Document of The World Bank FOR OFFICIAL USE ONLY Report No: PAD1418 INTERNATIONAL DEVELOPMENT ASSOCIATION PROJECT APPRAISAL DOCUMENT ON A PROPOSED GRANT IN THE AMOUNT OF SDR 29.2 MILLION (US$40 MILLION EQUIVALENT) TO THE REPUBLIC OF BURUNDI FOR A SOCIAL SAFETY NETS PROJECT (MERANKABANDI) November 22, 2016 Social Protection and Labor Global Practice Africa Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. CURRENCY EQUIVALENTS (Exchange Rate Effective October 31, 2016) Currency Unit = BIF BIF 1650 = US$1 SDR 1 = US$ 1.37 FISCAL YEAR January 1 – December 31 ABBREVIATIONS AND ACRONYMS ABUTIP Burundi Public Works Agency, Agence Burundaise de Travaux d’Intérêt Public AWPB Annual Work Plan and Budget BCC Behavior Change Communication BRB Banque de la République du Burundi BRAC Bangladesh Rehabilitation and Assistance Committee CAM Medical Assistance Card Carte d’Assistance Maladie CAS Country Assistance Strategy CDFC Community Family Development Center, Centre de Développement Familial Communautaire CNCA National Aid Coordination Committee, Comité National de Coordination des Aides CNPS National Commission for Social Protection, Commission Nationale de Protection Sociale CPM Commission de Passation des Marchés CPPS Provincial Commission for Social Protection, Commission Provinciale de Protection Sociale CRS Catholic Relief Services CSLP Poverty Reduction Strategy Framework, Cadre Stratégique de Réduction de la Pauvreté CTB Belgian Technical Cooperation, Coopération Technique Belge CT-CNPS Technical committee for the National Social Protection Strategy, Comité Technique de la Commission Nationale de Protection Sociale DA Designated Account ECD Early Childhood Development ECVMB Living Standards Measurement Study Survey 2013-14, Enquête sur les Conditions de Vie des Ménages au Burundi, de la Main d'Oeuvre et de la Protection Sociale FIDA International Fund for Agricultural Development, Fonds International pour le Développement Agricole FM Financial Management GDP Gross Domestic Product ICB International Competitive Bidding IDA International Development Association IEC Information, Education, Communication IFR Interim Financial Report IPC Integrated Food Security Phase Classification IPP Indigenous Peoples’ Plan IPV Intimate Partner Violence ISTEEBU Statistical Institute of Burundi, Institut de Statistiques et d’Études Économiques du Burundi M&E Monitoring and Evaluation MDC Ministry of Communal Development, Ministère du Développement Communal MDG Millennium Development Goals Merankabandi Name of the Burundi Cash Transfer Program, roughly translated from Kirundi as « Be equal to others » MFPTE Ministry of Civil Service, Labor and Employment, Ministère de la Fonction Publique, du Travail et de l’Emploi MIS Management Information System MSPLS Ministry for Public Health and Fight against AIDS, Ministère de la Santé Publique et de la Lutte contre le SIDA MDPHASG Ministry of Human Rights, Social Affairs and Gender, Ministère des Droits de la Personne Humaine, des Affaires Sociales et du Genre NGO Non-Governmental Organization PA Project Account PDO Project Development Objectives PIM Project Implementation Manual PIU Project Implementation Unit PMS Minimum Health Package Impact Evaluation Survey, Enquête Ménage pour le Suivi et l'Évaluation de l'Impact de l'Appui au Système de Remboursement du Paquet Minimum de Santé, 2012-13 PMT Proxy Means Test PNPS National Social Protection Policy, Politique Nationale de Protection Sociale PRSP Poverty Reduction Strategy Paper RFP Request for Proposal SDG Sustainable Development Goals SEP/CNPS Permanent Executive Secretariat for the National Social Protection Commission, Secrétariat Exécutif Permanent pour la Commission Nationale de Protection Sociale SNPS National Social Protection Strategy, Stratégie Nationale de Protection Sociale SSNA Social Safety Net Assessment SORT Systematic Operations Risk- Rating Tool SP Social Protection TA Technical Assistance TOR Terms of Reference UNDP United Nations Development Program UNICEF United Nations Children’s Fund WB World Bank WFP World Food Program WHO World Health Organization Regional Vice President: Makhtar Diop Country Director: Bella Bird Senior Global Practice Director: Michal Rutkowski Practice Manager: Dena Ringold Task Team Leader(s): Bénédicte de la Brière BURUNDI Social Safety Nets Project TABLE OF CONTENTS Page I. STRATEGIC CONTEXT .................................................................................................1 A. Country Context ............................................................................................................ 1 B. Sectoral and Institutional Context................................................................................. 2 C. Higher Level Objectives to which the Project Contributes .......................................... 