Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00004749 IMPLEMENTATION COMPLETION AND RESULTS REPORT IDA Credit Nos. 5004-BO, 5712-BO, and 5713-BO ON A CREDIT IN THE AMOUNT OF SDR68.2 MILLION (US$100 MILLION EQUIVALENT) TO THE PLURINATIONAL STATE OF BOLIVIA FOR THE Community Investment in Rural Areas Project (PICAR) August 27, 2021 Agriculture And Food Global Practice Latin America And Caribbean Region Official Use CURRENCY EQUIVALENTS (Exchange Rate Effective December 31, 2020) Currency Unit = Boliviano (Bs) Boliviano 6.86 = US$1.00 US$1.44027 = SDR1.00 FISCAL YEAR July 1 - June 30 Regional Vice President: Carlos Felipe Jaramillo Country Director: Marianne Fay Regional Director: Anna Wellenstein Practice Manager: Preeti S. Ahuja Task Team Leader(s): Julia Navarro ICR Main Contributor: Julia Navarro Official Use ABBREVIATIONS AND ACRONYMS AF Additional Financing CDD Community Driven Development CPF Country Partnership Framework CPP Community Participatory Planning EIRR Economic Internal Rate of Return ESMF Environmental and Social Management Framework EMPODERAR Emprendimientos Organizados para el Desarrollo Rural Autogestionario (decentralized unit of MRDT, dedicated to fostering locally managed rural development) FAO Food and Agricultural Organization of the United Nations FIRR Financial Internal Rate of Return FM Financial management GMIS Geo-referenced Management Information System (Sistema de Información Gerencial Geo Referenciado), EMPODERAR GoB Government of the Plurinational State of Bolivia IDA International Development Association IE Impact Evaluation IFR Interim Financial Report IRR Internal rate of return ISR Implementation Status and Results Report LA Latin America M&E Monitoring and evaluation MDRT Ministerio de Desarrollo Rural y Tierras (Ministry of Rural Development and Land) MEFP Ministerio de Economía y Finanzas Públicas (Ministry of Economy and Public Finance) MPD Ministerio de Planificación del Desarrollo (Ministry of Development Planning) O&M Operation and Maintenance PAD Project Appraisal Document PCU Project Coordination Unit PICAR Proyecto de Inversión Comunitaria en Áreas Rurales (Community Investment in Rural Areas Project) PND Plan Nacional de Desarrollo (National Development Plan) POM Project Operational Manual PPC Planes Participativos Comunitarios (Participatory Community Plans) RF Results Framework SIGG Sistema de Información Gerencial Geo Referenciado (Geo-referenced Management Information System), EMPODERAR TA Technical Assistance ToC Theory of Change TTL Task Team Leader Official Use TABLE OF CONTENTS DATA SHEET ........................................................................................................................... I I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 1 A. CONTEXT AT APPRAISAL ......................................................................................................1 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) ....................................6 II. OUTCOME ...................................................................................................................... 8 A. RELEVANCE OF PDOs ...........................................................................................................8 B. ACHIEVEMENT OF PDOs (EFFICACY) .....................................................................................9 C. EFFICIENCY ........................................................................................................................ 14 C. JUSTIFICATION OF OVERALL OUTCOME RATING ................................................................. 16 D. OTHER OUTCOMES AND IMPACTS (IF ANY) ........................................................................ 16 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 18 A. KEY FACTORS DURING PREPARATION ................................................................................ 18 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 21 A. QUALITY OF MONITORING AND EVALUATION (M&E) ......................................................... 21 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE.................................................. 23 C. BANK PERFORMANCE ........................................................................................................ 25 D. RISK TO DEVELOPMENT OUTCOME .................................................................................... 27 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 31 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 41 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 43 ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................... 44 ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ... 52 ANNEX 6. SUPPORTING DOCUMENTS .................................................................................. 53 ANNEX 7. PROJECT EVALUATIONS AND ASSESSMENTS ......................................................... 54 ANNEX 8. SUBPROJECT CYCLE .............................................................................................. 61 ANNEX 9. GEOGRAPHIC COVERAGE, PICAR: CATEGORY A COMUNITIES REACHED................. 62 ANNEX 10. PHOTOGRAPHIC EXAMPLES................................................................................ 63 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name P107137 BO PICAR Community Investment in Rural Areas Country Financing Instrument Bolivia Investment Project Financing Original EA Category Revised EA Category Partial Assessment (B) Partial Assessment (B) Organizations Borrower Implementing Agency Plurinational State of Bolivia EMPODERAR, Ministry of Rural Development and Lands Project Development Objective (PDO) Original PDO To improve access to sustainable basic infrastructure and services for the most disadvantaged rural communities selected in some ofthe poorest municipalities of Bolivia. Page i of v Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 40,000,000 39,759,466 36,325,256 IDA-50040 55,700,000 54,988,507 55,125,429 IDA-57120 4,300,000 4,299,931 4,279,914 IDA-57130 Total 100,000,000 99,047,904 95,730,599 Non-World Bank Financing 0 0 0 Borrower/Recipient 0 0 0 Local Communities 3,000,000 4,100,000 18,575,505 Total 3,000,000 4,100,000 18,575,505 Total Project Cost 103,000,000 103,147,905 114,306,103 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 21-Jul-2011 23-Nov-2011 22-Sep-2014 30-Apr-2017 31-Dec-2020 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 04-Aug-2015 26.51 Additional Financing 19-Jun-2019 82.49 Change in Loan Closing Date(s) Reallocation between Disbursement Categories 01-Apr-2020 95.14 Change in Loan Closing Date(s) 30-Jul-2020 96.45 Change in Loan Closing Date(s) Page ii of v Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) KEY RATINGS Outcome Bank Performance M&E Quality Satisfactory Satisfactory Substantial RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 05-Oct-2011 Satisfactory Satisfactory 0 02 22-Apr-2012 Satisfactory Satisfactory 1.00 03 02-Nov-2012 Satisfactory Satisfactory 1.21 04 10-Jun-2013 Satisfactory Satisfactory 1.76 05 13-Dec-2013 Satisfactory Satisfactory 12.67 06 03-Jul-2014 Satisfactory Satisfactory 16.22 07 03-Dec-2014 Satisfactory Satisfactory 20.31 08 26-May-2015 Satisfactory Satisfactory 23.71 09 26-Aug-2015 Satisfactory Satisfactory 28.51 10 09-Mar-2016 Satisfactory Satisfactory 31.99 11 21-Apr-2016 Satisfactory Satisfactory 31.99 12 04-Nov-2016 Satisfactory Satisfactory 33.62 13 23-Jan-2017 Satisfactory Satisfactory 36.05 14 05-Jun-2017 Satisfactory Satisfactory 38.57 15 27-Dec-2017 Satisfactory Satisfactory 44.52 16 24-Jun-2018 Satisfactory Satisfactory 55.35 17 18-Dec-2018 Satisfactory Satisfactory 72.17 18 10-Apr-2019 Satisfactory Satisfactory 74.39 19 20-Oct-2019 Satisfactory Satisfactory 87.83 Page iii of v Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) 20 10-Mar-2020 Satisfactory Satisfactory 95.14 21 30-Sep-2020 Satisfactory Satisfactory 96.45 22 22-Nov-2020 Satisfactory Satisfactory 96.45 SECTORS AND THEMES Sectors Major Sector/Sector (%) Agriculture, Fishing and Forestry 13 Public Administration - Agriculture, Fishing & Forestry 13 Public Administration 6 Sub-National Government 6 Social Protection 81 Social Protection 81 Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Social Development and Protection 0 Social Inclusion 50 Indigenous People and Ethnic Minorities 50 Urban and Rural Development 0 Rural Development 50 Rural Infrastructure and service delivery 50 Environment and Natural Resource Management 0 Climate change 75 Adaptation 75 Page iv of v Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ADM STAFF Role At Approval At ICR Regional Vice President: Pamela Cox Carlos Felipe Jaramillo Country Director: Carlos Felipe Jaramillo Marianne Fay Director: Laura Tuck Anna Wellenstein Practice Manager: Ethel Sennhauser Preeti S. Ahuja Julia Isabel Navarro Espinal, Task Team Leader(s): Renato Nardello Garry Charlier ICR Contributing Author: Julia Isabel Navarro Espinal Page v of v Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL 1. At the time of appraisal in 2011, the economy of Bolivia had experienced annual average growth of 4.5% since 2005, increasing per capita income by 18%. These changes were driven by high commodity prices and prudent fiscal and monetary policies. Current account surpluses had prevailed since 2003, and the fiscal balance had turned positive in 2006. Gross public debt had dropped from 96% of gross domestic product (GDP) in 2003 to 40% in 2010, while international reserves had increased ten-fold over the same period. 2. This strong economic performance was accompanied by widespread progress in reducing poverty and promoting inclusion, yet important gaps remained. Bolivia was one of the poorest countries in Latin America (LA), and had high levels of income inequality. With a Gini index1 of 46.1 in 2011, the richest 10% of the population held 33.7% of the total income, while the poorest 10% held only 1.1%. (World Development Indicators). 3. Poverty was still deeply entrenched in rural Bolivia, where 33% of the population lived. Around 52.7% of the rural population lived below the poverty line in 2011 ($5.5/day, 2011 PPP), compared to 15.8% in urban areas. Nearly one out of five lived in extreme poverty (19.3% using the international poverty line of $1.9/day, 2011 PPP, compared to 1.3% in urban areas). Moreover, the Gini index for rural areas was 0.53, compared to 0.41 within urban areas.2 Large numbers of rural (mostly indigenous) Bolivians lived in remote areas without access to safe water, electricity, road connections, and basic services such as education and healthcare. Indigenous rural communities were disproportionately affected by the uneven distribution and poor quality of infrastructure and basic services. 4. In 2009, the Government of the Plurinational State of Bolivia (GoB) enacted a new Constitution that sought to reverse the longstanding anti-rural bias by decentralizing the administration.3 The decentralization of administrative services required stronger capacity to establish priorities and to design, coordinate, and implement programs and public services to address the unequal distribution of public revenues across departments and municipalities. Even with the influx of resources to rural areas starting in 1994, poverty reduction remained modest. Most resources were allocated by municipalities to communities each year on a per capita basis, which left little scope for community infrastructure investments. Although local planning, decision-making, and investments benefited from strong participation by civil society, these local processes generally did not incorporate poverty targets or emphasize vulnerability reduction, the priorities of small-scale rural producers, and natural resource management. 5. The government was strongly committed to continue with decentralization, eradicate extreme poverty in rural areas, and promote the inclusion of marginalized groups. For example, the government reiterated this commitment in its Bolivia Digna National Development Plan (PND) for 2006–10 and 2010–15,4 which included the goal of strengthening the participation of communities in identifying, prioritizing, implementing, and monitoring integrated community development projects focused on the most vulnerable municipalities. In this context, the 1 World Development Indicators 2 World Bank LAC Equity Lab: https://www.worldbank.org/en/topic/poverty/lac-equity-lab1 3 The government introduced several reforms over the course of two decades, including the Popular Participation Law (PPL) in 1994, the Administrative Decentralization Law in 1995, the National Dialogue Law (NDL) in 2000, and the Hydrocarbons Law in 2005. 4 Bolivia Digna, Soberana, Productiva y Democrática para el Vivir Bien. The strategic aim of the Bolivia Digna PND was to reduce poverty and inequality and increase social inclusion to improve the well-being of all Bolivians. Page 1 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) GoB explicitly identified Community Driven Development (CDD)5 as an appropriate approach for implementing Bolivia Digna. 6. Bolivia became a major focus of World Bank support in LA and a model for macro-economic and decentralization policies during the 1990s when these reforms were underway. The Bank’s assistance to the reform process was embedded in a series of investments, structural adjustment operations, and support for preparing the Poverty Reduction Strategy, the principles of which were included in the National Dialogue Law of 2000. The Bank also maintained a strong dialogue informing the preparation of the National Strategy for Agriculture and Rural Development and supported rural development efforts mainly through decentralization and popular participation reforms, which transferred functions and resources to rural areas. 7. Bank-financed investments and Advisory Services and Analytics (ASA) supported decentralization and rural development into the second decade of the 2000s. The successful Rural Communities Development Project (P040085), follow-up Participatory Rural Investment Project (P1012998), Indigenous Peoples Development Project (P057416), and Rural Productivity Study established a solid rationale for the Bank to continue its support for rural development and decentralization into the 2010s. At appraisal for PICAR in 2011, the active portfolio supporting rural areas encompassed the Rural Alliances Project (PAR, P111863), Second Participatory Rural Investment Project (PDCR, P1012998), Land for Agricultural Development Project (Pro Tierra, P087925), and Innovation and Agricultural Services Project (PISA, P106700). In 2010 the Bank conducted an agriculture public expenditure review,6 which concluded that spending in agriculture alone had not led to reduced food insecurity and that a multi-sectoral and integrated approach was needed. 8. The Community Investment in Rural Areas Project (PICAR) was intended to build on and complement the active portfolio supporting rural development in Bolivia. A key design feature of PICAR was that it targeted some of the poorest, most vulnerable, and marginalized rural communities7 not reached by other projects. PICAR would address the needs of those communities (rather than producer organizations or municipalities) at a much more granular implementation scale, while working within the context of the broader development activities pursued by municipalities. Most importantly, the project was designed to be the first operation in Bolivia to operate by directly transferring investment funds to rural communities for them to manage. Theory of Change (Results Chain) 9. The Theory of Change (ToC) for PICAR was created retroactively from the Project Appraisal Document (PAD). Figure 1 illustrates how the provision of technical assistance (TA), training, capacity building, and investment grants to target communities would support their management of the entire subproject cycle and improve their access to sustainable infrastructure and services. Beyond those immediate achievements, the project’s CDD approach—which involved the target communities in participatory, gender-sensitive community 5 CDD operates on the principles of transparency, participation, accountability, and enhanced local capacity. The approach is usually based on partnering with communities and local units of government, including allowing organized communities to decide their priority investment needs and putting resources under their direct control, leading to more efficient delivery of basic services, and, when sustained over time, measurable reductions in poverty, particularly among the poorest populations and communities. 6 World Bank (2011), “Plurinational State of Bolivia: Agriculture Public Expenditure Review.” Report No. 59696-BO. Washington, DC. 7 In Bolivia, a “community” is a traditional form of territorial organization with recognized legal status and specific rights and capacities defined by law, including the capacity to manage public funds and prioritize public investments. Page 2 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) planning—would strengthen participating communities’ ownership of their development activities and empower8 them to continue acting as agents of their own development. 8“Empowerment” is the expansion of assets and capabilities of poor people to participate in, negotiate with, and hold accountable institutions that affect their lives. Empowering people means giving them greater access to voice and information, greater social inclusion and participation, greater accountability, and organizational strength. CDD aims to harness social capital through empowerment and increase social capital through scaling up. Page 3 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Figure 1. Theory of Change, PICAR Project Development Objectives (PDOs) 10. The objective of the Project, as stated in the Financing Agreement, is “to improve access to sustainable basic infrastructure and services for the most disadvantaged rural Communities selected in some of the poorest municipalities of the Recipient.”9 Key Expected Outcomes and Outcome Indicators 11. The expected Project outcome was accessible, sustainable infrastructure and services, properly targeted. 12. The original PDO-level indicators were: (i) Beneficiary communities show improvement in Social Capital 10 Index; (ii) Community subprojects (of which women’s) that satisfactorily achieve 75% of expected results; and (iii) Community subprojects (of which women’s) fully functional one year after their completion. These indicators were adjusted with the introduction of the Additional Financing (AF) (see “Revised PDOs and Outcome Targets”). 13. Targeted Beneficiaries: The Project sought to reach 870 communities across 131 of the poorest rural municipalities within the departments of La Paz, Oruro, Chuquisaca, Cochabamba, and Pando.11 These municipalities were selected based on the highest level of Vulnerability and Mapping12 (VAM)—a VAM of 4 or 5 at the community level—and the prevalence of extreme poverty, with more than 40% of the population in extreme poverty, as measured by the level of Unsatisfied Basic Needs (UBN)—a UBN of 4 or 5. PICAR targeted 42% of the Category A communities13 nationwide. Its direct beneficiaries were expected to include 350,000 people 9 The PDO as stated in the PAD varies slightly from the PDO in the Financing Agreement; the latter refers to the “Recipient,” and the former refers to “Bolivia.” 10 The subindicators of the Social Capital Index are: (1) His/her participation in the O&M has increased (25 points); (2) His/her priorities are better reflected in community decisions (25 points); (3) Women’s participation in community decisions has increased (25 points); (4) He/she is better informed about value and use of funds disbursed to the community (25 points). The Social Capital Index is 100 when all indicator are 75% or higher. 11 Pando was added under the AF, yet no Category A communities were identified. Consequently, Pando was replaced by the Department of Potosí as part of the geographical scope of PICAR. 12 Index developed by World Food Programme (FAO). 13 This categorization was used to select the communities reached by the AF. It was developed by the Ministry of Development Planning (MPD) in 2016 to support the formulation of comprehensive territorial development plans (Planes Territoriales de Desarrollo Integral para Vivir Bien). Data sources for this categorization included the 2012 National Population and Housing Census and the 2013 Agricultural Census. Page 4 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) living in some of the poorest rural regions, including the Bolivian highlands (Altiplano and Valles). The Project explicitly targeted women and Indigenous Peoples (IPs). Components 14. Component 1: Community Capacity Building (estimated cost US$13.5 million, of which Bank financing 100.0%; actual cost at closing was US$6.99 million, of which Bank financing 100.0%).14 Component 1 financed TA, training, and institutional strengthening to enable communities to identify, prepare, implement, operate, and maintain community investment subprojects under Component 2. Communities were involved in capacity- building activities (participatory planning and needs assessments, for example), consultations for preparing and implementing their subprojects, financial management (FM), procurement, operation and maintenance (O&M), and monitoring and evaluation (M&E). Component 1 also financed training and awareness campaigns on cross- cutting themes such as social inclusion, gender, nutrition, conflict resolution, and community management. It assisted communities formalize their status vis-à-vis other government entities, and it financed limited TA for local entities in participating municipalities to support and supervise subproject planning and implementation and ensure compliance with technical standards. 15. Component 2: Community Driven Development Investment (estimated cost US$80.10 million, of which Bank financing 91.14%; actual cost at closing was US$96.55 million, of which Bank financing 80.76%). Component 2 financed grants for participating communities to invest in development subprojects that they prepared in response to the priorities they identified under Component 1. Because of their demand-driven nature, these subprojects could include a wide range of activities, including an open menu of social and economic investments such as rural infrastructure, basic services, vulnerability reduction, food security, nutrition enhancement, and management of natural resources. 