Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00004849 IMPLEMENTATION COMPLETION AND RESULTS REPORT D0570-BI ON A GRANT IN THE AMOUNT OF SDR39.1 MILLION (US$55 MILLION EQUIVALENT) TO THE REPUBLIC OF BURUNDI FOR THE COFFEE SECTOR COMPETITIVENESS PROJECT March 2022 Agriculture and Food Global Practice Africa East Region CURRENCY EQUIVALENTS (Exchange Rate Effective Dec. 27, 2021) Burundian Franc Currency Unit = (BIF) BIF 2,001 = US$1 US$ 1.398 = SDR 1 FISCAL YEAR July 1 - June 30 Regional Vice President: Hafez M. H. Ghanem Country Director: Jean-Christophe Carret Regional Director: Catherine Signe Tovey Practice Manager: Shobha Shetty Task Team Leader(s): Amadou Alassane, Benjamin Billard ICR Main Contributor (s): Irene Bomani, Jean-Claude Balcet ABBREVIATIONS AND ACRONYMS AfD Agence Française de Développement (French Development Agency) ARFIC Autorité de Régulation de la Filière Café (Burundi Coffee Regulatory Agency) BLRRP Burundi Landscape Restoration and Resilience Project CAS Country Assistance Strategy CNAC Confédération Nationale des Associations de Caféiculteurs (National Consortium of Coffee Growers Associations) COCOCA Consortium des Coopératives de Café (Consortium of Coffee Cooperatives CWS Coffee Washing Stations ESF Environmental and Social Framework FA Financing Agreement FBU Franc Burundais (Burundian Franc) FFS Farm Field School FM Financial Management FCFA Fond Commun pour les Fertilisants et les Amendements (Common Fund for Fertilizers and Amendments) FW Fully Washed GAP Good Agricultural Practice GDP Gross Domestic Product GEF Global Environmental Facility GHG Greenhouse Gas Emissions GIS Geographical Information System GoB Government of Burundi GRM Grievance Redress Mechanism ICO International Coffee Organization IDA International Development Association IFC International Finance Corporation. IFDC International Fertilizer Development Center IFR Interim Financial Report IPM Integrated Pest Management IPR Implementation Progress Report IRR Internal Rate of Return ISA Institut Supérieur d'Agriculture (Higher Institute of Agriculture) ISABU Institut des Sciences Agronomiques du Burundi (Burundi Institute of Agro. Sciences) ITAB Institut Technique Agricole (Technical Agriculture Institute) ISR Implementation Status Report ISTEEBU Institut de Statistiques et d’Études Économiques du Burundi (Insitute of Statistics and Economic Studies of Burundi) MINEAGRIE Ministère de l’Environnement, de l’Agriculture et de l’Elevage (Ministry of Environment, Agriculture and Livestock) MFBPE Ministère des Finances, du Budget et de la Planification Économique (Ministry of Finance, Budget and Economic Planning) MIS Management Information System MoU Memorandum of Understanding NAIP National Agricultural Investment Plan M&E Monitoring and Evaluation NPV Net Present Value OCIBU Office des Cultures Industrielles du Burundi ((Burundi Industrial Crops Office) ODECA Office de Développement du Café (Coffee Development Office) PADZOC Projet d’Aménagement Durable des Zones Caféicoles au Burundi PACSC Projet d’Appui à la Compétitivité du Secteur Caféier (Coffee Sector Competitiveness Project) PCU Project Coordination Unit PDO Project Development Objective PIA Project Implementing Agency PIM Project Implementation Manual PNSEB Programme National de Subvention des Engrais (National Fertilization Subsidy Program) PPD Public-Private Dialogue PPP Public-Private Partnership P-RAM Procurement Risk Assessment and Management System PRDIG Projet Régional de Développement Agricole Intégré des Grands Lacs (Regional Integrated Agriculture Development of the Great Lake Project) PRSP Poverty Reduction Strategy Paper PSC Project Steering Committee RGCC Recensement Général des Caféiers et des Caféiculteurs (General Census of Coffee Trees and Growers, 2019) SODECO Société de Déparchage et de Conditionnement (Coffee Curing and Packaging Company) SOGESTAL Société de Gestion des Stations de Lavage (Company for Managing Coffee Washing Stations) STEP Systematic Tracking of Exchanges in Procurement VC Value Chain UNIPROBA Unissons-Nous pour la Promotion des Batwas (National Organization for the Promotion of the Batwa Population USAID United States Agency for International Development TABLE OF CONTENTS DATA SHEET .......................................................................................................................... 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 6 A. CONTEXT AT APPRAISAL .........................................................................................................6 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) ..................................... 14 II. OUTCOME .................................................................................................................... 16 A. RELEVANCE OF PDOs ............................................................................................................ 16 B. ACHIEVEMENT OF THE PDO’s (EFFICACY) ............................................................................... 17 D. JUSTIFICATION OF OVERALL OUTCOME RATING .................................................................... 24 E. OTHER OUTCOMES AND IMPACTS (IF ANY) ............................................................................ 24 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 27 A. KEY FACTORS DURING PREPARATION ................................................................................... 27 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 28 A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................ 28 B. FIDUCIARY, AND ENVIRONMENTAL AND SOCIAL COMPLIANCE .............................................. 29 C. BANK PERFORMANCE ........................................................................................................... 30 V. LESSONS AND RECOMMENDATIONS ............................................................................. 32 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 35 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 51 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 54 ANNEX 4. EFFICIENCY ANALYSIS ........................................................................................... 55 ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ... 61 ANNEX 6. MAP .................................................................................................................... 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name P151869 Burundi Coffee Sector Competitiveness Project Country Financing Instrument Burundi Investment Project Financing Original EA Category Revised EA Category Partial Assessment (B) Partial Assessment (B) Organizations Borrower Implementing Agency Ministry of Finance, Budget, Cooperation and Ministry of Environment, Agriculture and Livestock Economic Development Project Development Objective (PDO) Original PDO The project development objective (PDO) is to increase coffee productivity and improve its quality among small- scale coffee growersin Burundi. PDO as stated in the legal agreement The objective of the Project is to increase coffee productivity and improve its quality among small-scale coffee growers in the Recipient's territory. Page 1 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 55,000,000 39,056,858 38,018,542 IDA-D0570 Total 55,000,000 39,056,858 38,018,542 Non-World Bank Financing 0 0 0 Borrower/Recipient 0 0 0 Local Beneficiaries 17,250,000 0 0 Total 17,250,000 0 0 Total Project Cost 72,250,000 39,056,858 38,018,542 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 17-Jun-2016 12-Oct-2016 15-Nov-2019 30-Jun-2022 30-Jun-2021 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 30-Jun-2020 27.91 Reallocation between Disbursement Categories 25-Jun-2021 37.93 Change in Loan Closing Date(s) Cancellation of Financing Reallocation between Disbursement Categories KEY RATINGS Outcome Bank Performance M&E Quality Unsatisfactory Moderately Satisfactory Modest Page 2 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 18-Sep-2016 Satisfactory Satisfactory .37 02 28-Mar-2017 Satisfactory Satisfactory 3.21 03 18-Oct-2017 Satisfactory Moderately Satisfactory 4.70 04 07-Dec-2017 Satisfactory Moderately Satisfactory 5.00 05 29-Jun-2018 Satisfactory Moderately Satisfactory 10.19 06 09-Jan-2019 Satisfactory Satisfactory 14.33 07 18-Jul-2019 Satisfactory Moderately Satisfactory 18.72 Moderately 08 30-Jan-2020 Moderately Satisfactory 22.21 Unsatisfactory Moderately Moderately 09 12-Jun-2020 25.72 Unsatisfactory Unsatisfactory Moderately 10 17-Jul-2020 Unsatisfactory 27.91 Unsatisfactory 11 30-Mar-2021 Unsatisfactory Unsatisfactory 33.67 SECTORS AND THEMES Sectors Major Sector/Sector (%) Agriculture, Fishing and Forestry 71 Agricultural Extension, Research, and Other 18 Support Activities Crops 39 Public Administration - Agriculture, Fishing & 14 Forestry Page 3 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Industry, Trade and Services 29 Agricultural markets, commercialization and 15 agri-business Public Administration - Industry, Trade and 14 Services Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Private Sector Development 100 Jobs 100 Finance 12 Finance for Development 12 Agriculture Finance 12 Urban and Rural Development 44 Rural Development 44 Rural Markets 12 Rural Infrastructure and service delivery 27 Land Administration and Management 5 Page 4 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Environment and Natural Resource Management 82 Climate change 39 Mitigation 31 Adaptation 8 Environmental Health and Pollution Management 33 Air quality management 11 Water Pollution 11 Soil Pollution 11 Renewable Natural Resources Asset Management 10 Biodiversity 5 Landscape Management 5 ADM STAFF Role At Approval At ICR Regional Vice President: Makhtar Diop Hafez M. H. Ghanem Country Director: Philippe Dongier Jean-Christophe Carret Director: Juergen Voegele Catherine Signe Tovey Practice Manager: Tijan M. Sallah Shobha Shetty Juvenal Nzambimana, Paola Benjamin Billard, Amadou Task Team Leader(s): Agostini, Chakib Jenane Alassane ICR Contributing Author: Irene Bomani Page 5 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL 1. Burundi was, and remains, one of the poorest countries in the world. It is small, landlocked, densely populated (approximately 10 million inhabitants in 2015 , mostly rural1, with a population growth rate of 3 percent annually) with a total land area of 27,834 km. Burundi at project appraisal had many challenges: (i) the human capital base was low due to limited access to social services; (ii) most youth were under-employed because of lack of opportunities; (iii) less than 4 percent of the population had access to electricity; (iv) 52 percent of children under five were chronically malnourished; and (v) 64.6 percent of Burundians lived below the national poverty line.2 Poverty was overwhelmingly rural and most of the country’s poor were small-scale farmers. The 2015 Human Development Index ranked Burundi 184 out of 188 countries.3 2. After armed conflict between 1993 and 2003, at project appraisal, Burundi had enjoyed over a decade of relative political stability, security and economic recovery. Considerable progress towards macro-economic stabilization had been made thanks to a consolidated peace process that included expanded basic social services, modernization of the economic infrastructure and institutions, and reforms to improve governance, public finance management, and the business and investment climate. As a result, the World Bank had ranked Burundi as a high economic performer for streamlining business and improving its regulatory environment three years in a row (2012, 2013 and 2014). 3. Agriculture in Burundi has been central to economic growth, employment, poverty reduction and food security, but at a 2 percent agricultural production growth rate it had not kept pace with the 3 percent population growth rate of the decade preceding appraisal. At appraisal, agriculture contributed approximately 30 percent of GDP, employed about 90 percent of Burundi’s labor force and generated 90 percent of export revenues primarily from coffee exports, but the sector had not been modernized. It remained subsistence based with about 1.2 million rural households using unreliable and inefficient technology. This had aggravated the country’s food security situation. 4. Coffee was and remains Burundi’s main traditional export crop and key for its economic growth. At appraisal, coffee production was the key source of income for more than 600,000 rural households (about 30 percent of the population). The impact of the coffee value chain extended to other sectors through backward and forward linkages (input supply, processing, transport, financial services, taxes, etc.) and hence was vital for job creation, food security and poverty reduction. Burundi had the ideal biophysical conditions4 to produce high-quality Arabica coffee and potential to export more than 50,000 tons annually during peak production years, however production averaged about 15,000 tons at the time. World coffee consumption offered a ready outlet, including expanding niche markets for quality coffee. The global market price for coffee had increased at an average annual rate of 1.9 percent during the previous 50 years, with a high demand for specialty coffee, therefore the market outlook for the next 10-15 years was excellent. Given the existing opportunities in the international market for that type of coffee, Burundi 1 only 10.6 percent of the population lived in urban areas 22013-14 Household Survey, Institut des Statistiques et Etudes Économiques du Burundi's (ISTEEBU) 3 Human Development Index (2015) 4 including elevations of 1,500-2000 meters and an average rainfall of about 1,650 mm in coffee growing areas Page 6 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) had the potential to increase its coffee sales and secure better prices. In this context, in 2014 the Government of Burundi (GoB) in consultation with all coffee sector stakeholders5 had prepared a study6 to lay the groundwork for a new Coffee Sector Development Strategy (2015-2021) that would establish an institutional and regulatory framework aligned with its longstanding commitment7 to deregulation and privatization of the sector, state disengagement and private sector promotion. Approval and adoption of the strategy in 2015 coincided with project preparation and served as the backdrop for the project design. 5. Liberalization of the coffee sector was one of the lynchpins of the project design and the institutional structures and regulatory environment to support the process were in place. The country had opened the sector to private investors and farmer associations. Private-led governance structures were emerging, and existing business networks associated with coffee production were expanding. There was a well-established processing infrastructure to support increased production of high-quality coffee – more than 200 Coffee Washing Stations (CWSs), and substantive dry milling capacity. Privatization of State entities such as the Office of Industrial Crops of Burundi (OCIBU) and the Coffee Curing and Packaging Company (SODECO) had been completed and was ongoing for the State coffee washing stations (SOGESTALs). Other critical developments included, inter alia: (i) establishment of the Burundi Coffee Regulatory Agency - ARFIC8), an inter-professional association of coffee operators (InterCafé9), and the National Consortium of Coffee Growers (CNAC)10; (ii) deregulation of de-pulping, milling and export activities; and (iii) increased participation of the private sector through construction of new private CWSs and industrial stage-Dry Mills. The project design was predicated on the continued existence of this institutional and regulatory framework as critical for project implementation (see Theory of Change - TOC). One innovative project feature was the assigning of joint implementation responsibilities to ARFIC, InterCafé and CNAC, under the oversight of MINEAGRIE. While it was too early to assess the full impacts of the coffee privatization process and market reforms at project appraisal, there was evidence of positive impact on producer prices with a more equitable transmission of benefits. 6. At the macro-level, the project was to address the key constraints of Burundi’s coffee value chain (VC): (i) governance inefficiencies due to incomplete liberalization and privatization; (ii) low productivity and declining production levels due to insufficient technical and financial support to farmers, which had resulted in low and highly fluctuating production, poor maintenance and subsequent degradation of the coffee orchards, low fertilizer and pesticide use, and prevalence of pests and diseases; (iii) limited competitiveness with a decrease in quantity and quality caused by aging orchards (28 percent of trees were over 30 years old, i.e., beyond their productive stage) and limited technical producer know-how to support quality improvements due to weak technical and managerial capacity of farmers’ associations; (iv) imbalance of primary production with the expansion of processing capacity, where coffee berries were competitively sought after with little consideration for price differentiation based on quality; (v) weak infrastructure and services in the transport sector, bearing in mind that at production coffee berries 5 government and private sector representatives, farmer and women organizations, etc. 6 The study, supported by the World Bank and USAID, was conducted under the leadership of the Ministry of Environment, Agriculture and Livestock (MINEAGRIE). 7 Since the 1980s, the GoB had been committed to deregulation and privatization of the coffee sector as part of the structural adjustment program which aimed to limit the State’s involvement in the sector. 8 A new public regulatory authority responsible for ensuring that the basic “rules of the game” for managing the sector were adequately defined, transparently implemented, and, where necessary, enforced. 9 A stakeholder-driven coordinating body that has a mandate for most sector management decisions. Members include coffee producers, wet and dry millers, roasters and exporters. 10 composed of 3,226 associations of coffee growers representing about 109,000 individual members (of which 29 percent were women). Page 7 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) need to be transported immediately to washing stations, and that, on the export side, Burundi’s landlocked status subjected it to higher transportation costs than its competing neighbors (Kenya and Tanzania), and long shipment delays through the Indian Ocean ports; and (vi) sensitive political economy considerations related to agricultural production and land tenure. In Burundi, agriculture, in particular coffee, had been used for wealth and political influence, and access to arable land had always been a sensitive issue in the countryside given the high population pressure. 7. At the micro-level, the project was to support coffee growers’ farming strategies to increase food production for greater food security, as well as to expand income generation, and rural and urban employment opportunities. To best take advantage of small farm sizes (about 0.5 ha), Burundian producers typically intensify and combine intercropping of food crops and commercial crops (coffee) with livestock activities for improved food security and income diversification. This is done during the replanting and stumping process and 2–3-year gestation period for coffee trees. Improved subsidized production packages to increase productivity were available and ready for immediate dissemination to beneficiaries at project start. These included high yielding plants for coffee replacement and seeds for associated crops, improved implements, chemical inputs, etc., and Good Agricultural Practices (GAPs) for coffee production, including sustainable land management (SLM) practices, nutrition and fertilizer management, Integrated Pest Management (IPM), pruning and harvesting. 8. The project was to be aligned with the country’s higher - level vision and objectives as reflected in Burundi’s macro-economic blueprints, such as the second Poverty Reduction Strategy Paper (PRSP II) 2012-2015, the National Agricultural Investment Plan (NAIP11, 2012-2017), the FY13-FY16 World Bank Country Assistance Strategy (CAS), World Bank Strategy for Africa, and the National Coffee Strategy (2015-2021). In addition, the Project fell under the joint World Bank-International Finance Corporation (IFC) business plan for Burundi12. In line with the vision and objectives13 of these blueprints, the project was to help the coffee sector, the country’s largest employment sector, become more competitive, sustainable and private sector-led. It would support increased coffee production to expand income-generating activities, food security and employment opportunities for coffee growers. The project was to achieve these results through multiple interventions along the coffee VC, address productivity and quality bottlenecks, and open new market opportunities. Focus was on sector liberalization, empowering small-scale coffee growers, and ensuring more equitable sharing of the benefits from improved quality and market access amongst VC actors. 