Document of The World Bank Report No.: 89848 PROJECT PERFORMANCE ASSESSMENT REPORT BRAZIL FIRST PROGRAMMATIC DEVELOPMENT POLICY LOAN FOR SUSTAINABLE ENVIRONMENTAL MANAGEMENT (IBRD-76600) February 19, 2015 IEG Public Sector Evaluation Independent Evaluation Group i Currency Equivalents (annual averages) Currency Unit = real (BR$) March 5, 2009 US$1.00 BR$2.40 June 30, 2010 US$1.00 BR$1.87 December 15, 2010 US$1.00 BR$1.69 Abbreviations and Acronyms ABC Low-Carbon Agriculture ANA Water National Agency ARPA Amazon Protected Areas Project ASIBAMA Association of IBAMA staff BNDES National Bank for Economic and Social Development (of Brazil) BOD Biochemical Oxygen Demand CAS Country Assistance Strategy CDM Clean Development Mechanism CER Certified Emission Reduction CO2 Carbon dioxide COFA Amazon Fund Managing Committee CONAMA National Environmental Council CPE Country Program Evaluation CPS Country Partnership Strategy CSO Civil Society Organization DFID Department for International Development DPL Development Policy Loan DRM Disaster Risk Management EIA Environmental Impact Assessments Env PRL First Programmatic Reform Loan for Environmental Sustainability Env TAL Environmental Technical Assistance Loan EPs Equator Principles EPFI Equator Principles Financial Institutions ES Environmental and Social ESIA Environmental and Social Impact Assessment ESMP Environmental and Social Management Plan ESMS Environmental and Social Management System ESP Environmental and Social Policy ETS Emissions Trading Scheme EU European Union FIL Financial Intermediary Loan FSC Forest Stewardship Council FY Fiscal Year GEF Global Environmental Facility GHG Greenhouse Gas GOB Government of Brazil ii GP Green Protocol (Protocolo Verde) Ha Hectares IBAMA Brazilian Institute for the Environment and Renewable Natural Resources IBRD International Bank for Reconstruction and Development ICMBio National Institute for Biodiversity Conservation ICR Implementation Completion and Results Report ICTSD International Centre for Trade and Sustainable Development IDB Inter-American Development Bank IEA Integrated Environmental Assessment IEG Independent Evaluation Group IEGPS IEG Public Sector Evaluations IFC International Finance Corporation ILO International Labor Organization IMAZON Amazon Institute of People and the Environment INPE Brazilian Space Agency IPAM Amazonian Environment Research Institute ISA Social and Environmental Institute ISR Implementation Status Report Km2 Square Kilometers LDC Least Developed Countries M&E Monitoring and Evaluation MMA Ministry of Environment MOF Ministry of Finance MW Megawatts NCCAP National Climate Change Action Plan NEP National Environment Project NICFI Norway’s International Climate and Forest Initiative NGOs Non-Governmental Organizations NORAD Norwegian Agency for Development Cooperation NPCC National Policy for Climate Change NTS National Treasury Secretariat OPCS Operations Policy and Country Services PA Protected Area PAC Program for Accelerated Growth (Portuguese acronym) PAS Sustainable Amazon Plan PD Project Document PDO Project Development Objectives PNMC National Policy for Climate Change PNQA National Water Quality Evaluation Program PNRH National Water Resources Plan PPA Multi-Year Plan PPAR Project Performance Assessment Report PPCDAm Action Plan for the Prevention and Control of Deforestation in the Legal Amazon PPG7 Pilot Program to Protect the Brazilian Rain Forests PPO Public Prosecutor Offices PRODES Program for Depolluting River Basins iii REFLORESTA BNDES Forest Management Program RIMA Environmental Impact Assessment SEA Strategic Environmental Assessment SEM DPL Sustainable Environmental Management Development Policy Loan SFB Brazilian Forest Services SNUC National System of Conservation Units TJ Terajoules UNCBD Convention on Biological Diversity UNFCCC United Nations Framework Convention on Climate Change WWF World Wide Fund for Nature Fiscal Year Government: January 1 – December 31 Director-General, Independent Evaluation : Ms. Caroline Heider Director, IEG Public Sector Evaluation : Mr. Emmanuel Jimenez Manager, IEG Public Sector Evaluation : Ms. Marie Gaarder Task Manager : Mr. William R. Sutton v Contents Principal Ratings ............................................................................................................... vii Key Staff Responsible....................................................................................................... vii Preface................................................................................................................................ ix Overview ............................................................................................................................. x Management Response .................................................................................................... xix Chairperson’s Summary: Committee on Development Effectiveness.... Error! Bookmark not defined. 1. Background and Context................................................................................................. 1 Environmental Context ................................................................................................... 1 Development Context ..................................................................................................... 2 Operational Context ........................................................................................................ 3 2. Objectives, Design, and their Relevance ........................................................................ 7 Relevance of Objectives ................................................................................................. 7 Design ............................................................................................................................. 8 Relevance of Design ..................................................................................................... 12 3. Implementation ............................................................................................................. 16 4. Achievement of the Objectives ..................................................................................... 18 Objective A: Improve the overall Brazilian environmental management system. ....... 18 Objective B: Manage natural resources sustainably, reducing the degradation of agricultural lands, forests (in particular the Amazon), and water resources, and promoting renewable energy......................................................................................... 39 5. Ratings .......................................................................................................................... 44 Outcome ........................................................................................................................ 44 Risk to Development Outcome ..................................................................................... 45 Bank Performance ......................................................................................................... 46 Borrower Performance .................................................................................................. 54 Monitoring and Evaluation ........................................................................................... 57 6. Lessons .......................................................................................................................... 58 References ......................................................................................................................... 61 Annex A: Basic Data Sheet............................................................................................... 71 Annex B: SEM DPL Development Policy Matrix ............................................................ 73 Annex C: Summary of outcome indicators and achievements as reported in the World Bank’s ICR*...................................................................................................................... 79 vi Annex D: Poverty and Social Impacts, and Environmental, Forests, and other Natural Resources Aspects of DPLs .............................................................................................. 83 Annex E: Comparison of the Equator Principles with Brazil’s Green Protocol and BNDES’s environmental and social policy ...................................................................... 85 Annex F: Organization of Brazil’s Federal Ministry of Environment and related ........... 93 Agencies ............................................................................................................................ 93 Annex G: Project Information Available from BNDES’s Website .................................. 94 Annex H: History and Timeline of the Operation ............................................................ 95 Annex I: List of Persons Interviewed ............................................................................. 104 Annex J: Borrower Comments........................................................................................ 108 Boxes Box 1: Brazil’s National Policy on Climate Change ........................................................ 23 Box 2: BNDES financing for the Belo Monte Hydroelectric Dam .................................. 34 Figures Figure 1: Planned annual emission reductions in the Brazil CDM portfolio.................... 25 Figure 2: Area of Amazon deforestation over time .......................................................... 40 vii Principal Ratings ICR* ICR Review* PPAR Outcome Satisfactory Unsatisfactory Unsatisfactory Risk to Moderate Moderate Moderate Development Outcome Bank Performance Satisfactory Unsatisfactory Unsatisfactory Borrower Moderately Satisfactory Moderately Moderately Performance Unsatisfactory Unsatisfactory * The Implementation Completion Report (ICR) is a self-evaluation by the responsible Bank department. The ICR Review is an intermediate Independent Evaluation Group (IEG) product that seeks to independently verify the findings of the ICR. In this case, the ICR Review is being completed by IEG in parallel with the Project Performance Assessment Report. Key Staff Responsible Project Task Manager/Leader Sector Manager Country Director Appraisal Mark Lundell/Garo Batmanian Laura Tlaiye Makhtar Diop (Co-Task Manager) Completion Mark Lundell/Garo Batmanian Karin Kemper Makhtar Diop (Co-Task Manager) viii IEG Mission: Improving World Bank Group development results through excellence in evaluation. About this Report The Independent Evaluation Group (IEG) assesses the programs and activities of the World Bank for two purposes: first, to ensure the integrity of the Bank’s self-evaluation process and to verify that the Bank’s work is producing the expected results, and second, to help develop improved directions, policies, and procedures through the dissemination of lessons drawn from experience. As part of this work, IEG annually assesses 20-25 percent of the Bank’s lending operations through field work. In selecting operations for assessment, preference is given to those that are innovative, large, or complex; those that are relevant to upcoming studies or country evaluations; those for which Executive Directors or Bank management have requested assessments; and those that are likely to generate important lessons. To prepare a Project Performance Assessment Report (PPAR), IEG staff examine project files and other documents, visit the borrowing country to discuss the operation with the government, and other in-country stakeholders, and interview Bank staff and other donor agency staff both at headquarters and in local offices as appropriate. Each PPAR is subject to internal IEG peer review, panel review, and management approval. Once cleared internally, the PPAR is commented on by the responsible Bank department. The PPAR is also sent to the borrower for review. IEG incorporates both Bank and borrower comments as appropriate and the borrowers' comments are attached to the document that is sent to the Bank's Board of Executive Directors. After an assessment report has been sent to the Board, it is disclosed to the public. About the IEG Rating System for Public Sector Evaluations IEG’s use of multiple evaluation methods offers both rigor and a necessary level of flexibility to adapt to the lending instrument, project design, or sectoral approach. IEG evaluators all apply the same basic method to arrive at their project ratings. Following is the definition and rating scale used for each evaluation criterion (additional information is available on the IEG website: ieg.worldbankgroup.org). Outcome: The extent to which the operation’s major relevant objectives were achieved, or are expected to be achieved, efficiently. The rating has three dimensions: relevance, efficacy/achievement of objectives, and efficiency. Relevance includes relevance of objectives and relevance of design. Relevance of objectives is the extent to which the project’s objectives are consistent with the country’s current development priorities and with current Bank country and sectoral assistance strategies and corporate goals (expressed in Poverty Reduction Strategy Papers, Country Assistance Strategies, Sector Strategy Papers, and Operational Policies). Relevance of design is the extent to which the project’s design is consistent with the stated objectives. Efficacy is the extent to which the project’s objectives were achieved, or are expected to be achieved, taking into account their relative importance. Efficiency is the extent to which the project achieved, or is expected to achieve, a return higher than the opportunity cost of capital and benefits at least cost compared to alternatives. The efficiency dimension generally is not applied to adjustment operations. Possible ratings for Outcome: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Risk to Development Outcome: The risk, at the time of evaluation, that development outcomes (or expected outcomes) will not be maintained (or realized). Possible ratings for Risk to Development Outcome: High, Significant, Moderate, Negligible to Low, Not Evaluable. Bank Performance: The extent to which services provided by the Bank ensured quality at entry of the operation and supported effective implementation through appropriate supervision (including ensuring adequate transition arrangements for regular operation of supported activities after loan/credit closing, toward the achievement of development outcomes. The rating has two dimensions: quality at entry and quality of supervision. Possible ratings for Bank Performance: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Borrower Performance: The extent to which the borrower (including the government and implementing agency or agencies) ensured quality of preparation and implementation, and complied with covenants and agreements, toward the achievement of development outcomes. The rating has two dimensions: government performance and implementing agency(ies) performance. Possible ratings for Borrower Performance: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. ix Preface This Project Performance Assessment Report (PPAR) is a product of the Independent Evaluation Group (IEG). Following standard IEG procedures, an earlier draft of this evaluation was sent to World Bank Management, who provided extensive comments; and the revised report was then sent to the Government of Brazil, who submitted extensive comments. The PPAR was discussed at a meeting of the World Bank Board’s Committee on Development Effectiveness (CODE) on January 21, 2015. The Management Comments (found on page xix), Chairperson’s Summary of the CODE discussion (found on page xxxii), and Borrower Comments (found in Annex J on page 108) are an important part of this evaluation report. The Management and Borrower Comments flag there were significant differences in views with respect to the background, performance, and results of this operation, and the lessons learned from this PPAR. The Chairperson’s Summary requested that these divergences in views be flagged. The report was prepared by William Sutton (Task Team Leader), under the guidance of Emmanuel Jimenez (Director, Public Sector Evaluation) and Marie Gaarder (Manager, Public Sector Evaluation), and overall direction of Caroline Heider (Director General, Evaluation). Napoleão Dequech Neto and Estela Costa Neves provided inputs. The report was peer reviewed by Kenneth Chomitz and Jaime Jaramillo-Vallejo and panel reviewed by John Heath. Richard Scobey, James Adams, Jiro Tominaga, and Marcelo Selowsky also provided helpful comments. Marie Charles provided administrative support. This Project Performance Assessment Report (PPAR) presents findings based on a review of the Program Document, the Implementation Completion and Results Report, and documents from the World Bank project files, as well as from other World Bank sources and a literature relevant to the subject. In addition, IEG carried out a mission to Brazil in February-March 2013 that included interviews with government officials at the federal, state and municipal levels, World Bank staff, development partners, researchers, representatives of international NGOs and local civil society organizations, and community members. IEG extends its thanks to the Brazilian government officials, other external stakeholders, and current and former World Bank staff for their cooperation with the evaluation. IEG also thanks the World Bank office in Brasilia for logistical support and help in organizing the mission itinerary. x Overview This is the Project Performance Assessment Report (PPAR) for the First Programmatic Development Policy Loan for Sustainable Environmental Management (SEM DPL 1) (IBRD-76600; P095205). The assessment aims, first, to serve an accountability purpose by verifying whether the operation achieved its intended outcomes. Second, the report draws lessons that are intended to inform future environmental Development Policy Operations, either in Brazil or elsewhere. Third, the assessment will contribute to a forthcoming IEG learning product on environmental policy lending across World Bank client countries. The SEM DPL was designed as a programmatic series of two loans to the Federal Government of Brazil for a total of approximately US$2 billion. The first loan—the subject of this evaluation—was for a total of US$1.3 billion divided into two tranches of US$800 million and US$500 million on International Bank for Reconstruction and Development (IBRD) terms. The loan was approved by the World Bank Board on March 5, 2009. The SEM DPL 1 was signed and became effective on June 21, 2010. The first tranche of US$800 million was released on June 30, 2010, and the second tranche of US$500 million was disbursed on December 15, 2010. Thus the total loan amount for SEM DPL 1 of US$1.3 billion was disbursed. The operation was closed on December 31, 2010. The planned second loan in the series, SEM DPL 2, never materialized and was eventually canceled. To properly evaluate the SEM DPL, it is important to understand the history and evolution of the operation and the context in which it was prepared. The operation was initially proposed as a “BNDES PAC-Env DPL”—a development policy loan to the Brazilian National Bank for Economic and Social Development (BNDES) to support the government’s scaling-up of infrastructure investments while also improving BNDES’s environmental and social policies, which was considered a significant challenge. The Brazilian government and the World Bank next concluded that BNDES was not eligible for a DPL. Instead preparation began on a US$1 billion Financial Intermediary Loan—a type of investment loan—to BNDES called the “BNDES Environmental and Social Sustainability Project,” with 99 percent of the funds going to finance BNDES investment operations and 1 percent for technical assistance to strengthen BNDES’s environmental and social safeguards. Internal World Bank reviewers expressed serious concerns about BNDES’s ability to comply with requirements on environmental and social safeguards, interest subsidies, financial management, and procurement. At the time of the global financial crisis in 2008, an agreement was reached to change the design of the operation again by transforming it into an environmental DPL to the federal government. The reformulation incorporated some of the reform agenda contained in an earlier World Bank-financed environmental DPL—the 2004 First Programmatic Reform Loan for Environmental Sustainability (Env PRL). The Ministry of Environment was added as an implementing agency, the objective was framed at a national scale, and the size of the program doubled to US$2 billion to create the SEM DPL. BNDES remained an implementing agency and a major focus of the policy actions. BNDES was also on- lent the entire amount of the DPL funds from the federal government. xi The evolution of this operation—including significant changes to the financing instrument, the objectives, and the implementing agencies—created issues with the project logic, the delay in effectiveness, and the Bank’s reputation that are important for this evaluation. The objectives for the SEM DPL series were stated as follows: “The SEM DPL series supports the GOB’s [Government of Brazil’s] concerted efforts to strengthen environmental management, with particular attention to: improvements in the overall environmental management system, sustainable management of agricultural lands, forests, and water resources; reduction of deforestation in the Amazon; reduction of the environmental degradation of land and water resources that are key determinants of the well-being of the poor; and, promotion of renewable energy” (Program Document para. 172). There was no specific objective for the first operation in the series, the SEM DPL 1. The objectives of the SEM DPL could have been clearer—and they could have included a specific objective for the first loan—but they were undoubtedly relevant, given Brazil’s tremendous environmental wealth of global importance, its rapidly developing economy that is highly dependent on commodities, and the challenges it has faced in balancing the tradeoffs between the two. Although the environment was a priority for the World Bank’s engagement in the country for many years, the 2008-2011 Country Partnership Strategy (CPS) for Brazil notes that “there has been mixed success in dealing with deforestation and other major environmental challenges.” The CPS recognizes that for 2008-2011 the top priority of the Brazilian government was to accelerate economic growth in light of Brazil’s low growth rates when the CPS was prepared. The CPS also emphasizes that the government continues to be committed to environmental sustainability and maintains sustainability as one of four pillars in the Bank’s program. IEG hence rates relevance of development objective as Substantial. The design of the operation, as reflected in the Development Policy Matrix, suffered from disjointed project logic. This is likely a consequence of the rapid change in direction from preparing a loan to BNDES to preparing the broader SEM DPL and the limited opportunities for engagement on the broader reform agenda given the speed with which the operation had to be prepared. The Policy Matrix consists of nine policy areas and related sub-objectives that are generally in line with the program’s broad, national-level objectives. However, the series of three policy actions and the associated outcome indicator for each policy area inconsistently combine national-level policies and actions and outcomes specific to BNDES. For example, for the sub-objective “Improve sustainability of natural resources management,” the First Tranche Prior Action was to strengthen the federal legal framework by enacting specific forest management laws (for example, the Atlantic Forest); the Second Tranche Release Condition was focused exclusively on BNDES forest programs and guidelines; and finally the series outcome indicator was a general measure of the surface area of public and private forests sustainably managed, with no direct link to BNDES or the Atlantic Forest. As part of the relevance of design to the development objectives, IEG also considers the choice of the lending instrument. The World Bank had already used the programmatic DPL instrument to support very similar objectives in Brazil five years earlier under the Env PRL. That series was canceled after the first US$500 million loan. The design of the xii SEM DPL might have been more relevant if it had been informed by a formal self- assessment of the canceled series: what worked, what did not, what the outcomes of the policy actions were, why the Env PRL series was canceled, and whether a programmatic DPL series was the best instrument for supporting environmental reforms. But the full ICR that is required was not produced and lessons do not appear to have been learned because, once again, a World Bank environmental DPL series was canceled after the first loan. Relevance of design is therefore rated as Negligible. The SEM DPL had major shortcomings in the achievement of its objectives. For a loan— including a DPL—to be effective in achieving its objectives, it should successfully achieve not only the outputs represented by the various policy actions, but also the associated outcomes for the series to which those outputs are designed to contribute. 1 Moreover, these outputs and outcomes should be attributable to the World Bank’s related engagement under the loan with the client on the reforms. While Brazil has made substantial progress in strengthening environmental management in some areas, there is little evidence that the SEM DPL contributed to any of this. The ICR contains no presentation of specific information or analysis on whether or how the reforms that the SEM DPL was supposed to support contributed to the achievement of the overall series objectives. It also does not assess the scale of the DPL’s contribution, as one intervention among many by the GOB and development partners, to an environmental reform process that has been active for the last three decades. Many of the SEM DPL’s “prior actions” 2 were implemented before preparation of the loan even began—in a number of cases, years before. Some are the same as policy actions supported by the Env PRL DPL approved in 2004. Achievement of Objectives for the first objective of improving the overall Brazilian environmental management system is rated as Negligible. With regard to improving the effectiveness of government agencies in implementing mandated Brazilian environmental and social management procedures, the first policy area, the original indicator on the number of judicial challenges to environmental licenses was not being met. The indicator was changed to “number of environmental licenses issued at the Federal level.” This is not a good measure of the intended outcomes because a greater number of licenses could be issued if, for example, standards were lowered, which would not necessarily lead to improved environmental or social outcomes. No other relevant information on “effectiveness” or outcomes is provided in the SEM DPL ICR. Interviews with staff of both Federal and State Public Prosecutor Offices (PPOs) revealed that the number of lawsuits they file on environmental and social grounds has been increasing, due both to the poor quality of environmental impact assessments (EIAs) for investments, as well as a lack of implementation and monitoring of mitigation measures. The indicators also have no direct relationship with the majority of prior actions under the sub-objective, in 1 Note that in the case of DPL programmatic series such as the SEM DPL, even if the series is canceled before it is completed, whichever operations in the series that have completed are still evaluated against the original objectives for the entire series. 2 In a DPL series, the “prior actions” are the first actions to be implemented under each policy area, before the Program Document for the operation is submitted to the World Bank’s Board. xiii particular those related to the Ministry of Environment or the National Biodiversity Management Institute. The actions under this sub-objective related to restructuring and strengthening the Ministry of Environment (MMA) and its affiliates and IBAMA (Brazilian Institute for the Environment and Renewable Natural Resources) are very similar to actions under the 2004 Env PRL—including hiring hundreds of new staff and restructuring the two organizations—and were reported by the World Bank as achievements of that loan. It is possible that the actions were repeated in relation to the SEM DPL, but hiring new staff and restructuring an organization would not necessarily lead—on its own—to the desired outcome of “improved effectiveness of government agencies” with respect to environmental licensing. Indeed, in 2013 IBAMA, the Brazilian environmental licensing agency, issued Terms of Reference for consultancies paid for by another World Bank- financed project—The National Environment Program 2—to improve the environmental licensing system and capacity of its staff, citing serious problems that “can compromise the licensing quality” and that were resulting in “the amplification of the environmental conflicts.” With regard to the policy area on mainstreaming climate change in public and private sector investments under this objective, IEG found that progress has lagged behind expectations. The first actions for the two tranche releases of SEM DPL 1 required the drafting and approval of a National Climate Change Action Plan (or “NCCAP”). What was accepted as evidence of fulfillment of the action was, however, the approval by Law 12.187/2009 of a National Policy for Climate Change (NPCC). The NPCC stated that action plans would be prepared, one for each sector. The sectoral plans were supposed to be ready and the National Climate Change Plan revised before the end of 2011, but those deadlines were not met. At the time of this evaluation, the revised National Plan had still not been completed (though the Brazilian government reports that it expects it to be approved in 2014). The government also reports that a series of sector plans and Nationally Appropriate Mitigation Actions (NAMAs) have been produced, though evidence of their existence, status, added value, and relation to the SEM DPL is limited. There is no National Adaptation Plan. The outcome indicator set for this policy area was an “increase in planned signed reductions of 20 million tons of CO2 equivalent,” and was to result from BNDES developing clean development and carbon funds programs in accordance with the NCCAP and its own new environmental and social policy. Projects proposed for financing under the Clean Development Mechanism (CDM) were being prepared in Brazil since 2003, six years before the SEM DPL, and also well before any of the Brazilian national climate change plans were approved. While the ICR reports that “99% of the target of 20 million tons of CO2” was achieved, the actual target was for an additional (not total) 20 million tons of CO2 signed reductions. Moreover, “planned signed reductions” do not necessarily lead to actual reductions. With regard to improving the effectiveness of environmental and social management systems in BNDES and other financial institutions, the only action under this policy area related to ”other financial institutions” was the approval of the revised Green Protocol meant to improve environmental and social standards in Brazilian banks. But the revised Green Protocol ceased to be implemented and lost its relevance. As for the implementation of the new Environmental and Social Institutional Policy by BNDES, xiv there has been little if any discernible improvement in the performance of BNDES’s environmental and social system compared to the one described—and critiqued—by the World Bank team during preparation. One of the pillars of BNDES’s new environmental and social management system supported by the SEM DPL was supposed to be the development and application to investments of a set of at least 13 sectoral guidelines. But only three of the guidelines were ever developed, and they leave out critical sectors like hydropower, forests, soya, and water and sanitation. The SEM DPL Program Document also promised that BNDES would apply its new policies to all of its investments— including “indirect” investments financed through financial intermediaries—but the monitoring indicator for this policy area only covered direct investments, and BNDES’s Environmental and Social Institutional Policy only applies to about half of its disbursements. They also do not apply to BNDES’s increasing portfolio of investments in other countries, many of them with lower environmental standards than Brazil. High profile investments financed by BNDES after its environmental and social system was supposedly improved under the SEM DPL—such as Belo Monte and other hydroelectric dams in the Amazon—continue to experience an array of environmental and social problems, and a lack of transparency on environmental and social safeguards by international standards. Achievement of Objectives for the second objective of integrating principles of sustainable development in key sectors is rated as Modest. A high-profile area where the SEM DPL was supposed to support improvements was in reducing deforestation, both through strengthening of the general forest legal framework and specifically through improved regional planning to reduce deforestation in the Amazon. While Brazil has made major strides in reducing its deforestation rate over the past decade, it has been achieved through policies other than those supported by the SEM DPL. A number of the actions included in the SEM DPL Policy Matrix were already supported by the earlier Env PRL, including the Public Forest Management Law and the Atlantic Forest Law (both enacted in 2006); the preparation and establishment of the Sustainable Amazon Program (PAS); preparation and launch of the Water Resources National Plan; and strengthening of the PRODES sanitation program. With regard to the intended outcome of reducing Amazon deforestation, data show that it peaked in 2004, when the Brazilian government strengthened policies and enforcement on its own, well before the SEM DPL, and continuously decreased over the following years. With regard to improving management and quality of water resources, the ICR reports that 116,144 km of rivers were being monitored as of June 2011, surpassing the program target of 90,000 km. Based on interviews conducted by IEG and a visit to the federal water agency, ANA, Brazil does appear to have made good progress in terms of increasing the coverage of water monitoring sites around the country. ANA has been signing cooperation agreements with an increasing number of states. It is not clear to what extent this can be attributed to the SEM DPL. The first action under the SEM DPL, approval of the Water Resources National Plan (PNRH), was accomplished in 2006, several years before the SEM DPL, while the documentation for the plan was prepared in 2004 and was an action under the earlier Env PRL operation. The quality of the outcome indicator is questionable, as it does not specify which parameters are to be monitored. The most important aspect of the SEM DPL outcome indicator was that the results of water monitoring were to be used “for prioritization of investments for improved water xv quality.” On this count ANA could not cite any examples. Nor did the Bank’s ICR provide any information on the use of the water quality information. With regard to reducing environmental impacts through improved water, wastewater treatment, and solid waste services, the prior action on enacting the National Guidelines for Water Supply and Environmental Sanitation (Law 11,455) was approved on January 5, 2007, approximately two years before the SEM DPL. The ICR reports that 141,280 tons of pollution loads were being reduced by June 2011, surpassing the target of 110,000 tons. These pollution reductions were supposed to be achieved by a combination of BNDES-financed investments “reviewed under new BNDES social and environmental guidelines,” and investments financed under the “updated PRODES program.” Although BNDES financed wastewater treatment investments before the SEM DPL, they did not monitor the pollution reduction results, so the baseline indicator was inaccurately set at zero in the Program Document. BNDES credits the SEM DPL with introducing the practice of measuring pollution reductions, which is a positive sign. Support for the innovative PRODES program is important, but is credited to actions under the 2004 Env PRL, and the program is still growing much slower than expected. With regard to promoting renewable energy potential, the ICR reports that 50,102 terajoules per year were being generated from renewable energy sources supported by BNDES, just short of the 60,000 terajoule target, including wind power, mini hydropower and biomass cogeneration projects. Although BNDES had demonstrated commitment to these investments, the incremental change compared to pre-SEM DPL levels is relatively minor. By far, the biggest share of BNDES’s renewable energy investments continue to go to large hydropower dams. Meanwhile, the target for this policy area is specific to BNDES, and has little relation to the prior action on introducing the Integrated Environmental Assessment (IEA) methodology in the Brazilian National Handbook of the Electricity Sector’s Inventory. Overall, much more should have been expected in terms of outcomes from the largest ever single World Bank loan to Brazil for US$1.3 billion. A Substantial rating for Relevance of Objectives, one Modest and one Negligible rating for Achievement of the Objectives, and a rating of Negligible for Relevance of Design, leads to the SEM DPL Outcome being rated Unsatisfactory. Risk to Development Outcome is rated Moderate. The relevant outcomes that could be attributed to the SEM DPL operation were limited, so there are fewer achievements to sustain. In some environmental policy areas where actions began before the SEM DPL— such as reducing deforestation—there appears to be good government ownership and momentum. For others—such as climate change and the Green Protocol—progress has been slower than expected. As for the key actions aimed at improving BNDES’s environmental and social management system, while BNDES reports that it has maintained or continued to develop policies and institutions related to its environmental and social management system, there remains little evidence of progress on putting these changes into practice to improve the outcomes of its investments. World Bank performance is rated Unsatisfactory. A major shortcoming in the design of the SEM DPL was the lack of learning from the lessons of the 2004 Env PRL. By xvi proceeding with preparation of the SEM DPL without having properly evaluated the earlier environmental DPL through the required ICR, the World Bank repeated the same mistake, designing the SEM DPL as a programmatic series that was again canceled after the first loan. The World Bank team also included in the SEM DPL Policy Matrix prior actions that had been completed years before the Bank’s engagement on the SEM DPL began—a number of which were policy actions already supported under the Env PRL. This materially reduced the added value of the SEM DPL from the outset, and even more so once the remainder of the series was canceled. The World Bank carried out no analytical work on the critical area of BNDES’s environmental and social management system. In the interest of transparency, the World Bank team also could have mentioned in SEM DPL program documents that other instruments had been considered, and that the funds would be on-lent to BNDES. The incoherence of the DPL’s results framework is likely due to the rapid change in the financing instrument from a Financial Intermediary Loan (FIL) to a DPL, and a preparation process that afforded little time for the Bank SEM DPL team to engage outside of BNDES with the many government agencies implicated in the national-level reforms. There was also little collaboration or coordination with partners—particularly the International Finance Corporation (IFC)—which had been working with BNDES to improve its approach to environmental and social management by adopting the Equator Principles before the SEM DPL was prepared. The World Bank’s supervision also suffered from important shortcomings. The Bank should have been more candid in reporting on the performance of the SEM DPL, including in the Second Tranche Release Document and the ICR. This led to controversy and serious questions raised by civil society. Despite the lengthy effectiveness delay, the Bank held off on downgrading the operation’s implementation status for more than a year. Supervision also failed to ensure the implementation of the promised monitoring and evaluation system, or to provide the promised technical assistance to BNDES. The Bank’s ICR provided no information on feedback from stakeholders, and no beneficiary assessment. Borrower Performance is rated Moderately Unsatisfactory. The 13-month delay in effectiveness of the operation undermined part of the rationale for a DPL as fast- disbursing budget support, and was not dealt with in a timely manner. According to the World Bank ICR, the delay was mainly due to the government having forgotten to include the loan in its budget, and an “extraordinarily lengthy” senate approval process. The cancelation after the first loan of an environmental DPL series that the Government had committed to—for the second time in a matter of years—combined with the lack of progress on a number of environmental policy areas discussed in Section 4, indicates uncertainty and a degree of lack of commitment by the Government to the SEM DPL reform agenda. BNDES demonstrated little if any results in improving its environmental and social management system under the SEM DPL, and the limited progress has continued in the intervening years. There is no evidence that the Ministry of Finance used its influence to spur the promised BNDES reforms. The federal government lacked follow-up on implementation of the Green Protocol. xvii The Ministry of Environment has made notable progress in some areas like reducing Amazon deforestation (which began well before the SEM DPL). In other areas such as mainstreaming climate change action and improving the environmental licensing system, progress remains limited four years after the SEM DPL. Finally, the Ministry of Finance and, ultimately, the Ministry of Environment were responsible for implementing the monitoring and evaluation system for the SEM DPL. The system was never implemented. There is also no evidence that the implementing agencies engaged with Brazilian stakeholders as required by OP 8.60. The assessment includes a number of lessons, which are summarized below: • Certain aspects of the World Bank’s policies and guidance on Development Policy Lending should be clarified, including: o Both the policies supported and financing provided by DPLs can generate significant environmental and social impacts. There should be more specificity on how to approach an operation like the SEM DPL, where all of the funds are being on-lent to a bank and used to finance investments. While DPL funds are legally viewed as general budget support, the public may view such situations as using Bank funds for investments, thus introducing reputational risk to the Bank. o While environmental and social impacts and mitigation mechanisms are supposed to be identified in the Program Document, the requirements on how to monitor and evaluate them are not clear. • Particularly for DPLs focused on reforms in a sector—rather than on macro- economic stability—the impacts of the actions supported can often not be adequately perceived within the short timeframe of the loan, and the tight deadline for submitting an ICR after closure often does not allow for additional outcome evidence to be available and collected. It would be preferable to require that DPL operations wait for a reasonable period either before closing or before producing an ICR—at least one year—in order to allow for adequate monitoring and evaluation. • Back-loading of reforms in a DPL programmatic series can increase the risk of later loans in the series being canceled without full realization of the objectives of the series. • In the future, when attempting to support reforms in state-owned banks like BNDES that finance both public and private-sector investments, it will be important to take a “One World Bank Group” approach. The program would have benefitted from closer collaboration between IFC and World Bank. IFC had been working with BNDES to improve their environmental and social standards before the SEM DPL, but their efforts were later sidelined. • The level of government targeted by DPL reforms should be consistent with the outcomes intended and the client country’s institutional structure. In Brazil, although many national-level laws and policies are approved by the central government, implementation often depends in large part on states and xviii municipalities. But the SEM DPL limited itself to the federal-level implementing agencies. Caroline Heider Director-General Evaluation xix Management Response xx xxi xxii xxiii xxiv xxv xxvi xxvii xxviii xxix xxx xxxi xxxii Chairperson’s Summary: Committee on Development Effectiveness PPAR: Brazil – First Programmatic Development Policy Loan for Sustainable Environmental Management (SEM DPL) and Management Comments PPAR: Brazil – First Programmatic Reform Loan for Environmental Sustainability (Env DPL) and Management Comments Report to the Board of Executive Directors from the C ommittee on Development Effectiveness Meeting of January 21, 2015 The Committee on Development Effectiveness (CODE) met to discuss IEG’s Project Performance Assessment Report (PPAR): Brazil – First Programmatic Development Policy Loan for Sustainable Environmental Management (2009 SEM DPL) (CODE2014-0042), Management Comments on the SEM DPL (CODE2014-0043), IEG’s Project Performance Assessment Report (PPAR): Brazil – First Programmatic Reform Loan for Environmental Sustainability (2004 Env PRL) (CODE2014-0046), and Management Comments on the Env PRL (CODE2015-0001). The Committee welcomed the opportunity to discuss the two country project evaluations. Members noted the different views by Management and the Government of Brazil (GoB) on the IEG approach and findings of the PPAR on the SEM DPL. Several speakers commented that the differing views presented a challenge in assessing the findings and broader lessons. They reiterated the importance of collaboration and clarity in approach and scope among all parties before evaluations are undertaken, and hoped IEG and Management continue their constructive working relationship. The Committee focused on the two PPARs and the DPLs’ achievements and performance, and some members urged Management to consider relevant lessons to help inform future DPF operations; others urged caution given the factual disagreements. Members noted the progress Brazil has achieved in strengthening and mainstreaming environmental sustainability; they supported the Bank’s sustained efforts in this sector and the consolidated partnership between the Bank and the Government of Brazil. Members acknowledged that sector related DPOs raise particular issues related to environmental and social (E&S) risks, and several noted the difficulty in attributing impact and results in DPF operations in general. The Committee looked forward to consideration of the development policy financing retrospective planned for FY16 and, in particular, discussion of E&S risks, implementation, levels of due diligence and monitoring and evaluation requirements in DPLs. Members also looked forward to discussing IEG's broader review of DPFs, expected in the coming months. The Committee agreed that the PPARs will be disclosed with the Management Comments, Borrower Comments and Green Sheet Summary, and that the disclosure should reference these documents and flag that there were different views. ∗ This report is not an approved record 1 1. Background and Context Environmental Context3 1.1 The Federative Republic of Brazil is the largest country in South America and the fifth largest in the world, comparable in size with the US, and covering 8.5 million km2. Thanks in large part to its size and geographic diversity, Brazilian natural resources are diversified and abundant. Such diversity is expressed through a great variety of climates, from the tropical-wet to the semi-arid and temperate, as well as through varied soils, geomorphology and vegetation. Brazil also contains a wealth of mineral and energy resources, including aluminum ore, gold, iron ore, nickel, phosphates, platinum, tin, uranium, petroleum, and hydroelectric power. With more than 5 million km² covered by forests, Brazil also benefits from major wood and other forest product resources, and is home to the largest carbon sink on the planet. Water resources are abundant, but unevenly distributed. Owing to its large size and diverse geography, one of Brazil’s greatest natural resource endowments is its biodiversity. 