Document of The World Bank Report No: ICR00002088 IMPLEMENTATION COMPLETION AND RESULTS REPORT (IBRD-73260) ON A LOAN IN THE AMOUNT OF US$57.6 MILLION TO THE FEDERAL DISTRICT WITH THE GUARANTEE OF THE FEDERATIVE REPUBLIC OF BRAZIL FOR A BRASILIA ENVIRONMENTALLY SUSTAINABLE PROJECT June 27, 2012 Sustainable Development Department Brazil Country Management Unit Latin America and the Caribbean Region CURRENCY EQUIVALENTS (Exchange Rate Effective February 24, 2012) Currency Unit = Brazilian Real 1.00 Real = US$ 0.585 US$ 1.00 = 1.708 Reais FISCAL YEAR ABBREVIATIONS AND ACRONYMS ADASA Water, Energy and Sanitation Agencia Reguladora de Agua e Regulatory Agency Saneamento do DF AGEFIS Land Monitoring Agency of the Federal Agência de Fiscalização do Distrito District Federal APA Environmental Protection Area Area de proteção ambiental CAESB Environmental Sanitation Company of Companhia de Saneamento Ambiental the Federal District do Distrito Federal CAS Country Assistance Strategy CEB Electric Company of Brasilia Companhia Energética de Brasília CMU Country Management Unit CODHAB Housing Development Company of the Companhia de Desenvolvimento Federal District Habitacional do Distrito Federal COFIEX Commission for External Debt Comissão de Financiamentos Externos CPS County Partnership Strategy CRAS Social Assistance Centers Centros de Referência da Assistência Social DF Federal District Distrito Federal DO Development Objective EIRR Economic Internal Rate of Return ETE Sewage Treatment Station Estação de Tratamento de Esgoto GDF Government of the Federal District Governo do Distrito Federal GEF Global Environment Facility GIS Geographic Information System GTEC COFIEX Technical Group Grupo Técnico da COFIEX HABI Superintendent for Social Housing Superintendência de Habitação Popular IBAMA National Environmental Institute Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis IBRAM GDF’s Environmental Institute Instituto Brasília Ambiental ICR Implementation Completion and Results Report IP Implementation Performance IPTU Property Tax Imposto Predial e Territorial Urbano IQA Water Quality Index �ndice de Qualidade das �guas ISA Social Environmental Indicator Indicador de Salubridade Ambiental ISR Implementation Status Report M&E Monitoring and Evaluation MU Moderately Unsatisfactory NOVACAP Urbanization Company of the New Companhia Urbanizadora da Nova Capital of Brazil Capital do Brasil NPV Net Present Value O&M Operations and Maintenance OP Operational Policy PAD Project Appraisal Document PDO Project Development Objective PM Military Police Polícia Militar PMU Project Management Unit Unidade de Gestão do Projecto PPP Public-Private Partnership RAP Resettlement Action Plan RIDE Integrated Development Region Regiao Integrada de Desenvolvimento Economico SEAIN Department of International Affairs of Secretaria de Assuntos Internacionais the Ministry of Planning, Budget and Management SEAPA Secretariat for Agriculture, Fishing, and Secretaria de Agricultura e Pecuária e Supplies Abastecimento SEDEST Secretariat of Social Development and Secretaria de Desenvolvimento Social e Labor Trabalho SEDUMA Secretariat of Urban Development and Secretaria de Estado de Environment Desenvolvimento Urbano e Habitação SIL Specific Investment Loan SLU Urban Cleaning Service of the Federal Serviço de Limpeza Urbana do Distrito District Federal SUDESA Sub-secretariat for the Defense of Land A Subsecretaria de Defesa do Solo e da and Water �gua TERRACAP Brasilia’s Real Estate Company Companhia Imobiliária de Brasilia U Unsatisfactory WTP Willingness to Pay ZEE Economic–Environmental Zoning Zoneamento Ecológico – Econômico ZHIS Residential Zone of Social Interest Zona Habitacional de Interesse Social Vice President: Hasan A. Tuluy Country Director: Deborah L. Wetzel Sector Manager: Wambui G. Gichuri Project Team Leader: Sameh Naguib Wahba ICR Team Leader: Sameh Naguib Wahba   BRAZIL BRASILIA ENVIRONMENTALLY SUSTAINABLE PROJECT Datasheet 1. Project Context, Development Objectives and Design ................................................... 2 2. Key Factors Affecting Implementation and Outcomes .................................................. 6 3. Assessment of Outcomes .............................................................................................. 14 4. Assessment of Risk to Development Outcome ............................................................. 21 5. Assessment of Bank and Borrower Performance ......................................................... 21 6. Lessons Learned............................................................................................................ 24 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners............... 26 Annex 1. Project Costs and Financing .............................................................................. 27 Annex 2. Outputs by Component...................................................................................... 28 Annex 3. Economic and Financial Analysis ..................................................................... 30 Annex 4. Bank Lending and Implementation Support/Supervision Processes................. 34 Annex 5. Beneficiary Survey Results ............................................................................... 36 Annex 6. Stakeholder Workshop Report and Results ....................................................... 38 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR ......................... 39 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders ........................... 53 Annex 9. List of Supporting Documents .......................................................................... 54 MAP .................................................................................................................................. 55   A. Basic Information Brasilia Country: Brazil Project Name: Environmentally Sustainable Project Project ID: P089440 L/C/TF Number(s): IBRD-73260 ICR Date: 06/27/2012 ICR Type: Core ICR GOVERNMENT OF Lending Instrument: SIL Borrower: THE FEDERAL DISTRICT Original Total USD 57.64M Disbursed Amount: USD 44.88M Commitment: Revised Amount: USD 57.64M Environmental Category: A Implementing Agencies: ADASA (CAESB and NOVACAP co-executing agencies) Cofinanciers and Other External Partners: B. Key Dates Revised / Actual Process Date Process Original Date Date(s) Concept Review: 06/29/2004 Effectiveness: 05/16/2006 05/16/2006 12/17/2010 Appraisal: 03/07/2005 Restructuring(s): 03/31/2011 Approval: 08/25/2005 Mid-term Review: 08/05/2009 Closing: 03/31/2011 12/31/2011 C. Ratings Summary C.1 Performance Rating by ICR Outcomes: Unsatisfactory Risk to Development Outcome: Substantial Bank Performance: Moderately Unsatisfactory Borrower Performance: Unsatisfactory C.2 Detailed Ratings of Bank and Borrower Performance (by ICR) Bank Ratings Borrower Ratings Moderately Quality at Entry: Government: Unsatisfactory Unsatisfactory Implementing Moderately Quality of Supervision: Moderately Satisfactory Agency/Agencies: Unsatisfactory Overall Bank Moderately Overall Borrower Unsatisfactory Performance: Unsatisfactory Performance: i C.3 Quality at Entry and Implementation Performance Indicators Implementation QAG Assessments Indicators Rating Performance (if any) Potential Problem Project Quality at Entry No None at any time (Yes/No): (QEA): Problem Project at any Quality of Yes None time (Yes/No): Supervision (QSA): DO rating before Unsatisfactory Closing/Inactive status: D. Sector and Theme Codes Original Actual Sector Code (as % of total Bank financing) Central government administration 5 General water, sanitation and flood protection sector 70 77 Sub-national government administration 25 23 Theme Code (as % of total Bank financing) Environmental policies and institutions 29 4 Municipal governance and institution building 14 19 Other social development 14 3 Other urban development 14 12 Pollution management and environmental health 29 62 E. Bank Staff Positions At ICR At Approval Vice President: Hasan A. Tuluy Pamela Cox Country Director: Deborah L. Wetzel Vinod Thomas Sector Manager: Wambui G. Gichuri Abel Mejia Project Team Leader: Sameh Naguib Wahba Paula Dias Pini ICR Team Leader: Sameh Naguib Wahba ICR Primary Author: Catherine Lynch F. Results Framework Analysis Project Development Objectives (from Project Appraisal Document) The Project's Development Objective is to reduce regional inequalities and ensure quality water resources to meet the growing needs of the Federal District and the Brasilia ii Metropolitan Region, through environmental planning and management activities, integrated with poverty reduction interventions and environmental rehabilitation in the most critical river basins. Specifically, the expected outcomes of the project are: (a) improved water quality of the Vicente Pires and Descoberto Rivers; (b) improved living conditions in targeted low- income settlements by providing integrated urban environmental services; and (c) strengthened technical and institutional capacity, contributing to development of the Brasilia metropolitan region. Revised Project Development Objectives (as approved by original approving authority) (a) PDO Indicator(s) Original Target Formally Actual Value Values (from Revised Achieved at Indicator Baseline Value approval Target Completion or documents) Values Target Years Indicator 1 : Water quality of the Vicente Pires River improved and/or maintained to Class 2 Value No compliance with IQA No compliance quantitative or IQA class 2 class 2 with IQA class 2 Qualitative) Date achieved 12/31/2004 12/31/2011 12/31/2011 Sewerage works in Vicente Pires will benefit 70k low-income residents. Works, Comments financed by the counterpart, are 60% completed. Improved water quality will (incl. % only be measurable once completed. Expected that 80% of the connections will achievement) be completed by 12/2012. Improved living conditions in Vila Estrutural by providing integrated urban environmental services measured by improvement in the subindicators of the Indicator 2 : area Social Environmental Indicator (ISA) and by end-of-project satisfaction survey % improvement Survey conducted in in May 2012, but the indicator final results not yet compared to Value complete. However, baseline quantitative or Set of 46 indicators None significant Qualitative) improvements in 70% of living conditions families evident based on satisfied with preliminary data. Program results Date achieved 12/31/2004 03/31/2011 12/31/2011 05/31/2012 Comments 40k residents benefited from activities in Vila Estrutural. The ISA-F indicator (incl. % increased from 39% in 2008 to 88% in 2011. A beneficiary survey was carried iii achievement) out in May 2012, and the final results are forthcoming. (b) Intermediate Outcome Indicator(s) Original Target Actual Value Formally Values (from Achieved at Indicator Baseline Value Revised approval Completion or Target Values documents) Target Years Indicator 1 : ADASA strategic plan endorsed Value Plan (quantitative No plan n.a. completed and No plan or Qualitative) endorsed Date achieved 12/31/2004 03/31/2011 12/31/2011 12/31/2011 Comments (incl. % The procurement process for the strategic plan was canceled. achievement) Indicator 2 : SLU restructuring proposal adopted by the GDF Value Proposal (quantitative No proposal n.a. completed and Proposal completed or Qualitative) adopted Date achieved 12/31/2004 03/31/2011 12/31/2011 05/31/2012 Comments (incl. % achievement) Indicator 3 : New planning support instruments adopted by the GDF Value 4 instruments 3 instruments (quantitative 0 n.a. adopted adopted or Qualitative) Date achieved 12/31/2004 03/31/2011 12/31/2011 05/31/2012 Adopted instruments: (i) mgmt and protection of wetlands; (ii) survey and Comments treatment of sewage system liabilities; and (iii) strengthening of environ (incl. % licensing procedures at IBRAM. Final instrument is upgrading plan for San achievement) Sebastian (expected Nov 2012). Percent of population with access to paved roads, stormwater drainage system in Indicator 4 : Vila Estrutural Value (quantitative 0% n.a. 100% 84.6% or Qualitative) Date achieved 12/31/2004 03/31/2011 12/31/2011 05/31/2012 Comments Works were completed as of May 2012. About 40,000 low-income residents will (incl. % benefit from these improvements. achievement) Indicator 5 : Population connected to sewerage system Value (quantitative 0% n.a. 100% 92.2% or Qualitative) Date achieved 12/31/2004 03/31/2011 12/31/2011 05/31/2012 Comments Works were completed as of May 2012. About 40,000 low-income residents will iv (incl. % benefit from these improvements. achievement) Indicator 6 : Provision of additional educational and community facilities in Vila Estrutural Value All planned All except for 1 (quantitative No facilities n.a. facilities school or Qualitative) completed Date achieved 12/31/2004 03/31/2011 12/31/2011 12/31/2011 Comments Completed facilities include a pre-school, a community center, and a social (incl. % services center. Pending resettlement of 5 households has delayed the achievement) construction of one primary school. Indicator 7 : Number of scavengers benefiting from GDF-supported recycling program Value (quantitative 0 n.a. 800 0 or Qualitative) Date achieved 12/31/2004 03/31/2011 12/31/2011 12/31/2011 Comments It is estimated that support to the scavengers (training and equipment) and the (incl. % construction of the recycling center where they will work could all be completed achievement) by December 2012. The procurement process for the triage center is in progress. Indicator 8 : Number of household connections to the sewerage system in Vicente Pires Value (quantitative 0 15,700 15,700 0 or Qualitative) Date achieved 12/31/2004 03/31/2011 12/31/2011 12/31/2011 Comments 60% of works were completed as of December 31, 2011. It is expected that 80% (incl. % of the connections (12,560) will be completed by December 2012—benefiting achievement) about 56,000 residents. Environmental remediation activities of the unused areas of Joquei landfill Indicator 9 : launched Value 1/3 of landfill (quantitative No remediation n.a. No remediation remediated or Qualitative) Date achieved 12/31/2004 03/31/2011 12/31/2011 12/31/2011 Comments (incl. % This activity was cancelled. achievement) Procurement process of the new sanitary landfill of the Indicator 10 : GDF finished Value (quantitative No process n.a. Completed No process or Qualitative) Date achieved 12/31/2004 03/31/2011 12/31/2011 12/31/2011 Process launched several times (challenges to process and political instability). Comments Decision taken mid-2011 to build with public funds and tender operation and (incl. % maintenance to the private sector. Detailed designs and bidding documents achievement) completed May 2012. v G. Ratings of Project Performance in ISRs Actual Date ISR No. DO IP Disbursements Archived (USD millions) 1 05/30/2006 Satisfactory Satisfactory 0.14 2 12/27/2006 Moderately Satisfactory Moderately Satisfactory 0.14 3 06/13/2007 Moderately Satisfactory Moderately Satisfactory 0.14 4 10/26/2007 Moderately Satisfactory Moderately Satisfactory 3.95 5 04/14/2008 Moderately Satisfactory Moderately Satisfactory 3.95 6 11/20/2008 Satisfactory Satisfactory 21.43 Moderately 7 05/09/2009 Unsatisfactory 27.88 Unsatisfactory 8 10/19/2009 Moderately Satisfactory Moderately Satisfactory 32.97 Moderately 9 05/10/2010 Unsatisfactory 36.42 Unsatisfactory Moderately 10 02/16/2011 Unsatisfactory 37.06 Unsatisfactory Moderately 11 07/19/2011 Unsatisfactory 38.18 Unsatisfactory Moderately 12 12/26/2011 Unsatisfactory 44.88 Unsatisfactory H. Restructuring (if any) ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions The restructuring encompassed the following proposed changes: (i) transfer of the project implementation responsibility to a new agency; (ii) redefinition of the Project description to reflect the removal of some activities and reduction in the 12/17/2010 N MU U 36.42 scope of Part C of the Project; (iii) modification of the results indicators to reflect the changed activities; (iv) change in the reallocation of loan proceeds between categories and the percentage of eligible expenditures to be financed; (v) addition of Pregão Eletronico vi ISR Ratings at Amount Board Restructuring Disbursed at Restructuring Reason for Restructuring & Approved Restructuring Date(s) Key Changes Made PDO Change DO IP in USD millions to the applicable procurement methods; and (vi) extension of the implementation schedule for selected counterpart-funded activities. These proposed changes would allow the Federal District to meet the Project Development Objective (PDO) despite recent changes to the local government leadership and implementation challenges. Extension of closing date to 03/31/2011 U MU 37.06 December 31, 2011. I. Disbursement Profile vii   Executive Summary The objective of the Project was to reduce regional inequalities and ensure the availability of high quality water resources to meet the growing needs of Brasilia. It was envisioned that this would be accomplished through three primary actions: (i) installation of sewerage systems in key areas where water bodies were being polluted; (ii) closure and remediation of an overcapacity landfill; and (iii) upgrading of informal resettlement adjacent to a national park and water basin. While the design of the Project largely incorporated best practices in terms of a socially inclusive and integrated approach to environmental protection, there was insufficient consideration of the Borrower's limited institutional capacity and of the complexity of inter-agency coordination in a then uncharted area of intervention for the Borrower, namely slum upgrading. This led to multiple challenges during the implementation of the Project. First, part of the sewerage works located in an adjacent state needed to be dropped from the Project, thereby reducing its regional impact, and the remaining sewerage works faced major delays. Second, the closure and remediation of the landfill has not commenced due to political economy constraints related to the development of a new landfill under a concession regime. Finally, the slum upgrading works were impacted by recurrent problems with resettlement (including the inclusion of the Bank in a lawsuit filed by affected families). In the midst of these implementation issues, there were multiple changes in GDF leadership, including an impeachment of the Governor. After much deliberation and attempts at resolving the situation, the Bank was forced to suspend disbursements for more than 16 months, causing further implementation delays and institutional disarray. The above notwithstanding, the Project has had several important achievements, such as: (i) about 40,000 low-income residents in Vila Estrutural are benefitting from upgraded infrastructure and community facilities; (ii) about 70,000 low-income residents in Vicente Pires will benefit from sewerage services by the end of 2012 when the works are concluded; and, (iii) over 1 million residents of the DF will benefit from improved water quality by mid-2013 when the system enters into operation. The achievements related to these activities are aligned with the original Project targets. Both the Bank and the Project Implementation Unit (PIU) from the GDF put in extra effort to keep the Project moving forward. On the Bank side, the supervision budget was significantly increased to allow for expanded oversight and implementation support, senior management was actively engaged, the communications team from the CMU played a strong role in supporting the dialogue with complainants among affected families; and the project was twice restructured to adjust to evolving circumstances and to enable it to progress towards the PDO. On the GDF side, the PIU pressed forward with activities despite continuous institutional obstacles and uncertainties beyond its realm of control, and the current Governor (in office since late 2011) has committed to funding and completing all outstanding activities after the closing date using counterpart funds, including the Project Management Unit. The GDF is currently seeking external funding for Phase II, which is an indication of strong Government commitment to the Project and its objectives. Because of the multiple challenges faced by the Project, it has been possible to extract several important lessons for the design of future operations, including the following:  As part of project preparation, more emphasis should be put on assessing institutional capacity and designing arrangements and mitigation measures to carry out resettlement;  Although an integrated approach to environmental sanitation is considered good practice, it may be too complex for some clients;  The assessment of political economy issues should be a key part of the preparation in complex/integrated development projects; and 1  Urban upgrading projects are best managed by one agency with delegated authority for all activities within the target area. 1. Project Context, Development Objectives and Design 1.1 Context at Appraisal Urban environment issues in large urban areas of Brazil are imposing increasingly large costs to society in terms of additional infrastructure to guarantee reliable water supply, address health impacts of pollution, and reduce damage associated with flooding. To address these urban environment issues, a long-term comprehensive strategy is required. This is the case for the largest metropolitan areas of Brazil like São Paulo, Belo Horizonte, Curitiba, and Recife, but it is also an emerging demand from fast growing cities such as Brasilia, and medium size cities like Betim, Uberaba, and Cubatão that are experiencing similar urban environmental issues. At the time of appraisal, Brasilia was already the tenth largest urban concentration in Brazil, and despite its well-known plan for urban development, it showed high levels of socioeconomic inequality, uncontrolled urban expansion, and environmental degradation, which are found in most of the largest metro areas of the country. In addition, the urban sprawl in Brasilia occurs at the expense of the degradation of the most stressed ecosystem of the country, the Cerrado, which is compromising its natural capacity to provide key environmental services to support sustainable metropolitan expansion. High population growth, rising water demand, and unabated pollution in many unplanned settlements within the Federal District (DF) and its Integrated Development Region (RIDE) were adversely impacting the quantity and quality of water resources. Left unchecked, these impacts could potentially impose a high cost on the expansion of basic services, erode the quality of life and health of the most vulnerable population, and ultimately undermine the sustainable development of the region. Brasilia’s environmental importance lies in two main factors: a) its location at the head waters of three key river basins of national and regional importance (Parani, Tocantins - Araguaya, and São Francisco); and b) as a reservoir of biodiversity in the Cerrado ecosystem. Additionally, 43% of the DF territory is made up of environmental conservation units. Promoting socially and environmentally sustainable development in Brasilia has been part of a national development strategy to foster sustainable economic growth in the center-west region of the country. Undoubtedly, a substantial part of this strategy had been achieved over the fifty year history of Brasilia. At the time of appraisal, this strategy required a renewed policy consensus on urban and regional environmental issues to more effectively contribute to the control and abatement of high rates of environmental degradation affecting the capacity and cost of providing urban and regional services. The DF requested Bank lending assistance to formulate and implement a sustainable development strategy addressing priority environmental and social development issues. The DF also sought Bank assistance in order to have access to grant resources to deal with global environment issues associated with the Cerrado ecosystem and climate change. Specifically, the DF wanted to blend the proposed lending operation with a GEF grant that would seek sustainable productive landscape and biodiversity conservation in the Cerrado.1 Additionally, the DF was keen to build a partnership with the Bank to access carbon credits in exchange for improved solid waste management. 