Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00004015 IMPLEMENTATION COMPLETION AND RESULTS REPORT < IDA-53620-H3020> ON A IDA CREDIT IN THE AMOUNT OF SDR 15.2 MILLION (US$ 22.5 MILLION EQUIVALENT) TO THE Republic of Congo FOR THE AGRICULTURAL DEVELOPMENT AND RURAL ROADS REHABILITATION PROJECT ( P095251 ) October 25, 2017 Agriculture Global Practice Africa Region CURRENCY EQUIVALENTS (Exchange Rate Effective {Oct 05, 2017}) Currency Unit = XAF 559 = US$1 US$ 1.41 = SDR 1 FISCAL YEAR January 1 - December 31 Regional Vice President: Makhtar Diop Country Director: Ahmadou Moustapha Ndiaye Senior Global Practice Director: Juergen Voegele Practice Manager: Dina Umali-Deininger Task Team Leader(s): Amadou Oumar Ba Jeehye Kim ICR Main Contributor: Majid Benabdellah ABBREVIATIONS AND ACRONYMS ADRRP: Agricultural Development and Rural Roads Rehabilitation Project ADS: Departmental Monitoring Office AF: Additional Financing AIDS: Acquired Immune Deficiency Syndrome AWPB: Annual Work Plan and Budget CADS: Head of the Departmental Monitoring Office CC: Conversion coefficient CDTA: Center for Demonstration of Agricultural Technologies CNLP: National Poverty Reduction Committee CNSA: National Center for Improved Agricultural Inputs CNSEE: National Center for Statistics and Economic Studies CPPA: Planning Framework for Indigenous Populations CRAL: Loudima Agriculture Research Center CSA: Agricultural Sector Head CVTA: Center for Dissemination of Agricultural Technologies CVTE: Center for Dissemination of Livestock related Technologies DC: Departmental Councils DDEPW: Departmental Direction of Equipment and Public Works EIRR: Economic internal rate of return ERR: Economic rate of return ESMF: Environment and Social Management Framework FAO: Food and Agriculture Organization of the United Nations FIRR: Financial internal rate of return FRR: Financial rate of return GEF: Global Environmental Facility HIEQ: Highly-equipment intensive works HIPC: Heavily-Indebted Poor Countries HIV: Human Immunodeficiency Virus HLI: Highly-labor intensive IBRD: International Bank for Reconstruction and Development ICR: Implementation Completion and Results Report IDA: International Development Association IFAD: International Fund for Agricultural Development IGA: Income-generating activities IMF: International Monetary Fund INS: National Institute for Statistics IP: Initial Project IPPF: Indigenous People’s Planning Framework IR: Intermediate Results IT Information technology M&E: Monitoring and Evaluation MAEP: Rural Engineering Managers MALF: Ministry of Agriculture, Fisheries and Livestock MEER: Ministry of Infrastructure and Road Maintenance MFI: Microfinance Institution MI: Market infrastructure MPA: Ministry of Fisheries MTEF: Medium-Term Expenditure Framework MTR: Mid-Term Review NGO: Non-governmental Organization NPV: Net present value OAC: Support and Advisory Service Provider (Opérateur d’Appui Conseil) OCC: Opportunity cost of capital PAD: Project Appraisal Document PCR: Project Completion Report PCU: Project Coordination Unit PDO: Project Development Objective PID: Project Information Document PIPC: Interim Post-Conflict Program PMP: Pest Management Plan PND: National Development Plan PRSP: Poverty Reduction Strategy Paper RAP: Resettlement Action Plan RF: Road Fund RPF Resettlement Policy Framework TOR: Terms of Reference TTL: Task Team Leader VAT: Value-added tax XAF: Central African CFA Franc TABLE OF CONTENTS DATA SHEET .................................................................................... ERROR! BOOKMARK NOT DEFINED. I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ...................................................................7 A. CONTEXT AT APPRAISAL .........................................................................................................7 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) ..................................... 10 II. OUTCOME ................................................................................................................................ 13 A. RELEVANCE OF PDOs ............................................................................................................ 13 B. ACHIEVEMENT OF PDOs (EFFICACY) ...................................................................................... 14 C. EFFICIENCY ........................................................................................................................... 19 D. JUSTIFICATION OF OVERALL OUTCOME RATING .................................................................... 21 E. OTHER OUTCOMES AND IMPACTS (IF ANY) ............................................................................ 22 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME............................................. 23 A. KEY FACTORS DURING PREPARATION ................................................................................... 23 B. KEY FACTORS DURING IMPLEMENTATION ............................................................................. 24 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME ............... 26 A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................ 26 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE ..................................................... 28 C. BANK PERFORMANCE ........................................................................................................... 30 D. RISK TO DEVELOPMENT OUTCOME ....................................................................................... 31 V. LESSONS AND RECOMMENDATIONS ......................................................................................... 32 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................................ 34 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ...................................... 46 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................................ 48 ANNEX 4. EFFICIENCY ANALYSIS ....................................................................................................... 49 ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS ................. 60 ANNEX 6. SUPPORTING DOCUMENTS (IF ANY) .................................................................................. 63 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name AGRICULTURAL DEVELOPMENT AND RURAL ROADS P095251 REHABILITATION PROJECT ( P095251 ) Country Financing Instrument Congo, Republic of Specific Investment Loan Original EA Category Revised EA Category Partial Assessment (B) Partial Assessment (B) Related Projects Relationship Project Approval Product Line Additional Financing P145627-Additional 29-Jan-2014 IBRD/IDA Financing for the Agricultural Development and Rural Roads Rehabilitation Project Organizations Borrower Implementing Agency Minsitry of Finance Ministry of Agriculture, Livestock and Fisheries Page 1 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Project Development Objective (PDO) Original PDO The primary objective of the project is to assist the Republic of Congo to increase the ability of the rural poor to raise their income through the generation and promotion of improved agricultural technologies, the provision of market infrastructure, and the formulation and implementation of poverty-focused agricultural policies and expenditure programs. FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 20,000,000 20,000,000 20,858,942 IDA-H3020 2,500,000 2,500,000 2,202,047 IDA-53620 Total 22,500,000 22,500,000 23,060,989 Non-World Bank Financing Borrower 0 0 0 Total 0 0 0 Total Project Cost 22,500,000 22,500,000 23,060,989 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 05-Jun-2007 03-Apr-2008 21-Jan-2011 31-Dec-2012 30-Apr-2017 Page 2 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 30-Dec-2012 17.04 Change in Project Development Objectives Change in Loan Closing Date(s) Reallocation between Disbursement Categories Change in Safeguard Policies Triggered 10-Oct-2013 20.77 Change in Loan Closing Date(s) Reallocation between Disbursement Categories 29-Nov-2013 20.77 Change in Loan Closing Date(s) 17-Aug-2015 21.18 Change in Loan Closing Date(s) Reallocation between Disbursement Categories 24-Jun-2016 22.25 Change in Loan Closing Date(s) KEY RATINGS Outcome Bank Performance M&E Quality Moderately Satisfactory Moderately Satisfactory Modest RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 20-Dec-2007 Satisfactory Satisfactory 0 02 24-May-2008 Moderately Satisfactory Moderately Unsatisfactory 0 03 12-Dec-2008 Moderately Satisfactory Moderately Satisfactory 1.75 04 11-Jun-2009 Moderately Satisfactory Moderately Unsatisfactory 2.75 Moderately 05 15-Dec-2009 Moderately Unsatisfactory 3.29 Unsatisfactory Moderately 06 10-Jun-2010 Moderately Unsatisfactory 3.65 Unsatisfactory Moderately 07 29-Mar-2011 Moderately Satisfactory 4.98 Unsatisfactory 08 07-Jul-2011 Moderately Satisfactory Moderately Satisfactory 5.41 Page 3 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) 09 03-Jan-2012 Moderately Satisfactory Moderately Satisfactory 6.60 10 20-Mar-2012 Moderately Satisfactory Moderately Satisfactory 7.32 11 19-Oct-2012 Moderately Satisfactory Moderately Satisfactory 9.54 12 29-May-2013 Satisfactory Satisfactory 12.58 13 25-Nov-2013 Satisfactory Satisfactory 13.44 14 17-Jun-2014 Satisfactory Moderately Satisfactory 13.50 15 09-Jan-2015 Satisfactory Moderately Satisfactory 13.50 16 29-Jun-2015 Moderately Satisfactory Moderately Satisfactory 13.68 17 01-Feb-2016 Satisfactory Satisfactory 14.04 18 05-Jun-2016 Satisfactory Satisfactory 14.31 19 31-Dec-2016 Moderately Satisfactory Moderately Satisfactory 14.88 SECTORS AND THEMES Sectors Major Sector/Sector (%) Agriculture, Fishing and Forestry 100 Fisheries 4 Crops 27 Livestock 4 Public Administration 100 Central Government (Central Agencies) 17 Transportation 100 Rural and Inter-Urban Roads 40 Page 4 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Industry, Trade and Services 100 Other Industry, Trade and Services 8 Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Private Sector Development 100 Jobs 100 Finance 7 Finance for Development 7 Agriculture Finance 7 Public Sector Management 14 Public Finance Management 7 Public Expenditure Management 7 Public Administration 7 Transparency, Accountability and Good 7 Governance Urban and Rural Development 79 Rural Development 79 Rural Markets 36 Rural Non-farm Income Generation 14 Rural Infrastructure and service delivery 29 ADM STAFF Role At Approval At ICR Regional Vice President: Obiageli Katryn Ezekwesili Makhtar Diop Country Director: Pedro Alba Ahmadou Moustapha Ndiaye Senior Global Practice Director: Juergen Voegele Juergen Voegele Practice Manager: Joseph Baah-Dwomoh Dina Umali-Deininger Page 5 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Task Team Leader(s): Remi Kini Amadou Oumar Ba ICR Contributing Author: Jeehye Kim Page 6 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL Context 1. In 2000, the Government of the Republic of Congo prepared an Interim Post-Conflict Program (PIPC) aimed at the rehabilitation of basic infrastructure, and a strengthening of investment capacity in the social sectors, including job creation, as well as the initiation of an anti-poverty policy. However, over two million inhabitants (or 50 percent of the population) continued to face poverty, food insecurity and malnutrition issues1. Following on the actions envisaged in the PIPC, the 2004 Poverty Reduction Strategy Paper (PRSP) was developed with four pillars: ensuring the peace and supporting good governance; consolidation of macroeconomic conditions and reorganization of key sectors; improving access to basic social services and social welfare; and infrastructure development. In this context, the Government decided to revive the production sectors, in particular: agriculture, fisheries production, forestry, and livestock. In this context, the government sought to increase public investments with the aim of stimulating economic growth and reducing poverty. 2. The 2006 Agricultural Policy Note identified several constraints to sustained growth in the agricultural sector, specifically: the weak asset base of farm households and rural producers; inadequate market infrastructure and services; inadequate rural road infrastructure; ineffective sector policies and poor public expenditure management; and the weak implementation capacity of public institutions. The World Bank -supported Agricultural Development and Rural Road Rehabilitation Project (ADRRP) aimed at expanding the Government’s sphere of action in agricultural and rural development. Joint World Bank-Government cooperation in preparing the Poverty Reduction Strategy contributed to the development of the ADRRP. The ADRRP also benefited from a favorable institutional context and participatory process, made possible by the establishment of the National Poverty Reduction Committee (CNLP) in October 2001. Theory of Change (Results Chain) 3. The high-level objective of the project is increasing the ability of the rural poor to raise their income. The project aims to contribute to this objective by enabling poor farmers to access improved knowledge, technology, market infrastructure, and by addressing some of the most binding obstacles to agricultural development, including weak and uncoordinated sectoral policies. Project activities are designed to achieve the main intermediate results (IR) of the project, which ultimately contribute to its PDO. The IRs include: capacity building for officials responsible for 1 Congo Poverty Profile, Republic of Congo, 2005, Ministry of Planning, Spatial Planning and Economic Integration, CNSEE. Page 7 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) agriculture, fisheries and livestock production (IR 1); improving access to reliable, all-season production areas and marketing services (IR 2); and improving the farming methods and productivity of rural producers (IR 3). 4. Local actors, such as the administration or producers, are at the heart of the project intervention. Therefore, training, organizing and supporting actors is essential to increasing their capacities — and to changing behavior, relationships and actions necessary for the sustainable achievement of the PDO. In this respect, the project intervention logic revolves largely around capacity-building operations, on the principle that the key to the expected changes is held by local actors. 5. The project finances technical capacity-building activities. Specifically, it provides financial support and technical assistance to MALF staff and their decentralized services. The aim of these operations is to remedy the problems related to poor implementation capacity by public institutions, as well as poor management of public expenditures. The technical assistance also aims at addressing the inefficiency of sectoral policies as one of the factors limiting the development of agriculture in the Republic of Congo. 6. By providing targeted training to staff and technical support to government institutions, the project contributes to the improvement of technical and operational capacities. These improvements in the performance of administrative staff would in turn make it possible to ensure the sustainability of agricultural support and extension services, as well as strengthening the supervision of producer groups. 7. As the project aims to improve the ability of rural poor (farmers) to increase their income, the project finances the rehabilitation of input production centers, the supply of equipment and infrastructure for producers, and the training and financing of agricultural producers (through producer groups). The expected benefits from these project activities are numerous, and include: the adoption of improved production technologies by producers; increased areas under cultivation; and higher crop yields by producers and rural households. 8. Inadequate and inefficient infrastructure in the project area was the biggest constraint to increasing production and incomes. In order to eliminate this obstacle, the project was designed to target interventions exclusively for the rehabilitation and/or construction of rural roads and markets (storage, distribution, sales, and so on). Interventions in road and market infrastructure facilitate access to quality inputs, advisory services, and storage units. Through this support, it was hoped that production areas and villages would open and be connected through markets. They also help to provide markets so that producers/farmers can sell their crops at sustainable prices. In addition, the project would help to reduce travel time and costs, thereby reducing post-harvest losses. 9. The fulfillment of the project PDO is, however, dependent on certain conditions. These involve the assumptions related to the country's context, such as: the proper functioning of the road fund for the maintenance of rural road infrastructure; the reduction of police barriers at market entries; the increase in the autonomy of producer groups; and the political stability of the country (Annex 8). Project Development Objectives (PDOs) 10. The PDO of the ADRRP is to assist the Republic of Congo in increasing the ability of the rural poor to raise their incomes through the generation and promotion of improved agricultural technologies, the provision of market infrastructure, and the formulation and implementation of poverty-focused agricultural policies and expenditure programs. Page 8 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Key Expected Outcomes and Outcome Indicators 11. Outcome 1: Strengthened public capacity to formulate and implement poverty-oriented agricultural policies. This objective can be achieved through effective and efficient spending programs; improving the alignment of spending programs with government priorities; and increasing the effectiveness and efficiency of interventions. The key indicators under Outcome 1 of the Project are as follows: (i) the MALF’s annual budget execution rate increases from 45 to at least 80 by the end of the project; (ii) an increase in the rate of coverage of agricultural services from 5 to 58 percent; and (iii) 50% of annual budget is transferred to the departmental departments. 