1.2 In 1989, President Sarney announced Nossa Natureza, Brazil’s first attempt at formulating a national environmental policy, which announced a number of emergency measures. 4 This move was seen by many as a response to mounting domestic and international criticism over the country’s apparent lack of environmentally sound development policies. The main practical and immediate result of Nossa Natureza was the setting up of Brazil’s environmental control agency IBAMA. The Ministry of the Environment itself (MMA) was set up in 1993 (see Annex F for a diagram of Brazil’s federal environmental agencies). 1.3 Brazil’s environmental profile was heightened when it took on a leadership role in international environmental matters by hosting the UN Conference on the Environment and Development (UNCED) or “Earth Summit” in Rio de Janeiro in 1992, and also hosted the global Rio+20 summit in 2012. The country was the first to sign the Convention on Biological Diversity (UNCBD) and the United Nations Framework Convention on Climate Change (UNFCCC), during the conference. 1.4 Yet even as it was taking on a leadership role internationally, the idea of incorporating an environmental dimension into development policies and strategies was still highly controversial in Brazil. There had for example been strong resistance by Amazon state governors to the setting up of local environmental control agencies (OEMAs), because they saw this as a potential constraint on development. Environmental management was dominated by a centralized command-and-control approach. IBAMA was provided with statutory powers by the Environmental Crimes Law, which was finally approved in 1998 in the face of strong political opposition. It imposed large fines and prison sentences for a 3 This section draws heavily on IEG’s Brazil case study for the global evaluation of environmental management, in some cases verbatim. 4 These measures included attempts to curtail Amazon deforestation, limits on log exports, the creation of several national parks and setting up of a National Environment Fund (FNMA). 2 range of offenses including illegal logging, pollution, and illegal hunting. However, the law was poorly enforced and few fines were ever collected. IBAMA regularly comes under strong political pressure to relax restrictions where commercial interests are at stake; for example, by the steel industry in Pará which relies heavily on native forests for charcoal supplies in pig-iron smelting. 1.5 Brazil has placed a strong emphasis on conservation of the natural resource base. The National System of Conservation Units (SNUC) was established in 2000. According to a recent definitive study of Amazon deforestation, “From 2004 through 2012, protected areas and indigenous territories grew 68% to encompass 47% of the entire Brazilian Amazon region” (Nepstad and others 2014). However, problems of monitoring and enforcement of environmental laws remain. 1.6 While the federal government retains significant control over the ‘green’ environmental agenda through the MMA and IBAMA, responsibility for the ‘blue’ and ‘brown’ agendas has increasingly been devolved to state authorities. Rapid urban expansion made it difficult for the government to keep up with the demand for water, sanitation and solid waste disposal. Brazil’s 1988 Constitution decentralized responsibility for addressing water and sanitation-related pollution issues to the states and municipalities. 1.7 Environment has been one of the pillars of the World Bank’s program in Brazil for decades. The portfolio review for the recent IEG Brazil Country Program Evaluation (CPE) found more than one hundred World Bank operations in Brazil with an environment theme over the past decade. 1.8 The World Bank provided an earlier environmental policy loan to Brazil. Called the First Programmatic Reform Loan for Environmental Sustainability (Env PRL), it was designed as a programmatic series of three loans. The first, for US$ 500 million, was approved in 2004. Like the SEM DPL though, the subsequent loans never materialized and the rest of the series was canceled. A full Implementation Completion and Results Report (ICR) was never completed, and as a result IEG was not able to assign the loan an outcome rating. 5 Development Context6 1.9 Brazil made substantial achievements in fiscal adjustment and price stabilization in the late 1990s and early 2000s. But the resilience and continuity of that stabilization effort was tested in the 2000s by the global economic slowdown, a domestic energy crisis, spillovers from the Argentine crisis, and uncertainties related to the 2002 presidential election. The subsequent macroeconomic stability and a favorable external environment allowed Brazil to resume moderate growth from 2004. 5 The Env PRL is the subject of a separate IEG PPAR under preparation. 6 This section draws heavily on IEG’s 2013 Country Program Evaluation (CPE) of Brazil, in some cases verbatim (IEG 2013). 3 1.10 Gross Domestic Product (GDP) grew by nearly 5 percent per year between 2004 and 2008 with some fluctuations. Brazil was one of the last nations to fall into recession in 2008 and among the first to recover; the country’s sound fundamentals and prompt response helped mitigate the decline. Brazil has also made considerable progress in its long-term foreign currency sovereign credit ratings. Standard & Poor’s rating for Brazil improved by 4 notches from noninvestment grade BB- in 2003 to above investment grade of BBB in 2011. 1.11 Brazil’s population has been urbanizing rapidly. Rural out-migration has been encouraged by agricultural modernization and land concentration in the Southeast and Northeast with the expansion of major commercial and export crops such as sugar, cattle, wheat, and soybean. Development of transport, roads, and communications infrastructure has stimulated frontier settlement westwards, with implications for the environment. The growth of Brazilian manufacturing industry through import substitution and exports as well as expansion of the tertiary service sector has further catalyzed urban growth. 1.12 Brazil has made substantial progress in reducing poverty and income inequality. Poverty declined from 35.8 percent of the population in 2003 to 21.4 percent in 2009; and extreme poverty fell from 15.2 percent in 2004 to 7.3 percent in 2009. Non-income indicators of standards of living have also improved; for example, there have been reductions in child malnutrition and increases in primary school enrollment. Gender differences in enrollment have been eliminated. 1.13 Nevertheless, substantial development challenges remain. Although the growth rate during the past decade was higher than in the preceding two decades, it was lower than major emerging countries in most of the years, and has recently slowed significantly from 7.5 percent in 2010, to 2.7 percent in 2011 and 0.9 percent in 2012. 7 Much of the body of literature on this topic maintains that accelerating Brazil’s economic growth requires sharp increases in investment rates, particularly in infrastructure, which was low relative to other emerging market economies over the past decade. Key issues seem to be weak incentives for private investment and low savings rates. Regulatory reform is needed to encourage private investment in infrastructure and to reduce the cost of doing business in order to make Brazil more competitive (The Economist 2013b). The environment is also relevant to the economic situation, including the difficulty of combining agricultural growth and poverty reduction with protection of the environment in general, and forests in particular. Operational Context 1.14 To properly evaluate the SEM DPL, it is important to understand the history and evolution of the operation, and the context in which it was prepared. Over the course of the preparation process, there were significant changes to the design—including to the financing instrument, the objectives, and the implementing agencies. The changes in turn affected issues that arose later, including with the project logic, the delay in effectiveness, the inclusion of actions previously supported by the Env PRL, and the Bank’s reputation. Those 7 http://www.worldbank.org/en/country/brazil/overview 4 issues are important for this evaluation. A more complete history of the operation and a Timeline with the many steps involved are provided in Annex H. 1.15 Preparation of the operation began with a World Bank mission to the headquarters of the National Bank for Economic and Social Development (BNDES) in Rio de Janeiro in November 2007. BNDES is a wholly owned Brazilian federal government development bank that finances both private and public investments and is the main source of long-term financing in Brazil, including large infrastructure projects, and whose annual lending has grown to several times that of the World Bank globally (BNDES 2013a). 8 BNDES has also been increasing its presence internationally, particularly in other countries in Latin American, where it has become one of the most important financial institutions, and in Africa, where it recently opened its first representative office (BNDES; Financial Times 2010, 2012, 2013b, 2013c; BIZ 2011). 9 1.16 The operation was referred to as the “BNDES PAC-Env DPL” and “a potential PAC and environmental and social policy operation.” The “PAC” is the Portuguese acronym for the Brazilian government’s flagship Program for Accelerated Growth to support a major scaling-up of infrastructure investment, with “BNDES as the primary channel for federal financing of this ambitious program.” At the same time, BNDES had received criticism nationally and internationally for the adverse environmental and social impacts of some of its investments, including in large slaughterhouse operations in the Amazon (see for example, Greenpeace 2009), and both the BNDES and World Bank teams recognized that there were significant weaknesses in BNDES’s approach to environmental and social safeguards (see Program Document). 1.17 According to the World Bank’s 2008-2011 Country Partnership Strategy (CPS) for Brazil and the Country Director at the time, the World Bank consciously made a major change in direction at the time in deciding to engage in financing development in the Amazon—in sectors like energy, agriculture, mining, and infrastructure—and to do it in an environmentally sustainable way (World Bank 2008). The planned World Bank loan to BNDES was seen as a manifestation of the new focus. 1.18 In March 2008, the Government of Brazil External Financing Commission (COFIEX) authorized the preparation of a “DPL/BNDES-World Bank” program. It specified that the loan would be in the amount of US$1 billion, with BNDES as the only beneficiary, and the Federal Government of Brazil as the guarantor. The DPL is the World Bank’s main instrument for providing a client country with budget support through “rapidly disbursing policy-based financing”, and its use is governed by Operational Policy (OP) 8.60. 10 Although up until that point the idea was to prepare a DPL to BNDES, the first two Concept Notes 8 http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/The_BNDES_in_Numbers/ 9 http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/Press/Noticias/2013/20131206_a frica.html 10 See World Bank Operational Manual, available at: http://web.worldbank.org/WBSITE/EXTERNAL/PROJECTS/EXTPOLICIES/EXTOPMANUAL/0,, menuPK:64142516~pagePK:64141681~piPK:64141745~theSitePK:502184,00.html 5 issued for the operation in June and September 2008 were for a Financial Intermediary Loan (FIL) from the World Bank to BNDES titled the “BNDES Environmental and Social Sustainability Project” (though still with the same project code, P095205). The proposed US$1 billion in financing was to be divided into two Components: (A) Strengthening of Environmental and Social Screening and Monitoring would provide US$10 million for technical assistance and training to BNDES; and (B) a US$990 million Line of Credit to finance BNDES investments in infrastructure and green lines of business. 1.19 The Concept Notes explain that the originally proposed DPL instrument was rejected because a DPL directly to BNDES would require an exception to OP 8.60, and for the government to borrow from the World Bank and on-lend it to BNDES “is not consistent with the financial requirements of the country” due to concerns about the impact on external debt statistics. But internal World Bank reviewers of the Concept Notes expressed serious concerns about BNDES’s ability to comply with the requirements of a FIL—a type of investment loan—on environmental and social safeguards, interest subsidies, financial management, and procurement. 1.20 Meanwhile, Lehman Brothers filed for bankruptcy in September of 2008, sparking the global financial crisis. At the time there were concerns that “despite its increased resilience, Brazil has been hit hard by the global financial crisis” (World Bank 2009). In response, the Brazilian Government “announced the adoption of countercyclical fiscal policies,” including accelerated financing of infrastructure investments through BNDES. At the World Bank Annual Meetings in October 2008, a meeting was organized between the BNDES President, World Bank management, and Brazilian government officials to discuss this loan. BNDES needed the money to help increase its lending, but they objected to applying World Bank FIL requirements—particularly restrictions on interest rate subsidies, environmental and social safeguards, and procurement rules. So agreement was reached to abandon the idea of a FIL, and change instruments again. This time, the Brazilian Federal Government would borrow the money from the World Bank in the form of a broad, national- level environmental DPL, and on-lend the money to BNDES. 1.21 From that point, the operation was prepared very quickly. In November 2008, another Concept Review Meeting was held, this time for the “Brazil: First Programmatic Development Policy Loan for Sustainable Environmental Management”, or SEM DPL. A programmatic DPL was now proposed with nine policy areas and two objectives that were virtually the same as the objectives for the earlier Env PRL: improve the Brazilian environmental management system, and mainstream environment into targeted sectors. The lending amount for the series was doubled from US$1 billion to US$2 billion, and the Ministry of Environment was added as an implementing agency. Though the title and objectives no longer mentioned BNDES, BNDES remained an implementing agency and a major focus of the reforms the SEM DPL aimed to support. On the part of the Borrower, the plan to on-lend all of the SEM DPL funds to BNDES was made official soon after through Interim Measure 450 of the Brazilian Presidency in December 2008 (Presidency of the 6 Republic 2008; Senate of Brazil 2008), subsequently codified in Brazilian Law Number 11.943, Article 15 (Presidency of the Republic 2009). 11 1.22 The use of the DPL instrument allowed the World Bank to prepare the loan and commit the money much faster than would have been possible under a FIL—without having to address FIL requirements related to interest rate subsidies, environmental and social safeguards, and fiduciary oversight. Less than three months later, on February 3, 2009, the Program Document was finalized for the SEM DPL, and the SEM DPL 1 loan was approved by the World Bank Board on March 5, 2009. That contributed to the Bank reversing a steep decline in lending to Brazil, which had fallen from $3.2 billion over 2005-6 to less than $300 million in 2007. 1.23 The speed with which the operation was prepared once the decision was taken to change to a DPL to the federal government afforded little time for the Bank SEM DPL team to engage outside of BNDES with the many government agencies implicated in the national- level reforms. Moreover, despite the argument made in the Program Document that “the aim of ensuring adequate credit resources to the financial system is a key motivating factor of the GOB’s request for the SEM DPL” in response to the financial crisis, the loan then lingered for more than a year before the Loan Agreement was signed and the operation made effective. Based on interviews and documentation, IEG understands that the delay was due to the need for the loan to be approved by the Brazilian Senate, which took some time. Legal Opinions prepared prior to the Senate vote explain that all of the funds provided by the World Bank through the SEM DPL would be directed to BNDES, although they also emphasize that because the money is being lent under the World Bank’s DPL instrument, the proceeds from the loans “are not allocated for investments.” 1.24 The SEM DPL was signed and became effective on June 21, 2010. The first tranche of US$800 million was released on June 30, 2010, fifteen months later than what was planned when the loan was approved by the World Bank Board. The second tranche of US$500 million was disbursed on December 15, 2010, seventeen months late. Thus the total amount of US$1.3 billion for SEM DPL 1 was disbursed. Each of these disbursements was accompanied by a detailed Financing Contract signed between the Federal Government of Brazil and BNDES that in many respects passed on to BNDES the terms of the SEM DPL Legal Agreement signed between the federal government and the World Bank. In addition, each of the Brazilian Financing Contracts specified that “this contract aims at granting credit to BNDES…to provide it with the resources for application in its investment operations.” 1.25 The SEM DPL was closed two weeks after the second tranche was disbursed, on December 31, 2010. BNDES subsequently confirmed that the SEM DPL resources were provided as a concessionary loan by the Federal Treasury “to complement the BNDES disbursement budget.” Indeed, BNDES ramped up its lending in a major way at the time of 11 Available at: http://www.planalto.gov.br/ccivil_03/_ato2007-2010/2008/Mpv/450.htm : and http://www.planalto.gov.br/ccivil_03/_ato2007-2010/2009/Lei/L11943.htm . The Presidency website notes that Interim Measure 450 was revoked once Law No. 11.943 was approved (thus the strikethroughs). 7 the SEM DPL, reporting that its disbursements increased by 23% in 2010 (BNDES 2013a). 12 The planned second loan in the series, SEM DPL 2, never materialized. 1.26 To provide perspective on the scale of the SEM DPL 1, it was equivalent to the total average annual financing commitment for the environment for the entire World Bank from 2008-2012, including financing from the Global Environment Facility (GEF), and it was ten times larger than the World Bank’s total global GEF-financed commitments in 2011 or 2012. 13 Looking only at the non-DPL environment portfolio (since the SEM DPL itself is included in the overall calculation), SEM DPL 1 was more than 2.5 times bigger than the World Bank’s global average annual financing commitment for the environment from 2008- 2012. 2. Objectives, Design, and their Relevance 2.1 The objectives for the SEM DPL series are stated in the Program Document as follows: The SEM DPL series supports the GOB’s concerted efforts to strengthen environmental management, with particular attention to: improvements in the overall environmental management system, sustainable management of agricultural lands, forests, and water resources; reduction of deforestation in the Amazon; reduction of the environmental degradation of land and water resources that are key determinants of the well being of the poor; and, promotion of renewable energy. (Program Document, para. 172) 2.2 Although the SEM DPL was designed as a programmatic series of two loans, the series was canceled after the first loan. There do not appear to have been separate objectives for the individual loans in the series, including SEM DPL 1. For the purposes of this evaluation, IEG construes the objectives of the SEM DPL as follows: a) Improve the overall Brazilian environmental management system; and b) Manage natural resources sustainably, reducing the degradation of agricultural lands, forests (in particular the Amazon), and water resources, and promoting renewable energy. Relevance of Objectives 2.3 Improving Brazil’s overall environmental management system, and managing its natural resources sustainably, is relevant to Brazil’s national priorities and the Bank’s support program in the country. Brazil contains a wealth of natural resources, including some that are 12 In its 2013 Brazil Economic Survey, the OECD was critical of the significant increase in transfers from the national budget to BNDES, noting that “a large share of BNDES loans is extended to large companies… [which] would be the ones that would probably enjoy the easiest access to credit in private credit markets.” It recommends that “private entry will require levelling the playing field by phasing out all direct and indirect financial support to BNDES.” (OECD 2013) 13 Source: World Bank Business Warehouse. 8 of global importance, and as a rapidly developing middle-income country whose economy depends to a large extent on its natural resources, there are important environmental challenges that must be carefully managed in order to sustain Brazil’s progress over time. The Program Document highlights the inclusion in Brazil’s 2008-2011 Multiyear Plan (PPA) of “sustainable use of natural resources” as the fifth of eight themes, and charts various government programs and plans that relate to the different SEM DPL sub-objectives dealing with natural resource management. 2.4 Although the environment is clearly important for Brazil and was also a priority for the World Bank’s engagement in the country for many years, the 2008-2011 Country Partnership Strategy (CPS) for Brazil recognized that the top priority of the Brazilian government was to accelerate economic growth in light of Brazil’s low growth rates when the CPS was prepared. This effort included the PAC program to support infrastructure investment discussed in the Background section. At the same time, the CPS asserts that the government continues to be committed to environmental sustainability, and maintains sustainability as one of four pillars in the Bank’s program. 2.5 On the other hand, the statement of the SEM DPL series objectives is vague, and appears overly broad and ambitious. It is also formulated differently from what is presented as Sub-objective B in the Development Policy Matrix of the Program Document (see Annex B), which is “Integrating Principles of Sustainable Development in Key Sectors.” The Program Document also lacks a specific objective for the SEM DPL 1 loan, as distinct from the programmatic series objective, as recommended by OPCS. Overall, the relevance of objectives is rated Substantial. Design 2.6 The SEM DPL was designed as a programmatic series of two loans for a total of approximately US$2 billion. The first loan (SEM DPL 1) was for a total of US$1.3 billion divided into two tranches of US$ 800 million and US$ 500 million. The full US$ 1.3 billion of the SEM DPL 1 was financed by the World Bank on IBRD terms. 2.7 The key to understanding the design and intended program logic for any DPL is the policy matrix. For the SEM DPL, the “Development Policy Matrix” from the World Bank’s Program Document is reproduced in Annex B, including the outcome indicators that were to be achieved by the program by 2011. For a programmatic series such as this, outcome indicators are established for the entire series. 2.8 The policy actions under the two tranches of the SEM DPL were organized around nine “Key Issues and Objectives”, or policy areas, which covered a broad array of environmental issues: green, brown, and blue. The policy areas are here regrouped into two sub-objectives: A. improving the overall Brazilian environmental management system, and B. integrating principles of sustainable development in key sectors. Sub-objective B was further divided into four policy areas: natural resource management and conservation, water resource management, environmental sanitation, and renewable energy. The policy areas, as 9 originally designed in Annex 3 of the Program Document, are described in detail in the discussion that follows and summarized in Annex B. A. Improving the overall Brazilian environmental management system 2.9 The program sought to support policy reforms to improve and standardize the environmental management system by: (i) improving the effectiveness of government environmental agencies, including the MMA and IBAMA, in implementing mandated Brazilian environmental and social management procedures, through restructuring and staffing increases; (ii) mainstreaming climate change in public and private sector investments by drafting and approving a National Climate Change Action Plan, and by BNDES implementing clean development and carbon funds programs; and (iii) improving the effectiveness of environmental and social management systems in BNDES and other financial institutions, through approving a revised Green Protocol to be approved by all federal public Brazilian banks, the approval and application of a new Environmental and Social Institutional Policy by BNDES to all its directly financed operations, and finally the expansion of the application of this new policy to BNDES’s full portfolio. 2.10 By supporting these measures, the SEM DPL aimed to: (i) improve the environmental licensing process using as a proxy a 20 percent decrease in the number of judicially challenged licenses by the Public Prosecutor’s Office (this indicator was later changed; see below); (ii) increase planned signed reductions of 20 million tons of CO2 from CDM projects, BNDES projects, and other actions under the National Climate Change Action Plan; and (iii) achieve 100 percent of projects submitted directly to BNDES screened, approved and monitored according to the new Environmental and Social Institutional Policy. B. Integrating principles of sustainable development in key sectors B.1. Natural Resource Management and Conservation 2.11 The program sought to support policy reforms to implement an integrated strategy to address issues of deforestation, biodiversity loss and unsustainable agriculture, and livestock production by: (i) improving the sustainability of natural resource management, through strengthening of the forest legal framework, restructuring BNDES programs on forests and agriculture, and designing three sub-sectoral guidelines and the REFLORESTA Program based on the new legal framework and BNDES’s new Environmental and Social Institutional Policy; 10 (ii) improving Amazon regional planning for sustainable development and reduced deforestation, through approval of the National Sustainable Amazon Program (PAS), and completion of the Ecological Economic Zoning of the Amazon Region; and (iii) improving rainforest conservation, through issuance of a Presidential decree regulating the Amazon Fund, making the Amazon Fund operational, and issuance of a Presidential decree to regulate and create operational mechanisms for the Atlantic Forest Fund. 2.12 By supporting these measures, the SEM DPL aimed to: (i) expand sustainable natural forest management of private and public areas from 27,000 km2 to 50,000 km2; (ii) reduce Amazon deforestation by 20 percent; and (iii) promote sustainable use of natural resources by supporting 500,000 ha with the Amazon and Atlantic Forest Funds. B.2. Water Resource Management 2.13 The program sought to contribute to improved management and quality of water resources through the approval of the Water Resources National Plan and the National Water Quality Evaluation Program, and implementation of the National Water Quality Evaluation Program in ten states. 2.14 By supporting these measures, the SEM DPL aimed to achieve regular water quality monitoring for 90,000 km of main rivers, with the results publicly released and used to prioritize investments for improved water quality. B.3. Environmental Sanitation 2.15 The program sought to support reforms that would improve potable water, wastewater and solid waste service delivery by “supporting the development of related environmental and social regulations, the implementation of integrated solid waste management plans by local and state authorities, and the improvement and scale up of innovative results-based financing mechanisms.” The Program Document (p. 47) further promises that “these reforms would…improve the quality of life and health of the population, particularly in poor and marginal areas.” The specific reforms related to this policy area were: to enact Law 11,445/07 on National Guidelines for Water Supply and Environmental Sanitation, to ensure that BNDES programs are coherent with the new legal framework and BNDES’s new Environmental and Social Institutional Policy, for ANA (the Brazilian National Water Agency) to update and approve new PRODES (Program for Depolluting River Basins) rules and regulations governing payments for wastewater treatment (in accordance with the new legal framework), and to design two new BNDES sub-sectoral guidelines to ensure coherence with the new legal framework and with BNDES’s new Environmental and Social Institutional Policy. 2.16 By supporting these measures, the SEM DPL aimed to reduce pollutions loads by 110,000 tons of biological oxygen demand (BOD) per year from BNDES projects and the PRODES program. B.4. Renewable Energy 11 2.17 The program sought to promote the use of renewable energy in Brazil by diversifying energy sources and developing innovative technologies for alternative energy sources. The Program Document further stated that these reforms “would contribute to providing access to safe and renewable energy and related services, thereby reducing air pollution and mitigating its impacts on human health and associated costs related to treatment of respiratory diseases and the loss of income generation opportunities.” The specific reforms related to this policy area were: to include an Integrated Environmental Assessment (IEA) methodology for improving the environmental and social sustainability of the hydroelectric sector in the handbook of the Electricity Sector’s inventory and apply it in ten river basins, to ensure that BNDES programs for energy efficiency and renewable energy are coherent with BNDES’s new Environmental and Social Institutional Policy, and to ensure the coherence of six BNDES sub-sectoral guidelines for renewable energy with BNDES’s new Environmental and Social Institutional Policy. 2.18 By supporting these measures, the SEM DPL aimed to result in 60,000 terajoules per year produced by renewable energy sources or saved energy efficiency under projects supported by BNDES. Monitoring and Evaluation System Design 2.19 With regard to design of the M&E system, the Program Document includes a Results Framework Analysis (Table 5) with a Baseline, Intermediate Outcome Indicators, and Program Outcome Indicators for each of the Key Issues and Objectives in the Development Policy Matrix (Table 4). There are however problems with the indicators used, which are described in the next section. With regard to implementing a monitoring system in practice, the Program Document states: “In order to monitor this operation and the entire SEM DPL series, the Ministry of Finance and the Ministry of Environment will use the monitoring system developed through the Technical Assistance Loan for Environmental Sustainability project (ENV TAL) to monitor the previous PRL and the ongoing ENV TAL.” It further states that: “This system is web-based and open to the public, ensuring transparency and enabling social participation.” In terms of institutional responsibility for the monitoring system, the Program Document indicates that “this monitoring system is under the responsibility of the project management unit of the ENV TAL under the direct supervision of MMA’s Executive Secretariat.” There is no separate mention of how actions to be implemented by BNDES, which is not under MMA, would be monitored or evaluated. Normally MMA, through IBAMA, would be responsible for legally mandated environmental licenses in relation to BNDES-financed investments, but otherwise would not have an apparent role in the monitoring of the more ambitious changes to its environmental and social policies that BNDES aspired to under the SEM DPL. It would be up to BNDES to monitor the application of these new polices themselves, or make arrangements for someone else to do it. There is however no mention in the Program Document of such arrangements. Implementation Arrangements 2.20 The SEM DPL Program Document officially lists the Ministry of Finance (MOF) as well as the Ministry of Environment (MMA) and BNDES as implementing agencies. 12 However, the MOF was primarily responsible for the flow of funds on behalf of GOB, and there were no policy actions related to it. The ICR lists only MMA and BNDES as implementing agencies. 2.21 There were a number of policy actions that were meant to be carried out by agencies not listed as implementing agencies of the operation, such as IBAMA, ANA, ICMBio, and the Forestry Service (SFB). The description of monitoring arrangements in the text of the Program Document (p. 59) states that “the Ministry of Environment and its agencies (IBAMA, ANA, ICMBio, and the Forestry Service (SFB)), and BNDES will be responsible for implementing the proposed components of the operation.” As shown in Figure 1, most of these agencies are autonomous agencies linked to the MMA, except for the Forestry Service, which is directly subordinated to MMA. BNDES has no direct reporting relationship with MMA. It is a federal government-owned bank that reports to the Ministry of Development, Industry, and Foreign Trade, which is not a SEM DPL implementing agency (Colby 2012). IBAMA, linked to MMA, is responsible for the environmental licensing process for Federal- level (that is, spanning more than one state or related to multi-state water bodies) investments financed by BNDES or anyone else. Otherwise, MMA normally has no direct role in the governance of BNDES or in monitoring its internal practices or portfolio. 2.22 Although there is normally no reporting relationship between MMA and BNDES, the SEM DPL Program Document described MMA as being responsible for implementation M&E. In reality, BNDES provided reports such as their input to the ICR to the World Bank team separately from MMA. The Program Document does not explain who will be responsible for implementation and monitoring of the outcomes and mitigation measures related to the “poverty and social impacts” or “environmental aspects” of the operation described in the Program Document. 2.23 The Fiduciary Arrangements and Disbursement section of the Program Document (p. 64) states that “the proceeds of the loan would be deposited…into an account of the National Treasury of the Federative Republic of Brazil, established at the Banco do Brasil for the Borrower’s use.” There is no mention of BNDES, MMA, or any other agencies in this section. Relevance of Design 2.24 As noted in section 2.a, the statement of objectives in the Program Document was not very clear. The sub-objectives in the statement of objectives in the “Operation Description” section also differ somewhat from the sub-objectives listed in the Development Policy Matrix of the Program Document (see Annex B). For example, there is no mention in the Development Policy Matrix sub-objectives of the “sustainable management of agricultural lands, forests, and water resources” that is stated in the series objective. There was no clear statement of the specific objective of the first operation, or the planned second operation. Part (a) of the series objective in particular—“improve the overall Brazilian environmental management system”—is very broad and ambitious, and not clearly defined. It is not clear whether it is referring to the federal government alone, or whether it also includes the decentralized environmental systems of the states and municipalities as well, and whether the private sector is also included. 13 2.25 Inconsistencies in the design were created when the World Bank team switched from preparing a loan to BNDES to preparing the broader SEM DPL. Most of the policy areas are similar to those of the earlier, national-level Env PRL DPL, as are the objectives. But the Env PRL did not include BNDES. The relatively narrow focus of many of the policy actions included under the SEM DPL Policy Matrix is inconsistent with the broad scope and ambitious scale of the objectives. While the objectives promise improvements in the management of the entire country’s environment and main natural resources, six of the nine policy areas have actions that relate to a single enterprise – BNDES. 14 2.26 BNDES is the largest source of long-term finance for infrastructure investment in Brazil. BNDES’s lending volume is also larger than the World Bank’s global lending volume, and it is increasingly active not just in Brazil but in many other countries, particularly in Latin America and Africa. So improving the sustainability of BNDES’s investments, in particular through the sub-objective of improving their environmental and social management systems, has the potential to leverage increased sustainability throughout their substantial portfolio. 2.27 But BNDES is still only one of many public and private banks operating in Brazil, and according to the Program Document was only the fifth-largest bank in Brazil. BNDES represents less than seven percent of the Brazilian credit market, and even among state- owned banks (including other large state-owned banks like Banco do Brasil, Caixa Economica Federal) its share is only 14 percent (G1 2013a, Infolatam 2013). 15 It could be argued that BNDES’s role was relatively more important in the case of infrastructure investments. An analysis of sources of investments in industry and infrastructure in Brazil from 2001-2011 found that for the first half of the decade BNDES’s share of the total was around 20 percent (Colby 2012). 16 This increased substantially in the second half of the decade with the PAC and global financial crisis, to a peak of 52.5 percent in 2009, but has since fallen back again. However, for other sectors, BNDES’s presence is much smaller. For example for the agriculture sector, which is associated with some of the biggest environmental problems in Brazil, Banco do Brasil is by far the most important lender (MAPA 2013) 17, while Caixa Economica is the most important lender for housing construction, which clearly can also have important economic and social ramifications. 2.28 While improvement of BNDES’s environmental and social systems was no doubt important, to achieve the objective of “improving the overall Brazilian environmental management system” would have required a broader approach involving more banks and other institutions. One of the policy areas, A.iii, does state that it will improve environmental and social management systems in “other financial institutions” in addition to BNDES, and 14 Including actions related to the Amazon and Atlantic Forest Funds, which are managed by BNDES. 