9. The project design benefitted from global development lessons and World Bank analytical and operational experience in countries impacted by Fragility, Conflict and Violence (FCV). Global armed conflicts, including Burundi’s turbulent history, have demonstrated their inextricable impact on economic growth, political and social stability, the fragility of war-to-peace transitions, and the disproportionate impacts on the most vulnerable and marginalized populations, and the criticality of multi-pronged conflict sensitive development interventions. The project design incorporated key recognized elements for reducing conflict and fragility. Its focus on the coffee 11 prepared with support from the Comprehensive Africa Agriculture Development Program (CAADP) 12 During preparation, IFC provided Advisory Services for implementation of the third phase of the coffee sector privatization, including advice on financing options for coffee growers to buy shares of washing stations and development of international investors interested in entry points offered by the sector privatization. 13 PRSP II, 2012-2015 identified potential growth sectors with the highest potential impact on promoting sustainable and equitable economic growth and focused on modernizing the agricultural sector; the NAIP’s focus on: (i) raising productivity and ensuring optimal management of soil and water resources; and (ii) the strengthening of human resources capacities of national institutions and farmer organizations; CAS pillar on Improving Competitiveness; World Bank Strategy for Africa pillars on (i) ‘Competitiveness and Employment’; and (ii) ‘Vulnerability and Resilience’; Page 8 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) agriculture sub-sector as a key contributor to expanding Burundi’s economic growth, pro-poor, gender and climate sensitive investments, inclusive approaches to development, and increased private sector involvement was in line with the three - pillar approach for “conflict sensitive rural growth” 14: (i) improve macro-economic management and state capacity; (ii) strengthen communities by investing in human and physical capital; and (iii) develop the private sector in agriculture value chains. It would also align with the priority areas of the World Bank’s strategy for fragility, conflict and violence. The impacts of the re-eruption of political and social unrest following the 2015 presidential elections15 would be prevalent during project implementation and cemented the need for continued tailored conflict sensitive interventions. Theory of Change (Results Chain) 10. A key assumption for project implementation underlying the Theory of Change (ToC) was the Government’s continued strong commitment and political support to liberalize the coffee VC. The ToC’s long-term outcome is extrapolated from the Coffee Strategy as the sector blueprint at project design. Figure 1: Theory of Change Diagram for the Burundi’s Coffee Sector Competitiveness Project Burundi – Coffee Sector Competitiveness Project Activities Long-term Outputs Short-term outcomes PDO-level outcomes outcomes Outcome 1: Value chain institutions and governance strengthened Capacity building of Operational and institutional InterCafé, ARFIC and CNAC processes for VC management enhanced Social and economic Direct VC beneficiaries of Establishment of private- Legal framework for PPP dialogue empowerment of coffee VC public support increased public partnerships developed stakeholders Support to women and Women and youth trained/ sub- youth stakeholders projects implemented A professional and sustainable coffee VC Outcome 2: Coffee producers' productive capacity enhancement fairly compensating all actors and with a highly Rejuvenation of plantations Sustained coffee output, Provision of rejuvenation and recognized quality Fertilizer and insecticides maintenance of cash-flow and food Coffee productivity enhanced fertilization package and production extension of GAPs Good agriculture practices Production of improved planting Applied research operational Coffee research enhanced material and new varieties High quality coffee production increased Outcome 3 – Coffee quality improvement and market access Increased cupping capacity, CWSs Coffee VC stakeholders educated Share of specialty coffee Coffee quality improved operational management regarding quality increased improved, processing modernized, road access to CWSs enhanced Improved operational quality control procedures Critical Assumptions: 1. Strong government commitment and political will to fully Promotion of the Burundi Coffee liberalize the coffee V/C Marketing enhanced Brand, enhancement of market Development of market network 2. No major adverse climatic events intelligence and innovative and enhanced market intelligence 3. Political stability and no social unrest promotional activities 4. Good prices on international marketd 14See Background Paper “Breaking the Cycle: A Strategy for Conflict-Sensitive Rural Growth in Burundi”, World Bank, 2008. 15Large demonstrations and riots against the third term of the incumbent president led to violence, loss of property and productive assets, and displacement. Some 240,000 refugees fled to neighboring countries, mainly to Rwanda, the Democratic Republic of Congo and Tanzania. Regional and international communities expressed strong disagreement with the President’s third term and strongly urged renewed dialogue with the political opposition, the opening of private media and respect of human rights. The economy contracted by seven percent. Page 9 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Project Development Objective (PDO) 11. The PDO16 in the Project Appraisal Document (PAD) was to “increase coffee productivity and improve its quality among small-scale coffee growers in Burundi.” 12. The targeted primary beneficiaries were to be small-scale coffee growers. It was expected that at least 300,000 of these producers (from a total of about 600,000 at the national level) would directly benefit from project interventions (of which 30 percent were to be women). These beneficiaries were mostly hillside farmers with an average of 0.5 hectare of land, who cultivated about 150-250 coffee trees as an integral part of their livelihood base. The key expected benefits were to be access to improved planting materials and inputs, training and extension services and project-financed infrastructure, and increased small-holder incomes. Other beneficiaries would include the coffee industry in general and key public and private institutions overseeing the development of the coffee sector. Their benefits would include strengthened capacity; sustainable modernization of the processing and storage infrastructure of at least 30 percent of CWSs; rehabilitated rural access roads to the CWSs; and a nationwide awareness program on environmental standards for the sector, including education on labor practices and gender equality. Key Expected Outcomes and Outcome Indicators 13. The project had four PDO outcome indicators: (i) increased coffee productivity (kg/tree) in the project area; (ii) share of coffee production with AA/A/FW15+ quality (percentage); (iii) share of specialty coffee of total exports (percentage); and (iv) one core PDO indicator: direct project beneficiaries (number), of which women (percentage). 14. Intermediate results indicators for each component, as noted in the PAD and reformulated thereafter, are presented in Annex 1. 15. The project was also expected to result in a Net Greenhouse Gas (GHG) sink of 2.99 million tons of CO2 equivalent. The sink was expected to come mostly from improved practices, namely replanting of old unproductive trees, stumping of trees, introduction of good agricultural practices and systems (e.g., shaded coffee). The increase in carbon sequestration was expected to lead to other co-benefits including enhanced biodiversity and agro-ecosystem resilience. Components 16. The project was to be implemented over a period of six years with an original closing date of June 30, 202317 The project had four components (three technical and one on project management and coordination activities) as presented below. 16The PDO in the Financing Agreement is to “increase coffee productivity and improve its quality among small-scale coffee growers in the Recipient's territory”. The slight difference has no bearing on the PDO. 17 NB. There is a discrepancy in the closing date in the project documentation. In the datasheet it is June 30, 2022, and, in the PAD, and Financing Agreement it is June 30, 2023. The initial closing date was to be June 30, 2022. However, due to the political unrest following the 2015 presidential elections, project presentation to the Board was delayed, and it was agreed to postpone the closing date to June 30, 2023. However, the change was not updated in the Data Sheet. Page 10 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) 17. Component 1: Institutional strengthening and value chain governance (estimated cost at appraisal US$8.64 million, of which IDA US$8.5 million equivalent; actual cost at closing US$5.39 million) aimed to streamline and strengthen the sector’s governance structure through five sub-components. The bulk of the funds were allocated to sub-component 1.3 for capacity building of key institutions in the coffee sector. Sub-components 1.1, 1.2, 1.4 and 1.5 with a combined allocation of US2.5 million or 29 percent of C1, focused on ancillary governance and capacity-building activities. 18. Sub-Component 1.1: Streamlining of the institutional framework (US$0.69 million), focused on assessing the institutional framework and introducing competitive and growth-inducing sector regulations, preparing a sector census and an analysis of access to finance for coffee value chain actors, and reviewing the coffee pricing mechanism. 19. Sub-Component 1.2: Enhancing value chain coordination and Public-Private Dialogue (PPD) (US$0.37 million), included preparation of stakeholder mapping, establishment of a Private Public Dialogue (PPD) mechanism with training of the main stakeholders, and support for consultations among the value chain stakeholders. 20. Sub-Component 1.3: Strengthening the capacity of the key organizations overseeing the sector (US$6.11 million or 71 percent of C1; actual cost at closing US$3.49 million). The key organizations earmarked for support were ARFIC, InterCafé and CNAC. These organizations were to receive general support including training in group dynamics to promote good governance, assistance for strategic planning, business management and enterprise development, Monitoring and Evaluation (M&E), and marketing intelligence. The component was to finance technical assistance (TA) (long and short-term specialized expertise), training, forums and exchange visits (mainly South-South visits), transportation (vehicles, motorcycles, bicycles), and office IT and audio/video equipment. 21. Sub-Component 1.4: Women and youth empowerment (US$0.68 million), in recognition of the large gender gap in Burundi, included the preparation of a stand-alone comprehensive coffee gender and youth strategy and a matching grant program to promote women and youth entrepreneurial activities through sub-projects designed to support coffee productive activities, particularly those promoting land and resource preservation, climate-smart infrastructure and energy saving. 22. Sub-Component 1.5: Geo-referenced information system for the sector (US$0.79 million), was to be a tool for the coffee value chain actors for planning, M&E of VC activities, and provision of reliable information on the number of producers, number of trees/age, volume of production, coffee quality, washing stations’ performance, and other socio-economic and ecological factors. The system was to include a Geographic Information System (GIS) module to capture additional agro-ecological variables such as soil types, water sources and other factors that affect coffee quality. It would complement InterCafé’s Management Information System (MIS) on market data, market promotion initiatives, and was to be linked to ARFIC databases. 23. Component 2: Coffee growers’ productive capacity enhancement (estimated cost at appraisal US$47.04 million, of which was IDA US$31.26 million equivalent; actual cost at closing US$24.73 million) was the largest, representing 65 percent of project costs. Its objective was to enhance, on a sustainable basis, the productive capacity of small coffee growers in the targeted project areas to increase coffee production, reduce the cyclical swings of production and improve cherry quality. Low use of improved inputs (seeds, fertilizers, pesticides, etc.) and modern technologies, inadequate access to extension services, poor infrastructure, limited access to credit, and Page 11 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) marketing were among the key factors impacting the productive capacity of small-holder coffee growers. The component would provide extension services and supply inputs using InterCafé’s contractual system, MINEAGRIE’s technical services including its decentralized extension network, special programs (such as the national fertilization program), and specialized entities such as the Burundi Institute of Agronomy and Sciences - ISABU. The services of the territorial administration were to be used for awareness and outreach campaigns. Component 2 had the following four sub-components. 24. Sub-component 2.1: Rejuvenation of existing plantations (US$13.76 million) included two sub-programs: (i) uprooting of old trees and planting new ones using improved seedlings; and (ii) full stumping of existing unproductive trees. The project was to fund a rejuvenation “package” (planting material, fertilizers and small tools) -- for coffee and food crops inter-planted during the 2-3-year gestation period of the coffee trees. Farmers were to receive a subsidy for the package for three years on a declining schedule (100 percent the first year, 75 percent the second year and 50 percent the third year). The rejuvenation program, to be implemented by MINEAGRIE, InterCafé and CNAC would provide capacity building, extension support, and seedlings. The distribution of the seedlings and other elements of the package, as well as training and extension support, would be carried out by CNAC and MINEAGRIE through their network of extension staff. Communication campaigns for the rejuvenation program would be conducted annually in coordination with the territorial administration. 25. Sub-Component 2.2: Improving access to fertilizer and agro-chemicals (US$25.12 million) was the largest sub-component and had two sub-programs: (i) fertilizer input sub-program (US$19.04 million) was to help targeted farmers fertilize their fully productive coffee plantations, covering 50 percent of the country’s trees or 26 million trees. The project was to finance 40 percent of the fertilizer cost through the Government’s existing National Fertilizer Subsidy Program (PNSEB). The International Fertilizer Development Center (IFDC) would provide technical backstopping and monitoring. InterCafé and CNAC were to be strengthened to assess the needs of individual coffee growers prior to the cropping season and to provide relevant data to the PCU to order the required quantities from private agro-dealers; and (ii) pesticide treatment sub-program (US$6.08 million) was an emergency program (free for participating farmers) for the treatment of selected diseases such as Anthracnose and Antenstia that were negatively impacting coffee productivity at the time. Support would include provision of insecticides and small equipment such as sprayers. Corresponding extension advice was to be provided by CNAC and MINEAGRIE. 26. Sub-components 2.3: Generation and dissemination of sustainable agricultural practices (US$5.41 million) and 2.4: Applied coffee research (US$2.75 million). SC 2.3 would support investments to promote the adoption of GAPs for coffee production, including Sustainable Land Management (SLM) practices, nutrition and fertilizer management, Integrated Pest Management (IPM), pruning, harvesting, quality assessment and improvement, and best practices for intercropping and coffee under shade. It would also support the associated capacity-building of staff of public and private agricultural extension and research services. SC 2.4, was to implement a coffee research program, in cooperation with InterCafé, ISABU and academia to target issues such as coffee productivity, production fluctuation and quality (notably the ‘potato-taste’18). The program would achieve these objectives through the development of: (i) new varieties, including hybrid varieties that were more productive and resistant to diseases; (ii) fertilization formulas per agro-ecological zones; and (iii) methods for integrated management of the main coffee diseases and pests. 18The ‘potato taste’ occurs in coffees from Rwanda, Western Uganda, Burundi, and the Democratic Republic of Congo. It smells and tastes exactly as the name implies “raw potatoes” and needs to be eradicated if coffee is to keep its good quality. Page 12 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) 27. Component 3: Coffee quality improvement and market access (US$9.4 million, including an IDA contribution of US$8.08 million equivalent; actual cost at closing US$3.96 million) would provide TA, training, infrastructure and goods along the entire value chain (field, washing stations, dry mills, quality control laboratories, storage, etc.), and promote the Burundi coffee brand for increased market access. This was to be achieved through the following two inter-related sub-components. 28. Sub-Component 3.1: Coffee quality improvement (US$6.21 million) focused on educating value chain actors on quality issues and building a national testing (or cupping) capacity. The following investments were planned: (i) developing cupping capacity including a training program for “cuppers”, establishment of cupping laboratories in the main producing areas (within selected leader CWSs and dry mills19) and certification of “cuppers” by the Coffee Quality Institute (CQI) or other internationally-recognized institutions; (ii) improving CWS technical, operational and managerial capacities required to produce high quality coffee would assist the managerial staff to develop and implement customized Quality Improvement Plans (QIPs), and support training programs for CWS staff with a focus on financial, administrative and management topics, organization of cooperatives, and strategic business planning; and (iii) sustainable modernization of processing facilities would implement a Public-Private Partnership (PPP) program to facilitate access to eco-friendly technologies to reduce environmental degradation within the CWSs and their surroundings. The project also planned to rehabilitate 60 km of rural roads to improve coffee growers’ access to the CWSs and marketing facilities in project intervention areas and prepare a review of the existing rural roads in the vicinity of CWSs. 29. Sub-Component 3.2: Marketing and promoting the Burundi Coffee brand (US$3.19 million) would support: (i) building a recognizable image for Burundi coffee, including, a country logo, brochures, standardized washing station fact sheets, coffee maps, video clips, and other information that would be useful to coffee markets. The project would build on the InterCafé internet platform (www.cafeduburundi.com) developed under the USAID’s Burundi Agribusiness Project (BAP 2007-2012) to maximize the availability of coffee sector data (on production, sales and marketing, regulations, and sector developments) for use by the value chain actors and potential buyers; (ii) developing a reliable coffee industry intelligence and forecasting software-based system at InterCafé (linked to the existing coffee database and internet platform) to analyze key market information (daily world market prices for differing grades of coffee and the official US$/FBU exchange rate), increase transparency and facilitate free circulation of information; and (iii) introducing other innovative promotional activities such as buyer tours, coffee competitions, participation in trade and marketing events, and many other features that would place Burundi’s coffees in some of the very best markets in the world. 