16. Communities were expected to co-finance a minimum of 10% (in kind) of the cost of their subprojects. Each beneficiary community would be able to finance one subproject, based on an indicative ceiling of US$1,000 per family. This ceiling could be increased to US$1,500 per family for a second subproject identified and implemented by women. 17. Components 1 and 2 were implemented using Community Participatory Planning (CPP), in which empowered rural communities identified, prepared, implemented, and maintained subprojects responding to their most urgent needs, based on a CDD approach. 18. Component 3: Coordination, Monitoring, and Evaluation (estimated cost US$13.5 million, of which Bank financing 100.0%; actual cost at closing was US$10.76 million, of which Bank financing 100.0%). Component 3 provided financing to strengthen the institutional capacity of the Ministry of Rural Development and Land (MDRT) and its decentralized EMPODERAR15 unit, which was responsible for coordinating, monitoring, and evaluating the Project. PICAR established a Project Coordination Unit (PCU) at the central level with regional branches responsible for administration, FM, procurement, and social and environmental safeguards. Component 3 also financed the incremental operational cost of EMPODERAR and operational costs of the PCU, including salaries and travel costs for project staff (excluding civil servants); establishment and operation of the M&E system; technical and financial audits; O&M of project offices, including utilities and telecommunication; and acquisition, operation, and maintenance of office and field equipment, including vehicles for project activities. 14 See Annex 3 for details on original costs estimates before and after AF. 15 Emprendimientos Organizados para el Desarrollo Rural Autogestionario, dedicated to fostering decentralized, locally managed rural development. Page 5 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) 19. Reasons for cost deviations: Overall cost at closing was US$114.3 million, 6.73% above the cost estimates at appraisal and for the AF. The US dollar cost differences arose from several sources. First, variations in the Special Drawing Rights (SDR) exchange rate made additional resources available (approximately US$1,092,000). Second, under the AF Financing Agreement the Credit proceeds were reallocated between categories. Specifically, SDR5.55 million16 of operating cost savings (43% of Component 1 financing and 13% of the total AF) was transferred from Category (1) to Category (2) to finance the consolidation and completion of prioritized community subprojects. Finally, counterpart funding from beneficiaries exceeded appraisal estimates by almost US$11.5 million, and the Project also had unused Credit proceeds in the amount of US$716,633.77. B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) Revised PDOs and Outcome Targets 20. The PDO was not revised. Revised PDO Indicators 21. PDO Indicators were adjusted under the AF. Improvement in the Social Capital Index, comprising four subindicators (O&M, participation in subproject identification, women's participation, and social accountability) was removed as a PDO Indicator and converted into three standalone Intermediate Results Indicators, as summarized in Table 1. An indicator on the number of direct project beneficiaries, disaggregated by gender, indigenous groups, and female households, was added. The indicator on community subprojects (of which women) that achieve at least 75% of their expected results was replaced by improved access to basic infrastructure and services in targeted communities (measured by additional months of road access per year, new and improved hectares irrigated, additional safe drinking water available, and kilometers of fences constructed). The remaining PDO Indicators were unchanged to capture the sustainability dimensions of subprojects financed by the Project. Revised Components 22. The components were not revised. Other Changes 23. The AF introduced a number of other adjustments to the Project. Specifically, it: (i) financed about 1,100 additional community subprojects in existing and new communities;17 (ii) expanded the Project’s geographical coverage by adding Pando Department to the original four departments; (iii) revised the Results Framework (RF) to account for the corresponding new targets; (iv) triggered the safeguard policy on Projects in International Waterways (OP/BP 7.50)18 as a precautionary measure, because all major rivers in Pando Department are tributaries of international rivers within the Amazon Basin; and (v) updated the legal covenants, disbursement estimates, Project costs by component, and implementation schedule. The rationale for these changes and their implications for the ToC are discussed at the end of this section. 16 Approximately US$7.8 million. 17 Component 1 included about 750 new communities in five departments through consultations, subproject identification and implementation, while also financing - for 120 communities (from the original financing) - a second round of subprojects. Component 2 included the financing of approximately 1,100 additional subprojects to improve basic infrastructure and services in the selected communities. Component 3 included financing to set up an additional regional operation unit in Pando, and for operating costs in the central and two other regional units, including M&E as well as third party evaluation. 18 An exception to riparian notification was approved by the Regional Vice President on May 22, 2015, because: (a) drinking water and irrigation sector micro- projects would be of limited local scale and focused on the rehabilitation of existing drinking water systems and rehabilitation and cleaning of village canals for irrigation; and, (b) rehabilitation and improvement of small wooden docks and fishing facilities would be limited in scale and focused around enhancement of the existing facilities. Page 6 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Table 1: Changes in the Results Framework (Intermediate Results Indicators) under the PICAR Additional Financing Indicator Change Component 1: Capacity of beneficiary rural communities strengthened in needs assessments, prioritization, subproject preparation and implementation, and management of community organizations Number of training events implemented Deleted Share of community members (of which women) having Revised: Number of beneficiaries (of which women) trained in project attended at least one training management Share of families in targeted communities that are satisfied Revised: Percentage of community members (of which women) with the prioritized investments satisfied with the planning processes and subproject identification † Added: Number of beneficiaries engaged in subproject implementation committees (of which women) Component 2: Beneficiary families in targeted communities implement and maintain their priority social and economic investments Beneficiaries (of which women) that have benefited from at Replaced by mandatory beneficiary indicator introduced at the PDO least one subproject level Community (of which women’s) subprojects financed by Revised: Percent of subprojects satisfactorily completed project end implementation, including final accounting † Added: Percentage of community members satisfied with subproject outputs (infrastructure/service) (of which female) Share of technical and fiduciary audits undertaken each year Revised: Percentage of completed subprojects whose technical and that are unqualified (over total audits undertaken in same fiduciary audits have no substantial remarks † year), including women’s subprojects Source: Adapted from the AF Project Paper. † Derived from the Social Capital Index PDO Indicator. 24. Implementation of the AF was delayed by 16 months due to factors beyond the control of the implementing entity. The AF Credits became effective in February 2016 instead of December 2015, and the first disbursement of the US$2.0 million advance to the Designated Account did not take place until April 2017. The initial disbursement was delayed because it took longer than expected to conclude the Subsidiary Agreement required by GoB between the Ministry of Economy and Public Finance (MEFP), Ministry of Development Planning (MPD), and MDRT. The expanded geographic coverage of the Project had to be delineated in accordance with new national requirements and strategic approaches, including the new categorization of communities based on vulnerability and poverty criteria. 25. The AF was restructured three times (Level 2) for a cumulative 13-month extension of the original closing date of November 30, 2019. The extensions were granted to allow time to complete the final group of subprojects, which had been delayed by a series of events. • Restructuring #1 (June 2019). The very short implementation period of the AF left insufficient time to provide adequate technical guidance for the initial O&M of subproject investments. Community O&M was vital for sustaining local investments after the project ended but had to be timed in accordance with the agricultural calendar rather than the AF implementation period. The restructuring extended the closing date by five months and made it possible to strengthen TA and other support for community beneficiaries to ensure post-project sustainability of investments. An unexpected advantage of the extension and reallocation of proceeds was that the Project could widen its response to the expectations and demands generated by CPP and reach an additional 45 eligible communities. • Restructuring #2 (April 2020). The contested general elections and resulting political turbulence in late 2019 required the closing date to be extended by three months. The transition to a new administration delayed the transfer of Project financial resources from MEFP to beneficiary communities, slowing Page 7 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) implementation of the final group of 131 subprojects and putting their completion at risk. The extended closing date would make it possible to complete those subprojects; it also enabled the Project to support 87 additional communities (2,721 families) to fully complete subprojects. • Restructuring #3 (July 2020). Mobility restrictions and health risks arising from the COVID-19 pandemic restricted the delivery of supplies and provision of TA from adhering to the planned implementation schedule, requiring a third and final five-month extension of the Project closing date. This extension allowed the Project to implement national guidelines for protective equipment and other COVID safety measures to provide close supervision and TA. Despite these measures, some EMPODERAR staff (and their families) were directly affected by the pandemic. 26. Rationale for changes under the AF and implication for the original Theory of Change. The rationale for the AF was to scale up Project activities following satisfactory performance under the terms of the Original Credit. Under the AF, Pando Department was selected to expand the reach of the Project, based on the department’s high vulnerability to food insecurity and large indigenous population (determined by VAM and UNB measures) and the Project’s emphasis on reaching the most disadvantaged rural communities. 27. The changes to the PDO Indicators and Intermediate Results Indicators were justified by the difficulties that emerged in measuring the Social Capital Index subindicators and by the need to better capture the benefits of the heterogeneous array of subprojects financed by PICAR. As noted, one PDO Indicator, Beneficiary communities show improvement in Social Capital Index, was disaggregated into three individual Intermediate Results Indicators. A new PDO Indicator, Improved access to basic infrastructure and services in targeted communities, was added to improve measurability and focus specifically on two important project outputs (infrastructure and services) to which beneficiaries would gain “access.” The focus on capacity building and beneficiary engagement was similarly intensified through revisions to the PAD Intermediate Results Indicators and the addition of new ones for Component 1 under the AF. The new and revised Intermediate Results Indicators under Component 2 strengthened the emphasis on subproject completion, fiduciary soundness, and beneficiary satisfaction. The ToC was not affected by any of these changes, or by the extension of the closing date and related restructurings. II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating 28. Rating: HIGH. The PDO remained highly relevant to priorities identified in key Bolivian national plans and the World Bank Country Partnership Framework (CPF) FY16–2019 for Bolivia. The programs outlined in the government’s Agenda Patriótica 2025 and five-year Economic and Social Development Plan (Plan de Desarrollo Económico y Social 2016–20, which followed on the PND) seek to maintain a stable path toward continued growth and poverty reduction. The CPF was closely aligned with the aspirations of the Plurinational State of Bolivia to sustain progress in reducing poverty, and the PDO for PICAR specifically emphasizes progress on this path for poor and disadvantaged rural communities. 29. At appraisal, the PDO reflected the World Bank Group Interim Strategy Note for the Plurinational State of Bolivia for FY10–11,20 which was strongly oriented toward the GoB objective of eradicating extreme poverty 19 Report No. 100985-BO. 20 Report No. 48372-BO, discussed by the Executive Directors on June 2, 2009. Page 8 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) by focusing on good governance, job creation, improved public service provision, and sustainable development. 30. At closing, the PDO was aligned with CPF Pillar I, which supports Bolivia in preserving the gains in poverty reduction and shared prosperity, ensuring that broad-based and inclusive growth will continue. Encompassing a broader approach to poverty, which is viewed not just as a lack of income, education, or health facilities but as a lack of voice, empowerment, good governance, and security against shocks, the local empowerment envisioned in the PICAR PDO is a source of poverty reduction that is quite independent of income effects. PICAR provides an example of how projects can empower communities to promote decentralization through the acquisition of new skills based on capacity building and learning by doing. 31. The PDO also dovetails with CPF Pillar 2, which supports increased access to quality basic services for the poorest rural and urban communities. Through the participatory approach embedded in PICAR, some of the most disadvantaged rural communities in Bolivia succeeded in improving their access to electricity, water and sanitation, tertiary roads, and basic productive infrastructure, and their active involvement in the entire subproject cycle developed a sense of ownership of these improvements. 32. The Project’s objective, design, and methodology continue to be highly relevant to the current national context and the priorities outlined in the PND.21 These priorities include eradicating extreme poverty, providing universal access to basic services, and maintaining transparency in public management, in a national context oriented toward the complementary integration of sovereign indigenous peoples. PICAR also supports specific goals of the PND with respect to universal access to inputs, TA, and other services supporting agricultural production. Implementation of the AF centered on reaching Category A communities—communities defined by MPD as having the least access to basic services and lowest gross agricultural production value. 33. In line with its geographical focus on areas where indigenous people were extremely poor, at great risk of food insecurity, and in great need of basic infrastructure and services, PICAR was the first World Bank operation in Bolivia to work successfully with indigenous governments. All of these aspects of the Project remain highly relevant to Bolivia’s social and economic agenda. The country’s indigenous population is among the largest in the world, encompassing a tapestry of different groups with distinct historical, cultural, and economic features and a strong influence on policy decisions. The Project’s work with three Autonomous Indigenous Native Peasant Governments22 is significant, along with the fact that the overwhelming majority— 92%—of communities reached by the Project were indigenous. B. ACHIEVEMENT OF PDOs (EFFICACY) 34. The Project was innovative and ambitious. PICAR was the first operation in Bolivia to transfer funds directly to rural communities for them to invest in development subprojects under their own management. It implemented the CDD approach in more than 40% of the country’s poorest rural communities, which were historically marginalized by their geographical remoteness or otherwise excluded from top-down development approaches. The Project supported a wide array of basic infrastructure and services that these communities identified as priorities through the CPP, including the improvement of tertiary roads, access to electricity and potable water, micro-irrigation works, and other productive infrastructure. Because the CDD approach was successfully embedded in the local culture, traditions, and customs, PICAR also strengthened rural communities’ capacity for participatory decision-making and self-management, leading to improved social capital and long-term 21 Plan de Desarrollo Económico y Social 2016–20. 22 Gobiernos Autónomos Indígena Originario Campesinos. Page 9 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) sustainability of their investments. The Project’s implicit ToC was sound and based on rational operational and technical assumptions. 35. Discussion of results. The discussion that follows disaggregates the PDO by theme and presents evidence of the Project’s achievements using the PDO outcome indicators and other supporting information.23 PDO Outcome – Theme 1: Improved access to basic infrastructure and services24 for the most disadvantaged rural communities 36. Outcome Indicator 1: Direct project beneficiaries (target 350,000 beneficiaries). The Project improved access to basic infrastructure and services for about 42% of the poorest and most isolated communities in Bolivia, reaching 362,619 beneficiaries through 2,197 subprojects. These beneficiaries include all family members benefitting directly from all types of subprojects implemented in the participating communities (Table 2). Note that this number reflects some duplication of benefits, because (for example) communities could benefit from communal subprojects as well as women-led subprojects. A disaggregation of beneficiaries without duplication yields a total of 212,270 people reached by PICAR, equivalent to 46,585 families. Based on the overall number of beneficiaries (362,619), the achievement of the target for this indicator is 103.6%, given that the target is defined as the raw beneficiary number in the original RF and direct project beneficiaries under the AF. Table 2: Beneficiaries by type of subproject, PICAR Subproject type Subprojects (number) Beneficiaries (number) Irrigation 488 73,638 Road improvement 129 29,866 Agricultural production 1,303 206,970 Water and sanitation 199 40,500 Rural electrification with solar panels 29 3,709 Other† 49 7,936 Total 2,197 362,619 Source: EMPODERAR GMIS. † Ecological kitchens, tourist hostels, sewing and weaving facilities, and improved school and health post infrastructure. 37. Breakdown of Outcome Indicator 1: Female beneficiaries (target 45%). This outcome was exceeded by nearly 5%, as 49.7% of project beneficiaries are women. The gender strategy adopted by PICAR and included in the Project Operational Manual (POM) featured specific measures to: (i) increase the active participation of 23 This account of PICAR results draws on many sources. The Borrower Completion Report (2020) includes the final evaluation of PICAR, prepared by an independent third party based on extensive interviews with Project institutions, beneficiaries, and other stakeholders, as well field visits to 79 randomly selected subprojects, among other sources of information. The final evaluation used a participatory and integrated methodology that included Appreciative Inquiry, Theory of Change, and Most Significant Changes. Results were based on information triangulation with qualitative tools (secondary information, in- depth interviews, and testimonials) and quantitative tools (semi-structured interviews, beneficiary surveys, and reports issued by the EMPODERAR GMIS). Data are also available from three iterations of a Beneficiary Satisfaction Survey (2017, 2019, and 2020) conducted throughout implementation to inform the RF. The last round, which formed part of the final evaluation, was based on a random sample of 539 beneficiaries from all departments and covered all subproject types. The Gender and Social Final Evaluation (2020) involved the participation of nearly 500 beneficiaries and key informants and captured rich qualitative information through six different instruments and case studies. Additional sources of information included the Mid-Term Review Report (2018), project databases in the Geo-referenced Management Information System (GMIS) hosted by EMPODERAR, Aide Memoires, and insights gleaned from PCU staff, World Bank specialists, and field experiences. Annex 7 reports the findings of key studies. 24 The basic infrastructure and services supported through PICAR are not considered separate categories of investment; instead they are regarded as forming an investment continuum, in which the introduction of a unit of basic infrastructure leads to the introduction of a service derived from it. Examples include electrification, sanitation, and tertiary road infrastructure. Page 10 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) women throughout the entire subproject cycle, from identification to O&M; (ii) address the specific capacity- building needs of women through special training and workshops; and (iii) offer communities the possibility of applying for a second subproject on the condition that it was prioritized and implemented primarily by women. As a result, PICAR financed 973 subprojects managed exclusively by women, equivalent to almost 45% of investments supported by the Project. (See Section II.E for more details.) 