12. Outcome 2: Improvement of production and living conditions for beneficiaries in rural areas. This objective will be achieved through the provision of adequate road and market infrastructure, and subsequent increase in agricultural productivity and income. Key indicators under Outcome 2 are: (i) Increased productivity of key agricultural products with end target of cassava- 10.6 ton/ha, banana- 75 ton/ha, and maize- 3.5ton/ha; (ii) the income of the beneficiary households is increased by at least 50 percent2 at project end; (iii) at least 75 percent of the improved feeder roads are maintained at project end; and (iv) at least 90 percent of the newly-built market infrastructure is properly used and managed at project end. Components 13. The total project cost of the ADRRP is US$50.5 million. This amount was financed jointly by the Republic of Congo and the International Development Association (IDA) in two stages: 1) the parent project for US$40 million ($20 million (IDA-H3020), $20 million government) with a project period from June 5, 2007 to March 31 20143; and 2) an additional financing in the amount of US$ 10.5 million ($2.5 million (IDA-53620), $8 million government) with a project period from January 29, 2014 to April 30, 2017. The IDA-funded portion of the ADRRP cost totaled US $ 22.5 million. The project has four components, namely: 14. Component 1: Building the capacity of the MALF: This component aimed to strengthen the capacity of the MALF with regard to policy formulation, and budgeting and public expenditure management to increase the efficiency and effectiveness of poverty-focused spending programs. IDA financing for this component amounted to US $ 2.86 million, of which US $ 2.36 million was allocated for the initial phase and another US $ 0.5 million for the AF. This amounted to 40 percent of the total cost of the component, which was US$ 7 million, of which US$ 5.5 million was for the initial project and US$D 1.5 million for the AF. The main activities were to support the planning and monitoring and evaluation (M&E) units at the central level of the Ministry to ensure adequate monitoring of project activities. Targeted support was given to departments to provide agricultural services to beneficiaries. These activities also extended to the support of entities involved in research, production and distribution of seeds and cuttings (the National Center for Improved Agricultural Inputs [CNSA], and the Loudima Agricultural Research Center [CRAL]), as well as to entities from ministries involved in providing agricultural extension services to small 2 This target was formally revised from 10% to 50% at the additional financing stage. 3 Original closing date of the parent project was Dec 31, 2012. Page 9 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) farmers4. 15. Component 2: Rehabilitating rural roads and market infrastructure: The objective of this component is to provide and maintain roads that are reliable in all seasons, connecting rural communities with markets and other social services. To achieve this, the project financed two types of activities: the rehabilitation and maintenance of rural roads, and the construction/rehabilitation of rural market infrastructure. IDA financed the completion of this component for US$ 8.97 million, of which US$ 7.97 million was allocated to the initial phase and US$ 1.0 million to the AF phase. This corresponds to almost 38 percent of the total for this project component, which amounts to US $ 23.8 million, of which US $ 18.3 million is for the initial phase of the project and US $ 5.5 million for additional funding. 16. Component 3: Supporting productive and income-generating activities: The objective of this component is to enhance the capacity of rural producers to increase their productivity and income. In this context, the main activities of the project consisted of implementing a series of complementary operations in line with the needs of farmers5. In addition, activities also focused on supporting micro-projects of producer groups through a matching grant mechanism. IDA's contribution to this component amounted to US$ 4.8 million, or US$ 3.8 million and US$ 1 million, respectively, for the initial and AF project phases. The total cost of the component was US$ 10.7 million, of which US$ 8.7 million was allocated for the initial phase and US$ 2 million for the AF. 17. Component 4: Project management and coordination: This operational component aims to ensure the management, coordination, M&E and environmental and social monitoring of project activities. It was implemented by the PCU. The IDA-supported cost of this component amounted to US$ 2.98 million in the initial phase of the project. However, the entire component cost nearly US$ 9 million, including US$ 7.5 million for the initial project, and US$ 1.5 million for the AF. 18. In addition to these components, it should be noted that IDA mobilized US$ 2.89 million for project preparation (US$ 0.29 million), and an additional amount to guard against project risks and unforeseen events (US$ 2.6 million). B. SIGNIFICANT CHANGES DURING IMPLEMENTATION (IF APPLICABLE) Revised PDOs and Outcome Targets 19. Per the results framework of the parent Project Appraisal Document (PAD), the PDO was “to increase rural producers’ ability to raise their productivity and income by generating and disseminating improved agricultural technologies, providing market infrastructure, and implementing poverty focused agricultural policies and expenditure programs�. However, in the Financing Agreement, the PDO was stated as: “to enhance the Recipient's capacity to formulate and implement agricultural policies aimed at reducing poverty in the Recipient's rural areas 4 CAT (Centre d’Appui Technique), AGRI-CONGO, Center for Demonstration of Agricultural Technologies (CDTA), and the Center for Dissemination of Livestock-related Technologies (CVTE). 5 Creation and diffusion of improved production technologies, drying facilities, acquisition of small-scale processing and processing equipment for farmers, installation of collection and distribution platforms for agricultural and fishery products, and market information and capacity building of producer associations. Page 10 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) namely by: (a) preparing and implementing poverty-focused budget and expenditure programs; (b) developing adequate market infrastructure; and (c) generating and disseminating improved agricultural technologies.� 20. At the Project Level 1 Restructuring in December 30, 2012, the PDO was revised as follows: "The objective of the Project is to assist the Republic of Congo to increase the ability of the rural poor to raise their incomes through the generation and promotion of improved agricultural technologies, the provision of market infrastructure, and the formulation and implementation of poverty-focused agricultural policies and expenditure programs." Revised PDO Indicators 21. The PDO level indicators of the parent project were revised during the 1st project restructuring on December 20, 2012. One PDO level indicator regarding the number of beneficiaries was added at the Additional Financing approved on January 29, 2014. Original Revised during 1st project level 1 Revised and Added restructuring during Additional Financing (December 20, 2012) (January 29, 2014) 1. The production of at least Increased productivity of key Increased productivity of key three of the following crops: agricultural products by 80 agricultural products with cassavas, bananas, maize, and percent (cassavas, bananas and end target of peanuts are increased by 20 percent maize). Cassava : 10.6 ton/ha (compared to the 2008 baseline Banana : 75 ton/ha in project areas) at the end of Maize: 3.5ton/ha the project. 2. Producer income increased Increased income of project Increased income of project by at least 10 percent by the beneficiaries participating in beneficiaries participating in end of the project (in project micro-projects by 30 percent at micro-projects by 50 percent areas). the end of the project. at the end of the project. 3. Number of direct beneficiaries of the project totaled 321,600 at the end of the project, of which 51.6 percent were women. Page 11 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Revised Components 22. N/A Other Changes 23. Level 1 Restructuring (2012). The project underwent a level 1 restructuring, which was approved on December 30, 2012. Through this restructuring, the wording of the PDO between the PAD and the AF was harmonized (paragraph 19-21). It also triggered the policy of safeguarding the indigenous peoples, the planning framework for which was prepared from July12, 2011 until June 1, 2012. The restructuring consolidated the institutional foundations for the inclusion of aspects related to the socio-environmental safeguarding policy of the project. The original closing date of Dec 31, 2012 was extended to June 30, 2013. 24. Level 2 Restructurings (2013~2014). The project underwent a level 2 restructuring, which was approved on Oct 10, 2013. The restructuring entailed a change in loan closing date from June 30, 2013 to December 30, 2013, and a reallocation between disbursement categories to provide resources to Mid-Term Expenditure Framework. The project underwent another level 2 restructuring (approved on November 29, 2013) to extend the loan closing date from December 30, 2013 to March 31, 2014 to avoid intermission of the project until the AF is approved by the board. 25. Additional Financing (2014). The additional financing to ADRRP was approved on January 29, 2014. The additional financing scaled up project activities related primarily to infrastructure and micro-projects. It has therefore planned to finance the rehabilitation of 80 km of rural roads, 4 market infrastructures, as well as over 100 additional micro- projects. It sought to strengthen the synergies between activities by concentrating activities in certain geographical areas or “growth poles� with a high level of agricultural production potential. In addition, it introduced a value chain approach in the choice of micro-projects. Finally, performance indicators were revised to reflect the increased funding. 26. The results framework of the parent ADRRP project was revised when the additional financing was processed. Among the main changes related to the PDO indicators are the following:  The introduction of two new performance indicators, namely "the number of direct beneficiaries of the project (target: 321,600)" and the "percentage of female beneficiaries (target: 51.6 percent)�;  The revision of the indicator, "Increase in the income of the project beneficiaries registered at the end of the project", the target value of which increased from 15 percent in the PAD to 50 percent for the end of the project in 2015;  The indicator, "Increase in the execution rate of the annual budget of the MALF and the MPA from 45 to 80 percent," was reduced in rank from the PDO (initially in the PAD) to intermediate results 1. 27. Also noteworthy is the removal of a performance indicator for component 4 (project coordination), namely the indicator that "At least 95 percent of procurement (for goods, services and public works) processes are in compliance with IDA procedures.� 28. In summary, the main changes to the ADRRP results framework is as follows: i) two new outcome indicators were introduced at the PDO level; ii) two indicators related to the project coordination component's outcome indicators Page 12 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) were abandoned; and iii) the 17 intermediate outcome indicators, were revised. These revisions consisted mainly of the upward review of the indicators’ target values, which were mostly reached by the end of the initial phase of the project (2013). 29. Level 2 Restructurings (2015 ~2016). The project underwent a level 2 restructuring, which was approved on August 17, 2015. The restructuring entailed a change in loan closing date from October 31, 2015 to June 30, 2016, and a reallocation between disbursement categories. The reallocation of proceeds consisted of reducing IDA financing for the support to sustainable livelihoods (component 3) to reinforce the rural infrastructure component, which was deemed underestimated. The project underwent another level 2 restructuring, approved on June 24, 2016, to extend the loan closing date from June 30, 2016 to April 30, 2017 to ensure completion of project activities. II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating: Substantial 30. Relevance of objectives. The objectives of the project are in line with the country's Poverty Reduction Strategy Papers (PRSP I & II), and are well aligned with national strategies for poverty reduction and agricultural development (the National Development Plan [PND] of 2012-2016). The objectives of the ADRRP are also in line with the World Bank’s road map to support the Congolese Government's Interim Post-Conflict Program as part of its Transition Strategy (2003). This Strategy aimed to create jobs through economic growth and diversification, infrastructure rehabilitation, education and rural development. The ADRRP is also consistent with the World Bank’s Country Partnership Framework, which included support for agricultural and rural development, as well as the transformation of family farming as one of the priority areas for increasing public investment to benefit the poor who were affected by the country’s long-term socio-political instability. 31. Relevance of design. The ADRRP was conceived in the context of a socio-political crisis in the Republic of Congo, at a time when poverty and food insecurity prevailed throughout the country. During this time, there was also no real agricultural strategy for the country. The degradation of production factors and the inability of local actors to implement genuine development strategies were serious constraints to agricultural development and the revival of the country's economy. As a result, the design of the ADRRP aimed to restore the productive bases and solve the structural problems of agricultural development. Indeed, it corresponds well to the country's concerns about poverty reduction, diversification of production and attainment of food security. Similarly, the project focused on Page 13 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) strengthening the capacities of administrative actors, both at the central and decentralized levels, as well as the producers in their various roles. 32. Relevance to the needs of the main beneficiaries. The farming households targeted by the project live almost entirely on subsistence activities. The crisis experienced by the Republic of Congo has considerably reduced the economic prospects of these areas, with an attendant decrease in incomes. The isolation of production areas and the limited access to basic social services have led to a deterioration in living conditions. This has occurred in a context where the possibility of undertaking income-generating activities (IGAs) was already very limited. The project responds to these challenges in terms of upgrading productive potential and socio-economic infrastructure and improving access to markets. The ADRRP is relevant to the needs of these subsistent farming households, although some have not been sufficiently considered. The project’s relevance is further enhanced by the fact that particular attention is paid to women, who represent 51 percent of the country's population. 33. Relevance of implementation. The approach adopted for the implementation of the project was fully relevant. The context was characterized by the lack of capacity of the administrative actors to develop and implement coherent expenditure policies. Decentralized entities lacked the means to implement services without extension structures. As such, pending the capacity building of local actors, it was necessary to conduct the project through a team of international experts with attendant technical assistance. The PCU entrusted the execution of the operational activities of the project to various agencies and partners as well as technical providers of the public sector and the private sector, while ensuring the programming, control and monitoring-evaluation. This strategic partnership with private entities was beneficial for the successful implementation of the project, particularly for the construction of roads and market infrastructure. During the preparation of the FA, the MALF hired an international consultant who conducted a study that includes the stock taking of the status of all rehabilitated tracks for the 5 years prior to the additional financing by all donors combined and offered a draft of a government strategy to ensure optimization and sustainability of resources mobilized for rehabilitation of rural roads. Furthermore, through implementation, the project enabled the producers to organize in groups and work closely with local support services as well as in cooperation with the different ministries (the MALF, the MEER, and so on). B. ACHIEVEMENT OF PDOs (EFFICACY) Assessment of Achievement of Each Objective/Outcome: substantial 34. Achievement of project PDO. The ADRRP implemented its activities and achieved its objectives. Indeed, achieving results in terms of strengthening the capacity to formulate and implement poverty-oriented agricultural policies not only optimized the management of the MALF budget, but also the poor through the mobilization of decentralized services and better targeting of services. The project efficacy is assessed in relation to the achievement of its two outcome areas: 35. Outcome/ Objective 1: Strengthening the capacity of the MALF to formulate and implement poverty-oriented agricultural policies. To achieve this objective, 432 MALF staff were trained in: public budgeting and planning; Page 14 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) procurement and public financial management; monitoring and evaluation; results-based management; production- integrated protection; administrative and financial management; information technology (IT); survey techniques, data processing and analysis; and environmental and social safeguard measures. The project also supported and funded a study for a strategic program for agricultural sector budget reform (2012). With funding support for international consultants and one-year of technical assistance (2011-2012), the ADRRP also worked closely with the General Census of Agriculture. 