30. Component 4: Coordination, monitoring and knowledge management (US$7.16 million IDA; actual cost at closing US$4.47 million). This component supported: (i) project administration, technical, and financial management; (ii) coordination among all institutional partners to ensure an efficient flow of information and support to all value chain actors, in particular the small coffee growers; (iii) effective contractual arrangements with key official implementing partners and private sector operators; (iv) monitoring and evaluation of project performance and financial, environmental, and social impact management; and (v) communication activities to publicize and disseminate project results, best practices and success stories. 19 Burundi’s CWSs use water to remove the pulp of the cherries (‘washed process’). Once pulping is done, coffee beans are immediately dried to prevent spoilage. This is done by placing the beans on large beds/ clays exposed to sunshine. Subsequently, the job is to hull, sort and bag the green beans for shipment and export. This is done at dry mills privately owned or run by farmers cooperatives. Page 13 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) Revised PDO and Outcome Targets 31. The PDO and Outcome targets were not revised during implementation. Revised PDO Indicators 32. PDO indicators remained unchanged during the entire project duration. Targets were not revised following the Mid-Term Review (MTR) in November 2019 and the institutional and regulatory reforms decreed by the GoB (January 2020.) Revised Components 33. Components remained unchanged during the entire project duration. Other Changes 34. Change at MTR (November 2019). The major change was the World Bank’s and GoB’s decision to increase the level of fertilizer subsidies for producers under Component 2. This was a substantial change as it concerned the all- important Fertilizer Input Sub-Program (US$19 million --- 26 percent of total project cost). The subsidy was increased from 40 percent at design stage to 80 percent20 to stimulate low producer demand for project procured fertilizer deemed too costly by producers. Another reason was to make it more attractive than the PNSEB-procured fertilizer for food crops that carried a subsidy of 40 percent. As intended, the increase significantly raised producer demand for the project-procured fertilizer and all procured quantities were distributed before the project cancellation as programmed through the national PNSEB initiative. 35. Return of the coffee sector to State control under the January 2020 reforms. The single major change that derailed project implementation at mid-course shortly following the MTR and ultimately contributed to the eventual early closure of the project was the GoB’s unilateral decision to revert the coffee value chain back to State control. In January 2020, without advance warning or consultations with the World Bank and other donors, the GoB changed the agreed-upon institutional arrangements for the coffee sector21. ARFIC, Inter-Café and CNAC as the legally recognized Project Implementing Agencies (PIAs) per the FA, were replaced by a newly created Office for Coffee Development (ODECA), a State-controlled entity with a mandate to oversee the coffee value chain. 20 The PNSEB fertilizer subsidy (financed by the National Budget and other sources such as the Projet d’Appui à la Production Agricole-PAPAB) was 40 percent. PAPAB (2015-2019) funded by the Dutch Cooperation promoted market-oriented, climate-resilient and sustainable agricultural techniques through targeted fertilizer subsidies. 865,000 farmers enrolled in PNSEB in 2019. About 48% of Burundian farming households had access to fertilizers through PNSEB in 2019. Yearly quantities of fertilizer procured through the PAPAB program increased from 29,000 tons (2016) to 49,000 tons (2019). 21 Through legal measures: (i) Decree No. 100-001 dated January 7, 2020, created the Office of Coffee Development (ODECA), and defined its organization, mission, and financing mechanisms, and (ii) Decree No. 100-002 dated January 7, 2020, and revised Decree No. 100-12 dated January 14, 2005, which reformed Burundi’s coffee sector. Page 14 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) 36. Level 2 restructuring (June 2020). The Financing Agreement was amended on June 30, 2020, to reallocate US$10.97 million across disbursement categories. US$9.43 was reallocated from category 122 to categories 423 and 624. US$21,239 was reallocated from category 525 to 6. US$960,000 was reallocated from category 726 to 227. US$563,000 was reallocated from category 928 to categories 2, 329, and 6. The objective was to provide more resources to: (i) Component 1 for extension activities to the Department of Agricultural Extension (DGMAVAE) that had taken over extension activities following the institutional re-arrangements; and (ii) Component 2 to ISABU for implementation of additional research and demonstration plots deemed necessary to accompany the coffee rejuvenation programs. 37. Level 2 restructuring (June 2021). After numerous and lengthy discussions on the central impact of the 2020 reversal of the liberalization policy on development of the coffee sector, as well as the project’s institutional arrangements and regulatory framework which no longer conformed to the FA, and an impasse on the optimum way forward, the World Bank and GoB agreed to a level 2 restructuring30. The PDO was officially deemed no longer achievable. The project closing date was advanced by two years from the original June 30, 2023, to June 30, 2021. It was envisaged that: (i) component 2 activities under the coffee replanting, rejuvenation and fertilization programs and the related undisbursed funds would be transferred to the World Bank-financed Landscape Restoration and Resilience Project (LRRP, P160613); (ii) during the 4-month grace period following project closure (July 1 to September 30, 2021) PIU operating costs to oversee the project closing process (including processing of final payments for eligible activities completed by the new closing date, final project audit) would be funded by the ongoing World Bank-financed Great Lakes Regional Integrated Agriculture Project (PRDAIGL- P161781); and (iii) PRDAIGL would also provide the remaining funding for the academic training of technical staff (6 masters and 2 PhD level) who were in Kenya and had not yet finished their training. However, whilst support under (iii) was provided31, it proved impossible to transfer the funds for the coffee improvement programs to the LRRP and for the grace period PIU operating costs to the PRDAIGL. As a result, the undisbursed project funds were cancelled32. It is to be noted that 22 Goods, works, non-consulting services, and consultants’ services, Operating Costs, and Training for Parts 1(a); 1(b)(i) and (ii); 1(c)(i)(A) and (D); 1(c)(ii)(A), (G) and (H); 1(c)(iii)(A), (D), (E), (F), and (G); 1(d)(i); 2(a)(i); 2(d); 3(a)(i)(A) and (C); 3(a)(ii); 3(a)(iv); 3(b)(ii); and 4 of the Project 23 Goods, works, non-consulting services, and consultants’ services, Operating Costs, and Training for Parts 1(c)(iii)(B) and (C); 2(a)(ii); 2(b)(ii); and 2(c) the Project 24 CWS Matching Grants under Part 3(a)(iii) of the Project 25 Women and Youth Matching Grants under Part 1(d)(ii) of the Project 26 Fertilizer Subsidies under Part 2(b)(i) of the Project 27 Goods, works, non-consulting services, and consultants’ services, Operating Costs, and Training for Parts 1(b)(iii); 1(c)(i)(B) and (C); 1(e); 3(a)(i)(B); 3(b)(i), and 3(b)(iii) of the Project 28 Goods, works, non-consulting services, and consultants’ services, Operating Costs, and Training for Parts 1(a); (b)(i) and (ii); 1(c)(i)(A)and (D); 1(c)(ii)(A), (G) and (H); 1(c)(iii)(A), (B), (C), (D), (E), (F), and (G); 1(d)(i), 1(d)(ii); 2(a)(i)(ii); 2(b)(ii)(c)(d); 3(a)(i)(A) and (C); 3(a)(ii); 3(a)(iv); 3(b)(ii); and 4 of the Project 29 Goods, works, non-consulting services, and consultants’ services, Operating Costs, and Training for Parts 1(c)(ii)(B), (C), (D), (E), and (F) of the Project 30 A formal letter dated April 7, 2021, was sent by the Government requesting the Bank to proceed with the cancellation of funds and next steps. The GoB sent another letter dated June 11, 2021, to confirm to the WB the undisbursed IDA amount to cancel along with a detailed table for each disbursement category. 31 This was made effective as part of the PRDAIGL restructuring. It was doable since IITA (providing support and supervision of the students) and ISABU were involved in both PACSC and PRDAIGL, and support to curricula contributed to the PDO regional integration part of PRDAIGL as well as generated knowledge representing a regional public good. 32 World Bank confirmation of cancellation of project funding in the amount of about SDR 11.2 million (equivalent to about US$19.2 million) was sent to the GoB in a letter dated June 30, 2021. Page 15 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) the World Bank Restructuring Paper (Report No. RES47141) did not include a change in the Results Framework (RF) targets. 38. Extension of the Disbursement Date Deadline (DDD). To accommodate the payment of the final project accounting and financial audit and other miscellaneous expenditures, the project DDD was extended by two months from October 29 to December 29, 2021. This was necessary to allow completion of the final audit and processing of some remaining payments that were still pending at closing. Both have been completed as of the finalization of this report. Rationale for Changes and their Implication on the Original Theory of Change 39. The GoB’s January 2020 institutional and regulatory reforms had major implications for the TOC. The TOC was based on continued liberalization and privatization of the coffee sector as a critical assumption for project implementation. Legally, the GoB had committed to using ARFIC, InterCafé and CNAC as implementing agencies with project resources directly allocated to these entities. The subsequent transfer of project implementation responsibilities to another entity ran counter to the legal underpinnings of the ToC. II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDO and Rating 40. At preparation, the project was fittingly designed to address the key factors governing Burundi’s coffee value chain. It emphasized support to the value chain governance and required activities for orchard rejuvenation and enhanced quality, with increase participation of all actors, notably smallholder producers and private sector stakeholders for processing and marketing along the value chain. The project would not simply repeat earlier coffee support projects. It was conceived as an integrated operation focused on ensuring the long – term sustainability of the coffee sector by promoting value chain liberalization and policy dialogue and implementing the required institutional reforms to ensure participation of all actors. To achieve this, the project focused on the consolidation of coffee reforms as envisioned in the National Coffee Strategy (2015-21) and the institutionalization of associated mechanisms and organizational structures to ensure sustainable and strong support to participatory processes in the value chain, including support for the flagship institutions of ARFIC, InterCafé and CNAC. However, the GoB’s January 2020 institutional and regulatory reforms midway through project implementation negated the agreed-on sector liberalization process. The course change had far-reaching and significant negative consequences. It impacted achievement of the PDO and led to early project closure two years ahead of the initial closing date. In addition, project activities under components 1 and 2 previously under the joint responsibility of ARFIC, InterCafé and CNCA, were suspended as of the MTR and thereafter were not continued under the GoB’s new institutional implementation arrangements. Hence, the results linked to Outcomes 1 and 3 are based exclusively on the three years of implementation prior to the MTR. 41. At project design, the PDO was aligned with the three-pronged goals of the CAS (FY13-FY16), i.e.: (i) improving governance and service delivery; (ii) stimulating growth and economic diversification; and (iii) strengthening human capital. At MTR, the PDO remained highly relevant to the Country Partnership Framework (CPF) (FY19-FY23), specifically its two priority areas: (i) strengthening human capital; and (ii) strengthening the fundamental foundations Page 16 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) for economic and social resilience. The project was also fully aligned with the: (i) WBG Strategy for Africa (2019-23) that aimed to create sustainable and inclusive growth, strengthen human capital and build resilience to fragility and climate change; and (ii) PRSP II (2012-2015), Burundi’s major blueprint for poverty reduction and economic development at the time of project design. While the PDO remained relevant, the institutional and regulatory reforms distorted the agreed-on strategic orientations for the coffee sector and changed the project implementation modality which hindered achievement of the PDO. Ultimately, the PDO could no long respond to the higher objectives assigned to the project. Rating of Relevance of PDO Rating: “Modest” 42. The relevance of the PDO is rated as “Modest”. At project design and during the first half of project implementation, the project provided evidence of alignment of the PDO to the CAS, CPF, National Coffee Strategy objectives, and the GoB’s poverty reduction goals. However, the drastic change in the coffee sector institutional and regulatory framework mid-course in project implementation impeded achievement of the PDO. B. ACHIEVEMENT OF THE PDO’s (EFFICACY) Assessment of Achievement of Each Objective/Outcome 43. The achievement of the PDO objectives and outcome targets is summarized in Table 1 below. The presentation conforms to the ToC presented earlier. The assessment is made against the core outcome of “number of beneficiaries”, and against the three outcomes corresponding to the three project components, i.e.: (i) Outcome 1: Value chain institutions and governance strengthened; (ii) Outcome 2: Coffee growers productivity capacity enhanced; and (iii) Outcome 3: Coffee quality and market access improved. Since the RF was not changed during the June 2021 restructuring, the present project efficacy analysis is based on the original RF end-targets expected after the six-year implementation period (original closing date of June 30, 2023) specified in the PAD. Page 17 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Table 1: PDO-Level Results Indicators (Baseline, Target and Achievement) as of June 30, 2021 Achievement Rate of Target PDO-Level Results Indicators Baseline Target (June 2021) Achievement PDO Core Outcome: Direct beneficiaries reached by the project • OIC – Direct beneficiaries (disaggregated by gender) (Number/ 300,000 313,981 105% - percentage) (30%) (15.6%) (52%) PDO Outcome 1: Institutional strengthening and value chain governance Intermediate Results linked to Outcome 1 • IR 1.1 – Coffee Act adopted by the Ministry of Agriculture and - Yes No 0% Livestock (Yes/No) • IR 1.2 – Coffee cooperatives legally registered (percentage) 31% 70% 77% 118% • IR 1.3 – Coffee pricing formula accepted by InterCafé, CNAC and - Yes No 0% ARFIC (Yes/No) • IR 1.4 – Coffee sector MIS in place and operational - Yes No 0% • IR 1.5 – Beneficiary satisfaction with project services (including - 70% 97% 138% CNAC, InterCafé) (percentage) PDO Outcome 2: Coffee growers productivity capacity enhancement • OI 2.0 – Increased coffee productivity in the project area (kg/tree) 0.8 1.6 1.2 50% Intermediate Results linked to Outcome 2 • IR 2.1 – Coffee trees replanted (million) - 16.50 16.06 97% • IR 2.2 - Coffee trees rejuvenated through full stumping (million) - 18.75 18.92 101% • IR 2.3 – Total area established with improved intercropping - 7,050 6,498 92% systems (ha) • IR 2.4 – Clients who have adopted an improved agricultural - 90,000 97,700 109% technology promoted by the project (number) PDO Outcome 2: Coffee quality improvement and market access • OI 3.1 - Share of coffee production with AA/A/FW1+ quality 59% 69% 66% 70% (percentage) • OI 3.2 – Share of specialty coffee of total exports (percentage) 7.5% 15% - - Intermediate Results linked to Outcome 3 • IR 3.1 – Number of new cupping labs operational in major - 7 0 0% production areas (number) • IR 3.2 – CWSs and dry mills equipped with eco-friendly tech/ - 56 24 43% infrastructure and/or upgraded storage facilities (number) • IR 3.3 – Additional man/days offered by CWSs (by gender) - 240,000 0 0% (number) • IR 3.4 – Rural roads rehabilitated (km) - 60 33 55% • IR 3.5 – Coffee export procedure processing time (ARFIC) (days) 10 3 - - Core PDO Outcome: “Number of direct project beneficiaries” 44. In the project efficacy assessment, the number of direct project beneficiaries is the only indicator retained as a separate ‘core’ metric of project performance (Outcome Indicator Core- OIC). As indicated in the RF, this indicator is a global ‘core’ indicator predicated on full project implementation. It reflects the project impact on all the coffee sector stakeholders, based on the results achieved in completing all three components, bearing in mind that most of the primary beneficiaries were the small producers under the rejuvenation/ fertilization program. At closing, the Page 18 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) project had reached 313,981 people all along the value chain or 105 percent of the final project target. The fact that more beneficiaries were reached than the set target, even though the project closed earlier than planned, is testimony that the project was successful in reaching out to producers, in particular under Component 2’s rejuvenation/ fertilization program. On the other hand, the project achieved half of the final target (15.6 percent of the end target of 30 percent) for women which points to a highly skewed result in terms of gender outcome. This result seems to suggest the original target of women to be reached may have been overestimated, especially in light of the coffee census undertaken in 202033 as part of project implementation which revealed that on average women- headed coffee households comprised only about 20 percent of households. PDO Outcome 1: Value chain institutions and governance strengthened 45. Regarding Intermediate Results (IRs) linked to Outcome 1 (OI 1.0), the project achieved one positive result regarding registration of coffee cooperatives (IR 1.2). It achieved 77 percent of the rate of registration of cooperatives or 118 percent in incremental terms over the reference situation34. The project also produced significant outputs: (i) provision of training and technical assistance (both short term and long term), and transportation means (vehicles, motorcycles and bicycles), office IT and audio/video equipment to ARFIC, InterCafé, CNAC, UNIPROBA the National Association in charge of the defense of Batwas’35 interests), and MINEAGRIE36; (ii) partial renovation of ARFIC’s buildings intended to serve as the one-stop window for coffee export procedures (this window was never established), as well as the renovation and expansion of ISABU’s laboratories in the main coffee growing areas; and (iii) implementation of 28 sub-projects for vulnerable populations (women, youth and Batwas). In contrast, following the January 2020 coffee sector institutional and regulatory reforms, MINEAGRIE was unable to prepare the Coffee Act and regulatory documents (IR 1.1) that would have preserved the liberalization reforms agreed under the 2015 Coffee Strategy. Similarly, since project implementation responsibility was removed from them, ARFIC, InterCafé and CNCA were no longer able to achieve IR 1.3 and 1.4 related to the update of the Coffee Pricing Formula and implementation of a sector MIS. Regarding IR 1.4, although the coffee census including the physical identification of coffee trees and producers was prepared by the National Institute of Statistics (ISTEEBU) as mandated under project implementation, the MIS based on the geo-tagging of coffee areas using satellite imagery was not completed. Although a few of the expected outputs under Component 1 could not be met as planned, the degree of beneficiary satisfaction with project services (IR 1.5) was high (97 percent), indicating that project activities responded to stakeholders’ needs and were well received. 46. The overall achievement for Outcome 1 is rated as “Modest”. Results under Outcome 1 as reflected in the RF are to be attributed primarily to activities under Component 1, and partially to activities under Components 2 and 3. The high degree of beneficiary satisfaction takes into account the results attributed to Outcomes 2 and 3 as much as those of Outcome 1. Achievement of the other indicators that can strictly be attributed to Outcome 1 is low, including the failure to prepare and enact the all-important Coffee Act and to develop the MIS system. The fact that the project only partly achieved its objectives under Outcome 1 is due, in large part to the January 2020 institutional and regulatory reforms and the GoB halting the sector liberalization reforms agreed under the project. 