38. Outcome Indicator No 3: Improved access to basic infrastructure and services in targeted communities, as measured by the following subindicators: • Additional road access per year (target: 3 months per year). The improvement of tertiary roads has given 129 communities three additional months of access each year to this basic infrastructure, representing 100% achievement of the target. Of the 129 road investment subprojects, 70% consisted of investments to improve roads, 18% were vehicular bridges, 7.5% were pedestrian bridges, 4% were speed bumps, and 3% were retaining walls. • New and improved irrigated area (target: 1,400 hectares). The 488 community irrigation subprojects developed new or improved irrigation facilities on 2,611 hectares, representing 187% achievement of the target. This significant accomplishment partly reflects impressive community contributions (which were six times larger than expected) and partly reflects conservative target setting. Some 39% of these subprojects included the improvement and extension of irrigation systems as well as water-harvesting systems for agriculture; 10% included the construction of water reservoirs, and 9% were water- harvesting systems for animals. • Fences constructed (target: 480 kilometers). Access to this basic infrastructure was improved through 664 subprojects that built 2,809 kilometers of fences, representing 585% achievement of the target. Again, this accomplishment reflects larger than expected community contributions and the conservative targets set by the Project. Fencing subprojects include animal pens (about 54%) and fences to protect vegetable plots (27%); 10% included barbed wire to defend animals from predators. • Safe drinking water available for targeted communities (cubic meters of additional potable water per day) (target: 400 cubic meters per day). PICAR financed 155 sanitation subprojects that provided 3,375 liters of additional potable water per day per community on average. This supply is equivalent to 523 cubic meters of additional potable water per day, representing 130% achievement of the target. 39. Additional complementary results for Theme 1 include: • Intermediate Results Indicator: Percent of subprojects which satisfactorily completed implementation, including final accounting (target 80%). The Project required a final accounting of each subproject for the subproject to be considered officially completed. Based on data from the EMPODERAR Geo-referenced Management Information System (GMIS) at project closure, all 2,197 subprojects financed by PICAR completed their implementation satisfactorily, including the final accounting, representing 125% achievement of the target. • Intermediate Results Indicator: Percentage of community members satisfied with subproject outputs (infrastructure/service) (target 75%). Some 94% of beneficiaries were satisfied25 with subproject 25The degree of satisfaction was 4.2 on a 5-point Likert scale, measured as part of the Beneficiary Satisfaction Survey completed in 2020. Other results related to beneficiaries’ satisfaction with project activities included: activities with the Administration Committee (4.1 points); activities with the Social Control Committee (4.0 points); accountability activities conducted by the Social Control Committee (4.2 points); O&M activities conducted by the community (4.0 points); criteria used by the community to identify subprojects (4.2 points); benefits obtained from functioning subprojects (4.3 points); and community needs have been resolved thanks to the Project (4.4 points). Page 11 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) outputs, qualifying subproject outputs as either “very good “or “good,” representing 125% achievement of the target. 40. Breakdown of Intermediate Results Indicators: Percentage of women satisfied with subproject outputs (infrastructure/service) (target 75%). Of the women surveyed, 96% were satisfied with subproject outputs, representing 128% achievement of the target. PDO Outcome – Theme 2: Sustainability of basic infrastructure and services 41. The sustainability of basic community infrastructure and services financed by PICAR begins with efforts to build beneficiaries’ ownership of subprojects through active involvement and in-kind contributions. The active involvement of beneficiaries throughout the subproject cycle creates a strong incentive for them to maintain their investments by using the O&M training provided by the Project. Requiring beneficiaries to contribute at least 10% of the cost of their subprojects in kind, including labor and raw materials,26 strengthens beneficiaries’ ownership and commitment to sustain infrastructure and services. 42. The final evaluation confirmed that beneficiary communities and families achieved a high degree of ownership over their subproject investments and the resulting infrastructure, equipment, and services they accessed. When beneficiaries completed their subprojects, the works and equipment financed by the Project were formally transferred to them under the terms of an agreement signed by EMPODERAR and the communities. The strong sense of ownership exhibited by the communities was less a product of this formal agreement than of their direct involvement in selecting and implementing subprojects, however. By undertaking their subprojects, communities implicitly committed to maintaining their investments and making good use of them. By participating in building their infrastructure, beneficiaries developed a full understanding of its construction and operation. In turn, this involvement served as a space for training men and women in the proper use and operation of the works and equipment they now proudly own. 43. The Project defines the “sustainability” of completed subprojects in terms of their durability and the ability of communities to maintain the infrastructure and services over time, as captured in the following PDO Indicator: Subprojects fully functional 12 months after their completion (target: 75%). Of 2,197 subprojects financed by PICAR, 1,960 had been completed for 12 months or more at Project closure. Of these, 1,956 were fully functional at the time of the final evaluation. In other words, 99.79% of subprojects that had been completed for more than 12 months were fully functional, surpassing the target of 75%. The final evaluation attributed this achievement to the strong sense of empowerment and ownership generated by the Project, which enabled good O&M over time. A strong indication of the durability of these community investments is that, with the exception of only four subprojects, all investments financed by the Original Credit remained fully operational more than five years after their completion. 44. Notably, the Project learned that some types of infrastructure and services were inherently less sustainable than others. Through the Original Credit, for instance, PICAR financed electrification investments that soon proved to be less sustainable than others, given that their O&M depended on third parties with no clear connection to the communities. As a result, the Project ceased financing such subprojects under the AF. A similar phenomenon was observed with tertiary roads, although in this case there is anecdotal evidence that communities continue to hire third parties to keep roads in good condition, particularly during the rainy season when access to 26At Project appraisal, contributions from beneficiaries were estimated at US$7.10 million. By the time the Project closed, actual counterpart contributions had reached US$18,575 million, which was more than 260% percent of the initial estimate. Page 12 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) markets becomes difficult. 45. Beyond the narrow definition of sustainability used by PICAR, other relevant dimensions of sustainability identified in the CDD literature27 are relevant to the Project. For example, social sustainability is the bedrock on which other forms of sustainability rest. The Project was able to tap into social sustainability by being socially (and gender) inclusive, building on existing local institutions and forms of government and laying the groundwork for decentralized conflict resolution. As a result, PICAR was also able to deliver asset sustainability (infrastructure and services), because communities were sufficiently empowered and organized to assume responsibility for O&M, particularly for productive infrastructure and micro-irrigation subprojects. Other dimensions of sustainability identified in the literature, including fiscal and environmental sustainability, imply an additional transfer of responsibilities to communities to reinforce long-term management of resources within a decentralized approach to development. 46. Other relevant results for Theme 2 include: • The existence or creation of productive organizations (producer associations) in some of the beneficiary communities also contributed to the sustainability of subprojects. These associations are using the organizational and management skills acquired during the CPP to continue joint work, exercise social control, and look out for opportunities to make further collective investments to attain economies of scale. • Additional TA: During the last implementation tranche of the AF, PICAR strengthened TA and support to communities to boost the sustainability of Project investments, particularly investments in productive activities such as greenhouses and animal production/husbandry. To that end, the Project extended the contracts of community planners (acompañantes) to provide post-investment TA at the community level; strengthened ties with municipal governments28 so they could support the communities after the Project closing date; carried out specialized training and workshops for beneficiaries and community planners; and prepared educational materials that could easily be disseminated among beneficiaries (such as thematic technical videos for different types of subprojects), taking advantage of new information and communication technologies (such as WhatsApp groups) that also proved effective under the mobility restrictions linked to the pandemic. PDO Outcome – Theme 3: Reaching the most disadvantaged rural communities in some of the poorest municipalities of Bolivia 47. The Project ensured that it reached the most disadvantaged rural communities, especially the poorest communities and communities marginalized through geographical remoteness and isolation, by developing an effective process and methodology for mobilizing and targeting beneficiaries based on detailed, verifiable criteria. The Project began its outreach with an intensive, widespread communication and information campaign, carefully designed to ensure transparency, familiarize all potential beneficiaries with the goals and opportunities of the CDD program, and reduce the risk of elite capture. The campaign reached target groups with different cultural, gender, and other characteristics through native language radio (Aymara and Quechua). 48. By closing, PICAR had succeeded in reaching 1,284 of the 3,058 rural communities nationwide that the government has classified as the most disadvantaged. PICAR developed an initial list of potential target 27 World Bank (2003), “Scaling-up Community Driven Development, Theoretical Underpinnings and Program Design Considerations.” World Bank Policy Research Working Paper 3039. Washington, DC. 28 PICAR signed 89 agreements with municipal governments with respect to this support. Page 13 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) communities in municipalities that met specific poverty and vulnerability criteria, based on official data, as discussed in paragraph 13. Each municipality established a Prioritization Commission to carry out social validation of the target communities. The commissions included representatives of Municipal Councils, Vigilance Committees, Provincial Executives, public and private entities, Municipal Executives, and PICAR. Their active participation was a key factor enabling the Project to reach the most vulnerable rural communities. The risk of discretionary selection of beneficiary communities was reduced by developing the initial list of communities based on official poverty and vulnerability data. PICAR ultimately operated in 116 municipalities in La Paz, Oruro, Cochabamba, Chuquisaca, and Potosí Departments,29 located in the Altiplano and Valles regions, which are considered the poorest in Bolivia. Annex 9 shows the locations of the Project communities and municipalities. 49. PICAR reached a beneficiary population that was almost exclusively indigenous. The Project surpassed its target for indigenous beneficiaries as a share of all beneficiaries (85%). As noted, 92% of project beneficiaries were IPs, representing 108% achievement of the target. In the last iteration of the Beneficiary Survey, all respondents self-identified as IPs (34.3% as Aymara and 65.8% as Quechua). 50. Aside from reaching the poorest rural communities within the country, the Project also placed great emphasis on reaching the most marginalized members within these communities. Female-headed households are considered to be the most vulnerable households in the target communities. Among Project beneficiaries, 20% were female household heads, representing 100% achievement of the target. Justification of Overall Efficacy Rating 51. Overall Efficacy is rated SUBSTANTIAL. The Project achieved its PDO, as shown by the indicators and supplementary evidence. While a few early investments were not sustainable, the vast majority of infrastructure and services have shown durability and have been adapted to the context of each community. All beneficiary communities have developed a strong sense of ownership. The Project accurately targeted the most disadvantaged communities in the poorest municipalities. C. EFFICIENCY 52. The Project’s efficiency was assessed by applying the cost-benefit methodology and cost-effectiveness analysis to a sample of 64 representative subprojects.30 (See Annex 4 for the full analysis.) The sample included five subproject types: (i) sanitation, (ii) rural roads, (iii) electrification, (iv) micro irrigation, and (v) productive infrastructure. 53. The financial internal rate of return (FIRR) and economic internal rate of return (EIRR), for a 20-year evaluation period and capital cost of 12.81% and social discount rate of 12.67%, were estimated at 15.65% and 18.76%, respectively. The financial and economic incremental net present values (NPVs) were estimated at US$11.3 million and US$24.5 million, financial and economic benefit-cost ratios at 1.17 and 1.33, financial and economic investment payback periods at 10.8 and 10.6 years, and financial and economic incremental NPVs per beneficiary family at US$129.2 and US$296.1. These values reflect both direct and indirect project costs, including the cost of TA, and show positive returns on investment. These results are in line with economic and financial returns reported by similar projects implemented in the LA region and are higher than the estimates developed 29 Pando Department was included in the regional scope of the Project under the AF, but Pando no longer had Category A communities under the new classification introduced by MPD and used by PICAR once the AF became effective. 30 Out of a total of 2,197 subprojects. This sample encompassed 2,192 beneficiary families: 59 families reached with electrification subprojects; 1,008 families who benefited from productive agricultural subprojects; 246 families reached with sanitation subprojects; 341 families with micro-irrigation subprojects; 384 families with rural road subprojects; and 154 families with other types of subprojects. Page 14 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) for the Original Credit (13%) and AF (13.9%). 54. Sensitivity analysis shows how these results are affected by changes in key parameters. Over a 15-year evaluation period, the investment would remain profitable in economic terms, with a FIRR of 17.72% and EIRR of 21.05% (excluding project management costs). The sensitivity analysis shows that the internal rate of return (IRR) is more elastic to changes in the incremental benefits of subprojects and in levels of direct investment. The FIRR would be 19.92% in a scenario where the beneficiaries’ incremental benefits increase by 20%. These results emphasize the considerable impact of income fluctuations on the rural population in Bolivia. 55. The positive overall efficiency results conceal some variation by type of subproject. For example, community and women’s subprojects have positive returns on investment in economic and financial terms. So do investments in improved road infrastructure, micro-irrigation systems, and productive infrastructure, given their direct links to improved agricultural productivity and access to markets. On the other hand, sanitation and electrification subprojects did not have positive financial and economic returns on investment, given that they are not directly linked to agricultural activities in rural areas. The profitability of all investments was positive in all departments except Oruro, where economic and financial indicators are below the cost of capital and social discount rate, and fewer micro-irrigation subprojects were financed. 56. Cost-efficiency and cost-effectiveness indicators show that the Project had positive results compared with similar projects implemented by government institutions (see Annex 4, Table 9). Costs were 18% less for micro-irrigation subprojects (incremental hectares), 31% less for sanitation subprojects (connection to the sanitation system), and 69% less for electrification subprojects (grid connection) compared to similar investments in other projects. 57. Specific features of the Project’s design contributed to its efficiency. PICAR’s design was an effective approach to lighten the budget requirements set forth in the Municipal Expenditure Law,31 which stipulates that no more than 25% of the municipal budget can be allocated for operating expenses. The project design was also efficient in terms of public investment savings, given that the operating costs were 10% of project costs.32 Similarly, since municipal governments and governorships are expected to incorporate O&M expenses into their budgets, the O&M training provided by the Project to enhance the sustainability of subprojects also represents a cost savings. If the Project had been implemented without direct community execution of subproject investments, the O&M costs would have been an estimated 50% higher than those incurred by PICAR; the FIRR would have been only 14.74% and the EIRR 17.77%. 58. Implementation efficiency: The Project overcame a 16-month delay in implementing the AF, as discussed in detail in Section II (“Key Factors During Implementation”). The resulting savings in operational costs were reallocated to benefit an additional 45 communities, equivalent to 90 subprojects. 59. PICAR also achieved implementation efficiency by establishing an effective operational interface with communities of rural and indigenous people while respecting their traditions and customs. This interface included dialogue, coordination, and complementarity with the organizational processes embedded in these communities. As a result, the coordination and direct execution of subprojects was facilitated, while beneficiaries developed leadership, empowerment, and self-management capacities for subproject execution through the Project’s CDD focus on capacity building under Components 1 and 2. The implementation efficiency arises from putting communities in charge of subproject implementation and O&M: costs are reduced as beneficiaries are 31 Ley de Gastos Municipales (December 20, 2001). 