36. The training provided by the project has strengthened the capacity of MALF staff in the development of agricultural policies in relation to the country's development strategy. The results of the General Census of Agriculture also provided essential information to facilitate the design and distribution of resources dedicated to the agricultural sector for the poor. This is reflected in the preparation of the MALF’s Agricultural Policy Letter, s upported by the strategy and main points of the National Development Plan (2012-2016). Among other things, it should be noted that the implementation of the analytical and strategic budgetary reform program for the agricultural sector has enabled the MALF to improve its expenditure management system. Specifically, the MALF reviewed its spending in the agricultural sector and adopted a Medium-Term Expenditure Framework (MTEF). In addition, project support (through training and technical assistance for procurement) led to the upgrading of public procurement instruments and the creation of procurement management units within the MALF and the MEER6. 37. The following key results showcase the contribution of the above project activities to the strengthening of capacity within the MALF in the development, implementation and monitoring of agricultural policies, as well as the effective management of public expenditure:  Between 2013 and 2015, the MALF adopted three annual budgets with the objective of sectoral programs designed to fight poverty.  The implementation rate of the annual budget of the MALF increased from 45 percent in 2010 to 62 percent in 2012.  The percentage of public contracts awarded by the MALF, in accordance with accepted procedures, increased from 0 percent in 2008 to 8 percent in 2011, and to 28 percent in 2012. It reached 100 percent in 2016, thereby exceeding the target value of 80 percent at the end of the project.  The MALF made significant progress in terms of transferring the annual budget to Departmental Directorates in 2015, with a rate of 83.5 percent compared with 33.6 percent in 2013. However, the level of these transfers has fallen to 29 percent in 2017, a target achievement rate of only 60 percent7. 38. The project also supported decentralized services8 and specialized implementing entities9 in their outreach, research, and support to producers. This support consisted of the provision of modern operating equipment, rolling stock as well as the provision of financial assistance. These targeted interventions with the MALF’s decentralized entity structures resulted in an increase in the coverage of agricultural services and technical assistance in the project area from 5 percent in 2008 to 58 percent in 2017, as compared to the original forecast of 30 percent. 6 Decree n° 2009 - 160 of May 20, 2009 fixing the thresholds for public procurement, control and approval of procurement in the Republic of Congo. 7 The government faced a major fiscal crisis due to the drop in oil and commodity revenues, which constrained available fiscal resources to meet the budget plans. 8 This includes Departmental Directorates, AGRI-CONGO, the CDTA, and the CVTE. 9 This includes the CNSA, the CRAL, and other NGOs Page 15 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Beneficiary producers appreciated the assistance, as evidenced by subsequent surveys by service providers. Survey findings indicate that the percentage of small producers satisfied with managerial services increased, from 12 percent on average in 2008 to 55 percent in 2017. This represents an implementation rate of 92 percent10. 39. In sum, the project strengthened the capacity of the MALF in better targeting the rural poor through improved design and implementation of public policies tailored to their needs. This involved the allocation of technical and financial resources to the Departmental Directorates and agricultural research and extension entities. It also entailed technical capacity building for both central and decentralized MALF staff. In addition, the project improved coverage of agricultural services. The project also served to enhance coordination and collaboration between the various government departments (at the central and decentralized levels), as well as with project enterprises. 40. Outcome/Objective 2: Improvement of beneficiary production and living conditions in rural areas. The ADRRP's interventions to achieve this objective have focused on three main intervention areas: the development of rural roads; the development of market infrastructure; and the dissemination of improved agricultural technologies to rural populations, including agricultural producers. 41. Rural road infrastructure. In order to improve rural road infrastructure, the ADRRP supported the rehabilitation and building of 1,319 kilometers (km) of rural roads, representing an implementation rate of 96 percent of planned roads. The project also supported maintenance of 486 km of these roads (89 percent of the planned work), as well as the acquisition of 6 nautical vessels. This has made it possible for producers and people in the areas benefiting from the production sites to be connected to markets and other social services11. It helped to reduce the geographical isolation of 248 villages with a combined population of nearly 340,000 people. Rural roads provided access to market infrastructure (MI) for about 124,600 producers in 2017, or 140 percent of the forecasted number. 42. The road segments rehabilitated by the project are spread across almost all of the participating regions of the ADRRP12 (See Annex 7 map on road construction). As such, they fulfill the important social demand of such investments in terms of linking high potential production areas with markets, consumption centers, as well as facilitating access to inputs for production. The network was developed and scaled along the main network. 43. Farmers in the area are benefiting from new roads, with a 25 percent reduction in human transportation costs and a 15 percent reduction in the cost of freight transport. There was an additional reduction in post-harvest losses because of the ability to move agriculture products more quickly to market. The new roads also ensured easier access to social services. Moreover, the average number of days to find a vehicle to sell agricultural products decreased from 6.5 days in 2010 to 2.5 days in 2013.13 44. These achievements have served to underpin the development of local commerce, creating numerous economic opportunities and IGAs for thousands of producers and non-producers alike in the project area. They have already afforded access to market infrastructure for some 124,600 producers in 2017, or nearly 140 percent of the target number. In addition, adoption of the Highly Labor Intensive (HLI, as opposed to capital-intensive) strategy for the rural road maintenance has further created employment and increased incomes for the poor. Finally, engagement 10 Project Completion Report (PCU), 2017. 11 Sources of safe drinking water, latrines, and so on. 12 With the exception of the Likouala region. 13 Project Completion Report (PCU), 2017. Page 16 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) with the private sector14 has created about 1459 direct jobs, including 20 percent for women. 45. Market infrastructure (MI). 41 MI projects were built or rehabilitated between 2010 and 2017, including 9 built and rehabilitated in 2010-11, and 27 built/rehabilitated in 2012-2013. Only 5 markets were built in 2015-2016. These MIs were spread over many departments, as was the case for rural roads. The number of MIs exceeded forecasts (146 percent of the forecast number) through collaboration between PCUs, the MEER, and Departmental Councils (DCs). Per 2017 report, the construction of MIs and rural roads afforded 124,600 producers with access to markets15. 46. However, implementation of MI was not preceded by feasibility studies which imply that the location of MI has not been responsive to local demand, although the needs are enormous. Indeed, several constructed and rehabilitated markets were not well utilized by agricultural producers and livestock farmers initially. For example, an intermediate market survey conducted in 201316 on the sample of 32 markets indicates that in all the project areas, only 5.2 percent of surveyed agricultural workers are aware of a market constructed/rehabilitated by the ADRRP. This proportion is comparable for livestock farmers (6.6 percent). The 2013 survey also reported that only 4.0 percent of respondents with access to a market constructed/rehabilitated by the ADRRP exhibit their products in these markets. Although there was no follow up survey on the same markets, market use rate has gradually appreciated as evidenced by the project intermediate indicator on beneficiaries’ use of market infrastructure and completion report of 201717. 47. The inaccessibility to markets in terms of distance constituted a major obstacle for certain producers. The average distance between the producer and the rehabilitated/built-up market is 13.7 km on average. However, it can reach up to 50 km18. The nearest market rehabilitated or built is still very far from the locality (21.6 km on average)19. 48. The management committees of the rehabilitated markets are not operational mainly due to institutional difficulties with municipalities, DCs or village committees, even though these committees work on a voluntary basis, as well as a lack of training. Most of the MIs do not have nearby storage facilities. Therefore, the beneficiary communities will be required to make contributions of 15 percent of the total cost of the construction or rehabilitation of collection and storage centers and market infrastructure. 49. Agricultural Production. Project activities to improve productivity of rural producers have included training, donating plant material and equipment, and disseminating improved agricultural technologies to rural producers. This was mainly achieved through: the rehabilitation and support of two seed research and multiplication centers20; the organization of field demonstrations; the rehabilitation of two production stations; and the introduction of improved methods for fish farming. The project also supported the rehabilitation and expansion of the breeding stock, as well as the provision of various equipment for the Sheep Technical Support Center. Almost 68 percent of 14 These include local enterprises, suppliers of goods and services, private consultants, trainers, as well as non-governmental organizations (NGOs). 15 Government’s completion report and impact studies (2017) 16 This market survey was conducted as the parent project was closing in 2013. 17 Impact Studies (2017) 18 For example, in the La Cuvette region. 19 The longest distance to reach the market closest to the locality is observed in the basin (51.3 km), and the shortest is 7.2 km (in Bouenza). 20 These include the Loudima Agricultural Research Center (CRAL) and the National Center for Improved Agricultural Inputs (CNSA). Page 17 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) the beneficiary producers adopted the production and processing techniques developed by the project (equivalent to an achievement rate of 113 percent). 50. Support to the centers for seeding and seed production have facilitated the growth and free distribution to 403 agri- multipliers, and have increased the supply of improved maize seeds, groundnut seeds, high-quality banana plants and cassava cuttings. With the training and technical advice provided by agricultural extension agents, these agri- multipliers have increased their production capacity and enabled the supply to least 4,000 seed producers. 51. Financial support has also been provided to producer associations for the financing of various micro-projects in the main sub-sectors. In total, approximately 12,047 producer households (representing an achievement rate of 107 per cent) benefited from project activities. In addition, 952 micro-projects were financed (see table 1). These micro- projects related to crop production (49.5 percent), livestock production (20.2 percent), horticulture (17.5 percent), fish farming (6.6 percent), processing (3.8 percent), poultry farming (1.6 percent) and fishing (0.8 percent). Table 1: ADRRP Micro-Project Financing (US$ Millions) Sub-sector Number of ADRRP Promoters’ Total Cost micro- contribution contribution projects Agriculture 656 2,054.6 125.2 2,179.8 Livestock 225 320.6 16.0 336.7 production Fish 71 2,618.2 12.1 255.1 farming/Aqua- culture 21 TOTAL 952 2,618.2 153.4 2,771.6 Source: PCU, 2017. 52. The ADRRP contributed to the improvement of productivity of the main crops in the project area. The cassava yields increased from 5.6 tons per hectare (t/ha) in 2008 to 49.1 t/ha in 2016. Maize yields also increased from 0.16 t/ha to 1.3 t/ha between 2008 and 2016. In addition, the banana yields increased from 0.4 t/ha to 14.03 t/ha for the same period. The productivity of these crops increased by 8 to 35 times, respectively, for the same period. Production of major crops (cassavas, maize, groundnuts and bananas) increased from nearly 4,420 tons in 2010 to 6,240 tons by the end of 2013, and then again to more than 50,000 tons in 201622. As such, the results far exceeded the originally projected 20 percent increase in production. In addition, the average catch of a fisherman in the project area increased from 14.6 tons/year in 2010 to 26.4 tons/year in 2013. Likewise, the use of new techniques and improved inputs has also increased. The average area cultivated per household increased from 0.25 ha before the ADRRP to 5 ha per beneficiary household at the end of the project. 53. The project's achievements in terms of infrastructure and producer support facilitated not only increased 21 The total is the sum of 814 micro-projects of the parent project and 138 micro-projects implemented in the AF. 22 Impact Studies of the project, 2017 (Rapport d’achèvement et d’impact du PDARP, UCP, MAEP, 2017.) Page 18 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) agricultural production, but also improved mobility of population and producers of the micro projects. This helped to reduce post-harvest losses and allow for the sale of products at higher prices. Similarly, processing and other income-generating activities have developed in the project area, facilitating income diversification and improvements for the targeted rural populations. 54. The ADRRP benefited about 366,131 direct beneficiaries including 340,030 inhabitants in 248 remote villages, comprising 12,047 producers (60 percent youth and 51 percent women), and approximately 10,500 trainees23. These beneficiaries rehabilitated the marketing infrastructure, such as rural roads and markets. The project achieved its development goals of strengthening the capacity of MALF staff to develop and implement policies to address poverty and improving the ability of the rural poor to raise their income. Justification of Overall Efficacy Rating 55. The achievement of the PDOs is, overall, substantial. This is reflected in the main performance indicators, such as the number of direct beneficiaries (366,131), representing an implementation rate of nearly 111 percent, and the rate of increase in the income of households benefiting from the project, which far exceeds expectations. Indeed, with the implementation of the project, the average income of beneficiary households more than doubled, from Central African CFA Franc (XAF) 174,946 (US$ 313.15 equivalent) per year to XAF 391,152 (US$ 700.16 equivalent) per year between 2010 and 2013; and from XAF 500,000 (US$ 895 equivalent) to XAF 1 million (US$ 1790 equivalent) per producer per year between 2015 and 2016, depending upon the group24. In 2017,25 income increased by more than 50 percent for nearly 90 percent of the beneficiary producers, and 75 percent producers saw their incomes increase by more than 200 percent. This improvement in incomes is reflected in similar improvements in purchasing power, living conditions and social well-being. Indeed, 71 percent of beneficiaries say that the project has helped strengthen social cohesion in their communities. Seventy-five percent of producers interviewed during the impact survey26 believe that the ADRRP has helped them get out of poverty by enabling them to acquire material goods, provide schooling for their children, and diversify their productive activities. C. EFFICIENCY Assessment of Efficiency and Rating 56. Introduction. The analysis of the efficiency of the ADRRP focused on increasing the production of crop and animal products in the project areas. On the basis of surveys and studies carried out, as well as interviews with the PCU 23 From the MALF, Departmental Councils, Agricongo, the DAO, NGOs, consultants and private contractors. 24 The Project Impact Assessment was conducted in 2013 by the National Institute for Statistics (INS). 25 Project Completion and Impact Mission, 2017. 