15 http://g1.globo.com/economia/seu-dinheiro/noticia/2013/01/bancos-publicos-emprestam-quase-4- vezes-mais-do-que-privados-em-2012.html; and http://www.infolatam.com.br/2013/01/23/bndes- libera-valor-recorde-de-credito-em-2012/ 16 Colby, S. 2012, “Explaining the BNDES: what it is, what it does and how it works”, Centro Brasileiro de Relacoes Internacionais, Volume 3. http://www.cebri.org/midia/documentos/bndes.pdf 17 http://www.agricultura.gov.br/arq_editor/Pasta%20de%20Junho%20-%202013.pdf 14 this is meant to be achieved through the signing of the Green Protocol 18 by “all federal public Brazilian Banks”. But no other banks are mentioned by name either here or anywhere else in the policy matrix, the reforms are not carried through in subsequent stages of the series (for example, to implementation) the way they are for BNDES, and the Outcome Indicator for Policy Area A.iii only mentions BNDES, and not the other federal Brazilian banks. 2.29 Since the SEM DPL was a loan to the federal government, it is not immediately clear why—if its intention was to improve the sustainability of investments—it did not support policy actions that would have had a broader impact across all major lenders, or at least across all government-owned financial institutions. The focus on one bank appears particularly incongruous in relation to the very broad, economy-wide nature of the objectives. The Bank team justifies the heavy focus on BNDES by stating that “in view of its size and large number of financial agents, BNDES can also serve as a benchmark for other financial institutions.” (Program Document p. 15) However, BNDES’ financial agents, responsible for two-thirds of BNDES’ lending were ultimately not included in the environmental and social improvements targeted by the SEM DPL (the outcome indicator for improvement of BNDES’s environmental and social systems applied only to “projects submitted directly to BNDES”), and as several sources have pointed out more recently, BNDES trails many other Brazilian banks—including private sector banks—in its environmental and social standards (Reporter Brasil 2011; Widmer 2012). This was confirmed in IEG interviews with experts who assess investments in Brazil financed by both private and public sector Banks. So while improving BNDES’s environmental and social management system is important, given the broad objectives of the loan, the actions should have been broadened to include BNDES’s indirectly financed investments as well as other major public and private banks in Brazil. 2.30 Results framework and project logic: The causal chain from the policy actions and triggers to the intended outcomes, sub-objectives, and objectives was disjointed. The Development Policy Matrix from the Program Document is reproduced here in Annex B for convenience. Attempting to follow the logic across the rows—from sub-objective, to actions across each of the SEM DPL stages, to the outcome indicators—reveals that actions across different stages are often not closely related to one another, to the sub-objective, or to the associated outcome indicator. For many of the policy areas and sub-objectives, the actions under the first and sometimes second tranches of the first operation involve the development and approval of national-level policies or laws, but in later stages abruptly switch to actions and outcomes for BNDES with indicators that focus either only on BNDES or only on the other agency, but not both. This design feature meant that in those cases, the policies related to the non-BNDES agencies were not followed through to implementation, and insufficient information was collected to evaluate the outcomes. 2.31 An example of the disjointed project logic is for the Sub-objective B.1.i, “Improve sustainability of natural resources management”, the First Tranche Prior Action was to strengthen the Federal legal framework by enacting three laws that deal with specific aspects of forest management (e.g., the Atlantic Forest); the Second Tranche Release Condition was focused exclusively on BNDES forest programs and guidelines; there was no Trigger specified for the second operation in the series; and finally the series outcome indicator was a 18 See discussion below and Annex C for more information. 15 general measure of the surface area of public and private forests sustainably managed (apparently at the national level), with no direct link to BNDES, the Atlantic Forest, etc. 2.32 A second example is the “Promote renewable energy potential” sub-objective, which begins with an action to improve the environmental and social sustainability of the entire hydroelectric sector by including a methodology for Integrated Environmental Assessment in the electricity sector’s Inventory Handbook, which is the responsibility of the Ministry of Mines and Energy. The next two actions under the policy area relate only to BNDES improving its approach to renewable energy, and finally the indicator is only on the amount of renewable energy generated by BNDES projects. 2.33 A third example of the lack of coherence in the Policy Matrix is for the Sub-objective A.ii, to “mainstream climate change in public and private sector investments.” The first and second tranche actions under DPL 1 involve the drafting and approving of a National Climate Change Action Plan (NCCAP); the DPL 2 Trigger is based on BNDES alone establishing “clean development and carbon funds programs”; and finally the outcome indicator introduces the notion of increasing “planned signed” emission reductions under the Clean Development Mechanism (CDM) 19, which is a fairly narrow and marginal program in a large country like Brazil, and for which investments have been prepared in Brazil since 2003, six years before the SEM DPL, and six years before the NCCAP was approved, according to the ICR. 2.34 Appropriateness of level of government targeted: The SEM DPL implementing agencies were at the level of the federal government. Many overarching environmental laws and policies are approved at the national level, and the federal government is largely responsible for enforcement of laws related to the Amazon and for management of water basins that span more than one state. However, especially with the increasing emphasis on decentralization of government functions in Brazil, implementation of many environmental laws and policies is the responsibility of state and municipal governments. Examples relevant for the SEM DPL include environmental licensing for investments that do not cross state boundaries, and investments in potable water supply, wastewater treatment, and solid waste services. These local government agencies do not figure at all in the SEM DPL design. 2.35 Choice of instrument: As part of the relevance of design to the development objectives, IEG also considers the choice of lending instrument. In the case of this operation, the World Bank team explored several different options in terms of the instrument and implementing agencies as part of the design process. Serious questions can be asked about the final decision to use a programmatic series of national-level DPLs. 2.36 The World Bank had already used this instrument to support very similar objectives in Brazil five years earlier under the Env PRL. Under the Env PRL, the World Bank team made the argument that a programmatic series—accompanied by a technical assistance loan (the Env TAL)—was necessary to achieve the objectives of improving Brazil’s 19 The Clean Development Mechanism is provided for by Article 12 of the Kyoto Protocol, and is designed to assist developing countries in achieving sustainable development by allowing entities from Annex 1 Parties to participate in low-carbon projects and obtain Certified Emission Reductions (CERs) in return. (Carbon Finance, the World Bank, 2013) 16 environmental management system and mainstreaming environment into key sectors. However, the series was canceled after the first loan, and the Brazilian government continued to pursue many of the environmental reforms without the remaining two loans. 2.37 The Bank chose to repeat the use of the programmatic environmental DPL series without properly assessing the prior experience. The Bank did not produce the required full ICR of the Env PRL as required by World Bank policy 20 after the series was canceled. Thus there was no opportunity to come to an understanding of what worked and what did not, what the outcomes of the policy actions were, and why the series was canceled to inform the design of the SEM DPL. The World Bank team again came to the determination that a DPL programmatic series was necessary to achieve similar objectives with similar policy areas, and again the series was canceled after the first loan. 2.38 It could also be questioned whether a DPL in general was the best instrument for supporting the improvement of BNDES’s environmental and social management system. In the SEM DPL Program Document, the Bank team highlights the risk that “full implementation of the reform agenda within BNDES involves a challenging change in corporate culture regarding environmental and social sustainability” (Program Document, p. 66). For a large, complex institution like BNDES, significant time is needed to produce the changes to policies, institutions, and personnel capacity—not to mention culture—necessary for implementation of such a system. The Program Document also states that “BNDES’ management recognizes clearly that the reform steps to be carried out reduce BNDES’ reputational risks associated with adverse environmental and social impacts by improving screening and monitoring.” If BNDES management was already committed to improving its environmental and social system, the longer-term opportunity for engagement and support provided by an investment or technical assistance loan might have been more effective. 21 (The Program Document states that the risk of BNDES not carrying out the reform agenda would be “addressed through the substantial expansion of the Env TAL project to allow provision to BNDES of technical assistance and capacity building…”, but that never happened.) 2.39 Considering the shortcomings in the Results Framework and project logic, the appropriateness of the level of government targeted, and the choice of instrument, the relevance of design is rated Negligible. 3. Implementation 3.1 The Bank team blamed the delay in effectiveness of the operation on the Borrower and delays in its budget legislation. Since there were only three months planned originally between release of Tranche 1 and release of Tranche 2, the nearly fifteen-month delay in Tranche 1 should have provided more than adequate time for the Bank team to ensure that 20 See BP 8.60 – Development Policy Lending for the requirements on evaluation of DPOs (World Bank OPCS) 21 Another possibility might have been a single-operation DPL paired with a longer TA loan to BNDES. 17 Tranche 2 release conditions were also met, and the tranches could have been released at around the same time. Yet by the time of the Tranche 1 release, the Tranche 2 release conditions (see the column “Policy Actions—DPL 1: Second Tranche Release Conditions” in the Policy Matrix in Annex B) still had not been met, and required more than five months of additional time, resulting in a delay of more than seventeen months compared to the originally planned Tranche 2 release date—even greater than for Tranche 1. 3.2 Although the problems related to the Brazilian budget and loan approval process were eventually addressed, and both tranches of SEM DPL 1 were disbursed, the planned second loan in the series never materialized, and the SEM DPL series was eventually canceled. 3.3 The Bank team certified that the second tranche release conditions were met despite a lack of evidence presented in the ICR or made available to the IEG team and serious questions raised by civil society. A coalition of more than thirty national and international CSOs raised serious questions about the operation and about the evidence for disbursement of the two tranches. The CSOs sent letters to the Bank Regional Vice President and Country Director. They requested a dialogue with the Bank team on these issues, disclosure of the evidence, and a hold on disbursement of the second tranche. 3.4 Monitoring and evaluation were very limited. The monitoring system promised in the Program Document was never implemented. As reported in the ICR (p. 8), “due to unforeseen bureaucratic delays in contracting the Information Technology firm, the system was not updated during the course of the project.” Considering that the Environmental Technical Assistance Loan (Env TAL) became effective in April 2006, three years before the SEM DPL, and that SEM DPL effectiveness was delayed by over 14 months, this does not appear to be a valid justification for the lack of an M&E system. 3.5 There is no information in the ICR about M&E implementation with regard to BNDES. In an interview, BNDES confirmed to IEG that even today they have no environmental or social indicators with which to either monitor or evaluate the performance of their investments. 3.6 In the absence of the intended M&E system, the ICR states that “the Bank supervision team reviewed progress on actions and outcomes during the frequent meetings and supervision missions with the implementing agencies, and by directly tracking and reviewing publicly available monitoring systems (e.g., annual deforestation rate measured by Brazil Space Agency-INPE).” However, monitoring is primarily the responsibility of the client (in this case, MOF and MMA). Based on the information provided in the ICR, it is clear that the monitoring system promised in the Program Document was never updated. Since the SEM DPL team proposed using the system originally developed to “monitor the previous PRL and the ongoing ENV TAL”, and since BNDES was not directly involved in those operations, if the monitoring system was not updated for the SEM DPL, MOF and MMA would not have been able to regularly monitor progress at BNDES, which was implicated in so many of the actions. This is also evidenced in the progress reports provided to the World Bank for the release of Tranche 2, which were submitted separately by BNDES and MMA. 18 4. Achievement of the Objectives 4.1 Evaluating the Achievement of the Objectives involves assessing not only whether each of the key outcomes indicated in the operation’s statement of objectives has been achieved, but also whether the outcomes can be attributed to the actions supported by the operation. In the specific case of designing the results framework for a DPL, OPCS states: “Don’t include results that are not directly influenced by actions that are part of the operations or programmatic series of operations supported by the Bank.” (OPCS 2011) It is also important to bear in mind that for the evaluation of a DPL programmatic series, such as the SEM DPL, the operation is evaluated against the outcomes promised in the Program Document for the entire series, whether or not one or more of the loans in the series is canceled. 4.2 Brazil has made substantial progress in strengthening environmental management in a number of areas, including the impressive reduction in the rate of deforestation in the Amazon over the past decade. As an example, a definitive new article on the Amazon finds: “Deforestation—the clear-cutting of mature forest—declined from a 10-year average of 19,500 km2 per year through 2005 to 5,843 km2 in 2013, a 70% reduction” (Nepstad and others, 2014) But there is little if any evidence that the SEM DPL contributed to any of this. As noted in the previous section, there are flaws in the results framework and project logic that lead to a disjointed results chain and make it unclear how the series of actions specified—even if they were carried out—would result in achievement of the intended outcomes. The general nature of the objectives (especially Objective (a)) contributes to the difficulty in assessing its achievement. The ICR contains no presentation of specific information or analysis on whether or how any or all of the reforms that the SEM DPL was supposed to support contributed to the achievement of the overall series objectives. 4.3 An assessment of the achievement of each of the two objectives is provided below, with specifics for each of the policy areas under the objectives. A summary of the outcome indicators that were supposed to be achieved by the program by 2011, and the achievements as reported in the World Bank’s Implementation Completion Report (ICR) is provided in Annex C for convenience. Objective A: Improve the overall Brazilian environmental management system. 4.4 With regard to improving the effectiveness of government agencies in implementing mandated Brazilian environmental and social management procedures, as in the Monitoring and Evaluation section below, the original indicator on the number of judicial challenges to environmental licenses was not being met. The indicator was changed to “number of environmental licenses issued at the Federal level”, which is not a good measure of the intended outcomes. Moreover, these indicators have no direct relationship to the prior actions under sub-objective A.i related to the Ministry of Environment or the National Biodiversity Management Institute. 19 4.5 The actions under this sub-objective related to restructuring and strengthening MMA and its affiliates and IBAMA are virtually identical to actions under the 2004 Env PRL, which were reported as achievements of that loan in its ICR. An analysis of the Brazilian environmental licensing system carried out by the World Bank several years after the closing of the Env PRL found that despite the actions supported by that loan, “environmental agencies have to date failed to significantly bolster their institutional capacity” (World Bank 2008). With regard to the actions on increasing staff numbers at the agencies, that is only one of many inputs required, and would by no means necessarily lead—on its own—to the desired outcome of “improved effectiveness of government agencies.” Indeed, the 2008 World Bank analysis of the environmental licensing system found that simply increasing staffing numbers was not sufficient to improve the environmental licensing process, and identified a host of other areas where improvements were needed, including: • a major training operation for IBAMA staff; • changes in the staff mix, and particularly an increase in the number of professional staff with social science expertise; • the need for a law to clarify the responsibilities of the Federal and State Governments with regard to environmental licensing; • the establishment of a dispute resolution mechanism; • addressing the risk averse behavior of IBAMA employees caused by a law that renders them personally and criminally liable for licensing decisions. 4.6 Although the SEM DPL was prepared soon after the completion of the analysis of problems with the environmental licensing system, none of the issues above were addressed by its policy actions. Instead, it again included an increase in staffing numbers. While there is some evidence that IBAMA staffing numbers increased again around the time of the SEM DPL, this is only an input to the environmental licensing process, and was unlikely on its own to have much of an impact, for the reasons listed above. Interviews with IBAMA and with representatives of its staff (ASIBAMA) revealed an institution that continues to face major challenges. Although the number of staff has increased, because of high stress resulting from political pressure to approve licenses, and low salaries, there is a very high rate of staff turnover. One symptom of this is that only three staff members remain from the time when IBAMA was restructured in 2002. There are also reportedly 900 IBAMA staff set to retire shortly. Because of the high turnover, ASIBAMA reports that the licensing team has limited experience, and this has an adverse impact on the licensing process. 4.7 IBAMA and ASIBAMA representatives shared with IEG their belief that the revised indicator used by the SEM DPL team—the number of licenses issued—was a poor one. The staff representative stated that “The speed of licensing should not be the main objective. The objective should be to improve environmental outcomes.” To this end, the representative stated that ASIBAMA had suggested alternative indicators, such as post-licensing monitoring of environmental and social impacts, but that they had not been adopted. Meanwhile, IBAMA monitors the time it takes to issue environmental licenses, but does not monitor environmental and social impacts, which makes assessing the impacts of reforms difficult. Moreover, as a result of a restructuring in 2007, IBAMA has closed more than half of its regional offices, significantly reducing its already limited ability to monitor events on the ground. The idea was to decentralize this responsibility to local governments, which in 20 theory could have benefits in terms of local ownership, but the local governments do not have the capacity, and the SEM DPL contains no provisions for developing capacity at that level. 22 This has created potential gaps in Brazil’s environmental licensing system. 4.8 In 2013 IBAMA issued Terms of Reference for consultancies paid for by another World Bank-financed project—The National Environment Program 2—to improve the environmental licensing system and capacity of its staff (IBAMA 2013a; IBAMA 2013b). The Terms of Reference cite an array of problems that continue to adversely affect the environmental licensing system in Brazil several years after the SEM DPL, and which cannot be addressed by hiring additional staff. They include: “the need of greater transparence”; “a deficit in the management capacity of methods—standards, concepts and procedures—that makes the monitoring and control of results difficult”; “blanks, overlapping and ambiguities in the definition of concepts, standards and procedures related to the FEL, EIA, EC processes”; “low capacity to meet the demands and the lack of integration of standards and compatibility of procedures among the partner agencies”; and “deficiencies in the capacitation of environmental analysts”. The Terms of Reference warn that these problems “can compromise the licensing quality” and were resulting in “the amplification of the environmental conflicts.” 4.9 Restructurings of IBAMA that resulted in the creation of the independent ICMBio, another policy action, and the Brazilian Forest Service have also led to coordination problems among the various agencies, which used to all be under IBAMA. Moreover, through review of official documents and interviews with Brazilian government officials, IEG confirmed that the creation of the Brazilian Forest Service and ICMBio was completed well before the SEM DPL (March 2006 and August 2007, respectively), and had no relationship with it. 23 4.10 With regard to mainstreaming climate change in public and private sector investments, the first actions for the two tranche releases of SEM DPL 1 required the drafting and approval of a National Climate Change Action Plan (or “NCCAP”), which the Program Document (p. 20) promised would “present a balanced set of command-and-control and economic instruments covering both mitigation and adaptations” and that it would “address the need of cross linking actions carried out by different sector agencies and key actors in the economy” (World Bank 2009). In the ICR, the World Bank team reports that “the revised NCCAP was approved by Law 12.187/2009 to institute a National Policy for Climate Change” (World Bank 2011). . In fact, according to MMA’s website, what was established 22 In the section of the Program Document on Relationships to Other Bank Operations (p. 38), there is one line on the Brazil National Environmental Project (NEP) III, which it states was still under preparation, and which would support environmental capacity in states and municipalities. Presumably, it meant the second phase of the three-phase NEP II, which was approved in September 2009. That USD 24.3 million loan, which could have been important for supporting decentralization efforts, has only disbursed USD 1.5 million so far, and is rated by the Bank supervision team as “unsatisfactory” in the latest Implementation Supervision Report. 23 See for example : http://www.florestal.gov.br/menu-horizontal-de-internet/institucional/servico- florestal-brasileiro ; and: http://www.icmbio.gov.br/portal/quem-somos/o-instituto.html 21 by Law No. 12.187/2009 was a National Policy on Climate Change, which was to be implemented in part via a National Plan on Climate Change (see Box 1). 24 4.11 The National Policy on Climate Change is an important first step in establishing climate change as a priority in Brazil and laying out in broad strokes what Brazil would like to achieve, and how it intends to achieve it. But it is primarily a framework that establishes the instruments that will be used to achieve the policy aims, including the National Plan on Climate Change and the National Fund for Climate Change. It did include an overall, non- binding emissions reduction target, previously announced by Brazil in the run-up to the Copenhagen Conference of the Parties (COP) to the UNFCCC in December 2009: “the country shall adopt actions to mitigate greenhouse gas emissions with the purpose of reducing between 36.1% and 38.9% of projected emissions by 2020 as a national voluntary commitment.” This is a potentially significant—though non-binding—commitment that has been lauded by some, but criticized by others because rather than use an actual baseline, Brazil set its emissions reduction target against projected future emissions under a business- as-usual scenario, and did not make public its calculations (FAO 2011, WRI 2009). But it leaves it to the other instruments and organizations cited to develop specific actions. 4.12 With regard to “tools for implementation”, the National Plan on Climate Change lists a number of existing funds for financing investments relevant to Brazil’s climate change efforts, including some by BNDES. But in terms of developing new economic instruments, it states: “This item, which specifically addresses economic instruments…should be a detail in the second phase of the plan….” The National Plan also stated that actions plans for implementation were supposed to be sectoral, and that many plans would be prepared, one for each sector. Without the sectoral and action plans, both National Policy and Plan consist mainly of intentions, a few targets and broad guidelines without an implementation mechanism. The sectoral plans were supposed to be ready and the National Climate Change Plan revised before the end of 2011, but those deadlines were not met.. . At the time of this evaluation, the revised National Plan had still not been completed. The government also reports that a series of sector plans and Nationally Appropriate Mitigation Actions (NAMAs) have been produced, though evidence of their existence, status, added value, and relation to the SEM DPL is limited (see Box 1). There is no National Adaptation Plan. 4.13 Comparing what has been produced so far for Brazil reveals important contrasts with climate change action plans for other countries such as Turkey and the United Kingdom, 24 Source : http://www.mma.gov.br/clima/politica-nacional-sobre-mudanca-do-clima/plano-nacional- sobre-mudanca-do-clima An English version of the Law establishing the National Policy on Climate Change can be found at: http://www.preventionweb.net/files/12488_BrazilNationalpolicyEN.pdf National Climate Change Plan: http://www.mma.gov.br/clima/politica-nacional-sobre-mudanca-do- clima/plano-nacional-sobre-mudanca-do-clima 22 which include many more specifics on mitigation and adaptation objectives, actions, and targets for individual sectors. 25 25 Turkey: http://www.preventionweb.net/files/29675_turkeynationalclimatechangeactionpl.pdf ; UK: http://www.legislation.gov.uk/ukpga/2008/27/contents ; https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/209866/pb13942-nap- 20130701.pdf 23 Box 1: Brazil’s National Policy on Climate Change The Brazilian National Plan on Climate Change, released in December 2008, was considered a significant achievement in the international climate change agenda. According to the Brazilian government, the Brazilian Forum on Climate Change (FBMC) played a central role in the process. However, criticism has since been leveled against the Plan regarding its objectiveness, deadlines and strategies (ICTSD 2008). “On 29 December, 2009, the Brazilian government took a historic step by establishing the National Policy on Climate Change by Federal Law No. 12,187. Under this law, Brazil shall adopt measures to reduce between 36.1 percent and 38.9 percent of its potential emissions by 2020 (i.e. a reduction of 17 percent compared to 2005 levels). This goal includes a reduction target of 80 percent in Amazon deforestation, as defined by the National Plan on Climate Change. Two policy instruments are established under this Plan: the National Plan on Climate Change and the National Fund on Climate Change, established by Law No. 12,114. On December 9, 2010, the Government of Brazil approved Decree 7.390 to regulate the National Policy on Climate Change by providing for the incorporation of sectoral plans for climate change mitigation and adaptation to the National Plan on Climate that were previously established by Law 12.187, of December 29, 2009. The decree stated that the Sector Plans would be ready by 15 December 2011 and that the National Climate Change Plan would be revised by that year as well. It was also expected that the release of the Action Plans for prevention and control of deforestation in key biomes would take place in 2011. On December, 15, 2011 Decree 7643 was approved and the deadline for revision and for releasing all of the sectoral action plans was postponed to April 16, 2012. However, those deadlines were not met. Finally, in April 2013 preparation of the National Plan on Climate Change revision was initiated. Based on perceptions that the federal government had not provided any evidence of the implementation status of the 2008 National Plan on Climate Change, and that the revised National Plan would not represent a significant improvement, a large network of NGOs withdrew from the process, casting doubt on its future as well as its impacts to date.26 At the time of this evaluation, the revised National Plan had still not been completed, though the Brazilian government reports that it expects it to be approved in 2014. The government also reports that four sector plans were approved in June 2013 for industry, mining, transportation, and public health, and that in addition Brazil has since 2010 implemented Nationally Appropriate Mitigation Actions (NAMAs)27 to: reduce deforestation in the Amazon and Cerrado; promote low-carbon agriculture; support renewable energy and energy efficiency; and use charcoal more efficiently in the steel industry. The NAMAs were presented in the context of Brazil’s Copenhagen emission reduction 26 See: http://www.observatoriodoredd.org.br/site/index.php?option=com_content&view=article&id=1204:or ganizacoes-da-sociedade-civil-se-retiram-do-processo-de-revisao-do-plano-nacional-sobre-mudanca- do-clima ; and: http://www.axa.org.br/?p=1335 (last visited September 10, 2014) 27 NAMAs are prepared in the framework of the UNFCCC, which defines them as “any action that reduces emissions in developing countries and is prepared under the umbrella of a national governmental initiative.” See: http://unfccc.int/focus/mitigation/items/7172.php 24 commitment. 28 Two-thirds of the emission reductions are intended to come from reduced deforestation in the Amazon and Cerrado, through Action Plans for the Prevention and Control of Deforestation in the Amazon and Cerrado, respectively. 29 While the Amazon Plan—combined with other efforts—has been an important success, it was prepared in 2003, years before the National Plan on Climate Change, the Copenhagen COP, or the SEM DPL (see below for a more detailed discussion of the Amazon Plan, known as “PPCDAM”). The Cerrado Plan was launched in 2010, but the planned Cerrado Law has been delayed and there is no evidence yet of reduced deforestation in the Cerrado. The energy NAMA appears to consist of pre-existing government strategies related to the secto. At the time of this evaluation, IEG examined the Brazilian Ministry of Environment’s official webpage on the sectoral plans. 30 Although it lists eight sectoral plans, links to only four NAMAs are provided there: on the Amazon, Cerrado, agriculture, and energy. The other plans were unavailable at the time of this evaluation, and so IEG was unable to evaluate them. 31 And although the NAMAs are grounded in Brazilian policy and legislation, they do not appear to have been officially submitted to or accepted by the UNFCCC, as Brazil’s official UNFCCC NAMA Registry Country Page was empty at the time of this evaluation.32 The sectoral plans have also been criticized by Brazilian NGOs for not meeting minimum standards as they lack objectives, indicators, M&E frameworks, adequate policies, governance structures, and funding. IEG was not able to independently verify these claims, due to the fact that the documents were not publically available. With regard to climate change adaptation, there is currently no dialogue, and no National Adaptation Plan. Additional sources include: http://www.ipam.org.br/saiba-mais/abc/mudancaspergunta/O-que-sao-Servicos-Ambientais-possivel- compensar-economicamente-a-prestacao-destes-servicos-/41/21# ; http://ictsd.org/i/news/pontesquinzenal/33092/ 4.14 As part of GOB’s contribution to the preparation of the ICR, MMA’s report states simply that a preliminary low-carbon action plan was prepared for one sector—agriculture— but that it was not being implemented. It also refers to Brazil’s Second National Communication to the UNFCCC. It does not claim that a National Climate Change Action Plan was prepared or approved. This information from the MMA report was not reflected in the World Bank’s ICR. 4.15 The other actions for the “mainstream climate change” policy area, under SEM DPL 2, related to BNDES developing clean development and carbon funds programs in 28 See: http://mitigationpartnership.net/brazil-voluntary-namas-achieve-ghg-emission-reduction- between-361-and-389-2020-0 (last visited August 29, 2014) 29 See: http://www.mmechanisms.org/document/NAMA/NAMA_LCA15_brazil_EN.pdf (last visited August 29, 2014) 30 See: http://www.mma.gov.br/clima/politica-nacional-sobre-mudanca-do-clima/planos-setoriais-de- mitigacao-e-adaptacao (last visited August 29, 2014) 31 For the four sectoral plans that are not available, the MMA website includes a note for each stating: “presentation unavailable from 07.05.2014 to 10.26.2014, the election period….”. 32 See: http://www4.unfccc.int/sites/nama/SitePages/Country.aspx?CountryId=24 (last visited August 29, 2014) 25 accordance with the NCCAP and its own new environmental and social policy. The outcome indicator related to the promotion of projects under the CDM as well as BNDES’s programs and the NCCAP. As noted in the previous paragraphs, sectoral action plans with targets have not yet been developed. 4.16 The indicator selected for this policy area related to the CDM—“planned signed reductions”—has no meaning in the CDM lexicon or project cycle, which has very specific steps and requirements for each step. 33 So IEG was required to interpret the intentions behind this indicator in consultation with carbon finance experts (see footnote to Figure 1). Moreover, projects proposed for financing under the CDM were being prepared in Brazil since 2003, six years before the SEM DPL, and also well before any of the Brazilian national climate change plans was approved. Since then, Brazil has had 734 investments proposed under the CDM, the majority of them before 2009. 34 The ICR reports that 19.8 million tons of CO2 equivalent per year of planned emission reductions were signed “from CDM projects and selected BNDES programs”, and claims that “99% of the target of 20 million tons of CO2” was achieved. However, the target was for an increase in emission reductions of 20 million tons of CO2 (not a total figure of 20 million tons CO2) or an approximate doubling of annual emission reductions to 40 million tons CO2 compared to a baseline of 19.5 million tons from CDM projects as reported in the Program Document (footnote, p. 21). The information provided in the ICR on this policy area is limited. IEG conducted its own analysis and its results are presented in Figure 1. Figure 1: Planned annual emission reductions in the Brazil CDM portfolio Note: data as of August 2013. Source: UNEP RISOE : http://cdmpipeline.org 33 See UNFCCC: http://cdm.unfccc.int/Projects/pac/index.html 34 Based on official data from UNEP RISOE. See: http://cdmpipeline.org/ ; since, as noted above, “planned signed reductions” has no meaning for CDM, IEG used the “First Start Comment” date as the date when CDM projects were first proposed. 26 4.17 As the figure illustrates, CDM project preparation began in Brazil in 2003, and planned annual emission reductions from those projects peaked at about 15 million tons CO2 per year in 2005, before starting a steady decline to just over 3 million tons of CO2 in 2010, the last year of the SEM DPL, and the last full year that could have been included in ICR. The claim of either an increase or a total (depending on how one interprets the report from BNDES) in CDM CO2 reductions of 14 million tons of CO2 is overstated. 4.18 The figure shows that there was a sharp increase in planned emission reductions from proposed CDM projects in Brazil in 2011 and 2012. But as elsewhere in the world, “this did not reflect the demand side but was driven by the upcoming end of Phase II of the EU ETS and the start of Phase III, which sees the introduction of additional restrictions on the use of international credits” at the end of 2012 (Carbon Finance, the World Bank, 2013). Since then, planned emission reductions under the CDM have plummeted in Brazil, continuing their long-term downward trend, to 1.5 million tons CO2 as of August 1, 2013. In general, the evolution of the CDM portfolio in Brazil mirrors that of the CDM globally. Due to a lack of demand, the price of CERs crashed to 0.34 euros in December 2012, and shows little sign of recovering. The low prices do not even cover the costs of verification and issuance of CERs for CDM projects, “thereby questioning the role of the CDM as a catalyst for private sector investment in climate change mitigation” (Carbon Finance, the World Bank, 2013). Therefore, the SEM DPL does not appear to have had any positive effect on emission reductions under the CDM, and the significance of the CDM in general for “mainstreaming” of climate change in Brazil going forward is doubtful. 4.19 It should also be noted that a “planned signed” emission reduction does not necessarily lead to an actual reduction—which should be the real objective—and indeed in the majority of cases it does not. Of the 734 Brazil projects proposed under the CDM since its inception, only 135, or 20 percent, generated any CERs. Of the Brazil CDM projects proposed since 2008, by 2013 only two had so generated any CERs—for a total of 0.09 million tons CO2—and none since 2010. 35 4.20 In the report submitted by BNDES to the Bank in June 2011 as ICR input, BNDES noted that through its investments in wind power, mini hydropower plants, and biomass cogeneration, it expected to reduce an additional 5.7 million tons of CO2 equivalent per year. This is a notable increase that BNDES should be credited for. Nevertheless, even if combined with the reductions from CDM, it still represents only about a quarter of the target. 4.21 Brazil’s National Fund on Climate Change (Climate Fund), created under the NPCC, actually includes two funds: one reimbursable managed by BNDES, and a grant fund managed by the Ministry of Environment. 36 For the reimbursable fund, BNDES announced in 2012 a special line of credit with preferential interest rates “for projects that reduce 35 It is possible that CDM projects in Brazil generated CERs after 2010, but they would have been proposed before 2008. 36 http://www.mma.gov.br/apoio-a-projetos/fundo-nacional-sobre-mudanca-do-clima 27 emissions”, including for renewable energy. 37 This is a positive development that could help to further the goals of mainstreaming climate change in investments in Brazil, if designed and implemented well. But in an interview, a knowledgeable Brazilian government official stated that BNDES was still waiting for the first investments to be made by the fund, hoping that the first 2-3 projects would be initiated in 2013. BNDES attributes the lack of investments under the Climate Fund to the loss of attractiveness of the interest rates in the face of more competitive rates from other sources following the global financial crisis. However, BNDES reports that the Climate Fund is now on a better track after interest rates were reduced in 2013, with operations starting to be contracted in 2014. One area where BNDES does not appear to have increased its support is for energy efficiency, which has been shown to have some of the highest rates of return for emission reductions (IEG 2010). With regard to the grant fund managed by the Ministry of Environment, it was supposed to be financed by royalties from Petrobras, but according to recent reports, new royalty rules have eliminated that source of funding. 4.22 Ultimately, the essential indicator of progress on climate change mitigation is Brazil’s overall level of greenhouse gas emissions. The official greenhouse gas inventory recently released by the government provides a detailed stock-taking of this information (Ministry of Science, Technology and Innovation, Brazil, 2013). The main message is that Brazil has made dramatic progress, reducing its national greenhouse gas emissions by nearly 39 percent from 2005-2010 (Nature 2013). Yet it also underscores that significant challenges lie ahead. All of Brazil’s reduction in greenhouse gas emissions is due to the sharp decrease in deforestation in the country, particularly in the Amazon. This is impressive, but as discussed in detail below, appears to have started well before the SEM DPL and to have had little or nothing to do with that World Bank loan. Moreover, emission reductions from avoided deforestation are not eligible under the CDM that the SEM DPL targeted. 4.23 Other than avoided deforestation, the greenhouse gas inventory found that emissions from all of Brazil’s other main sources increased from 2005-2010. Agriculture is now Brazil’s biggest emitter, with greenhouse gases from the sector increasing 5.2% over the period. The fastest rate of growth of greenhouse gas emissions in Brazil is found in the energy sector. It is in these sectors—especially energy but also agriculture—where the CDM often supports investments to reduce emissions around the world. But in Brazil, emissions from the two sectors contributed 437 million and 399 million tons of CO2, respectively, over the period, dwarfing both the targeted and actual emission reductions under the SEM DPL. In 2010 the government launched a fund called “the Low-Carbon Agriculture (ABC) Program” 38, an ambitious plan to address the challenge of emissions from agriculture. But due to a lack of demand, the program “didn’t manage to lend a single penny of its initial US$1-billion endowment” in 2010-11, its first year (Nature 2012). The Nature article notes that disbursements did pick up considerably in 2012, but only after the ABC plan had been 37 http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/Press/Noticias/2012/20120213_f undoclima.html 38 This was apparently supposed to be a pilot of a sectoral plan under the National Plan on Climate Change discussed above. 