32 By project closure, operating costs were estimated at 15%, given the unused credit proceeds in the amount of US$716,633.77. Page 15 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) actively involved in providing labor and local materials for subprojects and carrying out O&M. Community contributions played a critical, outsized role in the efficiency outcome, as beneficiary contributions were six times higher than originally estimated. This level of participation is critical for reducing the large fiscal costs of such programs, even in contexts of extreme poverty. Subprojects were executed more rapidly than anticipated, knowledge was transferred, and communities increased their ability to administer and manage subprojects. Assessment of Efficiency and Rating 60. The rating for Efficiency is SUBSTANTIAL, considering that: • The ex-post cost-benefit analysis shows that the entire Project has positive returns on investment. The estimated IRRs are above the cost of capital and the social discount rate used in Bolivia. These results are higher than the economic and financial indicators calculated for the AF. • Community and women’s subprojects have a positive return on investment in economic and financial terms. In economic terms, most types of subprojects have positive returns on investment. Likewise, investments in all departments except Oruro have IRRs higher than the social discount rate. • The unit costs of the products delivered to the beneficiaries of most subprojects financed by PICAR are significantly lower than the costs of similar initiatives implemented by government institutions. • The results show the high efficiency achieved by the Project in relation to the costs incurred by each beneficiary. • Both the design and implementation approach of the Project led to implementation efficiencies. C. JUSTIFICATION OF OVERALL OUTCOME RATING 61. The overall outcome rating is SATISFACTORY. This rating reflects the Project’s: (i) HIGH rating for Relevance (there is clear evidence that the PICAR PDO is entirely relevant to and aligned with current CPF objectives); (ii) SUBSTANTIAL rating for Efficacy (PICAR fully achieved or surpassed all key targets, with important complementary achievements and only minor shortcomings in the sustainability of some types of investments among the diverse array of subprojects financed by the operation); and (iii) SUBSTANTIAL rating for Efficiency (positive economic and financial outcomes, in concert with implementation efficiencies resulting from the Project design and approach). D. OTHER OUTCOMES AND IMPACTS (IF ANY) Gender 62. PICAR is an exemplary story of rural women’s capacity to lift themselves and their families out of poverty when they gain opportunities to lead their communities and organizations.33 The Project created dedicated spaces for indigenous women to participate actively during the CPP. These spaces invited women to exercise new responsibilities and assume new roles by taking important positions on committees and in subprojects, in conditions of equality with their male counterparts. Women accounted for two-thirds of all active members of both the Social Control and Administration Committees. These committees were charged with acquiring goods for subprojects, paying suppliers, and accurately recording counterpart contributions and expenses to report back to the rest of the community. Women highly valued their participation in these committees. As noted in the Gender and Social Evaluation, through their participation women generated new 33See the World Bank blog Without empowered women, there is no future for rural areas and the related video presenting women’s accounts of the changes brought about by PICAR (available at: https://blogs.worldbank.org/latinamerica/without-empowered-women-there-no-future-rural-areas). Page 16 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) capacities to change not only their condition as women but their position in their communities and households, as they gained recognition for a new activity unrelated to their domestic work. Women now manage nearly 45% of subprojects financed by the Project. Poverty Reduction and Shared Prosperity 63. Broadly defined, poverty reduction requires processes that improve the capabilities and functioning of people along five dimensions: political, economic, social, lack of transparency, and lack of protective security.34 Seen in this light, PICAR communities, with support from local governments, were given the power and resources to choose and implement subprojects, and in dealing with the associated challenges they built capacity that was supplemented with training and TA. Evidence from the literature35 argues that intrinsic motivation is diminished by low levels of self-esteem and self-determination, which arise from lack of voice and empowerment. In this regard, PICAR succeeded in igniting the local interaction required to build capable, responsive, and responsible communities, in line with the dimensions pointed out in the literature on social capital.36 A noteworthy outcome of the Project, which has been reported by community groups during field visits by the World Bank and EMPODERAR teams but is hard to quantify, is the level of intra- and inter-community engagement with the various Project activities, and the way in which women used subprojects as a platform to bring elements of the community together. Beneficiaries surveyed for evaluation studies confirmed that PICAR has been a highly participatory Project that has improved community integration while promoting solidarity and cooperation. Participating communities built on/intensified the traditional ayni in the highlands and the minká in the valleys, an old tradition of practicing solidarity and reciprocity for the common good (te ayudo, me ayudas). Other Outcomes and Impacts 64. PICAR has confirmed the viability of the CDD approach in Bolivia by establishing, scaling up, and in some cases consolidating core CDD features, including real stakeholder participation and improved accountability, as outlined in the literature.37,38 The Project implemented CDD as an alternative approach to local development that transferred responsibility, resources, choice, and ownership to the beneficiary communities, boosted women’s engagement, and improved social and productive infrastructure at a lower cost than traditional top-down approaches. PICAR proved to be an effective means of promoting decentralized rural development that can be led successfully by communities with no prior experience. 65. The CPP implementation mechanism strengthened rural communities’ capacity for joint decision making and self-management. The CPP provided tools and training (such as learning by doing) for communities to become agents of their own development. This community-driven planning process supported participatory (and gender sensitive) decision-making to identify, prioritize, implement, and maintain subprojects; strengthened 34 Amartya Sen (1999), discussed in World Bank Group (2000), World Development Report 2000/2001: Attacking poverty. Washington, DC. http://documents.worldbank.org/curated/en/230351468332946759/World-development-report-2000-2001-attacking-poverty. 35 Ostrom (1999). 36 North (1990); Sola and Ebbe (1995); Burnside and Dollar (1997). 37 CDD is scaled up in three stages. The initiation stage entails a country setting the stage for decentralization and empowerment. The process of empowerment is initiated on three fronts: (i) enhancing real participation; (ii) targeting specific groups; and (iii) starting a dialogue with stakeholders on decentralization. The scaling-up stage requires that tools and logistics are refined and tested in a geographical area (e.g., district, province) to identify critical bottlenecks. Field-tested operational manuals, tools, and logistics are developed and adapted to local conditions to roll out a process that covers other districts/provinces. The consolidation stage includes: (i) integrating participation and decentralization; (ii) scaling up provincial programs to full national coverage; (iii) improving CDD design in light of experience; (iv) improving technical and organizational capability; and (v) expanding targeted programs to tackle issues that communities may have neglected. 38 World Bank (2003), “Scaling-up Community Driven Development, Theoretical Underpinnings and Program Design Considerations.” World Bank Policy Research Working Paper 3039. Washington, DC. Page 17 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) local capacities; and fostered transparent control and management of transferred resources and accountability.39 The creation and consolidation of Social Control and Administration Committees ensured high levels of transparency and social control throughout the subproject cycle. Through CPP, communities also became involved in contracting providers to deliver more complex works, supervising the works, and scheduling payments. III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 66. The Project’s approach and implementation mechanism were based on principles of community empowerment and self-management. Anchored in the CDD approach, the Project was designed to operate under the assumption that communities could develop their subprojects based solely on their needs, instead of having to choose from a pre-defined menu of options. That assumption would be put into practice through the CPP implementation mechanism and the structure of the Project components, in which community-directed investment planning and implementation would be accompanied by continuous capacity building and TA. 67. PICAR had a lean project design and well-defined institutional and implementation arrangements. PICAR benefited from its solid institutional and implementation arrangements at the national and regional level and well-contextualized and comprehensive POM. The POM included all procedures, rules, guidelines, and standards for implementing all aspects of the Project: the coordination team, planning, M&E, social and environmental reviews of subprojects, reporting, communication, human resources, community procurement, FM—including a model financing agreement to be entered into with beneficiary communities—and procedures for amending the POM. The Project also had a clear and realistic operational definition of the subproject cycle (Annex 8), which proved to be a useful tool to track implementation milestones at the required level of specificity, with the appropriate division of responsibilities. 68. The Project design was simple, but the direct transfer of funds to beneficiary communities (a relatively new operational approach in Bolivia) added an element of complexity and substantial FM risk. The operational arrangements, including FM arrangements, were sound. They were based on the existing agreement between EMPODERAR and the Rural Alliances Project and were sufficiently customized to address the specific needs of PICAR, including: (i) hiring key FM staff with terms of reference approved by the World Bank; (ii) tailoring specific procedures and arrangements to process disbursements to communities with adjustments to the GMIS, the Government’s integrated financial management system (SIGMA), and the Treasury Single Account. 69. The social and gender context of the Project was carefully considered to mitigate identified risks and guarantee appropriate targeting and inclusion of the most vulnerable populations, particularly women and IPs. EMPODERAR conducted a comprehensive Social Assessment, with extensive stakeholder consultations in rural areas. This assessment made it possible to identify: (i) the main social risks; (ii) mechanisms to target the poorest and most vulnerable communities; (iii) measures to enhance community ownership and empowerment; and (iv) important challenges associated with the participation of indigenous women, which were addressed in the gender strategy included in the POM. 70. The Project’s readiness for Implementation was noteworthy. All necessary implementation provisions and arrangements were in place, based on a comprehensive analysis and deep understanding of the Bolivian context and the proven operational and technical capacity of EMPODERAR, which was a key factor in the 39These benefits are in line with typical results reported from other, similar projects in Latin America, including almost two decades of Bank-supported CDD operations in northeastern Brazil. Page 18 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) achievement of Project objectives. B. KEY FACTORS DURING IMPLEMENTATION Factors Subject to Government Control and/or Implementing Entities 71. EMPODERAR’s notably effective coordination and engagement were key to delivering results on the ground and validating and scaling up PICAR’s CDD approach. EMPODERAR’s significant capacity in project implementation (accumulated through previous and concurrent projects) and the presence of all necessary human resources at the national and regional level were key factors in the successful implementation of the Original Credit and AF. The Original Credit (approved on July 21, 2011 and effective on November 23, 2011) was successfully implemented over a six-year period, closing on schedule on April 30, 2017. The AF successfully overcame two challenges that threatened implementation. First, start-up was delayed by 16 months. The Subsidiary Agreement required by GoB between MEFP, MPD, and MDRT took longer than expected to put into place, because shortly after the AF was approved, GoB revised the official methodology for classifying communities, which the Project used for targeting beneficiary communities. To ensure that the Project approach would be consistent with the new methodology, EMPODERAR conducted a comparative analysis and adjusted the criteria in the POM. The team responded to the delay in implementation with the first restructuring, which enabled a five-month extension of the Project closing date in response to the GoB request. The second challenge was the turbulent political transition in late 2019, which significantly delayed the transfer of funds from the MEFP to beneficiary communities. The second restructuring, entailing a three-month extension, compensated for the delays. Section I.B provides details on these restructurings. 72. The CDD approach presented challenges as participating communities gained experience with government procedures to implement their subprojects, but delays were overcome with dedicated support. Initial delays stemmed from the difficulty communities encountered in meeting specific administrative requirements (updating/obtaining their legal identity and tax identification number, opening a bank account), but these obstacles were gradually overcome with dedicated support from regional Project staff. 73. Commitment and leadership from the National Coordinator were also key to delivering results. A noteworthy feature of PICAR is that the initiative came from a national counterpart with a sound understanding of rural and institutional conditions in Bolivia. The same coordinator and team worked in close collaboration with the World Bank team to design the Project to reach isolated communities bypassed by other initiatives, and these personnel remained throughout project implementation, except for a hiatus in 2019–20, discussed later. 74. Human resources and organizational capacity were anchored in a merit-based, transparent, and independent recruiting and evaluation processes. In line with the parameters established in the Financing Agreement and POM, transparent, explicit, and rigorous technical criteria were applied in hiring and contracting personnel,40 and similarly rigorous metrics were applied to assess staff performance. Both hiring and evaluation were conducted by third-party independent entities, because at the national and regional levels this approach was proven to reduce political interference and encourage the retention of Project staff who performed well. 75. The formal provisions in the Credit Agreement to hire and evaluate staff according to a competitive, merit-based, and independent process were met, except during the political transition that took place in late 2019. As a result, the national coordinator and other key staff were removed and replaced, and certain aspects of project implementation, including the final evaluation, faced significant delays. After the 2020 elections, the 40 Project staff were hired under results-based contracts, and staff performance was evaluated annually. Page 19 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) reinstatement of the national coordinator allowed the Project to complete its final activities. A third and final restructuring, requested by GoB and rapidly advanced by the team, provided additional time for completion. 76. PICAR capably maintained accountability through a clear division of roles and responsibilities as contemplated in the Financing Agreement and POM. The POM defined the general activities of PICAR and established clear processes and procedures for the Project’s operational, administrative, and strategic management. The Project identified spaces and mechanisms for managing the social and environmental dimensions of its operations and institutional relationships at different levels. The roles and competencies of the Project operational units at the national and regional/departmental levels were clearly specified. Within this framework, each operating unit of PICAR established its own activity plan for each implementation period. These unit activity plans were not only designed to meet the unit’s general goals under the Project but detailed the mechanisms for coordinating and linking each unit’s work with that of the other operating units. 77. The definition, approval, and implementation of subprojects followed quasi-standardized procedures. These procedures were implemented with the participation of external consultants, companies, communities, and in some cases municipal governments. To compensate for delays related to the pandemic and post-electoral turmoil, the Project introduced changes in fiduciary procedures to optimize implementation of the last batch of subprojects. For example, given the delays in transferring resources to communities during 2019 –20, the Project requested Community Committees to make ex-ante calls for proposals and quotes valid for 60 days to speed the execution of subprojects once funds were disbursed to communities. Overall, PICAR had all necessary fiduciary arrangements in place, including procurement, financing, budgeting, and FM mechanisms. Factors Subject to World Bank Control 78. The World Bank team provided proactive and dedicated supervision throughout implementation, contributing to successful Project outcomes. PICAR relied on a responsive, technically, and operationally sound team from the Bank, including technical, fiduciary, and safeguard specialists located in the Country Office. Up to February 2020, full supervision missions were conducted in person twice each year, with additional technical and interim missions as needed. The Bank was responsive to new opportunities and emerging challenges, including preparation of the AF and three Level 2 restructurings of the AF. 79. Project targets were adjusted during preparation of the AF. The RF, including PDO Indicators and Intermediate-level Indicators, was adjusted to reflect changes in scale and geographical coverage, yet the level of ambition reflected in the AF targets is open to question, given that the Project significantly exceeded most targets at closing. Factors Outside of Government and/or Implementing Entity Control 80. The challenges associated with the COVID-19 pandemic coincided with political intervention that affected the composition of the EMPODERAR team. Mobility restrictions and health risks imposed by the pandemic prevented supplies and TA from being provided as scheduled, delaying implementation and requiring a third and final extension of the Project closing date by five months. Page 20 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 81. From a monitoring perspective, the M&E system was well designed and seamlessly integrated into the GMIS to track project activities, including fiduciary performance, safeguards, and technical and administrative processes. The M&E system enabled real-time, integrated financial and physical reporting that allowed informed decision-making and preserved institutional learning. The system included features to: (i) structure a database for each subproject, from identification to closure of the financing agreement with beneficiary communities; ii) use milestone logic to promote accountability and closely track and follow up on subproject planning, implementation, and procurement plans; (iii) geo-reference subproject locations to determine geographical coverage; (iv) identify the presence or lack of infrastructure in communities reached by the Project; (v) generate Project information at three tiers of analysis (community, regional, national);41 (vi) generate implementation tools for technical staff; and (vii) follow up on consulting contracts executed by communities. 82. From an evaluation perspective, the PDO was relatively unambitious, and some indicators were not aligned with the development objective. The level of ambition conveyed in the PDO was moderate. The PDO referred only to improving access to sustainable infrastructure and services rather than specifying measurable outcomes of improved access, such as empowerment, income, health benefits, and so on. Results indicators were not directly aligned with the PDO. Some targets were not sufficiently ambitious, although possibly they may have been underestimated, given the novelty of the Project’s delivery mechanism at the time of appraisal. The Social Capital Index was intended to capture results from the CPP, yet the index subindicators proved challenging to measure during implementation and were replaced during preparation of the AF, as noted in Section I.B. The design stipulated an ambitious Impact Evaluation (IE): the gradual roll-out of project interventions would be used to establish treatment and control communities and compare them using a difference-in-difference approach, with all target communities becoming treatment groups by the end of year 3. Because project implementation was demand driven, baselines were to be collected during implementation as part of the beneficiary entry profile. M&E Implementation 83. Implementation of all planned monitoring activities was outstanding, and the GMIS worked effectively. Information on all results indicators was collected consistently and was of sound quality, including entry profiles at the community level, with the caveat that the baselines were inadequate for the IE (see below). All Project information was managed through the GMIS, making it possible to monitor activities effectively and add to the voluminous repository of administrative, financial, and technical data on projects managed by EMPODERAR. As a result, a rich database of social, administrative, and financial information has been created. Both Mid-Term Reviews were informed by a Project evaluation, conducted by an independent third party with terms of reference acceptable to the World Bank. 84. Owing to technical factors (complexity, inadequate data), the IE was not implemented prior to the AF and was ultimately dropped. It became apparent during implementation that the IE would be conceptually and operationally complex. The Project’s innovative, demand-based approach introduced a high degree of uncertainty 41 First tier: Monitoring at the community/local level to track results around subproject implementation and generate operational information at the regional/departmental level. Second tier: Monitoring at the regional level to aggregate quantitative and qualitative data at the regional level for operational information required at the national level. Third tier: Monitoring at the national level to track progress toward the PDO and generate executive/managerial information. Page 21 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) that made it challenging and unrealistic to conceptualize and implement the IE. The heterogeneity of investments selected by communities would have required the Project to capture, a priori, comparable indicators defined across multiple dimensions (agricultural production, nutrition, health, transport, water use, and so on), which would have consumed time and resources. An inadequate data collection methodology disrupted efforts to establish a baseline. Open-ended questionnaires were used to collect descriptive information at the community level, whereas quantitative (and ideally household-level) data were required for the IE. The GoB was also reluctant to commit resources to a rigorous IE, given the costs and significant problems involved in completing the IE for the Rural Alliance Project.42 For these reasons, the IE was dropped when the AF was prepared. 