26 Project Completion and Impact Mission, March 2017 (151 producers were surveyed out of the 471 members of the 28 project- supported associations). Page 19 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) staff and producers, the mission identified seven micro-project models relating to fish farming, horticulture production, poultry, and banana, cassava, and maize production and processing. The expected increases in productivity have been generally achieved as a result of the ADRRP. The increases in production costs relate to expenses incurred by the ADRRP were due to the additional use of inputs. 57. The efficiency of the ADRRP is based on a several assumptions, of which the principal ones are as follows: (i) the benefits considered are those from the seven micro-project models; however, the benefits from the rural roads and market infrastructure were not included as the mission did not have objective elements upon which to calculate these benefits. The aggregation of profits from the models were made by comparing the financial and economic results to all micro-projects financed by the project; (ii) the economic costs of the ADRRP were derived from the financial costs after eliminating the value-added tax (VAT). A standard conversion coefficient (CC) of 0.90 for the prices of products traded was derived from profits of the models selected based on the CCs used in projects recently developed in the Republic of Congo; and (iii) the opportunity cost of capital (OCC) selected is 12 percent, reflecting the geographic isolation of the project areas and their low level of integration in the markets. A simulation using an OCC of 9 percent was also tested. 58. Increases in productivity and margins are much greater than would have been the case without the project. All the micro-project models developed by the ADRRP are cost-effective. Farmers' profit margins are nevertheless handicapped by the low sales prices of their products because the many intermediaries siphon off some of the added value. In addition, opportunities to generate profits are limited by persistent difficulties in accessing markets, information, improved inputs and credit. The level of transaction costs remains high. 59. Financial indicators for the seven micro-project models show good financial performance in terms of the financial rate of return (FRR) and net present value (NPV). The ADRRP's economic rate of return (ERR) is 12.96 percent, which is above the opportunity cost of capital (12 percent), and the NPV is positive (US$ 25.6 million). The same applies to the financial rate of return (12.4 percent). These indicators are satisfactory, given that some of the additional benefits have not been considered. For example, this is the case of the benefits of road rehabilitation and market infrastructure, and the benefits of reducing transaction costs due to facilitated access to technology, information, training, and technical support. These benefits resulted in the improvement of the living and food security conditions of the project's target group. 60. Estimates of the different cost-effectiveness indicators are lower than the project design estimates (with an ERR of 36 percent and a FRR of 26 percent). A reduction in the economic life of the project from 20 years to 10 years (design estimates) further reduces the ERR and NPV to values that make the project unprofitable and inefficient. 61. The ADRRP is not very sensitive to changes in aggregate costs and benefits. Its level of profitability remains acceptable despite simultaneous changes in costs and benefits (see annex 4). The Monte Carlo simulation analysis shows that there is a very high probability of cost-effectiveness for the ADRRP (90 percent), which is very satisfactory. It is the processing and poultry farming micro-projects that provide more value added to the beneficiaries of the project. 62. In terms of administrative efficiency, the rehabilitation and maintenance of rural roads helped reduce the cost of transporting people by 25 percent and freight by 15 percent. It also reduced post-harvest losses and long-term maintenance costs for infrastructure. The costs of road rehabilitation are 112 percent to 118 percent of the planned costs, and they are relatively smaller for market infrastructure. The amounts spent are significantly lower than those usually incurred by the MEER, and vary, depending on the terrain/vegetation, the way in which the work is carried Page 20 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) out,27 and the magnitude of the damage to the rural roads. 63. However, the cost of rural roads and market infrastructure was higher than estimated during project design. The parent project took 15 months to become effective. The AF project was expected to start in May 2014 and be completed in October 2015, that is, in 18 months. However, this phase lasted three years (until April 2017). Project coordination, M&E, staffing and some aspects of implementation have not been as effective as possible, thereby resulting in delays and cost overruns. This has resulted in a significant increase in project management cost (160 percent of the estimate and 25 percent of total costs). The delays in the implementation of the ADRRP had an impact on the project’s overall efficiency, especially when the significant resources (financial, human, and logistical) mobilized for the project are considered. These resources could have been used more efficiently for other productive activities. The delays are attributable to the inadequate capacity of some partners; poor access to certain project areas; and a lack of proficiency in procurement procedures or the slow pace of such procedures when implemented. In addition, the satisfactory performance rates need to be qualified in terms of quality measures. Administrative efficiency is modest. The efficiency of the ADRRP can be rated as overall modest. D. JUSTIFICATION OF OVERALL OUTCOME RATING 64. The objectives of the ADRRP are aligned with the policies and strategies of the Government and the World Bank, as well as the needs of the rural poor. Although the project scope exceeded the management capacity of the PCU during the first four years, during the AF the trend was toward greater involvement of project partners, including the PCU. This increased the geographical concentration of activities. The relevance of the ADRRP is therefore rated substantial. 65. The project has achieved its objectives despite a difficult start. The ADRRP helped increase agricultural productivity. The construction/maintenance of roads and market infrastructure has produced decisive results. The efficacy in achieving the PDOs of the ADRRP can be attributed in large measure to the proactive role of the PCU, the Task Team Leader (TTL) and the project partnership strategy. The project’s efficacy is rated substantial. 66. The ADRRP’s efficiency indicators can be rated as modest. The FRR, ERR, and NPV are significantly lower than at appraisal. Nevertheless, the project’s implementation period was increased from five to nine years, and effectively began only in 2010. It did not achieve real momentum until 2012. Similarly, some activities experienced delays of up to one year. The delays in the implementation of the ADRRP had an impact on the project’s overall efficiency. The project’s efficiency is therefore rated as modest. In sum, the overall outcome rating for the ADRRP is moderately satisfactory. 27 Road maintenance varies between XAF 2.2 million (US$ 3.938 equivalent)/km for highly equipment-intensive works (HIEQ), and XAF 0.5 to 1 million (US$ 895 ~ US$ 3,938 equivalent)/km for highly labor-intensive (HLI) works. Page 21 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) E. OTHER OUTCOMES AND IMPACTS (IF ANY) Gender 67. The project has helped to strengthen the capacity of national stakeholders, both governmental and non- governmental, as well as producer associations. Indeed, the training programs helped the associations to develop their management capacity, enhance their autonomy, and increase member financial capacity. These achievements should, in turn, lead to a transfer of skills to producers at the grassroots level. 68. The project has also helped women to attain greater financial autonomy. Although women are dominant in agricultural activities, they are generally less well off than men. However, the ADRRP has helped women28, who represent 51.8 percent of the 12,047 producers, to benefit from agricultural services. The increased incomes of rural women and young people (accounting for 60 percent) have been made possible through their equal access to advisory services, quality inputs, farm equipment and three-wheel delivery vehicles 29. Because of the additional income they earned, these women have gained more financial autonomy. Furthermore, women have contributed to decision-making within the household on the same basis as men, if not more30. In addition, the supply of wages and subcontracting is estimated at more than XAF 12 billion (US$ 21.4 million equivalent), paid by about thirty construction companies and NGOs. The construction companies and NGOs were mandated to pay more than 1,300 local laborers and suppliers of building materials, of which women account for more than 20 percent31. 69. Women benefited from the project in terms of a reduction in their workload. In addition to the facilitation of input and production transport, the provision of three-wheel delivery vehicles made it possible to reduce the workload of women, in particular with regard to the supply of firewood or transport of water cans from the source to households. This has allowed them to assume more responsibility in mixed groups,32 for example, by becoming presidents, treasurers, secretaries, and so on. The ADRRP has strengthened the involvement of women in the management of groups through training, information, awareness-raising and technical assistance. To this end, in 2016-2017 in collaboration with Agricongo, the project has trained 214 producers, 80 percent of whom are women. Specifically, the women received training in the rational protection of crops and the management of pesticides. Other Unintended Outcomes and Impacts 70. The ADRRP also mobilized young people because of the financial and material success of the first micro-project promoters who benefited from the project's support to relaunch/develop their farms. 71. The ADRRP attracted young people to agriculture by providing material support and promoting processing through improved methods and micro-projects. These measures have helped to simultaneously reduce the drudgery of work 28 This is against a target of 51 percent of women. 29 Other projects co-financed by the International Fund for Agricultural Development (IFAD) and the Food and Agriculture Organization (FAO) have also provided three-wheel delivery vehicles. However, the ADRRP’s contribution is consistent. 30 In the case of unemployed husbands or day laborers. 31 ADRRP Completion Report, World Bank, 2017. 32 This is the case of the Saint-Sérieux group in the district of Mabombo (department of Bouenza), which comprises 27 members, including 18 women and 9 men, aged 23 to 55; and the Sounguika marshy grouping, with 20 members, 55 percent of whom are women, constituting 19 households, aged 20 to 60 years. Page 22 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) and increase the financial viability of agricultural activities. Thus, by mobilizing young people, who account for as much as 60 percent of the direct beneficiaries, the ADRRP has the potential to reduce youth unemployment. 72. Beginning in 2015, the scope of the project also included the regular conducting of awareness-raising campaigns focusing on sexually-transmitted diseases (STDs), such as Human Immunodeficiency Virus (HIV)/ Acquired Immune Deficiency Syndrome (AIDS), with the assistance of community leaders33. 73. The project promoted active participation of private sector partners and local enterprises in both infrastructure activities and agricultural production activities. III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 74. Needs assessment and identification of activities to be implemented. The project design was based on the clearly identified needs of the target population and the Government of the Republic of Congo. In preparing the project, a consultative approach was taken with the local populations, and in cooperation with the deconcentrated state services. This made it possible to identify the real needs of beneficiaries, with the aim of ensuring that planned responses were appropriate. From the outset of the process, this made for a manageable approach to project implementation, especially at the local level. 75. Nevertheless, certain shortcomings affected project identification and design, which in turn have had a negative impact on the project. These include, among others, the failure to consider important needs of the producers (such as land tenure security, post-harvest transport, farm accounting management systems, and so on). These needs were identified only at the monitoring and evaluation stage. 76. Simple design. The definition of the components and activities was in line with challenges to be met. The aim was to ensure that the implementation of such components/activities would contribute to the achievement of the PDO, which is one of the key factors contributing to the success of the ADRRP. Emphasis must be placed on the all-encompassing approach of the project design, without which the achievement of the expected results would not have been possible. This was done mainly through the planning of activities all along the different agricultural sub-sectors both vertically, from the provision of inputs to marketing, and horizontally, with capacity-building activities targeted to stakeholders and support institutions. 77. Realistic objectives and outcomes. The project’s objectives are realistic. The project’s results chain is also quite clear and realistic in light of its activities and objectives. This justifies the fact that, apart from some adjustments to the performance levels for results indicators, the project’s results chain has hardly changed 33 These campaigns, which followed the “Communication to Change Behaviors� approach, reached around 10,000 individuals, most of them, young people. Page 23 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) from its initial to additional stages. The level of achievement of expected results and objectives at project’s end (90 percent to more than 100 percent) attests to the realistic nature of the project design. 78. Appropriate choice of stakeholders. From the project design phase, the decision to entrust the implementation of the project to a central Project Coordination Unit (PCU), which would recruit its staff on a competitive basis (for one-year renewable contracts, based on satisfactory performance), facilitated project implementation. In fact, it made it possible to make changes in the staffing of the PCU which was deficient at the beginning of the project. As a result, with increased experience, the PCU became more dynamic and effective. Similarly, the involvement of the ADRRP’s partners34 helped considerably in identifying beneficiaries and in ensuring the smooth progression of the project. Thus, the ADRRP created a favorable dynamic for revitalizing the agricultural sector in regions covered by the project. B. KEY FACTORS DURING IMPLEMENTATION 79. The approach to project implementation was based on the involvement of all stakeholders, and the capitalization of the combined achievements of national actors (producers, state bodies, the private sector, civil society, NGOs, and so on). This was the case with the training provided for staff of the Departmental Councils, who participated in monitoring and maintenance of the infrastructure works (rural roads). Strengthening the technical and material capacity of agricultural extension services, together with their participation in project implementation, was also one of the main factors contributing to the project’s success in promoting the adoption of improved farming methods by farmers (68 percent), as well as in increasing crop yields. 80. The participatory approach was also reflected in the process for selecting micro-projects, conducted under the auspices of the Support and Advisory Services Providers (OAC) (2009-2012) and the sector heads (2012- 2013), with the support of the departmental directorates of the MALF and the department prefect or their representative. Having recourse to the OAC is commendable since these agencies are for the most part operational bodies comprised of retired specialists (agronomists, animal husbandry professionals, and so on) and young people trained in various aspects of agriculture and livestock production. This has also been the case with sector heads. Thus, this approach made it possible to improve on-the-spot supervision of producers and monitoring of micro-projects. 81. Nevertheless, the project was adversely impacted by the dispersal of responsibilities between different agencies, as well as administrative and bureaucratic obstacles created by the opposition between local and central authorities. Despite training and awareness-raising activities undertaken late in the project’s existence, implementation of the project slowed. The coordination function concerned with the project as a 34 Partners include deconcentrated state services, communes, producers, producer associations, NGOs, and so on. Page 24 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) whole was adversely affected, in particular during the initial phase of the ADRRP. It also took longer than anticipated to replace the first project coordinator. 82. Shortcomings in the selection and organization of micro-projects. The micro-project selection process encountered some problems. These include insufficient field visits to investigate proposals, and the lack of reference data on project costs and comparative benchmarking, due mainly to the lack of baseline data and local cost documents. This has led to the disparity of costs among similar micro-projects, without any real justification. 628 out of 952 total micro-projects have encountered such cost justification problems. 