28 “stripped of some of its environmental character”, and there is no way to know whether the plan is having an impact because “the infrastructure to take the necessary measurements is not in place.” 4.24 The World Bank argues that the support it provided under the SEM DPL was important for getting Brazil to re-engage in a meaningful way in international climate discussions, such as under the UNFCCC process, for example at the 2008 COP-14 meetings in Posnan. Brazil did commit to cutting emissions by reducing deforestation by 70 percent by 2020, and its engagement in these processes is important. But as already noted, the cuts in deforestation began much earlier, in response to actions supported by the Env PRL. Other significant outcomes are not apparent. Thus, in general, the evidence points to a lack of impact of the measures supported by the SEM DPL in mainstreaming climate change mitigation in Brazil. 4.25 With regard to improving the effectiveness of environmental and social management systems in BNDES and other financial institutions, this policy area was significant for the SEM DPL because it represented a potentially new area for engagement under the operation, compared to the other policy areas which were based on engagement that was initiated under the 2004 Env PRL and other Bank activities. As the World Bank team stated in the ICR for the operation, “the results obtained on the environmental management front are particularly relevant for Brazil.” Nevertheless, the ICR includes only one sentence on this policy area in the Achievement of Program Development Objectives section (page 11): “Meanwhile, the BNDES screening of its proposals for compliance with its Environmental and Social Institutional Policy reached 100%, from 35% in 2007.” In response, IEG carried out a detailed assessment of the outcomes associated with this policy area under the operation has been carried out for this evaluation. 4.26 In order for the SEM DPL to have been effective in supporting the achievement of sub-sub-objective A.iii, and more importantly contributing in a significant way to the sub- objective of “improving the overall Brazilian environmental management system,” several conditions would have to be met: • The new environmental and social (environmental and social) policies adopted should be well-designed. • Improvements to environmental and social systems should go beyond BNDES to encompass other major financers of environmentally and socially investments. • The environmental and social policies need to be properly used and monitored by the financial institutions to manage their investment portfolios. • Implementation of the policies should result in improved environmental and social outcomes. 4.27 To begin with, the only action under this policy area related to “other financial institutions” is the first one, “Revised Green Protocol approved and signed by GOB and all federal public Brazilian Banks, including BNDES”. But neither the Program Document, nor the ICR, nor the BNDES or MMA inputs to the ICR provide adequate information on the signing, implementation, or results of the Green Protocol (GP). The “Green Protocol” (Protocolo Verde in Portuguese) is a colloquial term that was first applied to a “Letter of 29 Intentions” titled “General Principles for Sustainable Development” signed by BNDES and other Federal banks in 1995. 39 The 1995 letter contained a set of voluntary principles that were found lacking, so it was eventually decided to try and improve upon them (Widmer 2012). The result was in fact two letters of intention: one for Federal public banks, 40 and another for private banks. The main focus of the SEM DPL was the revised “Green Protocol” officially called the “Protocol of Intentions on Social and Environmental Responsibility”, and signed by the Minister of Environment, BNDES, and several other Federal banks on August 1, 2008. 41 The Program Document (p. 21) states that the provisions of the GP were supposed to inform the development of BNDES’s new environmental and social Institutional Policy. 4.28 Very little information is available on the current status or past implementation of the “revised Green Protocol”. For this policy area more generally, there is no information in the ICR on the outcomes of the Green Protocol, or on any banks other than BNDES, and all of the other actions and the outcome indicator for this policy area are addressed exclusively to BNDES. 42 One of the other signatories of the Green Protocol, the state-owned Banco do Brasil, has a webpage devoted to the Green Protocol. But the last entry there refers to a workshop organized in June 2010 by FEBRABAN—Brazil’s official banking federation, which was tasked with overseeing the Green Protocol—in order to work out implementation modalities. 43 None of the other signatory banks provided any information. So IEG enquired about the status of the Green Protocol with officials of FEBRABAN, the Brazilian banking federation, and were told that implementation of the 2008 revision of the Green Protocol that the SEM DPL was supposed to have supported is “on standby”, and moreover that “the Protocol has lost its importance as an instrument of voluntary and guiding actions of banks.” 4.29 Apart from the status of the Green Protocol, IEG conducted an assessment of the substance of the 2008 revised Green Protocol, along with BNDES’s new environmental and social policies, by comparing the two sets of policies with an international benchmark—the 39 http://www.mma.gov.br/estruturas/182/_arquivos/protocolo_verde_carta_de_intenes_1995.pdf 40 http://www.bndes.gov.br/SiteBNDES/export/sites/default/bndes_pt/Galerias/Arquivos/empresa/dow nload/ProtocoloVerde.pdf 41 The two versions are in fact quite similar. 42 See: http://www.bb.com.br/portalbb/page251,8305,3926,0,0,1,6.bb?codigoNoticia=28467 (last visited August 29, 2014). 43 As of April 2014, a new regulation has been issued by the Central Bank of Brazil, Resolution 4.327/2014. The resolution is a broad requirement for all financial institutions to develop a Policy of Social and Environmental Responsibility (PRSA) to manage environmental and social risks. In terms of its environmental and social content, the Resolution is less specific than and makes no mention of the Green Protocol and leaves the details of the PRSA up to the individual financial institutions. It requires that financial institutions approve PRSAs and action plans for implementing them by February or July 2015 (depending on the status of the institution). The diffusion of such a requirement to all Brazilian banks has the potential to broaden the use of environmental and social safeguards. But this action was not supported by the SEM DPL and was not mentioned in the World Bank’s ICR, and it is too early to assess what its impacts might be. See: http://www.bcb.gov.br/pre/normativos/busca/normativo.asp?tipo=Res&ano=2014&numero=4327 30 Equator Principles. 44 According to a recent assessment by the Harvard Law School, “in the last 10 years, the Equator Principles or EPs have emerged as the industry standard for financial institutions to assess social and environmental risk in the project finance market,” and have been adopted by “79 financial institutions in 32 countries …, reportedly covering over 70% of international project finance debt in emerging markets.” 45 Moreover, as discussed further below, BNDES had the opportunity to adopt the EPs through collaboration with IFC at the time that the SEM DPL was prepared, but instead chose to develop its own approach. Meanwhile, other major public and private Brazilian banks—including large state- owned banks like Caixa Econômica Federal and Banco do Brasil—have already signed on to the EPs. 4.30 The detailed comparison of the three approaches is provided in Annex E to this PPAR. But a summary of some of the main differences is as follows: • One general difference is immediately apparent upon examination of the three approaches. The Green Protocol and BNDES’s environmental and social policies are much less specific and detailed than the EPs. The elements of the GP and BNDES’s environmental and social policies are more like general principles, and therefore would require additional work to develop operational procedures to make them implementable. In the case of BNDES, this was apparently to be primarily through the promised sector-specific guidelines. • The EPs put a common global “floor” under environmental and social standards by establishing that if a country’s national environmental and social legislation and capacity are insufficient, then the financial institution will adopt the IFC’s Performance Standards. In contrast, BNDES allows the use of country systems even in countries with weaker environmental standards than Brazil. • The EPs include specific requirements on transparency; the GP’s and BNDES’s policies do not. • The EPs provide for a grievance mechanism; the GPs and BNDES’s policies do not. • The EPs include specific requirements on stakeholder consultation; the GPs and BNDES’s policies do not. 4.31 These differences are significant, and would already vitiate the effectiveness of the GPs and BNDES’s environmental and social policies even if they were fully implemented. 4.32 The BNDES aspect of this objective was one of the original objectives of the operation at the time when it was proposed as a DPL and then FIL to BNDES, and was an area where the Bank had substantial direct engagement with the client through the SEM 44 According to its website: “The Equator Principles (EPs) is a risk management framework, adopted by financial institutions, for determining, assessing and managing environmental and social risk in projects and is primarily intended to provide a minimum standard for due diligence to support responsible risk decision-making.” For more information, see Annex C to this evaluation or the Equator Principles website: http://www.equator-principles.com/index.php/about-ep/about-ep 45 https://blogs.law.harvard.edu/corpgov/2013/06/18/equator-principles-iii-enters-into-force-this-june/ 31 DPL. The Program Document Annex 8-Background on BNDES has a section on the “Current Status of BNDES Environmental and Social Screening”, including “Box A: Areas for potential improvement of the BNDES screening process for operations”. This discussion appears, for the most part, to be an accurate reflection of the significant shortfalls in BNDES’s environmental and social standards and procedures. 4.33 One point on which the Program Document states that BNDES should improve is by establishing “a mechanism for applying BNDES environmental and social policies to the projects financed by its financial intermediaries.” This would be important for achieving the BNDES portion of this sub-objective because more than half of BNDES’s lending is in fact through financial intermediaries, or “indirect”. And both the SEM DPL II trigger for this policy area and the Program Document text (p. 21, p. 57) indicate that BNDES’s new environmental and social policies would be applied to the full portfolio. But the SEM DPL outcome indicator—with reported 100 percent achievement—is instead only for “projects submitted directly to BNDES.” 46 Further, BNDES acknowledges that today it still does not apply its environmental and social policies to even directly financed investments in what it calls the “exports” sector. This sector was estimated at over US$ 5 billion in 2012 and includes BNDES’s large and growing investments outside Brazil—much of it for major infrastructure investments with serious environmental and social implications—in facilities like hydropower dams, mines, highways, and airports, including in Amazon regions of Bolivia, Colombia, Ecuador, and Peru (see for example the Wall Street Journal 2011). Overall then, even if the SEM DPL achieved its target indicator, it would have affected less than half of the investments of only one of Brazil’s public banks, diminishing from the beginning its ability to achieve the program’s objectives. This restricted scope was not made clear in the project documents. 4.34 There has been little if any discernible improvement in BNDES’s environmental and social (environmental and social) system. Based on IEG’s review of the evidence provided by the World Bank team and BNDES as well as third party sources, and interviews of key stakeholders, particularly the staff of BNDES, there has been little discernible improvement to BNDES’s environmental and social management system compared to the one described—and critiqued—by the World Bank team during preparation. For those BNDES investments that should have fallen under the provisions of the SEM DPL, the key action cited under both the DPL-I Tranche Release Conditions and the DPL-II Triggers was the approval of a new BNDES “Environmental and Social Institutional Policy”, and the application of the new policy to the screening, approval and monitoring of BNDES projects. In the Tranche Release Document (para. 18) certifying “full compliance” with the second tranche release conditions, the evidence provided by the Bank for compliance of this policy area is based on the approval by BNDES of three resolutions: Resolutions 2023/10, which 46 BNDES divides its portfolio into what it calls “direct operations”, “non-automatic indirect operations”, and “automatic indirect operations”. BNDES’s Socio-Environmental Policy states that it applies only to the direct and non-automatic indirect operations. Including the non-automatic indirect operations is a step in the right direction. However, for the non-automatic indirect operations, the policy states that the financial intermediaries “are responsible for verifying the social and environmental compliances of the borrower and the target project.” In 2011, the non-automatic indirect operations constituted 51 percent of BNDES’s disbursements (BNDES 2011). 32 the Tranche Release Document describes as the “Social and Environmental Responsibility and Governance (SERG) for the entire BNDES system”; Resolution 2025/10, which is described as “a new Environmental and Social Policy (ESP) governing BNDES operations” in the framework of the SERG; and Resolution 2022/10, described as “a supplementing resolution on the Structure and Use of Environmental and Social Guidelines. None of the actual resolutions is publicly available on BNDES’s website, although content consistent with the resolutions appears on the site where BNDES presents its environmental and social policies. 4.35 IEG has reviewed official copies of all three resolutions. In a slight deviation from what is stated in the Tranche Release Document, Resolution 2023 is actually titled “Social and Environmental Responsibility Policy of the BNDES System” (i.e., there is no “Governance”). It could best be described as a general statement of corporate values with regard to social and environmental issues, such as “To promote an integrated development that includes the economic, social and environmental dimensions”, and “Proactive attitude in line with the Brazilian public policies and rules while in observance of international norms of behavior.” It contains no information on how these policies are to be implemented or which parts of the BNDES portfolio they apply to. 4.36 Resolution 2025 states that it is intended “to approve the environmental and social policy of BNDES system.” Although the title of the resolution sounds similar to that of 2023, the Tranche Release Document explains that 2025 “presents an updated screening process regarding the social and environmental dimensions and guidance on operational supervision,” and also establishes requirements for supervision of operations to track environmental and social indicators and verify compliance with the environmental license. But the contents of Resolution 2025 are, like Resolution 2023, quite vague. It begins with a set of “Guidelines” that do not appear to commit the institution to anything more than adhering to Brazilian law, such as “Acting in line with government policies and legislations, in particular with the provisions of the National Environmental Policy.” Similarly, under the heading “socio-environmental analysis of projects”, it states that “For financial support, are observed: applicable laws, sector-specific norms,…” and so on. In terms of responses, it goes on to explain that BNDES “may” take various actions. There is nothing that resembles a specific, binding environmental or social safeguard. 4.37 Another important aspect of BNDES’s new environmental and social approach was supposed to be the introduction of “sectoral guidelines” governing the application of environmental and social good practices tailored to key sectors. Since neither Resolution 2023 nor Resolution 2025 establishes environmental and social safeguards or procedures for implementing them, it is presumed that those (the “sector-specific norms” referred to in Resolution 2025) would be established via the sectoral guidelines. The SEM DPL Program Document (Annex 8) lists 13 specific sectors for which it says BNDES will develop guidelines as part of the SEM DPL series (p. 104), including for hydroelectric plants. Three sets of guidelines—livestock, ethanol and sugar, and soy—plus a fourth for a forest management and plantation program (“REFLORESTA”) were supposed to be developed for SEM DPL 1. 33 4.38 On BNDES’s website today there are only three sets of sectoral guidelines presented: livestock, ethanol and sugar, and thermoelectric. That means that after five years, BNDES has still only developed three out of the thirteen sectoral guidelines that were promised to be developed under the SEM DPL series by 2011. And absent are guidelines for sectors that are critical from an environmental and social perspective, such as hydroelectric plants, forest management, soya, and water supply and sanitation. For the three sectoral guidelines that BNDES has posted, each has a different structure and approach, and varying degrees of detail. Although the specifics vary for each sector, in general all three simply require projects financed by BNDES to comply with relevant Brazilian legislation (for example, having appropriate licenses and complying with land use and anti-slavery laws). The one area where the sectoral guidelines appear to go beyond these legal requirements is for livestock, where the guidelines state that clients are required to “join a traceability system for the beef production chain, from birth to slaughter, to check the regularity of the environmental supply chain.” It made sense for BNDES to make the livestock sector a priority because, as mentioned in the Introduction, BNDES and others were the subjects of an international media campaign launched by Greenpeace aimed at raising awareness of the links between the sector and Amazon deforestation. 47 But overall, BNDES has made no progress on development of sectoral guidelines—one of the pillars of its new environmental and social policy—since SEM DPL 1 closed in 2010. 4.39 There is no specific evidence provided either in the SEM DPL ICR, or on BNDES’s website, or in the Brazilian government’s response to this PPAR that BNDES is actually applying the new environmental and social policies to investments, how they are being applied and monitored, and what the outcomes are. This lack of evidence was the basis of criticism leveled by thirty CSOs in a letter of complaint to the World Bank Regional Vice President (see below). In an interview with IEG, BNDES staff confirmed that the current extent of their project environmental and social screening and monitoring is to classify projects into categories “A, B or C”, which was something BNDES did even before the SEM DPL (see the Program Document, p. 101), and ensure that the investment has an environmental license if required by Brazilian law. To the extent that previously BNDES did not ensure that investments it financed were in compliance with Brazilian law, this could be seen as a step forward. But even though their own policy requires it, BNDES has no environmental or social monitoring indicators, there is no supervision or evaluation by BNDES of environmental and social aspects of projects under implementation, and BNDES cannot reject project proposals on environmental and social grounds alone (as long as they comply with Brazilian law). 4.40 BNDES points to changes it has made to its internal institutional structure as evidence of progress, such as the upgrading of its Environment Department to an Environmental Division. However, as illustrated by the “Credit concession process flowchart” on page 75 of its 2011 Annual Report, and confirmed by BNDES staff, the Environment Division is not 47 The resulting markets demands, combined with legal action taken by Brazil’s Public Prosecutor’s Office against irregular slaughterhouses, “led to a ‘Cattle Agreement’ in which the region’s largest beef processing companies agreed to exclude from their supply chain those livestock producers who deforested after October 2009” (Nepstad et al., 2014). The Nepstad article does not mention the role of BNDES. 34 involved in the analysis and approval of investments, including the environmental risk screening, unless it is specially requested to provide input (BNDES 2011). As a result, the impact of these institutional changes on environmental and social outcomes of BNDES’s investments is unclear. The lack of progress on BNDES’s environmental and social system is confirmed in multiple assessments by other parties since the closing of the SEM DPL (Reporter Brazil 2011; Ramos and Garzon, 2013; Widmer 2012; Lopes Pinto 2012; Fonseca and Mota 2013; Rainforest Foundation UK). 48 This falls far short of the ambitious goals of the SEM DPL, which among other things stated that “Commitment by BNDES to environmental and social outcomes is manifest by BNDES’ attempts to go beyond environmental and social guidelines focused solely on legal compliance” (Program Document, p. 102). 4.41 A critical case study for the application of BNDES’s environmental and social system, and an opportunity to showcase any improvements to it supported by the SEM DPL, is the Belo Monte hydroelectric dam complex in the Amazon rainforest of Pará State, Brazil. The site was the subject of an IEG field visit for the purposes of this evaluation, and is described in detail in Box 2. Reports indicate that Belo Monte will be the third largest dam in the world ever built, and is also the largest investment ever financed by BNDES, with a total cost of nearly USD 15 billion, Given the scale of the project, great environmental and social sensitivity of the location, and the experience Brazil has with the significant adverse impacts of large dam projects in the past, great care should have been taken to ensure that Belo Monte met the highest environmental and social standards before proceeding with the loan. Yet the Belo Monte investment has been widely criticized by other Brazilian government agencies, international human rights organizations, and local and international civil society organizations for adverse social and environmental impacts, and violations of Brazilian law, international agreements, and BNDES’s own environmental and social policies (Library of the European Parliament 2013; NORAD 2011). Box 2: BNDES financing for the Belo Monte Hydroelectric Dam A critical case study for the application of BNDES’s environmental and social system, and an opportunity to showcase any improvements to it supported by the SEM DPL, is the Belo Monte hydroelectric dam in the Amazon rainforest of Pará State, Brazil. Part of the Brazilian government’s PAC infrastructure investment program, 49 Belo Monte is a massive undertaking: the largest dam project under preparation anywhere in the world, once completed it will be the third largest dam ever built. It is also the largest investment ever financed by BNDES, with a total cost of nearly USD 15 billion, financed by BNDES at 80 percent, the limit for the bank. BNDES announced the first loan for the Belo Monte dam project well before the project had been granted the required license by IBAMA. According to BNDES’s environmental and social policy, an 48 See also: http://amazonwatch.org/work/bndes 49 http://norteenergiasa.com.br/site/ingles/norte-energia/ 35 IBAMA-issued Installation License 50 is required before an operation can be contracted.51 The first loan for the Belo Monte dam project was announced in December 2010, 52 well before IBAMA issued the Belo Monte consortium an Installation License (and only a week after the disbursement of the SEM DPL second tranche, raising the ire of the NGO community). At that point, only a Preliminary License had been issued based on a long series of conditions. The required Installation License was not issued until June 2011, and even then numerous conditions had still not been met. Many of the conditions were reported to still not have been met in 2013 53, and according to some analyses have even worsened. 54 The consortium building the dam complex, Norte Energia S. A., claims that the project will “generate energy constantly, at a low social and environmental impact” (Norte Energia 2014). BNDES states that it has “adhered to the decisions of the competent authorities” and made efforts to address the environmental and social impacts of the project. But with construction of Belo Monte still at a relatively early stage, there are reports of significant environmental and social consequences. These include involuntary displacement of thousands of families, including indigenous people; destruction of forests and fisheries; tripling of the population of the local town of Altamira without adequate services like healthcare or schools; labor unrest by dam construction workers; and a jump in social problems like drug use, crime, and sex trafficking. These and other problems are the subjects of fifteen separate class-action lawsuits related to Belo Monte filed by the Prosecutors of the Brazilian Federal Public Prosecutor’s Office (Ministério Público Federal). 55 According to the Federal Prosecutor’s Office, Federal Courts have already ruled against the Belo Monte consortium and BNDES in some important cases. They include halting construction at the end of 2013 due to the invalidity of environmental licenses issued by IBAMA, and a prohibition on BNDES financing, with the judge citing evidence of negative impacts on local communities and the environment. 56 In August 2014, a Federal Court confirmed that the environmental license 50 The licensing process has three distinct stages: 1. Preliminary License, 2. Installation License, and 3. Operating License. 51 See: http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/Apoio_Financeiro/Politicas_Tran sversais/Politica_Socioambiental/analise_ambiental.html ; and http://www.bancoldex.com/documentos/2615_Memorias_ALI14- 3_Claudia_Nessi_Seminario_Bogota_19.11.2010_Claudia_Nessi_resumido.pdf 52 http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/Sala_de_Imprensa/Noticias/201 0/energia/20101222_BeloMonte.html 53 For example, see: http://www1.folha.uol.com.br/mercado/2013/07/1318408-obra-de-belo-monte- pode-atrasar-por-problemas-com-ibama.shtml 54 For example, see: http://www.socioambiental.org/pt-br/noticias-socioambientais/ibama-afirma- que-o-cumprimento-de-condicionantes-de-belo-monte-so-piorou , and http://www.socioambiental.org/sites/blog.socioambiental.org/files/nsa/arquivos/quadro_condicioanan tes_2013_isa.pdf 55 http://www.prpa.mpf.mp.br/news/2013/processos-judiciais-do-caso-belo-monte-sao-publicados-na- integra-pelo-mpf 56 http://noticias.pgr.mpf.mp.br/noticias/noticias-do-site/copy_of_indios-e-minorias/norte-energia-e- obrigada-a-paralisar-obras-de-belo-monte-mais-uma-vez/?searchterm=Belo%20Monte (last visited September 17, 2014). 36 issued by IBAMA was void, citing faulty environmental impact assessments. 57 While some cases continue to be contested at various levels in Brazil’s legal system, the lawsuits provide a large amount of information on violations of Brazilian laws on the rights of indigenous peoples and on environmental protection, as well as violations of the terms of Belo Monte’s environmental licenses issued by the Brazilian environmental licensing agency, IBAMA. IBAMA itself has on more than one occasion cited numerous instances of non-compliance with the terms of the licenses issued for Belo Monte (IBAMA 2012, IBAMA 2013, ISA 2013). 58 As a result, IBAMA has imposed millions of dollars in fines on the Belo Monte consortium. 59 In the municipality of Altamira, which is the closest to the Belo Monte dam site, the Brazilian Institute for Applied Economic Research (Ipea), a Federal public foundation linked to the Strategic Affairs Secretariat of the Presidency, sponsored field research in 2011 and identified a host of serious urban environmental and social problems associated with the dam and influx of tens of thousands of workers, as well as increased pressure on surrounding protected areas and indigenous lands.60 Their general conclusion is: “Overall, Belo Monte repeats the same mistakes of other major historical hydroelectric projects, when it disregards the seriousness of the social consequences already experienced by other populations and the specific socio-cultural, economic and environmental conditions of the region.” This was echoed by multiple sources interviewed by IEG, who saw Belo Monte as a missed opportunity to finally do hydropower right from an environmental and social perspective. Internationally, the Inter-American Commission on Human Rights of the Organization of American States in April 2011 “granted precautionary measures for the members of the indigenous communities of the Xingu River Basin” and “requested that the State of Brazil immediately suspend the licensing process for the Belo Monte Hydroelectric Plant project and stop any construction work from moving forward until certain minimum conditions are met,” including carrying out a proper consultation process with indigenous communities and adopting “measures to protect the life and physical integrity of the members of indigenous peoples” (IACHR 2010) 61. The Committee of Experts on the Application of Conventions and Recommendations for the International Labor Organization (ILO) cited the Brazilian government for violating the rights of indigenous communities in the Xingu region while approving Belo Monte. The ILO reports allege that GOB violated the ILO’s Convention 169, which guarantees indigenous peoples the right to free, prior and informed consultation over projects that affect their lands and rights (ILO 2012). IBAMA itself has issued reports citing numerous violations of the conditions under which a partial environmental license was eventually issued for the project. The project and the licensing process have also been heavily criticized by a large number of national and international CSOs (ISA, Xingu Vivo, Survival International, WWF, Green Peace, Amazon Watch), who also document 57 http://noticias.pgr.mpf.mp.br/noticias/noticias-do-site/copy_of_meio-ambiente-e-patrimonio- cultural/tribunal-nega-recurso-da-norte-energia-e-confirma-prazo-de-90-dias-para-novos-estudos-de- belo-monte/?searchterm=Belo%20Monte (last visited September 17, 2014). 58 See also previous two footnotes and here: http://www.socioambiental.org/sites/blog.socioambiental.org/files/nsa/arquivos/3_parecer_ibama.pdf 59 For example, see: http://www1.folha.uol.com.br/mercado/2013/07/1318408-obra-de-belo-monte- pode-atrasar-por-problemas-com-ibama.shtml 60 http://www.ipea.gov.br/code2011/chamada2011/pdf/area7/area7-artigo19.pdf 61 http://www.oas.org/en/iachr/indigenous/protection/precautionary.asp 37 adverse environmental and social impacts, including through video evidence. 62 At the same time, some have questioned the economic viability of the project (Fearnside, P. 2012). 63 The general view of those familiar with the history of large dam projects in Brazil is that BNDES and the Belo Monte consortium have missed a great opportunity to improve environmental and social standards based on the lessons of past projects, but are instead repeating most of the same mistakes, and may be an effort to spearhead more dam construction in the Amazon (Fearnside, P. 2012). Similar complaints were mentioned during a meeting of dozens of community members attended by IEG. 4.42 Although Belo Monte is an emblematic example of the continuing weaknesses in BNDES’s environmental and social management system, it is not the only one. A recent investigation of the twenty largest BNDES-financed investments in the Amazon over the past five years found that at least sixteen were subject to legal action by federal prosecutors, states, or labor or civil society organizations due to environmental problems (Fonseca, B. and J. Mota, 2013). Approximately half of the twenty largest Amazon investments by BNDES were related to hydroelectric dams and transmission lines. All of the dams were the target of lawsuits for environmental and social problems. The problems include the poor quality of environmental impact assessments, involuntary resettlement, lack of consultation and disclosure with indigenous populations and other affected communities, and shortfalls in the environmental licensing process. An expert interviewed for the report stated that “compared to what the bank [BNDES] could do, what other international banks do, and what Brazilian law requires BNDES to do, it fails.” 4.43 Among the list of major investments financed by BNDES after the SEM DPL 1 is the Teles Pires hydroelectric dam (BNDES 2011). 64 Like Belo Monte, Teles Pires is accused of significant adverse environmental and social impacts, including lack of consultation with affected indigenous peoples, destruction of spawning grounds for fish that the local population depends on for its livelihood, and shortcomings in the environmental licensing process (Ministério Público Federal e Ministério Público no Estado de Mato Grosso 2012). 65 62 See for example: http://www.xinguvivo.org.br/2011/06/01/ibama-libera-licenca-de-instalacao-para- belo-monte-as-vesperas-do-dia-mundial-do-meio-ambiente/ ; http://oglobo.globo.com/economia/miriam/posts/2011/01/27/belo-monte-licenca-parcial-nao-existe- 359392.asp ; http://www.xinguvivo.org.br/2011/06/01/ibama-libera-licenca-de-instalacao-para-belo- monte-as-vesperas-do-dia-mundial-do-meio-ambiente/ ; http://www.internationalrivers.org/campaigns/belo-monte-dam; http://www.wwf.org.br/?28983/; www.socioambiental.org/esp/bm/index.asp ; http://amazonwatch.org/news/blog?start=55 63 http://www.economist.com/news/americas/21577073-having-spent-heavily-make-worlds-third- biggest-hydroelectric-project-greener-brazil 64 http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Hotsites/Relatorio_Anual_2011/Capitulos/de sempenho_operacional/infraestrutura/geracao_de_energia_de_base_hidreletrica.html 65 http://www.aneel.gov.br/arquivos/PDF/SUSPENSAO_UHE_TELES_PIRES_uniao_e_aneel_v3_bi p_junho_2012.pdf ; and http://www.internationalrivers.org/resources/comments-to-pjrces-on-the- teles-pires-hydropower-project-brazil-3056 38 In response to a recent civil society request to BNDES—under the Brazilian Law on Access to Information—for detailed information regarding the environmental and social safeguards applied in the case of Teles Pires, including the application of BNDES’s new environmental and social policy, BNDES stated only that the Teles Pires Hydroelectric Company was required to comply with Brazilian labor laws and environmental licensing requirements. These are Brazilian legal requirements regardless of BNDES’s involvement, and with or without the SEM DPL. 4.44 Another issue is BNDES’s continued lack of transparency with regard to environmental and social aspects of the investments it finances. The Equator Principles include Principle 10: Reporting and Transparency, which requires that, at a minimum, a summary of the Environmental and Social Impact Assessment is accessible and available online for relevant projects (see Annex E). The EP Principle 5: Stakeholder Engagement, also requires that financial institutions make assessment documentation readily available to affected communities and other stakeholders. BNDES’s Environmental and Social Policy contains no specific measures for disclosing information on projects. To its credit, in 2008 BNDES began posting on its website lists of investments it finances in Brazil (see Annex G for a sample). 66 But the list contains only very basic information such as the name of the client, date and financing amount of the loan, sector, and a one-line description of the project. The list also excludes “automatic indirect” investments and all investments if finances outside of Brazil (whether direct or indirect). Moreover, a project is included in the list only after the financing contract has been signed, and the only location information provided is the state, making it very difficult for affected stakeholders to be made aware in a timely fashion. BNDES still provides no information on environmental or social aspects of its investments, including the basic risk categorization, any safeguards applied, and actual impacts. 4.45 BNDES’s approach to reporting and disclosure of environmental and social information is limited compared to, for example, the World Bank, to whom Brazil turned for international best practice in this area under the SEM DPL. The World Bank discloses, during preparation of its investment operations, drafts of the Integrated Safeguards Data Sheet with detailed information on which safeguards are triggered (and which are not), why they are triggered, mitigation measures, and mechanisms for consultation and disclosure; and a Project Information Document with a detailed description of the operation. 67 Once an investment operation has been approved, the final Integrated Safeguards Data Sheet and full Project Appraisal Document or Program Document (in the case of DPLs) are typically disclosed. And during supervision, the World Bank now discloses information from the Implementation Supervision Reports for all operations. While it would not be realistic to expect BNDES to attain the level of safeguards of the World Bank and other international development banks right away, there should have been more progress in that direction given 66 http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/BNDES_Transparente/Consulta _as_operacoes_do_BNDES/operacoes_diretas.html 67 No Safeguards Data Sheet is prepared for DPLs since according to the World Bank’s definition, “safeguards” do not apply to DPLs. For DPLs, relevant analytic work conducted by the World Bank in relation to environmental and social aspects is supposed to be made public as part of the consultation process. 39 the intention to learn from the World Bank’s experience. As discussed above, BNDES’s policies also fall short of the Equator Principles, another international standard. 4.46 Given the dearth of evidence of progress in achieving Objective A and its sub- Objectives attributable to the SEM DPL, the Achievement of Objective A is rated as Negligible. Objective B: Manage natural resources sustainably, reducing the degradation of agricultural lands, forests (in particular the Amazon), and water resources, and promoting renewable energy. B.1. Natural Resource Management and Conservation 4.47 With regard to improving the sustainability of natural resource management, through strengthening of the forest legal framework, both the Public Forest Management Law and the Atlantic Forest Law were already supported by the 2004 Env PRL. The Public Forest Management Law was enacted in March 2006, and the Atlantic Forest Law was enacted in December 2006. Resolution 3545 of the National Monetary Council was enacted in February 2008, before preparation of the broader SEM DPL began. 4.48 With regard to improving Amazon regional planning for sustainable development and reduced deforestation, this has been one of the emblematic challenges and most important environmental goals for Brazil, and arguably for the planet. However, the preparation of the PAS and signing of an agreement to establish it was cited as an action taken under the 2004 Env PRL in its ICR. Ecological Economic Zoning in the Amazon and elsewhere (“area of invluence of the BR-163” and “North, North-East and Middle-West regions”) was also supported by the Env PRL. 4.49 As to the intended outcome of reducing deforestation in the Amazon, data show that Amazon deforestation peaked at 27,000 km2 per year in 2004, well before the SEM DPL, and began continuously decreasing over the following years. The evolution of Amazon deforestation is illustrated in Figure 2. 4.50 In fact, research has demonstrated that the greatest reductions in Amazon deforestation were likely due to the Brazilian government’s own Action Plan for the Prevention and Control of Deforestation in the Legal Amazon (PPCDAM) that was launched in 2004 (Nepstad and others 2014; IPEA and others 2011). Particularly impactful were increases in enforcement activities and a massive expansion of protected and indigenous areas, as well as a cutoff in agricultural credit to noncompliant farmers and ranchers via Resolution 3545 and the blacklisting of municipalities with high rates of deforestation in 2008 (Assunção and others 2013a, 2013b; Nolte and others 2013; The Economist 2013b). Interviews by IEG with key sources in Brazil confirmed that there is general agreement that PPCDAM has played a major role, and that it was primarily an initiative of the Brazilian government. Prior support by the Bank to the creation of protected and indigenous areas was also important, but these actions were largely in place by the end of 2006 Indeed, the Director General of the Federal Forest Service, who had worked with the service for decades, stated that he had never before heard of the SEM DPL. 40 4.51 Another action under this policy area was the macro ecological-economic zoning of the Amazon region, which has been completed. In principle the idea of planning which areas of the Amazon should be used for which purposes (including conservation) makes sense. It has been pursued for many years in Brazil. IEG interviews at the federal and state level found that the macrozoning effort was a paper exercise with little evident impact on interstate planning or coordination. 4.52 In short, Brazil has made major strides in reducing its deforestation rate over the past decade, through policies other than those supported by the SEM DPL. Maintaining and increasing the reductions, according to an analysis for IEG’s Country Program Evaluation of Brazil, will require a new set of policies to support Brazil’s newly revised Forest Code (IEG 2013). These will include implementing the comprehensive rural cadaster, finding ways to minimize the costs to landholders of complying with the Code, and addressing the challenge that residual deforestation is becoming increasingly concentrated among low-income smallholders. Figure 2: Area of Amazon deforestation over time Source: PRODES Project, INPE, Brazilian Ministry of Science and Technology 4.53 With regard to improving rainforest conservation, the main actions were related to the Amazon Fund (with the Atlantic Forest Fund planned to come in under SEM DPL 2). As an evaluation by the Norwegian Agency for Development Cooperation (NORAD) makes clear, “the creation of the Amazon Fund was a Brazilian initiative” with financial support provided by Norway’s International Climate and Forest Initiative (NICFI) providing “a significant stimulus” in February 2008, before preparation of the SEM DPL began (NORAD 2011). The NORAD report provides a detailed evaluation and history of the Amazon Fund, with no mention of a role for the World Bank. The evaluation finds that “NICFI has been effective in getting things started”, but that “what has been done so far has had limited effectiveness and has not been able to be particularly efficient due to procedural constraints” on the part of BNDES, who manages the fund. This has caused frustration amongst the CSOs 41 and potential beneficiaries of the fund. More seriously, the evaluation notes that “the bureaucratic and legal bottlenecks encountered…are moreover not only seen by indigenous organisations and community associations as a technical problem, but a mechanism that repeats failures to recognize their rights and importance in reducing deforestation.” At the time of the evaluation in 2011, there was as yet “no clear sustainability strategy for the Fund.” 4.54 More recently, there is evidence that activity has picked up under the Amazon Fund, with a reported 58 projects supported at a total commitment level of R$ 878 million—though only three projects have so far been completed (Amazon Fund). The Amazon Fund website, meanwhile, could serve as a model for BNDES, with more detailed information on projects, including geo-referencing (though still no information on environmental and social safeguards applied). 