85. PICAR undertook a series of evaluations to compensate for the IE. First, it commissioned a comprehensive third-party evaluation, which was conducted in two phases. The initial phase occurred mid-way through the implementation of the AF and involved communities covered by the Original Credit; the second phase occurred prior to closure of the AF and involved communities covered by the AF (both new and original communities).43 The evaluation used an integrated approach that relied on data collected through qualitative (secondary information, in-depth interviews, and testimonials) and quantitative tools (semi-structured interviews, beneficiary surveys, and reports issued by the EMPODERAR GMIS on progress towards the PDO, indicators, and factors related to social and governance changes). Second, the Project conducted two evaluations of performance and progress for the Mid-Term Reviews (the Original Credit and AF). Third, the Project commissioned an economic evaluation as part of the preparatory work for the AF. Ten detailed case studies formed part of this evaluation to assess the benefits derived from community subprojects. Fourth, a final evaluation focused on the gender and social dimensions of PICAR. M&E Utilization 86. The PICAR M&E system provided all the requisite information for implementation and reporting, in addition to amassing detailed geo-referenced data that will serve as a valuable resource moving forward. Data from commissioned evaluations and data routinely collected by the regional offices and PCU were used to prepare progress reports for submission to the line ministries and World Bank as stipulated in the Legal Agreements. The progress reports were also important inputs for implementation support missions. M&E data from the GMIS were used to monitor implementation progress, FM, procurement, disbursements, make course corrections, track results indicators and outcomes, update the economic and financial analysis, and support Project management in general. The Project was an early adopter of geo-referenced data for rural development M&E in Bolivia. Over its lifetime, PICAR assembled a major geo-referenced database as part of its M&E operations, including information on participating communities, beneficiary households, infrastructure, and other assets. This dataset is the most extensive and granular resource of its kind in Bolivia. Justification of Overall Rating of Quality of M&E 87. Rating: SUBSTANTIAL. Monitoring was exceptional. The M&E system was well designed to track project results in accordance with the RF and could be seamlessly integrated with the GMIS to monitor project activities across all domains of implementation. The data were adequate and used effectively for monitoring and coordination. Although the IE was dropped for practical reasons, considerable efforts were made to design and implement other types of evaluations, including an evaluation that collected and analyzed qualitative and 42 The total cost of the evaluation was three times higher than anticipated, and it required additional resources from the World Bank (such as additional funding to hire consultants) to be completed. 43 Project Paper Report No: PAD1420. Page 22 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) quantitative data at two separate stages of implementation, two Mid-Term Review evaluations, an economic evaluation with detailed case studies for the AF, and a final evaluation of gender and social dimensions. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE Environmental Safeguard Compliance 88. PICAR was classified as Category B project under OP/BP 4.01 based on the nature of the subprojects to be implemented under Component 2. The potential negative environmental impacts of these small-scale subprojects were assessed to be short-term, limited, localized, manageable, and reversible. They included soil erosion, water and soil pollution, loss of vegetation, generation of waste (plastic), pesticide poisoning, and pesticide residues in the food chain. EMPODERAR prepared an Environmental and Social Management Framework (ESMF) and Integrated Pest Management Manual. The Project completed local consultations on both instruments and publicly disclosed them in May 2015. 89. The Project triggered the following World Bank environmental safeguard policies: • Natural Habitats (OP/BP 4.04). Most subprojects were implemented in areas where agricultural production already occurred. Eighteen subprojects were implemented in buffer zones of National Protected Areas. They were simple, small, low-impact subprojects consisting mainly of horticultural gardens, perimeter fences, corrals for small livestock, metal silos and solar tents for use by individual households, sheds for camelids, household water reservoirs for livestock, and one rural community road improvement. The PCU managed the certification of compatibility of use based on the 2018 agreement with the National Service for Protected Areas (Servicio Nacional de Áreas Protegidas). • Pest Management (OP/BP 4.09). Additional actions for strengthening environmental management to avoid and/or minimize the risk of pesticide use and improve waste management were adopted, including: (i) development of national workshops for environmental specialists from the regional offices, technical assistants, and representatives of beneficiary organizations; (ii) management agreements with expert institutions44 on pesticide and toxic waste management plans; and (iii) publication and dissemination of educational material like brochures and manuals for community planners and communities. • Physical Cultural Resources (OP/BP 4.11). The Project did not finance any subproject with known potential negative impacts on physical cultural resources or sacred and/or religious sites, and the ESMF included measures to avoid works on cultural sites. • Forests (OP/BP 4.36). The Project screening mechanism had the capacity to identify subprojects with significant potential negative impacts on forests. • Safety of Dams (OP 4.37). This policy was triggered as a precautionary measure, even though the Project financed no construction or rehabilitation of any dam. Improvements and extensions of micro irrigation systems financed under PICAR covered less than 200 hectares per beneficiary community and relied on existing dams for their operation. 90. Overall, PICAR fully complied with World Bank environmental safeguard policies and the ESMF. EMPODERAR maintained the necessary organizational structure to implement the Project, with an environmental specialist in the central unit and four environmental specialists in the four regional offices of La Paz, Oruro, Cochabamba, and Chuquisaca. 44 Fundaciones Plaguicidas Bolivia and Promoción e Investigación de Productos Andinos. Page 23 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Social Safeguard Compliance 91. PICAR triggered OP/BP 4.10 (Indigenous Peoples), and its entire design reflected OP 4.10 principles, given that the vast majority of direct beneficiaries were indigenous. Consultations were held with Aymara and Quechua communities to prepare the Indigenous Peoples Plan, which included such considerations as cultural pertinence, participatory planning, and mechanisms for consultation during the subproject cycle, which were also included in the Social Management Plan. The consultations highlighted demands from indigenous communities to give equal attention to all communities, improve access to education and health infrastructure, build community capacity, and enhance women’s participation. The consultations also revealed general support for the Project and a special interest in infrastructure subprojects (roads and bridges). Community participation guidelines were included in the POM, laying out the community engagement strategy for subproject identification, prioritization, preparation, implementation, and O&M. 92. PICAR developed a gender strategy and action plan and ensured women’s participation in setting subproject priorities by organizing women-only consultations, which effectively channeled women’s priorities into subproject decision-making. Women´s specific needs, priorities, and interests identified in this way included animal husbandry (poultry, rabbits, pigs) and training to manage subprojects. Consultations ensured that the perceptions of men and women around impacts, risks, and mitigation measures were identified and later included to improve the Project design and POM. 93. No significant involuntary resettlement or land acquisition took place. The socio-economic impacts on land were limited to land donations by private parties, which under OP 4.12 (Involuntary Resettlement) did not exceed 10% of the area of any holding. The few cases of land donation were properly documented and filed for reporting and verification. No physical relocation or economic displacement was required, so no Resettlement Action Plans were prepared based on the Resettlement Policy Framework of the Project. 94. Grievance Redress Mechanism (GRM): Subproject grievance redress mechanisms (GRMs) were developed in line with the Project’s grievance redress provisions. They included physical mailboxes displayed in every departmental office and an SMS (messaging system). Following the emergence of COVID-19, text messaging was enabled to expand the GRM system. Three complaints45 were received from beneficiary communities and all were effectively resolved. Fiduciary Compliance 95. Financial Management: FM reviews conducted during regular supervision missions verified that an adequate FM system was in place to provide, with reasonable assurance, accurate and timely information that the World Bank Credit proceeds were being used for their intended purposes. The Project FM rating was consistently Satisfactory. The only exceptions occurred with Implementation Status and Results Reports (ISRs) 20 and 21, which coincided with the socio-political disturbances following the 2019 general election and the ensuing 45 (1) An anonymous complaint questioning the participation of subproject beneficiaries who were not part of the community and elevated material costs in Caravillque Community, Santiago de Huayllamarca Municipality, Oruro Department. The complaint was addressed by the Regional Coordinator, who visited the community to verify that all beneficiaries were indeed part of the community and that procurement of goods was conducted properly. (2) A complaint by Alejandra Amalia Escobar Apanqui and Blanca Escobar from Huayrapata Community, Chuma Municipality maintained that some designated beneficiaries had not received their supplies, and some had already left the community. The case was resolved after it was verified that the complainants were removed from the beneficiary list by the Community Assembly because they did not fulfill their counterpart contribution; the delivery of inputs was verified against the community procurement records. (3) A complaint by beneficiaries from Huerta Mayo Community, Potosi Department, about an irregular withdrawal of Bs125,000 by a community representative led MDRT to initiate legal proceedings. MDRT managed to recover Bs40,000, and although court proceedings continue in an effort to recover the balance of funds, the community managed to complete the subproject with its own resources, including the final accounting presented to the Regional Office. Page 24 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) transition to a new administration, which delayed the transfer of funds by MEFP to beneficiary communities. The FM reviews consistently recognized the adequacy of FM staffing, accounting, and internal control systems, maintenance of supporting Project documents, and implementation of recommendations from the annual audit. Quarterly financial reports of acceptable quality were submitted on time. Annual audited financial reports were submitted to the World Bank with unqualified opinions. The project accounting systems were observed to be in order and payments were well controlled. The independent performance audit was a good practice that provided a layer of control in addition to the checks on outputs performed by the Project and supervising consultants. 96. Procurement: Procurement activities were carried out in accordance with: (i) the provisions stipulated in the Financing Agreement; (ii) the World Bank “Guidelines: Procurement of Goods, Works, and Non-consulting Services under IBRD Loans and IDA Credits and Grants by World Bank Borrowers,” dated January 2011 and revised July 2014; and (iii) the “Guidelines: Selection and Employment of Consultants under IBRD Loans and IDA Credits and Grants by World Bank Borrowers,” dated January 2011 and revised July 2014. Procurement Guidelines and the POM were publicly available. Core procurement principles of economy and efficiency were considered in the POM. The procurement regulatory system as defined on paper was applied consistently in practice. EMPODERAR included a clear system of accountability in the POM, which defined responsibilities and delegation of authority related to procurement decisions. The procurement capacity of EMPODERAR had been confirmed prior to PICAR, and the unit was responsible for Project procurement activities and contracting processes, while also providing support to beneficiary communities. Realistic procurement plans were aligned with the Project budget, prepared on time, formally approved and followed during implementation. Turnover of procurement staff was minimal. EMPODERAR oversaw the development of technical specifications and terms of reference and participated in committees evaluating bids and proposals and managing each signed contract. The procurement plan on the Systemic Tracking of Exchanges in Procurement system (STEP) was up to date for contract execution, and only a few pieces of documentation were not yet uploaded. No cases of mis-procurement occurred, and the final procurement audit identified no issues. Procurement Post Reviews conducted during implementation found no ineligible contracts or serious issues that would indicate fraud or corruption. C. BANK PERFORMANCE Quality at Entry 97. Key elements relevant for assessing the Project’s quality at entry are: • The Project design and operational methodology benefited from two decades of World Bank support for community-driven operations throughout LA and beyond, as well as previous successful engagement with GoB through EMPODERAR. Specific lessons from experience in Bolivia and other relevant settings were incorporated into the Project design, including the need for: (i) community- driven mechanisms to improve local governance and social capital; (ii) capacity building and empowerment of communities as key contributors to the success and sustainability of investments; (iii) direct transfer of resources to communities, seen as the strongest mechanism for empowering beneficiaries and inducing their responsibility for (and ownership of) investments; and (iv) participatory processes, a proven and effective avenue for project implementation. • PDO and Results Indicators. The PDO, confined to improving access to sustainable infrastructure and services, conveyed a modest level of ambition. This type of PDO was standard for earlier CDD projects in LA, but at the time of appraisal for PICAR, the PDOs for CDD projects more commonly referred to specific improvements, such as improved well-being or quality of life of poor communities. The Page 25 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) selection of indicators did not fully capture Project objectives and outcomes. • The inclusion of an IE was unrealistic and overly ambitious. The design of the IE (see Section IV.A) sought to overcome negative perceptions of IEs in Bolivia, where a controversial randomized control trial by the Inter-American Development Bank in the water sector was regarded as unfair because communities were randomly selected to be treatment and control groups. This design did not take into account the difficulty of identifying comparable parameters for a heterogeneous array of demand- driven subprojects, however, and the potential reluctance of GoB to commit funds for the IE was not considered. • Poverty, gender, and social development. The importance of these issues was reflected in the Project design, which included specific, explicit criteria for prioritizing and selecting beneficiaries and provided a participatory process to address the needs of women and IPs. Poverty, gender, and social development were also considered in the risk assessment, and mitigation measures were put in place. • Environmental risks. Environmental risks were properly considered and assessed as part of Project preparation, and all necessary environmental safeguards were triggered. • Implementation arrangements. Implementation arrangements were exceptionally well considered, based on a deep understanding of the context and a solid operational proposal (see Section III.A). • Risk assessment. The Project team carried out all necessary preparation studies, considered the political and social context, and consulted continuously with representatives of indigenous and peasant organizations. The single level of implementation, with funds transferred directly to communities, was well suited to mitigate the risks of excessive bureaucracy and delayed implementation. Project design included important features to protect the Project against political interference, including clear and explicit criteria46 for Project staff to be hired by an independent party and for the selection of beneficiary communities to avoid elite capture. FM risk was considered Substantial, so all of the recommended risk mitigation measures were implemented (see Section III.A). The Project included measures to address limited capacity at the community level by investing in developing capacity at all levels of implementation and fostering greater inclusion of vulnerable groups within communities. Quality of Supervision 98. Key elements relevant for assessing the quality of Project supervision are: • The Bank’s focus was proactive, supportive, and results oriented. Missions were conducted twice per year,47 supervision teams possessed the appropriate skill mix, and sufficient time was allocated for field visits. Interim and technical missions were carried out as needed to provide additional training and guidance to the PCU, address impediments to implementation, and meet specific needs. Supervision was appropriately close during the early and final stages of implementation, ensuring the swift roll-out and adequate completion of activities, respectively. Turnover of Task Team Leaders (TTLs) was low; the Project had only three TTLs during its lifetime. The team was responsive to the need for restructuring, securing three extensions to the closing date to enable the final group of subprojects to be completed following delays that lay outside the control of PICAR (see Section I.B). • Supervision closely tracked Project targets in relation to results, and reports candidly assessed performance. Problems were reported and addressed promptly by the World Bank Task Team. Assistance was provided through site visits and, when necessary, by mobilizing national or international 46 Included as part of the Credit Agreement. 47 For a total of 17 regularly scheduled supervision missions with appropriately configured teams and field visits in participating departments. Page 26 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) expertise.48 Close coordination with GoB implementation teams, as well as dedicated support from an Agriculture Specialist based in La Paz, were effective in helping the Project to overcome the unanticipated delays in implementation described previously. The single lapse in the quality of supervision was the inability to conduct the IE, despite considerable effort by the Bank team to overcome its methodological flaws (see Section IV.A). The shortcomings affecting quality at entry, and the attenuating circumstances that prevented the rollout of the IE, were partially overcome by adjusting the RF at the time of the AF and by commissioning the evaluations described in Section IV.A. Results monitoring was strong overall, with the Bank Team providing extensive inputs that supported development of the comprehensive, innovative geo-referenced M&E system. • For technical support, the World Bank coordinated with the Cooperation Program of the Food and Agriculture Organization (FAO) to bring international specialists to Bolivia to provide technical support in areas such as agricultural productivity and M&E. • Supervision of fiduciary and safeguard compliance was well managed by experienced World Bank specialists stationed in-country. Procurement, FM, and environmental and social safeguards relied on experienced specialists based in Laz Paz, allowing dedicated support and close follow up. Justification of Overall Rating of Bank Performance 99. Rating: SATISFACTORY. Quality at Entry was Satisfactory. The Project benefited from a strong design and an operational methodology that was highly adapted to the Bolivian context, despite the drawbacks associated with the modest PDO, the imprecision of some results indicators, and the overly optimistic IE design. Supervision was timely and effective; supervision teams presented a strong mix of skills, and collaboration with the government teams was exceptional. Close follow-up in supervision supported government teams in reaching and exceeding targets by the end of the Project. D. RISK TO DEVELOPMENT OUTCOME 100. Although a number of potential risks could affect the development outcome, a number of strategies for mitigating these risks will help to sustain the Project’s achievements: • There is a substantial risk that the social capital and capacity of communities for local governance fostered under PICAR will fade away, if government policy moves away from CDD and decentralization. Community organization, empowerment for O&M, and integration of community governance processes into the broader bureaucracy were all strongly advanced under PICAR. Continuity in government policy on community empowerment and alleviating rural poverty is essential to keep capitalizing on these Project gains: • There is a moderate risk that during the next decade the working-age population will decline to the extent that communities can no longer maintain the infrastructure developed through PICAR. The share of the rural population aged 60 and over is increasing rapidly as youth migrate to cities to escape the difficult living conditions and lack of livelihood opportunities in rural areas. Investments financed by PICAR aimed to reduce these drivers of migration and mitigate the risk that it will accelerate. 