83. The quality of the work carried out by the project’s implementing entities (the PCU, the Steering Committee, the Government, and so on) has been critical to the smooth implementation of the project and the attainment of its objectives. In fact, the project’s initial phase was marked by the poor performance of the team in place at that time (2008-2010). This was due to constraints in providing the PCU with an adequate number of appropriately-skilled staff during the first three years of the program. The weaknesses of the PCU were noted during this phase, particularly in terms of operational management (procurement, administrative and financial management), organization, planning, coordination, and vision/leadership. In addition, officials responsible for certain technical components were sometimes unavailable. The weak operational management capacity of the PCU justified the use of public and private service providers in handling the institutional and implementation arrangements for project execution. 84. The project’s second phase was marked by an improved PCU (recruitment of a new coordinator, strengthening of the team through the addition of new skills, appropriate training and tools, and so on), which contributed to the turn around and success of the project. The changes were in accordance with the management documents and the preparation of the annual work plan and budget (AWPB). The semi-annual and annual reports on project implementation were prepared by the PCU, taking advantage of the quarterly reports from the Departmental Monitoring Office (ADS). These monitoring reports made it possible to speed the implementation of project activities, and contributed to the achievement of results. 85. Throughout implementation, the Government was fully committed to the project. The Government has honored all necessary commitments prior to the entry into force of the Financing Agreement by maintaining the oversight of the ADRRP by the MALF. It has also paid its share of counterpart funding (especially since 2011), albeit with significant delays in the initial phase, 35and implemented the recommendations of the various World Bank monitoring missions. 86. Government agencies have been involved in the implementation of the project as a stakeholder under the agreement protocols. The agreement provides for a budget for monitoring, oversight, maintenance, and so on. However, issues such as the weak capacity of state services to ensure support for the project and guarantee the ongoing supervision of project beneficiaries, as well as significant delays in the payment of national counterpart funding, were observed. The Government has not provided the deconcentrated services with all of the logistical and financial means required to achieve project objectives. This has resulted in a delay of up to one year in implementing the project’s operational activities, difficulties in maintaining rolling stock and water vessels, and the weak involvement of the service providers. However, when the timely 35 ADRRP has experienced slow release of counterparts’ funds due to the Congo’s lengthy process of the budget cycle. Page 25 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) funding, staff, equipment, and vehicles have been placed at their disposal, the contributions of these service providers have been relatively satisfactory. 87. Spatial coverage of project interventions. The ADRRP covered a broad expanse of territory, which the management teams found difficult to manage. The geographical distances between project sites where project activities were to be carried out posed serious problems for monitoring and evaluation purposes. Starting with the additional financing, the ADRRP took an important step toward a more relevant spatial targeting. It did so by focusing on the production areas (agricultural growth poles), and enhancing cooperation with state deconcentrated services and other partners in the field. In addition, the project’s operational means were often underestimated when considering the diversity of the operations and the areas covered by the project. However, the resources provided for investments were often overestimated in relation to the absorption capacity of the PCU and partner agencies. IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 88. The results framework for the ADRRP was designed in conjunction with the operational theory of change. Objectives and indicators were clearly defined at the time of design, which justified maintaining the objectives and indicators during the AF phase. The outcomes of the PDO statement are tracked by indicators. As shown in the results framework for both project phases, the indicators are specific, measurable, relevant, time-bound, and achievable. The target values of indicators are realistic. The final results show that most of the targets have been achieved except for two (of a total of over twenty). These included the targets that “at least 50 percent of the rehabilitated rural roads are regularly maintained at an acceptable standard�, and “at least 50 percent of the annual budget of the MALF are transferred to Departmental Directorates�. 89. At the end of the project, most of the target values of the indicators were met, or even exceeded. Regarding the results framework for the project’s initial phase, more than 80 percent of the indicators were reached at more than 50 percent; many even exceeded the 100 percent achievement rate. The revised indicators have maintained or upgraded the target value of indicators to reflect the increase in funding and additional activities of the ADRRP, with the exception of one indicator36. However, the results framework does not include indicators to assess the quality of design of rural roads and market infrastructure. 36 The indictor is: “Rehabilitated rural roads are maintained regularly�. Page 26 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) M&E Implementation 90. The budget allocated for M&E was quite limited because it was not fully considered to be a strategic activity that could minimize the operational costs of the PCU. The M&E manager recruited in 2013 had limited M&E experience, but since that time has benefited from supervision and training. Collaboration between the M&E manager and the other PCU management staff was relatively limited at the project start, due mainly to reduced staff in the M&E unit. In addition, the baseline was only completed in 2010. M&E suffers from weak coordination with departmental and central MALF services, as well as weak project team staffing. 91. It was only after the mid-term review that M&E became operational and was done on an ongoing basis, resulting in the improved use of M&E data to manage the project. The results framework of the project was updated after the AF. The M&E compiled the information from the ADS and the CSA. These entities had benefited from capacity- building efforts. They had also been consulted regarding the implementation of M&E; the expected needs for capacity building and information; the allocation of resources; and the management and dissemination of data. 92. However, there were several constraints related to M&E, such as the availability of certain actors, a lack of data, delays in sending data, and a poor understanding of certain M&E forms. The MALF has inadequate capacity to perform M&E, and does not centralize, disseminate and archive agricultural sector data at the Ministry level. 93. The field monitoring missions of the ADS and CSAs were limited, and the reports produced were more descriptive than evaluative. Also, they were not systematically shared among the field offices as part of promoting best practices. The size of the project area (over 20 villages per sector) and the age of the CSAs (an average of 47 years) made it difficult to effectively monitor the producers. Producer associations do not forward reports to the CSAs and Heads of the Departmental Monitoring Office (CADS). 94. The lack of a current data base on beneficiaries of the ADRRP (micro-project beneficiaries, seed multipliers) has hindered an analysis of the living conditions of project beneficiaries. This in turn has made it difficult to fully capture the results of the project based on documented indicators and analyses. Previously identified indicators could have been established in the Terms of Reference (TOR) for the providers of data collection. Moreover, the M&E specialist has no counterpart in the World Bank Office, unlike other units such as the environment or finance. However, the increase in the number of World Bank supervision missions (3 times per year), the ongoing relationship between the TTL and the PCU, and training made it possible for the M&E to perform at least some of M&E activities. In addition, M&E suffers from the absence of an online database, as well as a functional internet network to improve the speed of reporting. M&E Utilization 95. The M&E results were communicated to the stakeholders using informational forms, a bimonthly journal, televised presentations, local radio, and 20 bimonthly ADRRP informational letters. Meetings held by the MALF and the steering committees, the mid-term review, and the supervision missions facilitated communications about project performance. The AF was mainly the product of information collected through M&E during the initial phase of the project. Page 27 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Justification of Overall Rating of Quality of M&E 96. The M&E as designed, implemented and used made it possible to monitor the performance of the project. However, it did not allow a thorough evaluation of the achievement of the declared objectives, nor could the links in the results chain be tested. The use and impact of M&E were limited because of the small number of staff assigned to this activity at the PCU and the deconcentrated units of the State, as well as limited logistical and financial resources. At the same time, using the central and deconcentrated units of the MALF to monitor field activities may lead to M&E functions and processes being maintained after the project ends. The quality of M&E is rated modest. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE 97. The project was classified as ‘Category B’. Infrastructure activities as well as microprojects were expected to have potential risks of site-specific adverse environmental and social impacts. Therefore, the following safeguards were triggered: OP 4.01 Environmental Assessment; OP 4.09 Pest Management; and OP 4.12 Involuntary Resettlement. OP 4.10 Indigenous Peoples was retroactively triggered at the Mid-Term Review (MTR). OP4.04 Natural Habitats was not triggered in the original project. However, the issue was considered in the Environment and Social Management Framework (ESMF). The policy was later triggered in the Additional Financing. In compliance with Bank requirements, the Borrower prepared the following documents: an ESMF; an Agri-Environmental Guidance Guide in lieu of the Pest Management Plan (PMP); an Indigenous People Planning Framework (IPPF); and a Resettlement Policy Framework (RPF). 98. In accordance with the environmental safeguard policies initiated, several environmental safeguard instruments were developed for implementation. These included an ESMF, and about 70 environmental and social impact reports on rural roads, market infrastructure and other activities, Environmental safeguard instruments were also developed for the 114 preliminary screening trenches carried out on the runways to be maintained, as well as for other micro-projects. 99. The project has strengthened the technical and organizational capacity of civil servants, companies, oversight bodies and some producers in dealing with environmental and social management issues, as well as Resettlement Action Plans (RAPs). Thus, the project facilitated the development and satisfactory implementation of environmental and social management plans. Environmental and social studies preceded the activities conducted under the ADRRP, beginning in 201437. By resorting to this practice on a systematic basis, it has been possible to better anticipate environmental and social challenges — and to dispense with environmental and social impact studies when unnecessary. During the additional financing 37 Recommendation from the July 2014 Supervision Mission, which recommended that the ADRRP “systematically adhere to the screening form.� Page 28 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) stage, the project implemented environmental and social studies.38 Although the project respected the requirements with respect to environmental safeguards, the social dimensions were not always identified in sufficient detail, particularly with regard to the social impact of activities relating to market infrastructure, roads and micro-projects. 100. Economic and social advantages were provided to indigenous producer associations through outright grants for 31 micro-projects involving cassavas, bananas and yam growers and sheep-raising for approximately US$145,000. The ADRRP also supported indigenous populations by helping to improve access to remote communities,39 and by providing equipment for a NGO engaged in promoting the rights of indigenous peoples. 101. A total of 31 microprojects (29 micro-projects during initial project implementation and 2 during the additional funding phase) involving 620 indigenous people, 50 percent of whom were women, received technical and financial support from the project in six departments. These micro-projects for indigenous peoples, representing 3 percent of the 952 micro-projects financed by the project, enabled beneficiaries to produce better and more crops than before the project. As such, they generated income to improve their living conditions and to educate their children. 102. Fiduciary compliance: The financial management of the project improved over time. However, the non- optimal use of computerized financial and accounting tools made it difficult to manage the project. Moreover, the lack of knowledge about management and IDA rules led to the ineligibility of US $ 260,000, which had to be reimbursed by the Government. The overall financial performance of the project was considered to be satisfactory for the following reasons: (a) The general project accounting was kept up to date through the software; (b) a manual of administrative and accounting procedures was updated and used throughout the project implementation period; (c) the Project Financial Monitoring Reports were prepared on a quarterly basis and sent to the World Bank; and (d) the audit of the financial statements for the last few years was certified and transmitted in good time to the World Bank before the June 30 th deadline of each year. 103. Procurement: Procurement has been strengthened following the recommendations of the World Bank supervision missions and the completion report of the parent project. In addition to systematizing the distribution of markets in newspapers, procurement plans were regularly updated and submitted for IDA validation; and project activities were carried out in accordance with World Bank procurement procedures. 104. Disbursement: The overall disbursement rate of the project was high (105.7 percent) due in particular to the re-framing of the PCU by the Government and the World Bank, the provision of training, and the additional funding project. During Phase 2 of the project, the Government contributed additional funds of XAF 600 million, equivalent to US$ 1 million. 38 This included 6 Environmental and Social Impact Notices, together with Environmental and Social Management Plans relating to the rehabilitation of rural roads and the construction of market infrastructure, as well as 112 screenings. Model management plans have been developed for the micro-projects and road maintenance in cooperation with an environmental expert from the World Bank. 39 This included the Panda Boudouhou Road (15 km). Page 29 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) C. BANK PERFORMANCE Quality at Entry 105. Design: The project design brings to bear an integrated and holistic intervention to address the multi-dimensional issues in the rural environment. The component for rehabilitating rural roads and providing market access is critical for supporting sustainable agriculture and conducive to IGAs. Given the low level of capacity within national and local level institutions in the country, the design pursued the potential for active partnerships with donor community and/or within the NGO/private sector to support some of the main activities while allowing training and capacity building activities for both central and decentralized public entities. 106. Vague PDO and mismatch of language between PAD and FA: The PDO statement was not clearly articulated. It contained a description of components and activities that contributed to the objectives of the project. The ambiguity was later addressed through implementation of project activities and revision of PDO indicators. The inconsistency between the PDO statement of the PAD and the FA was reconciled at the first project level 1 restructuring. 107. Environmental aspects: Given the already identified regions and the proposed investments for roads and micro- projects in areas with Indigenous People (IP) populations, both the IP and Natural Habitat policies should have been triggered. This was rectified at the first project level 1 restructuring. 108. M&E: The M&E arrangements did not reflect the project’s expansive geographic coverage and the funding estimate was too low. Page 30 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Quality of Supervision 109. Proactive and close supervision of the project: The project implementation suffered long delays and poor progress in its initial phase. This was mainly due to weak leadership by both the Bank Supervision Team and the PCU. From the end of 2009 when a new TTL was posted to the region until project closure, there was greater proactivity and continuity in task management. Project leadership at the coordination unit level was also changed in 2010 following which implementation accelerated. These were instrumental in turning project performance around. The team carried out regular supervision missions and conducted dedicated meetings with PCU staff and other stakeholders to advise on project implementation issues and to ensure fiscal and safeguard compliance. 