68 BNDES also reports progress with the establishment of the BNDES Mata Atlantica Initiative, which is intended to provide grant support for restoration projects in the Atlantic Forest biome. The BNDES website notes that 15 projects have been approved so far, for a total of R$ 25 million, though as with BNDES’s other investments, minimal information is provided on the individual projects (BNDES 2014). 69 B.2. Water Resource Management 4.55 With regard to improving management and quality of water resources, the ICR reports that 116,144 km of rivers were being monitored as of June 2011, surpassing the program target of 90,000 km. Based on interviews conducted by IEG and a visit to the federal water agency, ANA, Brazil does appear to have made good progress in terms of increasing the coverage of water monitoring sites around the country. It is not clear to what extent this can be attributed to the SEM DPL. The first action under the SEM DPL, approval of the Water Resources National Plan (PNRH), was accomplished in 2006, several years before the SEM DPL, while the documentation for the plan was prepared in 2004 and was an action under the earlier Env PRL operation. The quality of the outcome indicator could also be questioned, as it does not specify which parameters are to be monitored. 4.56 ANA reports progress in signing cooperation agreements with states for implementing the National Water Quality Program (PNQA), the second tranche release condition. The PNQA is an important advance in Brazilian water resource management policy. It is clear that the ANA monitoring network was strengthened by the agreements established with the states, especially those with pre-existing monitoring systems. The benefits of the PNQA for the states are less clear. The actual implementation of the water monitoring, as well as the use of the information, should be carried out by local governments, which the SEM DPL did not involve. In interviews with state government water and environment officials, they reported that the relationship with the Federal ANA on water 68 See http://www.fundoamazonia.gov.br/FundoAmazonia/fam/site_pt/Esquerdo/Projetos_Apoiados/ (last visited August 18, 2014). 69 See http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Areas_de_Atuacao/Meio_Ambiente/Mata_Atla ntica/ (last visited August 18, 2014). 42 quality monitoring was mostly one-way: the states provide their water quality monitoring to ANA, but have thus far not received in return any support, for example to invest in necessary water monitoring infrastructure or to improve human capacity. The signing of the cooperation agreements is however an important first step. 4.57 The most important aspect of the SEM DPL outcome indicator was that the results of water monitoring were to be used “for prioritization of investments for improved water quality.” On this count ANA could not cite any examples. Nor did the Bank’s ICR provide any information on the use of the water quality information. B.3. Environmental Sanitation 4.58 With regard to reducing environmental impacts through improved water, wastewater treatment, and solid waste services, the prior action on enacting the National Guidelines for Water Supply and Environmental Sanitation (Law 11,455) was approved on January 5, 2007, approximately two years before the SEM DPL, and therefore could not be attributed to the World Bank loan. 4.59 The ICR reports that 141,280 tons of pollution loads (biochemical oxygen demand, or BOD) were being reduced by June 2011, surpassing the target of 110,000 tons. These pollution reductions were supposed to be achieved by a combination of BNDES-financed investments “reviewed under new BNDES social and environmental guidelines,” and investments financed under the “updated PRODES program.” The World Bank team set the baseline at zero, “since the new BNDES social and environmental guidelines and the updated PRODES program are not available yet” (Program Document). As discussed in detail above, the new BNDES environmental and social guidelines have major flaws and gaps. Moreover, the promised sector guidelines on solid waste management, and on water, sanitation and drainage, were apparently never finalized, or at least are not in use. No information is provided on pollution reductions from BNDES projects before the SEM DPL to establish a more valid baseline. No information is provided on how the pollution reductions were achieved (for example, through which investments) or on how they were measured. In their comments on this evaluation, BNDES notes that the 141,280 tons of BOD were the result of 15 projects in nine states. BNDES also indicates that they were financing wastewater treatment investments prior to the SEM DPL, but “did not systematically measure the quantities of BOD.” They credit the SEM DPL with introducing this practice, which is a positive sign. However, it also means that we do not know what incremental reduction in pollution loads can be attributed to the SEM DPL, since the baseline was not zero BOD as indicated in the Program Document and ICR, but rather some unknown quantity. 4.60 Improvement of the national PRODES program, which promised to be highly innovative by taking a payment for results type approach to sewage treatment, was already supported under the 2004 Env PRL. Although it was supposed to be a key aspect of this policy area, the ICR does not report any information on it. In its submission of a report to the Bank in June 2011 as input to the ICR, MMA and ANA did provide information on PRODES. They state that the program did not operate at all over the period of the SEM DPL due to a lack of funding. This is confirmed by the Water Agency (ANA) website, which 43 shows that no new contracts were issued in 2009 or 2010. 70 Even prior to the SEM DPL, as mentioned in the Program Document, up-take on this program had been uneven due to the inability of local governments to provide the required up-front financing. There is no indication that this problem has been addressed. Although there were contracts signed in 2011 and 2012, there were again no contracts issued in 2013. In interviews with IEG, the Rio state government (one of the highest-capacity in Brazil) cited the lack of up-front financing as a continuing major shortcoming, and the main reason why, to the best of their knowledge, no PRODES investments had been implemented in the state. B.4. Renewable Energy 4.61 With regard to promoting renewable energy potential, the ICR reports that 50,102 terajoules per year were being generated from renewable energy sources supported by BNDES, just short of the 60,000 terajoule target. The evidence for the renewable energy production comes from a report submitted by BNDES as input to the ICR in June 2011. BNDES reports that 50,102 terajoules per year were produced by renewable energy sources since January 2009. But to arrive at this figure, BNDES took into consideration the potential energy generation from wind power, mini hydropower and biomass cogeneration projects approved since 2007, as well as the projects considered “enquadrados”, or framed, but not yet approved. Over half of the total came from biomass cogeneration. The document mentions that 5 regions would be hosting 77 renewable energy projects. 4.62 Similarly to the environmental sanitation policy area, while no information is provided on renewable energy generation from BNDES projects before the SEM DPL, the Program Document sets the baseline at zero. However, evidence indicates that BNDES did support investments in renewable energy prior to the SEM DPL. BNDES’s 2012 Annual Report provides data showing that the bank disbursed R$ 5.7 billion for renewable energy and energy efficiency in 2008, and R$ 6.1 billion in 2012. While BNDES should be recognized for its significant support for these investments, the incremental change over the period is relatively minor (BNDES 2012, BNDES presentation to the IAEA 2013). The biggest share of BNDES’s renewable energy investments—totaling R$7.1 billion in 2012— continues to go to large hydropower dams, which as described elsewhere continue to suffer from important adverse environmental and social consequences as a result of weak safeguards. But BNDES deserves credit for increasing its investments in other forms of renewable energy as well. 4.63 It should be noted that the target for this policy area is specific to BNDES, and has little relation to the prior action on introducing the Integrated Environmental Assessment (IEA) methodology in the Brazilian national Handbook of the Electricity Sector’s Inventory. The IEA methodology was apparently included in the Handbook of the Electricity Sector’s Inventory in December 2007 (a year before the SEM DPL). The World Bank had been engaging on this issue through its analytical work under other activities (see for example World Bank 2008). It is not clear that the World Bank’s recommendations were fully captured by the new IEA methodology. The World Bank, as elsewhere in the world, had promoted a Strategic Environmental Assessment (SEA) approach, which would examine the 70 http://www.ana.gov.br/prodes/default.asp 44 potential environmental, social, and economic—as well as technical—impacts of hydroelectric dams at various sites along an entire river basin, and use the results to inform decisions about whether, where and what to build. The approach that Brazil continues to use has been criticized for selecting sites first based on technical engineering grounds, and only then attempting to estimate the environmental, social, and economic implications. 4.64 In some areas, Brazil has made good progress on this sub-objective, including significantly reducing deforestation and increasing water quality monitoring. It is often less clear whether this progress can be attributed to the SEM DPL, particularly in the case of the decreased deforestation. In other areas, such as using the water quality information to prioritize investments, or expanding the PRODES wastewater treatment program, intended outcomes have still not been achieved. Overall, the Achievement of Objective B is rated as Modest. 5. Ratings Outcome 5.1 The Relevance of Objectives is rated Substantial due to the importance of improving management of the environment for Brazil and consistency with the CPS that prevailed at the time of loan closing. But because of weaknesses in the results framework, project logic, and choice of instrument, the Relevance of Design is rated Negligible. 5.2 In its justification for assigning the loan an overall outcome rating of “satisfactory”, the ICR states that “the results obtained on the environmental management front are particularly relevant for Brazil. BNDES now systematically screens, approves and monitors all projects according to the BNDES new Environmental and Social Institutional Policy.” Yet IEG found no evidence either of improved screening by BNDES or more importantly of improved environmental and social outcomes from its investments. The ICR provided no information on other financial institutions. For other policy areas such as those related to mainstreaming climate change and improving water quality, the reforms seem to have stalled, with no evidence of outcomes. For remaining policy areas such as reducing deforestation, Brazil has achieved success, but this could not be attributed to the SEM DPL. The Achievement of Objective A under the operation suffered from severe shortcomings, with a consistent lack of evidence that it achieved its objectives, and was therefore rated Negligible, while the Achievement of Objective B suffered from major shortcomings and was rated Modest. 5.3 Overall, much more should have been expected in terms of outcomes from the largest ever single World Bank loan to Brazil, for US$ 1.3 billion. A Substantial rating for Relevance of Objectives, one Modest and one Negligible rating for Achievement of the Objectives, and a rating of Negligible for Relevance of Design, leads to the SEM DPL being rated Unsatisfactory for the Outcome. 45 Risk to Development Outcome 5.4 In general, as discussed in detail in the Achievement of Objectives section, the relevant outcomes that could be attributed to the SEM DPL operation were limited, so there are fewer achievements to sustain. For some of the policy areas, there appears to be good government ownership and momentum, because actions began well before the SEM DPL— including in some cases under the earlier Env PRL DPL–and continue today. But little of this could be attributed to the SEM DPL in the first place. Examples include reducing deforestation, increasing water quality monitoring, and providing BNDES financing for renewable energy investments. 5.5 In other areas, progress has been slower than expected. The revision of the National Plan on Climate Change that Government committed to is years behind schedule, progress on development and implementation of the sectoral action plans is lacking, and the withdrawal of CSOs from the consultations has cast doubts on the process. However, the Government now reports that some sectoral plans have been developed and that approval of the revised National Plan is imminent. The intermittent funding for and uptake of the PRODES program has meant that the growth of that innovative mechanism to finance sanitation has lagged behind expectations. For the financial sector, the implementation of the 2008 revision to the Green Protocol supported by the SEM DPL was put on hold, and its contents fell short of international standards. 5.6 With regard to some policy actions, there is greater uncertainty. For example, a very different, new Forest Code was approved by the Brazilian government in 2012. Though some of the changes are controversial, implementation mechanisms are still being developed and it remains to be seen how this will affect deforestation in the Amazon and elsewhere (Soares- Filho and others, 2014). With regard to the strengthening of environmental institutions, the evidence indicates that Brazil’s main Federal environmental agencies continue to suffer from institutional, legal, and human resources problems. 5.7 As for the key actions aimed at improving BNDES’s environmental and social management system, not only did progress fall short under the SEM DPL, it has continued to lag in the intervening years. A prime example is the sectoral guidelines that were supposed to be developed under the SEM DPL program in order to provide detailed instructions on application of BNDES’s new environmental and social policies. Of the thirteen guidelines promised in the Program Document, only three have been developed. In addition, the resolution BNDES approved to govern the application of the new guidelines (Resolution 2022) does not state that they are binding. Overall, there is almost no evidence that BNDES has improved its environmental and social management system in practice. At the same time, BNDES reports that it has maintained or continued to develop policies and institutions related to its environmental and social management system. The challenge remains for BNDES to put these changes into practice in order to improve the outcomes of its investments. 5.8 Risk to Development Outcome is rated Moderate. 46 Bank Performance Quality at Entry 5.9 Quality of program logic: In some cases, it can be seen as a positive for the World Bank to be responsive to the changing needs and priorities of Borrowers by being flexible in the design of operations and the instrument employed, within the requirements of World Bank policies and procedures. In the case of this operation, the objectives changed significantly from an effort to finance BNDES’s investments while improving its environmental and social standards, to a broad-based improvement in the national environment. In making that transition, the World Bank team tried to quickly merge elements of both concepts, and in the process ended up with a muddled program logic that made it unclear what the operation was trying to achieve. 5.10 Learning from previous operations: The Program Document has a section that explains how the SEM DPL team has learned from the experience of the first World Bank environmental DPL with Brazil, the First Env PRL that was approved in 2004. The SEM DPL Program Document goes so far as to say that “this operation is based to a large degree on the lessons extracted from the Env PRL I”. Nevertheless, as mentioned in the Relevance of Design section, a major shortcoming in the design of the SEM DPL was the lack of learning from the lessons of the Env PRL. The Env PRL had similar objectives, similar policy areas, and used the same instrument. But the World Bank team proceeded with preparation of the SEM DPL without having properly evaluated and completed the required full ICR of the Env PRL. Not having learned the lessons of why the Env PRL series was canceled after the first loan, they repeated the same mistake, designing the SEM DPL as a programmatic series that was again canceled after the first loan. 5.11 The World Bank team also designed the SEM DPL with many of the same policy areas—and some very similar policy actions—as the Env PRL. But without a comprehensive evaluation of the status of actions and outcomes from the Env PRL, there was a lack of determination of what had already been accomplished under the policy areas and what was yet to be completed, and more importantly what the outcomes were. 5.12 Inclusion of unattributable policy actions: As indicated in the Good Practice Note for Development Policy Lending: Designing Development Policy Operations, and confirmed in discussions with OPCS, actions included in a DPO policy matrix must be the result of the World Bank’s engagement in relation to the DPO, for example through dialogue, analytical work, or technical assistance provided. In the case of the SEM DPL, as established for many of the policy areas in the Achievement of the Objectives section above, the Bank team included multiple policy actions that took place well before the preparation of the SEM DPL began, or that were already supported by the Env PRL. 5.13 Timing and readiness: Once it was decided to change the loan instrument to a DPL, the operation was prepared very quickly, with the Program Document being finalized within four months and approved by the Board in less than five months. The loan then sat for nearly 15 months with no disbursements due to a delay in effectiveness. The ICR states that “the delay was caused by the Borrower not having formally requested the reflection of the receipt 47 of the Loan funds in its 2009 federal budget.” But given the large number of operations that the Bank has with Brazil, including DPLs, the Bank team should have been more aware of this risk at entry, and ensured that any Brazilian legal prerequisites were in place prior to approval of the loan. The additional time could have been usefully employed to improve the quality of the operation. The three months originally allotted between disbursement of Tranche 1 and Tranche 2 also appears inadequate given the many Tranche 2 release conditions, a number of which would have been challenging for the client to implement. 5.14 Sequencing and backloading of reforms: One difference in design between the SEM DPL and the Env PRL was that the SEM DPL divided the first loan into two tranches, so that even though there were long delays in the disbursement of the second tranche, eventually the Bank team certified that the tranche release conditions were met. Dividing the first loan into two tranches appears to have increased the motivation for the government to continue with the loan through the second tranche because they had already signed the loan agreement that covered both tranches. Nevertheless, full implementation of many of the reforms was not foreseen until SEM DPL 2, which was designed as a separate loan, and which was canceled. This is contrary to OPCS guidance, which advises teams to spread critical reform steps evenly across loans in a programmatic series. 5.15 Identification and mitigation of risks: The Bank team underestimated the risk of “the possibility of the reform agenda not being carried out by BNDES” rating it “moderate to substantial” (World Bank Program Document, 2009), and did not follow through on promised mitigation actions. The main mitigation response described in the Program Document was the provision of technical assistance and capacity building to BNDES through the “substantial expansion” of the separate Environmental Technical Assistance (ENV TAL) Project. However, the ICR for the ENV TAL, which is now also closed, makes no mention of any support provided to BNDES, and the ENV TAL was never expanded (World Bank 2011b). 71 The Bank team approached the risk as a technical problem (i.e., how to carry out the reforms), rather than as a risk of lack of political will both inside BNDES and in the Brazilian Government to fully carry out the reform agenda that could result in the slowing or limiting of certain investments at a time when the PAC was calling for major increases in BNDES financing for large infrastructure investments. The Program Document also states that an overall risk mitigation factor for the SEM DPL is “the factor of continuity with the Env PRL I of all policy areas.” This is not accurate, as the Env PRL I did not include policy actions related to BNDES’s environmental and social management systems, which is where the greatest risks lay. 5.16 As noted above, the SEM DPL was a loan to the federal government and was supposed to support policy actions primarily at the federal level, but implementation of many of the policies depends at least in part on state and municipal governments. It should have been considered important therefore for the SEM DPL team to coordinate closely with the National Environment Project, which was supposed to provide capacity building for state and municipal environment agencies, and to help ensure its successful implementation in tandem with the SEM DPL. As noted above, the SEM DPL Program Document makes only a single 71 In fact, the ENV TAL had only disbursed about 65 percent of its original commitments when it was closed. 48 sentence reference to the National Environment Project. Phase 2 of NEP II was approved in September 2009, but by the end of 2013 it had disbursed less than 7 percent of its funds and its progress was rated “unsatisfactory”. The potential failure of this project to provide needed support to local governments was never identified as a risk by the SEM DPL team, and coordination on the operations was apparently not adequate to ensure that they both progressed adequately. In interviews with IEG, stakeholders in Brazil frequently cited the continued lack of capacity at these local government levels as a significant weakness in Brazil’s environmental management system. 5.17 Collaboration and coordination with partners: “Coordination with Development Partners” is a requirement of OP 8.60. At the time of the preparation of this loan, the IFC had already been engaged with BNDES in a dialogue on improving BNDES’s environmental and social standards through the adoption of the Equator Principles, which are based on IFC’s Performance Standards, and IFC had invited BNDES to be part of IFC’s Performance Standards Community of Learning. Since BNDES finances both private and public investments, there would have been advantages to learning from IFC’s approach. Other public and private banks in Brazil have already adopted the Equator Principles. The SEM DPL Program Document makes no mention of IFC’s dialogue with BNDES on their environmental and social system. Adoption of the Equator Principles could have been a policy action under the loan, and would have saved BNDES from “reinventing the wheel” in developing its policies, and would have integrated them in an established community of practice. Instead, the Bank team neglected to incorporate IFC in its engagement with BNDES on environmental and social issues, and BNDES’s engagement with IFC on these issues stalled. There was in general a lack of coordination between IBRD and IFC on engagement with BNDES on its environmental and social management system. 5.18 More generally, there are numerous other organizations working on environmental issues in Brazil, including other IFIs as well as important bilateral and civil society donors. The SEM DPL could have been used as an opportunity to enhance coordination and partnership across these organizations. But according to the Program Document and ICR, no other partners were included. 5.19 Mitigation of adverse environmental and social impacts: In general, the SEM DPL Program Document attempts to address environmental and social issues by arguing that “the policies and reforms supported by this operation are likely to have a significant positive impact.” Nevertheless, the Program Document identified several potential social and environmental threats related to specific policy areas targeted by the SEM DPL (pp. 51-58). On the social (and poverty) side, the Program Document identified potential adverse impacts including restricted access resulting from the creation of protected areas; increased water tariffs as a result of policy reforms; and loss of “habitat” by indigenous communities as a result of hydroelectric sector development in the Amazon. With regard to the environment, potential adverse impacts identified in the Program Document include deforestation associated with the expansion of the ethanol sector; “habitat degradation” related to the expansion of hydropower; and deforestation arising from unclear guidelines for logging. 5.20 For the potential adverse environmental and social impacts identified, the Program Document describes mitigation measures that will be put into effect in association with the 49 loan. These include actions such as improvements to the environmental licensing process, increasing transparency and civil society participation, support for sustainable forestry practices, the adoption of improved BNDES environmental and social guidelines for a variety of sectors, and improvements to basin-wide integrated planning of the hydroelectric sector. 5.21 Of particular importance, the Program Document acknowledges “BNDES’ significant role in financing projects and enterprises which engage in activities with potential adverse environmental impacts” (p. 57). In response to this concern, the Program Document states that “the proposed operation is expected to have a leveraged impact: the full spectrum of BNDES over 70 financial agents is expected to observe the same improved environmental and social policies to be adopted by BNDES with support from this operation.” This seems to indicate that all of BNDES’s “indirect” investments through financial intermediaries, or “financial agents”, will also be covered by the improved environmental and social system, but in reality this was not the case. Both the SEM DPL policy actions described in the Program Document and the GOB’s Letter of Development Policy indicate that only BNDES’s directly financed operations would be covered. 72 In the same section, the Program Document also cites the adoption of a number of environmental and social sectoral guidelines by BNDES as mitigation actions, including for agriculture, sanitation, and renewable energy/hydropower, but there is no evidence (including on BNDES’s own website) that BNDES has approved any sectoral guidelines other than for livestock, ethanol, and thermoelectric (and even for those there is no evidence of implementation or results). 5.22 If this loan had continued to be prepared as a FIL—a type of investment lending instrument—as proposed in the two Concept Notes issued in June and September 2008, the operation would have had to adhere to the World Bank’s OP 8.30, which among other things requires removal of interest rate subsidies, application of World Bank policies on environmental and social safeguards, and financial management and procurement, to the loan and all sub-projects financed. 73 In their written comments, reviewers of the Concept Note were critical with regard to these requirements and BNDES’s ability to meet them, including because BNDES supplies loans at subsidized interest rates, and because one of the main rationales for the operation was that BNDES’s environmental and social policies were weak. 5.23 As a result of the SEM DPL loan, the resources were transferred in their entirety to BNDES and used to finance its investment operations, as discussed in the Background section and Annex H. The environmental and social requirements of OP 8.60 address only the potential impacts of the policy reforms supported by DPLs, and not the potential impacts of the financing, because it is assumed that under budget support operations, there are no investments with “footprints” to be concerned about. But in this case, the US$ 1.3 billion SEM DPL financing used by BNDES to finance its investment operations could have had a significant footprint, particularly considering the many large infrastructure investments that BNDES finances. According to the official interpretation of OP 8.60, all lending is fungible, 72 As discussed in the Achievement of Outcomes section, BNDES now reports that it applies the policies to “non-automatic indirect operations”, but not to “automatic indirect operations”. 73 OP 8.30 was replaced by OP 10.00 - Investment Project Financing, in April 2013. 50 and the government can do whatever it wants with the money, so there is no connection between the DPL and the money that the government on-lent to BNDES. But this is inconsistent with a common-sense view that the funds were in fact earmarked for BNDES, given that: a) there were explicit legal provisions and contracts to transfer the DPL resources to BNDES on similar terms to the World Bank’s; b) BNDES stated that these resources were used “to complement the BNDES disbursement budget”; and c) BNDES’s lending increased. The potential environmental and social impacts of the investments financed are unexamined, and this created reputational risks for the Bank, as evidenced by the many complaints this loan generated from a diversity of civil society organizations, and echoed by development partners (see discussion under Quality of Supervision below). 5.24 Inconsistency between results promised and actions supported: With regard to the policy area to “Improve the effectiveness of environmental and social management systems in BNDES and other financial institutions”, it was already explained in Section 2 that there was an inordinate focus in the policy actions on BNDES. Even for BNDES alone, the policy actions and indicator that were included in the policy matrix are substantially less ambitious than what was promised. At the time the Program Document was written, two-thirds of BNDES’s lending was done indirectly, through financial intermediaries. The Program Document (p. 44) promised that the SEM DPL would target “the promotion of greater facilitation of sustainable investments (directly through BNDES and indirectly),” and later (p. 21) states that “a trigger for the SEM DPL II would be that BNDES apply the new Environmental and Social Institutional Policy to its full portfolio, including operations managed by its financial intermediaries…” The actual outcome indicator for improvement of BNDES’s environmental and social systems, however, applied only to “projects submitted directly to BNDES.” So not even the majority of BNDES's investments were subject to the actions supported under the SEM DPL. 5.25 Analytic underpinnings: OP 8.60 states that “a development policy operation draws on relevant analytic work…”, including explicitly on “social impacts of proposed policies, environment and natural resource management”. The Bank team was able to draw on extensive analytical work produced over the years as a result of the Bank’s many environmental activities in Brazil over the years, as described in the SEM DPL Program Document section IV.E. Analytical Underpinnings. Much of this analytical work was on topics related to the management of forest resources and Amazon conservation, as well as additional work on water resources, sanitation, energy, and environmental licensing. A notable exception is the absence of any analytical work on BNDES or “other financial institutions” and their environmental and social management systems, despite the prominent role that BNDES in particular was supposed to play in the operation, and the significant risks involved. The Bank also lacked an overall environment strategy for Brazil that could help to prioritize reforms. As noted above, the design of the SEM DPL did not include the recommendations of the environmental licensing study that was completed not long beforehand. 5.26 Absence of consultation, transparency, and disclosure measures: An important aspect of environmental and social management at development Banks such as the World Bank is the establishment of policies and procedures for consultations, transparency, and disclosure of potential impacts and mitigation measures. Yet, despite the absence of such 51 measures at BNDES, the SEM DPL policy matrix did not contain any actions related to these important aspects of a modern environmental and social management system. And as analyzed in detail in Annex E, neither BNDES’s new environmental and social policy nor the Green Protocol contained such measures. 5.27 There are also requirements for “Consultations and Participation” spelled out in the Bank’s OP 8.60 for all DPLs. The SEM DPL Program Document provides a description of the GOB’s consultation process on its 2008-2011 Multiyear Plan, as well as other national processes and plans, as evidence for relevant consultations. But there is no evidence of consultations related to BNDES or its environmental and social systems, and no evidence that the Bank team supported a robust consultation process as part of preparation. OP 8.60 also contains a requirement that “Relevant analytic work conducted by the Bank, particularly on poverty and social impacts and on environmental aspects, is made available to the public as part of the consultation process” (World Bank OP 8.60). IEG could find no evidence that this was done in relation to the SEM DPL. The lack of consultation, transparency and disclosure were some of the main complaints leveled against the World Bank and the SEM DPL by an array of civil society organizations in letters to World Bank management. 5.28 Finally, none of the World Bank’s SEM DPL documents—including the Program Document and ICR—mentions that the operation was previously considered using different instruments and beneficiaries, and that all of the funds were on-lent by the Brazilian government to BNDES. In the interest of transparency, it would have been preferable for the World Bank team to provide that information in the SEM DPL program documentation. 5.29 Quality at Entry is rated Unsatisfactory. Quality of Supervision 5.30 Five supervision missions were conducted, with 30 weeks of Bank staff input. Eight months after approval, with zero disbursements, the Implementation Supervision and Results Report (ISR) still rated performance on DO as satisfactory and on IP as moderately satisfactory. It was only after more than one year without disbursements that these ratings dropped to Moderately Unsatisfactory and Unsatisfactory respectively. The early ratings now appear overly optimistic and not fully candid. 5.31 The Bank team blamed the delay in effectiveness of the operation on the Borrower and delays in its budget legislation. Since there were only three months planned originally between release of Tranche 1 and release of Tranche 2, the nearly 15 month delay in Tranche 1 should have provided more than adequate time for the Bank team to ensure that Tranche 2 release conditions were also met, and the tranches could have been released at around the same time. Yet by the time of the Tranche 1 release, the Tranche 2 conditions apparently still had not been met, and required more than five months of additional time, resulting in a delay of more than 17 months compared to the originally planned Tranche 2 release date—even greater than for Tranche 1. 5.32 Although the problems related to the Brazilian budget and loan approval process were eventually addressed, and both tranches of SEM DPL 1 were disbursed, the Bank team was 52 not able to continue the engagement with the client necessary to ensure continued momentum under the planned SEM DPL 2. The different explanations provided for this by the Bank team are not very convincing. The ICR states that because disbursement of SEM DPL 1 was so delayed, the preparation of SEM DPL 2 “was not started in time to meet [the] required follow-up preparation timeframe of two years from World Bank’s Board approval” (which would have meant March 2011). It continues that “a follow-up operation would fall into the next CPS implementation period, within which it would no longer be a priority activity to either the Bank or the Client”, but in the next paragraph states that GOB continues to implement the main policies. It is not clear how the improvement of Brazil’s environmental management system could go from being a major priority to justify the original operation and its disbursement to no longer being a priority in the course of two years. In interviews, Bank Brazil Country Office staff stated that the reason for the cancelation of SEM DPL 2 was that the client’s priority was now to provide budget and other support at the state level rather than the federal level, because that’s where the needs are the greatest. However, this was already known at the time of preparation of the 2008-2011 CPS (p. 10) and did not prevent both parties from agreeing to the SEM DPL program in the first place. Moreover, since the SEM DPL resources went to BNDES it meant that they were used to finance investments at the local level anyway. Overall, the ICR provides little analysis of why the second loan in a Brazil environmental DPL series was canceled for the second time in a row. 5.33 The Bank could have been more candid in reporting on the performance of the SEM DPL, including in the Second Tranche Release Document and the ICR. As described in detail in Section 4, Achievement of the Objectives, the SEM DPL in many cases fell short in implementing the reform program described in the Program Document. For example, the Tranche Release Document (para. 19) for the second tranche states that “a supplementing resolution on the Structure and Use of Environmental and Social Guidelines (Resolution 2022/10) gives binding force [emphasis added] to all the sectoral guidelines including those that BNDES has designed for three sectors under the scope of the SEM DPL I.” However, that is not stated in the resolution. On the contrary, the introduction to Resolution 2022 states: “The guidelines have an instructional character and their content does not create obligations additional to the existing Brazilian legislation and Resolutions from BNDES Directorate”. This seems to indicate that the sectoral guidelines are voluntary. 5.34 Also for release of the second tranche, BNDES was supposed to apply its new environmental and social policy to all of its directly financed operations. This is a critical action that was not addressed in the Second Tranche Release Document, and no evidence that BNDES was meeting it was found by IEG. 5.35 In the case of their ICR, in some cases the Bank team did not fully and accurately report information provided by the implementing agencies as input to the ICR. Examples include the failure to mention that the PRODES program had not been funded; that BNDES reported screening of projects according to its new environmental and social system for a period of only three months, and that it excluded indirectly financed investments and foreign investments with potentially serious environmental and social implications; and that the emission reductions from carbon investments were reported as a total instead of the increase over the baseline that the indicator targeted. 53 5.36 The incomplete evidence of full implementation of SEM DPL actions, combined with limited transparency and public consultations, led to civil society organizations raising serious questions about the loan. For example, a letter dated Dec. 17, 2010 was sent to the World Bank Regional Vice President, with copy to the World Bank Country Director and others, and signed by thirty Brazilian and international CSOs. It stated that “the purpose of this letter is to communicate the existence of serious problems and to request clarifications and other needed actions from the World Bank regarding Loan Agreement No. 7660-BR in the amount of USD 1.3 billion for the SEM DPL 1.” The letter goes on to request that “the second disbursement of the SEM DPL 1 for USD 500 million should not be approved by the World Bank…we convey our extreme concern [emphasis from original] with the lack of evidence regarding implementation of BNDES social and environmental safeguards policies in emblematic cases, such as the unprecedented subsidized loan for the USD 14.7 billion Belo Monte Hydroelectric Complex, whose approval was signaled prior to completion of technical analysis of the project.” The CSOs requested a dialogue with the Bank team on these issues, disclosure of the detailed evidence used to justify the disbursement of the loan, and a hold on disbursement of the second tranche. World Bank Management responded and in at least one case did provide essentially the same evidence that the World Bank team used to justify the second tranche release internally. But the evidence fell short of what the CSOs had expected, and Bank Management offered the possibility of dialogue only in the context of the preparation of the SEM DPL 2 loan, which never materialized. In the midst of this exchange, the Bank disbursed the second tranche of SEM DPL 1 without informing civil society, and quickly closed the operation a few weeks later. This generated a great deal of controversy and suspicion on the part of the CSO community that continues to this day, including in entire reports critical of the loan produced by CSOs, and echoed by important development partners (Library of the EU Parliament 2013; BICECA 2011; Rainforest Foundation UK). 5.37 Despite the broad, active CSO community in Brazil that has an interest in the environment in general and in the SEM DPL in particular, and the potential impacts of the operation on stakeholders, the World Bank team’s ICR left blank the annexes on “Beneficiary Survey Results”, “Stakeholder Workshop Report and Results”, and “Comments of Cofinanciers and Other Partners/Stakeholders”. 5.38 Supervision of environmental and social impacts was inadequate. As noted above, the SEM DPL Program Document describes an array of potential adverse environmental and social impacts from the reforms supported, as well as mitigation measures that will be put in place in association with the loan, as required by OP 8.60 (paragraphs 10 and 11). However, OP 8.60 is less clear on requirements for monitoring and evaluating the actual environmental and social impacts of DPLs and the efficacy of the mitigation measures. OPCS has confirmed to IEG that there is currently no explicit requirement for monitoring or evaluation of environmental and social aspects of DPLs. There is however OPCS Good Practice Note 4 on Environmental and Natural Resource Aspects of Development Policy Lending, which was prepared in 2004, and suggests that with regard to potential effects on the environment, DPLs should have “an explicit monitoring and evaluation strategy to review 54 progress during, as well as beyond, implementation” (OPCS 2004). 