1 The GEF grant for Brasilia did not occur because of a preference to focus on other areas in Brazil. 2 1.2 Original Project Development Objectives (PDO) and Key Indicators (as approved) The original PDO was to reduce regional inequalities and ensure quality water resources to meet the growing needs of the Federal District and the Brasilia Metropolitan Region, through environmental planning and management activities, integrated with poverty reduction interventions and environmental rehabilitation in the most critical river basins. The expected outcomes of the project were: (a) improved water quality of the Vicente Pires and Descoberto Rivers; (b) improved living conditions in targeted low-income settlements by providing integrated urban environmental services; and (c) strengthened technical and institutional capacity, contributing to development of the Brasilia metropolitan region. 1.3 Revised PDO (as approved by original approving authority) and Key Indicators, and reasons/justification In December 2010, the Project was restructured for the first time, including: (i) transfer of the project implementation responsibility to a new agency; (ii) redefinition of the Project description to reflect the removal of some activities and reduction in the scope of Part C of the Project; (iii) modification of the results indicators to reflect the changed activities; (iv) change in the reallocation of loan proceeds between categories and the percentage of eligible expenditures to be financed; (v) addition of Pregão Eletronico to the applicable procurement methods; and (vi) extension of the implementation schedule for selected counterpart-funded activities. In addition, a second restructuring in March 2011 provided for an extension of the closing date to December 31, 2011. The outcome indicators were revised to: 1. Water quality of the Vicente Pires River improved and/or maintained to Class 2; and 2. Improved living conditions in Vila Estrutural by providing integrated urban environmental services measured by improvement in the subindicators of the area Social Environmental Indicator (ISA) and by an end-of-project satisfaction survey. The justification for restructuring included: (i) delays in launching implementation due to legal and institutional issues; and (ii) the suspension of disbursement due to non-compliance with some provisions of the Loan Agreement related to resettlement safeguards and a unilateral change of implementing agency. Detailed explanation of these issues are included in Section 2, Key Factors Affecting Implementation and Outcomes. Table: Original and Revised Project Data Original Revised Outcome Indicators  Improved to IQA class 2 the water quality  Water quality of the Vicente Pires River of the Vicente Pires and Descoberto rivers improved and/or maintained to Class 2  Improved living conditions in Vila  Improved living conditions in Vila Estrutural by providing integrated urban Estrutural by providing integrated urban environmental services measured by environmental services measured by increasing the area Social Environmental improvement in the subindicators of the Indicator (ISA) area Social Environmental Indicator (ISA)  Generated technical and institutional and by end-of-project satisfaction survey capacity contributing to the Brasilia metropolitan region development 3 Intermediate Results Component 1: Component 1:  High quality technical assistance provided  ADASA strategic plan endorsed  Well targeted institutional capacity  SLU restructuring proposal adopted by the activities carried out GDF  New planning support instruments adopted Component 2: by the GDF  Vila Estrutural Social Environment Indicator (ISA) improved Component 2:  Joquei Club solid waste pickers working  Percent of population with access to paved in improved social and environmental roads, stormwater drainage system in Vila conditions Estrutural  Population connected to sewerage system Component 3:  Provision of additional educational and  Aguas Lindas: # households connected to community facilities in Vila Estrutural sewerage systems  Number of scavengers benefiting from  Vicente Pires: # households connected to GDF-supported recycling program sewerage systems  Closed Joquei Clube landfill Component 3:  Technically and Environmentally sound  Number of household connections to the sanitary landfill constructed sewerage system in Vicente Pires  Environmental remediation activities of the unused areas of Joquei landfill launched  Procurement process of the new sanitary landfill of the GDF finished Project Components Described in section 1.6 below. Procurement Methods Allowed Quality- and Cost-based Selection; Least-cost Quality- and Cost-based Selection; Least-cost Selection; Selection Based on Consultants’ Selection; Selection Based on Consultants’ Qualifications; Individual Consultants Qualifications; Individual Consultants; Pregão Eletronico Closing Date March 31, 2011 December 31, 2011 1.4 Main Beneficiaries The Project Appraisal Document indicates that the original project aimed to benefit about 175,000 low-income residents in Aguas Lindas and Vila Estrutural.2 It was also envisioned that it would contribute to the preservation of two important sources of potable water and recreation in the DF: (i) the lake formed by the dam on the Descoberto river, which supplies water to nearly 65% of the DF (about 1.34 million people); and (ii) the Paranoa lake, which constitutes a reserve source of potable water and recreation center. 2 Based on the data included in the economic analysis, the beneficiaries in Aguas Lindas, Vicente Pires, and Vila Estrutural totaled more than 200,000 residents. 4 1.5 Original Components (as approved) Component 1: Policy and Institutional Development (US$7.6 mil). The main objective of Component 1 is to support institutional strengthening and technical assistance for the enhancement of the technical and institutional environment enabling the Government of the Federal District (GDF) to undertake informed decision-making with respect to metropolitan issues, especially for promoting a more sustainable urban and environmental development. Subcomponent 1.1. Territory Development and Land Management (US$2.6 mil): This subcomponent encompasses four main thematic areas related to territory, urban, housing and land use policies and institutional enhancement. Subcomponent 1.2. Environmental and Water Resources Management (US$3.9 mil): This subcomponent addresses six thematic areas related environmental and water resources management. Subcomponent 1.3. Environmental Sanitation Management (US$l.0 mil): This subcomponent addresses four thematic issues related to environmental sanitation services. Component 2: Social Inclusion and Poverty Reduction (US$31.4 mil). The main objectives of Component 2 comprise reducing poverty and promoting social inclusion in localities presenting critical social conditions, coupled with actions to abate water resource pollution loads. Sub-component 2.1. Integrated Development Project for Vila Estrutural ($29.9 mil). This subcomponent is comprised of two activities: (i) planning, technical designs and community participation; and (ii) civil works and participation activities. Subcomponent 2.2. Support to the Joquei Clube Solid Waste Landfill Waste pickers (US$1.4 mil). Component 3: Water Resources Protection (US$71.5 mil). The main objective of Component 3 is to execute activities that will abate water resource pollution loads and improve the quality of life in the communities of Aguas Lindas and Vicente Pires, as well as protect the Brasilia National Park through solid waste disposal activities. Subcomponent 3.1. Improving sanitation services in Aguas Lindas (US$33.6 mil). Subcomponent 3.2. Improve Water and Sanitation Services in Vicente Pires (US$19.4 mil). Subcomponent 3.3. Closure of the Joquei Club Solid Waste Sanitary landfill (US$ 8.3 mil). Subcomponent 3.4. Construction of a solid waste sanitary landfill (US$3 mil). Subcomponent 3.5. Rehabilitating environmentally degraded areas (US$2.2 mil). Subcomponent 3.6. Waste water sludge treatment (US$5.0 mil). Component 4: Project Management, Monitoring and Evaluation (US$3.9 mil). The main objective of this component is support the appropriate technical, administrative and financial management conditions to ensure an effective project implementation and achievement of its development objectives. 5 1.6 Revised Components The following are activities that were modified or removed as part of the project restructuring: Component 1:  Subcomponent 1.1, include capacity building activities for ADASA and remove item related to the improvement of the Borrower’s territory development and land management policies.  Subcomponent 1.2, include the development of management tools for SEDUMA and replace IBAMA (the National Environmental Institute) by IBRAM (the Borrower’s Environmental Institute) following the decentralization of environmental responsibilities.  Subcomponent 1.3, include SEDUMA and ADASA as beneficiaries of technical assistance activities.  Subcomponent 1.4, remove the establishment of a sanitation company for adjacent areas of the Borrower’s territory; and include two new activities (i) technical assistance to improve the management and protection of water sources; and (ii) identify and remediate environmental impacts caused by the construction of sewerage systems in the Borrower’s territory. Component 3:  Removal of Subcomponent 3.1.  Subcomponent 3.3, revised to include the closure and environmental recuperation of only one-third of the existing Joquei Clube solid waste landfill, rather than its complete closure as initially planned.  Subcomponent 3.4, revised to include the launch of the procurement process for the construction of the new landfill, rather than an operational new landfill.  Removal of Subcomponent 3.5. 1.7 Other significant changes Key activities implemented with counterpart funds will be completed after the closing date and completion of the ICR. Specifically, the sewage treatment plant in Vicente Pires has an estimated completion date of December 31, 2012. This affects the achievement of the Outcome Level Indicator: Water quality of the Vicente Pires River improved and/or maintained to Class 2. It is estimated that the improved quality of the river will not be fully achieved until approximately 6 months after the completion of the works, or June 2013. 2. Key Factors Affecting Implementation and Outcomes 2.1 Project Preparation, Design and Quality at Entry The project design benefited from an integrated approach combining sanitation, solid waste management, and slum upgrading interventions in the objective of improving water resources quality, and the design aimed to integrate lessons learned in the country and sector. However, a key weakness in the design of the project was the underestimation of the risks related to the institutional complexity of the GDF, and the carrying out of an important resettlement program by a counterpart (Borrower and implementation agencies) who had no experience with urban upgrading which also lacked an adequate regulatory framework. The project design adequately integrated lessons learned in the country and sector, especially in regard to urban environmental protection, and social inclusion. At the time that the project was being prepared, the Bank had financed state and municipal projects over a number of decades in Brazil with the dual objective of reversing the lack of basic urban services and of combating the resulting negative impacts on the urban environment and on quality of life for city inhabitants. 6 These projects provided key lessons such as the importance of integrated basic service provision in densely populated and high risk areas, especially those along river courses where the poor invariably settle, in order to improve the living conditions of most vulnerable communities while reducing urban water pollution. There had also been a positive track record of urban upgrading projects in Brazilian municipalities at the time the project was designed, including among others the Recife Urban Project (Programa Prometropole), the São Paulo Water Quality and Pollution Control Project (Programa Guarapiranga), and Prosanear (targeting multiple municipalities). These projects provided lessons such as the importance of community participation, the need to create community safe spaces, and the importance of integrating education and health services, and income generation, youth, and leisure programs. Experience with urban projects in Brazil and other countries also highlighted the need to include the most excluded populations, in this case families with no regularized land titles who occupy high environmental risk areas and scavengers working at and living near the Joquei Clube dumpsite. The project team considered that institutional complexity would be a challenge in relation to metropolitan management and environmental licensing, but underestimated its impact on resettlement activities, the development of the new landfill, and the sewerage works in Aguas Lindas. The project team indicated that institutional issues would be the main challenge for the project proposal. First, implementation of the institutional component was viewed as a challenge in itself because it would deal with the complexity of the metropolitan region. Second, the complexity of the environmental institutional framework was foreseen as a challenge because obtaining the necessary environmental licenses could require significant effort and time. In hindsight, the project team underestimated the impact of the GDF’s institutional complexity on resettlement activities. The Bank’s experience with urban upgrading projects was primarily at the municipal level of government. The structure of the GDF is a unique case in Brazil because it functions as both a state and a municipality (it has 29 administrative regions). Urban upgrading projects typically require inputs from multiple municipal agencies or secretariats—water, sanitation, transport, works, and social development, and housing. The Bank’s experience with urban upgrading projects in Brazil relied on the power of the municipal executive (i.e., the mayor) to transfer the implementation authority for these activities to one secretariat or to coordinate the activities through the executive office. At the municipal level, this process is facilitated by the fact that (i) the executive directly appoints the secretaries and (ii) they are usually from the same political party. At the state level in Brazil, the executive (i.e., the governor) is often elected as the result of a political coalition if no single party can achieve a majority. The trade-off is that leadership of the secretariats is assigned to the political parties comprising the coalition. This can have the effect of making each secretariat an independent body that may not fully align with and respond to the executive as efficiently as at the municipal level, nor necessarily work effectively with peer agencies and Secretariats due to differences in political views and at times power disputes. This, unfortunately, became the situation in Brasilia after a change in gubernatorial leadership early in project implementation, even with the signing of agreements with each co- executing agency involved with project implementation. For instance, the slum upgrading activities were fragmented among multiple agencies and multiple works contracts. The upgrading of Vila Estrutural was an exception to the GDF’s policy on irregular settlements; the risk related to the absence of an institutional and legal framework for resettlement was not foreseen. The GDF’s policy for addressing unregulated low-income settlements was historically to demolish houses and resettle the population in regularized subdivisions. As a result, the DF had a lower percentage of population living in slums than any other Brazilian metropolitan region. At 7 project appraisal, Vila Estrutural was by far the largest and most critical settlement, in both social and environmental terms. Previous DF governments had tried to demolish the settlement without success, and the environmental agencies also were pushing for the full resettlement of Vila Estrutural to be carried out. Against such history, the decision to upgrade Vila Estrutural was an exception under GDF policy. A law passed in 2006 declared Vila Estrutural a Residential Zone of Social Interest (Zona Habitacional de Interesse Social—ZHIS). At the time of project design, it was believed that this provided a sufficient legal underpinning for the urban upgrading and resettlement activities. However, the creation of the ZHIS only provided a legal framework governing the urbanization of the area, namely the infrastructure works and public facilities. The GDF’s housing policy and regulatory framework (a chapter in the State’s Organic Law/constitution) served as the legal reference to implement resettlement activities, due to the absence of a policy and provisions related to involuntary resettlement and the inexistence of prior experience and de facto procedures to refer to. The DF does not have a policy on resettlement of those affected by development projects, and there is no legal framework with the specific objective of restoring sources of income and livelihoods of persons displaced by development projects. 3 Similarly, the DF does not have legislation that requires compensation at full replacement cost or the provision of assistance for resettling people who are physically displaced, as OP 4.12 requires. In addition, there was a culture in the GDF’s staff, especially at the housing company, against this option with the rationale of not creating incentives for new invasions. This type of situation is not unique to Brasilia. The law that designated Vila Estrutural as a ZHIS (Lei Complementar Nº 715, de 24 de janeiro de 2006) mandated the Executive Branch of the GDF to relocate occupants from areas of environmental risk, areas that would be affected by infrastructure works, or areas needed to complete the urban design plan, but did not stipulate any terms of compensation or resettlement. However, a unique aspect of the GDF is that there was a history of giving land and housing to residents as a means of attracting them to the new capital. In fact, about 1/3 of affected families in Vila Estrutural who were required to be resettled in accordance with the ZHIS had already received a house through other government programs. Under the GDF’s housing policy, these households were ineligible to receive another house from the GDF even though they needed to move from their current location. The Resettlement Action Plan (dated May 2008) envisioned that all affected households would be resettled to new houses (their compensation) before any works or clearing occurred at their previous dwelling, thereby avoiding any need for temporary resettlement. Yet, at the same time, social rental (the payment of a rental allowance for temporary displacement during construction) was also obliquely referred to in the RAP, but it did not have defined duration, eligibility terms, budgetary allocation, or responsible agency assigned with its implementation. Moreover, social rental was not at the time a defined instrument of the Secretariat that ended up assigned with its implementation. Finally, the fact that the various GDF agencies involved didn’t actually consider the RAP as the guiding document for resettlement created great difficulties. It had not been effectively discussed and incorporated into the program. The planned solid waste activities were overly ambitious and the institutional capacity and complexity were not adequately considered. The project was too ambitious regarding the construction of a new landfill and the closure of the existing dumpsite. It underestimated the 3 At the same time, there was no Federal-level regulatory framework that could apply in this case. 8 GDF’s complicated institutional arrangements, weak institutional capacity at SLU, and the complicated political economy of solid waste in Brazil. The time needed to prepare the design and finalize the ownership structure for the new land fill (use of PPP type, consortium, regular bidding, etc.), to level the commitment from the Government to follow through a PPP structure, and the projected costs were underestimated. Finally, the counterpart-funded works in Aguas Lindas proved very difficult to coordinate due to the fact that Aguas Lindas Municipality is administratively under the State of Goias, and thus falls outside of the Borrower’s jurisdiction. Specifically, land acquisition for the construction of the pumping stations and the waste water treatment plant could not be reasonably completed within the timeframe of the project, and the Borrower’s attempts to engage with the State and Municipal counterparts to resolve the delays were not successful. This activity was dropped from the Project as part of the restructuring. The improved quality of the Descoberto River was linked to sewerage works in Aguas Lindas, and was therefore also dropped from the Project. 