33 See National Coffee Census, ISTEEBU, Feb. 2021 34 Based on the Coffee Census (Feb. 2021), only about 3.8 percent of producers belong to a coffee cooperative, and 21 percent to a coffee producer association. These figures are very low in comparison with other coffee producing countries. 35 The Batwas are the original indigenous ethnic group occupying the national territory of Burundi. Over the years they have been marginalized as far as development efforts are concerned. 36 See Annex 4. Detailed disbursements for Sub-Component 1.4 are: ARFIC: US$1.27 million; InterCafé : US$0.56; CNAC: US$1.07; UNIPROBA: US$0.14 million; and MINEAGRIE (including DGMAVAE/BPEAEs): US$0.46 million. Page 19 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) PDO Outcome 2: Coffee growers productivity capacity enhanced 47. Project implementation under Component 2 proceeded as planned until the MTR. Thereafter, ad hoc arrangements were agreed on to continue project implementation directly between the PIU and MINEAGRIE’s operating structures without AFRIC’s, InterCafé’s and CNCA’s involvement. Implementation continued through ISABU for the supply of planting material, and DGMAVEA for outreach to producers in collaboration with IITA and UNIPROBA. 48. The all-important result under Outcome 2 (OI 2.0) is the increase in coffee productivity at farmer level. This is the core metric of success in project implementation in terms of increase of overall quantities produced each year and reduction of cyclical production levels over a period of years37. The increase in coffee productivity achieved was 50 percent of the final project target (increase of 0.4 kg/tree vs. 0.8 kg per tree expected). This result, however, should be qualified. The years 2019-2020 and 2020-2021 were ‘off-production’ years in terms of quantities produced in Burundi38. In addition, given the reduced project timeframe, the trees rejuvenated and replanted under the project had not yet had sufficient gestation time to reach full production at project closure. 49. The weak results for Outcome 2 on coffee productivity are to be contrasted with the substantial degree of achievement of the Intermediate Results linked to that outcome, corroborating the hypothesis that at project closure the trees which had been rejuvenated and fertilized had not yet reached full production. Over the shortened project implementation period, the project managed to get close to or surpass its targets on Intermediate Results linked to Outcome 2, with: (i) 97 percent of final target for trees replanted (IR 2.1); (ii) 101 percent for coffee trees rejuvenated (fully stumped) (IR 2.2); (iii) 92 percent of the targeted area under improved intercropping systems (IR 2.3) (coffee interplanted with common beans and/or soya beans); and (iv) 109 percent of beneficiaries having adopted the improved technologies supported by the project (IR 2.4). For the above purpose, 3,800 tons of NPK-S fertilizer39 and 307,000 tons of organic manure were provided by the project (see Annex 1-C). In addition: (i) the coffee shade growing system was introduced on 72 sites; and (ii) the ISABU’s laboratories for phytopathology and biotechnology, and the buildings for the training and documentation center were constructed and/or renovated and the required equipment was procured. 50. The above results are substantial since only 53 percent of the funds were disbursed under Component 2 (see Annex 3). It can be reasonably assumed that the project would have exceeded its final targets for the number of trees replanted and fully stumped, and area fertilized, if AFRIC, InterCafé and CNAC had continued project implementation responsibility after the MTR, if the original project duration of six years had been maintained, and if the available funding had been fully utilized. The question remains as to the eventual impact of the rejuvenation and enhanced fertilization programs on coffee yields. There was clearly insufficient lead time before project closure for the coffee replanting and rejuvenation programs to have full impact on coffee productivity. This said, there is no assurance that given two more years of implementation the target of 1.6 kg/tree could have been achieved. 37 Bearing in mind that there has been no area expansion under coffee in recent years given land scarcity, pressure on land for food crops and deterioration of soils. 38 Overall Burundi coffee production was 14,580 tons of green coffee in 2019 (Coffee Census 2021) and 11,700 tons in 2020 (data provided by MINEAGRIE, reported by “Further Africa” August 2020) as opposed to around 20,000 tons on average in the years preceding the project. This is due to decreased rainfall. During the decades preceding the Coffee Relaunch Strategy (2015) production averaged about 20,000 tons; but had been trending down with increased ‘cyclicity’ across years, due to the over ageing of the national coffee orchard. 39 vs. 10,500 tons or 40 percent of quantities estimated at project design. Organic fertilizer (manure and compost) was not envisaged at project design but was introduced during project implementation (NB. Animal manure and compost are widely marketed in rural areas of Burundi and highly appreciated by producers). Page 20 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) 51. In view of the above, the achievement of Outcome 2 is rated as ‘modest’ since the project, although it had substantial Intermediate Results, fell short of achieving the prominent target on productivity increase. PDO Outcome 3: Coffee quality and market access improved 52. PDO Outcome 3 was to be measured by two indicators based on activities under Component 3 which stagnated after the MTR. The first PDO Outcome 3 (OI 3.1) on the share of coffee production with AA/A/FW+ grade rose from 59 to 66 percent (70 percent of the target in incremental terms over the reference situation). This achievement is notable given the shortened project duration. There was no information on the second outcome indicator (OI 3.2: share of specialty coffee in total exports) after the removal of ARFIC, originally tasked with providing the data for this indicator. The new ODECA structure failed to provide any data. 53. The modest results for PDO Outcome Indicator OI 3.1 are further explained by the partial achievement of targets under Intermediate Results. After the changes in institutional implementation arrangements and reduced project timeframe, all intermediary results exhibited fractional achievement or no achievement at all, i.e.: (i) only 33 km of rural roads were built (out of a target of 60 km) or 55 percent of achievement (IR 3.4); (ii) no cupping labs were completed (IR 3.1); (iii) only 24 investment sub-projects for upgrading the environmental facilities and storage of CWSs and dry mills (IR 3.2) were implemented (43 percent of target), and no data on additional employment generated were provided by the CWSs (IR 3.3); one reason for the lack of appetite for CWS upgrade sub-projects is that the initial project subsidy (40 percent at project design) was insufficiently attractive; and (iv) ARFIC was unable to provide data on the timeframe for completing coffee export procedures; in this regard, despite the fact that ARFIC facilities were upgraded, the one stop window was never formally established, and no data were available on whether or not processing time for export procedures had been reduced from the referenced 10 days to the project target of 3 days (IR 3.5). 54. Based on the above, the achievement of Outcome 3 is rated as “Modest” as the project only partly achieved its objectives under Component 3. Justification of Overall Efficacy Rating Rating: “Modest” 55. The overall project efficacy is rated as ‘modest’. Most of the results indicators linked to PDO Outcomes 1 and 3 are assessed as being “under achieved” or “not achieved”. This is in contrast with the positive results indicators for the number of beneficiaries’ satisfaction, and Outcome 2 in terms of the progress of the rejuvenation/ fertilization program. The limited results under Outcome 1 and 3 following the institutional implementation arrangement changes impacted the entire project and were the key reasons for early project closure. In addition, activities under Component 2, which were on track at project closure and needed to continue to yield their full impact, could not be transferred to another World Bank financed project as envisaged during the June 2021 restructuring due to internal World Bank procedural challenges. Although it is yet to be supported by statistical data, coffee productivity was bound to increase substantially if the planting/rejuvenation/fertilization programs were to be continued. 56. The dismantling of the original institutional set-up of the coffee value chain and the domino effects on project implementation is the key rationale for the project efficacy rating of “modest”. As indicated in the ToC, maintaining the institutional set-up as mandated under the Coffee Strategy and agreed on in the FA was the cornerstone for Page 21 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) project implementation. It was also the building block for the liberalization of the value chain as agreed between the GoB and development partners. The low rating of Outcomes 1 and 3 because of non-pursuit of the coffee VC liberalization process, significantly diminished all positive developments made under Outcome 2. Hence there is a risk that the PDO in terms of productivity increase may fail to be achieved and pose a corresponding high risk to the long-term project development outcome. C. EFFICIENCY 57. Project benefits. An ex-post economic and financial analysis was conducted as part of this report. Following the methodology used at appraisal, the analysis is based on the quantified benefits generated from the increased production of coffee trees, as well as the production of the food crops (e.g., red beans, soybeans and bananas) that were inter-planted with coffee trees under the rejuvenation program. The benefits (decrease in production losses and enhanced coffee quality) indirectly capture the impact of the upgrade of the CWSs and dry milling infrastructure, and improved access roads. They also capture improvements in marketing through price gains associated with improved coffee quality. As done at appraisal, the benefits derived from capacity building of producers and other targeted actors at various levels of the value chains were not included. However, these benefits are important especially for the poorest and most vulnerable farmers, since these groups were expected to be better equipped to produce and market coffee efficiently, and, in turn, improve their economic status as a result of project implementation. This applies particularly to women who represented about 20 percent of project beneficiaries at project closing. The project benefits computed ex-post, as in the ex-ante case at appraisal, are therefore conservative. 58. Project financial returns. At closing, the project had provided benefits and was expected to yield future benefits to coffee growers, mainly through the trees replanted and fully stumped under the coffee rejuvenation and access to inputs programs that were meant to result in higher coffee yields. These programs were subsidized as follows: (i) rejuvenation program: participating farmers received a subsidy for coffee replanting and stumping (a package of seedlings, fertilizers and tools) for three years on a declining schedule (100 percent the first year, 75 percent the second year and 50 percent the third year), and (ii) program of facilitation of access to inputs: fertilizers and pesticide treatment for diseases and pests (e.g., Anthracnose, Antestia); farmers received a subsidy for fertilizer initially to the tune of 40 percent and increased to 80 percent at MTR. The farmer’s gross margin at full production was computed in the ‘with’ and ‘without’ subsidy situation for two typical models of coffee with food crops (beans and soybeans) and coffee in mono-cropping under full fertilization. The margin without subsidy yielded an ex-ante financial internal rate of return of 61 percent and 76 percent respectively, clearly higher in the scenario with subsidy (123 percent and 117 respectively). Ex-post this rate of return was estimated at 92 and 96 percent without subsidy and 135 and 126 percent with subsidy, due essentially to the drastic increase in crop yields. This is testimony that coffee growing following rejuvenation and fertilization under the project compared favorably with ex-ante estimate, and that the activity is a financially attractive proposition including under the scenario without subsidy. Even if it was not necessary from a strictly financial internal return perspective, the subsidy was justified in that it enabled producers to cope with their negative cash-flow during the gestation period where replanted or stumped coffee trees were not generating any amount of production. 59. Economic profitability measures. Table 2 below presents a comparison of the results of the overall profitability measures computed at project design (ex-ante) and at project closure (ex-post). The project economic benefits computed over a 15-year period of analysis were expected ex-ante to yield a NPV of US$19.3 million and generate an economic rate of return of 22.7 percent. Based on ex-post estimates at project closure, the project economic Page 22 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) benefits were expected to yield a NPV of US$ 2.3 million and generate an economic rate of return of 6.9 percent. These results ex-post are marginally positive and denote a modest overall efficiency measure for the project. 60. Sensitivity analysis. In the ex-ante situation, the sensitivity analysis showed that projected benefits were robust as far as the gross margin was concerned. If the gross margin were to drop by 30 percent, either from price or quantity produced, the IRR was expected to drop to 12.4 percent. In the ex-post situation, the IRR is expected to drop to 1.6 percent if benefits are reduced by 30 percent. This means that there is a high risk that the project impact would not be positive as far as economic profitability is concerned, and, hence, overall efficiency would be jeopardized. This is not an unreasonable scenario given the volatility of the price of coffee on the international market. The sensitivity to price means that every effort should continue to be made to facilitate production collection, processing and marketing, as well as partnerships between farmers, processors and offtakers, to produce quality coffee that commands a higher and more stable price. The sensitivity to production (yield) level implies that continued close support should be given to producers so that they adhere strictly to technology prescriptions and adopt the new technologies developed by research. Every effort should be made to ensure that input delivery services are readily available. Table 2: Comparison of ex-post vs. ex-ante Project Economic Benefits Ex-Ante Ex-Post Profitability Measures % • IRR basic scenario 22.7 6.9 • IRR 30% decrease in benefits 12.4 1.6 US$ million • NPV 19.3 2.3 61. Project fiscal impact and foreign exchange earnings. The project was expected to generate US$6.3 million in taxes for the Government’s Treasury and US$51.7 million in foreign exchange earnings from coffee sales during its implementation period. The ex-post analysis reveals that only US4.1 million was generated in taxes or 65 percent of the ex-ante estimate, and US$22.6 million in foreign exchange earnings or 44 percent of the ex-ante estimate. 62. GHG Accounting. Burundi’s climate is tropical, moist with low activity clay soils. The computing modules used under the FAO EX-ACT model include land use change, crop production and inputs, with a project actual implementation phase of four years and capitalization phase of 20 years. The net carbon balance quantifies GHGs emitted or sequestered as a result of the project compared to the without project scenario. Over the analysis period of 20 years, the project generated gross GHG emissions -1,457,416 tCO2-eq, constituting a net GHG emissions (a carbon sink) of -1.52 tCO2-eq. On that basis, the project provided a sink of -63,493 tCO2-eq per year, equivalent to - 1.2 tCO2-eq per ha per year. The sink resulted largely from improved practices, namely, rejuvenation of old unproductive trees, stumping of old trees and shaded coffee practices. The above results come short of what was estimated ex-ante, i.e., a net GHG sink of 2.99 million tons of CO2 equivalent, corresponding to 4.0 tons of CO2 equivalents per hectare per year. These results were to be expected since the rejuvenation/replanting program was cut short by two years due to the project cancellation. 63. Administrative efficiency. At the beginning, the project benefitted from the administrative set-up of the then ongoing Agro-Pastoral Productivity and Markets Development Project (PRODEMA, P107343). Therefore, it became operational quickly and did not face the usual administrative difficulties that projects typically experience at project launch, and which can delay the start of initial activities, effectiveness and disbursement. Up until the MTR, there Page 23 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) was also effective support from MINAGRIE through the Steering Committee, and the active involvement of ARFIC, InterCafé and CNAC as implementing agencies, which led to swift execution of project activities. One exception was procurement of fertilizer. Contractual arrangements with the supplier proved complex and difficult to implement and are still under investigation by the Bank. These difficulties were exacerbated by red-tape and somewhat rigid administrative procedures of the PENSEB program. Up until the MTR, overall, decisions on project implementation including execution of remedial actions and coordination typically led to improved project management. Unfortunately, the changes in implementation arrangements after the MTR created hurdles for the PIU to operate efficiently as it navigated working modalities with a newly-created agency and directly with MINAGRIE’s services using temporary and sometimes unclear procedures. On this basis, the overall administrative efficiency is rated as ‘Modest.’ Justification of Overall Efficiency Rating Rating: “Modest” 64. The above efficiency analysis indicates the project was not as fully efficient as it should have been in terms of generating net benefits for farmers and using external resources at least cost. Without environmental benefits (including the GHG carbon sink generated) which are significant, the overall project efficiency performance was below expectation in comparison with both the expected benefits and recognized norms in the coffee industry. The project was also not fully efficient on the administrative front after the institutional implementing arrangement changes. The key factors are threefold: (i) a reduced implementation period; (ii) limited use of available funding resources (54 percent disbursement level); and (iii) changes in institutional implementation arrangements midway through implementation. Hence, the overall project efficiency is rated as ‘Modest’. D. JUSTIFICATION OF OVERALL OUTCOME RATING Rating: “Unsatisfactory” 65. Given the major shortcomings in the project’s relevance, efficacy and efficiency as indicated by the rating of ‘modest’ for all three aspects of project performance, the overall outcome rating is ‘unsatisfactory’. E. OTHER OUTCOMES AND IMPACTS (IF ANY) Gender 66. Given the significant gender gap in the coffee value chain, the project was gender-tagged. Its objective was to reach 30 percent of women as part of its direct beneficiaries. Activities specifically designed to address the specific needs of vulnerable groups, including women in terms of coffee productivity with a focus on environmental concerns, were implemented through a small matching grants program under Sub-Component 1.4. Twenty - eight sub-projects totaling US$423,000 were funded and benefitted 776 people of whom 573 were women. To deliver on its gender program, the project recruited a gender specialist in 2018 and commissioned a Gender Strategy for the coffee value chain. However, no action plan was prepared under the strategy, and no specialized data was collected. While the specific efforts had some visibility and were commendable, they were insufficient in scope and scale to adequately address gender and other gaps and make a marked difference across components. At closing, the project had only reached about 16 percent women, short of its target of 30 percent. This is a mediocre result even if, in hindsight, the Page 24 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) target may have been overestimated in view of the overall women’s participation in the coffee sector estimated at 20.3 percent under the recent coffee census commissioned by the project. Institutional Strengthening 67. The project institutional building activities were at the core of project support for public and private entities. Overall, the results on capacity building are mixed. The project devoted substantial resources, i.e., about US$3.5 million under Sub-Component 2.3, to strengthen ARFIC, InterCafé and NCCA. They received the bulk of the project funding earmarked for capacity building up until the MTR. However, due to the change in implementation arrangements, the capacity building earmarked for them was no longer relevant under the new State-dominated institutional framework and was discontinued. On the other hand, the project devoted additional resources to strengthen farmers’ groups and cooperatives (exceeding its target in that area), as well as MINEAGRIE’s services, including the DGMAVEA decentralized services for advisory support and extension at the grassroots level, and ISABU for research and production of planting material. The capacity building of MINEAGRIE’s services helped to continue support under Component 2, as they had to operate without the assistance of the InterCafé and CNAC networks. On support to quality improvement, Component 3 achieved lower results than planned for institutional building for ‘cupping’ capacity (no cupping laboratories were established) and upgraded processing facilities to address environmental concerns (only 24 CWSs benefited from project support vs. a target of 56 units). Mobilizing Private Sector 68. In line with the national coffee strategy, the project provided support to private operators working alongside coffee producers through a PPP program both upstream (provision of inputs and services) and downstream (processing, transport and marketing). Through Intercafé, and in partnership with ARFIC, the project aimed to support a program (56 sub-projects for US$680,000) for CWSs to access eco-friendly technologies40 to reduce environmental degradation and impact within the CWS facilities and their surroundings, and to boost CWS operational efficiency. The program had limited results since the project subsidy (40 percent) proved insufficient to elicit substantial interest from CWSs. It disbursed about US$280,000 (36 percent of allocated funds) for 24 CWS sub-projects (out of the target of 56) before it was stopped in early 2020. Poverty Reduction and Shared Prosperity 69. The project was rightly focused on the small coffee producers who are among the poorest segments of the population. Component 2 was designed to boost their incomes through the coffee rejuvenation program. Significant efforts were made to meet the needs of the vulnerable population in project areas, including those of the women coffee farmers (see para 67 above), and the Batwa population. For the latter, the project implemented a specific action plan to improve their well-being in close collaboration with UNIPROBA. Unfortunately, these efforts were cut short by the early project closure. This said, although the project did not specifically monitor the increase in farmer income generated by the program, the significant enhancement of coffee productivity (although below the project’s target) and high degree of satisfaction of smallholder project beneficiaries indicates their well-being was on course to being improved markedly under the project. This positive trend would certainly have been reinforced if the original full implementation project period had continued. 40water-conserving eco-pulpers, systems to treat waste-water, solar energy application, equipment to improve sanitation and hygiene, etc. Page 25 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Environmental benefits 70. There is evidence that the improvement in the coffee orchards, including the GAPs introduced under Component 2, have had positive environmental benefits. These include: (i) better soil coverage that reduced run-off and protected fertility; (ii) treatment of trees (pruning and cleaning) that reduced the incidence and spread of coffee plant diseases; (iii) the intercropping of food crops during the coffee tree regrowing stage that provided shade (bananas) and contributed to production of additional food for producers; and (iv) the introduction of shade-grown coffee following the results of projects such as the Sustainable Landscape Coffee Project (P127258 FY2013). The project also introduced environmentally friendly GAPs for livestock production under the small investment sub- projects which it financed particularly for women. The associated beneficial environmental impacts included practices such as keeping animals in sheds as opposed to the traditional practice of free roaming, and production of organic manure and compost, hence lessening the need to use chemical fertilizer. Other Outcomes and Impacts 71. Geo-referenced Information System/General Coffee Census. Under Sub-Component 1.5, the project financed in 2018 a contract with a consortium headed by the firms Earth I and Weathersafe from the UK. The intermediary results from the contract were not satisfactory and the contract was cancelled. As a partial alternative, in 2019, the project financed through ISTEEBU the preparation of the 3rd General Census of Coffee Trees and Growers (RGCC 2019)41. The census provided valuable information against which to assess project performance, inter alia: (i) 2019 was an “off” year (production of only 64,000 tons of coffee cherries or 10,600 tons of green coffee beans vs. 15,000- 20,000 tons on average in a normal year42); (ii) most coffee trees (94 percent) are grown in association with other crops43; (iii) only a small fraction of the coffee producers (3.8 percent) are members of a cooperative and 17 percent use NPK+S fertilizer; (iv) 20 percent of coffee producers are female producers; and (v) 43 percent of coffee producers are satisfied with extension services. 72. Grievance Redress Mechanism (GRM). The project implemented a GRM with related procedures detailed in a specialized manual. The required committees were established, and their members were trained in the office and in the field. This allowed them to keep abreast of project activities including the completion of civil works, and collection of information on beneficiary grievances. Project activities did not generate any major complaints during implementation. A comprehensive updated report was produced by the PIU and shared with the Bank prior to project closure. The report stated that 18 minor complaints were officially received, of which 9 were resolved and closed, and at project closure, 6 were being processed and 3 were not yet resolved pending further investigation. The communities, producer organizations, as well as local level committees in the project area were well informed through stakeholder engagement activities carried out by the PIU and trained on use of the grievance mechanism. 41 The previous census had been undertaken in 2007. 42 See Coffee Relaunch Strategy (2015); the fact that 2019 was a bad year had an impact on the project indicators on productivity for that year (NB: production was 93,000 tons of cherries in 2018 or 50 percent more than in 2019, and productivity per tree was 1.19 kg in 2018 as opposed to 0.82 in 2019); hence, the necessity to take into account time series data for that indicator; the conversion ratio coffee cherries vs. beans (6 to 1) see the Coffee Guide (International Trade Center) (https://www.thecoffeeguide.org/QA-108/). 43 Hence the importance to address intercropping and livestock management as was done under the GAPs supported by the project under Sub-Component 2.3. Page 26 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 73. The key factor that positively impacted project preparation is the prior preparation of the National Coffee Strategy in late 2014. The strategy was formally approved beginning 2015. Project design dovetailed closely on the strategy’s recommendations. However, project preparation and subsequent Board approval were delayed by the 2015 political events including a failed coup attempt, and widespread social unrest and violence which killed hundreds of people, and tens of thousands fled the country. The Board presentation date was delayed to June 2016 and the project closing date was adjusted to June 2023 (as opposed to June 2022 as originally planned). 74. Another positive factor at preparation was the prior and successful testing of GAPs and packages for coffee improvement under other World Bank funded projects such as the Sustainable Landscape Coffee Project (P127258, FY2013-18). Most of its recommendations were already available for dissemination at the project preparation phase and were incorporated into the design of the project’s coffee rejuvenation program. B. KEY FACTORS DURING IMPLEMENTATION 75. The project implementation years were a turbulent period politically. The political turmoil and subsequent violence that occurred at project preparation stage delayed Board presentation. Thereafter, following several years at the helm under a controversial third term, the incumbent President died suddenly on June 9, 2020, and a new rocky political environment ensued. All the above constituted an unstable context for project implementation. The project years were also not favorable in terms of project impact on productivity as they coincided in part with a low productivity period in the multi-year coffee production cycle. The 2019 campaign was a particularly low production year for the entire country. This indirectly reflected negatively on the productivity performance of the project. Finally, the Covid 19 global pandemic in 2020 exacerbated the already difficult circumstances for project implementation as field work and the periodic visits of international experts supporting project implementation, were impacted by global and local travel restrictions. Factors subject to the control of the Government 76. The one key factor under the Government’s control that negatively affected project implementation was its January 2020 decision to suspend the liberalization of the coffee value chain and unilaterally change the agreed-on project institutional implementation responsibilities legally entrusted to ARFIC, InterCafé and CNAC. The Government argued they were not effectively performing their roles and functions. However, there are no project records officially documenting the Government’s concerns or specific actions to redress them. To the contrary, the project had provided substantial resources, close to US$4 million under Sub-Component 1.4 to strengthen the capacity of these institutions which would have addressed what could have been perceived inefficiencies and shortcomings, and right- coursed the situation. The unfortunate decision by the Government led to the agreement between the World Bank and GoB in June 2021 to close the project two years earlier than anticipated. As a result, components 1 and 3 came to a halt. Implementation of component 2 continued sub-optimally with other institutions under MINEAGRIE’s oversight (DGMAVAE and ISABU in particular) not affected by the sector institutional changes. GoB discussions with the World Bank and other partners prior to its decision would have provided the opportunity for in-depth dialogue, Page 27 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) including an assessment of the legal implications on project implementation and identification of alternative options to minimize project disruption and potentially avoid the eventual project closure and cancellation of unused funds. Factors under World Bank control 77. The Bank project team continued consistent follow-up on implementation of the project areas considered still feasible despite the challenging environment at key points during implementation, especially after January 2020. The Bank conducted field implementation support missions at regular intervals twice a year (including the MTR in November 2019). In addition to regular communications and videoconferences, including during 2020 and 2021 when field work and the normal mission schedules were disrupted by the Covid 19 pandemic, three implementation support missions were completed virtually. Missions were composed of Bank technical and management staff, supported by other technical experts who helped to proactively identify and resolve issues. These missions were useful to provide timely advice and support to the PIU until the MTR. Thereafter, the missions focused more on operational matters related to the coffee orchard rejuvenation activities under Component 2. However, the lack of constructive dialogue with the Government on institutional matters prevented progress under the components 1 and 3. Factors outside the control of the Government and the Bank 78. Two major negative factors outside the Borrower’s and the World Bank’s control which were major hurdles to project implementation, were the 2020 political events surrounding the presidential elections, the unexpected death of the incumbent president in June 2020 and accompanying political upheaval, and the onset of the Covid 19 global pandemic. The pandemic prevented in-person Bank support missions for the project. The PIU was unable to do field visits due to mobility restrictions on the ground and adherence to government sanitary measures. Covid 19 also disrupted the academic training of coffee specialists from Burundi who were in Kenya for graduate degree programs. The length and the cost of training increased. Finally, the pandemic prevented international specialized experts to provide face to face support to the project stakeholders, for example to develop and train the concerned actors on the digitalized payment of coffee growers during the commercialization season. IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) 79. M&E Design. The project’s M&E activities were designed to inform and monitor progress toward achieving the PDO using effective data collection and processing arrangements. As demonstrated in the Efficacy section, the RF focused on: (i) core indicators for assessing project impact on beneficiaries; and (ii) specific outcome and intermediate indicators to assess the impact of activities per component and for the entire project. The project M&E system was supposed to evolve into an integrated MIS for the entire sector for use by the entities in charge of sector regulation and control, and those representing stakeholders, to elicit their participation and inform the dialogue of the different parties within the sector. The M&E system was to integrate geo-referenced data and associated GIS mapping outsourced to specialized bodies to gauge the evolution of coffee production. 80. The RF adequately covered all project activities, with M&E indicators devised on a Specific, Measurable, Achievable, Relevant and Time-Bound (SMART) basis. It was relatively simple and to the point, and easy to inform based on appropriate data collection procedures. It was meant to collect information on key variables such as the Page 28 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) increase in coffee productivity and beneficiary satisfaction. In hindsight, however, the RF could have been improved. First it could have gone one step beyond collecting data on yields to assessing project impact on the income of targeted producers. The systematic follow-up of producers’ income could have been devised with the collection of benchmark data at project inception, on-going data at Mid-Term and final data at project closure. Another flaw identified during the ex-post evaluation of efficacy was that the RF did not have any outcome indicator for Component 1. An indicator could have been devised to assess the degree of efficacy of the pursuit of the liberalization and privatization reform and the continued existence of the related institutions, beyond the simple number of beneficiaries and the degree of stakeholder satisfaction. This would have been very useful as a basis for dialogue with authorities at the critical time prior to the GoB’s decision to revert the sector back to State control. 81. M&E Implementation. At appraisal, M&E implementation was entrusted to the PIU and ARFIC, InterCafé and CNCA. Each entity was to collect data on their respective contribution to project implementation. ARFIC was to produce information on the quality of exports and the processing time to complete coffee export procedures. InterCafé working with MINEAGRIE’s extension directorate was to collect data on the evolution of the various project-supported coffee enhancement activities; and CNAC would help undertake the satisfaction surveys using its networks with producers on the ground. The system worked reasonably well up until the MTR. The PIU’s M&E unit was able to coordinate data collection across the relevant institutions and process and summarize data adequately. What was missing were qualitative surveys that would have further probed into keys issues such as the additional employment generated by coffee processing facilities and the impact of the increase in producer income. Clearly, M&E implementation was negatively affected by the institutional changes which disrupted the arrangement. 82. M&E Utilization. The M&E system was appropriately utilized to collect data to inform the RF and to tally and assess project impacts on a periodic basis, and therefore adequately served project implementation. However, the system failed to be turned into a proper MIS system for the sector as anticipated under project design. Justification of Overall Rating of Quality of M&E Rating: “Modest” 83. The quality of the M&E system is rated as “modest”. The M&E system as designed and implemented was generally sufficient to assess the achievement of the project objectives and inform the RF. However, it displayed shortcomings, with significant weaknesses in a few areas such as the failure to produce a functioning sector MIS, test key causal links in the results chains and provide evidence-based information on the evolution of producer incomes. B. FIDUCIARY, AND ENVIRONMENTAL AND SOCIAL COMPLIANCE 84. Financial management (FM). Though some progress was recorded in FM during the latter part of project implementation, FM performance remained “Unsatisfactory” at project closure due to: (i) the delay by the PIU in transmitting the internal auditor's project audit reports to the World Bank; (ii) reports that the PIU was forced to sign an acceptance report for 4,000 tons of fertilizers and pay the supplier even though the initial acceptance commission stated that the fertilizers did not meet the agreed contractual specifications44; (iii) some payments that were not yet processed 44 This matter is currently part of a complaint submitted to the World Bank’s Integrity (INT) unit and remains to be resolved. Page 29 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) at project closure, and which required an extension of the DDD; and (iv) issues related to potentially ineligible expenditures45 that have been resolved as of the completion of this report. 85. Procurement Performance as recorded in the final Procurement Risk Assessment and Management System (P- RAMS report June 12, 2021) was rated “Moderately Satisfactory” at project closing with a substantial risk associated to activities that needed to be completed by other executing entities such as completion of the bidding process for civil works of the remaining 27 km of rural roads serving the CWSs, and procurement of 365 tons of fungicides. During project implementation, there were significant delays in the procurement of fertilizers and fungicides. Other weaknesses which the Bank and PIU teams worked to address included: (i) the PIU’s limited experience in World Bank procurement procedures; (ii) a lack of proper procurement planning and contracts packaging (especially for large contracts); (iii) weak management of the procurement process to minimize delays and limited follow-up on implementation of the procurement plan; (iv) inadequate planning in sequencing project activities; (v) insufficient proactivity in dealing with the procurement process; (vi) inadequate record keeping and document management system especially in archiving documents in the World Bank’s Systematic Tracking of Exchanges in Procurement (STEP) and to need to establish separate and strengthened procurement records and documents in line with the Bank’s guidance on filing and archiving procurement documents, especially for activities subject to post review.; Per the World Bank’s recommendation, the PIU recruited an additional Procurement Specialist with experience in World Bank procedures beginning in 2020 which improved. 