110. MTR: The preparation of the MTR was well done and allowed for key corrections, especially with regard to the safeguard policy on indigenous peoples. Retroactive application of this safeguard policy necessitated a level 1 restructuring of the project. 111. Additional Financing: Having achieved tangible results on the ground by the end of the original closing date, the government requested for additional financing to scale up the project activities. The additional financing helped to: (a) expand on the successful implementation of project activities (micro projects and infrastructure component); (b) better target interventions in defined geographic areas (called growth poles); and (c) revise performance indicators upwards. Justification of Overall Rating of Bank Performance 112. Based on the above argumentation, and given that the outcome is rated moderately satisfactory, the overall Bank Performance is rated as Moderately Satisfactory. D. RISK TO DEVELOPMENT OUTCOME 113. The modest scale and simplicity of the facilities financed by the project allows for easy maintenance. The agricultural measures promoted by the ADRRP appear to be well established at the local level because of the involvement of the producer associations and the service providers. Consultations with the associations promoted by the project contributed to their buy-in by the communities, an essential condition for sustainability. However, there are still a number of threats to the sustainability of these achievements, namely: (a) limited soil fertility management; (b) poor water management within the small market gardens; (c) low management capabilities among the associations, in particular with regard to replacing material and the deteriorated sections of the facilities; and (d) land tenure insecurity. 114. Institutional Sustainability. The project strengthens the capacities of the service providers, thereby contributing to the development of local expertise that can be called upon as new partnerships evolve. However, in a context of insufficient public and private investment in rural areas, sustainability is Page 31 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) endangered by the weak capacity of these providers who depend heavily on the business that comes to them from the project. Capacity building was less impressive with regard to state technical units, and true partnerships were not established. The post-project continuity of services is also endangered by limited operating resources. The partnership with the communities was insufficiently developed, although it could have contributed to the sustainability of the outcomes given the responsibilities recently granted to them. It is expected that some of these issues around continuity and resource constraints will be addressed in the new WB project, Commercial Agriculture Project (P159979). 115. A number of factors that favor sustainability are present in terms of the rehabilitation of rural roads40 and market infrastructure. In addition, good farming practices were adopted, such as using improved seeds. This helped to increase productivity and incomes. Also, the capacity of both public and private stakeholders was strengthened as a result of the project. However, these qualities are eclipsed by unfavorable factors such as the lack of an appropriate seeding strategy, dependence on community and public institutions for foreign assistance, and the operational deficiencies of the infrastructure management and maintenance committees. The sustainability of the achievements of the ADRRP is also tested because of failures accentuated by the isolation of many production areas and basins, as well as distance from population centers. Nevertheless, the future World Bank Commercial Agriculture Project (P159979) provides an opportunity to ensure the sustainability of the achievements of the ADRRP. The sustainability of the project is therefore rated moderately satisfactory overall. V. LESSONS AND RECOMMENDATIONS 116. The most important lessons that can be drawn from the project are reflected in recommendations for the various project stakeholders. These recommendations are intended to improve the design, implementation, performance and impact of future World Bank interventions in the Republic of Congo. 117. Involvement of stakeholders. Faced with the inadequacy of the means of intervention by MALF staff, private actors identified the problems encountered by producers, and transmitted information about these constraints to the PCU. The private sector has been engaged in project implementation through suppliers of inputs and services, trainers, processing, as well as through the rehabilitation and maintenance of infrastructure. The degree of private involvement, when carefully matched to the objectives and needs of the project, can bring stronger ownership and participation of stakeholders in project implementation. 40 The ADRRP developed low-cost, labor-intensive road maintenance techniques which are well suited to the local context. Page 32 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) 118. Consideration of the local conditions. Numerous delays with infrastructure work were due to climate risks, and the delivery of inputs during off-season disrupted the farming season. In addition, group producers had numerous shortcomings, including a lack of knowledge of procedures and insufficient capacities. Failure to account for the local conditions led to the inactivity of certain producers’ groups that had been formed solely to take advantage of the project, and who lacked any shared vision. Mapping of stakeholders and profiling of different local conditions during project preparation would help avoid delays due to insufficient capacities. 119. Value chain approach. The project essentially targeted the agricultural upstream value chain. The focus on downstream has been modest. However, opportunities in these areas are important. The link between actors along the entire value chain could be strengthened to ensure efficient matching of buyers and sellers especially in the outputs market. This will provide producers sufficient confidence to invest over the long term in productivity enhancing technologies that can improve yields, profitability and farm income. In this context, the new World Bank project on commercial agriculture is coming at the right time. 120. The following five recommendations are proposed: 121. Recommendation 1: Project design. The project design should focus on a limited number of achievable outcomes, outputs and components. M&E should be considered as a strategic activity. Similarly, support and supervision missions should facilitate prompt adjustments to the project. Synergy must be created among stakeholders to ensure the scaling up of project activities. 122. Recommendation 2: Length of implementation and geographic coverage of the project. Considering the limited absorption capacity of project partners, a longer implementation period should be considered with phased approach and geographically-concentrated activities. This would also allow for the phased expansion of the project area, should the need arise. 123. Recommendation 3: Value chain. The organizational capacities of producers in terms of a holistic approach to the value chain should be improved to facilitate the search for new markets and the negotiation of better prices. Environmental issues should be integrated into future projects to encourage sustainable farming practices. Similarly, there is a need to promote innovative activities to mobilize savings and credit. However, building such capacity of producer organizations is a long- term undertaking which requires long term engagement. The new WB project, Commercial Agriculture Project (P159979), is expected to sustain and build upon ADRRP’s early achievement, especially through its continued engagement with producer groups as well as small and medium agribusiness enterprises. 124. Recommendation 4: Project Sustainability can be ensured by empowering associations and local actors. Infrastructure management committees should be given the means to collect and manage funds. Producer organizations need to be professionalized to become sustainable entities. Continued support to the MALF will enable it to carry out the necessary reforms initiated by World Bank support. . Page 33 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: The primary objective of the project is to assist the Republic of Congo to increase the ability of the rural poor to raise their income through the generation and promotion of improved agricultural te Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Direct project beneficiaries Number 0.00 0.00 321600.00 366131.00 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 Female beneficiaries Percentage 0.00 0.00 51.60 51.30 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 Comments (achievements against targets): Indicator was added at additional financing. Data is from the final impact study Page 34 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Increased in income of Text Agr: 635,365 10 50% 200 project beneficiaries Horticulture: participating in micro- 1,992,366 projects by end of project Livestock: 374,987 Fisheries/Aqua: 1,466,844 Processors: 2,013,864 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 Comments (achievements against targets): Data is from the final impact study Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Increased productivity of key Text cassava: 5.6 ton/ha C:10.0 cassava: 10.6 ton/ha Cassava: 49,11 agricultural products banana: 40 ton/ha banana: 75 ton/ha (cassava, banana and maize) maize: 1.6 ton/ha B: 0.7 maize: 3.5 ton/ha Banana: 14,03 M: 0.3 Maize: 1,3 Page 35 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 Comments (achievements against targets): Data is from the final impact study A.2 Intermediate Results Indicators Component: Component 1: Building Capacity in the Ministry of Agriculture and Livestock Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Increase rate of annual Text 45 MAE: 80 MAE: 80% Ministry of Agri sectoral budget MPA: 64 MPA: 70% (2014): 52% execution/use of the MAE Ministry of Fisheries and MPA (2014): 36% 30-Apr-2008 31-Dec-2012 29-Jan-2014 22-Dec-2016 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion After PY2, the Ministry of Text 0 2 3 3 Agriculture and Livestock has Page 36 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) prepared and adopted result- 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 based sub-sector programs to guide the preparation and execution of the sectoral annual budget. Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion At least 60% of farmers are Text Agr.: 2.7% 60 Agric. 20% Agriculture : 42 satisfied with the services Livestock: 8.9% Livestock: 20% Livestock : 75 received from the extension workers and the Ministry of Fisheries: 24% Fisheries: 50% Fisheries : 50 Agricuture by the end of the Horticulture: 13% Horticult.: 50% Horiculture : 89 project (EOP) Transformation : 20 30-Apr-2010 31-Dec-2012 30-Jan-2014 22-Dec-2016 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Page 37 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) By EOP, at least 80% of the Text 0 80% 60 % procurement of goods and services is handled by the 30-Apr-2008 31-Dec-2012 30-Mar-2017 Ministry's procurement unit in conformity with public procurement guidelines. Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion 50% of the annual budget of Text 5% 50 29.3 the Ministry's regional offices is transferred to regional 30-Apr-2008 31-Dec-2012 30-Mar-2017 offices Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of farmers Percentage 5.00 30.00 40.00 58.30 associations/groups receiving agricultural extension 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 services from MAE and MPA Page 38 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Comments (achievements against targets): Component: Component 2: Rehabilitating Rural Roads and Market Infrastructure Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Roads rehabilitated, Rural Kilometers 0.00 0.00 1371.00 1319.00 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of market Text 0 24 40 41 infrastructure rehabilitated 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion At least 90% of built market Text 0 90 90 Page 39 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) infrastructure is properly 23-May-2008 31-Dec-2012 30-Apr-2017 30-Mar-2017 used and maintained by beneficiaries Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion At least 50% of upgraded Kilometers 0.00 75.00 50.00 33.60 rural roads are maintained at appropriate standards 03-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of producers with Text 0 22742 88396 124600 access to built or rehabilitated market 30-Apr-2008 31-Dec-2012 30-Apr-2017 30-Mar-2017 infrastructure Comments (achievements against targets): Component: Component 3: Supporting Productive Activities and Sustainable Livelihood in Rural Areas Page 40 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of beneficiairies Text 5% 60 67.90 adopting production and processing 30-Apr-2008 31-Dec-2012 30-Apr-2017 22-Dec-2016 advice/technology package Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of farmers Text 0 9600 11247 12047 (of which 51.3% associations/households of which 51% are female and 30.02% benefitting from matching female are youth) grants (% of which are femal and youth) and 25% are youth 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of microprojects Text 0 614 900 952 Page 41 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) financed by the project 30-Apr-2008 31-Dec-2012 30-Jan-2014 30-Mar-2017 Comments (achievements against targets): Component: Component 4: Coordination and Management Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Periodic progress reports and Text 0 4 4 IFRs are produced and submitted timely 30-Apr-2008 31-Dec-2012 30-Apr-2017 30-Mar-2017 Comments (achievements against targets): Page 42 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) A. KEY OUTPUTS BY COMPONENT Component 1: Building Capacity in the Ministry of Agriculture and Livestock * Objective 1: Improvement of beneficiary production and living conditions in rural areas 1. Three results-based sub-sector programs were prepared and adopted by the Ministry of Agriculture and Livestock to guide the preparation and execution of the annual budget. 2. Fifty-five percent of farmers were satisfied with the services received from the extension workers and the Ministry of Agriculture by the end of the project. Intermediate Results Indicators 3. Sixty percent of the procurement of goods and services is handled by the Ministry's procurement unit in conformity with public procurement guidelines. 4. Fifty percent of the Ministry’s annual budget is transferred to regional offices. 5. About twenty-nine percent of farmers’ associations/groups receive agricultural extension services from the MALF 1. 432 MALF staff were trained in public budgeting and planning, procurement and public Key Outputs by Component financial management, monitoring and evaluation, results-based management, production- (linked to the achievement of integrated protection, administrative and financial management, information technology, Objective/Outcome 1) survey techniques, data processing and analysis, and environmental and social safeguard measures Page 43 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) 2. Three rural radio stations were rehabilitated. 3. The offices of the National Center for Soil Studies and the MFA’s Department of Studies and Planning were rehabilitated. 4. The operational capacities of the MALF’s departments were reinforced through the provision of equipment. Objective 2: Improvement of beneficiary production and living conditions in rural areas Component 2: Rehabilitating Rural Roads and Market Infrastructure 1. 248 villages (including 340,000 inhabitants) are less isolated. Intermediate Results Indicators 2. 124,600 producers have access to newly-built or rehabilitated market infrastructure. Key Outputs by Component 1. 1319 kilometers of roads have been rehabilitated. (linked to the achievement of 2. 41 markets have been rehabilitated. Objective/Outcome 2) 3. 446 km of built market infrastructure is properly used and maintained by beneficiaries. Component 3: Supporting Productive Activities and Sustainable Livelihoods in Rural Areas 1. Project beneficiaries increased their incomes by 200 percent in participating micro-projects by the Intermediate Results Indicators end of project. 2. Increased productivity of key agricultural products (cassavas, bananas and maize). Page 44 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) 3. 67.9% of beneficiaries adopted the production and processing advice/technology package. 1. 12,047 farmers’ associations/households benefitted from matching grants (51.3 percent of whom Key Outputs by Component are female and 30 percent youth). (linked to the achievement of 2. 2 rehabilitated centers for fish farming. Objective/Outcome 3) 3. 57 fish farmers trained. 4. 952 micro-projects financed by the project. Page 45 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Remi Kini Task Team Leader Simeon Ehui Lead Economist Jean-Charles Amon Kra Financial Management Specialist Thimotee Mpoy-Kamulayi Lead Counsel Korotimi Sylvie Traore Program Assistant Mohamed Arbi Ben-Achour Social Safeguard Specialist Amadou Konare Environmental Safeguards Specialist Luc Lapointe Team Member Supervision/ICR Amadou Oumar Ba Task Team Leader(s) Clement Tukeba Lessa Kimpuni Procurement Specialist(s) Bella Diallo Financial Management Specialist Alexandre K. Dossou Team Member Antoine V. Lema Social Safeguards Specialist Carine-Reine Mbedo Ngassia Team Member Karine N. MOUKETO-MIKOLO Team Member Amadou Nchare Team Member Abdoulaye Gadiere Environmental Safeguards Specialist Page 46 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) A. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation Total 0.00 0.00 Supervision/ICR FY10 0 122,425.95 FY11 0 168,588.43 FY12 0 133,392.59 FY13 0 187,686.01 FY14 0 46,410.15 FY15 9.381 78,505.30 FY16 13.316 80,420.09 FY17 7.082 74,851.94 FY18 7.275 48,689.71 Total 37.05 940,970.17 Preparation cost is manually insered below Preparation No. of Staff Weeks US$ FY06 26.16 166,769 FY07 24.59 134,453 Total 50.75 301,223 Page 47 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) ANNEX 3. PROJECT COST BY COMPONENT Actual at Percentage Amount at Additional Revised Cost Project Closing of Components Approval Financing (US$M) (US$M) Approval (US$M) (US$M) (US$M) Component 1: Building Capacity in the Ministry 5.35 1.30 6.64 5.23 78.8% of Agriculture and Livestock Component 2: Rehabilitating Rural 23.09 4.29 27.38 27.48 100.3% Roads and Market Infrastructure Component 3: Supporting Productive Activities and 9.47 0.16 9.63 8.70 90.3% Sustainable Livelihood in Rural Areas Project Component 4: Coordination and 6.26 2.85 9.11 14.36 157.6% Management Total 44.17 8.60 52.76 55.76 105.7% Project Financing Appraisal Actual at Project Percentage of Type of Source of Funds Closing financing Estimate Appraisal (US$M) (US$M) (%) Borrower_ Initial project Counterpart 22.08 22.13 100.2% Borrower - Additional financing Counterpart 6.68 8.95 133.9% IDA (WB) IDA Credit 22.08 22.13 100.2% IDA (WB) IDA Credit 1.91 2.57 134.1% Total 52.76 55.76 105.7% Page 48 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) ANNEX 4. EFFICIENCY ANALYSIS 1. Introduction. The economic analysis of the ADRRP focused on increased crop production and the by- products of the plant and animal sectors. The project targeted categories of populations living in precarious conditions, including those relatively poor populations living in vulnerable areas from the perspective of access to resources. Such populations are handicapped by the high levels of transaction costs. One of the specific objectives is to strengthen the technical and financial capacities of the target groups to improve their productivity and incomes. 2. On the basis of surveys and studies carried out under this project, and discussions and surveys held with PCU managers, producers and cooperatives, the mission identified seven models of micro-projects related to fish farming, market gardening, poultry farming, and banana, cassava, and maize production and processing. 3. Financial Analysis. The anticipated increases in production and productivity were generally achieved following implementation of the ADRRP. This was due to: (i) improved and more intensive farming techniques; (ii) better selection of crops and an increase in the use of improved inputs; (iii) construction and rehabilitation of rural roads; and (iv) the transformation of market infrastructure into profitable businesses. Increased production costs related to ADRRP expenditures occurred because of the rational and increased use of inputs. The profitability of the project is therefore assessed by comparing the additional benefits that are derived in the implementation areas relative to the increased costs. 4. The ADRRP provided support for technologies to project beneficiaries, enabling their gross margins to be much more substantial than would have been the case without the project. The yields of several crops improved with the use of better production techniques, proper supervision, and cooperation with other projects. Producers also had lower-cost access to improved plants and cuttings. Consequently, cassava production grew from an average 8.4 metric tons/hectare in 2013 to 49.11 metric tons/hectare in 2016, whereas the national average was only 5.6 metric tons/hectare (2008). Banana production rose from 0.4 metric tons/hectare pre-project to 2.96 metric tons/hectare in 2013, and 14.03 metric tons/hectare in 2016. Maize grew on average to 0.25 metric ton/hectare in 2013 and 1.3 metric tons/hectare in 2016. Prior to the project, it had not been higher than 0.16 metric ton/hectare. Yams were introduced during the Additional Financing phase of the project. Productivity of yams increased from 5 metric tons/hectare pre-project to 27.5 metric tons/hectare in 2016 (Project Completion Report, 2017). 5. A sensitivity analysis would allow for a relaxation of these assumptions in order to take into account, as much as possible, plausible scenarios in the project areas. The selected models became fully effective between years 11 and 14, and the economic life of the micro-projects is considered to be 20 years. The Page 49 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) project-supported increases in production (see table 1 are notable and vary according to the micro-project model, from 26 percent (in fish farming) to 799 percent (in banana production). Table 1: Productivity Assumptions (main products) for Micro-project Models Average Yield during Average Pre- Increase Micro-projects Unit Project project Yield (%) (year 20) Model 1. Fish farming Pond (0.17 ha) 1,631 2,061 26 Model 2. Market Kilogram (Kg) 2,000 3,332 67 gardening Model 3. Poultry farming Eggs 93,100 193,488 118 Model 4. Bananas Bunch 1,000 8,990 799 Model 5. Cassavas Kg 6,700 9,625 44 Model 6. Maize Kg 160 783 389 Model 7. Processing 50 Kg bag 5,535 9,560 73 Source: 6. As the principal objective is to raise and sustain the farm incomes of small producers, the net additional incomes from these crops show that they significantly contributed to small producers’ income as a result of the ADRRP. All the micro-project models are profitable under current financing and market conditions. The farmers’ annual profit margins are nevertheless handicapped by the low sales prices of their crops because of problems along the value chains. In this context, a myriad of intermediaries siphoned off some of the added value created, and there have been persistent difficulties in accessing markets, obtaining information, and accessing improved inputs and credit. Also, the level of transaction costs on the market is still high. 7. Financial indicators for the seven micro-project models show good financial performance in terms of financial internal rate of return (FIRR) and net present value (NPV). The FIRR varies between 12.4 percent and 16.2 percent for these models, and the positive NPV of the opportunity cost of capital (OCC) varies between 6 and 9 percent. See table 2. Table 2: Financial Return of Various Micro-Project Models FIRR NPV @ 9% OCC NPV @ 6% OCC Models (%) (US$) (US$) Model 1. Fish farming 12.4 70.20 172.30 Page 50 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Model 2. Market gardening 16.0 15,664.50 29,052.90 Model 3. Poultry farming 16.2 30,372.70 59,147.50 Model 4. Bananas 13.8 14,844.10 35,860.00 Model 5. Cassavas 12.9 793.90 1,954.30 Model 6. Maize 14.7 493,081.10 1,081,566.00 Model 7. Processing 13.1 222,396.80 544,493.20 Source: 8. Economic analysis. The economic analysis of the ADRRP is based on a number of assumptions, of which the principal ones, include: i) An economic life of 20 years, an exchange rate of XAF 530/US$1 (2008-2013) and XAF 600 /US$1 (2014-2017), and a zero-residual project value was considered. The profits of the production from the seven micro-project models were taken into account. However, the return from the rural road and market infrastructure was not included because the mission did not have the objective elements upon which to calculate this return. The aggregation of profits from the models was done by comparing the financial and economic results to all micro-projects financed by the project. ii) The economic costs of the ADRRP were derived from the financial costs after eliminating the VAT. A standard conversion coefficient (CC) of 0.90 for the prices of products traded was derived from profits of the models selected, based on the CCs used in projects recently developed in the Republic of Congo. iii) The economic values of annual recurrent costs have been carried over for the length of the economic life of the project, with 25 percent of the operating costs starting in year 11 of the project – provided that, at the end of the project, the operating costs of the ADRRP will be covered by the Government, the cooperatives, the professional organizations and those local communities interested in the sustainability of the project benefits. iv) The OCC selected is 9 percent, reflecting the geographic isolation of the project areas and their low level of integration in the capital markets, as well as in the markets for goods and services. A simulation using an OCC of 6 percent was tested. The production pace of the project will be gradually increased to take into account the capacity of the producers in the project to adopt new technologies. 9. On the basis of these conservative assumptions, the economic internal rate of return (EIRR) of the project is close to 13 percent. The EIRR is significantly higher than the opportunity cost of capital (9 percent) and the NPV is positive. This is also true for the ADRRP’s financial rate of return , which is 12 percent. These economic analysis indicators are very satisfactory, given that part of the additional benefits have not been taken into consideration. This refers to the benefits of rehabilitating rural roads and market infrastructure, as well as other “soft� project actions. Moreover, the advantages that result from the transaction cost reductions due to easier access to technology, information, training, and technical assistance — as well as the resulting improved living conditions and diet of the project’s target groups —were not taken into account as they are difficult to quantify. The estimates of the different Page 51 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) profitability indicators are lower than the estimates provided when the project was designed (with an EIRR of 36 percent and a FIRR of 26 percent). 10. Sensitivity Analysis. The EIRR sensitivity analysis in a deterministic environment shows that a drop in aggregate profits – under constant costs – of up to 90 percent makes a project economically profitable because the EIRR is above 6 percent. This shows the robustness of the project if only a “variable -by- variable� analysis is done. However, an analysis of scenarios, meaning a simultaneous variation in costs (increase) and profits (decrease), shows that the EIRR is lower than the 9 percent OCC — with decreasing aggregate profits of 70 percent associated with a 50 percent aggregate increases of costs (see table 3). The same is true of the 80 percent (and more) drop in aggregate profits associated with increases in aggregate costs of 10 to 30 percent. A simulation of recurrent costs after the end of the project life shows that the EIRR would be positive and comparable to the base EIRR (that is, about 13.4 percent). Table 3: Sensitivity of the ADRRP’s Economic Internal Rate of Return (EIRR) to Certain Cost/Benefit Variations (%) Cost/Benefit -50% -60% -70% -80% -90% +10% 11.93 11.49 10.82 9.60 6.73 +30% 11.61 11.12 10.35 9.00 5.88 +50% 11.30 10.76 9.92 8.44 5.10 Source: Authors 11. Project Risk Analysis. In terms of the quantification of risk, a prior sensitivity analysis was performed to detect the input variables that have the most influence on the project’s level of profitability. The results showed that the input variables with the greatest influence on project profitability are related to the following factors: the sale of feed for livestock (quantity and price) and the costs of raw materials; the cumulative number of “processing� micro-projects; and the sale of poultry products (price and quantity of eggs), as well as charges for laying and starter feed (price and quantity). See figure 2 Thus, the processing and poultry farming micro-projects and their corresponding variables provided more added value to the project beneficiaries. In this context, particular attention should be paid to this aspect to ensure the sustainability of the ADRRP. Figure 1: Tornado Graph of the ERR Page 52 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Tornado Graph of ERR Impact by Input sac de 50 Kg / 2015 (N6) Matière première / Prix total (E13) œufs / prix/cout (AB6) œufs / Quantité (D6) Sac de 50 Kg / 2 000 (C15) Sac de 50 Kg / Quantité (D15) œufs / 2009 (H6) nbre de MP cumulés / 2015 (AK24) Sac de 50 Kg / 2 000 (C13) Sac de 50 Kg / Quantité (D13) 0.04 0.06 0.08 0.12 0.14 0.16 0.18 0.22 0.1 0.2 Value of ERR Source: Authors 11. The spider graph in Figure 3 shows that it is possible to explain the effect of the relatively major variations in the risk variables. As an example, when the sale of livestock feed (50 kg bags in the “processing� micro-project) increases by 10 percent, the EIRR of the project can increase by almost 60 percent. If this variable increases by only 5 percent, the EIRR increases by 28.7 percent. Conversely, when the variable decreases by 10 percent, the EIRR of the ADRRP could drop by almost 54.4 percent. Figure 2 : Spider Graph of the EIRR œufs / Quantité (D6) Spider Graph of ERR 80% œufs / 2009 (H6) 60% œufs / prix/cout (AB6) 40% Sac de 50 Kg / 2 000 (C13) Output % Change 20% Sac de 50 Kg / Quantité (D13) 0% Sac de 50 Kg / 2 000 10% 15% 0% 5% -15% -10% -5% (C15) -20% Sac de 50 Kg / Quantité (D15) -40% sac de 50 Kg / 2015 (N6) -60% Matière première / Prix t (E13) -80% nbre de MP cumulés / 2015 (AK24) Input % Change Source:authors Page 53 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) 12. Risk analysis using the Monte Carlo simulation makes it possible to estimate the probability of the ADRRP’s profitability. An analysis of the probability distributions (figure 4) shows that the level of risk remains very low in terms of the non-profitability of the project. The probability of registering an EIRR that is lower than the opportunity cost of capital (6 percent) is only 5 percent. In other words, in 90 percent of the situations, the EIRR would be higher than the OCC. The same applies to the NPV. Figure 4: Probability Distributions of the Project’s EIRR Source : authors Note : TRE=Economic Rate of Return 13. In summary, the ADRRP is not very sensitive to variations in aggregate costs and benefits. Its level of profitability remains acceptable, despite the simultaneous variations in costs and benefits. The Monte Carlo simulation analysis shows that the probability of the ADRRP achieving profitability is very high (90 percent), which can be considered very satisfactory. Costs of rural roads and market infrastructure. The average cost per kilometer for the rehabilitation of rural roads is close to XAF 6 million (US$ 10,740 equivalent) /Km. It varies from XAF 5.2 million (US$ 9308 equivalent)/Km to about XAF 9 million (US$ 16,110 equivalent) /Km (Table 4). This takes into account the recommendations of the MEER concerning developing rural roads on the priority network and the scaling of construction works While it is difficult to estimate the area served by rural roads, the ratio of the amount of the rural road rehabilitation works per inhabitant is relatively high (table 4) during the initial phase of the project (XAF 0.3 to 1.1 million (US$ 537 ~US$ 1969 equivalent) between 2010 and 2012). However, it was weak in 2015-16 ( XAF 0.04 to 0.08 million (US$ 71.60 ~ US$1432 equivalent). Table 4: Changing Costs of Rehabilitating Rural Roads in the ADRRP Year of Number of Amount (XAF) Number of Amount per Amount per Page 54 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Rehabilitation kms Inhabitants Inhabitant km 2010 275 1,642,195,705 4,543 361,578 5,971,621 2011 557 2,905,739,000 6,845 424,501 5,216,767 2012 437 2,712,811,000 2,387 1,136,278 6,688,355 2015-2016 50 451,365,000 5,609 80,479 9,027,300 Total 1,319 7,712,110,705 188,997 40,805 5,846,938 Source: Compiled from data provided by the Head of Component 2. 14. The costs of rehabilitated rural roads varied from 112 percent to 118 percent compared to the forecasts, whereas they were relatively lower for rehabilitated market infrastructure. Taking into consideration the work completed during the ADRRP, the cost of rehabilitation to be incorporated into future projects can be estimated at between XAF 7 and 12 million (US$ 12,530 ~ US$ 21,480 equivalent) per kilometer (table 5).41 Table 5: Comparison of Actual and Estimated Costs of Rural Roads and Market Infrastructure Works Actual/ Estimated No. Works Actual Cost Estimated Cost (XAF) Cost (%) 1 Rural Roads 1.1 Les saras-Gare Ntessi (41 km) 407,000,000 465,000,000 114% 1.2 Kingonda - Kitomba / Nzansa (65 km) 328,500,000 368,000,000 112% 1.3 Kengué-Kimboto (40 km) 179,251,000 210,727,000 118% 1.4 Odziba - Ndzion Ndzion – Ingah (52 km) 175,000,000 206,391,250 118% 2 Market Infrastructures 2.1 Ikongono Market 29,242,200 33,412,200 114% 2.2 Bohoulou Market 57,001,700 63,849,700 112% 2.3 Ngbala Market 36,880,175 42,213,075 114% 2.4 Makotimpoko Market 118,442,500 127,898,500 108% 41 According to the Ministry of Infrastructure and Road Maintenance. Page 55 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) 2.5 Nsoungui Offices 90,000,000 99,500,000 111% 2.6 Mossaka Market 109,152,350 124,377,100 114% Source: Compiled from data provided by the Head of Component 2. 