2.2 Implementation Delays in launching implementation. The Project was approved by the Board on July 5, 2005. Following the signing of the Loan Agreement on February 23, 2006, the Project became effective on May 16, 2006. Project implementation was slow during 2006 because of elections in the GDF. Moreover, as a result of the elections, a major institutional restructuring was carried out by the new Government. Therefore, the Borrower asked for a first amendment of the loan agreement to make the project implementation arrangement consistent with its institutional reorganization. The amended and restated loan agreement was signed on May 21, 2007 and a new effectiveness letter was sent on July 27, 2007. The project had recurrent resettlement issues during implementation, due largely to institutional coordination problems. Five different GDF secretariats were responsible for various aspects of the resettlement program, which was forecasted to involve 903 households and businesses. As per governor decree (27.918 of May 3, 2007), the State Secretariat for Urban Development and Environment (SEDUMA) was responsible for the general supervision of the project. Follow-up decrees stipulated that aspects of the project related to upgrading, environment, and housing (including resettlement) would be coordinated by the Housing Development Company of the Federal District (CODHAB), social aspects would be coordinated by the State Secretariat for Social Development and Income Transfers (SEDEST), and infrastructure works would be coordinated by the State Secretariat for Works through the works company (NOVACAP), as well as, the land regularizing role of TERRACAP. Finally, the processing of the resettlement of the rural residents to Monjolo would be coordinated by the State Secretariat for Agriculture, Fishing, and Supplies (SEAPA). By contrast, in most Brazilian municipalities, one agency has the delegated authority to oversee design and works in the upgrading area (e.g., the Superintendency for Popular Housing (HABI) in Sao Paulo). Initially, the Project Management Unit (PMU) only had a relationship with one of these, SEDUMA, which did not have the institutional power to influence or even monitor the work of the four other secretariats (despite formal agreements signed between the entities to implement the project). Many of the resettlement problems were a reflection of this institutional failure. Volatility/uncertainty in GDF leadership exacerbated the risks related to the multi-agency implementation model, resulting in further delays and coordination problems. Coinciding with this period of implementation difficulties, a period of political uncertainty and transition started in November 2009 following the corruption charges against the GDF Governor and some members of the government leadership and their eventual impeachment. Since then, three changes took 9 place in the leadership of the GDF and of the main implementing agency, SEDUMA, and the PMU coordinator which contributed to further Project implementation delays, particularly the resolution of pending resettlement issues. In March 2010, the Bank was informed that the Borrower had unilaterally changed the Project implementing agency from SEDUMA to ADASA—the Water, Energy and Sanitation Regulatory Agency. The change of implementing agency created legal ambiguity in implementation, especially fiduciary responsibilities, and thus required Project restructuring and an amendment to the loan agreement. In addition, once a new government was established in 2011, all staff in the secretariats dealing with resettlement were changed, leading to a transition period and a learning process when the Project was in a critical period. Suspension of disbursement was necessary due to non-compliance with the safeguards policy and the unconsulted change in implementing agency. The possibility of suspension of disbursements linked to safeguards non compliance was flagged as early as April 2009. After multiple action plans had been prepared by and agreed with the Borrower to resolve implementation issues, and in light of the continuing delays for well over 12 months in completing the resettlement houses in Monjolo and given the unilateral change in Project implementation responsibilities, the Bank was left with no option but to exercise the remedies under the legal agreement, namely through the suspension of disbursements as of May 7, 2010. Suspension would be lifted when the construction and delivery to the affected households of the resettlement houses in Monjolo (in two phases) and the assessment of ADASA’s capacity to implement the Project had been completed and found satisfactory to the Bank. The suspension was partially lifted (on the goods and consultants services) on October 13, 2010 and fully lifted (for works) on September 29, 2011. Overall, the suspension may have had some impact on implementation although overall this is deemed limited. The suspension did not have an impact on the Borrower's ability to meet pending resettlement activities and the sewerage works in Vicente Pires, as these were counterpart funded. The suspension may have had an impact on the Borrower's ability to finalize works in Vila Estrutural, but these works were largely held up by delays in resettling households from the right- of-way of works rather than a lack of funds4. It is important to note that the suspension was partially lifted in the interim to recognize the Borrower's progress towards meeting the conditions (which were partially met) and to enable the Borrower to maintain its ability to supervise and coordinate resettlement activities, since project management support and supervision were funded under the loan. After the suspension was lifted, the disbursement process was changed from advances to reimbursement based on SOEs. This has had an impact on the counterpart’s flow of funds for carrying out project activities. 2.3 Monitoring and Evaluation (M&E) Design, Implementation and Utilization The project’s monitoring and evaluation design and implementation was largely satisfactory. It was envisioned that the project M&E system would require three sets of information: (a) gathering water quality data for the rivers benefiting from the project intervention; (b) evaluating the urban standard achieved in Vila Estrutural and Aguas Lindas in order to estimate the amount of phosphorus abated; and (c) registering the project results achieved per activity. The PMU and CAESB were given the responsibility for implementing the project M&E system. CAESB already 4 During the same period, GDF’s contract signed with the firm responsible for building the houses was questioned by the Public Hearing’s Office, and a percentage of the payments were retained while the process was being evaluated. As a result, the firm passed through financial difficulties and the contract was reduced from the construction of the original 1,360 houses to 586. 10 monitored the DF rivers’ water quality through an efficient monitoring system, and agreed to provide timely information to the PMU with respect to the water quality of rivers that would benefit from the Project. The second item would be achieved by an assessment at the end of the interventions proposed for Vila Estrutural and Aguas Lindas. The existing ‘Water Resource Master Plan’ provides the comparative baseline for this ex-post assessment. The third item would be covered by the regular project implementation monitoring under the PMU’s responsibility. The project design relied on the Social Environmental Indicator (ISA-F) as the tool for measuring the increase in the quality of life in Vila Estrutural. Prior to the Mid-term Review, it was determined that using the ISA-F would be too complicated and data intensive to replicate. The PMU extracted 46 indicators (later reduced to 14 indicators) from the ISA-F that it would use to monitor the quality of life, including information from service providers, installed infrastructure, and citizen surveys. The project design included the contracting of a consultancy firm with expertise in managing internationally financed development projects to assist the PMU in the implementing agency (SEDUMA and ADASA). The consultant firm performed the broad specialized tasks needed for managing the multisectoral project activities, including M&E. 2.4 Safeguard and Fiduciary Compliance While fiduciary compliance has been satisfactory throughout project implementation, compliance with World Bank safeguards policies has been a significant challenge, especially in regard to involuntary resettlement. Financial Management and Procurement have remained Satisfactory or Moderately Satisfactory throughout project implementation. Despite some initial problems, the project’s financial management system efficiently and reliably provided timely information required to manage and monitor the implementation of the project. Appropriate internal controls were in place and functioned effectively. Procurement implementation under the project was adequate overall, although changes to the procurement team contributed to delays and some shortcomings in processing a few packages. As described in earlier sections of this ICR, compliance with OP 4.12 was an ongoing challenge during project implementation. Overall Safeguard Compliance has been rated Unsatisfactory since May 2010 and Moderately Unsatisfactory since May 2009. The two primary safeguard compliance issues were the use of temporary resettlement arrangements not explicitly included and codified in the RAP (described earlier) and the decision to not resettle occupants from area originally deemed as environmentally sensitive by IBAMA (i.e., Santa Luzia and Quadra 12). During Project preparation, it was determined that 903 households and businesses would require resettlement due to their location in environmentally sensitive areas or in the path of needed infrastructure improvements. Specifically, 346 houses were located on top of a former dumpsite, 213 in the buffer zone of the National Park, and 3 in proximity to the Cabeceira do Valo Creek. In addition, 115 commercial workshops required relocation from an area that drains into the National Park. The resettlement of households and businesses from environmentally sensitive areas was required by the license permitting the upgrading of Vila Estrutural (IBAMA, Licença de Instalação n.º 008/2007), and constituted about 81% of the resettlement cases. In regard to resettlement due to infrastructure and resettlement housing construction, it was estimated that 169 lots would be affected. 11 In August 2008, the Bank started receiving complaints from community members regarding resettlement actions taken by the GDF. Specifically, there were concerns about households being temporarily resettled into tents and wooden shacks—both of which were not options included in the Resettlement Action Plan (May 2008). The GDF explained that the main causes of delays in the implementation of the resettlement plan were problems with the procurement process for the construction of the 1,290 houses in Vila Estrutural and 70 houses in Monjolo, which were counterpart funded —the process had been delayed, canceled for a lack of bids, restarted (the awarded bidder faced financial difficulties resulting in further delay). Families affected by infrastructure works were given the option of receiving temporary housing or a rental allowance. In the case of the households who moved to Monjolo, the GDF explained that the farmers requested to move there quickly because they were fearful of losing the right to occupy the area. To fulfill their request, the GDF accelerated the process of relocating the families despite the lack of infrastructure and appropriate housing. In the case of four householders, they were relocated there against their will as the GDF argued their lives were endangered due to risk of flooding. During a mission in October 2008, which included participation of sector and country management, the first of multiple action plans was agreed between the Bank and the Borrower to resolve the resettlement issues. Unfortunately, problems with the construction of the resettlement housing persisted, the families in Monjolo remained in temporary conditions for many months, and their complaints to the GDF and the Bank continued. Eventually, the delay in implementation and non-compliance with the RAP and revised action plan resulted in the suspension of disbursement (see Section 2.2). In addition, in February 2010, a lawsuit was filed before a Brasilia court by a group of 20 resettled families against the Bank and GDF who claimed compensation for material and moral damages resulting from the resettlement plan under the Project. The law suit against the Bank was later dismissed on lack of merit. Managing the list and documentation of the 903 households and businesses was a challenge throughout project implementation. As of May 2012, it was estimated that about 491 of the 903 affected parties had not been resettled—249 have been deemed to be no longer affected, including all of the commercial workshops, rural residents in Cabeceira do Valo, and households in areas where infrastructure was redesigned to avoid demolition of structures. In addition, there are 242 households that were not yet attended to by the project for two primary reasons. First, there are 39 households located in the right-of-way of works that have not yet commenced (i.e., Rua Nove de Julho, elementary school, the Urban Park, and the area where future resettlement housing will be developed). These households will be moved after resettlement housing is completed. The second category consists of 203 households located in Santa Luzia and Quadra 12. Both of these areas were originally included for resettlement as a requirement of the license issued by IBAMA—Santa Luzia is in the buffer zone of the National Park and Quadra 12 is located on top of an old dump site. It was later agreed that these areas would not be resettled in order to reduce the scope of resettlement under the Project. However, a point of concern is the failure to monitor land occupation and address further invasions in these areas. A structure is in place in the GDF to deal with land invasions—AGEFIS is responsible for monitoring land occupation and SUDESA has police power to remove people once invasions have occurred. However, there is a long history of land grabbing in the Federal District, and there doesn’t currently appear to be strong political will to address the issue. In Vila Estrutural, specifically, signs have been posted to warn against additional invasions, but they have not proven very effective. 12 Table 1: Reason for Household and Business Resettlement Reason for Resettlement Households Commercial Total Environmentally Sensitive Areas 619 115 734 Right-of-way 169 0 169 TOTAL 788 115 903 Table 2: Status of Resettlement (as of 5/31/12) Status Households Commercial Total Resettled 408 4 412 Removed from RAP (desaffectados) 138 111 249 Environmentally Sensitive Areas 19 111 130 Right-of-way 119 0 119 Not yet attended 242 0 242 Right-of-way, resettlement in 2012 39 0 39 No resettlement (Santa Luzia, Qd. 12) 203 0 203 TOTAL 788 115 903 In regard to environmental safeguard compliance, a portion of the Environmental Management Plan for the Project, excluding all activities related to solid waste, was implemented. In addition, the Environmental Manual for Works, in general, was satisfactorily complied with, and the Protected Areas Management Plan for Vila Estrutural is currently in progress. Environmental licensing was a challenge during Project implementation. Responsibilities between IBAMA (the Federal Government, with a regional office in the DF) and IBRAM (GDF) were not clearly defined, and this impacted the analysis of applications for environmental permits and their renewals. For example, the issuance of the environmental license for the works in Monjolo was a critical obstacle for the Project. Initially, it was understood that a license was not necessary because it is a rural area. Later, the works were suspended until conditions for receiving a license were met. Moreover, the issuance of environmental permits for the sewerage works in Aguas Lindas and Vicente Pires also took a significant period of time. In Vila Estrutural, the original license issued by IBAMA-DF covered all activities in the area and included more than 40 conditions, touching upon multiple sectors and including the closure of the landfill and activities with the scavengers. The GDF, through TERRACAP, had the obligation to monitor compliance with the conditions. However, even with all the Project interventions in Vila Estrutural, the conditions have not been met. During supervision missions, the Bank requested that the conditions be monitored and recommended that the borrower initiate discussions with IBRAM to issue partial operating licenses for works already completed such as the sewerage system. The operating licenses for all works supported by the Project have yet to be issued. On the other hand, consulting activities focused on important environmental issues, such as the ZEE (Economic–Environmental Zoning), the assessment of CAESB’s environmental liabilities, and the strengthening of IBRAM’s environmental licensing process, among others, have been undertaken or are being completed. This will enable the GDF to plan in a more environmentally sustainable manner and to expedite the licensing process in the future. 2.5 Post-completion Operation/Next Phase The Borrower has committed to complete all project activities not completed by the closing date. By the closing date, several activities under the project (including counterpart-funded ones) have not been completed. The GDF, via a letter from the Governor, has committed to continue the 13 implementation of these remaining activities and to assume the financial responsibility for payment for any activity carried out after the closing date.5 Budget for the PMU and counterpart funding was included for 2012. A letter from the Country Director to the Governor listing the activities that were still ongoing after the closing date, and which were therefore the responsibility of the GDF, was sent on January 4, 2012. These included: Under Execution  Preparation of an upgrading plan for San Sebastian  Economic-environmental zoning of the Federal District  Proposal for restructuring the SLU (now complete)  Strengthening of environmental licensing process of IBRAM (now complete)  Preparation of the design for the new sanitary landfill (now complete)  Management plan for protected areas in Vila Estrutural (now complete)  Road system and drainage in Vila Estrutural (2nd stage) (now complete)  Sewage system in Vicente Pires  External Audit (now complete) Not Under Execution (including in Procurement)  Strategic plan for ADASA (cancelled)  Preparation and implementation of the public agency recycling program (cancelled)  Rezoning of the environmental protection area in Sao Bartolomeu  Survey of Vila Estrutural beneficiaries  Construction of bridge in Monjolo (cancelled)  Repairs to houses in Monjolo (now complete)  Construction of elementary school in Vila Estrutural  Construction of a solid waste triage center  Equipment for the solid waste triage center  Purchase of 4 vehicle for selective collection of solid waste  Training of scavengers in Joquei landfill The Bank will continue to provide implementation support to the Borrower for an additional 12 months after the project closing date. A commitment to extend Bank supervision for an additional 6 months was formalized in the January 4 letter from the Country Director to the Governor. Later, it was agreed with the GDF that the Bank would continue to provide support until December 31, 2012. This action better aligns the Bank’s support with the GDF’s internal closing date for the Project and the operation of the PMU. The primary focus of the support will be the resettlement of households that occupy sites earmarked for infrastructure improvements, and whose resettlement would enable the works to be completed. 3. Assessment of Outcomes 3.1 Relevance of Objectives, Design and Implementation Rating: Substantial 5 Letter dated January 4, 2012. 14 The relevance of the project objectives is assessed as substantial. The project addressed issues that have been and remain a priority for the Government of Brazil and relevant to the country’s strategy. At the time of the project design, these objectives were central to the Country Assistance Strategy (CAS) of FY2004-2007, namely the development of “A More (Environmentally) Sustainable Brazil.