector /Forest Spec (1) 09/02/2003 Operations Analyst (1); ICR 10/2004 1 Task Manager (1) HS HS (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation 40 66 Appraisal/Negotiation 73 210.4 Supervision 156 311.2 ICR 14 16 Total 273 603.6 - 43 - Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Macro policies H SU M N NA Sector Policies H SU M N NA Physical H SU M N NA Financial H SU M N NA Institutional Development H SU M N NA Environmental H SU M N NA Social Poverty Reduction H SU M N NA Gender H SU M N NA Other (Please specify) H SU M N NA Community Organization Private sector development H SU M N NA Public sector management H SU M N NA Other (Please specify) H SU M N NA Civil Society development - 44 - Annex 6. Ratings of Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating Lending HS S U HU Supervision HS S U HU Overall HS S U HU 6.2 Borrower performance Rating Preparation HS S U HU Government implementation performance HS S U HU Implementation agency performance HS S U HU Overall HS S U HU - 45 - Annex 7. List of Supporting Documents Azevedo, R. A. B. (Março, 1999) ­ Programa Piloto de Proteção às Florestas Tropicais do Brasil ­Projetos Demonstrativos ­ PDA ­ Estrutura conceitual para a avaliação e monitoramento dos projetos do PD/A com base em indicadores, Brasília Banco Mundial (Agosto, 2001). Estudos sobre os fundos sociais e ambientais apoiados pelo Banco Mundial no Brasil (Relatório Final do "Estudo dos Fundos de Pequenas Doações em Projeto Apoiados pelo Banco Mundial no Brasil). Banco Mundial ­ Programa Piloto de Proteção às Florestas Tropicais do Brasil ­ Projetos Demonstrativos ­ PD/A ­ Missão de Supervisão 11 a 15 de setembro, 2000 (Ajuda Memória) Banco Mundial (Outubro, 2001) ­ Programa Piloto de Proteção às Florestas Tropicais do Brasil ­ Subprograma Projetos Demonstrativos ­ PDA ­ Avaliação de aspectos da sustentabilidade ambiental dos sistemas agro-florestais: Um estudo de caso Banco Mundial (1998) ­ Programa Piloto de Proteção às Florestas Tropicais do Brasil ­ Subprograma Projetos Demonstrativos ­ PDA ­ Manual de Operações Execução Técnica e Financeira de Subprojetos Banco Mundial ­ Programa Piloto de Proteção às Florestas Tropicais do Brasil ­ Projetos Demonstrativos ­ PD/A ­ Missão de Supervisão do Banco Mundial e Avaliação do Aumento da KfW - 27 a 30 de agosto, 2000 (Ajuda Memória) Banco Mundial ­ Programa Piloto de Proteção às Florestas Tropicais do Brasil ­ Subprograma Projetos Demonstrativos ­ PDA ­ Missão de Supervisão e Pré-Appraisal do componente "Mata Atlântica ­ 23 de setembro a 3 de outubro, 2002 (Ajuda Memória) Ghazala Mansuri and Vijayendra Rao, Community Based (and Driven) Development: A Critical Review; World Bank - Research Working Paper 3209, 2004 http://papers.ssrn.com/sol3/cf_dev/AbsByAuth.cfm?per_id=365960 Guimarães, Alice (2002) - Projetos Demonstrativos ­ PDA ­ Relatório de Acompanhamento ­ AMA Leroy, J-P. & Toledo, G. T. (Maio, 2001) ­ Programa Piloto de Proteção às Florestas Tropicais do Brasil ­ Experiências PDA Nº 1 ­ Projeto Frutos do Cerrado ­ Novo Rumo para o desenvolvimento regional. Lisansky, Judith, and Loretta Sprissler. 2002. "Lessons from the Rain Forest: Participation in the First Decade of the Pilot Program to Conserve the Brazilian Rain Forest." In Moji Anderson, ed., Thinking Out Loud III: Innovative Case Studies on Participatory Instruments. Latin America and the Caribbean Region, World Bank, Washington, D.C. - 46 - Ministério do Meio Ambiente/ SCA / PPG7 /PDA (1998) ­ Oficinas de intercâmbio de experiências ­ lições dos projetos apoiados pelo PDA-PPG7 ­ Resultados. Ministério do Meio Ambiente/SCA/PPG7/PDA. (Junho, 2001). Proposta de Consolidação dos Projetos Demonstrativos ­ PDA, Brasília. Ministério do Meio Ambiente / SCA / PPG7 (2000) ­ Subprograma de Projetos Demonstrativos - Relatório de Atividades 2000 Ministério do Meio Ambiente / SCA / PPG7 / AMA / WB (2002) - Lessons from the Rain Forest ­ Experience of the Pilot Program to conserve the Amazon and Atlantic Forests of Brazil Ministério do Meio Ambiente/ SCA / PDA (2001) ­ Projetos Demonstrativos no Estado do Rio Grande do Sul: Experimentos Locais e Possibilidades de Consolidação ­ Uma contribuição para a Missão de Avaliação do PDA de 27 a 31 de agosto de 2001 (Ricardo Verdum) Ministério do Meio Ambiente / SCA / PPG7 (Maio, 2004) Experiências PDA Nº 4 ­ Estudos da Mata Atlántica Avaliação de doze projetos PDA. Ministério do Meio Ambiente / SCA / PPG7 / PDA ­ Relatório de Avaliação ­ Subprograma Projetos Demonstrativos, Abril de 2004. Ministério do Meio Ambiente / SCA / PPG7 - Subprograma de Projetos Demonstrativos Projeto BRA/93/044 ­ Relatório Anual de Progresso, Janeiro a Dezembro de 2003. Ministério do Meio Ambiente / SCA / PPG7 - Subprograma de Projetos Demonstrativos Projeto BRA/93/044 ­ Relatório Anual de Progresso, Janeiro a Dezembro de 2002. Ministério do Meio Ambiente / SCA / PPG7 (Maio, 2004) Experiências PDA Nº 5 ­ Estudos da Amazônia tlántica Avaliação de doze projetos PDA: Avaliação de Vinte Projetos PDA. PDA em Campo (2003) ­ Projetos monitorado 2002/2003 PPG7 & GTZ (2003) ­ Subprograma de Projetos Demonstrativos (SPD) - Cooperação Técnica Brasil-Alemanha ­ Avaliação do Desenvolvimento do Programa (PFK) ­ Relatório Final - Caio Marcio Silveira, Carlos Amaral, Han-Jürgen Fiege & Maria Emilia Lisboa Pacheco. PPG7 / Subprograma PDA ­ Cinco Anos de PDA: Lições aprendidas no processo de avaliação, Outubro 2001 PPG7 / Subprograma PDA ­ Projetos Econômicos no PDA: Levantamento e situação quanto a obtenção de créditos comerciais, Outubro de 2001. PPG7 / Subprograma PDA ­ Oficina de auto-monitoramento do Sub-programa Projetos Demonstrativos - Paulo Sérgio Ferreira Neto, Novembro de 2003 - 47 - SCHERR, S. J. & WHITE, A. & KAIMOWITZ, D. (2001) ­ Making Forest Markets Work for Low-Income Producers ­ Policy Brief ­ Forest Trends (Washington, DC) & Center for International Forestry Research (Bongor, Indonésia) ­ October 23, 2001 SCHÄFFER, W. B. ­ O mecanismo PDA ­ Lição de Política Pública ­ Como fazer parcerias entre Poder Público e Organizações da Sociedade Civil de forma ágil, trasnparente e responsable. Vivan, J. L. & Monte, N. L. & Gavazzi, R. A. (Agosto, 2002) ­ Programa Piloto de Proteção às Florestas Tropicais do Brasil ­ Experiências PDA Nº 3 ­ Comissão Pró-Índio do Acre ­ Implantação de tecnologias de manejo agroflorestal em terras indígenas do Acre. Wagner B. de Almeida, A., "The Growing Pains of an Unprecedented Civil Society ­ Government Partnership in the Brazilian Amazon: The Case of the Amazon Working Group (GTA)", in Thinking Out Loud, Innovative Case Studies on Participatory Instruments, LCR Civil Society Team (PREM/ESSD), World Bank, 2001 Weiss, J. (Março, 2000). Estudos sobre os fundos sociais e ambientais financiados ou administrados pelo Banco Mundial no Brasil (Relatório Geral I "Lições aprendidas da experiência comparada entre fundos no Brasil e no exterior). World Bank (2003). Implementation Completion Report (CPL ­ 34440) - Rondônia Natural Resources Management Project - (Loan Number 3444-BR). World Bank (2001). Implementation Completion Report (TF ­ 21949; TF - 21948) ­ Science and Technology Subprogram: Science Centers & Directed Research ­ Phase I - And Emergency Assistance - (Report Number 22533). World Bank & Banco do Brasil S. A. (1994) ­ Rain Forest Trust Fund Grant and CEC Grant Agreement (Demonstration Project). World Bank (1994) Pilot Program to Conserve the Brazilian Rain Forest ­ Memorandum and Recommendation of the Director of the Latin American and Caribbean Department I to the Regional Vice President on a Proposed Grant from the Rain Forest Trust Fund to the Bank of Brazil for Demonstration Projects ­ Report nº 12602-BR - 48 - - 49 -