74 In the case of the SEM DPL, despite the importance of the mitigation measures proposed in the Program Document and the environmental and social impacts of the policy areas, there is no follow-up discussion in either the ISRs or ICR of any of the potential adverse impacts identified in the Program Document, mitigation measures implemented, or their efficacy. As discussed above, the extent of the implementation of a number of the mitigation measures is questionable. Moreover, the potential adverse impacts generated by the financing of BNDES investments were never acknowledged in any Bank documents, and therefore not mitigated or monitored. 5.39 The Bank supervision team also appears to lack the participation of a qualified social expert for many of the missions, even though one of the objectives was to improve BNDES’s social as well as environmental management systems, and other policies promoted by the loan (such as the increase in protected areas) have important social implications. For example, ISR #3, dated 05/17/2010, states that “The Bank team has held several meetings…related to (i) development of an overall BNDES social policy to be integrated to its environmental policy, (ii) environmental and social traceability requirements related to cattle ranching, …” [emphasis added], but the mission did not include a social development specialist. The World Bank’s ICR also lists an environmental and social specialist as part of the supervision team, but when contacted by IEG, the staff member reported never having participated in supervision. 5.40 The ISRs also regularly neglected to report on the “Progress to Date” of many or even any of the outcome indicators for the SEM DPL, as required. They also did not flag for management attention the lack of progress on the M&E system for the loan (see M&E section below), or take action to ensure that the M&E system promised in the Program Document was actually implemented. 5.41 Quality of Supervision is rated Unsatisfactory. 5.42 In light of the many Quality at Entry problems and the lack of success in addressing the problems or continuing the engagement during Supervision, Overall Bank Performance is rated Unsatisfactory. Borrower Performance Government Performance 5.43 Although the SEM DPL project documents list the Ministry of Finance (MOF) as well as the Ministry of Environment and BNDES as implementing agencies, MOF was primarily responsible for the flow of funds and there were no policy actions related to it. 5.44 Delay in effectiveness: The ICR (pages 8, 16) attributes the delay in effectiveness of more than 14 months for the SEM DPL 1 to the government (World Bank 2011). The long delay undermined part of the rational for a DPL as fast-disbursing budget support. The 74 According to OPCS, there is no requirement in OP 8.60 to monitor and evaluate the social and environmental impacts of DPLs or the efficacy of mitigation measures. 55 reported cause relating to the lack of inclusion of the SEM DPL in the budget indicates inadequate planning on the part of the government, and the issue was not dealt with in a timely manner, with the ICR (page 8) criticizing the “extraordinarily lengthy” senate approval process. 5.45 Cancelation of the series: The ICR (page 16) also lays responsibility for cancelation of SEM DPL 2 on the Brazilian government and the effectiveness delays that it caused, stating that “the 14 month delay of Effectiveness affected the possibility of carrying out the second operation of the SEM DPL series as designed.” The cancelation after the first loan of an environmental DPL series that the Government had committed to—for the second time in a matter of years—combined with the lack of progress on a number of environmental policy areas discussed in Section 4, indicates uncertainty and a degree of lack of commitment by the Government to the SEM DPL reform agenda. As discussed in the Risk to Development Outcome section, this appears to have had an adverse impact. In particular, though the Ministry of Finance is one of the most important sources of BNDES’s capital, there is no evidence that it used its influence to encourage BNDES to improve its environmental and social management system in a timely manner.. 5.46 In comments, the Brazilian government stated that SEM DPL 2 did not materialize due to their decision to prioritize borrowing for subnational units (states and municipalities) rather than for the federal government. That was the same reasoning given in interviews for the cancelation of the earlier Env PRL series. Moreover, the World Bank-Brazil Country Partnership Strategy for 2008-2011, approved in May 2008, almost a year before the SEM DPL, already stated that “the federal government wants the focus of the IBRD program to change to one in which there is…a major lending program with states, on state priorities,” and SEM DPL 1 was approved as a loan to the federal government nearly a year later anyway (World Bank 2008). 5.47 Monitoring and evaluation: According to the design of the SEM DPL as described in the Program Document, “in order to monitor this operation and the entire SEM DPL series, the Ministry of Finance and the Ministry of Environment will use the monitoring system developed through the [Env TAL] project….” The Program Document also promised that the system would be “web-based and open to the public”. As indicated in the ICR, the promised system was never implemented “due to unforeseen bureaucratic delays in contracting the information technology firm.” The Borrower’s comments on the ICR consisted of only three sentences (ICR Annex 5). 5.48 Progress on the Green Protocol: One SEM DPL area where the federal government was supposed to play a central role was on the revision and implementation of the Green Protocol. However, as discussed in Section 4, there was little follow-up on the revised Green Protocol and all indications are that it has now been abandoned. The lack of follow-up is an indication of insufficient federal government ownership. 5.49 Government Performance is rated Moderately Unsatisfactory. Implementing Agency Performance 56 5.50 The implementing agencies responsible for carrying out the policy actions were MMA and BNDES. MMA does not appear to have had much direct engagement with the World Bank on the SEM DPL. But it does appear to have carried out, or is continuing to carry out, a number of the policy actions that appear in the SEM DPL Policy Matrix—some in relation to the previous Env PRL, many of its own accord—though these could not be attributed to the SEM DPL in most cases. 5.51 Lack of progress at BNDES: BNDES was the agency responsible for most of the actions attributable to the SEM DPL, and was the recipient of all of the funds. One of the main justifications for the loan was BNDES’s reported commitment to improving its environmental and social systems. However, in reality BNDES’s progress in this regard has been much slower and more limited than promised. As just one example, one of the pillars of BNDES’s new and improved environmental and social management system was to be the development of 13 sectoral guidelines under the SEM DPL. Five years later, BNDES has still only developed three guidelines. BNDES also remains highly opaque with regard to sharing information on its investments in general and environmental and social management aspects in particular. 5.52 Areas of shortfalls at the Ministry of Environment: MMA also demonstrated less- than-promised progress in some areas under their aegis. Those included mainstreaming action on climate change adaptation and mitigation in public and private sector investments, and improving the environmental licensing process. Issues with the Brazilian environmental licensing process in particular are long-standing (they were also highlighted in the Env PRL Program Document), yet evidence of progress in addressing them is scant. 5.53 Monitoring and evaluation: As noted above, the monitoring and evaluation system promised in the SEM DPL Program Document never materialized. That was supposed to be primarily the responsibility of the MMA, which was also the Implementing Agency for the Env TAL that was supposed to finance the monitoring and evaluation system. 5.54 Absence of consultation and participation: As discussed in the Bank Performance section above, OP8.60 includes requirements on consultations and participation. As part of this, the Bank team is supposed to advise the Borrower “to consult with and engage the participation of key stakeholders in the country….” But the Borrower also has an important role to play. There is no evidence that this was done by BNDES in the context of the SEM DPL. This was especially important because well before the SEM DPL CSO had been highly critical of BNDES’s environmental and social systems, its lack of consultations with civil society, and its lack of transparency. 75 But there is no evidence in the project documents of efforts by BNDES to address these concerns through consultations with stakeholders. BNDES reports no system for monitoring its environmental and social policy implementation. More generally, as mentioned in the Bank Performance section above, the ICR annexes on “Beneficiary Survey Results”, “Stakeholder Workshop Report and Results”, and “Comments of Cofinanciers and Other Partners/Stakeholders” were all left blank. 75 See for example Spink 2013 for a recent history of the mostly fruitless efforts of CSOs to get BNDES to become more transparent and engage with civil society. 57 5.55 Implementing Agency Performance is rated Unsatisfactory. 5.56 Given that Government Performance is rated Moderately Unsatisfactory and Implementing Agency Performance is rated Unsatisfactory, overall Borrower Performance is rated Moderately Unsatisfactory. Monitoring and Evaluation 5.57 There were problems with the design of the M&E system to begin with, since responsibility was assigned to the project management unit from another project under MMA, and there is no information that BNDES was included in the plans. Indicators were flawed and did not reflect outcomes. The monitoring system promised in the Program Document was never implemented. Since no M&E system was implemented, it does not appear that appropriate data were used to inform decision-making. The lack of an M&E system also hindered the measurement and communication of any SEM DPL results to stakeholders. The important potential adverse environmental and social impacts and mitigation measures described in the Program Document appear not to have been monitored. The MMA was designated as the sole agency responsible for monitoring, even though it has no direct responsibility over the monitoring of BNDES’s internal processes and portfolio environmental and social performance, beyond the licensing process. 5.58 The “outcome indicators” for the SEM DPL series in many cases are not sufficient to measure achievement of the operation’s objectives. An important example is in relation to sub-objective A.iii, “improve the effectiveness of environmental and social management systems in BNDES and other financial institutions.” The outcome indicator target was that “100 percent of the projects submitted directly to BNDES screened, approved and monitored according to the new Environmental and Social Institutional Policy”. First, measuring the way projects are screened, approved and monitored is not the same as measuring environmental and social outcomes. Second, while the sub-objective targets “other financial institutions” in addition to BNDES, the indicator only relates to BNDES. Another example is in relation to sub-objective A.i, “improve effectiveness of government agencies in implementing mandated Brazilian environmental and social management procedures.” The original indicator on the number of judicial challenges to environmental licenses by the Public Processor’s Office was not being met (and, as reported in the ICR, was ultimately not met). The indicator was changed to “number of environmental licenses issued at the Federal level”. This is not a good measure of the intended outcomes. For example, a greater number of licenses could be issued if standards were lowered, which would not necessarily lead to improved environmental or social outcomes. No other relevant information on “effectiveness” or outcomes is provided in the ICR. Interviews with staff of both Federal and State Public Prosecutor Offices (PPOs; in Portuguese, Ministério Público, or literally "Public Ministry" ) revealed that in fact the number of lawsuits they file on environmental and social grounds has been increasing, due both to the poor quality of environmental impact assessments (EIAs) for investments, as well as a lack of implementation and monitoring of mitigation measures. The Brazilian PPO is credited with being one of the most important and effective institutions in representing the public interest and promoting the enforcement of environmental legislation (Crawford 2009). The public attorneys interviewed by IEG were critical of the original SEM DPL indicator on environmental licensing. Moreover, these 58 indicators have no direct relationship with the majority of prior actions under sub-objective A.i, in particular those related to the Ministry of Environment or the National Biodiversity Management Institute, since Federal environmental licenses are the responsibility of IBAMA. 76 5.59 Another example of a poorly designed indicator is for sub-objective A.ii on mainstreaming climate change. As discussed above, the indicator for this was based on “planned signed” greenhouse gas emission reductions, firstly from CDM projects. The scale of the CDM compared to the emission reduction needs of a large economy like Brazil’s is marginal. While potentially useful as an approach to piloting incentive programs for emission reduction investments, it was from the beginning limited by its inclusion under the Kyoto Protocol, and even more so by the fact that the dominant buyer of credits generated by the CDM is the European Union Emissions Trading System (EU ETS). The EU announced some time ago that after the end of 2012, only Certified Emission Reductions (CERs) from projects hosted by Least Developed Countries (LDCs) would be eligible for the EU ETS, meaning that Brazil is no longer eligible (Carbon Finance, the World Bank, 2013). Moreover, the indicator used by the team—“planned signed reductions”—has no meaning in the CDM lexicon or project cycle, which has very specific steps and requirements for each step. 77 In interviews with IEG, carbon finance specialists found the use of CDM “planned signed reductions” an odd indicator for national climate change policy implementation. A more meaningful indicator for Brazil’s progress in climate change mitigation would have been, for example, reductions in national emissions, which are inventoried. 78 Finally, “mainstreaming” climate change should include adaptation, which should be a priority for Brazil, as well as mitigation, but there is no indicator for adaptation. 5.60 In a number of cases, the Bank team did not fully and accurately report important monitoring information in the ICR, even when the information was provided as input to the ICR by the implementing agencies. Specific examples are discussed in the Quality of Supervision part of Section 5 above. The lack of candor contributes to a misleading representation of the achievements of the operation. 5.61 Overall, Monitoring and Evaluation is rated Negligible. 6. Lessons 6.1 Certain aspects of the World Bank’s policies and guidance on Development Policy Lending should be clarified. The lack of clarity with regard to certain aspects of the 76 http://www.mma.gov.br/governanca-ambiental/portal-nacional-de-licenciamento- ambiental/licenciamento-ambiental/compet%C3%AAncias-para-o-licenciamento 77 See UNFCCC: http://cdm.unfccc.int/Projects/pac/index.html 78 See Ministry of Science, Technology and Innovation, Brazil, 2013: http://www.mct.gov.br/upd_blob/0226/226578.pdf 59 World Bank’s OP 8.60 and associated guidance notes led to confusion on the part of the World Bank team, the Borrower, and civil society, as described at various points in this evaluation. The following questions governing the design, implementation and evaluation of DPLs should be clarified: • According to the official interpretation, all lending is fungible, and the government can do whatever it wants with the money, so there is no connection between the DPL and the money that the government on-lent to BNDES. This is inconsistent with the common-sense view of the money being transferred to BNDES to complement their disbursements. There is a reputational risk to the Bank in such a case, particularly in relation to the potential environmental and social impacts of the financing. Such impacts are not covered by OP 8.60, which addresses only the potential impacts of policy reforms. • The requirements for monitoring and evaluation of potentially adverse social and environmental impacts of policies supported under DPLs are also not clear. While those impacts and mitigation mechanisms are supposed to be identified in the Program Document, the requirements on how to monitor them after the reforms are put in place and how to evaluate them in the ICR are not clear. 6.2 Particularly for DPLs focused on reforms in a sector—rather than on macro- economic stability—the impacts of the actions supported can often not be adequately perceived within the short timeframe of the loan, making monitoring and evaluation of outcomes difficult. The tight deadline for submitting an ICR after closure of an operation often does not allow for additional outcome evidence to be available and collected. It would be preferable to require that DPL operations wait for a reasonable period either before closing or before producing an ICR—at least one year—in order to allow for adequate monitoring and evaluation. 6.3 Back-loading of reforms in a DPL programmatic series can increase the risk of later loans in the series being canceled without full realization of the objectives of the series. This appears to be the case with the SEM DPL. “Back-loading” refers to the practice of placing reforms that are more meaningful and have greater value-added—but are also often more difficult—later in a DPL series. 79 In this regard, the SEM DPL does not appear to have benefitted from the lessons of the earlier Brazil environmental DPL, the 2004 Env PRL, which was planned as a three-loan programmatic series, but was also canceled after the first loan. 6.4 In future, when attempting to support reforms in state-owned banks like BNDES that finance both public and private-sector investments, it will be important to take a “One World Bank Group” approach. BNDES combines characteristics of the World Bank/IBRD and IFC: it lends both for large public sector infrastructure projects, as 79 For example, a “back-loaded” DPL series would typically have actions related to carrying out a study or drafting a law under loan/tranche 1, approving a law under loan/tranche 2, and implementing the loan under loan 3. For a law to have any impact, it has to be effectively implemented, but this is typically more difficult than drafting it. 60 well as to the private sector. IFC was engaged in high-level discussions with BNDES on improving their environmental and social safeguards—including a visit to BNDES by IFC’s then-Vice President for Business Advisory Services—at the time the Bank began preparation of the SEM DPL. They were encouraging BNDES to adopt the Equator Principles, derived from IFC’s Performance Standards, as other Brazilian banks have. By working together with the IFC, the World Bank could have helped ensure that BNDES adopted environmental and social standards that were considered as good practice internationally, and that were appropriate for both public and private investments. It also would have saved BNDES from having to “re-invent the wheel” by developing its own approach. This is consistent with IEG’s often-repeated assertion that the IFC and the World Bank should work together more to increase development effectiveness. 6.5 The level of government targeted by DPL reforms should be consistent with the outcomes intended and the client country’s institutional structure. Brazil has a federal system of government. Although many national-level laws and policies are approved by the central government, implementation depends in large part on lower levels of government— particularly states and municipalities. That was the case with a number of the policy areas targeted by the SEM DPL, including improving the environmental licensing process, improving water quality, and implementing wastewater and solid waste investments. By limiting itself to federal-level implementing agencies, the SEM DPL had reduced scope to influence the achievement of objectives related to these policy areas. 61 References AIDA. “Belo Monte hydroelectric dam”. Available at http://www.aida-americas.org/en/project/belomonte (last visited 10 September 2013) Amazon Fund (Fundo Amazonia). “Funded Projects”. Available at: http://www.fundoamazonia.gov.br/FundoAmazonia/fam/site_pt/Esquerdo/Projetos_Apoiados/ (last visited August 18, 2014). Amazon Watch. 2010. “Banking Amazon Destruction: BNDES” Available at http://amazonwatch.org/work/bndes (last visited 21 August 2013) Asian Development Bank (ADB). 2012. “Accountability Mechanism: Summary Brochure”. Available at: http://www.adb.org/sites/default/files/pub/2012/accountability-mechanism-summary-brochure.pdf ASIBAMA. 2012. “Carta Aberta do V Congresso de Servidores Federais da Carreira de Especialistas em Meio Ambiente”. Available at: http://www.asibama.org.br/editor/arquivos/Carta%20aberta_final_V%20CONGRESSO%20ASIBAMA%2 0NACIONAL.pdf (last visited 20 September 2013) Banco Central do Brasil.. “Relatorio de Inflacao”. Bacen website. Available at http://www.bcb.gov.br (last visited 20 September 2013) BBC News. 2011. “Brazil grants building permit for Belo Monte Amazon dam”. Available at “http://www.bbc.co.uk/news/world-latin-america-13614684”. (last visited 27 July 2013) BBC News. 2012. “Work to resume on Brazil's Belo Monte dam”. Available at “http://www.bbc.co.uk/news/world-latin- america-19404740”. (last visited 27 September 2013) BICECA. 2011. “Bulletin October 2011”. Available at http://www.bicusa.org/wp-content/uploads/2013/08/Boletin- October-2011.pdf (last visited 15 September 2013) BIZ. 2011. “BNDES increases disbursements to US$ 870 million for Latin America”. Available at http://www.bizusaperu.com/news_details.php?cat=¬i=2458 (last visited 25 September 2013) Bloomberg Business Week. 2013. “Batista Collapse Has Prosecutors at BNDES’s Door: Brazil Credit”. Available at http://www.businessweek.com/news/2013-07-18/batista-collapse-has-prosecutors-at-bndes-s-door-brazil- credit (last visited 26 September 2013) BNDES .2013a. “The Evolution of the BNDES’ Disbursements”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/The_BNDES_in_Numbers(last visited 20 September 2013) BNDES and Renewables: 60 Years, presentation to the IAEA, 2013. Available at: http://www.iea.org/media/workshops/2013/scalingupfinancingtoexpandrenewables/6BNDES%28MariaHel ena%29IEARenewableEnergy%2809.04.13%29.pdf (last visited Sept. 28, 2013). BNDES. “Análise socioambiental de projetos”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/Apoio_Financeiro/Politicas_Transversa is/Politica_Socioambiental/analise_ambiental.html (last visited 24 August 2013). BNDES. “BNDES aprova empréstimo-ponte de R$ 1 bilhão para usina de Belo Monte”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/Sala_de_Imprensa/Noticias/2010/energi a/20101222_BeloMonte.html (last visited 24 August 2013) BNDES. “BNDES e o Protocolo Verde”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Hotsites/Relatorio_Anual_2011/Capitulos/atuacao_i nstitucional/o_bndes_e_protocolo_verde.html (last visited 10 June 2013). 62 BNDES. “Consulta às operações do BNDES”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/BNDES_Transparente/Consulta_as_op eracoes_do_BNDES/ (last visited 24 August 2013) BNDES. “Critérios socioambientais para apoio ao segmento de geração elétrica”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/Apoio_Financeiro/Politicas_Transversa is/Politica_Socioambiental/criterios_socioambientais_geracao.html (last visited 24 August 2013) BNDES. “Diretrizes da Politca Socioambiental”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/Apoio_Financeiro/Politicas_Transversa is/Politica_Socioambiental/diretrizes.html (last visited 24 August 2013). BNDES. “Diretrizes e critérios ambientais para apoio ao setor de açúcar e álcool”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/Apoio_Financeiro/Politicas_Transversa is/Politica_Socioambiental/diretrizes_acucar_alcool.html (last visited 24 August 2013) BNDES. “Diretrizes socioambientais para pecuaria bovina”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/Apoio_Financeiro/Politicas_Transversa is/Politica_Socioambiental/diretrizes_pecuaria_bovina.html (last visited 24 August 2013) BNDES. “Roteiros e Manuais”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Ferramentas_e_Normas/Roteiros_e_Manuais/ (last visited 24 August 2013) BNDES. “Social and Environmental Responsibility”. BNDES website. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/Social_and_Environmental_Responsibi lity/ (last visited 24 August 2013). BNDES. “Socioenvironmental Policy ”. BNDES website. Available at (last visited 24 August 2013).http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/Social_and_Environmental_Res ponsibility/environmental_policy.html (last visited 24 August 2013). BNDES. 2011. “Annual Report 2011”. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/The_BNDES_in_Numbers/Annual_Re port/ (last visited 10 September 2013) BNDES. 2012. “Annual Report 2012”. Available at: http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/The_BNDES_in_Numbers/Annual_Re port/ (last visited August 18, 2014) BNDES. 2013. “BNDES and Renewables” IEA website. Available at http://www.iea.org/media/workshops/2013/scalingupfinancingtoexpandrenewables/6BNDES(MariaHelena) IEARenewableEnergy(09.04.13).pdf (last visited 27 September 2013) BNDES. 2014. “BNDES Mata Atlantica. Available at: http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Areas_de_Atuacao/Meio_Ambiente/Mata_Atlantica/ (last visited August 18, 2014). Central Bank of Brazil. 2014. Resolution 4.327, of April. 25, 2014. Available at: http://www.bcb.gov.br/pre/normativos/res/2014/pdf/res_4327_v1_O.pdf (last visited August 18, 2014). Chico Mendes Institute for Biodiversity Conservation (ICMBio). http://www.icmbio.gov.br/portal/quem-somos/o- instituto.html (last visited August 14, 2014) Colby, S. 2012. “Explaining the BNDES: what it is, what it does and how it works”, Centro Brasileiro de Relacoes Internacionais, Volume 3. Available at http://www.cebri.org/midia/documentos/bndes.pdf . (last visited 20 September 2013) Crawford, C. 2009. “A Review of Lesley Mcallister’s Making Law Matter: Environmental Protection and Legal Institutions in Brazil”. The George Washington International Law Review. Vol. 40, pages 619-647. Available at: http://docs.law.gwu.edu/stdg/gwilr/pdfs/40-3/40-3-2-crawford.pdf (last visited January 15, 2014). 63 DW. 2013. “Construção de Belo Monte já rende 19 processos na Justiça”. Available at http://www.dw.de/constru%C3%A7%C3%A3o-de-belo-monte-j%C3%A1-rende-19-processos-na- justi%C3%A7a/a-17087946 (last visited 20 September 2013) El Saifi, Samira; Sampaio Dagnino, Ricardo de. 2011. “Grandes projetos de desenvolvimento e implicações sobre as populações locais: o caso da usina de Belo Monte e a população de Altamira, Pará”. IPEA. Available at http://www.ipea.gov.br/code2011/chamada2011/pdf/area7/area7-artigo19.pdf (last visited 20 September 2013) The Economist. 2013a. “The rights and wrongs of Belo Monte”. Available at http://www.economist.com/news/americas/21577073-having-spent-heavily-make-worlds-third-biggest- hydroelectric-project-greener-brazil (last visited 20 September 2013) The Economist. 2013b. “Trees of Knowledge”. Available at http://www.economist.com/news/special- report/21585096-how-brazil-using-education-technology-and-politics-save-its-rainforest-trees (last visited 20 September 2013) The Economist. 2013c. “Brazil’s development banks: A ripple begets a flood”. October 19, 2013. Available at: http://www.economist.com/news/finance-and-economics/21588133-politically-inspired-surge-lending- weakening-state-owned-banks-latin (last visited January 9, 2014)Febraban. 2009. “Protocolo Verde”. Available at http://www.febraban.org.br/protocoloverde/PROTOCOLO%20DE%20INTEN%C3%87%C3%95ES.PDF (last visited 25 August 2013) Equator Principles. “The Equator Principles III – 2013”. Equator Principles website. Available at http://www.equator-principles.com/resources/equator_principles_III.pdf (last visited 24 August 2013). European Bank for Reconstruction and Development (EBRD). 2011. “Project Complaint Mechanism: A User’s Guide”. Available at: http://www.ebrd.com/downloads/research/guides/pcm_eng.pdf Fearnside, P. 2012. “Belo Monte Dam: A spearhead for Brazil’s dam-building attack on Amazonia?”. GWF Discussion Paper 1210, Global Water Forum, Canberra, Australia. Available at: http://www.globalwaterforum.org/wp-content/uploads/2012/04/Belo-Monte-Dam-A-spearhead-for-Brazils- dam-building-attack-on-Amazonia_-GWF-1210.pdf (last visited January 16, 2014) FEBRABAN. 2010. “Política Ambiental e Instrumentos de Colaboração Financeira para Meio Ambiente”. Available athttp://www.febraban.org.br/7Rof7SWg6qmyvwJcFwF7I0aSDf9jyV/sitefebraban/Raphael%20Stein%20T reinamento%20BID-Febraban%20nov10.pdf (last visited 18 August 2013) Financial Times. 2010. “Brazil accelerates investment in Africa”. By Richard Lapper. February 9, 2010. Available at: http://www.ft.com/intl/cms/s/0/e9550408-1519-11df-ad58-00144feab49a.html#axzz2g1RBGLq5 . (last visited September 26, 2013). Financial Times. 2011. “Pristine forest is finite resource”. By John Paul Rathbone. December 1, 2011. Available at: http://www.ft.com/intl/cms/s/0/c4a347f8-110e-11e1-a95c-00144feabdc0.html#axzz2g1RBGLq5 (last visited September 26, 2013). Financial Times. 2012. “Brazil: A bank too big to be beautiful”. By Samantha Pearson. September 23, 2012. Available at: http://www.ft.com/intl/cms/s/0/983f1bca-0234-11e2-b41f- 00144feabdc0.html#axzz2g1RBGLq5 . (last visited September 26, 2013). Financial Times. 2013a. “Brazil: The creaking champions”. By Joe Leahy. April 21, 2013. Available at: http://www.ft.com/intl/cms/s/0/8fbc8b50-a391-11e2-8f9c-00144feabdc0.html#axzz2g1RBGLq5 . (last visited September 26, 2013). Financial Times. 2013b. “Questions surround Brazil bank in Carrefour deal”. By Joe Leahy. July 13, 2011. Available at: http://www.ft.com/intl/cms/s/0/46fda736-ad71-11e0-bc4f- 00144feabdc0.html#axzz2g1RBGLq5 . (last visited September 26, 2013). Financial Times. 2013c. “Brazil: showing Africa some love”. By Samantha Pearson. May 29, 2013. Available at: http://blogs.ft.com/beyond-brics/2013/05/29/brazil-showing-africa-some-love/ (last visited December 12, 2013). 64 Financial Times. 2013d. “Brazil’s development bank heads for Africa”. By Adam Green. December 5, 2013. Available at: http://blogs.ft.com/beyond-brics/2013/12/05/brazils-development-bank-heads-for-africa/ (last visited December 12, 2013). Financial Times. 2013e. “BNDES, Caixa: robbing Peter to pay Paul”. By Joe Leahy. March 21, 2013. Available at: http://blogs.ft.com/beyond-brics/2013/03/21/bndes-caixa-robbing-peter-to-pay-paul/#axzz2pjtvpLgG (last visited January 9, 2014). Financial Times. 2013f. “Brazil slows down: The time has come for Dilma Rousseff to change tack”. December 4, 2013. Available at: http://www.ft.com/intl/cms/s/0/fe25491c-5cdf-11e3-a558- 00144feabdc0.html?siteedition=intl#axzz2qbG0uLQR (last visited January 16, 2014). Folha de Sao Paulo. 2008a. “BNDES terá R$ 5 bilhões do Banco Mundial para crédito a empresas”. Available at http://www1.folha.uol.com.br/folha/dinheiro/ult91u469808.shtml. (last visited January 11, 2014) Folha de Sao Paulo. 2008b. “Lula assina medida que repassa US$ 2 bilhões do Bird ao BNDES”. Available at http://www1.folha.uol.com.br/folha/dinheiro/ult91u477337.shtml (last visited January 11, 2014) Folha de Sao Paulo. 2013. “Obra de Belo Monte pode atrasar por problemas com Ibama”. Available at http://www1.folha.uol.com.br/mercado/2013/07/1318408-obra-de-belo-monte-pode-atrasar-por-problemas- com-ibama.shtml (last visited 20 September 2013) Folha de Sao Paulo. 2014. “Disputa entre banco e Controladoria espera parecer da AGU”. January 19, 2014. Available at http://www.estadao.com.br/noticias/impresso,disputa-entre-banco-e-controladoria-espera- parecer-da-agu--,1120268,0.htm (last visited January 22, 2014) Fonseca, B. and J. Mota. 2013. “The Footprints of BNDES in the Amazon”. Agencia Publica and O Eco. October 15, 2013. Available at: http://www.apublica.org/2013/10/investimentos-bndes-em-infraestrutura-na- amazonia-caso-da-hidreletrica-de-dardanelos/ (last visited January 17, 2014). Forest Service of Brazil. http://www.florestal.gov.br/menu-horizontal-de-internet/institucional/servico-florestal- brasileiro (last visited August 14, 2014). G1. 2013. “Bancos públicos emprestam mais do que privados em 2012”. Available at http://g1.globo.com/economia/seu-dinheiro/noticia/2013/01/bancos-publicos-emprestam-quase-4-vezes- mais-do-que-privados-em-2012.html (last visited 20 September 2013) Global Environment Facility Evaluation Office 2012. “Avaliação de Portfólio de Projetos do GEF:Brasil (1991– 2011), volume 2” available at http://www.thegef.org/gef/sites/thegef.org/files/documents/GEF%20Brazil%20CPE%20- %20Final%20Report%20-%20Volume%202_0.pdf September 2012. Greenpeace. 2009. “Slaughtering the Amazon”. Available at http://www.greenpeace.org/international/en/publications/reports/slaughtering-the-amazon/ (last visited June 5, 2014) The Guardian. 2011. “Lehman Brothers: financially and morally bankrupt” available at http://www.theguardian.com/commentisfree/cifamerica/2011/dec/12/lehman-brothers-bankrupt (last visited 25 September 2013) The Guardian. 2011. “What's behind the Belo Monte dam” available at http://www.theguardian.com/commentisfree/cifamerica/2011/feb/15/brazil-energy (last visited 25 September 2013) The Guardian. 2012. “Belo Monte dam construction halted by Brazilian court” available at http://www.theguardian.com/world/2012/aug/16/belo-monte-dam-construction-suspended (last visited 25 September 2013) The Guardian. 2013 “Brazil's hydroelectric dam boom is bringing tensions as well as energy”. February 12, 2013. Available at: http://www.theguardian.com/world/2013/feb/12/brazil-hydroelectric-jirau-dam Guia dos Bancos Responsaveis (GBR). “Casos Polemicos” . Available at http://www.gbr.org.br/casos-polemicos- integra-2.php (last visited 10 June 2013) 65 Hall, A. (ed.), Amazonia at the Crossroads: the challenge of sustainable development, Institute of Latin American Studies, School of Advanced Studies, University of London, 2000: 99-114. Hall, A. ‘Environment and Development in Brazilian Amazonia: From Protectionism to Productive Conservation’ Hall. A. 1997 Sustaining Amazonia: Grassroots Action for Productive Conservation, Manchester University Press, Manchester. IBAMA. 2013a. “Term of Reference No. 01”. National Environment Program – NEP (Phase 2). February 2013. Available at: http://www.ibama.gov.br/phocadownload/ibama/tr%20ingles%20mi%2001-2013%20- %20laf%20aia.pdf (last visited June 4, 2014) IBAMA. 2013b. “Term of Reference No. 02”. National Environment Program – NEP (Phase 2). February 2013. Available at: http://www.ibama.gov.br/phocadownload/ibama/tr%20ingles%20mi%2002-2013%20- %20mapeamento%20de%20competencias.pdf (last visited June 4, 2014) IBAMA. 2013. “Análise do 3° Relatório Consolidado de Andamento do PBA da UHE Belo Monte”. PAR. 004933/2013. Available at: http://www.socioambiental.org/sites/blog.socioambiental.org/files/nsa/arquivos/3_parecer_ibama.pdf (last visited October 3, 2013) IBAMA. 2011. “Ibama autoriza instalação de Belo Monte”. Available at http://www.ibama.gov.br/publicadas/ibama-autoriza-a-instalacao-da-usina-de-belo-monte (last visited 20 June 2013) ILO 2012, Report of the Committee of Experts on the Application of Conventions and Recommendations, International Labour Conference, 101st Session, ILC.101/III1A. Available at: http://www.ilo.org/wcmsp5/groups/public/---ed_norm/--- relconf/documents/meetingdocument/wcms_174843.pdf Independent Evaluation Group (IEG). 2011. The World Bank Group’s Response to the Global Economic Crisis, Phase 1. Washington, DC: The World Bank Group. Independent Evaluation Group (IEG). 2012. The World Bank Group’s Response to the Global Economic Crisis, Phase II. Washington, DC: The World Bank Group. Independent Evaluation Group (IEG). 2013. Brazil Country Program Evaluation: FY2004-11. Washington, DC: The World Bank. Draft. International Monetary Fund (IMF). “Guidelines for Fiscal Adjustment”. Available at http://www.imf.org/external/pubs/ft/pam/pam49/pam4902.htm (last visited 20 September 2013) International Partnership on Mitigation and MRV. “Brazil: Voluntary NAMAs to achieve a GHG emission reduction between 36.1% and 38.9% by 2020”. Available at: http://mitigationpartnership.net/brazil-voluntary-namas- achieve-ghg-emission-reduction-between-361-and-389-2020-0 (last visited 29 August 2014) Infolatam. 2012. “BNDES lidera recorde de credito em 2012”. Available at http://www.infolatam.com.br/2013/01/23/bndes-libera-valor-recorde-de-credito-em-2012/ (last visited 24 September 2013) Instituto de Pesquisa Ambiental da Amazonia. “Qual o objetivo do Plano e da Politica Nacional de Mudanca do Clima criados no Brasil?” Available at http://www.ipam.org.br/saiba-mais/abc/mudancaspergunta/O-que- sao-Servicos-Ambientais-possivel-compensar-economicamente-a-prestacao-destes-servicos-/41/21# (last visited 10 September 2013) Instituto de Pesquisa Econômica Aplicada (Ipea). 2012. “O Bndes e a “Copa da Fifa” no Brasil” Available at http://www.ipea.gov.br/desafios/index.php?option=com_content&view=article&id=2719:catid=28&Itemid =23 (last visited 21 August 2013) Instituto Socio Ambiental (ISA). 2013. “Análise do atendimento de condicionantes da Licença de Instalação da Usina de Belo Monte”. Available at http://www.socioambiental.org/sites/blog.socioambiental.org/files/nsa/arquivos/quadro_condicioanantes_2 013_isa.pdf (last visited 21 August 2013) 66 Instituto Socio Ambiental (ISA). 2013. “Ibama afirma que o cumprimento de condicionantes de Belo Monte só piorou”. Available at http://www.socioambiental.org/pt-br/noticias-socioambientais/ibama-afirma-que-o- cumprimento-de-condicionantes-de-belo-monte-so-piorou (last visited 21 September 2013) Instituto Socio Ambiental (ISA). 2013. “Responsabilidade socioambiental do BNDES em Belo Monte é “para inglês ver””. Available at http://www.socioambiental.org/pt-br/blog/blog-do-xingu/responsabilidade- socioambiental-do-bndes-em-belo-monte-e-para-ingles-ver (last visited 21 August 2013) Inter-American Development Bank (IDB). 2010. “Policy Establishing the Independent Consultation and Investigation Mechanism”. Available at: http://idbdocs.iadb.org/wsdocs/getdocument.aspx?docnum=35074768 (last visited September 21, 2013). International Center for Trade and Sustainable Development. 2008. “Plano Nacional sobre Mudanca do Clima recebe criticas de ONGs ambientais” Available at http://ictsd.org/i/news/pontesquinzenal/33092/. (last visited 25 September 2013) International Labour Organization (ILO). 2012. “Report of the Committee of Experts on the Application of Conventions and Recommendations”. Available at http://www.ilo.org/wcmsp5/groups/public/---ed_norm/-- -relconf/documents/meetingdocument/wcms_174843.pdf (last visited 27 August 2013). Law 11.943, May, 2009. Available at https://www.planalto.gov.br/ccivil_03/_ato2007-2010/2009/lei/l11943.htm Le Monde. 2013. “Au Brésil, chantier de titan à Belo Monte” Available at : http://www.lemonde.fr/international/article/2013/05/21/bresil-chantier-de-titan-a-belo- monte_3414793_3210.html (last visited 21 September 2013) Leitao, Miriam. 2011. “Belo Monte: licença parcial não existe”. O Globo Blogs. Available at http://oglobo.globo.com/economia/miriam/posts/2011/01/27/belo-monte-licenca-parcial-nao-existe- 359392.asp (last visited 10 September 2013). Library of the European Parliament. 2013. “Brazil’s Belo Monte Dam project: Financial impact, indigenous peoples’ rights & the environment”. Library Briefing. June 20, 2013. Available at: http://www.europarl.europa.eu/RegData/bibliotheque/briefing/2013/130579/LDM_BRI%282013%2913057 9_REV1_EN.pdf (last visited January 16, 2014). Lopes Pinto, João Roberto (editor). 2012. Ambientalização dos Bancos e Financeirização da Natureza: Um debate sobre a política ambiental do BNDES e a responsabilização das Instituições Financeiras. Brasilia: Rede Brasil sobre Instituições Financeiras Multilaterais. Available at: http://www.scribd.com/doc/140772638/ambientalizacaodosbancosefinanceirizacaodanatureza Lopez-Calva, Luis F. and Nora Lustig. 2010. “Explaining the Decline in Inequality in Latin America: Technological Change, Educational Upgrading, and Democracy”. Brookings. Available at: http://www.brookings.edu/~/media/press/books/2010/declininginequalityinlatinamerica/declininginequality inlatinamerica_chapter.pdf Merriam-Webster. 2014. Definition of “earmark”. Britannica Digital Learning. Unabridged. Available at: http://unabridged.merriam-webster.com/unabridged/earmark (last visited January 11, 2014) Ministério da Agricultura Pecuária e Abastecimento. 2013. “Estatisticas e Dados Basicos de Economia”. Available at http://www.agricultura.gov.br/arq_editor/Pasta%20de%20Junho%20-%202013.pdf (last visited 20 September 2013) Ministerio da Ciencia e Tecnologia. 2013. “Estimativas Anuais de Emissoes de Gases de Efeito Estufa”. Available at http://www.mct.gov.br/upd_blob/0226/226578.pdf (last visited September 21 2013) Ministerio do Meio Ambiente (Ministry of Environment). “Planos Setoriais de Mitigação e Adaptação”. Available at: http://www.mma.gov.br/clima/politica-nacional-sobre-mudanca-do-clima/planos-setoriais-de-mitigacao- e-adaptacao (last visited 15 September 2014) Ministerio do Meio Ambiente (Ministry of Environment). “ Plano Nacional Sobre Mudancas do Clima”. Available at http://www.mma.gov.br/clima/politica-nacional-sobre-mudanca-do-clima/plano-nacional-sobre- mudanca-do-clima (last visited 15 September 2014) 67 Ministerio do Meio Ambiente (Ministry of Environment). “Competencias para o Licensiamento”. Available at http://www.mma.gov.br/governanca-ambiental/portal-nacional-de-licenciamento-ambiental/licenciamento- ambiental/compet%C3%AAncias-para-o-licenciamento(last visited 24 September 2013) Ministerio do Meio Ambiente (Ministry of Environment). “Politica Nacional Sobre Mudanca Climatica”. Available at http://www.mma.gov.br/clima/politica-nacional-sobre-mudanca-do-clima (last visited 18 September 2013) Ministério do Meio Ambiente (Ministry of Environment).2008. “Protocolo de Intenções pela Responsabilidade Socioambiental”. Available at http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Ferramentas_e_Normas/Roteiros_e_Manuais/(last visited 10 August 2013) Ministério Público Federal (Federal Public Prosecutor). 2010. “Análise de pedido de financiamento para AHE Belo Monte”. Procuradoria da Republica do Pará. Available at http://www.xinguvivo.org.br/wp- content/uploads/2010/10/Oficio-2-BNDES.pdf (Last visit 20 August 2013) Ministério Público Federal (Federal Public Prosecutor). 2010. “MPF questiona BNDES sobre financiamento de Belo Monte”. Procuradoria da Republica do Pará. Available at http://www.prpa.mpf.mp.br/news/2010/noticias/mpf-questiona-bndes-sobre-financiamento-de-belo-monte (Last visit 20 August 2013) Ministério Público Federal e Ministério Público no Estado de Mato Grosso. 