� The objectives remain strongly relevant today, with regards to two of the four strategic objectives of the 2012-2015 Country Partnership Strategy (CPS), namely (i) improving the quality and expanding the provision of public services for low income households and (ii) improving sustainable natural resource management. In particular, integrated urban water resources management, improved access to sanitation and scaling up of slum upgrading are especially important development issues in Brazil. Ensuring the sustainable use of water resources in the long run is a top priority of the Brazilian Government. In terms of sanitation, in 2009, only 72.3% of Brazilian households had access to wastewater collection and barely 34.6% of the total produced wastewater was treated. This meant that some 16.2 million households still lacked access to sanitation, of which 60% live in urban areas. In terms of slum upgrading, some 3.2 million households lived in precarious housing within favelas in Brazil in 2000, and the country’s total housing deficit is estimated today at 7 million units. About 80% of households living in precarious conditions and one third of the housing deficit are concentrated in Brazil’s 11 largest metropolitan areas, including the Federal District. The relevance of the project design is assessed as substantial. The project design was based on an integrated approach to urban water resources management, linked to slum upgrading, expanded access to sanitation and solid waste management, and strengthening of policy and implementation capacity. This approach integrated lessons learned from the sector, but underestimated the complexity of the design vis-à-vis limited implementation capacity and weak inter-agency coordination. Implementation was significantly impacted by recurrent resettlement problems due to this lack of inter-agency coordination. 3.2 Achievement of Project Development Objectives The PDO was to reduce regional inequalities and ensure quality water resources to meet the growing needs of the Federal District and the Brasilia Metropolitan Region, through environmental planning and management activities, integrated with poverty reduction interventions and environmental rehabilitation in the most critical river basins. The results of the Project’s intervention would be measured by the following three indicators: (i) improved water quality of the Vicente Pires and Descoberto Rivers; (ii) improved living conditions in targeted settlements by providing integrated urban environmental services; and (iii) strengthened technical and institutional capacity contributing to the sustainable development of the Brasilia metropolitan region. Although the PDO is quite broad, the planned Project activities had clear linkages to these three primary indicators. The project has had a positive impact on the quality of life in Vila Estrutural, but has largely not achieved the environmental goals envisioned in the PDO. The project resulted in an overall improvement of livelihood of inhabitants of Vila Estrutural in terms of access to infrastructure and community services. In addition to the investments carried out under the project, the GDF also established several community facilities including, among others, a community restaurant, a large sports center, and police facilities. However, the improvements to the neighborhood are less than what was anticipated due to the fact that at the time of project closing there were still a number of minor works to be finalized to complete the infrastructure improvements and public facilities in Vila Estrutural (drainage, road paving, and water and sanitation networks, and one school remains to be built), the dumpsite was not closed or remediated (hence the negative externality from proximity to the landfill was not addressed), and the resettlement of households 15 to new houses was not fully completed (some houses remained to be built, including most importantly for 39 families in front of works and family occupancy of completed houses remained incomplete). However, between 2004 and 2011, there was marked improvement in the level of access to basic services like water, sewerage, and solid waste collection, as well as the condition of housing, in Vila Estrutural. For example, in 2004 only 9.2% of households had access to sewerage services compared to 92.2% of households in mid-2012. The expectation is that by the end of 2012, 99% of households will benefit from network connections. Similarly, there was significant improvement in urban infrastructure. In 2004 only 3.5% of households were located in paved streets, and by mid-2012 this percentage rose to 84.6%. The rates of street lighting and stormwater drainage also improved, rising from 93.6% to 95.1% and 4.5% to 85%, respectively. Finally, there were positive improvements to the quality of housing structures in Vila Estrutural. In 2004 about 55% of households were classified as shacks—constructed with improvised or not permanent material. By 2011, this percentage dropped to 14.1%. The percentage of masonry built homes jumped from 48.1% to 85.1%, and the percentage of households with documentation of ownership (land tenure) increased from 0% in 2004 to 79.1% in 2011. In addition, several new facilities were completed, including a preschool, a community center, and a social assistance center. The GDF also completed several facilities in the neighborhood separate from the project, including a community restaurant, police stations, and a sports center. It is important to note that a key driver of the upgrading works in Vila Estrutural was to reduce the quantity of untreated waste and run-off reaching the national park and water resources through the installation of water and sewerage lines and a stormwater drainage system. This has largely been achieved. The other primary investment to abate water resource pollution loads, namely the construction of the sewage system in Vicente Pires, will not be finished until December 2012.6 It is anticipated that improvements to the water quality will not be evident until mid-2013. The primary cause of delay to the sewerage works was a failed land acquisition for the construction of one of two initially planned pumping stations, which led to the redesign of the system to rely on only one pumping station. A secondary cause was delay in environmental license issuance. However, it is highly likely that these works will be completed, and once they are, they will directly benefit about 70,000 low-income residents in Vicente Pires and improve the overall water quality for over 1 million residents in Brasilia. Although it is highly likely that many of the objectives of the Project will be met once all activities are completed, due to their incompletion at the closing date (and time of ICR writing), the achievement of the PDO is rated overall as Unsatisfactory. 3.3 Efficiency 6 When the extension of the closing date was proposed by the GDF in March 2011, the timeline presented by the GDF suggested that all activities could be completed by the revised closing date (this is after the restructuring that adjusted and reduced the scope of the project). Given that the project had DO and IP ratings that were not satisfactory, the extension of the closing date was therefore an exception to Bank policy. The extension was granted for the period requested by the GDF with an explicit prior agreement that no second extension would be granted, and this was agreed with the GDF. 16 Economic Analysis. For the economic analysis, the project's benefits were defined as: (i) improved water quality in the principal sources for the DF, reducing water treatment costs and avoiding contamination and long-term compromise of these sources; (ii) improved water quality of the Paranoa lake for recreation and leisure; and (iii) expanded delivery of potable water, sewerage collection and treatment in the DF and the surrounding areas. The Economic Internal Rate of Return (EIRR) of the Program was estimated at 21.2%, with a net present value (NPV) of R$168.9 million, confirming the viability of the proposed investments. A project horizon of 20 years was assumed. Present Value (R$ million) EIRR Investment Components Costs Net Benefits Investment O&M Benefits Water Source Preservation 214,200 57,995 42,429 113,836 36.1% Paranoa Lake Preservation 32,360 20,316 5,267 6,778 15.9% Aguas Lindas – Water Supply 151,432 92,610 29,241 29,581 16.3% Aguas Lindas – Sanitary Sewerage 31,145 24,758 6,057 330 12.2% Vicente Pires-Sanitary Sewerage 40,397 20,116 3,587 16,695 21.8% Vila Estrutural-Sanitary Sewerage 6,149 4,684 850 615 13.8% Vila Esturural-Microdrainage and Paving 10,306 8,609 634 1,063 13.8% Total 485,989 229,088 88,065 168,898 21.2% Without replicating the detailed model that was prepared as part of the project’s preparation, it is clear that the actual EIRR is much lower than projected due to the fact that “Water Source Preservation� was the primary driver of the net benefits. These economic benefits were primarily anticipated from the removal of sewerage discharges from Vila Estrutural and Vicente Pires (partially completed); drainage run-off from Vila Estrutural (partially completed); export and secondary treatment of sewerage from Aguas Lindas (cancelled); and the deactivation and environmental recuperation of the Joquei Clube sanitary landfill (cancelled). Activities in Aguas Lindas were excluded from the project after restructuring in March 2011; such activities had accounted for about 18% of the net benefits. Financial Analysis. The financial analysis was conducted with the objective of assessing project investments in terms of their viability and to determine the need for subsidies for those components with less than full cost recovery as required for operation and maintenance over the project horizon of 20 years. Revenue generation from investments in water supply and sanitary sewerage were estimated using the actual tariff structure used by CAESB, which will be responsible for the management of these investments. Revenue from draining and pavement investments in Vila Estrutural were estimated based on expected collection of property taxes (IPTU). In the case of the restoration of the Joquei Clube landfill and the construction of the new sanitary landfill, carbon credits were factored into the estimation of revenue. Overall cost recovery was estimated at 51.1% of the total program cost, with a negative NPV of R$180.2 million. Incremental fiscal revenue overall from the project was estimated at R$30.4 million, against a required subsidy of R$75.7 million. Present Value (R$ million) % Cost Investment Costs Net Revenue recovery Investment O&M Revenue Aguas Lindas/Vicente Pires-Water Supply 87.7 94.4 35 -41.7 67.8% Aguas Lindas-Sewerage 25.5 93.4 25.1 -93.0 21.5% Vicente Pires-Sewerage 21.2 30.9 8.3 -18.0 54.1% Vila Estrutural – Sewerage 4 10.7 2.9 -9.6 29.4% 17 Vila Estrutural –drainage/paving 3.4 13.7 5.2 -15.5 18.0% Joquei Clube landfill 38.1 11 0 27.1 346.4% New sanitary landfill 9 7.1 32 -30.1 23.0% Total 188.9 261.2 108.5 -180.8 51.1% Note: The table doesn’t exactly match the table provided in the PAD because the original table includes some arithmetic errors. Keeping the assumptions on demand, tariffs, investments, and O&M fixed, the exclusion of investments not completed under the project reduced the cost recovery to 47.6% of the total program cost, with a negative NPV of R$65.4 million. Present Value (R$ million) % Cost Investment Costs Net Revenue recovery Investment O&M Revenue Vicente Pires-Water Supply 30.9 39.6 13.6 -22.3 58.1% Vicente Pires-Sewerage 21.2 30.9 8.3 -18.0 54.1% Vila Estrutural – Sewerage 4 10.7 2.9 -9.6 29.4% Vila Estrutural –drainage/paving 3.4 13.7 5.2 -15.5 18.0% Total 59.5 94.9 30 -65.4 47.6% In summary, efficiency is rated overall modest. 3.4 Justification of Overall Outcome Rating (combining relevance, achievement of PDOs, and efficiency) Rating: Unsatisfactory With a substantial relevance of objectives and design, modest efficiency, but an overall unsatisfactory implementation and achievement of the project development objectives (at the closing date), the overall outcome rating is considered to be Unsatisfactory. In particular, there were significant shortcomings in the operation’s achievement of its objectives. Specifically, several critical activities were not concluded by the closing date, including: (a) Construction of the sewerage system in Vicente Pires (works underway). This meant that the reduction in the pollution load discharged to the lake and the resulting improvement in water quality did not materialize; (b) The package of activities related to solid waste management and environmental improvement, namely the closure and environmental recuperation of one-third of the Joquei Clube solid waste landfill (not started), the launch of the procurement process for the construction of a new sanitary landfill (aborted after multiple attempts at concessioning the new landfill – is being relaunched under a new PPP model of public constructing and private management), the implementation of a recycling program for government offices (under design); and the construction of a solid waste triage center and training of scavengers at the Joquei Clube landfill (under procurement), has failed to materialize. In particular, the multiple delays faced during the procurement process of the concession for the new landfill of the DF have meant the inability to close the existing landfill. After launching the call for bids, the GDF postponed and suspended the process on more than one occasion. This was a missed opportunity to improve the environmental conditions in Vila Estrutural and improve the overall solid waste management system for the DF. 18 3.5 Overarching Themes, Other Outcomes and Impacts (a) Poverty Impacts, Gender Aspects, and Social Development During the program implementation period, 2004 to 2011, improvements to educational attainment, the formality of employment, and income of residents in Vila Estrutural was observed. For example, the percentage of children not enrolled in school declined from 1% to 0.1% and the percentage of illiterate people decreased from 4.4% to 2.0%. The overall set of investments in Vila Estrutural has improved access to education for the residents, as well as overall access to services (social services, community facilities, etc). There was an increase in the number of schools in the neighborhood—two schools were built by the program and four more schools by the GDF, as well as centers for early children education. In addition, vocational courses are offered at the newly built community center. In 2004, the average income per capita was 0.4 times the minimum wage and family income was 1.9 minimum salaries. In 2011, these indicators rose to 0.6 and 2.3, respectively. The percentage of people with access to formal employment also increased. In 2004 about 34% of the residents of Vila Estrutural worked with documented salaries, in 2011 this percentage rose to 46.7%. Unfortunately, it is not possible to determine whether these improvements are due to project interventions or overall market conditions. (b) Institutional Change/Strengthening N/A (c) Other Unintended Outcomes and Impacts (positive or negative) Awareness of the importance of putting in place a regulatory framework for involuntary resettlement in the DF has increased significantly at the technical level among staff of the housing, social development, agriculture and environment agencies and secretariats, which is a necessary step (although alone insufficient) towards the eventual introduction of a regulatory framework governing involuntary resettlement. 3.6 Summary of Findings of Beneficiary Survey and/or Stakeholder Workshops The indicator used to measure the improvement in living conditions of residents in Vila Estrutural is the Social Environmental Indicator for slums (ISA-F). This indicator was developed for measuring social and environmental condition in unregulated low income settlements denominated slums. The ISA-F derives from a broad ISA methodology used to monitor urban environmental improvements in a municipal territory. It consists of the following sub-indicators:  Percent coverage of water supply;  Percent coverage of sewerage;  Percent coverage of garbage collection;  Percent coverage of stormwater drainage;  Percent coverage of street paving;  Geological and geotechnical stability;  Percent coverage of residential electricity;  Percent coverage of street lighting;  Percent coverage of regularization; 19  Population density;  Access to community and recreation equipment;  Access to education (improving educational levels); and  Access to income (the improvement of income levels). To measure these indicators, the team used information available in the Sample Household Survey (Pesquisa Distrital por Amostra de Domicílios) conducted by the Society of Planning of the Federal District (Codeplan) in 2004, i.e. before the Program, and in 2011 after completion of the main interventions. This was supplemented by data collected in the satisfaction survey of Program conducted in May 2012. Between 2004 and 2011, due to Program interventions, a marked improvement in the ISA-F indicators was observed. There was a significant increase in access to basic services like water, sewage and garbage collection. Additionally, as described in section 3.5 (a) above, improvements to educational attainment, the formality of employment, and income of residents in Vila Estrutural was observed. The following table presents a summary of the status of the ISA-F sub-indicators and therefore the degree of improved quality of life achieved through the implementation of the Program. It is observed that the average result is greater than 85%, which is considered very good. 20 Comparision of ISA-F Indicator for Vila Estrutural, 2008 to 2011 Physical Goal Nº Sub-Indicator 2008 2011 Forecast Actual Forecast Actual 1 Water supply 50% 93% 100% 98% 2 Sewerage 35% 33% 100% 92% 3 Garbage collection 50% 80% 100% 98% 4 Stormwater drainage 100% 10% 100% 85% 5 Street paving 50% 38% 100% 85% 6 Geological and geotechnical stability 100% 7.90% 100% 90% 7 Population density 0% 0% 100% 70% 8 Residential electricity 100% 79% 100% 96% 9 Regularization 35% 0% 100% 79% 10 Urban cleaning service 50% 80% 100% 98% 11 Public lighting 50% 80% 100% 97% 12 Community and recreation facilities 50% 47% 100% 80% 13 Income 0% 0% 100% 70% 14 Access to education 0% 0% 100% 85% Total result 25% 39% 100% 88% 4. Assessment of Risk to Development Outcome Rating: Significant Site visits have shown that the GDF is not carrying out land use control in Vila Estrutural, which is resulting in an increased number of invasions specifically in the Santa Luzia area and near the Jocquei Clube landfill. There are many new low-income families that have arrived to settle in the area, especially in the period of political uncertainty and coordination challenges. Without concerted effort to limit further invasions, it is likely that access to urban services, operation and maintenance of infrastructure, and overall quality of life will be negatively impacted. In addition, the invasion of environmentally protected lands (and re-occupation of areas that were vacated by households relocated to the resettlement housing) will continue to put water source quality at risk. Similarly, the solid waste management activities continue to face risks towards its sustainability. The studies indicated the feasibility of concessioning the new landfill of the DF to the private sector, which was intended to guarantee financial sustainability of quality service provision. Yet, the obstacles faced in the procurement process—arising from an underestimation of the political economy risks associated with such a concession—necessitated a change from the PPP model to one wherein the landfill is built with public funds and its operation and maintenance tendered for private management. Such a change adds further fiscal burden on the DF. 5. Assessment of Bank and Borrower Performance 5.1 Bank Performance (a) Bank Performance in Ensuring Quality at Entry Rating: Moderately Unsatisfactory The Bank adequately integrated lessons learned in the country and sector in the design of the project. Specifically, the holistic approach to addressing environmental protection in a socially 21 inclusive manner was and still is a best practice. However, the design did not consider the political economy of the GDF, and no institutional analysis of capacity was done. The RAP was prepared and proved comprehensive except it did not largely address the issue of temporary resettlement (in terms of putting in place safeguards and eligibility criteria and confirming the responsible Secretariat’s ability to implement such instruments), as per the concerns raised by the Safeguards Advisory Team. (b) Quality of Supervision Rating: Moderately Satisfactory Early in project implementation, a gap developed between what was stated in the RAP and what was being implemented on the ground, especially with regard to temporary resettlement. Although the RAP indicated that works would commence only after the resettlement of affected families, the first major works contract (the perimeter road) was initiated prior to the procurement process for the resettlement houses, putting 34 households in temporary accommodation not contemplated and appropriately governed by the RAP. In terms of task team composition, during early supervision the team did not have a dedicated social safeguards specialist. This role was being primarily filled by the Team Leader (whom – as a social specialist – had the necessary qualifications, but faced an increasing challenge with the increase of resettlement-related challenges in project implementation). However, once complaints began to surface, the task team took a proactive approach to supervision, especially in regard to resettlement. In