86. Environmental and social safeguards (E&) performance is rated “Satisfactory”. An audit of the project E&S implementation was commissioned to assess compliance of project activities to the World Bank’s and Burundi’s E&S safeguards policies. The audit concluded that overall, project activities conformed to the environmental and social standards, and, hence, could be safely transferred, as was envisaged at some point, to another Bank project such as the LRRP or to some other specialized government institution. C. BANK PERFORMANCE Quality at Entry 87. Quality at entry was “Satisfactory”. The project was designed in line with the long-standing Bank experience with rural development projects in Burundi, value chain development, private sector promotion in other African countries and domestic and international lessons from implementing development operations in FCV contexts. Specifically, the project design came on the heels of the approval of MINEAGRIE’s National Coffee Strategy (2015), and remnant impacts of the country’s turbulent past including lingering political sensitivities. The project incorporated the recommendations of the coffee strategy, including support to governance to further liberalize the sector, coffee orchard rejuvenation to increase productivity, and enhancement of coffee quality and marketing to take advantage of international markets, particularly niche markets. The World Bank provided timely technical support to ensure a sound project design. The main risks were identified, and appropriate mitigation measures and M&E modalities were proposed. The risk related to the possible change of course by GoB vis-à-vis the sector governance was duly flagged as high, although it was not anticipated that the GoB would completely reverse the liberalization course at MTR. Procurement, financial management and safeguards were duly analyzed and taken into account as part of project preparation and appraisal. Overall, the RF 45It appears the project accepted some deliverables (on at least one contract related to the delivery of 385 tons of fungicides) after the closing date and still paid some custom related fees. Page 30 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) captured all pertinent outcomes and results with SMART indicators; however, a shortcoming was the lack of an indicator on the degree of efficacy of the pursuit of the coffee sector liberalization and privatization reforms. Quality of Supervision 88. Quality of Supervision is rated as “Moderately Satisfactory”. The World Bank team closely monitored project progress and provided regular implementation support and training to reinforce the capacity of the PIU. Supervision missions were organized systematically at six-month intervals per World Bank guidelines. The supervision missions in 2020 and 2021 were conducted virtually due to the Covid 19 pandemic restrictions. All missions were conducted in close collaboration with key sector stakeholders, notably MINEAGRIE and ISABU, as well as ARFIC, InterCafé and CNAC when they were still on board. This facilitated supervision coordination and maintained the momentum of political/strategic dialogue among all the different stakeholders, especially when significant coordination was required to keep some project implementation momentum after enactment of the new reforms. In between support missions, the Bank project team was responsive and proactive. It systematically responded to the client’s project team requests (for non-objection or technical advice) in a timely manner as mandated under World Bank procedures. This was particularly evident in the guidance provided to the client during the MTR review and in the aftermath of the new sector reforms. Providing implementation support required navigating around major decisions among them the sector institutional changes and related confusion from the new and oftentimes unclear government procedures and processes, the corresponding use of budget resources, the increase in the fertilizer subsidy, the realities of the COVID-19 pandemic limitations, as well as the difficult socio-political landscape further impacted by the turbulent and extended election process (April to August 2020) and the sudden death in June 2020 of President Pierre Nkurunziza under whom the strategic changes in the coffee value chain had been initiated. Aide-memories, Implementation Status and Results Reports (ISRs) and other records of exchanges with the client show the World Bank was candid with the national counterpart team on project implementation bottlenecks and proposed remedies, despite the sometimes-fractious discussions with the client. 89. Notwithstanding the above, this report notes it took more than a year for the World Bank and GoB to reach the final decision to close the project early. While continuing to support the implementation of project activities that were still feasible, the World Bank project team with the support of the Country Management Unit, continued to engage46 with government authorities at the highest level to attempt to reach a workable option that would fully ring-fence the project to allow achievement of the PDO as originally designed. However, in the absence of a revised PDO, the ideal option would have been to adjust the results targets such as during the first project restructuring in June 2020 which only focused on a reallocation of funds. Furthermore, both parties could have immediately embarked on a practical course to use the remaining project funds for the ongoing technical programs under Component 2 that were proving extremely useful as indicated at MTR (e.g., coffee orchard replanting, full stumping and fertilization programs). Measures47 to make this option possible were considered late in the process. Given the impasse, lost time and the Bank’s complex and protracted procedures48, there was insufficient time to implement alternatives. That 44 percent of the project grant funding had to be cancelled was clearly not an optimal outcome for the World Bank and GoB, and more importantly for all coffee operators in Burundi, in particular the small-scale coffee growers. This proved a clear downside for the sustainability of project activities. 46 exchanges between the GoB and World Bank management in the Burundi office, numerous discussions and exchanges between the World Bank Task Team and project team and aide-memoire indicating the changes posed considerable risks to the Project. 47 reallocation of remaining funds to the Burundi Landscapes Resilience and Restoration Project (BLRRP) (series of internal Bank meetings between January and April 2021), preparation of Additional Financing (March 2021 meeting with OPCS to clarify Additional Financing processing and restrictions). 48 Including requirements for application of the new Environmental and Social Safeguards policies and COVID – 19 related waiver Page 31 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Justification of Overall Rating of Bank Performance 90. The overall World Bank performance is rated “Moderately Satisfactory”. While project quality at entry is rated “satisfactory”, the overall quality of supervision is rated “moderately satisfactory” despite the significant implementation efforts deployed by the World Bank team under the prevailing difficult circumstances. The impact of the impasse on the coffee sector governance dialogue and the protracted process and eventual failure to find workable measures is significant as it ultimately led to early closure of the project and cancellation of 44 percent of the grant funding. D. RISK TO DEVELOPMENT OUTCOME 91. The major risk to the development outcome stems from the 2020 reversal of the liberalization policy of the coffee sector which was critical for its development. The subsequent regulatory and institutional changes, the June 2021 project restructuring, and early project closure resulted in non-completion of key component 2 activities. These activities were critical to the long-term productive status of the coffee sector and were on track both at MTR and at project closure. Their completion and impact assessment would have been key. This risk is assessed as substantial in the long run in the absence of a policy readjustment and if MINEAGRIE does not have minimum capacity to directly handle project operations through DGMAVEA and ISABU, after the removal of ARFIC, InterCafé and CNCA. Also, in the short-term, on the Bank side, the Landscape Restoration and Resilience Project (PZDOC P160613) and Great Lakes Regional Integrated Agriculture Development Project (P161781) to which it was envisaged to transfer project activities at some point, could not be used as a relay project to pursue some key related activities due to procedural difficulties. 92. Other specific risks to the project development outcome are as follows: a) Non completion of the small investment sub-projects programs for vulnerable groups (women, youth and Batwas) and upgrade of CWSs regarding environmental features: although they were of small scope and size, these programs were critical to the project development outcome as they addressed both the plight of vulnerable groups and environmental issues surrounding coffee processing; b) At closure, the project was still supporting 8 students (6 at master’s level and 2 PhD candidates), whose training curricula had been delayed by the Covid-19 pandemic. The completion of their curricula spans a period up to September 2021 for the earliest and August 2022 for the latest. Having already spent more than 2.5 years under the project financing, the cancellation of the funds would jeopardize all the gains accrued so far and prevent them from obtaining their degrees in various scientific topics related to the coffee sector. Arrangements to fund the remaining costs under the Bank-funded Great Lakes Regional Project as of completion of this report were approved; and c) The project is currently under investigation for possible mis-procurement of 4,000 tons of fertilizers (completed end 2019), which generated complaints. In the event the investigation reveals any ineligible expenditures, this would result in another concern regarding the project because GoB would likely be required to reimburse them. V. LESSONS AND RECOMMENDATIONS 93. Notwithstanding the implementation challenges, shortened project timeframe in a FCV context and partial fund disbursements, useful lessons can be drawn that can inform future operations in similar contexts as well as serve as useful tools for continued Bank sector dialogue in Burundi. The general lessons center around the criticality of ensuring: Page 32 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) (i) a multi-faceted project team49 composition with a strong presence on the ground; (ii) adequate budgets and support to offset the additional and different challenges of working in a FCV context; (iii) sound communication strategies to better communicate with officials and beneficiaries during implementation where political sensitivities are high and delicate; (iv) better coordination and communication within Bank projects teams so that cross-cutting lessons can be shared and applied in real time; and (v) flexible procedures to allow the Bank to respond quickly in fast-changing environments. 94. In a FCV context where the political environment can be uncertain and government policies can change abruptly, such as in the case in Burundi, it is necessary to have flexible procedures in place under World Bank projects to allow a quick transfer of undisbursed funds across the portfolio when conditions for a particular operation to continue are not met. The aim is to continue supporting worthwhile activities which otherwise would stop, and, hence, keep project benefits accessible for ultimate beneficiaries such as smallholder farmers. While fund reallocations and Contingent Emergency Response Components (CERCs) are suitable existing options for use within ongoing projects, it would be beneficial for the World Bank to explore additional agile FCV tailored procedures that in case of an emergency, would allow easy access and quick transfer of funds within World Bank country portfolios. This of course does not eliminate the risk that fund transfers under this procedure would lead to the desired outcome completely, and, clearly, does not preclude the need to proceed swiftly when adverse circumstances arise. Success is always predicated on acting swiftly as soon as obstacles arise. This did not happen under the project. The envisaged transfer of the undisbursed project funds to other operations proved impossible because procedures were not in place for that purpose and the process was initiated too late. As a result, unfortunately, the undisbursed funds of US$19.2 million, more than a third of the initial grant funding, had to be canceled. 95. Producer incentives: a necessary condition for the coffee sector (or any other crop sector predicated on smallholder production) to thrive, and for benefits to eventually accrue to all parties, is for farm level prices to be set at a sufficiently high level to incentivize producers. This can only be achieved if the sector is liberalized and managed under rules established jointly by all stakeholders of the entire value chain. This was requested under the project as a metric to gage progress in establishing an enabling environment for coffee producers. It was to be done by setting producer prices at an attractive level using a formula agreed by all parties as part of regulatory changes aimed at liberalizing sector policies. The main end-result was to incentivize farmers to invest more of their resources and efforts into coffee production. The related changes did occur during the first half of the project (before they were reversed under the January 2020 sector reforms). They aimed at considering smallholders as full partners in the value chain and passing on to them an adequate fraction of international prices. These changes were expected to help Burundi reverse course and embark upon a virtuous circle benefiting all coffee value chain actors. Enhanced farmer incentives would result in higher yields and improved coffee quality. Higher volumes of quality, fully washed coffee would, in turn, lead to lower per unit processing costs and to more stable and higher payments by off-takers. Thus, higher prices would contribute to more generous margins across the board, enabling a renewed cycle of investments. This would again place Burundi’s coffee sector on a steady growth path similar to other countries that have liberalized the sector — a path that would generate higher foreign exchange earnings and raise the country’s prospects for a sustainable coffee sector. This is a worthwhile course that may merit further pursuit with the Burundi authorities. 96. Financial transactions: the project demonstrated that timely payments need to be made along the value chain between actors, especially at upstream level from CWSs to producers, to keep the required level of incentives with 49Not only agronomists and ag. Economists to vet technical propositions, but also political analysts to assess the socio-political environment, sociologists/anthropologists to assess issues such as gender, youth and poverty, cultural dimensions, etc., that are critical for project implementation. Page 33 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) the producers. At certain points in the past, producers had to wait for extended periods of time to get paid by CWSs once they delivered their cherries. This is partly because the CWSs were also encountering difficulties to get paid by final off-takers down the value chain. In addition, they could not access seasonal credit from commercial banks to cover their financial shortages. Producers should be paid promptly in cash or by e-transactions. At MTR, the Bank launched analytical work to identify options (e.g., e-banking) to improve financial transactions. This work was not completed unfortunately given the change in government sector policy. 97. In the context of a coffee rejuvenation program, intercropping during the gestation period when new coffee plants and stumped trees are growing and not yet fully productive, is a good agricultural practice that needs to be systematically disseminated and adopted. Intercropping ensures continued agriculture production during the critical transition period. It allows producers to maintain some measure of cash generation to feed and meet their basic households needs. In the context of Burundi, the project’s intercropping packages (including under shade grown coffee) obtained from the Sustainable Coffee Landscape Project were ready for immediate dissemination to farmers. The recommendation regarding intercropping is valid in countries other than Burundi under projects for tree crop plantations in the hands of smallholders such as cocoa, cashews or mangoes that need to be created, rehabilitated or expanded. 98. Last mile good rural infrastructure is critical to quickly link producers to the point of first treatment and local markets following crop harvest. In Burundi, coffee cherries need to be brought to CWSs within 6 hours after harvest to avoid a drastic decline in quality as many producers do not have proper storage facilities to maintain fresh berries. However, although distances in Burundi are short, rural roads, especially tracks at producer level are usually in very bad condition and hardly passable during the rainy season which coincides with the timing of the coffee harvest50. Typically, therefore, coffee farmers transport their coffee cherries to the CWSs either by bicycle or on their heads. Doing so increases the time devoted to post-harvest and transport activities, and/or raises transportation costs. Producers were particularly satisfied with the 33 km (out of the planned 60 km) of roads constructed/rehabilitated under the project. The remaining 27 km could not be completed given the project cancellation. When hinterland transport infrastructure is as poor as it is in Burundi, projects which address production at the grassroots must make provisions to finance ‘last mile’ rural roads to open producing areas and decrease the time and cost of transporting fresh primary production to . processing points and markets. 50 The coffee cherries are harvested over a period of about five months from March to July (medium altitude) and April to August (high altitude). The period coincides with the rainy season (February to June) during which the rural roads are often impassable. Page 34 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: PDO indicators Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Increased coffee productivity Number 0.80 1.60 1.20 in the project area 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Share of coffee production Percentage 59.00 69.00 65.93 with AA/A/FW15+ quality 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Page 35 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Direct project beneficiaries Number 0.00 300,000.00 313,981.00 31-Mar-2015 30-Jun-2023 30-Jun-2021 Female beneficiaries Percentage 0.00 30.00 15.62 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Share of specialty coffee of Percentage 7.50 15.00 7.50 total exports 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Exports were done during the project period however data is not available. The data was to have been provided by the original implementing agencies that were removed from the project after the Government of Burundi's decision in January 2020 to revert the coffee sector to State control. A.2 Intermediate Results Indicators Component: Component 1: Institutional strengthening and value chain governance Page 36 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Coffee Act adopted by the Yes/No No Yes No Ministry of Agriculture and Livestock 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Coffee cooperatives legally Percentage 31.00 70.00 77.08 registered 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Coffee pricing formula Yes/No No Yes No No accepted by InterCafé, CNAC and ARFIC 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Page 37 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Coffee sector MIS in place Yes/No No Yes No and operational 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Coffee trees replanted Number 0.00 16.50 16.06 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Coffee trees rejuvenated Number 0.00 18.75 18.92 through full stumping 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Page 38 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Total area established with Hectare(Ha) 0.00 7,050.00 6,498.14 improved inter-cropping systems 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Clients who have adopted an Number 0.00 90,000.00 97,687.00 improved agricultural technology promoted by the 30-Apr-2015 30-Jun-2023 30-Jun-2021 project Comments (achievements against targets): Component: Component 2: Coffee growers productive capacity enhancement Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Coffee trees replanted (in Number 0.00 16.50 16.06 million) (Thousand) 31-Mar-2015 30-Jun-2023 30-Jun-2021 Page 39 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Coffee trees rejuvenated Number 0.00 18,750,000.00 18,920,000.00 through full stumping 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Total area established with Hectare(Ha) 0.00 7,050.00 6,498.14 improved inter-cropping systems 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Clients who have adopted an Number 0.00 90,000.00 97,687.00 improved agricultural technology promoted by the 31-Mar-2015 30-Jun-2023 30-Jun-2021 project, Page 40 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion 12. Roads rehabilitated, Kilometers 0.00 60.00 33.00 Rural (km) 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Component: Component 3: Coffee quality improvement and market access Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of new cupping labs Number 0.00 7.00 0.00 operational in major production areas, (Number) 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion CWSs and dry mills equipped Number 0.00 56.00 24.00 Page 41 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) with eco-friendly tech/ 31-Mar-2015 30-Jun-2023 30-Jun-2021 infrastructure and/or up- graded storage facilities Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Additional man/days offered Number 0.00 240,000.00 0.00 by CWS (by gender), 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Data is not available. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Roads rehabilitated, Rural Kilometers 0.00 60.00 33.00 (km) 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Indicator Name Unit of Measure Baseline Original Target Formally Revised Actual Achieved at Page 42 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Target Completion Coffee export procedure Days 10.00 3.00 0.00 processing time (ARFIC) - (Days) 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Beneficiary satisfaction with Percentage 0.00 70.00 97.00 project services (including CNAC, Inter Café) 31-Mar-2015 30-Jun-2023 30-Jun-2021 Comments (achievements against targets): Page 43 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) B. KEY OUTPUTS BY COMPONENT Outcome Indicators Description (OIs) Core Outcome Indicator: Direct beneficiaries reached by the project Outcome Indicator OIC – Direct beneficiaries (disaggregated by gender) (Number/ percentage) (Core) (OIC) Key output OCI – 313,981 direct beneficiaries reached (15.62% women) Objective/Outcome 1: Institutions and value chain governance strengthened IR 1.1 – Coffee Act adopted by the Ministry of Agriculture and Livestock (Yes/No) IR 1.2 – Coffee cooperatives legally registered (percentage) Intermediate Results IR 1.3 – Coffee pricing formula accepted by InterCafé, CNAC and ARFIC Indicators (IRs) (yes/no) IR 1.4 – Coffee sector MIS in place and operational (yes/no) IR 1.5 – Beneficiary satisfaction with project services (including CNAC, InterCafé) (percentage) OI 1.0 -- Direct project beneficiaries: 313,981 (52%) IR 1.1 – The Coffee Act was adopted by the Ministry of Agriculture and Key Outputs by Livestock: No Component IR 1.2 – Percent coffee cooperatives legally registered: 77.08% (linked to the IR 1.3 – Coffee pricing formula was accepted by InterCafé, CNAC and achievement of the ARFIC: No Objective/Outcome 1) IR 1.4 – Coffee sector MIS was place and operational: No IR 1.5 – Beneficiary satisfaction with project services (including CNAC, InterCafé): 97% Objective/Outcome 2: Coffee growers productivity capacity enhanced Outcome Indicators OI 2.0 – Increased coffee productivity in the project area (kg/tree) (OIs) IR 2.1 – Coffee trees replanted (million) IR 2.2 - Coffee trees rejuvenated through full stumping (million) Intermediate Results IR 2.3 – Total area established with improved intercropping systems Indicators (IRs) (ha) IR 2.4 – Clients who have adopted an improved agricultural technology promoted by the project (number) Page 44 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) OI 2.0 – Productivity increased from 0.8kg/tree to 1.2kg/tree Key Outputs by IR 2.1 – Coffee replanted: 16.06 million Component IR 2.2 - Coffee trees rejuvenated through full stumping: 18.92 million (linked to the IR 2.3 – Total area established with improved intercropping systems: achievement of the 6,498.14 ha Objective/Outcome 2) IR 2.4 – Producers who have adopted an improved agricultural technology promoted by the project: 97,687 Objective/Outcome 3: Coffee quality and market access improved OI 3.1 - Share of coffee production with AA/A/FW1+ quality Outcome Indicators (percentage) (OIs) OI 3.2 – Share of specialty coffee of total exports (percentage) IR 3.1 – Number of new cupping labs operational in major production areas (number) IR 3.2 – CWSs and dry mills equipped with eco-friendly tech/ Intermediate Results infrastructure and/or upgraded storage facilities (number) Indicators (IRs) IR 3.3 – Additional man/days offered by CWSs (by gender) (man/days) IR 3.4 – Rural roads rehabilitated (km) IR 3.5 – Coffee export procedure processing time (ARFIC) (days) OI 3.1 - Share of coffee production with AA/A/FW1+ quality: 65.93% OI 3.2 – Share of specialty coffee of total exports: unavailable Key Outputs by IR 3.1 – Number of new cupping labs operational in major production Component areas: 0 (linked to the IR 3.2 – CWSs and dry mills equipped with eco-friendly tech/ achievement of the infrastructure and/or upgraded storage facilities: 24 Objective/Outcome 3) IR 3.3 – Additional man/days offered by CWSs (by gender): unavailable IR 3.4 – Rural roads rehabilitated: 33 km IR 3.5 – Coffee export procedure processing time (ARFIC): unavailable C. OTHER PROJECT OUTPUTS Component 1 Sub-Component 1.1: • Workshop for 60 participants to review and validate the institutional framework • Study of cost of production Sub-Component 1.2 • Two fora to assess the coffee sector organization Sub-Component 1.3 • Capacity building of ARFIC, InterCafé and CNAC: o Technical assistance (including funding of two M&E specialists, one accountant, two procurement specialists, and one marketing/ communication specialist) o Staff training (198 staff trained), Page 45 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) o Office equipment and renovation/ expansion of office buildings (including ARFIC’s building for the one-stop window, central laboratory, three regional laboratories), o Studies: organizational audit of InterCafé, and CNAC M&E manual • Capacity building of MINEAGRIE staff, including staff decentralized offices and staff of ISABU: 502 staff trained; preparation of kits for GAPs; footage/ clip on GAPs as extension tool Sub-Component 1.4 • Study regarding gender strategy • Operating manual and attendant training of beneficiaries for gender sub-projects • Financing of 28 sub-projects (including 50 ha of shade coffee) for about US$675,000 million (including beneficiary contribution of US$300,000 for 776 beneficiaries including 573 women Sub-Component 1.5 • National Coffee Census (2019) Component 2 Sub-Components 2.1 • Rejuvenation of coffee trees: 16 million (replantation), 19 million (full stumping) • Associated crops: fertilizer (1,660 tons DAP and 37 tons urea), seeds (1,038 tons of common beans and 54 tons of soya beans) and tool kits to cover 14,000 ha Sub-Component 2.2 • Fertilizer: NPK-S: 424 tons (subsidy at 60%); et 3,400 tons (subsidy at 80%) total 3,824 tons (vs. 10,500 tons planned at appraisal) • Organic manure: 307,000 tons • Pesticides: Insecticides (120 liters) and fungicides (470 tons), plus attendant spraying equipment and producer training Sub-Component 2.3 • Training on GAPs for 14,315 man/days based on FFS • Demonstration sub-projects: shade coffee (18 sub-projects) and intercropping (20 sub-projects) • Demonstration plots: 72 plots • Kits to producers: 2560 Sub-Component 2.4 • Production of seeds by ISABU: 13 tons, including rehabilitation of seed plots, upgrade of seed laboratory and staff training • Agronomic trials: 3 ha of trial plots established by ISABU • Rehabilitation of ISABU Kayanza research station comprising biotechnology and phytopathology laboratories, and training and documentation center • Academic training: 1 post-graduate, 2 PHDs and 6 masters (to be completed): Component 3 Sub-Component 3.1 • Training on quality protocols (‘cupping’) by UNIDO (100 trainee including 10 women for US$60,000) • Technical audit and action plans for 261 CWSs • Sub-projects (24) to upgrade CWSs environmental facilities for total subsidy of about US$361,000 • Rural Roads (33 km) to provide better access to 20 CWSs for US$1.7 million Sub-Component 3.2 Page 46 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) •Participation of Burundi delegation in international coffee events and promotion of Burundi ‘Coffee of Excellence’ Component 4 • Environmental and social audit • Satisfaction survey with beneficiaries • Study of project impact at mid-term D. ALTERNATIVE EFFICACY ASSESSMENT BASED ON YEAR 4 VS. YEAR 6 TARGETS The table below compares the project efficacy performance assessment based on the PAD targets for Year 4 as opposed to the basic scenario reflected in the Results Framework final targets in Year 6 (see indicators under Section A. above). This last scenario is used in the report to assess the efficacy performance even though the project duration was reduced from six to four years under the June 2021 restructuring. The performance was assessed on that basis because the Results Framework at restructuring was not revised to reflect the shorter project duration and reduced amount of funding available. The results of the assessment based on the PAD targets for Year 4 are not substantially different from those under the scenario for Year 6. The performance using this approach is ‘substantial’ for the core outcome of number of beneficiaries and Outcome 2 on the rejuvenation program, and ‘modest’ for outcome 1 and 3 on project governance and marketing/ quality respectively, implying a performance of ‘modest’ overall (see Table 3 below). Page 47 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Rate of Target Achievement PDO-Level Results Indicators Baseline Achievement (June 2021) Year 4 Year 6 Year 4 Year 6 PDO Core Outcome: Direct beneficiaries reached by the project • OIC – Direct beneficiaries (disaggregated by gender) 210,000 300,000 313,981 150% 105% - (Number/ percentage) (25%) (30%) (15.6%) (62%) (52%) PDO Outcome 1: Institutional strengthening and value chain governance Intermediate Results linked to Outcome 1 • IR 1.1 – Coffee Act adopted by the Ministry of - Yes Yes No 0% 0% Agriculture and Livestock (Yes/No) • IR 1.2 – Coffee cooperatives legally registered 31% 65% 70% 77% 135% 118% (percentage) • IR 1.3 – Coffee pricing formula accepted by InterCafé, - Yes Yes No 0% 0% CNAC and ARFIC (Yes/No) • IR 1.4 – Coffee sector MIS in place and operational - Yes Yes No 0% 0% • IR 1.5 – Beneficiary satisfaction with project services - 60% 70% 97% 162% 138% (including CNAC, InterCafé) (percentage) PDO Outcome 2: Coffee growers productivity capacity enhancement • OI 2.0 – Increased coffee productivity in the project 0.8 1.4 1.6 1.2 67% 50% area (kg/tree) Intermediate Results linked to Outcome 2 • IR 2.1 – Coffee trees replanted (million) - 11.55 16.50 16.06 139% 97% • IR 2.2 - Coffee trees rejuvenated through full stumping - 13.13 18.75 18.92 144% 101% (million) • IR 2.3 – Total area established with improved - 4,900 7,050 6,498 133% 92% intercropping systems (ha) • IR 2.4 – Clients who have adopted an improved agricultural technology promoted by the project - 48,000 90,000 97,700 204% 109% (number) PDO Outcome 3: Coffee quality improvement and market access • OI 3.1 - Share of coffee production with AA/A/FW1+ 59% 65% 69% 66% 117% 70% quality (percentage) • OI 3.2 – Share of specialty coffee of total exports 7.5% 12.5% 15% - - - (percentage) Intermediate Results linked to Outcome 3 • IR 3.1 – Number of new cupping labs operational in - 45 45 0 0% 0% major production areas (number) • IR 3.2 – CWSs and dry mills equipped with eco-friendly tech/ infrastructure and/or upgraded storage facilities - 60 70 24 40% 34% (number) • IR 3.3 – Additional man/days offered by CWSs (by - 80,000 240,000 0 0% 0% gender) (number) • IR 3.4 – Rural roads rehabilitated (km) - 60 60 33 55% 55% • IR 3.5 – Coffee export procedure processing time 10 5 3 - - - (ARFIC) (days) E. SUCCESS STORIES AND BEST PRACTICES Page 48 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) The project though cut short by two years, has some success stories as summarized below. Component 1. Enhancement of women, youth and Batwa empowerment. The project developed a stand-alone comprehensive coffee gender and youth strategy (Women and Youth Strategy, 2018), to enhance the participation of women and youth in the full coffee value chain. Emphasis was on facilitating their access to assets, training, and employment. In addition, the project established a matching grant program to specifically promote the entrepreneurial activities of women, youth, and the often-marginalized Batwa people. The program supported activities: (i) linked to coffee production which led to land and resource preservation; and/or (ii) which promoted climate-smart and eco-friendly technologies for household use such solar panels, improved cooking stoves. Successful elements of the program included the introduction of labor-saving production and processing tools by women’s groups which contributed to reducing their workloads and improving their well-being. The program also targeted youth and increasing their production, processing, marketing and management capacity, to among others, give them an entry point as service providers in contract spraying on farmers’ fields, transport services, and equipment maintenance. The highly visible program financed 28 sub-projects and was very much appreciated by the participating beneficiaries. However, its size and scope were small. At project closure US$0.42 million of the allocated US$0.68 million had been utilized. Component 2. Increase in access to fertilizers through a voucher system under the existing National Fertilizer Subsidy Program (PNSEB). The PNSEB demand-driven program was originally conceived for food crops only. The project was successful in obtaining approval to channel its funds for the fertilizer sub-program through the PNSEB. The inclusion of coffee within the PSNEB subsidy package expanded access to fertilizers for smallholder farmers who also cultivate coffee for income diversification while also allowing them to purchase fertilizers for food crops. The introduction of organic fertilizer during implementation not foreseen at project design stage was also a success and widely appreciated by producers. The specific operating arrangements summarized in the diagram below, although seemingly complex, worked well under the project. Process flow of the fertilizer subsidy program Component 3. Sustainable modernization of processing facilities to promote responsible environmental management. The project supported Public-Private Partnership (PPP) program facilitated access to eco-friendly Page 49 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) technologies to reduce water use and environmental degradation within the CWSs and their surroundings. The investments funded on a cost-sharing basis (60 percent by the private sector and 40 percent by the project) included, among others, water-conserving eco-pulpers, systems to treat waste-water and remove pulp from waste- water for use as fertilizer, solar energy application, equipment to improve sanitation and hygiene, etc. These investments aimed at substantially reducing production costs led to a reduction of environmental resources (particularly water use) degradation. The PPPs also supported implementation of environmental standards for coffee washing stations, enhanced awareness of eco-certification schemes for access to niche markets, education on labor practices, social responsibility, gender equality awareness and adoption of environmental management systems such as ISO 14000. In addition, for highly performing washing stations, the project supported the introduction of traceability systems from the plantations to destination markets to enable buyers and sellers to track each lot from the time it enters the washing station (or even the field) until it is delivered to its final destination, an essential element that distinguishes specialty coffee production from the production of standard grade coffee. Responsible environmental management is a recognized best practice for coffee processing and an obligation for certification to several preferred supplier programs such as the Starbucks’ C.A.F.E. Practices Program. Page 50 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Juvenal Nzambimana, Paola Agostini, Juvenal Nzambimana Task Team Leader(s) Melance Ndikumasabo Procurement Specialist(s) Bella Diallo Financial Management Specialist Grahame Beaumont Richard Dixie Peer Reviewer Wilhelmus Gerardus Janssen Peer Reviewer Marie-Claudine Fundi Team Member Gayle Martin Team Member Cheikh A. T. Sagna Social Specialist Oliver Braedt Team Member Jean Okolla Owino Team Member Omar Lyasse Team Member Stephen Ling Social Specialist Srilatha Shankar Team Member Mamadou Ndione Team Member Luz Berania Diaz Rios Team Member Kaliza Karuretwa Team Member Ademola Braimoh Team Member Nneoma Veronica Nwogu Counsel Faly Diallo Team Member Page 51 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Shingira Samantha Masanzu Counsel Hugo De Vries Team Member Christian Simbananiye Team Member Alice Museri Team Member Supervision/ICR Amadou Alassane, Benjamin Billard Task Team Leader(s) Arcade Bigirindavyi, Rahmoune Essalhi, Clement Tukeba Procurement Specialist(s) Lessa Kimpuni Hugues Agossou Financial Management Specialist Yeo Yenemanyan Financial Management Specialist Yves Bertrand Koudjou Tatang Team Member Blaise Pascal Dushime Team Member Ghislaine Kouedi Nombi Team Member Alexis Manirambona Environmental Specialist Saba Nabeel M Gheshan Team Member Boyenge Isasi Dieng Social Specialist Alice Museri Procurement Team Philippe Eric Dardel Team Member Nneoma Veronica Nwogu Counsel Clarette Rwagatore Team Member Nancy Musibega Visavilwa Program Manager Enagnon Ernest Eric Adda Team Member Marie Lolo Sow Team Member Amadou Konare Environmental Specialist Marie Roger Augustin Team Member Nora Kaoues Team Member Juvenal Nzambimana Team Member Tracy Hart Environmental Specialist B. STAFF TIME AND COST Page 52 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY15 41.290 286,804.46 FY16 15.191 100,861.37 FY17 16.934 84,604.91 FY18 0 175.62 FY20 0 523.34 Total 73.42 472,969.70 Supervision/ICR FY17 .527 8,689.65 FY18 26.197 177,102.13 FY19 23.130 146,898.08 FY20 20.915 162,935.38 FY21 0 - 6.00 Total 70.77 495,619.24 Page 53 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) ANNEX 3. PROJECT COST BY COMPONENT Amount at Approval Actual at Project Closing Percentage of Approval (US$M) (US$M) (%) World Benefi- World Benefi- World Benefi- Components Total Total Total Bank ciaries Bank ciaries Bank ciaries Component 1 – Institutional strengthening and value chain governance Sub-Component 1.1: Streamlining 0.69 - 0.69 0.10 - 0.10 14 - 14 of the institutional framework Sub-Component 1.2: Enhancing value chain coordination and 0.37 - 0.37 0.05 - 0.05 14 - 14 public-private dialogue Sub-Component 1.3: Strengthening the capacity of 6.11 - 6.11 3.49 - 3.49 57 - 57 sector institutions Sub-Component 1.4: Women and 0.54 0.14 0.68 0.57 0.186 0.76 106 132 111 youth empowerment Sub-Component 1.5: Geo- reference information system for 0.79 - 0.79 0.99 - 0.99 126 - 126 the sector Sub-Total 8.50 0.14 8.64 5.20 0.19 5.39 62 132 62 Component 2 – Coffee growers’ productive capacity enhancement Sub-Component 2.1: Rejuvenation 11.73 2.03 13.76 8.30 - 8.30 71 a/ 60 of existing plantations Sub-Component 2.2: Improving access to fertilizer and agro- 11.60 13.52 25.12 10.92 1.75 12.67 94 13 50 chemicals Sub-Component 2.3: Dissemination of sustainable 5.41 - 5.41 1.31 - 1.31 24 - 24 agricultural practices Sub-Component 2.4: Applied 2.52 0.23 2.75 2.45 - 2.45 97 - 89 coffee research Sub-Total 31.26 15.78 47.04 22.98 1.75 24.73 74 11 53 Component 3 – Coffee quality and market access Sub-Component 3.1: Coffee 4.89 1.32 6.21 2.15 0.38 2.53 44 29 41 quality improvement Sub-Component 3.2: Marketing 3.19 - 3.19 1.43 - 1.43 45 - 45 and promoting the Burundi coffee Sub-Total 8.08 1.32 9.40 3.58 0.38 3.96 44 29 42 Component 4 – Coordination, monitoring and knowledge management Sub-Total 7.16 - 7.16 4.57 - 4.57 64 - 64 TOTAL 55.00 17.25 72.25 36.33 2.33 38.66 66 13 53 Page 54 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) ANNEX 4. EFFICIENCY ANALYSIS I. CONCEPTUAL FRAMEWORK 1. The project targeted major coffee growing areas in six provinces of Burundi (Kayanza, Ngozi, Karuzi, Gitega, Muyinga and Kirundo). These provinces represent more than 50 percent of the country’s coffee growers and production. In these provinces, the project supported a coffee rejuvenation program (replanting and stumping of coffee trees, Sub-Component 2.1), and a program of access to inputs (fertilizer and pesticides, Sub-Component 2.2), as well as support to research and extension (Sub-Components 2.3 and 2.4). The project targeted not only coffee but also the food crops associated with the rejuvenation program. In addition, it supported coffee processing and marketing to enhance coffee quality and obtain better prices (Component 3), and the institutional strengthening of the coffee sector (Component 1). Finally, it funded expenses related to the management and coordination of project activities (Component 4). 