2012. Ação Civil Pública no. 0003947- 44.2012.4.01.3600. Available at: http://www.aneel.gov.br/arquivos/PDF/SUSPENSAO_UHE_TELES_PIRES_uniao_e_aneel_v3_bip_junho _2012.pdf (last visited December 16, 2013). Ministério Público Federal (Federal Public Prosecutor). 2013. “Processos judiciais do caso Belo Monte são publicados na íntegra pelo MPF”. Procuradoria da Republica do Pará. Available at http://www.prpa.mpf.mp.br/news/2013/processos-judiciais-do-caso-belo-monte-sao-publicados-na-integra- pelo-mpf (Last visit 20 August 2013) Ministério Público Federal (Federal Public Prosecutor). 2013b. http://noticias.pgr.mpf.mp.br/noticias/noticias-do- site/copy_of_indios-e-minorias/norte-energia-e-obrigada-a-paralisar-obras-de-belo-monte-mais-uma- vez/?searchterm=Belo%20Monte (last visited September 17, 2014). Ministério Público Federal (Federal Public Prosecutor). 2014. http://noticias.pgr.mpf.mp.br/noticias/noticias-do- site/copy_of_meio-ambiente-e-patrimonio-cultural/tribunal-nega-recurso-da-norte-energia-e-confirma- prazo-de-90-dias-para-novos-estudos-de-belo-monte/?searchterm=Belo%20Monte (last visited September 17, 2014). Ministry of External Relations and Ministry of Environment. “Brazil’s Nationally Appropriate Mitigation Actions”. Available at: http://www.mmechanisms.org/document/NAMA/NAMA_LCA15_brazil_EN.pdf (last visited August 29, 2014) Ministry of Science, Technology and Innovation, Brazil, “Estimativas anuais de emissões de gases de efeito estufa no Brasil”, 2013: http://www.mct.gov.br/upd_blob/0226/226578.pdf Nature 2012, “Brazil’s fund for low-carbon agriculture lies fallow”, http://www.nature.com/news/brazil-s-fund-for- low-carbon-agriculture-lies-fallow-1.11111 Nature 2013, “Brazil reports sharp drop in greenhouse emissions”: http://www.nature.com/news/brazil-reports- sharp-drop-in-greenhouse-emissions-1.13121 Nepstad, Daniel and others. 2014. “Slowing Amazon deforestation through public policy and interventions in beef and soy supply chains”. Science 344, 1118. Available at: http://www.sciencemag.org/content/344/6188/1118.full.html (last visited June 8, 2014) New York Times. 2012. “The Dam Boom in the Amazon”. Available at http://www.nytimes.com/2012/07/01/opinion/sunday/the-dam-boom-in-the-amazon.html?_r=3& New York Times. 2010. “Amazon Dam Project Pits Economic Benefit Against Protection of Indigenous Lands”. By Alexei Barrionuevo. Published April 16, 2010. Available at: http://www.nytimes.com/2010/04/17/world/americas/17brazil.html?_r=0 (last visited 21 August 2013) 68 Nolte, C., A Agrawal, K Silvius, and B. Soares-Filho. 2013. “Governance regime and location influence avoided deforestation success of protected areas in the Brazilian Amazon”. PNAS Early Edition. Available at: http://www.pnas.org/content/110/13/4956 (last visited August 20, 2013). Norad. 2011. “Real-Time Evaluation of Norway’s International Climate and Forest Initiative Contributions to National REDD+ Processes 2007-2010.Country Report: Brazil”, Evaluation Report 13/2010. Available at: http://www.oecd.org/derec/norway/48086441.pdf Norte Energia. 2014. “Belo Monte”. From the company’s website. Available at: http://norteenergiasa.com.br/site/ingles/belo-monte/ (last visited January 16, 2014). O Estadao. 2013. “Belo Monte leva o maior financiamento da história do BNDES: R$ 22,5 bilhões”. Available at http://www.estadao.com.br/noticias/impresso,belo-monte-leva-o-maior-financiamento-da-historia-do- bndes-r-225-bilhoes-,965721,0.htm (last visited 21 August 2013) O Globo. 2013. “BNDES não cumpre exigência ambiental no crédito a Belo Monte”. Available at http://oglobo.globo.com/economia/bndes-nao-cumpre-exigencia-ambiental-no-credito-belo-monte-8295084 (last visited 21 August 2013) OECD. 2013. OECD Economic Surveys: Brazil 2013. OECD Publishing. Available at: http://www.oecd- ilibrary.org/economics/oecd-economic-surveys-brazil-2013_eco_surveys-bra-2013-en (last visited June 4, 2014) OPCS. 2004 (?). “Good Practice Note 4 – Environmental and Natural Resource Aspects of Development Policy Lending”. OPCS. The World Bank. (undated) OPCS 2008. “Good Practice Note: Using Poverty and Social Impact Analysis to Support Development Policy Operations”. OPCS. The World Bank. August 2008 OPCS. 2010. “Good Practice Note: Addressing Financial Management Issues in Development Policy Lending”. Financial Management Sector Board. The World Bank. October 2010. OPCS. 2011. “Good Practice Note for Development Policy Lending: Designing Development Policy Operations”. OPCS. The World Bank. January 2011. OPCS. 2011, “Good Practice Note for Development Policy Lending: Results in Development Policy Lending”. The World Bank. January 2011. OPCS 2012. “2012 Development Policy Retrospective”. The World Bank. Organization of American States (OEA). 2011. Inter-American Commission on Human Rights (IACHR): Precautionary Measure MC-382-10. Available at http://www.oas.org/en/iachr/decisions/precautionary.asp (last visited 21 August 2013) Oxford English Dictionary. 2014. Definition of “earmark”. Usage example from 1928 Manch. Guardian Weekly 31 Aug. 179/2. Digital version (last visited January 11, 2014). Presidency of the Republic 2008. Civil Affairs Ministry, Sub-office of Legal Affairs. Interim Measure Number 450. December 9, 2008. Available at: http://www.planalto.gov.br/ccivil_03/_ato2007-2010/2008/Mpv/450.htm (last visited January 12, 2014). Presidency of the Republic 2009, Civil Affairs Ministry, Sub-office of Legal Affairs, Law No. 11.943, Article 15. 28 May 2009. Available at: http://www.planalto.gov.br/ccivil_03/_ato2007-2010/2009/Lei/L11943.htm (last visited August 12, 2014). Previ. 2011. “O BNDES e o Meio Ambiente: DiretrizesSocioambientais e Instrumentos de ApoioFinanceiro”. Available at http://www.previ.com.br/pls/portal/docs/PAGE/PREVI_DESENV/INVESTIMENTOS/GOVERNANCA/G OV_CONSELHEIROS/ENCONTRO_DE_CONSELHEIROS_2010/ENCONTRO_2009_APRESENTAC OES/14H15_SERGIO%20WEGUELIN.PDF (last visited 24 July 2013) R.Guimarães.1991. The Ecopolitics of Development in the Third World: Politics and the Environment in Brazil, Lynne Rienner, Boulder, Colorado and London. 69 Rainforest Foundation UK. “The Brazilian National Development Bank”. Available at: http://www.rainforestfoundationuk.org/files/The%20Brazilian%20National%20Development%20Bank%20 Fact%20Sheet.pdf (last visited October 3, 2013) Ramos, Adriana, and Biviany Rojas Garzon. 2013. “The lack of an effective safeguards policy at the Brazilian Development Bank”. Forest Peoples Programme. Available at http://www.forestpeoples.org/topics/other- institutions/news/2013/04/lack-effective-safeguards-policy-brazilian-development-bank (last visit 14 August 2013) Reporter Brasil. 2011. “BNDES and its environmental policy: A critique from the perspective of organised civil society”. São Paulo. Available at: http://reporterbrasil.org.br/documentos/BNDES_English.pdf Reuters. 2012. “Brazil's BNDES approves $10.8 bln loan for Amazon Belo Monte dam”. Available at http://www.reuters.com/article/2012/11/26/bndes-belomonte-idUSL1E8MQ76I20121126 (last visited 21 August 2013) Senate of Brazil. 2008. Available at http://www.senado.gov.br/atividade/materia/detalhes.asp?p_cod_mate=88646 (last visited 20 September 2013). Soares-Filho, B., R. Rajão, M. Macedo, A. Carneiro, W. Costa, M. Coe, H. Rodrigues, A. Alencar. 2014. “Cracking Brazil’s Forest Code”. Science 25 April 2014: Vol. 344 no. 6182 pp. 363-364. Available at: http://www.sciencemag.org/content/344/6182/363.full.pdf (last visited August 19, 2014). Soares-Filho, B. et al. 2006. “Modelling conservation in the Amazon Basin”, Nature, 440, 23 Spink, P. 2013. “The Art of Getting in the Way: Five Years of the BNDES Platform”. Social Science Research Network, and International Budget Partnership. Study No. 13. June 2013. Available at: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2324454 (last visited August 7, 2014) UNEP RISOE. “CDM/JI Pipeline Analysis and Database” Available at http://cdmpipeline.org/ (last visited 15 September 2013) UNFCCC. NAMA Country Page: Brazil. Available at: http://www4.unfccc.int/sites/nama/SitePages/Country.aspx?CountryId=24 (last visited 15 September 2014) UNFCCC. “CDM Project Activity Cycle”. Available at http://cdm.unfccc.int/Projects/pac/index.html (last visited 20 September 2013) Veja. “Notícias com a tag "belo monte". Available at http://veja.abril.com.br/tag/belo-monte (last visited 21 August 2013) Verdum, Ricardo. 2013. “Systematized Information on BNDES Funded Projects in the Amazon Region with Implications for Deforestation”. Derecho, Ambiente y Recursos Naturales (DAR). Available at http://www.vigilamazonia.com/uploads/files/3be77d3f622d8b65fc1e3d379e406e6e.pdf (last visited 10 September 2013) Widmer, Roland. 2012. “The Brazilian Safeguard Regime, Its Application, and Recommendations for the Future”. OneAdvisory, funded by the World Resources Institute. Available at: http://www.internationalrivers.org/files/attached-files/case_study_hydropower_development_in_brazil.pdf The Wall Street Journal. 2011. “Indian Protests Spur Bolivia Crisis”. The Wall Street Journal Online. October 10, 2011. World Bank 2008, “Environmental Licensing for Hydroelectric Projects in Brazil: A Contribution to the Debate”, Summary Report, World Bank Report No. 40995-BR, March 28, 2008. World Bank 2009, “Program Document for a Proposed First Programmatic Development Policy Loan for Sustainable Environmental Management in the Amount of US$1.3 billion to the Federative Republic of Brazil,” Report No. 47215-BR, The World Bank. World Bank. 2009. “World Bank Group Response to Financial Crisis”. Available at http://www.worldbank.org/financialcrisis/pdf/WBGResponse-VFF.pdf (last visited 24 September 2013) World Bank. 2011. “Implementation Completion and Results Report on a First Programmatic Development Policy Loan for Sustainable Environmental Management in the Amount of US$1.3 billion to the Federative Republic of Brazil.” Report No. ICR00001982. The World Bank. 70 World Bank. 2011b. “Implementation Completion and Results Report: Brazil Environmental Sustainability Agenda Technical Assistance Project”. Report No. ICR00001997. The World Bank. September 23, 2011. World Bank. 2012. OP 8.60 – Development Policy Lending. World Bank Operational Manual. Revised March 2012. WWF. 2011. “Serious risks expected to the Belo Monte hydroelectric dam project due to climate change impacts in the Amazon” Available at http://www.wwf.org.br/?28983/(last visited 23 September 2013) Xingu Vivo. 2011. “Ibama libera Licença de Instalação para Belo Monte às vésperas do Dia Mundial do Meio Ambiente”. Available at http://www.xinguvivo.org.br/2011/06/01/ibama-libera-licenca-de-instalacao-para- belo-monte-as-vesperas-do-dia-mundial-do-meio-ambiente/ (last visited 10 September 2013) Xingu Vivo. 2011. “Com Belo Monte, violência sexual contra criança e adolescente cresce 138%”. Available at: http://www.xinguvivo.org.br/2011/10/11/com-belo-monte-violencia-sexual-contra-crianca-e-adolescente- cresce-138/ . (last visited September 26, 2013) Yale. 2012. “Brazil’s Belo Monte Dam: A Conflict in the Amazon”. Available at http://e360.yale.edu/feature/video_belo_monte_dam_conflict_in_the_brazilian_amazon/2536/. (last visited 20 September 2013) 71 ANNEX A Annex A: Basic Data Sheet FIRST PROGRAMMATIC DEVELOPMENT POLICY LOAN FOR SUSTAINABLE ENVIRONMENTAL MANAGEMENT (LOAN IBRD-76600) Key Project Data (amounts in US$ million) Appraisal Actual or Actual as % of estimate current estimate appraisal estimate Total project costs 1,300.0 1,300.0 100 Loan amount 1,300.0 1,300.0 100 Cofinancing Cancellation Cumulative Estimated and Actual Disbursements FY09 FY10 FY11 Appraisal estimate (US$M) 800,000.0 1,300.0 1,300.0 Actual (US$M) 797,750.0 1,297,750.0 Actual as % of appraisal 61.3 99.8 Date of final disbursement: 12/08/2010 Project Dates Original Actual Initiating memorandum 11/11/2008 Negotiations 12/18/2008 Board approval 03/05/2009 Signing Effectiveness 05/12/2009 06/21/2010 Closing date 12/31/2010 12/31/2010 Staff Inputs (staff weeks) Staff Time and Cost (Bank budget only) Stage US$ 000s (including travel and Staff Weeks (number) consultant costs) Lending FY05 1 0.47 FY06 1 10.13 FY07 1 2.37 FY08 15 109.45 FY09 38 344.74 ANNEX A 72 Total: 55 467.16 Supervision/ICR FY09 3 42.10 FY10 15 132.27 FY11 12 196.60 Total: 30 370.97 Task Team Members Other Project Data Name Title (at time of appraisal and Unit Responsibility/ closure, respectively) Specialty Lending Marcos T. Abicalil Sr Water & Sanitation Spec. LCSUW Alexandre Moreira Baltar Water Resources Mgmt. Spec. LCSEN Garo J. Batmanian Sr Environmental Spec. LCSEN Regis Thomas Cunningham Sr Financial Management Spec. LCSFM Estanislao Gacitua-Mario Lead Social Development Spec. LCSSO Augusto Juca Sr. Energy Specialist LCSSD Mark R. Lundell Sector Leader LCSSD Isabella Micali Drossos Senior Counsel LEGLA Alberto Ninio Chief Counsel LEGEN Miguel-Santiago da Silva Senior Finance Officer CTRFC Oliveira Paula Silva Pedreira de Freitas Operations Analyst LCSEN Luis R. Prada Villalobos Senior Procurement Specialist MNAPR Paul Procee Senior Infrastructure Spec. EASCS Cristina Oliveira Roriz Operations Analyst LCSRF Jennifer J. Sara Sector Manager EASVS Eduardo Martin Urdapilleta Sr Financial Economist LCSPF Supervision/ICR Marcos T. Abicalil Sr Water & Sanitation Spec. LCSUW Alexandre Moreira Baltar Water Resources Mgmt. Spec. LCSEN Garo J. Batmanian Sr Environmental Spec. LCSEN Fernando Andres Blanco Cossio Senior Economist AFTP4 Estanislao Gacitua-Mario Lead Social Development Spec. LCSSO Daniella Ziller Arruda Program Assistant LCC5C Karagiannis Alberto Ninio Chief Counsel LEGEN Paula Silva Pedreira de Freitas Operations Analyst LCSEN Paul Procee Senior Infrastructure Spec. EASCS Cristina Oliveira Roriz Operations Analyst LCSRF Jennifer J. Sara Sector Manager EASVS Eric Shayer Sr Environmental Spec. CESI1 Barbara Farinelli Economist LCSSD Abdoulaye Sy Economist LCSSD Borrower/Executing Agency: Follow-on Operations Operation Credit no. Amount Board date (US$ million) 73 ANNEX B Annex B: SEM DPL Development Policy Matrix Policy Area Policy Actions – DPL I: First Policy Actions – DPL I: Second DPL II - Triggers Outcome Indicators for Tranche Prior Actions Tranche Release Conditions Program for 2011 Sub-objective A: Improving the Overall Brazilian Environmental Management System Improve effectiveness of · MMA and its Formal selection and Improve the government agencies in affiliates restructured to hiring processes to fill environmental licensing implementing mandated support the implementation of 600 vacancies to staff process using as a proxy the Brazilian environmental environmental policies, ICMBio and AFB decrease by 20 percent the and social management processing of environmental concluded. number of judicially procedures licenses and enforcement of challenged licenses by Public environmental regulations Prosecutor`s Office compared through: i) the establishment with the nenumber of of a new institutional structure environmental licenses issued for MMA; ii) the creation of at the Federal level during the National Biodiversity 2002-2007. Management Institute (Institute Chico Mendes – ICMBio) and the Brazilian Forest Service (SFB); iii) the restructuring of IBAMA to focus on environmental licensing and enforcement; and, iv) increase staffing of MMA and IBAMA. Revised: Number of environmental licenses issued per year at the Federal level. ANNEX B 74 Mainstream Climate · National Climate · National Climate BNDES clean Increase planned signed Change in public and Change Action Plan drafted Change Action Plan, approved development and carbon reductions of 20 million tons private sector investments by the Inter-ministerial by the Inter-ministerial funds programs (in equity of CO2 equivalent/year from: Committee for Climate Committee, after public investment funds) (i) CDM projects; (ii) other Change and submitted for consultation approved and operating in selected BNDES projects; public consultation accordance with the and (iii) other actions National Climate Change monitored under the National Action Plan and its New Climate Change Action Plan. Environmental and Social Institutional Policy. Improve the effectiveness · Revised Green · New Environmental BNDES` new 100 percent of the of environmental and social Protocol approved and signed and Social Institutional Policy, Environmental and Social projects submitted directly to management systems in by GOB and all federal public which includes the provisions of Institutional Policy BNDES screened, approved BNDES and other financial Brazilian Banks, including the Green Protocol, is approved applied to its full and monitored according to institutions BNDES by BNDES Board of Executive portfolio through the new Environmental and Officers and applied to BNDES’ procedures specified Social Institutional Policy. directly financed operations diferentially for each major type of operation. Sub-objective B: Integrating Principles of Sustainable Development in Key Sectors B.1. Natural Resources Management and Conservation 75 ANNEX B Improve sustainability of · Forest legal · BNDES programs on Sustainable Natural natural resources framework strengthened forest management, forest Forest Management of management through the enactment of three plantations, agriculture and private and public areas key legal acts: Public Forest associated processing industries expanded from 27,000 km2 to Management Law which restructured to provide 50,0002. promotes forest management incentives for long-term forest in public land; Atlantic Forest management and sustainable Law which promotes land use, and three sub-sectoral conservation of this highly guidelines and the endangered biome; and, REFLORESTA Program are Resolution 3545 of the designed to ensure coherence National Monetary Council with the new forest legal that regulates bank lending to framework and BNDES’ new agribusiness Environmental and Social Institutional Policy Improve Amazon regional · National Sustainable Ecological Reduction in average planning for sustainable Amazon Program (PAS) Economic Zoning at the annual rate of deforestation development and reduced approved by GOB and the macro scale (1 : 1 un the Amazon for the period deforestation governors of the Amazon million) of the Amazon 2008-1020 to 20 percent region Region, based on PAS, below 2005-2007 average concluded. annual rate of deforastation (14,800 km2). ANNEX B 76 Improve Rainforest · Presidential decree Amazon Fund An area of 500,000 Conservation regulating the Amazon Fund operational policies hectares receiving support issued, to support grant approved by its managing from the Amazon and activities that promote committee (COFA), Atlantic Forest Funds for sustainable use of natural Amazon Fund activities that promote resources, rehabilitated operational, and sustainable use of natural degraded areas or, Presidential decree resources and biodiversity prevent/combat deforestation regulating the Atlantic conservation, rehabilitate Forest Fund, including its degraded areas or, prevent/ operational mechanisms, combat deforestation. issued. B.2. Water Resource Management Improve management and · The Water · ANA Signs ANA sigs Water quality quality of water resources Resources National Plan cooperation agreements with cooperations agreements monitoring and evaluation for (PNRH) and the National five states for the purpose of with additional five states 90,000 km of main rivers Water Quality Evaluation implementing the National for the purpose of being executed and publicly Program (PNQA) approved by Water Quality Evaluation implementing the released on a regular basism GOB Program (PNQA) at the state National Water with results used for level QualityEvaluation priorization of investments Program (PNQA) at the for improved water quality. state level. B.3. Environmental Sanitation 77 ANNEX B Reduce environmental · Law 11,445/07 on · BNDES ensures that PRODES rules and Projected reduction of impacts through improved National Guidelines for Water programs for water and regulations governing 110,000 tons of pollution water, wastewater Supply and Environmental environmental sanitation are payments for wastewater loads (tons of bio-chemical treatment, and solid waste Sanitation enacted coherent with the new legal treatment updated and oxygen demand - BOD per services framework and with BNDES’ approved by ANA, in year) discharged into rivers: new Environmental and Social accordance with the new i) from approved BNDES Institutional Policy legal framework. environmental sanitation BNDES` new projects reviewes under new Environmental and Social BNDES social and Institutional Policy. environmental guidelines; and, ii) from the updated PRODES program. B.4. Renewable Energy Promote renewable energy • Integrated • BNDES ensures that BNDES six sub- 60,000 tera joules per potential Environmental Assessment programs for energy efficiency sectoral guidelines for year (TJ/year) to be produced (IEA) methodology for and renewable energy are renewable energy by renewable energy sources improving the environmental coherent with BNDES’ new designed to ensure or saved by energy efficiency and social sustainability of the Environmental and Social coherence with BNDES projects supported by hydroelectric sector included I Institutional Policy new Environmental and BNDES, once they are fully the handbook of the Social Institutional operational. Electricity Sector’s inventory Policy. and applied in ten river basins Source: World Bank Program Document, 2009 ANNEX C 79 Annex C: Summary of outcome indicators and achievements as reported in the World Bank’s ICR* Policy Area Outcome Indicators for Program for Achievements as Reported 2011 in the Bank's ICR Sub-objective A: Improving the Overall Brazilian Environmental Management System Improve effectiveness of government Improve the environmental Estimated 30 percent agencies in implementing mandated licensing process using as a proxy the judicially challenged Brazilian environmental and social decrease by 20 percent the number of licenses by Public management procedures judicially challenged licenses by Public Prosecutor`s Office. Prosecutor`s Office compared with the number of environmental licenses issued at the Federal level during 2002- 2007. Revised: Number of Revised: From 375 to environmental licenses issued per year 475 at the Federal level. Mainstream Climate Change in public Increase planned signed 19,762,768 tons of CO2 and private sector investments reductions of 20 million tons of CO2 equivalent/year. equivalent/year from: (i) CDM projects; (ii) other selected BNDES projects; and (iii) other actions monitored under the National Climate Change Action Plan. Improve the effectiveness of 100 percent of the projects 100 percent. environmental and social management submitted directly to BNDES screened, systems in BNDES and other financial approved and monitored according to institutions the new Environmental and Social Institutional Policy. Sub-objective B: Integrating Principles of Sustainable Development in Key Sectors B.1. Natural Resources Management and Conservation ANNEX C 80 Improve sustainability of natural Sustainable Natural Forest 33,415 km2 in private resources management Management of private and public and public land. areas expanded from 27,000 km2 to 50,0002. Improve Amazon regional planning for Reduction in average annual rate Annual average rate in sustainable development and reduced of deforestation in the Amazon for the 2008-2010 is 8,942 km2, a deforestation period 2008-2010 to 20 percent below decrease of the annual 2005-2007 average annual rate of average rate of deforestation deforestation (14,800 km2). of 40.3%, Improve Rainforest Conservation An area of 500,000 hectares 20,250 km2. receiving support from the Amazon and Atlantic Forest Funds for activities that promote sustainable use of natural resources and biodiversity conservation, rehabilitate degraded areas or, prevent/ combat deforestation. B.2. Water Resource Management Improve management and quality of Water quality monitoring and 116,141 km. water resources evaluation for 90,000 km of main rivers being executed and publicly released on a regular basis with results used for prioritization of investments for improved water quality. B.3. Environmental Sanitation Reduce environmental impacts through Projected reduction of 110,000 141,280 tons of improved water, wastewater treatment, tons of pollution loads (tons of bio- pollution loads. and solid waste services chemical oxygen demand - BOD per year) discharged into rivers: i) from approved BNDES environmental sanitation projects reviews under new BNDES social and environmental guidelines; and, ii) from the updated PRODES program. B.4. Renewable Energy 81 ANNEX C Promote renewable energy potential 60,000 terajoules per year 50,102 TJ/year (TJ/year) to be produced by renewable energy sources or saved by energy efficiency projects supported by BNDES, once they are fully operational. Source: World Bank Program Document and ICR. * Note that the information contained in this table was transposed as accurately as possible from the relevant Bank documents, and do not reflect IEG’s assessment. ANNEX D 83 Annex D: Poverty and Social Impacts, and Environmental, Forests, and other Natural Resources Aspects of DPLs The World Bank’s environmental and social safeguard policies are designed to ensure that potentially adverse impacts of Bank-supported programs on the environment and people are avoided or minimized, and unavoidable adverse impacts are mitigated. When applied correctly, they also help to reduce risks to the Bank’s reputation. Under the World Bank’s OP 8.60 governing DPLs, the safeguards and fiduciary requirements for typical investment operations do not apply. The reasoning in both cases is essentially the same: DPLs are for general (in this case national) budget support, budget resources are fungible, and expenditures are too diffuse to be tracked or to have a significant footprint through any specific investment. This does not however mean that there are no risks or impacts related to these issues. As noted above, when the operation P095205 was proposed as a Financial Intermediary Loan to BNDES, commenters on the Concept Notes raised serious doubts regarding the ability and willingness of BNDES to meet World Bank safeguards and fiduciary requirements on the investments financed by the loan, and also about the possibility of adhering to World Bank restrictions on the subsidization of interest rates under OP 8.30 on FILs. After all, one of the main justifications for the operation was the need to improve BNDES’s environmental and social safeguards. After the instrument for the loan was switched from a FIL to a DPL, those responsible were able to achieve the same goal of providing financing for BNDES’s investments, but without having to apply any of the World Bank’s requirements on environmental and social safeguards, fiduciary oversight, or interest rate subsidization. Although the World Bank’s definition of environmental and social “safeguards” does not technically apply to the financing of DPLs, the “Design of Development Policy Operations” section of OP 8.60 does include requirements for determining whether the policies supported by DPLs could have significant environmental or social consequences. These requirements are described in paragraphs 10 and 11 of the OP, titled “Poverty and Social Impacts,” and “Environmental, Forests, and other Natural Resources Aspects.” (World Bank Operational Manual 2012) In both cases, the OP states: “If there are significant gaps in the analysis of shortcomings in the borrower’s systems, the Bank describes in the Program Document how such gaps or shortcomings would be addressed before or during program implementation, as appropriate.” The potential environmental and social consequences and the borrower’s systems for mitigating them are supposed to be described in the Program Document, and any potential shortcomings are supposed to be identified along with measures for addressing them before or during implementation. 85 ANNEX E Annex E: Comparison of the Equator Principles with Brazil’s Green Protocol and BNDES’s environmental and social policy This section compares the provisions of these three sets of policies. The information used for the Equator Principles was based on the initiative’s website. 80 The information for the Green Protocol was based on the protocol’s document available in BNDES’s website. The information regarding BNDES was based on the institution’s website and on BNDES’s resolutions 2023/10 and 2025/10. The latter two carry different names but are very similar in content. Equator Principles “The Equator Principles (EPs) are a risk management framework, adopted by financial institutions, for determining, assessing and managing environmental and social risk in projects and is primarily intended to provide a minimum standard for due diligence to support responsible risk decision-making. The EP apply globally, to all industry sectors and to four financial products 1) Project Finance Advisory Services 2) Project Finance 3) Project-Related Corporate Loans and 4) Bridge Loans. The relevant thresholds and criteria for application is described in detail in the Scope section of the EP. Equator Principles Financial Institutions (EPFIs) commit to implementing the EP in their internal environmental and social policies, procedures and standards for financing projects and will not provide Project Finance or Project-Related Corporate Loans to projects where the client will not, or is unable to, comply with the EP. While the EP are not intended to be applied retroactively, EPFIs apply them to the expansion or upgrade of an existing project where changes in scale or scope may create significant environmental and social risks and impacts, or significantly change the nature or degree of an existing impact. Currently 78 adopting financial institutions (77 EPFIs and 1 Associate) in 35 countries have officially adopted the EPs, covering over 70 percent of international Project Finance debt in emerging markets”. Green Protocol 80 Since the time of the SEM DPL and this analysis, the EPs were further strengthened. See: http://www.equator-principles.com/index.php/ep3 ANNEX E 86 “The Protocol of Intentions on Environmental and Social Responsibility, informally known as the Green Protocol is a charter of principles for sustainable development signed by official banks in 1995 (Banco do Brazil, Banco do Nordeste, Banco da Amazônia, BNDES, Caixa Economica Federal and Banco Central Brazil) in which they propose to undertake policies and practices that are always and increasingly in harmony with the objective to promote development that does not compromise the needs of future generations. In May 2008, as a result of discussions on the impacts of deforestation in the Amazon involving both government and federal banks, it was established an informal working group to review and revision of the Green Protocol. The group was composed of representatives from the Ministry of Environment, Ministry of National Integration, Ministry of Finance, Bank of Northeast Brazil, the National Bank of Economic and Social Development, Banco da Amazonia, Caixa Economica Federal and Banco do Brazil. The result of this effort was to propose a new wording which argues that banks can play a role in inducing fundamental pursuit of sustainable development that assume responsibility for environmental preservation and continuous improvement in social welfare. To do so, they are provided principles that involve the commitment of banks: the promotion of sustainable development, the environmental assessment of projects to be funded, the eco-efficiency of administrative practices, the development of policies and practices aimed at sustainability, and forecasting mechanisms governance and monitoring of commitments made by signatories” 81. BNDES “Economic development on sustainable bases and support for environmental conservation initiatives and for investments of a social nature are part of the BNDES’ commitment to present and future generations. The BNDES not only respects social and environmental principles when granting credit, but also maintains its commitment to providing adequate funding to foster socially and environmentally sustainable efforts. Fostering sustainable development, proactively and in all the projects supported, is the main objective of the Bank’s Socioenvironmental Policy, focusing on the integration of economic, social, environmental and regional aspects. 82” 81 http://www.bb.com.br/portalbb/page251,8305,3926,0,0,1,6.bb?codigoNoticia=28467 82 http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/Social_and_Environmental_Resp onsibility/environmental_policy.html 87 ANNEX E BNDES’s present Social and Environmental Policy and BNDES’s Social and Environmental Responsibility were signed in November 2010 according to BNDES’s resolution 2023/10 and 2025/10. Equator Principle as a global reference The criteria used in this analysis took as reference the Equator Principles. During “the last 10 years, the Equator Principles or EPs have emerged as the industry standard for financial institutions to assess social and environmental risk in the project finance market. The EPs – which are based on the International Finance Corporation or IFC’s performance standards on social and environmental sustainability and the World Bank’s environmental, health and safety guidelines – have significantly increased attention on social/community responsibility, including as related to indigenous peoples, labor standards, and consultation with locally affected communities. They have also promoted convergence in the market: at present, 79 financial institutions in 32 countries have officially adopted the EPs, reportedly covering over 70% of international project finance debt in emerging markets” 83 1. Review and Categorization of Investments Equator Principles: According to the Equator Principles III (third version) “when a Project is proposed for financing, the EPFI [Equator Principles Financial Institution] will, as part of its internal environmental and social review and due diligence, categorize it based on the magnitude of its potential environmental and social risks and impacts”. The categories are A to C, having A as the projects with highest risks of potential impacts and C with the lowest or minimal adverse impacts. Green Protocol: The protocol’s guideline is “to consider the impacts and socio-environmental costs of asset management (owned and Third party) and the analysis of customer and investment risks based on Brazil’s National Environment Policy, subject to the following : (i) The funding of potential polluter projects or effectively using natural resources in the production process should be conditioned to the environmental legislation and licensing (ii) Incorporate socio-environmental criteria in the process of analyzing and conceding credit for investment projects, considering the magnitude of their impacts and risks and the necessity of mitigating and compensating measures; (iii) Carry out socio-environmental analysis of clients whose activities require environmental licensing and / or representing significant adverse social impacts (iv) to take into consideration during the credit analysis the recommendations and restrictions mentioned in the agro-ecological zoning or, preferably, the ecological-economic zoning during the credit analysis”. The Green Protocol does not provide specific criteria for the categorization for its projects. 83 https://blogs.law.harvard.edu/corpgov/2013/06/18/equator-principles-iii-enters-into-force-this-june/ ANNEX E 88 BNDES: According to the BNDES’s environment and social policy “the categorization of environmental and social risks will be defined according to the sector and type of activity, its location, attributes and the magnitude of environmental impacts inherent to the project. The Environmental category established for a certain investment will determine the procedures during the Analysis and Monitoring phases of the operation.” The categories are A to C, having A as the projects with highest risks of potential impacts and C with the lowest or minimal adverse impacts. 2. Environmental and Social Assessment Equator Principles: Based on the Equator Principle III “for all Category A and Category B Projects, the Equator Principle Financial Institutions (EPFI) will require the client to conduct an Assessment process to address, to the EPFI’s satisfaction, the relevant environmental and social risks and impacts of the proposed Project. The Assessment Documentation should propose measures to minimize, mitigate, and offset adverse impacts in a manner relevant and appropriate to the nature and scale of the proposed Project”. Green Protocol: According to the protocol’s guidelines there are no suggested pre-established procedures to assess environment and social impacts of the projects. Still the protocol suggests that “all activities should consider the impacts and socio-environmental costs in asset management (own and third party) and the analysis of customer and investment’s risks based on Brazil’s National Environment Policy on Environment. This includes incorporating socio-environmental criteria in the process of analyzing and conceding credit for investment projects, considering the magnitude of their impacts and risks and the necessity of mitigating and compensating measures. Last, for investments with potential social and environmental risks the protocol suggests the compliance with the environmental legislation”. According to BNDES’s website in order “to grant financial assistance, the following are respected: applicable legislation; the beneficiary’s policy of social and environmental responsibility; environmental compliance; environmental risk of the undertaking, in addition to social and environmental practices that raise the level of competitiveness of organizations and economic sectors and contribute to the improvement of social and environmental factors, not only of undertakings, but also of the country.” 3. Applicable Environmental and Social Standards Equator: According to the Equator Principles III, “the Assessment process should, in the first instance, address compliance with relevant host country laws, regulations and permits that pertain to environmental and social issues. EPFIs operate in diverse markets: some with robust environmental and social governance, legislation systems and institutional capacity designed to protect their people and the natural environment; and some with evolving technical and institutional capacity to manage environmental and social issues. For projects in countries with robust environmental regulation and governance, the process evaluates 89 ANNEX E compliance with the relevant country law and for project in countries without a robust environmental governance, the process evaluates compliance with the then applicable IFC Performance Standards on Environmental and Social Sustainability (Performance Standards) and the World Bank Group Environmental, Health and Safety Guidelines(EHS Guidelines)” Green Protocol: Based on the protocol’s guidelines “the environmental assessments will be taken under the guidance of Brazil’s National Environmental Policy”. The protocol does not mention the environmental & social standards to be applied in investments abroad. BNDES: According to the BNDES’s environmental and social policy, the institution requires environmental obligations established by Brazilian law. There is no information regarding environmental and social assessments applied to BNDES projects in other countries. 4. Environmental and Social Management System Equator Principles: According to the Equator Principle’s III document “for all Category A and Category B Projects, the EPFI will require the client to develop or maintain an Environmental and Social Management System (ESMS). Further, an Environmental and Social Management Plan (ESMP) will be prepared by the client to address issues raised in the Assessment process and incorporate actions required to comply with the applicable standards. Where the applicable standards are not met to the EPFI’s satisfaction, the client and the EPFI will agree an Equator Principles Action Plan (AP). The Equator Principles AP is intended to outline gaps and commitments to meet EPFI requirements in line with the applicable standards” Green Protocol: The protocol’s orientation is “to guide the borrower in order to induce the adoption of production practices and sustainable consumption”. BNDES’s environmental and social policy mentions that “the institution will promote and guide the adoption of preventive and mitigation of social and environmental impacts.” Still according to its website, BNDES “may, for example, recommend the project be revised; offer resources to strengthen mitigation measures to stimulate the achievement of social and environmental investments by the beneficiaries; and even refuse financial support due to non- compliance or social and environmental risks.” 5. Stakeholder Engagement Equator Principles: According to the Equator Principles’ III document “for all Category A and Category B Projects, the EPFI will require the client to demonstrate effective Stakeholder Engagement as an ongoing process in a structured and culturally appropriate manner with Affected Communities and, where relevant, Other Stakeholders. For Projects with potentially significant adverse impacts on Affected Communities, the client will conduct an Informed Consultation and Participation process. The client will tailor its consultation ANNEX E 90 process to: the risks and impacts of the Project; the Project’s phase of development; the language preferences of the Affected Communities; their decision-making processes; and the needs of disadvantaged and vulnerable groups. This process should be free from external manipulation, interference, coercion and intimidation”. Green Protocol: The protocol’s orientation is “to inform, sensitize and continuously engage stakeholders in policy and sustainable practices of an institution by enabling the workforce to develop the skills necessary to implement the principles and guidelines of this protocol; to develop mechanisms for consultation and dialogue with stakeholders, and to publish annually the results of the implementation of the principles and guidelines set forth in this protocol.” BNDES’s environmental and social responsibility guidelines say that the institution will concentrate efforts to “develop partnerships and share experiences with other organizations in order to foster social and environmental responsibility and to strengthen transparency and dialogue among stakeholders while reinforcing citizen participation in public management”; 6. Grievance Mechanism Equator Principles: The Equator Principles III mention that “for all Category A and, as appropriate, Category B Projects, the EPFI will require the client, as part of the ESMS, to establish a grievance mechanism designed to receive and facilitate resolution of concerns and grievances about the Project’s environmental and social performance. The grievance mechanism is required to be scaled to the risks and impacts of the Project and have Affected Communities as its primary user. It will seek to resolve concerns promptly, using an understandable and transparent consultative process that is culturally appropriate, readily accessible, at no cost, and without retribution to the party that originated the issue or concern. The mechanism should not impede access to judicial or administrative remedies. The client will inform the Affected Communities about the mechanism in the course of the Stakeholder Engagement process”. Green Protocol: The protocol’s orientation is to develop mechanisms for consultation and dialogue with stakeholders. No mention is made of grievance mechanism in the protocol. BNDES: Similarly to the green protocol, there is no guidance in this specific topic in the BNDES environmental social policy. 