6 II. PROJECT DEVELOPMENT OBJECTIVES ................................................................7 A. PDO............................................................................................................................... 7 Project Beneficiaries ........................................................................................................... 7 PDO Level Results Indicators ............................................................................................. 8 III. PROJECT DESCRIPTION ..............................................................................................8 A. Project Components ...................................................................................................... 8 B. Project Financing ........................................................................................................ 15 Project Cost and Financing ............................................................................................... 15 C. Lessons Learned and Reflected in the Project Design ................................................ 15 IV. IMPLEMENTATION .....................................................................................................18 A. Institutional and Implementation Arrangements ........................................................ 18 B. Results Monitoring and Evaluation ............................................................................ 18 C. Sustainability............................................................................................................... 19 V. KEY RISKS ......................................................................................................................19 A. Overall Risk Rating and Explanation of Key Risks.................................................... 19 VI. APPRAISAL SUMMARY ..............................................................................................20 A. Economic and Financial Analysis ............................................................................... 20 B. Technical ..................................................................................................................... 23 C. Financial Management ................................................................................................ 23 D. Procurement ................................................................................................................ 24 E. Social (including Safeguards) ..................................................................................... 25 F. Environment (including Safeguards) .......................................................................... 26 G. World Bank Grievance Redress .................................................................................. 26 Annex 1: Results Framework and Monitoring .........................................................................27 Annex 2: Detailed Project Description .......................................................................................36 Annex 3: Implementation Arrangements ..................................................................................52 Annex 4: Implementation Support Plan ....................................................................................72 Annex 5: Economic Analysis .......................................................................................................75 . PAD DATA SHEET Republic of Burundi Social Safety Nets Project (Merankabandi) (P151835) PROJECT APPRAISAL DOCUMENT . AFRICA Report No.: PAD1418 . Basic Information Project ID EA Category Team Leader(s) P151835 B - Partial Assessment Benedicte Leroy De La Briere Lending Instrument Fragile and/or Capacity Constraints [ ] Investment Project Financing Financial Intermediaries [ ] Series of Projects [ ] Project Implementation Start Date Project Implementation End Date 15-Dec-2016 31-Mar-2022 Expected Effectiveness Date Expected Closing Date 15-Mar-2017 30-Jun-2022 Joint IFC No Practice Senior Global Practice Country Director Regional Vice President Manager/Manager Director Dena Ringold Michal J. Rutkowski Bella Bird Makhtar Diop . Borrower: Ministry of Finance and Planning for Economic Development Responsible Agency: Ministry of Human Rights, Social Affairs and Gender Contact: Michel Nyabenda Title: Permanent Executive Secretary, CNPS Telephone No.: +257 22273081 Email: nyabendamic@gmail.com . Project Financing Data(in USD Million) [ ] Loan [X] IDA Grant [ ] Guarantee [ ] Credit [ ] Grant [ ] Other Total Project Cost: 40.00 Total Bank Financing: 40.00 Financing Gap: 0.00 . Financing Source Amount IDA Grant 40.00 Total 40.00 . Expected Disbursements (in USD Million) Fiscal 2017 2018 2019 2020 2021 2022 Year Annual 2.90 10.61 11.81 10.01 3.53 1.14 Cumulati 2.90 13.51 25.32 35.33 38.86 40.00 ve . Institutional Data Practice Area (Lead) Social Protection & Labor Proposed Development Objective(s) The objectives of the proposed Project are to provide regular cash transfers to extreme poor and vulnerable households with children in selected areas while strengthening the delivery mechanisms for the development