2. Direct project benefits. The project provided direct socio-economic benefits to coffee growers, and to several other actors involved in the coffee industry, for a total of about 314,000 households, thereby contributing to Burundi’s national socio-economic welfare. As part of the project ex-post evaluation for this ICR, an attempt has been made to measure some of the direct incremental benefits that are expected from project implementation, by comparing the ‘with’ and ‘without’ project situations. Direct project benefits have been estimated for activities under the rejuvenation and access to inputs programs. For the other project activities (support to extension and research, upgrade of coffee processing, access to markets, institutional strengthening and project coordination and management), no cost-benefit analysis was prepared. Indeed, the related benefits, although they are substantial, are often intangible and difficult to quantify. 3. ‘Without’ and ‘with’ project situations. In the ‘without project’ (or reference) scenario, coffee trees are low yielding (average of 0.8 kg of cherries per tree). Producers use some manure but virtually no chemical fertilizer, and outdated implements. They get little or no outside advisory support neither for cropping, nor for post-harvest or marketing activities. They typically practice intercropping which is a practice well adapted to low input agriculture. In the ‘with project’ scenario, producers are given access to an improved production package (high yielding plants for coffee replacement and seeds for associated crops, improved implements, chemical treatments and fertilizer supply, etc.) Provision of these inputs is subsidized: (i) for the improved package of seedlings, seeds and implements to the tune of 100 percent the first year, 75 percent the second year and 50 percent the third year; and (ii) for fertilizer to the tune of 40 percent until MTR and 80 percent for the remainder of the project period. Producers are also provided with advisory services to apply the technical packages, as far as good production and soil fertility management practices are concerned. They get access to better processing facilities (CWSs and dry-milling units) supported through public-private partnerships. These facilities add value to their products. Finally, the project funds marketing advice that improves producers’ knowledge of markets, increases their negotiating power and allows them to obtain better prices. The replanting and stumping of coffee trees, and the adoption of the improved packages for coffee trees in production were expected to result in a doubling of coffee yields in final Year 6, and a 70 percent to 100 percent yield increase for associated crops (beans and soybeans selected in the analysis as typical intercrops). 4. Benefit streams. Project activities have generated three main benefit streams: (i) increased production of coffee and associated crops due to rejuvenation (replanting and stumping), use of improved planting material, fertilizer and pesticides, as well as improved cultural practices (pruning, weeding, mulching, use of compost, etc.); (ii) returns to investments on product quality enhancement mainly through processing facilities, as well as Page 55 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) investments in access roads linking producing areas to Coffee Washing Stations ; and (iii) benefits from capacity building for farmers’ groups and cooperatives, and other organizations along the value chain. These benefit streams lend themselves either fully or less readily to quantification. For purposes of the economic and financial analysis, the returns are fully estimated for benefit stream (i), in both the ‘with’ and ‘without’ subsidy situations. For benefit stream (ii), the returns are partially captured through the coffee price used that incorporates premiums for quality. Benefits are not computed for category (iii). These benefits are positive and enhance substantially overall project returns. Overall, as computed, the benefits are therefore conservative. 5. Economic and financial analysis. The financial analysis was prepared for crop enterprises concerning coffee, and intercrops (red beans and soybeans). For the economic analysis, benefits from individual enterprises were aggregated using the number of hectares covered each year of project implementation in the five targeted provinces. The total benefit streams were compared to project costs nets of transfers to derive the Net Present Value (NPV) and compute the project economic Internal Rate of Return (IRR). II. BURUNDI COFFEE FARMS AND PEASANT STRATEGIES 6. Farm characteristics. At appraisal, with about 10 million people (90% rural) and a small land mass of about 28,000 km2, Burundi’s had a rural population density amongst the highest in Africa. The targeted coffee areas were in the highlands of the country which exhibit cool temperatures (15 to 18 degrees Celsius year-round) and abundant rainfall (average 1650 mm). These areas have the highest population density exceeding 500 inhabitants per km2 in many areas. Smallholdings are the norm with farm size about 0.5 ha. Average rural household size is about six persons. 7. Given the small farm sizes, Burundian agriculture is very labor intensive. It is characterized by low management practices and little use of improved inputs. This applies to all crops equally, including coffee which receives little chemical fertilizer and pesticide application. All farms keep one or several animals, not only for milk and meat that are important for food security and nutrition, but also for manure used to fertilize crops. Livestock is also important for social status and is kept for emergency situations when cash is needed. 8. Subsistence and cash crops in peasant strategies. In Burundi, peasant strategies are governed by a twin objective of subsistence combined with profit maximization. Farmers typically grow both (i) food crops (e.g., beans, soybeans, maize, tubers) with the view to both obtaining the desired degree of food security and placing some surplus production on the market; and (ii) commercial crops (coffee essentially) to serve basic monetary needs and generate increased monetary proceeds to recapitalize means of production and embark on a sustainable growth path. Bananas are ubiquitous in all parts of Burundi and a major contributor to the peasant twin objective. In addition to cropping activities, farmers typically keep some animals (cows and/or goats). There are strong linkages between cropping and livestock activities through fodder crops and manure. Animals produce milk for home consumption and sale on the market, and manure used to fertilize crops. 9. Intercropping and low productivity. To best take advantage of their tiny farms, Burundian producers typically increase cropping intensity, by intercropping or undertaking two or more cropping cycles yearly on part of their farms. They practice intercropping when they replace or stump coffee trees. During the 2–3-year gestation time during which the trees grow or re-grow, farmers plants food crops in association with coffee to generate income at the time the coffee trees are not yet productive. In the quasi-absence of chemical fertilizer, manure is used extensively; but it is insufficient to maintain soil fertility. In recent decades, the depletion of soil fertility has accounted for the downward trend of food crops and coffee yields. Page 56 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) III. GROSS MARGIN AND FINANCIAL ANALYSIS AT FARMER LEVEL 10. Gross margin from coffee and associated crops. The project-generated gross margin has been computed for individual crop enterprises (coffee and associated crops), based on estimated incremental expenditures and revenues over the reference situation. With the project, as compared to the initial situation, the expenditures increase substantially. But revenues increase much further so that the gross margin is substantially higher. Coffee performance is very dependent on the linkages that exist between production and the other stages of the value chain (post-harvest, processing and marketing). Benefits will only materialize if these linkages exist. Hence, the project focuses on improving not only production, but also storage, processing and marketing of coffee. It supports public-private partnerships for that purpose. 11. With and without subsidy situation. To test the impact of the project support, the gross margin has been computed in both the ‘with’ and ‘without’ subsidy scenarios. Based on the ex-ante analysis, the results show that even without the subsidy the returns at farmer level were significantly positive: (i) for the rejuvenation program the Internal Rate of Return IRR was 36 percent; and (ii) for the fertilization program, the IRR was 76 percent. With the subsidy, the rate of return for participating farmers clearly increased markedly: (i) 123 percent for the rejuvenation program, and (ii) 117 percent for the fertilization program. The justification of the subsidy is to enable producers to cover the negative cash-flow they are likely to experience in the first years. Given their tight cash flow situation, producers would not be able to pay for the expenditures related to the replacement or stumping of trees. Indeed, they are left with a much-reduced cash income during the 2–3-year gestation period required for the trees to get to full production, even with the returns generated by associated crops. Similarly, they are expected to have a negative cash-flow the first year under the fertilization program. Hence, they were given a subsidy under the project: (i) for the rejuvenation program, the subsidy was on a decreasing schedule 100 percent the first year, 75 percent the second year and 50 percent the third year); (ii) for the fertilization program, the subsidy was 40 percent for three years; it was subsequently adjusted to 80 percent after the Mid-Term Review (MTR); and (iii) the pesticide treatments were provided free of charge. The ex-post analysis reveals that for the above two programs the IRR was as follows: (i) without subsidy: 92 and 96 percent respectively; and (ii) with subsidy 135 and 125 percent respectively. Owing to this substantial increase in profitability vs. the situation without project, the farmers were clearly highly motivated to embark on the project-supported programs. IV. OVERALL PROJECT PROFITABILITY MEASURES 12. Overall project benefits. Project-supported investments were expected generate substantial financial benefits for rural households in areas served by the project, as well as substantial economic benefits for Burundi’s society with additional taxes for the Treasury and foreign exchange for the Central Bank. As was done at project design, only the benefits generated from increased coffee production and associated crops have been quantified ex-post. These benefits capture indirectly the impact of CWS infrastructure and access roads through decrease in production losses and higher prices due to enhancement of quality. The benefits derived from capacity building of producers and other actors targeted at various levels of the value chains are not included. These benefits are important especially for the poorest and most vulnerable since these groups are expected to be better equipped to produce and market efficiently, and, in turn, improve their economic status. This result applies particularly to women and youth expected to represent 30 percent of project beneficiaries. 13. Economic returns. Table A4.1 below present a comparison of the results of the overall profitability measures computed at project design (ex-ante) and at project closure (ex-post). The project economic benefits computed over a 15-year period of analysis were expected ex-ante to yield a NPV of US$19.3 million and generate an economic rate of return of 22.7 percent. Based on ex-post estimates at project closure, the project economic benefits Page 57 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) computed were expected to yield a NPV of US$2.3 million and generate an economic rate of return of 6.9 percent. The results ex-post are marginally positive and denote a modest overall efficiency measure. 14. Sensitivity analysis. In the ex-ante situation at project design the sensitivity analysis showed that projected benefits were robust as far as benefits were concerned. If the gross margin were to drop by 30 percent, either from price or quantity produced, the IRR was expected to drop to 12.4 percent. In the ex-post situation, the IRR is expected to drop to 1.6 percent if benefits are reduced by 30 percent. This means that there is a high risk that the project would not be positive as far as economic profitability is concerned, and, hence, that its overall efficiency would be jeopardized if benefits decrease significantly. This is not an unreasonable scenario given the volatility of coffee price on the international market. The sensitivity to price means that every effort should continue to be made to facilitate production collection, processing and marketing, as well as partnerships between farmers, processors and offtakers, to produce quality coffee that commands a higher and more stable price. The sensitivity to production (yield) level implies that continued close support should be given to producers so that they adhere strictly to technology prescriptions and adopt the new technology developed by research. Every effort should be made to ensure that input delivery services are readily available. Table A4.1 - Comparison of Economic Profitability Measures in the Ex-post vs. Ex-Ante Situation Ex-Ante Ex-Post Profitability Measures % • IRR basic scenario 22.7 6.9 • IRR 30% decrease in benefits 12.4 1.6 US$ million • NPV 19.3 2.3 15. Project fiscal impact and foreign exchange earnings. The project was expected to generate US$6.3 million in taxes for the Government’s Treasury and US$51.7 million in foreign exchange earnings over its implementation period. The ex-post analysis reveals that only US4.1 million were generated in taxes for the Government or 65 percent of the ex-ante estimate, and US$22.6 million in foreign exchange earnings on the basis of coffee sales or 44 percent of the ex-ante estimate. Page 58 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) ANNEX 5. GREENHOUSE GAS ACCOUNTING 1. Corporate mandate. In its 2012 Environment Strategy, the World Bank adopted a corporate mandate to conduct greenhouse gas (GHG) emissions accounting for investment lending in relevant sectors. The ex-ante quantification of GHG emissions is an important step in managing and ultimately reducing GHG emissions and is becoming a common practice for many international financial institutions. 2. Methodology. To estimate the impact of agricultural investment lending on GHG emissions and carbon sequestration, the World Bank has adopted use of the Ex-Ante Carbon-balance Tool (EX-ACT), developed by the Food and Agriculture Organization of the United Nations (FAO) in 2010. EX-ACT allows the assessment of a project’s net carbon-balance, defined as the net balance of CO2 equivalent GHG that were emitted or sequestered because of project implementation compared to a without project scenario. EX-ACT estimates the carbon stock changes (emissions or sinks), expressed in equivalent tons of CO2 (tCO2-eq) per hectare and year. 3. Project boundary. The EX-ACT modules used included land use change, crop production, and inputs. Project planned interventions are summarized below. (a) Rejuvenation of old unproductive coffee trees – new trees (Sub-component 2.1-Rejuvenation of existing plantations) Baseline = 0 ha. Project by 2021 = 6554.17 ha. (b) Rejuvenation of old unproductive coffee trees – stumping of old trees (Sub-component 2.1- Rejuvenation of existing plantations) Baseline = 0 ha. Project by 2021 = 7648.75 ha. (c) Introduction of coffee under shade in existing coffee plantations (Sub-Component 2.3 Generation and dissemination of sustainable agricultural practices) Baseline = 0 ha. Project by 2021 = 253 ha. (d) Intercropping with beans and soybeans (Sub-Component 2.3 Generation and dissemination of sustainable agricultural practices). Baseline = 0 ha. Project by 2021 = 37041 ha. (e) Introduction of Good Agricultural Practices (Sub-components 2.3: Generation and dissemination of sustainable agricultural practices). These include Sustainable Land Management (SLM) practices, nutrition and fertilizer management, Integrated Pest Management, pruning, harvesting, quality assessment and improvement. (f) Application of fertilizers (Sub-Component 2.2: Improving access to fertilizer and agro-chemicals). Baseline: 1200 tons of fertilizers are used, while it increased to 6951.55 tons in 2021. Table A5.1: Total volume of fertilizer provided by the project Year 2017 2018 2019 2020 2021 NPK+S (ton) 1,300.00 0.00 0.00 3,953.05 0.00 DAP (ton) 0.00 1,661.00 0.00 0.00 0.00 Urea (tons) 0.00 37.50 0.00 0.00 0.00 (g) Application of pesticides (Sub-Component 2.2: Improving access to fertilizer and agro-chemicals). Baseline: All plantations are treated again bugs. Insecticide Lambdalm and Imidacropne have been applied. In addition, the project also used fungicides to suppress the spread of fungi. 4. Baseline scenario. Coffee plantation baseline. Coffee plantations were estimated at 70,000 ha with 122 million coffee trees. Baseline yield per tree was 0.8 kg; the yield increased to 1.2 kg/tree by 2020. Page 59 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) 5. Data sources. Project team. 6. Key assumptions. The project region in Burundi has tropical climate with moist moisture regime. The dominant soil type is low activity clay. The project implementation phase is 4 years (2017-2020), and the capitalization phase is assumed to be 20 years. The 20 years capitalization period is standard in the use of EX-ACT if there is agroforestry in the project. 7. Results. The net carbon balance quantifies GHGs emitted or sequestered as a result of the project compared to the without project scenario. Over the project duration of 20 years, the project generating gross GHG emissions -1,457,416 tCO2-eq, constituting a net GHG emissions (a carbon sink) of -1,523,832 tCO2-eq. On that basis the project provides a sink of -63,493 tCO2-eq per year, equivalent to -1.2 tCO2-eq per ha per year. The sink results largely from improved practices, namely, rejuvenation of old unproductive trees, stumping of old trees and shaded coffee practices. Intercrops (annual) also contribute to net GHG sink (518,574 tons of CO2 equivalents). However, increased input use will create extra 310,981 tons of CO2 equivalent carbon emission. Finally, the increase in C sequestration will lead to other co-benefits including improved biodiversity, reduced soil erosion and enhanced agro-ecosystem resilience. Table A5.2: Results of the ex-post GHG analysis Over the economic project lifetime (tCO2 eq) Annual average (tCO2 eq/ year) GHG emissions GHG emissions of Gross GHG emissions Net Gross emissions of without without Net GHG emissions Project activities with project scenario GHG emissions project scenario project scenario project scenario (4-3) (2) (2-1) (4) (1) (3) Land use change 0 1,849,143 1,849,143 0 77,048 77,048 Intercrops 0 -518,574 -518,574 0 -21,607 -21,607 Improved practices 0 -3,165,382 -3,165,382 0 -131,891 -131,891 Inputs use 66,416 377,397 310,981 2,767 15,725 12,958 Total 66,416 -1,457,416 -1,523,832 2,767 -60,726 -63,493 Per hectare 1 -28 -30 0.1 -1.2 -1.2 Page 60 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS As of March 28th, 2022, the World Bank has not received any borrower comments on this report. Page 61 of 62 The World Bank Burundi Coffee Sector Competitiveness Project (P151869) ANNEX 6. MAP Page 62 of 62