fact, all of the core team members became fully versed in involuntary resettlement policy with implementation progress. A highly-experienced social development and safeguards consultant joined the team, and senior management from social development and safeguards legal teams were enlisted for support. The Bank team met with complainants several times and referred them to the GDF. Overall, the sudden onset of resettlement-related issues was promptly met with commensurate attention from the task team and Bank management. Starting with the first wave of complaints, the Bank team worked with the GDF to prepare time- bound actions plans to resolve the issues. Community engagement and communications were strengthened, Bank supervision was intensified with increased budget, and the project was put on management’s watch list. A clear message was conveyed to and formally agreed with the GDF that no temporary resettlement of affected families would be carried out under the Project. As a result of these measures, GDF started addressing the issues raised in the complaints. After witnessing multiple delays on the part of GDF in resolving some of the items in the agreed action plans and with the occurrence of some deviations from the provisions of the loan agreement related to implementation and safeguards, the Bank exercised its remedies under the loan agreement through, initially, suspension of loan disbursements. Subsequently, as progress towards the implementation of the conditions to lift the suspension was being observed, the Bank provided a partial lifting of the suspension related to the goods and consultancy categories to enable the GDF to maintain its efforts to supervise the implementation of resettlement actions. Despite implementation difficulties, the team continued to provide implementation support to the GDF to enable the achievement of the project objectives and complete the resettlement and safeguards obligations, and such support remains in place for six months after the closing date of the loan. Since the emergence of implementation challenges, the Bank significantly increased the frequency of the supervision missions to about 4-6 such missions per year, with additional in- between meetings with the implementing unit to maintain constant engagement and ensure monthly follow-up. Senior management and leading Bank experts in social development and resettlement safeguards (including the Chief Counsel for safeguards, and the Sector Manager for 22 Social Development) participated in key missions. The quality of supervision is rated as Moderately Satisfactory. (c) Justification of Rating for Overall Bank Performance Rating: Moderately Unsatisfactory Despite the intensive supervision efforts demonstrated by the task team and attention from senior management in both the Country and Sector Management Units (CMU and SMU) after complaints from community members started to surface and delays in critical works occurred, the lack of focus on institutional issues in the design and in the early implementation period together with the gap in the RAP concerning temporary resettlement created the foundation for the implementation issues and challenges that affected the project until its closing date. 5.2 Borrower Performance (a) Government Performance Rating: Unsatisfactory During project preparation, the government’s commitment to the project was strong. However, initial implementation was slow due to local elections and then a restructuring of the government, resulting in a change of implementing agency. The commitment of the new government proved less evident than its predecessor in particular in ensuring. inter agency coordination. Overall, the GDF was unable to provide sufficient executive coordination for the implementation of the program. The engagement and participation of multiple secretariats was needed for the successful implementation of the project. There was institutional fragmentation and a lack of coordination in the implementation of resettlement and infrastructure works. The four involved agencies, CODHAB, SEDEST, SEDUMA and SEAPA, did not coordinate their actions nor adequately deliver on their individual mandates—CODHAB faced major delays in habilitating (screening, vetting and obtaining administrative authorization for occupation) the families; SEDEST faced major delays in payment of the social rental payment for temporarily resettled families and needed to retool its regulatory instruments accordingly; SEAPA faced major delays in preparing the beneficiary list and the delineation of the lots; and SEDUMA’s coordination role was deficient (an issue that was not helped by the lack of high level support from the GDF) and the issue related to the environmental license for resettlement housing in Monjolo further complicated implementation. Houses were built, but families were not yet authorized to occupy them. Works were progressing and families resettled, but to temporary solutions. After complaints started to surface, the Governor did respond personally and directed the authorities to address the problem expeditiously, resulting in some progress. However, the GDF failed to comply with the agreed time-bound action plans, resulting in a suspension of disbursement. This situation was exacerbated by the political scandal that took place in November 2009 and led to the Governor’s impeachment, and the repeated changes in leadership between then and until the election of a new Governor in January 2011. In addition, the level of execution of counterpart funding seriously lagged the Bank loan. Despite the project being 50-50 between loan and counterpart funds, the execution of counterpart funded activities was three times slower, and activities such as the Vicente Pires sewerage system have faced serious delays. 23 (b) Implementing Agency or Agencies Performance Rating: Moderately Unsatisfactory The Project Management Unit, under the leadership of SEDUMA and ADASA, made great attempts to keep the project activities moving forward despite institutional coordination problems and the suspension of disbursements, which made for a particularly difficult situation for ADASA in terms of implementation environment. However, there was a failure to comply with covenants and agreements, especially related to safeguards policies, which persisted throughout project implementation and especially in the period up to the suspension of the disbursements. A particular challenge was the poor management and oversight of the list of households requiring resettlement, especially due to insufficient support from the responsible DF entities – CODHAB in Vila Estrutural and SEAPA in Monjolo. Data provided in status reports were also on various occasions inconsistent with previous reports or incomplete. (c) Justification of Rating for Overall Borrower Performance Rating: Unsatisfactory Overall, the GDF’s performance in ensuring quality of preparation and implementation, and complying with covenants and agreements, toward the achievement of development outcomes was Unsatisfactory, mainly on account of the involuntary resettlement issues and the implementation delays of the solid waste management program. 6. Lessons Learned As part of project preparation, more emphasis should be put on assessing institutional capacity and designing arrangements and mitigation measures to carry out resettlement. For most of the agencies involved in the implementation of the Project, this was their first experience with urban upgrading activities, including resettlement. While the institutional challenges regarding environmental issues and metropolitan planning were addressed during Project preparation, there was not a sufficient assessment of the capacity of multiple agencies involved in resettlement to carry out the RAP. Urban upgrading projects are best managed by one agency with delegated authority for all activities within the target area. In a multi-agency implementation arrangement, the primary implementing agency needs authority (directly or in-directly through a high-ranking public official) over others. In this Project, the impact of the GDF’s institutional complexity on the coordination of works was underestimated. Urban upgrading projects require inputs from multiple municipal agencies (i.e., water, sanitation, transport, works, and social development), and therefore there is a need for strong central leadership or authority. Formal agreements between the agencies may be necessary. A better option is the delegation of responsibility for design, works, and social issues to one agency. The presence of an effective grievance redress mechanism is critical to improve development impact by minimizing potential impact on beneficiaries especially in cases of resettlement. The lack of a channel to present full and timely responses to complaints led affected families to contact the Bank directly in hopes of forcing the GDF to take action. 24 The construction of resettlement housing should be prioritized in the chronogram of major works and under the control of the implementing agency. The construction of the resettlement housing in the Project was uncoordinated with the upgrading works dependent on a source of funding that was not under the control of the PMU. A key lesson learned is that the construction of resettlement housing should be prioritized in the chronogram of major works and under the control of the implementing agency. Moreover, the separation of the financing sources between works and the related resettlement created a situation of disconnect between works under construction and resettlement houses that were delayed. The construction of resettlement housing should be strongly linked to the upgrading works contracts. Resettlement options should enable choice differentiated resettlement solutions, and open and frequent consultation with affected families is essential. Resettlement issues were the greatest challenge for Project implementation. A lesson learned is that resettlement options should enable choice and provide differentiated solutions. In this case, the resettlement options were either (a) a new house for resettled households designated as urban (in situ, as per best practice) or (b) a lot with a house in the Monjolo rural area for families that were designated as rural by SEAPA (a 2- ha or 4-ha lot with house). These options did not address the needs of those with commercial lots or those with houses larger than the resettlement units. Temporary resettlement could be a viable option, but should be limited to temporary rental payments (aluguel social), rather than temporary structures, and needs to be clearly defined in the Resettlement Action Plan. It is essential that communication with affected families is open and frequent. Once the project appointed a social expert to the PMU, established an on-site resettlement office, conducted frequent meetings with families, and distributed leaflets, the number of complaints diminished greatly. Resettlement activities must be closely linked to the control of new irregular settlements. The GDF has not been able to manage additional invasions in the target area, compromising all the resettlement work that has been accomplished. More importantly, the full impacts of the project interventions are not likely to be perceived, even for the resident population. The Bank’s safeguards policy is not always easy to implement in middle income countries that have their own policies which do not necessarily coincide with the Bank's. At the time of project design, it was believed that the law declaring Vila Estrutural a Residential Zone of Social Interest (Zona Habitacional de Interesse Social—ZHIS) provided a sufficient legal underpinning for the urban upgrading and resettlement activities. However, this proved to be insufficient because it only allowed for compensation with a physical housing unit. In addition, the GDF’s housing policy excluded families that had received a social housing unit in the past, even if they were being dispossessed/resettled under the Project–the case for about 1/3 of affected families in Vila Estrutural. Although the RAP stipulated that all affected families should receive compensation, the judicial arm of the GDF was firm about applying its policy to not allow double housing benefits. Although an integrated approach to environmental sanitation is considered good practice, it may be too complex for some clients. To address water quality in an integrated manner, the Project included activities in a broad range of areas, including water and sewerage, solid waste, programs for wastepickers, a new recycling program, urban upgrading, social support, urban planning, and institutional capacity building. The design also required the participation of many agencies, which may be too complex in some settings. The assessment of political economy issues should be a key part of the preparation in complex/integrated development projects. The component of the Project dealing with solid waste faced delays and opposition. Solid waste management involves multiple stakeholders, 25 including scavengers and political agencies, making it important to discuss options for dealing with waste and landfills openly and through various media. In addition, when a decision is made, multiple stakeholders must be in agreement—political buy-in is essential. 7. Comments on Issues Raised by Borrower/Implementing Agencies/Partners (a) Borrower/implementing agencies Please see Annex 7: Summary of Borrower's ICR and/or Comments on Draft ICR. (b) Cofinanciers (c) Other partners and stakeholders (e.g. NGOs/private sector/civil society) 26 Annex 1. Project Costs and Financing (a) Project Cost by Component (in USD Million equivalent) Actual/Latest Appraisal Estimate Percentage of Components Estimate (USD (USD millions) Appraisal millions) Component 1 7.6 7.3 96% Component 2 31.2 70.3 225% Component 3 71.5 33.2 46% Component 4 3.5 5.1 146% Total Baseline Cost 114.50 115.9 101% Physical Contingencies 0.00 0.00 Price Contingencies 0.50 1.20 Total Project Costs 0.50 1.20 240% Front-end fee PPF 0.00 0.00 Front-end fee IBRD 0.30 0.10 Total Financing Required 115.30 117.20 102% (b) Financing Appraisal Actual/Latest Type of Estimate Estimate Percentage of Source of Funds Cofinancing (USD (USD Appraisal millions) millions) Borrower 57.64 29.61 51% International Bank for Reconstruction 57.64 44.74 78% and Development 27 Annex 2. Outputs by Component Component/Activity Status Borrower IBRD Financing Financing (US$) (US$) Component 1 1.1.2.1 GIS system for urban projects Complete - 363,318.55 1.1.2.2 Preparation of an upgrading plan for San Sebastian In progress - 371,038.25 1.2.1.2 Preparation and implementation of the public Cancelled agency recycling program - - 1.2.2.1 Strengthening of environmental licensing process of Complete IBRAM - 239,344.26 1.2.2.2 Remote sensors for environmental licensing by Complete IBRAM - 49,331.15 1.2.2.3 Software for the processing of environmental Complete licenses by IBRAM - 272,867.37 1.2.2.4 Equipment for the processing of environmental Complete licenses by IBRAM - 35,332.24 1.2.4.1 Rezoning of the environmental protection area in In progress Sao Bartolomeu - - 1.2.4.2 Economic-environmental zoning of the Federal In progress District - 921,186.89 1.2.4.3 Master plan for the preservation of the Botanical Complete Garden - 313,358.09 1.3.1.1 Proposal for restructuring the SLU Complete 33,084.70 317,833.33 1.3.1.2 Strategic plan for ADASA Cancelled - - 1.3.1.3 Legal framework for resettlement in the GDF Excluded - - 1.3.1.4 Study of service routes for recycling program in Complete public agencies 23,039.74 - 1.3.2.1 Technical assistance to CAESB for the protection of Complete wetlands - 167,595.63 1.3.2.2 Survey and treatment of the liabilities of sewage Complete systems - 114,411.40 1.3.2.3 Preparation of licensing studies for CAESB Complete - 252,818.85 Component 2 2.1.a.1 Management plan for protected areas in Vila Complete Estrutural - 160,413.11 2.1.a.2 Construction of CRAS in Vila Estrutural Complete - 969,075.61 2.1.a.2 Construction of community center in Vila Estrutural Complete 48,566.15 539,164.12 2.1.a.3 44 resettlement houses in Monjolo Complete 837,627.67 - 2.1.a.3 Construction of elementary school in Vila Estrutural Not bid 2.1.a.3 460 housing units in Vila Estrutural Partially completed 3,045,205.14 - 2.1.a.3 70 housing units in Vila Estrutural Partially completed - - 2.1.a.3 270 housing units in Vila Estrutural Partially completed 1,211,736.53 - 2.1.a.3 560 housing units in Vila Estrutural Partially completed 1,636,478.39 - 2.1.a.4 Construction of preschool in Vila Estrutural Complete - - 2.1.b.1.1 Perimeter road system in Vila Estrutural Complete 727,118.75 5,544,669.88 2.1.b.1.2 Road system and drainage in Vila Estrutural Complete - 7,011,352.37 2.1.b.1.3 Road system and drainage in Vila Estrutural Complete 5,789,110.66 11,451,544.83 2.1.b.2.1 Sewage system in Vila Estrutural Complete 15,016.90 4,688,046.92 2.1.b.2.2 Pumping stations in Vila Estrutural Complete 218,124.92 2,579,673.80 2.1.b.3.1 Water system in Vila Estrutural Complete 687,722.50 - 2.2.1 Training of scavengers in Joquei landfill Not bid - - 28 2.2.2.1 Construction of solid waste triage center Bidding - - 2.2.2.2 Equipment for the solid waste triage center Not bid - - 2.2.2.3 Repairs to houses in Monjolo Complete 30,696.82 - Seminar on resettlement Excluded - - - Purchase of 4 vehicle for selective collection of Not bid solid waste - - - Construction of bridge in Monjolo Not bid - - - Community center in Monjolo Excluded - - Component 3 3.2.1 Sewage collection system in Vicente Pires - West In progress 4,579,134.51 - 3.2.2 Sewage collection system in Vicente Pires - East In progress 4,144,972.72 - 3.2.3 Sewage collection system in Vicente Pires - Central In progress South 2,964,366.01 - 3.2.4 Sewage collection system in Vicente Pires - Central In progress North 7,555,202.02 - 3.2.4.1 Sewage collection system in Vicente Pires - Central In progress North 634,779.00 - 3.2.5 Pumping station in Vicente Pires In progress 1,323,102.79 - 3.3.1 Closure and remediation of 1/3 of the Joquei Cancelled Landfill - - 3.3.2 Preparation of the design for the new sanitary In progress landfill - 64,961.75 3.3.3 Drilling and surveying services Complete - 66,449.21 - Seminar on the concession of the new sanitary Excluded landfill - - 3.4.1 Removal of macrophytes from Paranoa Lake In progress - 1,336,447.21 Component 4 4.1.1 Management and M&E services In progress 528,491.80 6,728,537.89 4.1.2 Independent audit Complete - 46,377.05 4.1.2.1 Independent audit Complete - 18,204.67 4.1.2.2 Independent audit Complete - 18,800.55 4.1.3 Survey of Vila Estrutural beneficiaries In progress - - TOTAL 36,033,577.72 44,642,154.97 * Exchange Rate: US$1.00 = R$1.83 29 Annex 3. Economic and Financial Analysis (including assumptions in the analysis) Economic benefits of the project The principal benefits of the Brasilia Sustainable Program were intended to be the improvement of environmental quality in the Federal District and the surrounding area (entorno), and the improvement of health and quality of life for the urban populations of the municipality of Aguas Lindas, in Goias, and the populations of Vicente Pires and Vila Estrutural. The Program would specifically reach about 175,000 low-income residents of Aguas Lindas and Vila Estrutural. Particularly, the program would also contribute to the preservation of two important sources of potable water and recreation in the DF: (i) the lake formed by the dam on the Rio Descoberto, which supplies water to nearly 65% of the DF (about 1.34 million people); and (ii) the Paranoa lake, which constitutes a reserve source of potable water and recreation center. Specifically, the Project’s benefits envisioned as follows: (i) improved water quality in the principal sources for the DF, reducing water treatment costs and avoiding contamination and long-term compromise of these sources; (ii) improved water quality of the Paranoa lake for recreation and leisure; and (iii) expanded delivery of potable water and sewerage collection and treatment in the DF and the surrounding area. Estimation of benefits Preservation of Potable Water Sources: The preservation of Descoberto lake, the primary water source for the DF, and of the Corumba system, the principal source of future water supply, would generate direct benefits for approximately 1.4 million individuals, who would be integrated as users into the Brasilia water supply system. These benefits would result from the actions financed under the project associated with the export and secondary treatment of sewerage from Aguas Lindas and the transport of the treated effluent to the Verde river basin, the construction of a new sanitary sanitary landfill for solid waste in the same river basin, and the territorial management of the basin itself, in terms of containment of discharges, such as urban and rural run-off. Benefits were calculated applying the Continent Valuation method and the transfer of benefits through the adoption of a functional form estimated as a model for natural resources management in the Alto Iguacu Basin under the Water Quality Program (PQA) in the Curitiba Metropolitan region. The research, conducted in 1996, sought to determine the Willingness to Pay (WTP) for improved environmental quality in the Alto Iguacu region, the principal water source for the Curitiba Metropolitan region. Given that the PQA sought to (i) install sanitary sewerage and (ii) control erosion in order to recoup the water source and preserve water quality - both expected outcomes of the proposed project - the transfer of benefits based on this previous research was justified. Based on the research, a WTP for the DF was estimated at R$6.57 per family per month. Preservation of Paranoa Lake: Some 255,000 dwellers in Brasilia - particularly Lago Norte and Lago Sul—will directly benefit from the preservation of water quality in Paranoa lake in leisure and recreational activities. These