48 Implementation delays were not reflected in project ratings in ISRs, as the overlap between the Original Credit and AF allowed for a sustained pace of implementation on the ground. PICAR was rated Satisfactory throughout its lifetime (e.g., ratings for Implementation Progress and Progress towards achievement of PDO). The delays experienced during the implementation of the AF were flagged in the ISRs, but they did not translate into a downgrading of the Project’s ratings, given that the overlap between the two Credits enabled PICAR to sustain implementation at the community level. Page 27 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) • There is moderate potential that market risk (related to prices, quality of production, and the capacity to honor contracts to deliver produce on time) will hamper beneficiaries’ ability to maintain infrastructure developed through PICAR, threatening sustainability. Although the Project cannot control market risk, small-scale farmers have developed a deeper awareness of market demands and more integrated production systems through PICAR, and they are better equipped to react and adapt to changing markets. • There is a moderate risk that climate will affect the sustainability of the infrastructure and services. Climate change scenarios for Bolivia reveal patterns of increasingly variable rainfall and a greater incidence of drought, which have important implications for production systems. Scenarios predict that crop yields could decline owing to water shortages and hot spells during critical crop growth stages. The Project worked with beneficiary communities to increase awareness of the risks associated with extreme weather events and climate change, and it provided opportunities to increase local capacity to respond to climate shocks. As a result, communities prioritized and implemented subprojects specifically aimed at improving their adaptation to climate change and management of natural resources (water reservoirs, micro-irrigation infrastructure, greenhouses for vegetable production, and improved structures and practices for agricultural and animal production). V. LESSONS AND RECOMMENDATIONS 101. The capacity accumulated at EMPODERAR was a key factor contributing to project outcomes and should be leveraged going forward. The model of a permanent independent agency tasked with project implementation proved successful due to the gradual accumulation of capacity, including the leadership of the national coordinator, the unit’s relative autonomy, technical soundness, and stable staffing over time. 102. Building a geo-referenced management information system can be useful for project management and is likely to have other valuable applications. The major geo-referenced database of communities, beneficiary households, infrastructure, and other assets assembled as part of M&E operations for PICAR has value beyond project management. Further applications of this dataset should be explored, including its use in evidence-based policy making and the design of future programs. 103. Poverty targeting mechanisms should be simple, explicit, and monitorable; based on objective criteria; foster transparency; and minimize political interference in project resource allocation. The Project placed great emphasis on the methodology and process to reach some of the poorest rural municipalities in Bolivia, thus assuring the targeted, verifiable, and effective selection of beneficiaries who had been highly marginalized owing to geographical remoteness and other factors. The initial list of prospective beneficiary communities was verified by Prioritization Commissions. These commissions, established at the municipal level, provided a forum for social and institutional actors to arrive at a consensus on which communities would participate in the Project. The risk of discretionary selection of beneficiary communities was reduced by developing the initial list of communities based on official poverty and vulnerability data. PICAR also mounted an intensive, widespread, and carefully designed communication and information campaign to ensure transparency, inform all prospective beneficiaries about PICAR’s objectives and opportunities, and hedge the risk of elite capture. 104. Projects should facilitate dedicated spaces for women’s participation, so that women themselves analyze and make decisions about their needs and the execution of community and women's projects. The Project adopted a gender strategy to increase the active participation of indigenous women throughout the subproject cycle and address their specific capacity-building needs. Gender needs were considered as early as Page 28 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) project preparation, during the social consultations. The consultations created an essential space for women to incorporate their specific needs into CPP and fostered their participation in the subproject cycle. The flexibility to increase the subproject ceiling for a second subproject, if identified and implemented by women, also incentivized women’s participation. Projects should be designed with opportunities for women to assume new roles and responsibilities in which they can acquire and exercise authority. For example, in the Administration and Social Control Committees created under PICAR, women demonstrated their capacities, work, and leadership in tasks unrelated to their domestic work. Anecdotal evidence collected during field visits and findings from the final social and gender evaluation suggest that women’s participation strengthened their confidence and self-esteem, and their participation is acknowledged and valued by their households and the wider community. 105. Rural and poor communities with no prior experience can gain the capacity to assume fiduciary responsibility for major works as well as O&M if sufficiently empowered. PICAR piloted the direct transfer of project resources and fiduciary responsibility to the community level in Bolivia. This shift in power from the top to the bottom is fundamental in CDD.49 Anticipated risks related to insufficient capacity did not materialize. The results were overwhelmingly positive, with technically sound and impactful interventions implemented under active community supervision and with private sector participation, yielding a greater degree of sustainability. The participation of community-based organizations in financing subprojects generate a sense of ownership as well as a willingness to share responsibility for the future O&M of project investments. Active participation in the selection, execution, and supervision of subprojects ensures that investments respond to perceived needs, generates cost savings, increases accountability at the local level, and ensures proper O&M. 106. Beyond the definition of sustainability used by PICAR, other relevant dimensions of Project sustainability were identified in developing this report.50 Social sustainability is the bedrock on which other forms of sustainability rest. The Project was able to tap into social sustainability by building on existing local-level institutions and forms of government, laying the groundwork for decentralized conflict resolution. PICAR also delivered asset sustainability (infrastructure and services), because communities were sufficiently empowered and organized to assume responsibility for O&M, particularly for productive infrastructure and micro-irrigation subprojects. Since such responsibility has been assumed by communities, it is recommended that the authority to levy user fees and local taxes to finance maintenance is considered in the future, particularly given the fact that different types of investments require different conditions to promote their sustainability. Fiscal and environmental sustainability imply additional transfer of responsibilities to communities to reinforce long-term management of resources within a decentralized approach to development. For example, the management of land, water, forests, pastures, and other environmental resources should be embedded in powerful incentives for environmental sustainability to be managed by communities. 107. Realistic, more flexible evaluation frameworks may be needed where systemic resistance toward formal, classic IE is evident. The World Bank can avoid building up expectations by considering other options if evidence suggests systemic aversion to IE based on cost and other considerations. There may be trade-offs between the importance of organized, methodologically sound data to measure project achievements and attribution on the one hand, and the institutional learning to be gained from coordination units having managed the M&E function themselves, on the other. M&E technical and systems needs must be analyzed upstream, the 49 Narayan, Deepa (2002). “Empowerment and Poverty Reduction: A Sourcebook.” Washington, DC: World Bank. https://openknowledge.worldbank.org/ handle/10986/15239. 50 World Bank (2003), “Scaling-up Community Driven Development, Theoretical Underpinnings and Program Design Considerations.” World Bank Policy Research Working Paper 3039. Washington, DC. Page 29 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) options well-justified and defined in the PAD, and established no later than the first year, backed by close Bank monitoring and agreed Bank enforcement options if M&E commitments start to slip. 108. It remains unclear whether projects such as PICAR can have a transformative impact in slowing or reversing flight from rural to urban areas. The RF and impact measurements undertaken for PICAR do not permit an answer to this question. The evidence of continued rural–urban flight cited previously would seem to suggest otherwise, yet PICAR might well have slowed internal migration, even as the magnitude of Project investments was insufficient to reverse the broader trend. Future engagements in Bolivia should seek to elucidate this question, as it could provide a potent additional rationale for CDD investments in rural communities. . Page 30 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: Improve access to sustainable basic infrastructure and services for disadvantaged rural communities Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Direct project beneficiaries Number 0.00 150000.00 350,000.00 362,619.00 01-Nov-2011 13-Jun-2011 02-Jul-2015 31-Dec-2020 Female beneficiaries Percentage 0.00 45.00 49.70 Indigenous beneficiaries Percentage 0.00 85.00 92.00 Female heads of household Percentage 0.00 20.00 20.00 (considered most vulnerable in targeted communities) Page 31 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Comments (achievements against targets): Target exceeded: 103,6 percent of the target reached. The number of direct beneficiaries corresponds to the total number of people (family members) who are direct beneficiaries of the subprojects that were executed in the communities (including communal subprojects and subprojects of women beneficiaries). The project has reached 362,619 direct beneficiaries (DB) through 2,197 subprojects. The total number of families reached is 46,585 according to the registers of the Geo-referenced Management Information System (SIGG). According to the subproject typology, PICAR reached: i) 73,638 direct beneficiaries (DB) with 488 irrigation subprojects; ii) 29,866 DB through 129 roads subprojects; iii) 206,970 DB with 1,303 productive investments; iv) 40,500 DB with 199 water and sanitation subprojects; v) 3,709 DB with 29 electrification subprojects; and vi) 7,936 DB through 49 other type of investments. Targets for both, women beneficiaries and indigenous people, were exceeded by 5 and 7 percent respectively. The last iteration of the Beneficiary Survey conducted in December 2020, confirmed that all participants self-identified as IP (34.3 percent Aymara and 65.8 Quechua). The target for Female heads of household (considered most vulnerable in targeted communities) was achieved at 100 percent. This indicator was added as a PDO indicator during the AF to replace the Social Capital Index. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Subproject fully functional 12 Percentage 0.00 70.00 75.00 99.70 months after their completion 01-Nov-2011 13-Jun-2011 02-Jul-2015 31-Dec-2020 Comments (achievements against targets): Target exceeded: Out of 2,197 subprojects financed by PICAR, 1,960 subprojects had been completed for 12 months or more at the time of project closure. Out of these, 1,956 were fully functional at the time of the evaluation. This means that 99.79 percent of the subprojects that were completed for more than 12 months were fully functional, surpassing the final target of 75 percent.There are 4 subprojects from the original financing that are no longer in Page 32 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) operation, including Taypi Uta Choquemarca Community (tourist hostel); Junthuta Community (drinking water well); Tazapaya Community (ecological latrines), and; Mataral Puente Loma Community (sewing center). Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Improved access to basic Text 0.00 Additional 3 months 3.0 Additional months infrastructure and services in of road access per of road access per targeted communities year; 1,400 ha year (corresponding to irrigated (new and 129 communities with improved); 480 km of improved rural road fences constructed; access during the additional 400 m3 per rainy season); 2,611 day safe drinking ha irrigated (new and water available for improved); 2,809 km targeted communities of fences constructed; additional 523 m3 per day of safe drinking water available for targeted communities 01-Nov-2011 02-Jul-2015 31-Dec-2020 Comments (achievements against targets): Fully achieved and/or Exceeded. The level of achievement per sub-targets per type of investments is as follows: Page 33 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) (a) Road access: 100 percent of the target achieved. A total of 129 communities have improved their access through tertiary roads which has enabled 3 additional months of road access per year. Out of the 129 subprojects under this typology, 70 percent are investments in road improvement; 18 percent in vehicular bridges; 7.5 percent in pedestrian bridges; 4 percent in road bumps; and 3 percent in retaining walls. (b) Hectares irrigated: 187 percent of the target achieved. Investments of 488 community subprojects translated into 2,611 hectares with new or improved irrigation (the target was 1,400 ha). Thirty-nine percent of these subprojects include the improvement and extension of irrigation systems as well as water-harvesting systems for agriculture; 10 percent included the construction of water reservoirs, and 9 percent were water-harvesting systems for animals. (c) Kilometers of fence constructed: 210 percent of the target achieved. This target was met through the execution of 664 subprojects equivalent to 2,809 km of fence constructed. Fifty-four percent were animal pens; 27 percent were perimeter fences for plot protection, and 10 percent included barbed wire. (d) Cubic meters (m3) of additional water available/day: 131 percent of the target achieved. PICAR financed 155 sanitation subprojects resulting in an average of 3,375 liters of additional potable water per day per community. This is equivalent to 523 cubic meters of additional potable water per day delivered by the Project, achieving 130 percent of the target. This indicator was added with the AF. A.2 Intermediate Results Indicators Component: 1 - Community Capacity Building Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of beneficiaries Number 0.00 14000.00 23,282.00 trained in project management (of which 01-Nov-2011 02-Jul-2015 31-Dec-2020 Page 34 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) women) Comments (achievements against targets): Target exceeded:166 percent of the end-of-project target reached. The beneficiaries included as part of this indicator refer to the training provided to committee members in areas that include financial management, procurement, safeguards, subproject management, and Operation and Maintenance. The target of reaching at least 40 percent of women was exceeded:132.5 percent of the target reached). Fifty-three percent of trained beneficiaries are female. This indicator was included in the AF. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of community Percentage 0.00 70.00 92.40 members satisfied with planning process and sub- 01-Nov-2011 02-Jul-2015 31-Dec-2020 project identification (of which women) Comments (achievements against targets): Target Exceeded: 132 percent of the end target was reached. This value is based on the results of the Beneficiary Satisfaction Survey, completed in December 2020. 91.5 of women are satisfied with the planning process and sub-project identification, exceeding the target of 70 percent of women satisfied with the process. The indicator was modified in the AF. Page 35 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Component: 2 - Community-Driven Development Investment Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of beneficiaries Number 0.00 7800.00 8,966.00 engaged in sub-project implementation committees 01-Nov-2011 02-Jul-2015 31-Dec-2020 (of which women) Comments (achievements against targets): Target exceeded: 115 percent of target achieved. This indicator was measured among the beneficiaries participating in the Administration and Social Control Committees. The target for women's participation was also exceeded: 132 percent of the target was achieved) . Sixty-six percent of beneficiaries engaged in sub- project implementation were women. Both of these indicators were measured as part of the Beneficiary Satisfaction Survey completed in December 2020. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent of subprojects Percentage 0.00 80.00 100.00 satisfactorily completed implementation, including 01-Nov-2011 02-Jul-2015 31-Dec-2020 Page 36 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) final accounting Comments (achievements against targets): Target Exceeded: 125 percent of the target achieved). A total of 2,197 subprojects have completed their implementation satisfactorily, including the final accounting. Verified in the GMIS upon project closure. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of completed Percentage 0.00 90.00 98.00 subprojects whose technical and fiduciary audits have no 01-Nov-2011 02-Jul-2015 31-Dec-2020 substantial remarks Comments (achievements against targets): Target Exceeded: 108.9 percent of the target reached achieved). According to the financial audit carried out in July 2020, 98 percent of the subprojects did not have substantial observations. Component: 3 - Coordination, Monitoring and Evaluation Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Page 37 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Percentage of community Percentage 0.00 75.00 94.10 members satisfied with sub- project outputs 01-Nov-2011 02-Jul-2015 31-Dec-2020 (infrastructure/service) (of which female) Comments (achievements against targets): Target Exceeded: 125 percent of the target reached. Based on ub-project outputs as reported in the last Beneficiary Satisfaction Survey, completed in December 2020 (of which 95.9 percent of women reported being satisfied with subproject outputs. Page 38 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) B. KEY OUTPUTS BY COMPONENT 109. Objective/Outcome 1 Improved access to basic infrastructure and services for the most disadvantaged rural communities 1. Direct project beneficiaries – Final target: 350,000 beneficiaries. The Project reached 362,619 beneficiaries. 2. Breakdown of Outcome Indicator No 1: Female beneficiaries (percentage) – Final target: 45%: This outcome was exceeded by nearly 5%, as 49. 7% of Project beneficiaries were women. 3. Improved access to basic infrastructure and services in targeted communities, as measured by the following subindicators: Outcome Indicators o Additional road access per year (months) – Final target: 3 months per year: 100% achieved. o New and improved irrigated area (hectares) – Final target: 1400 hectares: 187% achieved o Fences constructed (kilometers) - Final target: 480 kilometers: 210% achieved o Safe drinking water available for targeted communities (cubic meters of additional potable water per day) – Final target: 400 cubic meters per day: 130% achieved. Component 2: Community Investments 1. Percent of subprojects satisfactorily completed implementation, including final accounting (80%): 125% achieved. Intermediate Results Indicators 2. Percentage of community members satisfied with sub-project outputs (infrastructure/service) (75%): 125% achieved. 3. Percentage of women satisfied with subproject outputs (infrastructure/service) (75%): 128% achieved. Component 1: — 996 Community Participatory Plans (PPC by its Spanish acronym) Component 2: Key Outputs by Component — 2,197 subprojects under the following typology: (linked to the achievement of the o 73,638 beneficiaries with 488 irrigation subprojects. Objective/Outcome 1) o 29,866 beneficiaries with 129 road improvement subprojects. o 206,970 beneficiaries with 1,303 agricultural productive investments. o 40,500 beneficiaries with 199 water and sanitation subprojects. Page 39 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) o 3,709 beneficiaries with 29 rural electrification subprojects (solar panels). o 7,936 beneficiaries with 49 other types of investments, such ecological kitchens, tourist hostels, sewing and weaving facilities, and improved school and health post infrastructure. Objective/Outcome 2: Sustainability of improved access to basic infrastructure and services: 1. Subproject fully functional 12 months after their completion (percentage)– Final target: 75%: 133% Outcome Indicators achievement — 92.4% of community members were satisfied with the planning process and subproject identification, corresponding to 132% of the target achieved. Beneficiaries were satisfied with the planning and subproject identification process because decisions were reached by general consensus as part of the community Intermediate Result Indicators assemblies that took place during the CPP workshop, and the Project fulfilled their expectations by addressing the problems identified by the communities. — 92.4% of community members were satisfied with the planning process and subproject identification. 91.5% of women were satisfied with this process. Components 1: — 23,282 beneficiaries that were part of Social Control and Administration Committees trained in topics such as financial management, procurement, safeguards, subproject management, and O&M. 53% of trained beneficiaries were women. Component 2: Key Outputs by Component (linked — 8,966 beneficiaries actively engaged in Implementation Committees. 66% of beneficiaries engaged in to Outcome 2) subproject implementation were women. Component 3: — Three rounds of the Beneficiary Satisfaction Survey. — Two evaluations for Mid-Term Reviews. — Final Project Evaluation. — Gender and Social Final Evaluation. Page 40 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Supervision/ICR Garry Charlier Task Team Leader(s) Julio Sanjines Gonzales Procurement Specialist(s) Enrique Antonio Roman Financial Management Specialist Juan Carlos Serrano-Machorro Financial Management Specialist Tatiana Perez Guerra Financial Management Specialist Mario I. Mendez Team Member Jorge Trevino Team Member Maria Pia Cravero Counsel Griselle Felicita Vega Team Member Julia Isabel Navarro Espinal Team Member Ximena Rosio Herbas Ramirez Environmental Specialist Angela Maria Caballero Espinoza Social Specialist Karla Janeth Vera Aguirre Team Member Anna Roumani Consultant B. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY09 7.425 38,760.89 FY10 22.509 101,456.28 Page 41 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) FY11 21.898 130,366.68 FY12 .900 5,638.74 Total 52.73 276,222.59 Supervision/ICR FY12 14.373 83,525.21 FY13 14.025 63,360.68 FY14 14.258 89,514.67 FY15 34.733 167,412.94 FY16 21.808 78,999.34 FY17 25.510 143,720.73 FY18 20.551 138,476.15 FY19 23.674 159,648.59 FY20 32.124 206,342.37 Total 201.06 1,131,000.68 Page 42 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 3. PROJECT COST BY COMPONENT Initial AF Planned Initial Closing Closing AF Closing Appraisal Planned and AF Initial and Percentage of Component Amount Amount Amount Amount Amount AF Amount Approved (%) (US$M) (US$M) (US$M) (US$M) (US$M) (US$M) 1 - Community 5.0 2.68 8.5 4.31 13.5 6.99 51.77 Capacity Building 2 - Community- Driven Development 33.0 37.66 47.1 58.9 80.1 96.55 120.5 Investment Of which contributions from 3.0 8.88 4.1 9.69 7.1 18.57 261.54 communities 3 - Coordination, Monitoring, and 5.0 4.87 8.5 5.88 13.5 10.76 79.7 Evaluation Total 43.0 45.21 64.1 69.09 107.1 114.3 106.7 Page 43 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 4. EFFICIENCY ANALYSIS I. Methodology The cost-benefit methodology was used to assess Project efficiency. Economic and financial indicators were estimated to evaluate the return on the investment operation. Sampling Table A4.1 shows the total number of PICAR subprojects, disaggregated by type, category (community and women’s subprojects), and geographical location (department). A statistically representative, stratified random sample of these subprojects (n = 64, with a confidence level of 90% and sampling error of 10%) was used for the efficiency analysis. The sample was allocated proportionally across subproject types, categories, and geographic locations, as shown in Table A4.2. Ideally, a larger number of subprojects would have been analyzed, resulting in a higher level of confidence with a lower sampling error. That approach was not possible, however, because the additional fieldwork required by a larger sample was precluded by the post-election disturbances and advent of the pandemic in late 2019 and early 2020. Table A4.3 shows the distribution of direct beneficiaries in the sample by subproject type. Table A4.1: Subprojects supported under PICAR Subproject type Subproject category Department Total Community Women’s Total Chuquisaca Cochabamba La Paz Oruro Potosí Sanitation 199 168 31 199 69 44 60 26 Rural roads 129 122 7 129 18 16 42 30 23 Electrification 29 27 2 29 19 10 Micro-irrigation 488 420 68 488 87 98 185 33 85 Others 49 7 42 49 35 9 4 1 Productive 1303 480 823 1,303 275 180 424 186 238 Total 2,197 1,224 973 2,197 503 357 711 279 347 Table A4.2: Sample distribution by subproject type, category, and location Subproject type Subproject category Departments Total Community Women’s Total Chuquisaca Cochabamba La Paz Oruro Potosí Sanitation 9 8 1 9 3 3 2 1 0 Rural roads 6 5 1 6 1 1 2 1 1 Electrification 2 1 1 2 1 1 0 0 0 Micro-irrigation 11 9 2 11 3 3 3 1 1 Other 5 2 3 5 2 1 0 1 1 investments Productive 31 15 16 31 7 6 7 6 5 Total 64 40 24 64 17 15 14 10 8 Table A4.3: Number of direct beneficiaries in the subproject sample, by type of subproject Subproject type Subproject sample size Direct beneficiaries Electrification 2 59 Other investments 5 154 Productive 31 1,008 Sanitation 9 246 Micro-irrigation 11 341 Rural roads 6 384 Total 64 2,192 Page 44 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Data Collection and Analysis The PICAR team collected field data from February to May 2020 through direct interviews with technical assistants in each department. A specific questionnaire was used to collect data on each type of subproject to develop “with Project” and “without Project” scenarios for the efficiency analysis.51 Data were collected on production volumes, prices, willingness to pay for infrastructure services, productive parameters, technical characteristics of infrastructure, numbers of beneficiaries, O&M costs, production costs, and other variables for each evaluated subproject. This information was complemented with data from the PICAR M&E system to estimate the incremental benefits and costs for: (i) evaluated subprojects, (ii) subproject categories, (iii) subproject types, (iv) subproject locations, and (v) the entire Project. The analysis estimated the incremental net benefits valued at private and social prices, including the incremental NPV over a 20-year period, economic and financial IRRs, benefit/cost ratio, payback period, and incremental benefits per family in terms of present value. Hypothetical changes in key variables such as benefits, costs, and investment were used to conduct the sensitivity analysis. Assumptions and Parameters Individual and aggregate cash flows were estimated using nominal prices over a 20-year period. A 12.81% cost of capital and 12.67% social discount rate were the reference discount rates52 used to evaluate the feasibility of private and public investment, respectively, in the country.53 The exchange rate (Bs:US$) was 6.68 (the average value in 2020). Factors for converting market to social prices were: (i) Currency conversion factor: 1.24,54 (ii) domestic tradable goods: 0.885,55 and (iii) output conversion factor: 1.00. The analysis could not take into account all potential benefits (for example, environmental and health benefits) related to the positive externalities generated by the subprojects focused on sanitation, electricity, education, and health infrastructure. For some types of investments, beneficiary contributions were assumed to represent willingness to pay for the service and used to estimate the incremental cash flows. The O&M costs were based on the type of infrastructure financed by the Project. For example, estimates of O&M costs for drinking water and irrigation systems considered the costs of cleaning the distribution system, repairing pipes, and replacing pressure mechanisms for water distribution. The lifespan of each type of subproject was established during the pre-investment analysis. Table A4.4 shows the hypothetical incremental benefits assumed to arise from each type of subproject. The efficiency analysis also incorporated the parameters and assumptions used in the detailed case studies of 10 subprojects conducted in preparing the AF. 51 Data collection included questions related to key parameters such as yields before and after the completion of micro-irrigation subprojects. 52 http://www.iidee.net/archivos/Inversion/%20Fconv%20Bolivia.pdf 53 http://www.iidee.net/archivos/Inversion/%20Fconv%20Bolivia.pdf 54 http://www.iidee.net/archivos/Inversion/%20Fconv%20Bolivia.pdf 55 http://www.iidee.net/archivos/Inversion/%20Fconv%20Bolivia.pdf Page 45 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Table A4.4: Incremental benefits by type of subproject Subproject type Benefits Hypothesis Sanitation • Willingness to pay for the service† Survey to determine willingness to pay and continue paying for access to drinking water and sanitation services (sewerage and silos) by each of the project beneficiaries. • Disease prevention The use of sanitary infrastructure prevents diseases and infections such as diarrhea. Case study information from AF preparation. • Time savings Domestic access to drinking water allows beneficiaries to spend less time collecting water. Case study information from AF preparation. Roads • Willingness to pay for the service Survey to determine willingness to pay for the use off-road infrastructure. • Time savings Good road infrastructure reduces travel time. • Increased volume of product sales Access to better road infrastructure increases opportunities to market agricultural products. Electrification • Willingness to pay for the service Survey to determine willingness to pay for access to electricity. • Time saving Beneficiaries who have access to electricity may have a greater chance of doing an activity in less time. • Increase in hours worked per day Beneficiaries who have access to electricity may have more hours per day for working. Case study information from AF preparation. Micro-irrigation • Increased yields Irrigation increases crop yields. • Increased production volumes Irrigation increases crop yields. • Time savings Beneficiaries with direct access to irrigation systems spend less time irrigating crops. Other investments • Willingness to pay for the service Survey to determine willingness to pay for health and education infrastructure. • Disease prevention Access to modern health and education infrastructure prevents disease and infection. Case study information from preparation of the AF. • Increase in sales revenue Rural businesses such as tourism generate income. Productive • Improvement of production Improvement of productive infrastructure such as shelters has positive parameter effects on different agriculture activities (e.g., reduction of animal mortality). • Increased yields Adoption of agricultural technologies increases crop and animal yields. • Increased production and sales Adoption of agricultural technologies and practices increases production volume and sales volumes. • Price improvement Improvement of harvest infrastructure and adoption of post-harvest practices increase the quality and price of agricultural products. • Reduction of losses in the Adoption of good post-harvest practices reduces food loss and waste. production process † For each type of subproject, “willingness to pay” refers to the maximum price at or below which a beneficiary would definitely buy one unit of a product or service. The sensitivity analysis incorporated hypothetical changes of -20% to 20% in some variables such as incomes, costs, financial investment, and beneficiary contribution. In addition, the direct participation of beneficiaries in implementing subprojects was estimated to reduce the O&M costs borne by government agencies by 50%. Cost-Efficiency and Cost-Effectiveness Analysis The cost-efficiency analysis was conducted for each subproject type by dividing the total cost for each product unit delivered to beneficiaries. These results were compared with the unit cost of the other similar projects executed by Bolivian government institutions. To compare the marginal cost incurred by each subproject type, the following cost-effectiveness indicators were estimated: (i) cost present value (CPV), (ii) equivalent annual cost (EAC); and (iii) cost- effectiveness ratio (CER). Page 46 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) II. Results Aggregate Results Table A4.5 shows the aggregate economic and financial indicators for two periods of evaluation (20 and 15 years). The estimated FIRR was 15.65% and the EIRR was 18.76%, indicating that the total investment yielded returns higher than the cost of capital and the social discount rate. The incremental financial NPV was US$ US$11,280,614, and the economic NPV was US$24,478,184. These estimates reflect both direct and indirect project costs (including the cost of TA). If project management costs are not considered in the analysis, the FIRR is 17.72% and the EIRR is 21.05%. These results are in line with economic and financial returns reported by similar projects implemented in the LA region and are higher than those estimated for the Original Credit (13%) and AF (13.9%). Table A4.5: Aggregate economic and financial indicators Indicator type 20 years 15 years Financial indicators FNPV 11,280,614.2 (6,245,939.14) FIRR 15.65% 9.30% B/C 1.17 0.90 Payback period 10.78 FNPV/beneficiary family 152.69 --- Economic indicators ENPV 24,478,184.5 1,607,098.9 EIRR 18.76% 13.35% B/C 1.33 1.02 Payback period 10.63 ENPV/beneficiary family 331.32 6.3 In economic terms, the estimated return on investment is positive for both the 20- and 15-year evaluation periods. These results indicate that community infrastructure investments implemented by communities themselves are likely to yield positive returns in the medium and long term, and the investment made by PICAR would be recovered around year 11. Sensitivity Analysis The sensitivity analysis shows that the return on investment is sensitive to changes in beneficiaries’ income and therefore to changes in production and prices (Table A4.6). An income reduction of 5% would reduce the FIRR to 14%. This significant elasticity of returns occurs mainly because all subprojects in both phases financed very poor, small-scale producers who are highly affected by any variation in income. If the level of investment were to increase by 10%, the IRR would be reduced by up to 14.1%. A 10% increase in the beneficiary contribution would reduce the IRR to 14.9%. These results reveal the importance of conducting adequate technical studies before implementing rural infrastructure subprojects. If the Project had been implemented without direct community participation, the FIRR would have been only 14.74% and the EIRR would have been 17.77%. Page 47 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Table A4.6: Sensitivity analysis, PICAR Hypothetical changes EIRR change (%) (%) Benefits Cost Beneficiary contribution Financial investment (%) (%) (%) (%) 20 23.71 17.81 18.09 15.63 10 21.28 18.29 18.42 17.10 5 20.03 18.53 18.59 17.90 0 18.76 18.76 18.76 18.76 -5 17.45 18.99 18.93 19.67 -10 16.11 19.23 19.11 20.66 -20 13.27 19.69 19.46 22.86 Disaggregated Economic and Financial Indicators Table A4.7 shows the estimated economic and financial indicators by subproject category (community and women’s subprojects), without considering indirect costs, over two evaluation periods. The results indicate that the return on investment is positive in economic and financial terms. Community projects are estimated to have a better return on investment than women's subprojects, however, probably because community subprojects financed more irrigation and road infrastructure. Women's subprojects tended to finance very small-scale production activities with low profitability, such as vegetable and poultry production, which are nevertheless important for household consumption and improved nutrition. Table A4.7: Economic and financial indicators by subproject category, PICAR Subproject Financial indicators Economic indicators category NPV IRR B/C Payback NPV/family NPV IRR B/C Payback NPV/family Community subprojects 20 years 60,936,168 24.74% 1.63 4.17 1,352 88,715,661 27.75% 1.82 3.76 1,968 15 years 49,340,891 24.03% 1.52 73,739,559 27.18% 1.69 Women’s subprojects 20 years 5,746,314 15.31% 1.08 6.2 199 14,893,873 18.06% 1.19 5.3 517 15 years 2,254,057 13.97% 1.03 10,071,167 16.95% 1.52 Table A4.8 shows the estimated economic and financial indicators for each subproject type without considering indirect costs. Investments related to rural road improvement, implementation of micro- irrigation systems, and construction of productive infrastructure show positive returns in economic and financial terms. These results can be explained by the fact that these types of infrastructure investments improve the productivity of agricultural systems by: (i) increasing animal and crop production; (ii) reducing crop losses and animal mortality; (iii) improving product quality; and (iv) allowing producers to access markets. These results highlight the importance of investing in strategic rural infrastructure to boost small- scale agriculture in isolated areas. Investments in sanitation and electrification did not yield positive returns on investment in economic and financial terms, probably because these types of infrastructure are not directly linked to typical agricultural activities, and because positive health externalities (arising from improved sanitation, for instance) were not considered in the analysis. Page 48 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Table A4.8: Economic and financial indicators by subproject type, PICAR Subproject Financial indicators Economic indicators type NPV IRR B/C Payback FNPV/family NPV IRR B/C Payback ENPV/family Sanitation 20 years (435,651.70) 12.03% 0.97 7.46 (55.97) 1,164,491.97 14.47% 1.08 149.62 6.45 15 years (1,153,855.50) 10.35% 0.91 203,325.35 13.04% 1.01 Rural roads 20 years 9,659,412.31 33.63% 2.14 2.96 1629.73 13,272,146.58 38.00% 2.41 2239.27 2.63 15 years 8,307,971.85 33.28% 1.99 11,540,116.19 37.75% 2.24 Electrification 20 years 163,944.50 16.71% 1.21 5.71 191.97 322,761.98 19.50% 1.38 377.94 5.02 15 years 95,807.09 15.49% 1.13 235,163.52 18.49% 1.28 Micro-irrigation 20 years 34,797,092.86 31.06% 1.89 3.82 1978.68 49,423,567.41 35.67% 2.15 2810.39 3.19 15 years 29,383,912.09 30.60% 1.77 42,392,524.81 35.34% 2.02 Others 20 years 758,436.91 20.12% 1.35 4.84 474.91 1,274,051.72 23.59% 1.54 797.78 4.18 15 years 564,224.70 19.16% 1.27 1,021,510.11 22.84% 1.44 Productive 20 years 39,811,706.82 20.71% 1.36 5.04 991.97 66,354,796.15 24.35% 1.55 1653.33 4.16 15 years 30,030,328.13 19.78% 1.28 53,548,090.27 23.63% 1.45 Table A4.9 summarizes results by geographical location (department). Subproject investments were profitable in all departments except for Oruro, where the economic and financial indicators are below the cost of capital and social discount rate. This result may be explained by the fact that fewer micro-irrigation subprojects were financed in Oruro. Investments in La Paz had the best estimated returns on investment. Table A4.9: Economic and financial indicators by department, PICAR Financial Indicators Economic Indicators Department NPV IRR B/C Payback FNPV/family NPV IRR B/C Payback ENPV/family Cochabamba 20 years 1,469,435.42 14.12% 1.06 6.58 134.39 4,133,996.97 15.76% 1.15 6.00 134.39 15 years (145,814.91) 12.66% 0.99 2,025,943.63 14.46% 1.08 Chuquisaca 20 years 26,461,640 26.14% 1.72 5.80 1555.83 37,636,884.75 28.94% 1.91 4.13 2,212.89 15 years 21,587,307 25.46% 1.61 31,329,693.51 28.39% 1.78 La Paz 20 years 34,439,006 30.88% 1.59 6.78 1396.04 50,909,131 35.9% 1.78 5.22 2,063.69 15 years 29,299,509 30.45% 1.52 44,038,403 35.6% 1.70 Oruro 20 years 3,671,112.36 16.46% 1.17 5.79 333.34 7,321,254.57 18.97% 1.31 5.11 664.78 15 years 2,056,484.64 15.21% 1.10 5,214,818.09 17.93% 1.22 Potosí 20 years 22,994,199 28.2% 1.83 6.48 2241.81 32,197,394 31.4% 2.03 4.83 3,139.07 15 years 19,276,847 27.7% 1.71 27,424,018 31.0% 1.89 Page 49 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Cost-Efficiency/Cost-Effectiveness Analysis Table A4.10 compares the unit costs for some types of subprojects financed by PICAR with the costs of other similar projects executed by Bolivian government institutions. Most of the PICAR subprojects had lower costs per unit of product compared to similar projects implemented by government institutions. Table A4.10: Unit costs of PICAR subprojects and similar projects executed by Bolivian government institutions Subproject type Measurement unit PICAR subprojects Other projects (efficiency Efficiency parameters) range Number Unit cost Unit cost (US$) Difference Range (US$D) Electrification Connection to electricity grid 29 1.380 4.500 - 69% ± 20% Sanitation Connection to sanitary grid 199 590 854 - 31% ± 20% Micro-irrigation Incremental hectares 488 2.875 3.500 - 18% ± 15% Fences and farmyards Kilometers 664 3.025 5.500 - 45% ± 15% Rural roads Kilometers 99 5.407 9.000 - 40% ± 20% Vehicle bridges Meters 24 4.684 4.000 + 17% ± 20% Pedestrian bridges Meters 6 1.097 700 + 57% ± 15% TOTAL 2,197 Source: Final evaluation report. † Reference efficiency ranges. Table A4.11 shows the cost-effectiveness indicators at present value: (i) for all subprojects; (ii) by subproject category (community and women’s subprojects); (iii) by subproject type; and (iv) by subproject location (department). The average annual cost per beneficiary family over the entire project evaluation period is US$136 at private prices and US$142 at social prices. This value reflects the low O&M costs of the PICAR model, which indicate that the investment operation has an appropriate level of efficiency. Table A4.11: Cost-effectiveness analysis Subproject Private price (US$) Social price (US$) characteristic CPV EAC CER CPV EAC CER a. Aggregation Total subproject 71,840,121.8 10,110,132.3 136.8 74,912,287.4 10,453,182.1 141.5 b. Subproject category Community 96,731,161.0 13,613,073.2 302.0 108,521,900.4 15,143,032.2 336.0 Women’s 70,357,264.9 9,901,448.4 343.7 80,280,732.4 11,202,289.3 388.8 c. Subproject type Sanitation 13,509,297.0 1,901,176.9 244.3 15,185,317.8 2,118,943.3 272.3 Rural roads 8,500,621.4 1,196,301.0 201.8 9,438,665.2 1,317,061.4 222.2 Electrification 774,554.5 109,003.8 127.6 1,439,101.9 200,810.8 235.1 Micro-irrigation 39,077,156.4 5,499,367.4 312.7 43,014,350.7 6,002,177.4 341.3 Others 2,157,025.7 303,560.4 190.1 2,372,005.5 330,987.2 207.3 Productive 109,978,341.8 15,477,362.2 385.6 121,005,270.8 16,884,948.7 420.7 d. Department Cochabamba 23,424,181.8 3,296,508.6 301.5 26,997,787.5 3,767,243.0 344.5 Chuquisaca 36,508,096.2 5,137,821.0 302.1 41,170,545.6 5,744,894.8 337.8 La Paz 58,492,565.1 8,231,717.3 333.7 65,310,444.8 9,113,351.0 369.4 Oruro 21,145,276.8 2,975,796.0 270.2 23,713,949.0 3,309,019.6 300.5 Potosí 27,727,364.7 3,902,099.8 380.4 31,234,640.9 4,358,449.1 424.9 Page 50 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Community subprojects show greater efficiency than women’s subprojects because they benefit a larger number of families. For productive and micro-irrigation subprojects, the annual cost per beneficiary is higher than it is for other types of subprojects. These results can be explained by the higher O&M costs of productive and micro-irrigation investments. The estimated cost per beneficiary is lower in Oruro than in other departments. The case studies of 10 types of subprojects carried out in preparing the AF also indicated the PICAR had a high level of efficiency owing to the low unit costs achieved in implementing each type of subproject. These findings support the conclusion that PICAR was a cost-efficient intervention and that the PICAR model should be scaled up. The Project was able to overcome an initial delay of 16 months of the AF, as discussed in Section II of the main text. The resulting savings in operating costs were reallocated to benefit an additional 45 communities, equivalent to 90 subprojects. PICAR also achieved implementation efficiency by establishing an effective operational interface with communities of rural and indigenous people while respecting their traditions and customs. This interface included dialogue, coordination, and