15. The cost of maintaining rural roads is a function of the level of deterioration of the roads and the method chosen to perform maintenance work (that is, highly-equipment intensive or highly-labor intensive). The amounts spent during the ADRRP are considerably lower than those of the MEER. They vary according to the geographic characteristics/vegetation: 84-90 percent in hills, 13-23 percent in mountains, and 10-32 percent on flat land. 16. The costs of upkeeping rural roads are higher, depending on the mode of maintenance (table 6). It costs on average XAF 2.2 million (US$ 3,938 equivalent) under the highly-machine-intensive method, and XAF 0.5 to 1.1 million (US$895 ~US$ 1969 equivalent) /km using the highly-labor-intensive method. The latter method has a good record in terms of community participation in the form of workforce engagement, the use of local construction materials, the quality and speed of construction and completion. The equipment-intensive method has support from a mainly skilled, yet very limited labor force. Table 6: ADRRP and MEER:s Comparison of Rural Road Upkeep Costs Geographic ADRRP N° Construction Department Km Vegetation MEER Amount feature amount RURAL ROADS UPKEEP – Highly-equipment-intensive method (2015-2016) 1 Mavitou-lac cayo Kouilou 11 Flat Savana 34,240,000 327,124,983 2 Pilikondi-Bilala-Cacao Kouilou 38 Mountain Forest 96,000,000 759,742,518 3 Les Saras-Ntessi/Aviation-Bondika Kouilou 41 Mountain Forest 119,800,000 521,583,900 4 Malolo-Dihessé Niari 27 Flat Savana 73,700,000 229,178,500 5 Kidzoumba-Kibamba Bouenza 28 Hill Savana 73,250,000 - Bouansa-Ngouedi- 6 Bouenza 30 Hill Savana 73,340,000 - Kinzaba/Brettelles 7 Massa-Maty centre-Binokina Pool 70 Flat Savana 87,567,750 219,165,100 8 Odziba-dzion dzion Pool 39 Flat Savana 81,260,000 183,079,248 9 Boko-Ngondzo Pool 18 Mountain Savana 67,550,000 - Western 10 Baya – Mbama 45 Hill Forest 80,812,500 89,606,625 basin Page 56 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Western 11 Dzouama-doubandzo-Adingo 37 Hill Forest 91,620,000 109,678,000 basin 12 Yonon-Ossa-Kialé Plateau 17 Flat Savana RURAL ROADS UPKEEP – Highly-Labor Intensive Method (2015 - 2016) - Phase 1 1 Inga-Ingolo Pool 13 Flat Savana 4,500,000 2 Yonon-Ossa-Kialé Pool 17 Flat Savana 9,981,000 3 Djambala-Abalandolo Plateau 10 Flat Savana 5,508,000 RURAL ROADS UPKEEP – Highly-Labor Intensive Method (2016) - Phase 2 1 Inga-Ingolo Pool 13 Flat Savana 6,000,000 2 Ingolo-Mah-Imvouba Pool 22 Flat Savana 32,886,500 3 Djambala-Abalandolo Plateau 10 Flat Savana 12,500,000 Total 45 51,386,500 Source: Cthompiled from data provided by the Head of Component 2. 17. Aspects of ADRRP implementation. Despite the fact that the project entered into effect in 2007, implementation did not truly begin until 2010. The project started receiving good support in 2012, notably with the change of the TTL, the PCU Coordinator, and the implementation of recommendations by the supervision missions. Many activities, in particular infrastructure construction, had seen delays of up to one year. Indeed, until 2010, the completion rate of the project’s Annual Work Plan Budgets (AWPBs) averaged 46 percent (table 7). The physical execution of the AWPBs significantly improved starting in 2011, with a completion rate of 79 percent, climbing to 81 percent in 2012 and 94 percent in 2013. Following a decline that coincided with the end of the initial phase of the project and the start-up of the additional financing phase, the rate reached 100 percent in 2017. Table 7: Changes in the Rate of the Project’s Physical Execution of the AWPBs 2009 2010 2011 2012 2013 2014 2015 2016 2017 51% 40% 79% 81% 94% 29% 74% 76% 100% Source: Compiled from data provided by the PCU. 18. The implementation difficulties and the delays incurred are attributable to the changes in the PCU between 2007 and 2013. In addition to replacing the Project Coordinator, following the governmental and World Bank decision in 2010, the Head of Monitoring and Evaluation left his job in February 2013— as did other members of the operational components and contracting team. Moreover, the weak technical skills of some of the administrative staff contractors, as well as difficulties in accessing a few target zones Page 57 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) — especially during the rainy season,42 complicated project work and contributed to delays in completing the project. Additionally, the insufficient experience of certain providers in the areas of design, surveys and/or monitoring increased the workload for the project team. 19. Delays can also be partly explained by a lack of knowledge of contracting procedures and slowness in performing this work. The withdrawal request had very long processing delays (averaging about 88 days43). These administrative difficulties demonstrate the need to strengthen capacities among the members of the PCU, especially in the absence of a self-evaluation system or peer evaluations of the member’s performance. Moreover, this situation has led to adjustments in estimates, as well as in the plan to complete the rural roads work. 20. Nevertheless, a number of factors combined to produce the expected results (table 8). Indeed, there was a dynamism injected into the project by the new TTL, the new PCU team, the effective involvement of the various stakeholders,44 as well as the strengthening of capacities of those involved in the implementation of the project. In addition, the revision of the procedures manuals and the creation of a list of efficient providers helped to achieve the project results. The numerous difficulties encountered by project managers led to a significant cost overrun in completion of component 4, (conducting project coordination and management), compared to what had originally been forecast. Table 8: Estimated versus Actual Costs of Project Components Estimated amount Actual amount Project Components Achievement rate (%) (US$) (US$) 1. Building capacity in the Ministry of Agriculture, Fisheries, and Livestock- (MALF) 6,642,488 5,232,711 79% 2. Rehabilitating rural roads and market infrastructure 27,382,800 27,476,275 100% 3. Supporting productive activities and sustainable livelihoods in rural areas 9,628,582 8,698,314 90% 42 Rehabilitation works on rural roads were often disrupted by rain, which delayed deliveries and required resumption of some construction works at the beginning of the dry season in June. 44 Steering Committee, Government, and World Bank. Page 58 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) 4. Conducting project coordination and management 9,109,024 14,357,259 158% Total project 52,762,894 55,764,559 106% Source: PCU (Monitoring-Evaluation Unit), 2017. 21. The successive updates to the “SUCCESS� accounting management software and the training of PCU staff on how to best use this tool — as well as recommendations and support from World Bank missions and outside auditors — helped resolve the financial management difficulties encountered at the beginning of the project. Improvements to the procedures manual covering the administrative and financial management accounting of the project also helped to streamline management and internal control. These successes are confirmed by the disbursement rate, which despite the difficulties encountered, reached 95.7 percent in 2013 and 98.48 percent in 2017. 22. These figures illustrate the success of the project in terms of the satisfactory financial and physical completion rates achieved. However, these rates are in contrast with the results of the interventions in qualitative terms. Indeed, the project results framework does not include indicators to facilitate an assessment of the results of interventions in qualitative terms. Consequently, an evaluation of the efficiency of investments, notably in rural roads and market infrastructure, cannot be objectively made — despite the shortcomings of a number of providers, particularly during the project startup phase. Page 59 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) ANNEX 5. BORROWER, CO-FINANCIER AND OTHER PARTNER/STAKEHOLDER COMMENTS MINISTERE DU PLAN, DE LA STATISTIQUE ET REPUBLIQUE DU CONGO DE L’INTEGRATION REGIONALE Unité * Travail * Progrès --------------- ------------ CABINET ------------ Brazzaville, le N°______________/MPSIR-CAB.- Le Ministre A Monsieur le Directeur des Opérations de la Banque Mondiale pour la République du Congo, -Kinshasa- Réf. : V/L n°….. du … octobre 2017 Objet : Coopération Congo/Banque mondiale Projet de Développement Agricole et de Réhabilitation des Pistes Rurales (PDARP) Page 60 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Monsieur le Directeur des Opérations, Faisant suite à votre correspondance sus référencée relative au projet cité en objet, j’ai l’honneur de vous faire savoir les commentaires formulés par la partie congolaise sur le rapport d’achèvement du Projet de Développement Agricole et de Réhabilitation des Pistes Rurales (PDARP). Ces commentaires et observations se présentent comme suit : - la concentration des activités dans des zones spécifiques dites pôles de croissance agricole a permis de le circonscrire et d’en avoir un meilleur impact du fait de synergie entre les composantes opérations 1 et 2. Mais cela ne s’est fait que pendant le financement additionnel (supra 22) ; - l’efficacité de l’ODP est noté substantiel (point 2.2). Nous pourrions en convenir avec vous, malgré les importants changements qui ont été induits par la mise ne œuvre du PDARP dans des zones du projet qui auraient peut-être permis de mieux noter l’efficacité de ce projet ; - le rapport affirme qu’il n’aurait pas d’éléments d’évidence, pour appuyer l’impact réel du projet en termes d’interventions du MAEP orientées vers l’amélioration des capacités des populations pauvres à accroitre leurs revenus (supra 38) : le projet a justement participé à améliorer les capacités des administrations notamment déconcentrés dans leur travail quotidien consistant à encadrer et suivre les producteurs pour leurs activités génératrices de revenus. Les moyens matériels et financiers, ainsi que les formations dont ils ont bénéficié grâce au projet n’ont pas servi qu’aux activités soutenues par ledit projet ; - la pérennisation des activités du projet (supra 53) ne devrait pas se limiter à celle ayant été le plus rentable (transformation et aviculture) mais à toutes les activités ayant survécu à la clôture du projet dont certaines pourraient s’avérer plus rentables que pendants la mise en œuvre du projet. Il s’agit notamment de certaines microprojets de pisciculture, mais aussi de maraichage qui sont déjà parmi les plus rentables ; - contrairement à ce qu’affirme le rapport, les dépassements des délais de mise en œuvre de certaines activités du financement additionnel n’ont nullement été dus à l’inefficacité des équipes du projet (supra 56). En effet, le projet a été souvent été confronté aux problèmes exogènes tels que les retards dans le décaissement des fonds de contrepartie et la capacité d’absorption des très petites entreprises souvent adjudicatrices des marchés du PDARP. En réalité le FA n’a pas été exécuté sur 36 mois, mais sur 29, dont une bonne partie consacrée à l’attente des fonds de contrepartie (2015) ; - nous convenons avec le rapport d’achèvement que la pertinence du projet est hautement satisfaisante (supra 58); - la performance globale du PDARP jugée modérément satisfaisante (supra 60) contraste quelque peu à l’ensemble des avancées marquées par ce projet, en termes d’innovations, d’appui aux Page 61 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) producteurs et même en renforcement des capacités des administrations qui ont eu un impact réel et durable. Ce projet pourrait raisonnablement être jugé satisfaisant ; - il est vrai que la pratique du screening a été systématisée à partir de 2014, lors de la mise en œuvre du financement additionnel (supra 63), mais les études d’impact environnemental et social (EIES) ont toujours été réalisées depuis le début du projet et ont permis de mettre en application des plan de gestion environnementale et sociales (PGES) par les entreprises adjudicatrice des marchés d’infrastructure ; - la stratégie de faire faire (supra 78) a été prévue par le document d’évaluation du projet (PAD) et non commandée par une certaines insuffisance du personnel du projet. Cette stratégie est même une grande innovation du projet et a permis une appropriation du projet par les acteurs nationaux ; ce qui constitue un point très positif ; - le rapport juge la durabilité du projet modérément insatisfaisante (supra 97), contrastant ainsi avec l’ensemble des innovations apportées par le PDARP (financements à coûts partagés, faire- faire, entretien des pistes rurales en HIMO, mise en œuvre des PGES, autonomisation de certains groupements bénéficiaires…). Pour notre part, cette durabilité devrait, à défaut être jugée satisfaisante ; - la question de la durée du projet jugée trop courte (supra 114) pourrait être échangée en interne par les équipes de la Banque mondiale pour voir ce qui est mieux profitable pour le pays. Nous n’avons pas plus de commentaires que cela sur cette question ; - nous partageons avec les auteurs du rapport, les recommandations sur la durabilité des acquis du projet et les aspects environnementaux (supra 115 et supra 116) qui devraient justement conforter les acquis du projet en matière d’infrastructures, voire de financement des activités des producteurs. Certaines activités devraient être même amorcées dabs le cadre du projet d’appui au développement de l’agriculture commerciale que le Congo vient de signer avec la Banque mondiale. Sous réserve des informations que nous n’aurions pas bien cernées dans le rapport d’achèvement du PDARP, Veuillez agréer, Monsieur le Directeur des Operations, l’expression de ma considération distinguée. Ingrid Ghislaine Olga EBOUKA-BABACKAS Page 62 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) ANNEX 6. SUPPORTING DOCUMENTS (IF ANY) A). Bank 1. Aide Mémoire, Additional Financing. Management letter. April 2016. The World Bank, Washington D.C. 2. Amendment to Financing Agreement. August 2015. The World Bank, Washington D.C. 3. Project Appraisal Document, Agricultural Development and Rural Roads Rehabilitation Project (P095251), Report No. 39425-CG, May 9, 2007. 4. Financing Agreement for the Agricultural Development and Rural Roads Rehabilitation Project, Grant Number H302-COB, 2007. 5. Financement Additionnel du PDARP, Procès-verbal des négociations, 17 décembre 2013. 6. Global Environmental Facility (GEF). Project Appraisal Report. 2016. 7. Project Information Document (PID), Appraisal Stage, Report No: AB2256, 2007. 8. Project Information Document (PID), Additional Financing, Report No.: PIDA2461, 2013. 9. Project Paper on a Proposed Additional Credit, Report No: PAD 709, January 6, 2014. 10. Memorandum and Recommendation of the President of the International Development Association to the Executive Directors on a Proposed Additional Credit to the Republic of Congo for the Additional Financing for the Agricultural Development and Rural Roads Rehabilitation Project, January 6, 2014. 11. Memorandum and Recommendation of the President of the International Development Association to the Executive Directors on a Proposed Restructuring of the Agricultural Development and Rural Roads Project (Grant No. H302-COB) for the Republic of Congo, July 2013. 12. Implementation Status and Results Reports (a total of 16 reports during the project implementation period). 13. Financing Agreements and Restructuring documents. 14. Supervision reports. 2008-2017. 15. Project Concept Note. 2005. The World Bank, Washington D.C. (B). Client 16. Note de politique agricole, MAEP, 2009. 17. Plan National de Développement, 2012-2016. 18. Plan national d'investissement agricole et de sécurité alimentaire et nutritionnelle (PNIASAN), 2015. 19. Enquête de base pour le suivi et l’évaluation du PDARP, Rapport d’analyse, MAEP, Centre National de la statistique et des études économiques, 2011. 20. Enquête de clôture du projet de développement agricole et de réhabilitation des pistes rurales, Rapport final, MAEP, Institut national de la Statistique, 2013. 21. Lettres d’informations bimestrielle du Projet de développement agricole et de réhabilitation des pist es rurales, No 01 à No 20, PDARP, Décembre 2011 – Janvier 2012 à Mars – avril 2017. 22. Manuel d’exécution du PDARP pour le Financement Additionnel, UCP, MAEP, 2014. 23. État du marché des produits maraîchers et suivi économique des exploitations maraîchères, Rapport, Observatoire Économique des Filières Maraichères, Février 2017. Page 63 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) 24. Évolution du décaissement et de l’exécution financière du PDARP, Cellule de suivi -évaluation, UCP, 2017. (C). Completion Reports and ICR Guidelines: 25. Guidelines for Reviewing World Bank Implementation Completion and Results Reports, A Manual for Evaluators, IEG, World Bank, August 2014. 26. Implementation Completion and Results Report (ICR) for Investment Project Financing (IPF) Operations, Bank Guidance, OPSPQ, July 2017. (D). Impact Evaluations: 27. Rapport d’achèvement du PDARP, UCP, MAEP, 2014. 28. Rapport d’achèvement et d’impact du PDARP, UCP, MAEP, 2017. 29. Étude d’impact du projet. Institut National de la Statistique du Congo (INS), 2013. 30. Rapport de suivi environnemental et social de clôture du PDARP, UCP, 2017. 31. Rapport circonstancié sur l’impact de l’instabilité socio-politique sur les réalisations du PDARP dans le département du Pool, Antenne Départementale de Suivi de Kinkala, 2017. (E). Others 32. Project Appraisal Document of the Agricultural Development and Rural Road Rehabilitation Project, GEF, March 2016. 33. Repères sur les théories du changement, Synthèse bibliographique réalisée dans le cadre du groupe de travail F3E – COTA « Agir pour le changement », Mai 2012. 34. Entretenir une croissance plus inclusive en République du Congo, Rapport, FMI, 2015. 35. Bernard PECCOUD. Stratégie d'entretien des pistes. Octobre 2013. 36. Observatoire économique des filières maraichères. Agricongo. Février 2017. 37. Expérience des bénéficiaires de financement du PDAC. PCU. 2017. Page 64 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) ANNEX 7. Map of the Achievements of the Agricultural Development and Rural Roads Rehabilitation Project Page 65 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Annex 8. Theory of Change Page 66 of 68 The World Bank Agricultural Development and Rural Roads Rehabilitation Project ( P095251 ) Page 67 of 68