economic benefits flow from the removal of sewerage discharges from Vila Estrutural and Vicente Pires, by transferring these effluent discharges to treatment stations (ETE Norte and ETE Melchior, respectively); drainage run-off from Vila Estrutural; and the deactivation and environmental recuperation of the Joquei Clube landfill. In estimating these economic benefits, as in the previous case, benefits transfer was performed. Under the Basic Sanitation Program for the DF, financed by the IDB in 1997, a WTP equivalent to R$3.57 per family per month was calculated for the use of Paranoa lake in terms of leisure and 30 recreation. This estimation applied a Discrete Choice model which incorporated primary and secondary treatment options. Despite the fact that in both Vicente Pires and Vila Estrutural, (i) tertiary treatment of sewerage would be undertaken and (ii) the absence in the model of diffuse discharge removal and the closing of the landfill, a conservative decision was made to use the results from the secondary treatment in the 1997 model. Updating the functional form based on the mean income for Brasilia, the WTP was estimated at R$4.49 per family per month. Sanitary Drainage in Aguas Lindas and Vila Estrutural: The benefits associated with investments under the project in sanitary drainage were estimated using Contingent Valuation, based on household surveys in Aguas Lindas to determine the maximum WTP. Focus groups were conducted to pre-test and calibrate the household questionnaire, and a subsequent pilot field study in which 30 households were surveyed. The actual household survey sampled 1,027 households, yielding 961 observations, and from which 858 heads of household indicated a willing to pay for a domestic sewerage connection and associated service. The mean WTP observed from the field research was R$12.06 per family per month. Sanitary Sewerage in Vicente Pires: Based on field research conducted in other areas of the DF with similar physical and socioeconomic characteristics, the benefits from sanitary sewerage provision in Vicente Pires were estimated following a transfer of benefits methodology. Specifically, total economic benefits were estimated through adoption of a WTP estimated in 1996 for Recanto das Emas and Vale do Amanhecer during the preparation of the Basic Sanitation Program for the DF. The resulting revised WTP totaled R$36.06 per family per month. Micro drainage and Street Paving: Under the Basic Sanitation Program for the DF, 502 households were surveyed in three localities which, at the time of the research in 1996, shared economic and demographic characteristics with the proposed project area. The WTP estimation updated the monetary value of the constant, drawing on the variation in mean income for the D.F. population between 1996 and 2004, and applying this revised household income to the project area yielded a WTP of R$25.05 per family per month. Economic Costs The direct economic costs of the project include initial investment costs, operation and maintenance, supervision, environmental costs, and acquisition of land and equipment. Also included are costs associated with individual household hook-ups for sanitary sewerage and water supply. A project horizon of 20 years was assumed. Conversion factors for transforming financial costs into economic costs were extracted from “Metodologia para Avaliacao Economica e Financeira de Projetos: a experiencia do PMSS�, as follows: Unskilled labor 0.46 Skilled labor 0.81 Construction materials 0.88 Equipment 0.80 Chemical products 0.83 Electricity 0.97 31 Results of Economic Evaluation Overall, the Economic Internal Rate of Return (EIRR) of the Program was estimated at 21.296, and a net present value (NPV) of R$168.9 million, confirming the viability of the proposed investments. The EIRR and NPV results for the seven investment subprojects assessed are presented below. Present Value (R$ million) EIRR Investment Components Costs Net Benefits Investment O&M Benefits Water Source Preservation 214,200 57,995 42,429 113,836 36.1% Paranoa Lake Preservation 32,360 20,316 5,267 6,778 15.9% Aguas Lindas – Water Supply 151,432 92,610 29,241 29,581 16.3% Aguas Lindas – Sanitary Sewerage 31,145 24,758 6,057 330 12.2% Vicente Pires-Sanitary Sewerage 40,397 20,116 3,587 16,695 21.8% Vila Estrutural-Sanitary Sewerage 6,149 4,684 850 615 13.8% Vila Esturural-Microdrainage and Paving 10,306 8,609 634 1,063 13.8% Total 485,989 229,088 88,065 168,898 21.2% Without replicating the detailed model that was prepared as part of the project’s preparation, it is clear that the actual EIRR is much lower than projected due to the fact that “Water Source Preservation� was the primary driver of the net benefits—representing 67% of the net benefits. These economic benefits were primarily anticipated from the removal of sewerage discharges from Vila Estrutural and Vicente Pires (partially completed); drainage run-off from Vila Estrutural (partially completed); export and secondary treatment of sewerage from Aguas Lindas (cancelled); and the deactivation and environmental recuperation of the Joquei Clube sanitary landfill (cancelled). Activities in Aguas Lindas were excluded from the project after restructuring in March 2011. Financial Analysis The financial analysis was conducted with the objective of assessing project investments in terms of their viability and to determine the need for subsidies for those components with less than full cost recovery as required for operation and maintenance over the project horizon of 20 years. Revenue generation from investments in water supply and sanitary sewerage were estimated using the actual tariff structure used by CAESB, which will be responsible for the management of these investments. Revenue from draining and pavement investments in Vila Estrutural were estimated based on expected collection of property taxes (IPTU). In the case of the restoration of the Joquei Clube landfill and the construction of the new sanitary landfill, carbon credits were factored into the estimation of revenue. Overall cost recovery was estimated at 63.4% of the total program cost, with a negative NPV of R$182.8 million. Incremental fiscal revenue overall from the project was estimated at R$30.4 million, against a required subsidy of R$75.7 million. 32 Present Value (R$ million) % Cost Investment Costs Net Revenue recovery Investment O&M Revenue Aguas Lindas/Vicente Pires-Water Supply 87.7 94.4 35 -41.7 67.8% Aguas Lindas-Sewerage 25.5 93.4 25.1 -93.0 21.5% Vicente Pires-Sewerage 21.2 30.9 8.3 -18.0 54.1% Vila Estrutural – Sewerage 4 10.7 2.9 -9.6 29.4% Vila Estrutural –drainage/paving 3.4 13.7 5.2 -15.5 18.0% Joquei Clube landfill 38.1 11 0 27.1 346.4% New sanitary landfill 9 7.1 32 -30.1 23.0% Total 188.9 261.2 108.5 -180.8 51.1% Note: The table doesn’t exactly match the table provided in the PAD because the original table includes some arithmetic errors. Keeping the assumptions on demand, tariffs, investments, and O&M fixed, the exclusion of investments not completed under the project reduced the cost recovery to 47.6% of the total program cost, with a negative NPV of R$65.4 million. Present Value (R$ million) % Cost Investment Costs Net Revenue recovery Investment O&M Revenue Vicente Pires-Water Supply 30.9 39.6 13.6 -22.3 58.1% Vicente Pires-Sewerage 21.2 30.9 8.3 -18.0 54.1% Vila Estrutural – Sewerage 4 10.7 2.9 -9.6 29.4% Vila Estrutural –drainage/paving 3.4 13.7 5.2 -15.5 18.0% Total 59.5 94.9 30 -65.4 47.6% 33 Annex 4. Bank Lending and Implementation Support/Supervision Processes (a) Task Team members Responsibility/ Names Title Unit Specialty Lending Stefania B. Abakerli B Social Development Specialist SASDS Social Roberto Gabriel Aiello Senior Energy Specialist LCSEG Carbon Finance Musa S. C. Asad Sr Institutional Dev. Specialist SASDA Fernando Andres Blanco Cossio Senior Economist AFTP4 Econ. Analysis Edward William Bresnyan Senior Rural Development Specialist LCSAR Econ.Analysis Jose Augusto Carvalho Consultant LCSPT Legal Flavio Chaves Natural Resources Mgmt. Specialist AFTEN GEF Dean A. Cira Lead Urban Specialist EASVS Urban Martin P. Gambrill Senior Water Engineer SASDU Water Ann Jeannette Glauber Sr Natural Resources Mgmt. Specialist AFTEN Safeguards Andrea C. Guedes Senior Operations Officer ECSH2 Waste Pickers Werner L. Kornexl Senior Climate Change Specialist EASIS Solid Waste Claudio Mittelstaedt Consultant LCSFM Financial Mgmt Alexandre Borges de Oliveira Senior Procurement Specialist LCSPT Procurement Lilian Pena Pereira Weiss Water & Sanitation Specialist LCSUW Operational Support Horacio Cristian Terraza Sr Environmental Specialist LCSEN Solid Waste Supervision/ICR Marcos T. Abicalil Sr Water & Sanitation Specialist LCSUW Water Susana Amaral Financial Management Specialist LCSFM Financial Mgmt Guang Zhe Chen Sector Manager LCSUW Oversight Frederico Rabello T. Costa Procurement Specialist LCSPT Procurement Regis Thomas Cunningham Sr Financial Management Specialist LCSFM Financial Mgmt Marcus Vinicius Ferreira Da Consultant LCSUW Infrastructure Silva Clarisse Torrens Borges Dall E T Consultant LCSEN Environment Acqua Eduardo Franca De Souza Consultant LCSFM Financial Mgmt Juliana Menezes Garrido Pereira Infrastructure Specialist LCSUW General supervision Maninder S. Gill Sector Manager LCSSO Safeguards Catherine Lynch Operations Officer OPCIL Gen supervision/ICR Soraya Simao Melgaco Consultant LCSSO Social safeguards Alexandre Borges de Oliveira Senior Procurement Specialist LCSPT Procurement Suspension/ Reynaldo Pastor Chief Counsel LEGAL restructuring Lilian Pena Pereira Weiss Water & Sanitation Specialist LCSUW Operational Support Paula Dias Pini Senior Urban Development Specialist AFTUW TTL to 06/2009 Suspension/ Catarina Portelo Senior Counsel LEGLA restructuring Sameh Naguib Wahba Sector Leader LCSSD TTL since 07/2009 Monica Rivera Senior Counsel LEGIA Law suit Luis R. Prada Villalobos Senior Procurement Specialist MNAPR Procurement 34 Paul Procee Senior Infrastructure Specialist EASCS Solid waste Jennifer J. Sara Sector Manager EASVS Oversight Marcus J. Wishart Sr Water Resources Specialist AFTWR Operational Support (b) Staff Time and Cost Staff Time and Cost (Bank Budget Only) Stage of Project Cycle USD Thousands (including No. of staff weeks travel and consultant costs) Lending FY05 198.04 FY06 13.56 Total: 211.60 Supervision/ICR FY06 62.44 FY07 77.54 FY08 15.16 61.53 FY09 33.49 125.42 FY10 31.70 152.72 FY11* 14.27 47.48 FY12 * 17.13 71.08 Total: 598.21 * The Staff Time and Costs for FY11 and FY12 do not include procurement, financial management or safeguards supervision time as these were funded off-the-top, nor do they reflect the time spent by Management on Project oversight nor for legal services as these were covered by the CMU. In addition, the TTL relocated to the Brazil Country Office in Brasilia, which significantly reduced travel expenses. 35 Annex 5. Beneficiary Survey Results Please note: a beneficiary survey was conducted in May 2012. The final results of the survey were not yet available at the time the ICR was completed. The survey measured improvement in living condition based on access to urban services as well as beneficiary satisfaction. The following is based on preliminary results. The indicator used to measure the improvement in living conditions of residents in Vila Estrutural is the Social Environmental Indicator for slums (ISA-F). This indicator was developed for measuring social and environmental condition in unregulated low income settlements denominated slums. The ISA-F derives from a broad ISA methodology used to monitor urban environmental improvements in a municipal territory. It consists of the following sub-indicators:  Percent coverage of water supply;  Percent coverage of sewerage;  Percent coverage of garbage collection;  Percent coverage of stormwater drainage;  Percent coverage of street paving;  Geological and geotechnical stability;  Percent coverage of residential electricity;  Percent coverage of street lighting;  Percent coverage of regularization;  Population density;  Access to community and recreation equipment;  Access to education (improving educational levels); and  Access to income (the improvement of income levels). To measure these indicators, the team used information available in the Sample Household Survey (Pesquisa Distrital por Amostra de Domicílios) conducted by the Society of Planning of the Federal District (Codeplan) in 2004, i.e. before the Program, and in 2011 after completion of the main interventions. This was supplemented by data collected in the satisfaction survey of Program conducted in May 2012. The following table presents a summary of the status of the ISA-F sub-indicators and therefore the degree of improved quality of life achieved through the implementation of the Program. It is observed that the average result is greater than 85%, which is considered very good. Comparision of ISA-F Indicator for Vila Estrutural, 2008 to 2011 Physical Goal Nº Sub-Indicator 2008 2011 Forecast Actual Forecast Actual 1 Water supply 50% 93% 100% 98% 2 Sewerage 35% 33% 100% 92% 3 Garbage collection 50% 80% 100% 98% 4 Stormwater drainage 100% 10% 100% 85% 5 Street paving 50% 38% 100% 85% 6 Geological and geotechnical stability 100% 7,90% 100% 90% 7 Population density 0% 0% 100% 70% 8 Residential electricity 100% 79% 100% 96% 36 Physical Goal Nº Sub-Indicator 2008 2011 Forecast Actual Forecast Actual 9 Regularization 35% 0% 100% 79% 10 Urban cleaning service 50% 80% 100% 98% 11 Public lighting 50% 80% 100% 97% 12 Community and recreation facilities 50% 47% 100% 80% 13 Income 0% 0% 100% 70% 14 Access to education 0% 0% 100% 85% Total result 25% 39% 100% 88% The ISA-F indicator reflects the fact that, between 2004 and 2011, there was marked improvement in the level of access to basic services like water, sewage, and solid waste collection, as well as the condition of housing in Vila Estrutural. For example, in 2004 only 9.2% of households had access to sewerage services compared to 92.2% of households in mid-2012. The expectation is that by the end of 2012, 99% of households will benefit from network connections. Similarly, there was significant improvement in urban infrastructure. In 2004 only 3.5% of households were located in paved streets, and by mid-2012 this percentage rose to 84.6%. The rates of street lighting and stormwater drainage also improved, rising from 93.6% to 95.1% and 4.5% to 85%, respectively. Finally, there were positive improvements to the quality of housing structures in Vila Estrutural. In 2004 about 55% of households were classified as shacks— constructed with improvised or not permanent material. By 2011, this percentage dropped to 14.1%. The percentage of masonry built homes jumped from 48.1% to 85.1%, and the percentage of households with documentation of ownership (land tenure) increased from 0% in 2004 to 79.1% in 2011. In addition, several new facilities were completed, including a preschool, a community center, and a social assistance center. The GDF also completed several facilities in the neighborhood separate from the project, including a community restaurant, police stations, and a sports center. During the program implementation period, 2004 to 2011, improvements to educational attainment, the formality of employment, and income of residents in Vila Estrutural was observed. For example, the percentage of children not enrolled in school declined from 1% to 0.1% and the percentage of illiterate people decreased from 4.4% to 2.0%. The overall set of investments in Vila Estrutural has improved access to education for the residents, as well as overall access to services (social services, community facilities, etc). There was an increase in the number of schools in the neighborhood—two schools were built by the program and four more schools by the GDF, as well as centers for early children education. In addition, vocational courses are offered at the newly built community center. In 2004, the average income per capita was 0.4 times the minimum wage and family income was 1.9 minimum salaries. In 2011, these indicators rose to 0.6 and 2.3, respectively. The percentage of people with access to formal employment also increased. In 2004 about 34% of the residents of Vila Estrutural worked with documented salaries, in 2011 this percentage rose to 46.7%. Unfortunately, it is not possible to determine whether these improvements are due to project interventions or overall market conditions. 37 Annex 6. Stakeholder Workshop Report and Results N/A 38 Annex 7. Summary of Borrower's ICR and/or Comments on Draft ICR [Translation of document provided by the Borrower.] 1. PROJECT GOALS, IMPLEMENTATION AND EXPERIENCE. In January 2004, the carta consulta for the Sustainable Brasilia Program was sent to the Department of International Affairs of the Ministry of Planning, Budget and Management (SEAIN). After analysis by GTEC and COFIEX, the carta consulta was approved on April 26, 2004, at which point negotiations for obtaining external financing from the World Bank began. The carta consulta presented a proposal for a large-scale program focused on territorial management in the Federal District and its surrounding area – the metropolitan region of Brasília. As a result of impasses in the conditions of indebtedness existing at the time, the loan authorized by SEAIN for Phase I of the Program restricted the focus of the land management interventions to the watersheds of the Federal District—Descoberto and Paranoá. Phase I of the Program would support the planning of activities that would be carried out, in due time, under Phase II of the Program, both for the benefit of the population of the DF and the surrounding area. Thus, the preparation was carried out with the idea of a Program APL (Adaptable Project Lending), or in stages—with the second step occurring if the GDF met a number of conditions, the main one being the ability to pay the debt of the Federal District. The carta consulta provided a complete diagnosis of the DF and its surroundings in relation to economic, social and environmental dimensions, as illustrated in the following passage: "Designed as a sectorized administrative city with a set boundary, Brasilia has attracted a large flow of migrants since its construction. The population settlement and occupation of the land surrounding the city still occur without planning, outside the original intent. The region's population is growing at a higher rate than that of the country. There are now 2.0 million people in the Federal District, with more than 900,000 in the surrounding areas, living in conditions of uncontrolled growth... The housing units planned Brasilia have still not been fully constructed, which has contributed to the urbanization of new areas. In the Federal District, irregular condominiums proliferate. In the surrounding areas, there are 22 cities with no prospect of economic sustainability, unplanned and unequipped, with chronic needs and imposing permanent pressure on the original urban center, which was declared a World Heritage and National Heritage site." The Program was design with two concepts that were relatively innovative for the time: Sustainable Development and Human Development. The first was based on as desire to transform the future and the second was based on the understanding given by the United Nations when it created the Human Development Index - HDI as an objective of moving towards a cross-sectional analysis of development, valuing the social dimension. Thus, the Sustainable Brasilia Program was born as a new and great challenge, with the Federal District and its surrounding area formed by RIDE (Integrated Environmental Development Region) as its focus, defined based on the understanding that solutions to preserve the quality of life in the Federal District, as well as a real promotion of sustainable development, will need to occur at a regional level. To address this challenge, institutional strengthening actions for specific sectors of the GDF were designed, as well as plans and projects for environmental sanitation, housing and urban development; and land management and water resources. The 39 planning for watershed management was adopted based on the recognition of water as a strategic environmental attribute of the DF due to its scarcity, depending on the geographical location of water sources. The Project Appraisal Document (PAD), approved in July 2005, highlighted the following Program objective: "Ensuring the quality of water resources of the Federal District of Brasilia and the metropolitan region, promoting the improvement of living conditions of the population and sustainable land management." The program, endowed with a strong social and environmental nature, was designed primarily to: 1. Preserve the quality of surface water of the Federal District, most notably the Descoberto Lake which supplies more than 60% of the entire population of the Federal District, since its quality is being compromised by the rapid and unplanned population growth in areas of environmental fragility without urban infrastructure caused by the large demand for housing. 2. Improve the quality of life and social inclusion of the low-income population of Vila Estrutural, who currently survive in an unhealthy environment, close to the landfill and without appropriate urban services. 3. Improve the working conditions, survival, and human dignity of garbage collectors who currently use the Vila Estrutural landfill. 