complementarity with the organizational processes embedded in these communities. As a result, the coordination and direct execution of subprojects was facilitated on the one hand, while leadership, empowerment and self-management capacities for subproject execution were developed among project beneficiaries through the Project’s CDD focus on capacity building under Components 1 and 2. The implementation efficiency arises from putting communities in charge of subproject implementation and O&M: costs are reduced as beneficiaries are actively involved in providing labor and local materials for subprojects and carrying out subsequent O&M. III. Conclusions The ex-post cost-benefit analysis shows that the entire project has positive returns on investment. The Project FIRR is 15.65%, and the EIRR is 18.76%. Considering PICAR’s social objective, the overall Project rate of return would be 18.76% (social prices). The estimated IRRs are above the cost of capital and the social discount rate used in Bolivia. They are also higher than the economic and financial indicators estimated for the Original Credit (13%) and AF (13.9%) Community and women's subprojects have positive returns on investment in economic and financial terms. In economic terms, all types of subprojects have positive returns on investment. Likewise, the investments made in all departments show IRRs above the social discount rate. The unit costs of the products delivered to the beneficiaries of most of subprojects financed by PICAR are significantly lower than costs in similar initiatives implemented by government institutions. Together, these results show the high efficiency achieved by the Project in relation to the costs incurred by each beneficiary. Page 51 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS Letter from the Project Coordination Unit The draft ICRR was shared with the borrower on June 18,2021. No additional comments were received prior to submission of this report. Further communications will be archived in WBDocs if received. Page 52 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 6. SUPPORTING DOCUMENTS 1. Project Appraisal Document, Report No. 59877-BO, June 13, 2011 2. Financing Agreement, Credit Number 5004-BO 3. Project Paper, Report No. PAD1420, July 2, 2015 4. Financing Agreement, Credit Numbers 5712-BO and 5713-BO 5. Project Information Documents (PID) - Concept Stage 6. Project Information Documents (PID) – Appraisal Stage 7. Supervision Aide Memoires 8. Mid-Term Review Report, 2018 9. Implementation Supervision Reports (ISR) 10. Procurement Post Reviews (PPR) 11. Procurement Supervision Reports 12. Financial Management Supervision Reports 13. Restructuring Papers 14. Semester Progress Reports Studies and Surveys 1. Borrower’s Completion Report, Project Coordination Unit, 2020 2. Beneficiary Satisfaction Surveys completed in 2017, 2019 and 2020 3. Gender and Social Final Evaluation, Hoffmann Vignaud, 2020 4. Assessment of PICAR’s benefits for the AF, Juan Morelli from FAO, 2015 Page 53 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 7. PROJECT EVALUATIONS AND ASSESSMENTS Summary of Findings under the Borrower’s Completion Report 1. Final Report on Project Performance and Assessment of Subproject’s Impact Methodology 1. The evaluation followed a participatory approach by identifying relevant institutional, social, and individual actors that were related to project implementation. The scope was systemic to better understand the different dimensions of beneficiaries at the individual and collective levels. The study used mixed methods as follows: (i) Appreciative Inquiry: assessment of life experiences to identify lessons learned around participation and satisfaction; (ii) Theory of Change: causal relationships between financial and human resources, project activities, outputs, results and impacts; and (iii) Most Significant Changes: identification and collection of experiences at the collective level to highlight project’s key elements and associated changes. Quantitative analysis included semi-structured interviews covering 79 subprojects, triangulated with the data generated by the SIGG. Qualitative analysis was based on in-depth interviews and testimonials. The Beneficiary Satisfaction Surveys applied the same questionaries in three different years to a random sample of beneficiaries. Main Conclusions 2. The relevance of the Project was verified based on the methodology used for the selection of participating municipalities. The Project reached a significant percentage of communities with high levels of vulnerability to food insecurity and communities classified as Category A based on high levels of unsatisfied basic needs and low levels of gross production value. The formation of Prioritization Commissions, with the participation of different institutional and social actors of the municipality, contributed to the selection of the most impoverished communities, although the risk of using subjective criteria by committee members was identified. This risk was minimized with a preliminary selection of communities made by PICAR based on technical criteria and official data on poor communities. 3. The design of the Project was coherent given that its operative and administrative mechanisms were articulated (e.g., dialogue and complementarity) with the organizational and management procedures developed by indigenous and peasant communities, according to their traditions and customs. As a result, project activities related to execution of subprojects were organized while local capacities and leadership were developed as part of the community participatory planning process. 4. Project design included a cross-cutting gender approach. Consequently, the vision of community development was embraced and shared by men and women, from the identification of their needs up to the execution of subprojects. This approach proved to give legitimacy to women’s decisions as their vision was clearly privileged by the Project. PICAR has provided space for communities to change their perspective around gender roles. 5. In terms of efficiency, budget execution exceeded 90% by project closure. A closer look at budget execution per year showed that it was highest in 2014 (115%) and 2020 (321%). Execution was high in 2014 because of the need to extend the life of the Project, and it was high in 2020 owing to the social crisis precipitated by the political transition in late 2019 and the pandemic in 2020. Page 54 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) 6. The creation of Social Control and Administrative Committees was key to promote the active participation of beneficiaries through the entire the subproject cycle. These committees brought legitimacy and transparency to subproject implementation and the use of monetary resources. As such, it is unlikely that subprojects could be implemented with fewer resources under a different intervention model. The Project did not manage to record implementation times due to weaknesses in data collection, yet beneficiaries’ perception is that the established timelines were met both in terms of efficiency and administration of resources. 7. Communities were able to meet their counterpart requirements, which positively affects the sustainability of the subprojects by generating a feeling of ownership. 8. The assessment of subproject impact varies by type of subproject. Productive subprojects are highly valued by beneficiaries because they have helped them to improve their livelihoods, take better advantage of their limited agricultural productive potential, reduce their vulnerability, improve their resilience, generate additional income, and improve the institutional capacities of indigenous and peasant communities, including women’s participation in the decision-making process. 9. Benefits from water sanitation subprojects are related to significant improvements in accessing potable water and lowering the risk of water-borne diseases due to improved solid waste management. Road subprojects have improved market access for a longer period during the rainy season. Electrification subprojects have improved the living conditions of beneficiaries, enabling them to perform different activities at night and facilitating better access to information and modern communication technologies. The sustainability of the subprojects financed by the Project varies by type of subproject. 2. Beneficiary Satisfaction Survey 2020 Methodology 10. The third and final iteration of the survey was applied to 539 beneficiaries. The objective was to measure the degree of satisfaction with the subproject cycle, including the degree of satisfaction with CPP and subproject execution. The degree of satisfaction was measured on a 5-point Likert scale. Main Findings 11. Results from this survey are reflected in the relevant indicators captured in the RF. Tables A7.1 and A7.2 summarize results for each of the variables included in the survey instrument. Table A7.1: Community Participatory Planning Perception of Beneficiaries (Percentage) Variables Included in the Survey Very Good Good (%) Regular Poor Very Poor Way of explaining the CPP by relevant technicians 8.3 83.3 8.3 0.0 0.0 Work delivery by the entity hired to conduct the CPP 7.2 85.5 7.2 0.0 0.0 Work deliver by Community Planners 10.9 81.6 7.4 0.0 0.0 Ability to learn how to identify community needs and 17.6 76.6 5.8 0.0 0.0 solutions Participation in the selection of subprojects 23.0 67.5 9.3 0.0 0.2 Total Average 13.4% 78.9% 7.6% 0.0% 0.1% Page 55 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Table A7.2: Subproject execution Perception of Beneficiaries (Percentage) Variables Included in the Survey Very Good Good Regular Poor Very Poor Activities carried out by the Administration Committee 18.4 78.1 3.5 0.0 0.0 Activities carried out by the Social Control Committee 12.6 79.2 8.0 0.2 0.0 Accountability by the Administration Committee 28.9 64.6 6.5 0.0 0.0 O&M carried out by the Community 13.7 76.4 9.6 0.2 0.0 Criteria used by the Community to select subprojects 27.1 70.7 2.2 0.0 0.0 Evaluation of the results of the subproject operation 31.2 64.0 4.8 0.0 0.0 Resolution of Community needs by the Project 47.5 46.8 5.8 0.0 0.0 Total Average 25.6% 68.5% 5.8% 0.1% 0.0% Summary of Findings under the Gender and Social Evaluation Methodology 1. Mixed methods were used to evaluate and systematize the sociocultural dimensions of the Project, including the learning process experienced by beneficiaries as well as women’s empowerment and participation. The systematization focused on answering specific questions related to the initial situation/problems before PICAR, the process of the experience (what happened, when, which methods, who participated), and the current situation, as well as arriving at the main findings, conclusions, and lessons learned. The data collection instruments included: (i) 297 surveys applied to beneficiaries; (ii) 167 interviews with beneficiaries; and (iii) 3 interviews with regional Project staff. Owing to the mobility restrictions related to COVID-19, the different instruments were applied at three different times, and the methodology was adapted to collect data via mobile phone, and community members were trained to use the instruments with other beneficiaries. The resulting data were verified with a random sample from the SIGG of 18 communities (out of 68). Main Findings 2. Results were disaggregated by sociodemographic characteristics and subproject category and type. Seventy percent of the evaluated sample were communal subprojects, and thirty percent were subprojects identified and implemented by women. Disaggregated by type of subproject, 55% of the sample consisted of productive subprojects, 17% were water and sanitation subprojects, 16% were micro- irrigation investments, and 4% related to tertiary road infrastructure. The study reached beneficiaries from all five departments; 53% of the surveyed beneficiaries were men and 47% were women. Beneficiaries ranged in age from 21 to 75 years, averaging 49 years old. In terms of education attainment, 60% of men and 78% of women surveyed had completed elementary education, 37% of men and 15% of women had no formal education, and only 1.6% of men and 7% of women reported having some secondary education. 3. PICAR had a significant social impact at the community level, including social and economic empowerment. Committees learned how to manage resources, hold each other accountable, and monitor subproject execution Communities gained the experience and ability to undertake new subprojects. They recognized and valued women’s role in the community while promoting their empowerment and greater participation within the community. In addition, there was observed improvement at the household level Page 56 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) in agricultural production, access to basic services, earnings from selling agricultural surplus, and social cohesion. 4. Social Control and Administration Committees were the keys to transparency and accountability with subproject execution. They exercised an internal control among beneficiaries that was central to completing subprojects according to the principles established in the POM. 5. The main benefits derived from subprojects can be summarized as follows: • Productive subprojects: Yield improvements allowed beneficiaries to access markets (e.g., rural fairs), thus improving income-earning opportunities. • Water and sanitation subprojects: Improved access to water through pumping systems and water-harvesting structures has allowed beneficiaries to access water for human and animal consumption and irrigate their household parcels. Other works of this kind include family bathrooms, which have improved living conditions, reduced environmental pollution, and curbed water-borne diseases. • Micro-irrigation subprojects: Irrigation has helped to optimize agricultural production and guarantee the harvest. Beneficiaries can irrigate their crops in a much simpler and more efficient way. The benefits derived from this type of investment are highly valued, to the extent that beneficiaries have organized themselves to ensure their O&M. • Road subprojects: Beneficiaries agreed that road improvements have brought better integration of communities with surrounding areas. Before the Project, communities did not have road access in the rainy season, which severely limited their mobility and ability to take their products to nearby markets. 6. The following extracts from testimonials captured through the qualitative survey instruments show satisfaction with subproject outputs: “The community’s need for household electrification has been addressed .” –Zenobio Rafael Coaquira, Community of Puca Huasi “With the subprojects financed by PICAR we have prioritized our needs and improved our living conditions in our homes and in our community.” –Sayda Solano Castillo, Community of San Francisco “Because of the animal fences we are now able to take better advantage of our time, we spend more time with our families and we even have free time.” –Adela Ajata Layme, Ayllu Llanquera Central Community 7. In conclusion, PICAR has been a vehicle for communities to achieve economic, social, and political empowerment. This assessment is based on the elements of empowerment shown in Figure A7.1 and defined in the following paragraphs. Page 57 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) Figure A7.1: Empowerment elements 8. Economic empowerment: The Project has reversed traditional inequalities in access to the means of production and unequal distribution of income between men and women. Beneficiaries have addressed these issues by improving income-earning opportunities through productive investments identified and managed exclusively by women. 9. Political empowerment: Conventional exclusive power relations have been transformed. The Project has provided resources for communities to democratically determine, with full consideration of gender equality, which subprojects to undertake to improve access to productive infrastructure and sustainable basic services. 10. Social empowerment: The Project has promoted social empowerment by developing spaces (consultations, committees) to motivate horizontal dialogue between the various actors to make their own assessments of the current state of their communities, participate in the search for solutions, mange the implementation of those solutions, and foster accountability and transparency throughout the process. Beneficiaries strengthened and formalized their organizations and achieved unity to pursue shared goals. Additional personal and sociocultural development has stemmed from participatory social management. 11. Women’s empowerment is the process by which women, in a context in which they are disadvantaged by structural gender barriers, acquire or reinforce their capacities, strategies, and leadership, both individually and collectively, to achieve an autonomous life in which they can participate, on equal terms, in access to resources, recognition, and decision-making in all spheres of personal and social life. In this regard, PICAR did not pursue women’s empowerment through workshops on gender. Instead, it facilitated spaces where women’s participation was encouraged and where women themselves analyzed and made decisions regarding their needs and the execution of community and women's projects. The new opportunities for women to learn and exercise new positions of authority in the community, such as the Project Administration and Social Control Committees, enabled women to demonstrate their capacities and work, and allowed them to be recognized in a new activity unrelated to their domestic work. Through these actions, women have developed greater confidence in themselves, and their participation is now considered and valued within their households and the wider community. Change can be perceived from the following testimonials: Page 58 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) “‘We are now responsible for our own subproject demonstrating that [we] can do other activities.” -Costa Flores Iquise, Comunidad Rancho Juaniquina “Now we talk and express our opinions more at the community meetings.” -Victoria Escobar, Kuri “Thanks to PICAR we have been able to unite as women and cooperate among ourselves. That is why we managed to increase [our] participation in the community meetings and assumed communal positions.” -Sayda Solano Castillo, San Francisco Community Summary of Findings under the Economic and Financial Analysis for the AF Methodology 1. The analysis was based on case studies of subprojects financed by the Original Credit that benefited 489 families in total. Main Findings 2. PICAR’s interventions contributed to improving the living conditions, productive capacities, and access to markets for small or subsistence farmers and rural families. For example, the improvement in yields and crop diversification through the construction of greenhouses, sheds for animals, corral fences, silos, and sprinkler irrigation systems, among structures, have also improved the opportunities for Project beneficiary families to increase incomes and food and nutritional security. These interventions can also be considered important avenues for adaptation to climate change and improved management of natural resources, given the improved availability of water and improved structures and practices for agricultural and animal production. 3. Summary of case studies: • Fences for animal production in the Ayllu Originario Sullca Uta Salla Collana Community, Corque Municipality (Oruro Department). The community prioritized the construction of fences to improve animal husbandry. The total cost of the investment was US$70,350, of which 88% was financed by the Project. The fences enabled the 45 families reached by this subproject to save, on average, 4 hours per day, which is equivalent to 8,100 daily wages per year, or a value added of US$34,020 for the community.56 The reduction in animal mortality brought about by better fencing was estimated at US$13,500 per year. • Micro-irrigation improvement/extension in the San José de Paredón Community, Tarabuco Municipality (Chuquisaca Department). In this community, 39 families chose to improve the three communal irrigation systems at an estimated cost of US$79,341, of which 21% was financed by the community. Gross production value of crops increased by 60% due to the expansion of irrigated plots and earlier planting dates, which also allowed beneficiaries to receive better prices from the surpluses obtained with irrigated agriculture. 56 Assuming an opportunity cost of US$6 per daily wage. Page 59 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) • Improved silos for grain storage in the Cruz Mayu Community, Tarvita Municipality (Chuquisaca Department). Before PICAR, this community produced wheat and maize, stored under suboptimal conditions at the household. In this community, 55 families chose to invest in two silos per family, at a cost of US$58,268, with beneficiaries contributing 15% of the cost. The storage capacity of each silo is 1,300 pounds. As a result of this investment, families reduced their post-harvest losses from 50% to only 5%. The reduction in losses enabled households to reduce the area planted to maize and wheat, giving them scope to diversify into other crops such as potatoes and vegetables. Page 60 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 8. SUBPROJECT CYCLE Source: Project Appraisal Document Page 61 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 9. GEOGRAPHIC COVERAGE, PICAR: CATEGORY A COMUNITIES REACHED Page 62 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ANNEX 10. PHOTOGRAPHIC EXAMPLES i) Productive Investments Greenhouse Beekeeping Fence for animal keeping Fence for animal keeping Page 63 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) i) Productive Investments Pen and fence for animal production Greenhouse Chicken house and fence Page 64 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) i) Productive Investments Before and after- grain silo for storage Water catchment infrastructure and greenhouse Page 65 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) ii) Water and sanitation infrastructure Before PICAR After Page 66 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) iii) Water catchment and micro-irrigation infrastructure Page 67 of 68 Official Use The World Bank BO PICAR Community Investment in Rural Areas (P107137) iv) Road infrastructure Photo credits: EMPODERAR. All photos taken during supervision and Project period (2013–20) and archived in the EMPODERAR GMIS. Page 68 of 68 Official Use