7. Independent Review Equator Principles: The principle’s document says that “for all Category A and, as appropriate, Category B Projects, an Independent Environmental and Social Consultant will carry out an Independent Review of the Assessment. For project-related corporate loans an Independent Review by an Independent Environmental and Social Consultant is required for Projects with potential high risk impacts including, but not limited to, any of the following: 91 ANNEX E (i)adverse impacts on indigenous peoples; (ii) Critical Habitat impacts; (iii) significant cultural heritage impacts and (iv) large-scale resettlement”. Green Protocol: There is no reference to independent review in this protocol. BNDES: Independent reviews are not mentioned in BNDES environmental and social policy. 8. Covenants Equator Principles: According to the principle’s document “an important strength of the Equator Principles is the incorporation of covenants linked to compliance. For all Projects, the client will covenant in the financing documentation to comply with all relevant host country environmental and social laws, regulations and permits in all material respects”. Green Protocol: The protocol does not mention the incorporation of covenants. Still, the protocol mentions that potential polluters projects or effectively using natural resources in the production process should be conditioned to legislation compliance. BNDES: The BNDES does not mention the incorporation of covenants. 9. Independent Monitoring and Reporting Equator: The Equator Principle’s document says that “to assess Project compliance with the Equator Principles and ensure ongoing monitoring and reporting after Financial Close and over the life of the loan, the EPFI will, for all Category A and, as appropriate, Category B Projects, require the appointment of an Independent Environmental and Social Consultant, or require that the client retain qualified and experienced external experts to verify its monitoring information which would be shared with the EPFI. On the Project-Related Corporate Loans For Projects where an Independent Review is required under Principle 7, the EPFI will require the appointment of an Independent Environmental and Social Consultant after Financial Close, or require that the client retain qualified and experienced external experts to verify its monitoring information which would be shared with the EPFI”. Green Protocol:”No independent monitoring mechanism and report is mentioned in this protocol. BNDES: Still and according to the BNDES’s social and environmental policy, BNDES will provide “internal socioenvironmental guides, assessment methodologies of beneficiaries, assessment of credit risk and monitoring and impact assessment of projects supported” . However, the institution does not mention the use of an independent mechanism/consultant to monitor its projects. ANNEX E 92 10. Reporting and Transparency Equator Principles: According to the principles’ document “for all Category A and, as appropriate, Category B Projects (1) The client will ensure that, at a minimum, a summary of the ESIA is accessible and available online; (2) The client will publicly report GHG emission levels (combined Scope 1 and Scope 2 Emissions) during the operational phase for Projects emitting over 100,000 tons of CO2 equivalent annually. Refer to Annex A for detailed requirements on GHG emissions reporting”. “The EPFI will report publicly, at least annually, on transactions that have reached Financial Close and on its Equator Principles implementation processes and experience, taking into account appropriate Except in cases where the client does not have internet access”. Green Protocol: The protocol’s orientation is “to commit to publish annually the results of the implementation of the principles and guidelines set forth in the protocol”. BNDES: The BNDES expressed intention is “To continuously expand and update knowledge concerning sustainable development as well as social and environmental responsibility while sharing information and experiences with beneficiaries, financial institutions and other organizations, seeking dialogue and fostering the integration of efforts to strengthen the approach to social and environmental dimensions as a strategic issue. Still, according to BNDES’ social and environmental responsibility the institution considers “ethics and transparency as the pillars of relations with all stakeholders, ensuring dialogue and accounting for its decisions and efforts” However there is no expressed commitments on annual reports or on individual projects. ANNEX F 93 Annex F: Organization of Brazil’s Federal Ministry of Environment and related Agencies Source: GEF Evaluation Office, 2012 94 ANNEX G Annex G: Project Information Available from BNDES’s Website Data da Val Contratado Cliente CNPJ Descrição do Projeto UF Porte da Empresa Ramo/Gênero de Atividade Área Operacional Modalidade de Apoio Contratação R$ CONTRIBUIR PARA O DESENVOLVIMENTO ECONOMICO DO ESTADO DE SANTA CATARINA POR MEIO DO APOIO A ESTADO DE SANTA CATARINA 82951229000176 EMPREENDIMENTOS PRODUTIVOS, SELECIONADOS POR UM OU MAIS EDITAIS, PARA A INCLUSAO DE PESSOAS SC 1/8/2013 10,000,000 ADMINISTRAÇÃO PÚBLICA DIRETA COMERCIO E SERVICOS/ADMINISTRAÇÃO PÚBLICA AREA AGROPECUARIA E DE INCLUSAO SOCIAL NÃO REEMBOLSÁVEL DE BAIXA RENDA. IMPLANTACAO DE UM PARQUE EOLICO, COM CAPACIDADE INSTALADA TOTAL DE 30 MW, ENVOLVENDO PEDRA BRANCA S/A 12709996000198 BA 1/8/2013 50,654,100 GRANDE/MÉDIA-GRANDE COMERCIO E SERVICOS/ELETRICIDADE E GÁS AREA DE INFRA-ESTRUTURA REEMBOLSÁVEL MONTAGEM DE TURBINAS EOLICAS, CONSTRUCAO DA INFRAESTRUTURA E INSTALACOES ELETRICAS. IMPLANTACAO DE UM PARQUE EOLICO, COM CAPACIDADE INSTALADA TOTAL DE 30 MW, ENVOLVENDO SAO PEDRO DO LAGO S/A 12709813000134 BA 1/8/2013 51,746,800 GRANDE/MÉDIA-GRANDE COMERCIO E SERVICOS/ELETRICIDADE E GÁS AREA DE INFRA-ESTRUTURA REEMBOLSÁVEL MONTAGEM DE TURBINAS EOLICAS, CONSTRUCAO DA INFRAESTRUTURA E INSTALACOES ELETRICAS. IMPLANTACAO DE UM PARQUE EOLICO, COM CAPACIDADE INSTALADA TOTAL DE 30 MW, ENVOLVENDO SETE GAMELEIRAS S/A 12710327000136 BA 1/8/2013 55,146,700 GRANDE/MÉDIA-GRANDE COMERCIO E SERVICOS/ELETRICIDADE E GÁS AREA DE INFRA-ESTRUTURA REEMBOLSÁVEL MONTAGEM DE TURBINAS EOLICAS, CONSTRUCAO DA INFRAESTRUTURA E INSTALACOES ELETRICAS. VIABILIZAR A EXECUCAO DE PROGRAMAS DE DESENVOLVIMENTO INTEGRADO CONSTANTES DO PLANO ESTADO DO MARANHAO 06354468000160 MA 1/14/2013 1,001,340,520 ADMINISTRAÇÃO PÚBLICA DIRETA COMERCIO E SERVICOS/ADMINISTRAÇÃO PÚBLICA AREA DE INFRAESTRUTURA SOCIAL REEMBOLSÁVEL PLURIANUAL E LEIS ORCAMENTARIAS ANUAIS DO ESTADO DO MARANHAO AMPLIACAO, MANUTENCAO E EXPLORACAO DOS 84 PRIMEIROS MESES DA CONCESSAO DO AEROPORTO INFRAMERICA CONCESSIONARIA DO AEROPORTO DE BRASILIA S/A 15559082000186 INTERNACIONAL DE BRASILIA, DENOMINADO PRESIDENTE JUSCELINO KUBITSCHEK, LOCALIZADO EM BRASILIA, DF 1/14/2013 488,000,000 GRANDE/MÉDIA-GRANDE COMERCIO E SERVICOS/TRANSPORTE AÉREO AREA DE INFRA-ESTRUTURA REEMBOLSÁVEL DISTRITO FEDERAL, OBJETO DO EDITAL DO LEILAO Nø 2/2012 DA ANAC, REALIZADO EM 06/02/12. DESENVOLVIMENTO DE UM CIRCUITO INTEGRADO (CHIP) DO TIPO ASIC (APPLICATION SPECIFIC INTEGRATED FUNDACAO CPQD CENTRO PESQ. DESENV. EM TELECOMUNICACOES 02641663000110 CIRCUIT) COM A FINALIDADE DE PROCESSADOR OTN (OPTICAL TRANSPORT NETWORK), PARA APLICACOES EM SP 1/16/2013 29,328,080 MPME COMERCIO E SERVICOS/ATIV IMOBIL, PROFISSIONAL E ADM AREA INDUSTRIAL NÃO REEMBOLSÁVEL REDES OPTICAS OPERANDO A 100 GBIT/S. FINANCIAMENTO NO AMBITO DO PROINVESTE, DE DESPESAS DE CAPITAL RELATIVAS A AMORTIZACAO DO PRINCIPAL DA DIVIDA DO ESTADO DE GOIAS DECORRENTE DO CONTR.DE FINANCIAMENTO MEDIANTE A ESTADO DE GOIAS 01409580000138 GO 1/18/2013 260,416,043 ADMINISTRAÇÃO PÚBLICA DIRETA COMERCIO E SERVICOS/ADMINISTRAÇÃO PÚBLICA AREA DE INFRAESTRUTURA SOCIAL REEMBOLSÁVEL ABERTURA CREDITO N.10201371(PEF2), CELEBRADO COM BNDES EM 01/06/2010, E APORTE PARA AUMENTO DE CAPITAL DA SANEAGO CONTRAPART IMPLANTACAO DO ACORDO DE COOPERACAO ENTRE BNDES E MINISTERIO DA INTEGRACAO NACIONAL - ACOES CIA DE DESENVOLV DOS VALES DO SAO FRANCISCO E PARNAIBA 00399857000126 IE 1/22/2013 38,000,000 MPME COMERCIO E SERVICOS/OUTRAS ATIV SERVIÇOS AREA AGROPECUARIA E DE INCLUSAO SOCIAL NÃO REEMBOLSÁVEL EMERGENCIAIS DE CONVIVENCIA COM A SECA. IMPLANTACAO DO ACORDO DE COOPERACAO ENTRE BNDES E MINISTERIO DA INTEGRACAO NACIONAL - ACOES DEPARTAMENTO NACIONAL DE OBRAS CONTRA AS SECAS 00043711000143 IE 1/22/2013 19,000,000 MPME COMERCIO E SERVICOS/OUTRAS ATIV SERVIÇOS AREA AGROPECUARIA E DE INCLUSAO SOCIAL NÃO REEMBOLSÁVEL EMERGENCIAIS DE CONVIVENCIA COM A SECA. VIABILIZAR A EXECUCAO DO PROGRAMA MULTISSETORIAL DE APOIO A INTERIORIZACAO DO DESENVOLVIMENTO ESTADO DE PERNAMBUCO 10571982000125 SUSTENTAVEL DO ESTADO DE PERNAMBUCO, COMPOSTO POR PROJETOS CONSTANTES DO PLANO PE 1/23/2013 423,613,990 ADMINISTRAÇÃO PÚBLICA DIRETA COMERCIO E SERVICOS/ADMINISTRAÇÃO PÚBLICA AREA DE INFRAESTRUTURA SOCIAL REEMBOLSÁVEL PLURIANUAL E LEIS OR‡AMENTARIAS DO POSTULANTE. DESENVOLVIMENTO DE UM SISTEMA OPTICO DE VARREDURA E LEITURA A LASER, BASEADO NA TECNOLOGIA INSTITUTO ATLANTICO 04614281000123 DE FILMES FLUORESCENTES DE FOSFORO (PSPL), INCLUINDO UM SOFTWARE DE GERACAO DE IMAGENS PARA CE 1/24/2013 5,627,273 MPME COMERCIO E SERVICOS/ATIV IMOBIL, PROFISSIONAL E ADM AREA INDUSTRIAL NÃO REEMBOLSÁVEL USO EM RADIOLOGIA DIGITAL ODONTOLOGICA, NO AMBITO DO BNDES FUNDO TECNOLOGICO - BNDES FUNTEC. Source: BNDES’s website, direct contracts, posted by 8/10/2013, accessed on 10/1/2013; available at: http://www.bndes.gov.br/SiteBNDES/bndes/bndes_pt/Institucional/BNDES_Transparente/Consulta_as_operacoes_do_BNDES/operacoes_diretas.html ANNEX H 95 Annex H: History and Timeline of the Operation To properly evaluate the SEM DPL, it is important to understand the history and evolution of the operation, and the context in which it was prepared. Over the course of the preparation process, there were significant changes to the design—including to the financing instrument, the objectives, and the implementing agencies—that affected issues that arose later, including with the project logic, the delay in effectiveness, the inclusion of actions previously supported by the Env PRL, and the Bank’s reputation. Having the complete story up-front in one place makes it more coherent, and allows the report to refer back to certain aspects of it later in the evaluation where relevant. To summarize the many steps involved, a Timeline is provided at the end of this section. According to documentation and World Bank management, preparation of the operation, with the same project ID (P095205) began with a World Bank mission to the headquarters of the National Bank for Economic and Social Development (BNDES) in Rio de Janeiro in November 2007. BNDES is a wholly owned Brazilian federal government development bank that finances both private and public investments, including large infrastructure projects, and whose annual lending has grown to several times that of the World Bank globally (BNDES 2013a; Colby 2012). 84 BNDES has also been increasing its presence internationally, particularly in other countries in Latin American, where it has become one of the most important financial institutions (Financial Times 2012, 2013b; BIZ 2011; Rainforest Foundation UK). It also finances investments in Africa, and recently opened its first representative office in Africa in order to expand its presence there (Financial Times 2010, 2012, 2013c, 2013d). 85 The World Bank memo with the objectives of the first mission referred to the operation as the “BNDES PAC-Env DPL”, and said that the World Bank team would “discuss with the main counterparts key design aspects of a potential PAC and environmental and social policy operation.” The “PAC” is the Portuguese acronym for the Brazilian government’s flagship Program for Accelerated Growth to support a major scaling-up and acceleration of infrastructure investment, and the World Bank team’s back-to-office report highlighted “BNDES as the primary channel for federal financing of this ambitious program,” and noted that as a result, “infrastructure lending is projected to more than double in 2007-2010.” At the same time, BNDES had received criticism nationally and internationally for the adverse environmental and social impacts of some of its investments, including in large slaughterhouse operations in the Amazon (see for example, Greenpeace 2009), and both the BNDES and World Bank teams recognized that there were significant weaknesses in BNDES’s approach to environmental and social safeguards (see Program Document). BNDES therefore wanted to make “the widespread strengthening of BNDES social and 84 http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/The_BNDES_in_Numbers/ 85 http://www.bndes.gov.br/SiteBNDES/bndes/bndes_en/Institucional/Press/Noticias/2013/20131206_ africa.html ANNEX H 96 environmental screening policies and of their implementation throughout the BNDES portfolio” one of the main objectives of the operation. According to the Country Director at the time, the World Bank consciously made a major change in direction at the time in deciding to engage in financing development in the Amazon. Apart from “boutique” environmental projects, the World Bank had not engaged in lending for economic development in the Amazon since the 1980s, and wanted to support investments in sectors like energy, agriculture, mining, and infrastructure, and to do it in an environmentally sustainable way. This was reflected in the World Bank’s 2008-2011 Country Partnership Strategy (CPS) for Brazil, which states that “…the Bank Group will reverse its de facto decades’ long withdrawal from the financing of development in the Amazon…” through what it called the Amazon Partnership Program, which among other things would support “…the planning and implementation of major energy and transport infrastructure…” (World Bank 2008). In view of BNDES’s extensive financing of projects in the Amazon, including in the context of the PAC, the planned World Bank loan to BNDES was seen as a manifestation of the new focus. In March 2008, the Government of Brazil External Financing Commission (COFIEX), authorized the preparation of a “DPL/BNDES-World Bank” program. It specified that the loan would be in the amount of US$ 1 billion, with BNDES as the only beneficiary, and the Federal Government of Brazil as the guarantor. The DPL is the World Bank’s main instrument for providing a client country budget support through “rapidly disbursing policy-based financing”, and its use is governed by Operational Policy (OP) 8.60. 86 Although up until that point the idea was to prepare a DPL to BNDES, at the first formal stage in the World Bank project preparation cycle, the Concept Note issued for the operation in June 2008 was not for a DPL, but rather for a Financial Intermediary Loan (FIL) from the World Bank to BNDES titled the “BNDES Environmental and Social Sustainability Project” (though still with the same project code, P095205). The proposed US$ 1 billion in financing was to be divided into two Components: (A) Strengthening of Environmental and Social Screening and Monitoring would provide US$ 10 million for technical assistance and training to BNDES; and (B) a US$ 990 million Line of Credit to finance BNDES investments in infrastructure and green lines of business. The June 2008 Concept Note contains a section wherein the Bank team discusses alternative lending instruments that they had considered but rejected in favor of the FIL instrument. In particular, they discuss at some length the originally proposed DPL instrument—which the Bank team reports was the preferred option of BNDES—and why the Bank team rejected it for the operation. The June 2008 Concept Note states: “Though BNDES deemed that a policy based loan is the most appropriate instrument, this would require that either BNDES receive DPL funding through on-lending through the Federal Government or through a DPL directly to BNDES which would entail the granting of an exception to OP 8.60 by the World Bank’s Board” because, they explain, “OP 8.60…states that only Bank members and their sub- 86 See World Bank Operational Manual, available at: http://web.worldbank.org/WBSITE/EXTERNAL/PROJECTS/EXTPOLICIES/EXTOPMANUAL/0,, menuPK:64142516~pagePK:64141681~piPK:64141745~theSitePK:502184,00.html 97 ANNEX H national divisions may benefit from a DPL.” The Bank team expressed uncertainty about whether such an exception would be granted. The Concept Note further states that “an approach for this sort of operation in which the government borrows from the Bank and then on-lends to BNDES is not consistent with the financial requirements of the country.” In detailing why the government did not want to borrow from the Bank and on-lend to BNDES, the Concept Note refers to concerns by the client about the impact on the external debt statistics of the Federal government. If this loan had continued to be prepared as a FIL as previously proposed, the operation would have had to adhere to the World Bank’s OP 8.30, which among other things requires removal of interest rate subsidies, application of World Bank policies on environmental and social safeguards, and financial management and procurement, to the loan and all sub- projects financed. 87 In their written comments, reviewers of the Concept Note were critical with regard to these requirements and BNDES’s ability to meet them, including because BNDES supplies loans at subsidized interest rates, and because one of the main rationales for the operation was that BNDES’s environmental and social policies were weak. The Bank team did not receive clearance to proceed with preparation of the operation at the Concept Review Meeting in June and, unusually, Decision Notes were never circulated. A revised Concept Note was issued—again for a FIL to BNDES—and a Concept Review Meeting held in September 2008. As reflected in the Decision Notes from the meeting issued on October 8, 2008, many of the same issues remained as in June, but nevertheless the Bank team received clearance to proceed with preparation of the FIL—with guidance to take care to address the many outstanding issues that remained, including with interest rate subsidies, safeguards, and procurement. 88 In the meantime, Lehman Brothers filed for bankruptcy in September of 2008, sparking the global financial crisis. Although we now know that Brazil emerged from the crisis relatively unscathed, at the time there was a great deal of uncertainty. In a special annex to the SEM DPL Program Document on “The Brazilian Economy in the World wide Financial Crisis”, the World Bank highlighted that “despite its increased resilience, Brazil has been hit hard by the global financial crisis,” which had “already shocked the real economy” (World Bank 2009). In response, the Brazilian Government “recently announced the adoption of countercyclical fiscal policies,” including through accelerated financing of infrastructure investments through BNDES. In this context, the World Bank Annual Meetings were held from October 11-13, 2008. The BNDES President took the unusual step of traveling to Washington, DC to attend, and a meeting was organized between him, World Bank management, and Brazilian government officials to discuss the loan. According to World Bank team members and those familiar with the meeting, BNDES needed the money to help increase its lending, but they objected to requirements of OP 8.30 on FILs. According to those familiar with the discussions, BNDES objected to applying World Bank investment lending requirements—particularly restrictions on interest rate subsidies, the application of World Bank policies on environmental and social 87 OP 8.30 was replaced by OP 10.00 - Investment Project Financing, in April 2013. 88 The Decision Notes contained no recommendation to change the instrument to a DPL. ANNEX H 98 safeguards, and procurement rules—to the sub-investments financed. So agreement was reached to abandon the idea of a FIL, and instead pursue other options. Brazilian government representatives noted that since the earlier discussions on a DPL, its debt situation had improved (due in part to exchange rate depreciation; see Program Document Annex 9). Moreover, the Brazilian officials noted that there was still an unfinished environmental policy reform agenda in Brazil from the earlier Env PRL series, and that many of the 15 policy areas from that earlier environmental DPL could be revisited to help justify a broader national-level policy operation. So agreement was reached that the Brazilian Federal Government would borrow the money from the World Bank in the form of an environmental DPL, and on-lend the money to BNDES. This allowed the World Bank, to prepare the loan and disburse the money much faster than would have been possible under a FIL—without having to deal with FIL requirements related to interest rate subsidies, environmental and social safeguards, and fiduciary oversight. It also allowed the Bank to reverse a steep decline in lending to Brazil, which had fallen from $3.2 billion over 2005-6 to less than $300 million in 2007. World Bank Brazil team members reported that they felt pressure from senior management, at the time, to increase lending. From that point, the operation was prepared very quickly. In November 2008, another Concept Review Meeting was held, this time for the “Brazil: First Programmatic Development Policy Loan for Sustainable Environmental Management”, or SEM DPL. Although the project ID remained the same, a programmatic DPL was now proposed with nine policy areas and two objectives that were virtually the same as the objectives for the earlier Env PRL: improve the Brazilian environmental management system, and mainstream environment into targeted sectors. The lending amount for the series was doubled from that originally requested by COFIEX for the BNDES DPL and proposed by the World Bank in the FIL Concepts: from US$ 1 billion to US$ 2 billion. The Ministry of Environment was added as an implementing agency. Though the title and objectives no longer mentioned of BNDES, BNDES remained an implementing agency and a major focus of the reforms the SEM DPL aimed to support. Less than three months later, on February 3, 2009, the Program Document was finalized for the SEM DPL, and the SEM DPL 1 loan was approved by the World Bank Board on March 5, 2009. None of the World Bank documentation for the SEM DPL—including the Program Document, Loan Agreement, and the Implementation Completion and Results Report (ICR)—mentioned that all of the financing would be on-lent to BNDES, or that other instruments and beneficiary arrangements had been considered. On the part of the Borrower, the plan to on-lend the SEM DPL funds to BNDES was made official through Interim Measure 450 of the Brazilian Presidency in December 2008, several months before the SEM DPL was submitted to the World Bank Board of Directors (Presidency of the Republic 2008; Senate of Brazil 2008). 89 The measure authorizes the Brazilian federal government to transfer up to US$ 2 billion from the World Bank (IBRD) to 89 Available at: http://www.planalto.gov.br/ccivil_03/_ato2007-2010/2008/Mpv/450.htm . The Presidency website notes that Interim Measure 450 was revoked once Law No. 11.943 was approved (thus the strikethroughs). 99 ANNEX H BNDES. 90 The Interim Measure later became Brazilian Law number 11.943, Article 15, which authorizes the transfer of up to the full US$ 2 billion planned for the SEM DPL series, in the form of a loan from the Federal Government to BNDES (Presidency of the Republic 2009). 91 The law is dated May 28, 2009, not long after the SEM DPL was approved by the World Bank Board, and specifies that “the Government will pass on resources to BNDES under the same financial terms offered by the IBRD.” Despite the speed with which the SEM DPL was prepared and approved, and the argument made in the Program Document that “the aim of ensuring adequate credit resources to the financial system is a key motivating factor of the GOB’s request for the SEM DPL” in response to the financial crisis, the loan then lingered for more than a year before the Loan Agreement was signed and the operation made effective. Based on interviews and documentation, IEG understands that the delay was due to the need for the loan to be approved by the Brazilian Senate, which took some time. In preparation for the vote in the Senate, detailed official Legal Opinions (“Parecer” in Portuguese) were prepared by the Brazilian Treasury, Ministry of Finance, on March 31, 2010 (Parecer PGFN/COF/No. 589/2010) and by the Senate on April 20, 2010 (Parecer No. 410 of 2010). The Legal Opinions explain that all of the funds provided by the World Bank through the SEM DPL would be directed to BNDES. Moreover, they say that the Brazilian Treasury entered into the SEM DPL on behalf of BNDES, with the purpose of on-lending the funds to BNDES on the same terms that the World Bank lent the money to the Government of Brazil. The Legal Opinions also emphasize that because the money is being lent by the World Bank under its DPL instrument, the proceeds from the loans “are not allocated for investments.” The SEM DPL was signed and became effective on June 21, 2010. The first tranche of US$800 million was released on June 30, 2010, fifteen months later than what was planned when the loan was approved by the World Bank Board. The second tranche of US$500 million was disbursed on December 15, 2010, seventeen months late. Thus the total amount of US$1.3 billion for SEM DPL 1 was disbursed. Each of these disbursements was accompanied by a detailed contract signed between the Federal Government of Brazil and BNDES (Financing Contracts 544 and 590, respectively). The Financing Contracts were quite detailed and in many respects passed on to BNDES terms that were specified in the SEM DPL Legal Agreement signed between the Federal Government and the World Bank (IBRD), stating that the “financing must observe the same financial conditions agreed in the IBRD Loan Contract”, and integrating copies of the World Bank Loan Agreement and The General Conditions for IBRD Loan Contracts, including the interest rate and the amortization schedule. There was however at least one notable difference: each of the Brazilian Financing Contracts specified that “this contract aims at granting credit to BNDES…to provide it with the resources for application in its investment operations.” 90 Although the SEM DPL is not mentioned by name, there were no other World Bank loans to BNDES around that time, and the amount of US$ 2 billion quoted is the same as the total planned for the SEM DPL series. As noted below, a subsequent legal review by the Brazilian Treasury does name the SEM DPL, and links it to the earlier Interim Measure and Law cited here. 91 Available at: http://www.planalto.gov.br/ccivil_03/_ato2007-2010/2009/Lei/L11943.htm . ANNEX H 100 The SEM DPL was closed two weeks after the second tranche was disbursed, on December 31, 2010. The planned second loan in the series, SEM DPL 2, never materialized. To provide perspective on the scale of the SEM DPL 1, it was equivalent to the total average annual financing commitment for the environment for the entire World Bank from 2008-2012, including financing from the Global Environment Facility (GEF), and it was ten times larger than the World Bank’s total global GEF-financed commitments in 2011 or 2012. 92 Looking only at the non-DPL environment portfolio (since the SEM DPL itself is included in the overall calculation), SEM DPL 1 was more than 2.5 times bigger than the World Bank’s global average annual financing commitment for the environment from 2008-2012. Not long after the SEM DPL 1 closed, the Brazilian Federal Prosecutor's Office (Ministério Público Federal) enquired about the loan with BNDES, and BNDES confirmed that the full US$1.3 billion “were transferred to BNDES for complementing its investment budget”. BNDES’s response stated further that the resources were provided as a concessionary loan by the Federal Treasury “to BNDES in order to provide resources to be applied in its investment operations. That way, the resources from the financing were utilized to complement the BNDES disbursement budget.” Indeed, BNDES ramped up its lending in a major way at the time of the SEM DPL, reporting that its disbursements increased by 23% in 2010 (BNDES 2013a). The large amounts of money that the Brazilian government has been channeling through BNDES was recently criticized by the OECD. 93 Timeline for World Bank operation P095205 (“SEM DPL” and other manifestations) Date Event Comments 2007 Nov. 13-14 World Bank mission to BNDES “PAC” is the Portuguese acronym for for preparation of a “BNDES Brazil’s “Program for Accelerated PAC-Env DPL” Growth”, a major government infrastructure investment program. There is no indication that Brazilian agencies other than BNDES were involved. 2008 Mar. 27 COFIEX sends fax to World The fax specified that the beneficiary of Bank authorizing the the loan should be BNDES, with the preparation of a “DPL/BNDES- Federal Government of Brazil as the World Bank” guarantor, for US$ 1 billion. June 4 Concept Note for a “Brazil: The Concept was for a US$ 1 billion loan BNDES Environmental and to BNDES, with US$ 10 million for 92 Source: World Bank Business Warehouse. 93 In its 2013 Brazil Economic Survey, the OECD was critical of the significant increase in transfers from the national budget to BNDES, noting that “a large share of BNDES loans is extended to large companies…[which] would be the ones that would probably enjoy the easiest access to credit in private credit markets.” It recommends that “private entry will require levelling the playing field by phasing out all direct and indirect financial support to BNDES.” (OECD 2013) 101 ANNEX H Social Sustainability Project” environmental and social TA, and US$ Financial Intermediary Loan 990 million to finance BNDES (FIL) formally issued by World investments. It argued that a DPL would Bank team for comment not be appropriate. Serious issues were raised by reviewers. No Decision Notes were ever issued. Sept. 6 A second Concept Note for a The Concept is similar to the one issued in “Brazil: BNDES Environmental June 2008. and Social Sustainability Facility” FIL was formally issued by World Bank team. Oct. 8 The Decision Notes issued for a The Decision Notes reflect serious issues Concept Review Meeting for that continued to be raised by reviewers; “BNDES Environmental and instruct Bank team to pay attention to Social Sustainability Facility” these during project preparation. No FIL held on Sept. 12. mention of a decision to change to the DPL instrument. Oct. 11-13 BNDES President attends A meeting is held between the BNDES Annual Meetings of the World President, Brazilian government officials, Bank in Washington, DC and World Bank management to discuss the loan. Agreement is reached to prepare DPL to the Federal Government, which would pass the financing to BNDES. Nov. 3 Concept version of a Program The Program Document is now for a Document for “Brazil: First programmatic series of two loans totaling Programmatic Development US$ 2 billion; SEM DPL1 for US$1.3 Policy Loan for Sustainable billion. Implementing agencies: Ministry Environmental Management” of Finance, Ministry of Environment, and (SEM DPL) completed BNDES. No mention of earlier efforts to prepare a loan to BNDES. No mention of financing going to BNDES. Nov. 11 Concept Review Meeting for Decision Notes state “Despite the key role “Brazil: First Programmatic to be undertaken by BNDES, it was Development Policy Loan for reiterated clearly that the loan is designed Sustainable Environmental as a DPL.” And “highlight the importance Management” (SEM DPL) of BNDES without generating the erroneous perception that we are directly supporting investments.” Dec. 9 Brazilian President issues The news is covered in Brazilian media, Interim Measure 450 that including the largest paper in the country. authorizes the government to transfer up to US$ 2 billion from the World Bank to BNDES. Dec. 18 Begin Negotiations ANNEX H 102 2009 Feb. 3 Program Document for SEM There is no mention of the earlier efforts to DPL finalized prepare a loan to BNDES, or that other instruments were considered. No mention of financing going to BNDES. Mar. 5 World Bank Board approval No discussion of earlier efforts to prepare a loan to BNDES, or consideration of other instruments. No mention of financing going to BNDES. May 28 Law 11,943 enacted by The Law replaces Interim Measure 450. Brazilian government authorizes the government to transfer up to US$ 2 billion from the World Bank to BNDES. 2010 April 13 Bank receives Legal Opinion The 10-page Opinion explains in detail (Parecer) 589 from Ministry of plans by GOB to on-lend the SEM DPL Finance funds in their entirety to BNDES, on IBRD terms. States: “According to IBRD DPL procedures, the funds will not be destined to new investments...” April 20 Federal Senate issues Legal Similar to the Ministry of Finance Legal Opinion (Parecer) 410 Opinion. Also confirms that the funds are not intended for new investments, and will be “allocated to BNDES to finance actions for the management of Brazilian environmental sustainability.” June 21 SEM DPL 1 loan for US$ 1.3 Effectiveness is 13 months behind billion becomes effective schedule. The Bank blames this on GOB forgetting to put the loan in the budget the previous year. June 30 SEM DPL 1, tranche 1, is The disbursement is 15 months late. disbursed for US$ 800 million June 30 BNDES and Federal The contract states that the purpose is “to Government sign Financing provide [BNDES] with the resources for Contract 544, granting BNDES application in its investment operations.” a loan for US$ 800 million Other content of the contract is similar to from the SEM DPL. the World Bank Loan Agreement. Dec. 15 CSO representatives meet with The CSOs seek clarification on Bank Management in Brasilia achievement of SEM DPL policy actions, and request that SEM DPL tranche 2 not be disbursed in the meantime. Dec. 15 SEM DPL 1, tranche 2, is The disbursement is 17 months late. disbursed for US$ 500 million 103 ANNEX H Dec. 15 BNDES and Federal The contents appear identical to Contract Government sign Financing 544. Again confirms that funds are for Contract 590, granting BNDES BNDES’s investment operations. a loan for US$ 800 million from the SEM DPL. Dec. 17 More than 30 international and Stated that there were “serious problems” national CSOs write to World with the SEM DPL, and expressed Bank LAC VP about SEM “extreme concern” regarding lack of DPL. transparency, consultations, and evidence of implementation of reforms. Raises specific questions about BNDES. Requested dialogue with the Bank and hold on disbursement of tranche 2. Dec. 31 SEM DPL 1 is closed 2011 Jan. 13 World Bank Brazil Country States: “the documents presented by the Director responds to CSO letter borrower as evidence of compliance with of 12/17/10 (on behalf of VP) the second tranche release conditions of the SEM DPL 1 were received in confidence and cannot be disclosed by the Bank…” Feb. 14 Brazilian Federal Prosecutor’s Specific questions on the SEM DPL relate Office writes to BNDES to BNDES’s use of the funds, and actions President to enquire about the taken to improve BNDES environmental SEM DPL. and social management. April 25 BNDES responds to the Federal States that the funds from the SEM DPL Prosecutor’s Feb. 14 enquiry. were completely transferred to BNDES, on concessionary terms, “to be applied in its investment operations…to complement the BNDES disbursement budget.” ? SEM DPL 2 is canceled The planned second loan in the programmatic series never materializes. Aug. 12 SEM DPL Implementation There is no mention of funds being on-lent Completion and Results Report to BNDES. The Bank team rates the completed. operation “Satisfactory”. ANNEX I 104 Annex I: List of Persons Interviewed Brazilian Officials Name Designation Rodrigo Vieira General Coordinator for External Financing, SEAIN (Secretaria de Assuntos Internacionais), Ministry of Planning Leny Maria Corazza Substitute Coordinator of Social Projects, SEAIN, Ministry of Planning Tania Ribeiro Manager, Environmental Projects, SEAIN, Ministry of Planning Marcus Barreto Coordinator of Institutional Development Projects and State Reform, SEAIN, Ministry of Planning Lilia Maya Cavalcante Coordinator of Social Projects and Substitute coordinator of External Financing, SEAIN, Ministry of Planning Paulo Lopes Varella Director, ANA - Agência Nacional de Águas (National Water Agency) Marcos Neves Advisor, ANA - Agência Nacional de Águas (National Water Agency) Antônio Carlos Hummel General Director, SFB (Serviço Florestal Brasileiro) Daniel Tristão Governmental Manager – Head of International Cooperation Division, SFB (Serviço Florestal Brasileiro) Ana Maria Evaristo Cruz President, ASIBAMA (Associação Nacional dos Servidores do IBAMA) Volney Zanardi Junior President, IBAMA (Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis) Carlos Klink Secretary of Climate Change and Environmental Quality, MMA (Ministério do Meio Ambiente / Ministry for the Environment) Artur Lacerda Deputy Assistant Secretary for International Affairs / Coordinator-General for International Financial Institutions, Ministry of Finance Juliana Torres da Paz Operation Specialist, Treasury, Ministry of Finance Roberto Sainz Researcher at EMBRAPA, Brazil / Professor at the Department of Animal Science, University of California, Davis Roberto Ricardo Vizentin President, ICMBio (Instituto Chico Mendes de Conservação da Biodiversidade) Gustavo Luedemann CDM Designated National Authority, Ministry of Science and Technology Mario Sérgio Vasconcelos Director, International Relations, FEBRABAN (Brazilian Banking Federation) Alessandra Panza FEBRABAN (Brazilian Banking Federation) Paulo Araujo Head of Department of International Division – Institutional Funding and International Relations Department, BNDES Vivian Machado Manager for International Organizations, BNDES 105 ANNEX I Marcio Macedo Head of Environment Department, BNDES Rafael da Silva Andrade BNDES Daniel Soeiro BNDES Ana Paula BNDES Fabricio Barreto Former Program Coordinator, MMA (Ministério do Meio Ambiente / Ministry of Environment) Fátima Soares Manager, Water Quality Department, SEA/RJ (Secretaria do Ambiente do Estado do Rio de Janeiro) Anselmo Frederico Environmental Specialist, SEA/RJ (Secretaria do Ambiente do Estado do Rio de Janeiro) Luiz Firmino Undersecretary Executive, SEA Secretariat for the Environment of Rio de Janeiro State Denise Lobato Superintendent of International Relations, SEA Secretariat for the Environment of Rio de Janeiro State Victor Zveibil Superintendent of Sanitation Policy, SEA Secretariat for the Environment of Rio de Janeiro State Fátima Soares Manager of Water Quality, SEA Secretariat for the Environment of Rio de Janeiro State Anna Cristina Henney Director of Environmental Licensing, INEA/ SEA Secretariat for the Environment of Rio de Janeiro State Justiniano Netto Secretary, Pará State Secreatary for Programa Municípios Verdes Meliza Alves Barbosa Prosecutor, Ministério Público Federal Pessoa Felício Pontes Júnior Prosecutor, Ministério Público Federal, Pará Ubiratan Cazetta Prosecutor, Ministério Público Federal, Pará Helena Palmquist Assessora-Chefe de Comunicacao, Ministério Público Federal, Pará Lilian Regina Furtado Promotora de Justica, Ministério Público, State of Pará Braga Eliane Moreira Promotora de Justica, Ministério Público, State of Pará World Bank Name Designation Mark Lundell Country Director, AFCS2; former Brazil Sector Leader and SEM DPL TTL Garo Batmanian Lead Environmental Specialist, EASCS; former SEM DPL co-TTL Deborah Wetzel Country Director, LCC5C Gregor Wolf Sector Leader, LCSSD Laura Tuck Vice President, ECA Region; former Sector Director, LAC Region John Briscoe Former Country Director, Brazil (retired) Adriana Moreira Senior Environmental Specialist, LCSEN ANNEX I 106 Alberto Ninio Chief Counsel for Environment (currently on leave) Jennifer Sara Sector Manager; former Sector Leader, Brazil Miguel Santiago da Silva Senior Finance Officer Oliveira Isabella Micali Drossos Senior Counsel, LEGAM Kirk Hamilton Lead Environmental Economist Erick Fernandes Adviser Klaus Oppermann Team Leader, Policy and Methodology Team, Climate Policy and Finance Department, Carbon Finance Unit Mauricio Athie Senior Environmental Specialist, IFC, CESI2 (formerly based in Brazil) Alexander Indorf Principal Environmental Specialist, IFC, CESI2 (formerly based in Brazil) Eric Shayer Senior Environmental Specialist, IFC, CESI2 (currently based in Brazil), team member Judith Lisansky Senior Social Specialist (retired) Isabel Braga Senior Environmental Specialist (retired) Robert Schneider Sector Leader (retired) Stephen Lintner Senior Adviser, OPSOR Manuela Francisco Economics Adviser, OPSPQ Qays Hamad Senior Operations Adviser, OPSOR Katia Madeiros Environmental Specialist supporting World Bank (and other IFI) operations, FAO Investment Center Civil Society Organization Representatives Name Designation Oriana Rey Lawyer and advisor of the Eco-Finance Program, Amigos da Terra Amazônia. Member of Amazon Fund Advisory Committee Pedro Bara Neto Infrastructure Strategy Leader, Living Amazon Initiative, WWF Carlos Tautz Coordinator, Instituto Mais Democracia, Rio Adalberto Veríssimo Senior Researcher, IMAZON (Instituto do Homem e Meio Ambiente da Amazônia) Cássio Pereira Researcher, IPAM (Instituto de Pesquisa Ambiental da Amazônia); formerly of MMA and Agency for Development of the Amazon Vincent McElhinny Senior Policy Advisor, Bank Information Center Brent Millikan Amazon Program Director, International Rivers Maíra Irigaray International Finance Advocate and Amazon Watch Brazil Campaign Coordinator, In Casa do Indio Sarah Freeman Amazon Watch Antonia Melo Coordinator, Movimento Xingu Vivo 107 ANNEX I Marcelo Salazar ISA- Instituto Socioambiental, Adjunct Coordinator of Xingu Program Leonardo José Amorim Lawyer, ISA - Instituto Socioambiental, Programa Xingu Jammilye Sales Lawyer, International Rivers/AIDA (Interamerican Association for Environmental Defense) Roberta Amanajás Lawyer, Programa de Acesso à Justiça Internacional SDDH (Sociedade Paraense de Defesa dos Direitos Humanos) Walter Silva Santos CSP – CONLUTAS (Central Sindical e Popular) Attendees at Group Meeting with Community Members, Field Visit, Altamira, Para State Name Designation Juma Xipaia Indigenous leader Maria do Socorro Community member Ivana Community member Moacyr Community member Maria das Graças Fisher, saleswoman Vanusia Brito Community member Ivoneia Community member Mariano Community member Silvia Community member Antonia Fisher, Arara ethnicity Cecilio Kayapó Fisher and Kaiapó Representative Leonardo Kayapó Indigenous leader Maria de Nazaré Barroso Women representative. Movement of Women from BR-1 63 and Midwives Santa Duarte Vieira Quilombola representative Helenilda Community member Soror Maria Vitti Community member Cileanto CIMI (Conselho Indigenista Missionário) Helena Instituto Xingu Vivo Milene Simone Instituto Xingu Vivo Benedita Franciscan Sister Irmã Terezinha Jesus de Franciscan Sister Brito José Bernardo Representative of Altamira’s potters Adam Fest Altamira’s Parish Articulator Caroline Militão Community member Ana Paula Community member Mazeth Community member Sarah Freeman Amazon Watch ANNEX J 108 Annex J: Borrower Comments 109 ANNEX J ANNEX J 110 111 ANNEX J ANNEX J 112 113 ANNEX J ANNEX J 114 115 ANNEX J ANNEX J 116 117 ANNEX J ANNEX J 118 119 ANNEX J ANNEX J 120 121 ANNEX J ANNEX J 122 123 ANNEX J ANNEX J 124 125 ANNEX J ANNEX J 126 127 ANNEX J ANNEX J 128 129 ANNEX J ANNEX J 130 131 ANNEX J ANNEX J 132 133 ANNEX J ANNEX J 134 135 ANNEX J ANNEX J 136 137 ANNEX J ANNEX J 138 139 ANNEX J ANNEX J 140 141 ANNEX J ANNEX J 142 143 ANNEX J ANNEX J 144 145 ANNEX J ANNEX J 146 147 ANNEX J ANNEX J 148 149 ANNEX J ANNEX J 150 151 ANNEX J ANNEX J 152 153 ANNEX J ANNEX J 154 155 ANNEX J ANNEX J 156 157 ANNEX J ANNEX J 158 159 ANNEX J ANNEX J 160 161 ANNEX J ANNEX J 162 163 ANNEX J ANNEX J 164 165 ANNEX J ANNEX J 166 167 ANNEX J ANNEX J 168 169 ANNEX J ANNEX J 170 171 ANNEX J