4. Adequately address the disposal of urban waste in the Federal District, closing the Vila Estrutural landfill, and developing a landfill designed with modern engineering techniques. The Program was structured with four components that contribute to the achievement of specific goals, as indicated below. Part A – Policy and Institutional Development: the Program’s support to the creation and strengthening of strong regulating institutions to allow the Federal District Government to have commitments and guarantees that the concept of sustainable development will be effectively deployed in the region. The main objective is to seek the environmental and institutional technical development of entities involved, in such a way that the GDF has tools and efficient means to provide itself with information to enable quick and timely decision making related to metropolitan issues, especially to promote sustainable urban and environmental development. Part B – Social Inclusion and Poverty Reduction: consists of the financing of works, activities, and sub-activities that are not only technically feasible and economical but also viable from an institutional, social and environmental point of view and that serve as models for new interventions in other target areas of the region (Descoberto and Paranoá basins). The main objectives include reducing poverty and promoting social inclusion of the population living in areas with critical social conditions, coupled with actions to reduce the pollution loads in water resources. Part C – Water Resources Protection: includes the promotion of an integrated approach to water resources management, oriented to future generations to ensure the balance between supply and demand for drinking water and protection of water sources from polluting activities. The objective is the financing of interventions aimed at reducing the pollution loads of water resources in the area of influence of the Program, with priority to the basins of the Descoberto River and Paranoá Lake. Part D – Project Management, Monitoring and Evaluation: refers to the financing of management activities, coordination and control of the physical and financial implementation of the program, 40 with the aim of ensuring that actions are carried out as prescribed in the Loan Agreement No. 7326-BR and the PAD. There is not space to repeat in this annex the entire structure of the Program and its various components and subcomponents. The goal here is to highlight some aspects of the context in which the various changes and evaluations were done to the Program. In this sense it is worth highlighting the reasons that led to the prioritization of the intervention in Vila Estrutural: a) Vila Estrutural is located in a strategic position within the urban fabric of the Federal District, in its "core," only 12 km from the capital of Brazil; b) It is surrounded by productive areas on one side and environmentally protected areas on the other, which requires setting clear urban boundaries; c) It is located in the sub-basin of the Riacho Fundo/Vicente Pires, and therefore, drains into waterways that feed the Paranoá Lake with a risk of degradation of water resources; d) Its location near the Brasilia National Park represents a vector of degradation to ecosystems it contains and puts at risk its biodiversity; e) It was oldest invasion that had not yet been regularized. The social organization of its residents had great importance for their ability to remain on site because they always resisted any attempt to remove the invasion; f) To be considered a Residential Zone of Social Interest (ZHIS) required an appropriate solution for its integration into the urban fabric and the economy of the city; g) The environmental conditions aggravated by the unplanned occupation of Vila Estrutural and the need for public investment to improve the living conditions of its residents qualified it for a more effective and priority intervention, with far-reaching social, economic, urban and environmental actions. Thus, Vila Estrutural entered as a beneficiary of urban designs, studies, and works that aimed to change the local conditions, including resettlement actions. The history of Vila Estrutural, plus the recurring structural problems related to land tenure in the DF, foretold that the implementation of the Program would encounter significant challenges. The "Invasion of Vila Estrutural" dates back to the early 1970s, with the installation of substandard housing by scavengers who survived on the waste deposited in the Jóquei Clube landfill, better known as the "Jóquei Dump." During the 1970s and 1980s, the invasion increased slightly, not significantly affecting the environment of the Brasilia National Park. Since then, the occupation began consolidated due to the increasing number of scavengers and the establishment of small farms. In 1993, 393 households were registered, of which 149 belonged to scavengers. By 1994, the number of households had doubled. Despite several attempts to remove the invasion, Vila Estrutural continued to grow and consolidate, from about 2,800 people in 1994 to 25,000 in 2003. In 1998, part of the area was designated/zoned as the Industry and Warehouse Sector (SCIA), with about 980 commercial lots. The interest and need for its regularization led to approval in the Legislative Assembly of DF in 2002 of Law No. 530, which administratively created Vila Structural and declared it as a 41 Residential Zone of Social Interest (ZHIS). In December 2005, IBAMA issued the Preliminary License for regularization of Vila Estrutural. On January 25, 2006, Complementary Law No. 715 established the Special Zone of Social Interest (ZEIS) for Vila Estrutural, making it possible to regularize existing housing units under special urban planning conditions and standards, as well as public facilities needed to service its population. In 2007, at the beginning of the Program, the population of Vila Estrutural was estimated at 35,000 people and had precarious urban infrastructure, as in any irregular settlement: a. The urban core was composed of 21 blocks, two special areas, a workshops area, and a central area; b. The blocks were formed by subnormal housing, i.e. erected without adequate construction and sanitary facilities, often with only one room; c. There were families occupying hazardous areas and in areas defined for public use, with the aim of readjust the internal road network or implanting urban infrastructure. As such, an Involuntary Resettlement Plan was required; d. The streets did not meet the official standards for a subdivision, were not paved, and produced excessive dust in the dry season. In the rainy season, they became muddy and almost impassable due to a lack of drainage; e. There was a need to solve the stormwater drainage problem due to, among others, the risk of contamination of nearby aquifers and to define the internal and external road network to connect it with the surrounding road system in order to facilitate accessibility to Vila Estrutural and its integration into the urban area of Brasília; f. As of 2003, the water supply in Vila Estrutural was done through a distributor network, only in Blocks 3, 4 and 5. In other blocks, the supply was made by tanker trucks from the Sanitation Company of the Federal District (CAESB). g. There was no sanitary sewer collection system. The domestic sewage was deposited in cesspools installed in the back of the houses, where bathrooms are also located. It was common dispose of liquid and solid waste into the public roads; h. Garbage was collected by the Cleaning Urban landscaping Company of the Federal District, every two days, in a few blocks (20% of households), because the trucks did not have access to all of the blocks due to the insufficient width of the streets and the proliferation of illegal connections to electricity facilities. Household garbage was burned in 16.6% of households and thrown in open areas by 62.5% of households. i. The urban and community facilities serving the local population were virtually nonexistent:  There was only one school "Escola Classe nº 1 da Vila Estrutural ", which served about 1,400 students. Some students were driven by public transport to schools in Guará;  Health care: The residents had a poor medical care unit (Posto de Saúde da Vila Estrutural), with rooms for children and women, a pharmacy, immunization, and dentistry.  There was no preventative policing or fire department.  Spaces for leisure, recreation, and culture were practically inexistent. j. Vila Estrutural was one of the most violent areas of the Federal District. According to its residents, deaths and nights shootings were constant. Cars were stolen in the light of day and drug trafficking occurred without limits. Residents reported the existence of two handmade weapons factories on site. There was only one police kiosk with only three officers. 42 Even though Vila Estrutural had a population of low income people and precarious housing, infrastructure, and urban services, the privileged location and imminent regularization and urbanization of the area increased its attractiveness each day, and consequentially the arrival of new residents. These observations regarding Vila Estrutural, combined with the fact that implementation of the Program depended on the direct action of different institutions of the GDF, as well as the State Government of Goias, foretold the complexity of the challenge that the GDF and IBRD assumed upon signing the Loan Agreement. 2. IMPLEMENTATION AND OPERATION EXPERIENCE This section briefly addresses the main facts which constitute the history of the Program. The complete reports can be found in the biannual reports prepared by the Program and sent to the Bank in the form of "Progress Reports." Although the Loan Agreement was signed in February 2006, the Program only became effective in 2007, due to delays in establishing the PMU, due to difficulties that are typical during the period of government transition (beginning of the government of José Roberto Arruda). In April 2007, by Decree No. 27.833, it was determined that the responsibility for the overall coordination and implementation of the Program would be transferred to the Secretary of State for Urban Development and Environment of the Federal District (SEDUMA), revoking Decree. 25.845 of May 17, 2005 that delegated authority to ADASA. As a result, SEDUMA, through Ordinance No 31 of 4/23/2007, began the process of revising the organizational structure for the implementation of the Program, as well as the provisions of the agreements with the co-executors and in the loan agreement. In addition, it promoted a series of actions to enable the structuring of the Program, such as: (i) assessment of the procurement process for the company to provide consulting services for the institutional strengthening and capacity building for implementation and monitoring of the Program, (ii) revision and signature of the loan agreement, mainly to adapt it to the new administrative structure of the GDF, (iii) organization of the sources and uses of funding for the Program, reordering the activities within the new administrative structure of the GDF, (iv ) revision of the timeline for the physical and financing implementation of the Program, and (v) revision of the procurement plan for the Program, primarily to streamline the procedures for hiring consultants and works. One year after effectiveness, the Program already had 60% of the loan contracted or under procurement, especially those allocated to infrastructure in Vila Estrutural. Although this seems positive, it also had a troubling side: the early works led to the removal of families before the completion of houses under the resettlement plan (many faced security risks if they remained in their homes). The families were placed in temporary accommodation and/or social rental (two options planned as temporary solutions in the resettlement plan). The solution generated complaints to the Bank from some residents, causing a series of meetings and visits that resulted in commitments between the GDF and the Bank to expedite the resettlement of the population and immediately improve conditions in the temporary accommodation. This disconnect between the timelines and the difficulty in finding an appropriate mechanism that would constitute a consensus among the Bank's rules and GDF were the trigger for an exhausting process that held up the Program on different occasions. 43 In this first year of effectiveness, two important actions were taken: (i) SEAIN approved the carta consulta for additional financing US$57.6 million and (ii) the Governor identified the location for the new sanitary landfill and the PMU already had the land documentation necessary to start the designs. At the same time, the institutional strengthening component became a major challenge for the PMU once it established that the benefiting agencies should have a leading role in deciding the type of services to be contracted to ensure the internalization of the results consultancy contracts. However, these agencies presented difficulties in defining priorities and draw up the Terms of Reference, which made the PMU seek greater support in the area of procurement to assist in the preparation of TORs. In 2008, the Program completed the designs, economic feasibility study, and the first version of the bidding package for the new landfill. In 2009, the Program continued efforts with the landfill and it showed encouraging results: in March a public hearing was held with acceptance of the terms of the concession by the community of the Federal District; a law was passed allowing the concession and defined the partnership between SEDUMA and ADASA (the regulator of solid waste) by editing the Ordinance to finalize the joint announcement of the landfill. In September, the first attempt to receive proposals for hiring a specialized company was held. A strategy was also prepared for social inclusion of scavengers in Vila Estrutural. At the request of scavengers, a recycling triage center would be built in the neighborhood. Also in Vila Estrutural, a contract for construction of the houses was signed and the first 32 homes were delivered to families who were in temporary accommodation. Also in 2009, an additional 37 houses were handed over in Vila Estrutural, totaling 69 families attended. 16 houses were completed in the rural area of Monjolo. The consultancies in the institutional development component were also moving forward satisfactorily. In the second half of 2009, significant actions were taken and/or initiated: completion of the Supplemental Environmental Impact Report (RIAC) for Monjolo; contacting of the company to do the PTTS in Vila Estrutural; identification of the area and the elaboration of the designs for the Recycling Triage Center; signing of Contract No. 7869/2009 for the initiation of consulting services for environmental certification in CAESB; signing contracts 14/2009, 21/2009, 22/2009 and 25/2009 to give first, respectively, the following consulting services: strategic planning for environmental management, ecological and economic zoning of the DF; land regularization of São Sebastião and master plan for the conservation unit of the Botanical Garden; signing of Contract No. 7931/2009 concerning the implementation of the sewerage system in the western portion of Vicente Pires. However, in 2009 the problems relating to Monjolo began: the construction of resettlement housing for families with activities in the rural area were stopped due the absence of environmental licensing. The institutions involved had different positions on whether or not the works required an environmental licensing process. This impasse led to the suspension of disbursements by the Bank until a resettlement solution for rural households was provided as well as a set of demands that were difficult to carry out considering that the various institutions involved were having serious problems working in an integrated manner, especially due to political positions. Among these demands was the finalization of the bidding package for the new landfill, which was in its sixth revision as a result of discussions with the Public Prosecutor, Court of Accounts, and Attorney General of the GDF and the incorporation suggestions from the public hearing; the 44 solution to land disputes related to the installation of sewage pumping stations in Aguas Lindas and Vicente Pires; obtaining the environmental license for the implementation of the landfill closure and remediation services and the implementation of the recycling triage center designed; and the need to prevent further invasions in Vila Estrutural and Monjolo. The year 2010 was marked by events that greatly influenced the progress of the Program. Administrative changes and political problems in the GDF led to the Decree No. 31.419 of March 16, 2010, which determined that the overall coordination and implementation of the Program would be transferred from SEDUMA to the Regulatory Agency for Water, Energy and Sanitation District Federal (ADASA). On May 6, 2010, the Program received a letter informing that the Bank suspended disbursement of activities not contracted prior to the date of issue of the correspondence. The legal nature of ADASA makes it less susceptible to occasional changes and the maintenance of the Program team during the change in coordination enabled the PMU to continue Program, but the difficulties were many. The implementation of the Program involved the participation of ten different institutions of the GDF. This decentralized arrangement, coupled with small number permanent technical staff and the constant political-administrative changes, hampered the maintenance of the technical teams involved in the implementation of projects affecting the continuity of actions, and thus compromised the deadlines and targets agreed between the GDF and the Bank. Besides these factors, there was interference from residents and the housing developed suffered financial problems, which caused repeated delays in construction at Vila Estrutural. At the same time, it was not possible to resettle families in the houses that were already built due to an injunction issued by Circuit Judge of the Environment Court of the Federal District and Territories (TJDFT), which prevented the GDF to resettle any residents until concerns regarding the health/safety of inhabiting the area was clarified/resolved. The financial operation of the Program was added to this list of difficulties due to the suspension of disbursements of external resources, pursuant to the Letter of Suspension of the World Bank. In October 2010, the Bank issued a letter allowing the disbursements to be partially restored for the categories of goods and consultants in the form of reimbursements and/or direct payments. This was due to the good financial management of the Program which kept all the financial arrangements in order, reflecting the good administration of the loan proceeds, and the submission of quarterly reports (IFRS). In these circumstances, the actions taken during the period related primarily to monitoring and technical support to resettlement activities (in Vila Estrutural and Monjolo); continuity of procurement processes for consultancies; and continuity of actions and dealings with the Secretary of State of Finance and Treasury of the GDF in order to enable the new procedures on "direct payment" and "reimbursement" by the Bank. Due to the suspension of disbursements by IBRD, and considering the proximity of the end of the 2010 fiscal year, reallocations of resources from the GDF were made to meet resource demands for work completed. In December, the Program was restructured to reflect possible achievements. The scope of the Program and the indicators were revised. 45 In 2011, the activities continued to reflect the operational situation of the Program considering the partial suspension of disbursements. The PMU tried to keep the focus on core activities in the resettlement at Monjolo (construction of 40 housing units); the continuity of sanitation activities being developed by CAESB as counterpart; and the bidding processes and contracts for consulting and purchasing goods. The long period of suspension of disbursements and problems in resettlement activities resulted in the slowdown of some strategic activities. Consequently, the GDF on 1/13/2011 requested an extension the closing date until December 2011. The request was approved by COFIEX on 1/19/2011 and accepted and formalized by the World Bank on 3/31/2011. It was also noted that loan proceeds could not cover costs of eligible activities carried out after the closing date of December 31, 2011. These would become the responsibility of the Federal District. Additionally, the Executive Coordinator of the Program developed strong institutional links with the various technical and political segments of the "new administration of the Federal District," which began on January 1, 2011. The activities developed during 2011 were impacted by the restrictions of the partial suspension of disbursements, which was finally lifted on September 29, 2011. In this sense, the extension period granted to the Program—from March to December 2011— was largely undermined by the constraints imposed, hindering the work of the PMU/ADASA. Still, it can be stated that despite the difficulties, a series of positive actions were developed, namely:  Recognition by the World Bank mission in the aide memoire that the PMU was well trained, motivated, and persistent. Recognition, in addition, of important advances such as the completion and delivery of the second phase of houses Monjolo, with a major impact on improving the lives of the beneficiary population. The Program has also demonstrated good financial management and procurement.  Contracting of the following: o Restructuring of the SLU – contract signed with Ernst &Young Terco. o Management Plan for the Protected Areas of Vila Estrutural – contract signed with GREENTEC Tecnologia Ambiental.  Comlpetion of contracts: o Completion of the contract for “Study on Implementing an Environmental Management System in CAESB, aimed at ISO 14.001 certification of ETE Norte e ETA Descoberto� CT 7869/09; o Completion of the contract with a company specialized in remote sensoring images to support environmental licensing activities at IBRAM; o Completion of the acquisition of equipment for strengthening the environmental licensing process of IBRAM; o Completion of the acquisition of software for strengthening the environmental licensing process of IBRAM; o Completion of the consultancy on “Survey and treatment of the liabilities of sewage systems�; o Completion of consultancy on “Management and preservation of wetlands�.  Consolidation of the Works Monitoring Committee at Monjolo consisting of six residents. The committee was created to track the progress of the works of the second phase of housing to communicate any problems to the building inspector (Novacap), through weekly meetings. 46  Meeting with community members. The PMU, through weekly visits to Monjolo, was in direct contact with the residents and the Works Monitoring Committee. For this activity, the PMU provided a car and two full-time employees.  Integrated action by AGEFIS, SUDESA, CAESB, PM, CEB and ADASA to remove invasions from Monjolo. On that occasion, 16 squatter families were removed and their shacks demolished.  Works in Monjolo o On August 29, 2011, the final phase 2 housing units were completed, totaling 41 units built in that phase; o The installation of electricity and water services were completed for phase 2; o Curbs and containment boxes were installed to control erosion, primarily at the entrance to Monjolo; o Cleaning and removal of construction materials and shacks from protected areas with the participation of SEAPA and SLU; o TERRACAP installed fences around the protected areas, 80% completed; o The construction of 44 housing units was completed in Monjolo. 12 were delivered to the beneficiary families on a provisional basis.  In August 2011, 316 housing units were delivered to beneficiaries in Vila Estrutural, totally 583 beneficiary families.  Satisfactory financial implementation of the Program through reimbursements in the categories of consulting and acquisition of goods despite the suspension of disbursements;  Inclusion of necessary resources in the 2012 budget for the completion of Program activities;  Completion and submission to the Bank of a report on the status of involuntary resettlement activities. 3. RESULTS OF THE OPERATION [This section has been incorporated into the main body of the ICR—specifically, sections 3.2, 3.5 (a), and 3.6.] 4. EVALUATION OF BORROWER PERFORMANCE AND LESSONS LEARNED Taking advantage of the benefits of post implementation reviews and comments, it appears that the challenges presented by the Program can be considered relative. It was designed based on input from people heavily involved in the process and the level of political will that was presented at the time, especially the coordinating agency. Therefore, strategies for dealing with political and administrative changes that would alter the power relations in the GDF were not considered. In the Bank’s evaluation, it is recognized that “a key weakness in the design of the project was the underestimation of the risks related to the institutional complexity of the GDF... The project team considered that institutional complexity would be a challenge in relation to metropolitan management and environmental licensing, but underestimated its impact on resettlement activities.� The borrower also had this error of perception, and the importance of this fact could have been addressed without much gravity if, upon the early detection of this error, the Bank and borrower came together to deliberate joint strategies for addressing the problems. But actions were taken individually, in a design in which the roles are well defined and separated. 47 One consequence of this form of management, which caused continuous problems with the implementation of the Program, was the relationship that the beneficiaries of the resettlement actions created with the Bank and that the Bank began to play the role of receiving complaints. This role could have existed, but not as an intermediary between the beneficiaries and the GDF. Certainly, the housing problems in the DF involve serious social problems. But it is also true that there is a strong politicization of invasions and it would have been for the Bank and the GDF to quickly solve the problems that arose with the existing rules so that the Program could come to a definite stand in regard the complainants, thus reducing the possible interference from local leaders more interested in personal gain and imprinting their political positions and less on meeting the real needs of the community involved. It is difficult in an ex-post evaluation to separate the borrower and the Bank. It is evident that many of the successes and failures during the implementation of a Program the size of this are directly linked to how the relationship developed between the teams. In this sense, it is of fundamental importance to review the lessons learned during the development of the Program. The lessons that follow are part of a process of reflection carried out throughout the implementation of the Program and endorsed during the evaluation workshop held in January 2012 with the GDF technical institutions involved: 1. A program implemented in a decentralized manner, like the Brasilia Environmentally Sustainable Program, necessarily involves a large number of actors. Thus, it is essential that these actors participate in the preparation of the program, particularly in defining goals and actions that will be the responsibility of each co-executor. Participation must be qualified, in the sense that participation of technical staff from the institution are given priority, in order to promote organizational memory, successful development and implementation of projects, as well as the continuity of the program. 2. During preparation, the program should include mechanism to include participation of community organizations that are representative of the direct beneficiary population. 3. Another key item of learning is the need to develop and adopt legal instruments and legal frameworks for implementing the program during its preparation, so as not to preclude the implementation of programmed activities. 4. The institutional arrangement adopted for the Brasilia Environmentally Sustainable Program was based on a flat hierarchy, in which different institutions would work in an integrated manner in support of the goals. During the implementation, several complications to this arrangement occurred, namely: (i) constant political changes that have resulted in the lack of direct dialogue between the central core of the Government and the Program coordination team; (ii) weakness of institutions due the lack of effective technical staff, which meant that the positions were filled by political appointees; (iii) weaknesses in coordination between the agencies. 5. Technical staff involved in the implementation noted that the institutional arrangement for similar programs should include a planning system with preventive and corrective actions, such as: (i) continual disclosure regarding the Program for the different GDF institutions involved including: goals, actions, rules (Bank safeguards), specific laws, difficult points; and (ii) continual training of the technical staff directly involved in implementing the Program. 48 6. Overall, the lesson learned is that the institutional arrangement needs to have a deliberative body that, based on technical studies, is capable of making decisions in a timely manner to ensure the results contractually defined for the program. It is also necessary to maintain and enhance the role of the Local Government Units (UGLs) and PMU that were essential in implementing the Program. 7. The Program went through a large number of administrative and political changes that affected some activities from conception to implementation. These changes have brought challenges to the establishment of priorities and the pace of implementation. The need for redefinition and readjustment of the activities brought direct impact on costs, disbursements, and schedule delays and affected the performance and measurement of project results. A lesson learned from this case is the need to create mechanisms to ensure the continuity of the actions agreed, even in scenarios with political changes, as occurred in the DF. In addition, there is a need for greater flexibility on the part of the Bank to renegotiate terms and actions, due to situations outside the control of the program, such as embargoes or judicial cases of contractors who delay the delivery of works due internal management issues. 8. The Program faced several challenges related to resettlement—identifying families once the reality of occupation, removal and reoccupation of areas for resettlement started happening at a speed that was difficult to control and the process of applying the legal/statutory requirements the rules of the Bank. The GDF had no legal instruments that would allow the government to offer resettled residents options for compensation. Thus, the GDF used the law applicable to housing programs, which includes a series of demands for which some families had difficulty complying. Despite the challenges, the resettlement was done, with families having only one compensation option according to their profile—urban or rural. The main lesson learned in this item refers to the need dimension resettlement during the preparation of the Program, freezing or controlling the areas identified for resettlement in order to maintain the initial design and development of a legal instrument covering different options to compensate different types of affected people. 9. Regarding the execution of works, in the case of predicted removal, whether of people or physical obstacles such as power grids, etc., it is necessary to condition the start of construction on the clearance of the affected areas. 10. In the case of not being able to clear the area in advance, the start of construction in areas of apparent ease should be avoided in order to avoid the temptation to start works at a pace that is faster than what is real. 11. Avoid changing the entity that is responsible for monitoring physical works because this could hinder effective performance due to the lack of continuity and full background knowledge, resulting in additional charges. If the change is necessary, ensure that the knowledge and facts are transferred to the new manager. 12. Have permanent schedule of technical monitoring meetings with a defined frequency and participation from the executing units, and representatives from the executing companies, regional administration, and other bodies involved. The main goal of this type of meeting is to check progress and problems, and to seek immediate solutions or ways forward. 49 Given this set of observations and the results achieved, the borrower's performance was considered satisfactory. 5. EVALUATION OF BANK PERFORMANCE AND LESSONS LEARNED As previously explained in this report, it is difficult to separate the actions of the Bank of the borrower. Reviewing the history of the Program, the progress reports, aide memoires and the ICR prepared by the Bank, different findings can be made, some as lessons learned and some as suggestions for new programs to achieve higher level of success. First, the format of the ICR does not lead the reader to be interested in the analysis. The first pages already announce the end of the story without explaining the method of evaluation, which only starts to be detailed in item 3.4 of the report. It is a bulky document, in English, full of analysis in an institution where the analyst early on receives information that the Program was evaluated as unsatisfactory, and, for the vast majority of readers, this is the final note showing that that there was disapproval. In this sense, it would be a good idea for the report to have an introduction that outlines the objectives, concepts used, and some important information on the formulation and implementation of the project. This suggestion aims to encourage readership of the evaluation reports so that successes and failures are analyzed and help in shaping and implementing new programs. In the case of the Brasilia Environmentally Sustainable Program, it would be of utmost importance that an initial summary would encourage the reader to understand why, even with all expertise involved, the borrower and the Bank underestimated the risks associated with the institutional complexity of the GDF. Or why the program was classified as unsatisfactory by the Bank considering the obvious social improvements that have occurred in Vila Estrutural and other good results of the Program in Vicente Pires and with consultant deliverables. It would be of particular interest to the GDF and future programs that readers analyze and understand why there were so many revisions to the bidding package for the new landfill. In addition, they could form an opinion on whether or not to invest public money in solid waste projects knowing that the area has a deficit of institutional and technical means to allow the appropriation of the results in favor of effective public policy. Why was a Terms of Reference not approved in a program the size of Sustainable Brasilia? Capacity to prepare a good TOR exists or is hired. But the continuity of actions? How can the GDF invest in the training of the Jóquei landfill scavengers without the supporting projects? These are questions that lead to the questioning of the responsibilities of each technical specialist and each manager in implementing a project. Specifically in relation to the landfill, presented below are some considerations on the subject that try to give a general idea about the difficulties faced in conducting the bidding process. Firstly, it should be noted that the nature and complexity of the project, which is deploying the first landfill in the Federal District, in itself attracted the attention of several entities, including the Public Ministry of the Federal District and Territories (Ministério Público do Distrito Federal e Territórios) through its Attorney for Environmental Defense (Promotoria de Defesa do Meio Ambiente) who, upon learning of the publication of the draft bidding documents, ordered an examination and subsequent manifestation. 50 This fact has motivated a series of meetings with prosecutors that required clarification and discussion of technical issues, leading to suggestions for revision of the draft in question. Likewise, the Attorney General of the Federal District (Procuradoria Geral do Distrito Federal), as part of its mandate, should also review the legal aspects, from those relating to specific legislation as well as permission or restriction on the participation of consortia and enforcement regime. It is worth adding, though, the participation of other units of the organizational structure of the Federal District like the Department of Works, Urban Cleaning Service (SLU), Environmental Institute Brasilia (IBRAM), TERRACAP, the Department of Housing and Urban Development (SEDHAB), IBAMA; representatives from the sector, civil society, and the bidders themselves, each with their comments and suggestions which required great amounts of analysis and persuasion, with implications for the finalization of the draft bidding package. One cannot fail to also mention the active participation of the Court of the Federal District (Tribunal de Contas do Distrito Federal) in the review process, with several requests for clarification and answers, as a way of ensuring the smoothness of the act and the greater efficiency of the entire process. Another factor that may give rise to reflections is the relationship between the Bank and the borrower. In the specific case of Sustainable Brasilia, the dialogue between technical staff was always friendly and professional, creating a climate conducive to work. However it would be very helpful to rethinking the process for measuring progress and addressing difficulties during the implementation process. The missions have become bureaucratic and unproductive, a forum where the Bank identifies issues and the borrower looks for ways to defend itself, when they should be periodic evaluations by partners of the same agreement, of a plan with successes and mistakes where a meeting between the parties can make all the difference in finding solutions and correcting the path forward. Technical assistance received from the Bank also deserves mention—starting from the principle that it is a development program and not just an investment project. Here, the competence and commitment of management and the technical team is not being called into question, but rather the consequence of the amount of projects that the Bank has in its portfolio, making a small group responsible for attending to and training teams in various states of a country like Brazil which is of continental dimensions. This reality can also generate uncomfortable situations between the borrower and Bank teams. An example of which is the aide memoire for the mission held in February 2010 when PMU approved a text that was then denied and modified by Bank Management. Another difficult period in which the team felt the Program needed greater guidance and greater clarity regarding the rules and procedures of the Bank was the suspension of disbursement, during which conditions were changed throughout the period of suspension. It is also important to mention the need for training. A program should not be initiated without the Bank, PMU, UGLs and staff from the coordinating institution coming together for training on the standards and rules of the Bank. There is resistance in legal and procurement areas of public institutions to work with rules for which they have not been trained, which inevitably leads to delays in the process. It is critical that the PMU, Bank, and the institution that houses the project are in full agreement with the rules and regulations and procedures to be followed. The borrower 51 should be aware that it needs to arrange training, together with the Bank, whenever there are staffing changes. More specifically, in relation to the monitoring of physical works executed in the Program, it is necessary to define the Bank and borrower monitoring report template, deciding by consensus, a model that can reflect stage of work at anytime, as well as indicating of the progress of activities and any significant deviations. Given the observations made in this chapter on the performance of the borrower and the Bank, observations that are intended to contribute to the design of new programs, believing in the richness of actual experiences to replicate successes and mistakes, the Program coordinating entity and the PMU of Sustainable Brasilia believe that the performance of the Bank and the borrower, considering the context of implementation and the results, was satisfactory. 52 Annex 8. Comments of Cofinanciers and Other Partners/Stakeholders N/A 53 Annex 9. List of Supporting Documents  Project Appraisal Document (Report No: 32941-BR) – July 6, 2005  Loan Agreement (Loan Number 7326-BR) – May 21, 2007  Resettlement Action Plan (Plano Físico e Social de Relocação e Reassentamento das Famílias Beneficiárias do Projeto Integrado Vila Estrutural) – May 2008  Indicators Monitoring Report (Relatório de Monitoramento do Programa Brasília Sustentável) – August 2009  Restructuring Paper (Report No: 58442-BR) – December 8, 2010  Restructuring Paper (Report No: 60256-BR) – March 30, 2011  Status of Resettlement Report, December 2011  Borrower Progress Reports Aide Memoire – May 2004 Aide Memoire – September 2004 Aide Memoire – December 2004 Aide Memoire – February 2005 Aide Memoire – August 2006 Aide Memoire – May 2007 Aide Memoire – September 2007 Aide Memoire – October 2008 Aide Memoire – February 2009 Aide Memoire – August 2009 Aide Memoire – October 2009 Aide Memoire – November 2009 Aide Memoire – February 2010 Aide Memoire – May 2010 Aide Memoire – November 2010 Aide Memoire – June 2011 Aide Memoire – September 2011 Aide Memoire – May 2012 Implementation Status Report 1 – May 30, 2006 Implementation Status Report 2 – December 27, 2006 Implementation Status Report 3 – June 13, 2007 Implementation Status Report 4 – October 26, 2007 Implementation Status Report 5 – April 14, 2008 Implementation Status Report 6 – November 20, 2008 Implementation Status Report 7 – May 9, 2009 Implementation Status Report 8 – October 19, 2009 Implementation Status Report 9 – May 10, 2010 Implementation Status Report 10 – February 16, 2011 Implementation Status Report 11 – July 19, 2011 Implementation Status Report 12 – December 26, 2011 54 To Ciudad Guayana 70°W 60°W 50°W 40°W R.B. DE GUYANA VENEZUELA SURINAME French Guiana BRAZIL r i n oc (Fr.) AT LA NT IC O o COLOMBIA Boa Vista O CEA N AMAP� STATE CAPITALS RORAIMA NATIONAL CAPITAL Macapá 0° 0° RIVERS Ne zon gro Ama MAIN ROADS A m a z o n Belém São Luís Manaus RAILROADS Am az o jó s n Fortaleza STATE BOUNDARIES pa a Ta eir ad B a s i n Teresina RIO GRANDE M PA R � MARANHÃO INTERNATIONAL BOUNDARIES CEAR� DO NORTE AMAZONAS Natal u s Pur u Xing PARA�BA João Tele PIAU� Pessoa s Porto Velho PERNA Tocantins Pir es MBUC O Recife ACRE Maceió ia Rio Branco o Palma sc Aragua 10°S RONDÔNIA Juruena ALAGOAS ci an Aracaju São Fr TOCANTINS BAHIA MATO SERGIPE PERU GROSSO Mato Grosso B r a z i l i a n Salvador Plateau Cuiabá F.D. BRAS�LIA BOLIVIA GOI�S Goiânia MINAS GERAIS PA C IF IC To H i g h l a n d s Santa Cruz MATO GROSSO Belo ESP�RITO OC E AN CHILE DO SUL Grande Horizonte SANTO Paragu y 20°S Campo 20°S Grande Vitória To a Santa Cruz SÃO PAULO á an RIO DE r Pa PARAGUAY São Paulo Rio de JANEIRO Janeiro PARAN� AT L AN T I C Curitiba BRAZIL STA CATARINA OCE AN Florianópolis RIO GRANDE DO SUL Porto Alegre 0 200 400 600 Kilometers 30°S 30°S ARGENTINA SEPTEMBER 2009 IBRD 33377R This map was produced by the Map Design Unit of The World Bank. To Buenos The boundaries, colors, denominations and any other information Aires URUGUAY 0 200 400 Miles shown on this map do not imply, on the part of The World Bank Group, any judgment on the legal status of any territory, or any endorsement or acceptance of such boundaries. 70°W 60°W 50°W 40°W To Montevideo