of a basic social safety net system. . Components Component Name Cost (USD Millions) 1. Support to the cash transfer program 26.70 2. Support to the key delivery mechanisms of a basic social 7.50 safety net system 3. Project management 5.80 . Systematic Operations Risk- Rating Tool (SORT) Risk Category Rating 1. Political and Governance High 2. Macroeconomic Substantial 3. Sector Strategies and Policies Moderate 4. Technical Design of Project or Program Substantial 5. Institutional Capacity for Implementation and Sustainability Substantial 6. Fiduciary Substantial 7. Environment and Social Moderate 8. Stakeholders Substantial 9. Other OVERALL Substantial . Compliance Policy Does the project depart from the CAS in content or in other significant Yes [ ] No [ X ] respects? . Does the project require any waivers of Bank policies? Yes [ ] No [ X ] Have these been approved by Bank management? Yes [ ] No [ ] Is approval for any policy waiver sought from the Board? Yes [ ] No [ X ] Does the project meet the Regional criteria for readiness for implementation? Yes [ X ] No [ ] . Safeguard Policies Triggered by the Project Yes No Environmental Assessment OP/BP 4.01 X Natural Habitats OP/BP 4.04 X Forests OP/BP 4.36 X Pest Management OP 4.09 X Physical Cultural Resources OP/BP 4.11 X Indigenous Peoples OP/BP 4.10 X Involuntary Resettlement OP/BP 4.12 X Safety of Dams OP/BP 4.37 X Projects on International Waterways OP/BP 7.50 X Projects in Disputed Areas OP/BP 7.60 X . Legal Covenants Name Recurrent Due Date Frequency Annual Work Plans and Budgets X Yearly Description of Covenant No later than October 30 in each calendar year (or one month after effective date for the first year of Project Implementation), the Recipient shall prepare and furnish to the Association: (i) a draft annual work plan and budget for the project (including training and operating costs) for the subsequent calendar year of project implementation; as well as (ii) any IPP which may be required for the implementation of the activities included in the draft annual work plan and budget. Name Recurrent Due Date Frequency Project reports X SemiAnnual Description of Covenant Each Project Report shall cover the period of one calendar semester, and shall be furnished to the Association not later than forty-five (45) days after the end of the period covered by such report. Name Recurrent Due Date Frequency Interim Financial Reports X Quarterly Description of Covenant Recipient shall prepare and furnish to the Association not later than forty-five (45) days after the end of each calendar quarter, interim unaudited financial reports for the Project covering the quarter, in form and substance satisfactory to the Association. Name Recurrent Due Date Frequency Internal auditor 15-Jul-2017 Description of Covenant The Recipient shall recruit, no later than four (4) months after the Effective Date, an internal auditor in accordance with the provisions of Section III.C of this Schedule 2 to the Financing Agreement. Name Recurrent Due Date Frequency External auditor 15-Jul-2017 Description of Covenant The Recipient shall recruit, no later than four (4) months after the Effective Date, an external auditor in accordance with the provisions of Section III.C of this Schedule 2 to the Financing Agreement. Name Recurrent Due Date Frequency Financial and accounting system 15-Feb-2017 Description of Covenant No later than two (2) months after the Effective Date, the Recipient shall establish, within the Project Implementation Unit, a computerized financial and accounting system satisfactory to the Association, and successfully train relevant staff in the use thereof. . Conditions Source Of Fund Name Type IDAT Payment of the first Cash Transfer Installment Disbursement Description of Condition No withdrawal shall be made for payment of the First Cash Transfer installment under Category (1)(a) unless the Recipient has engaged the Payment Agent in accordance with the provisions of Schedule 2 Section I.A.4 of the Financing Agreement and (b) the Recipient has adopted aspects of the Project Implementation Manual pursuant to Schedule 2 Section I.D.2(b) of the Financing Agreement Source Of Fund Name Type IDAT Payment of the second Cash Transfer Installment Disbursement Description of Condition No withdrawal shall be made for payment of the Second Cash Transfer installment under Category (1)(b) unless the Recipient has submitted to the Association the technical audit report, in form and substance satisfactory to the Association, referred to in Schedule 2 Section II.B.6 of the Financing Agreement on the use of 20% of the proceeds of the First Cash Transfer Install installment. Source Of Fund Name Type IDAT Establishment of PIU Effectiveness Description of Condition The Recipient has established the PIU and recruited to the PIU, a project coordinator, a procurement specialist, a financial management specialist and an accountant in accordance with the provisions of Section I.A.2 of Schedule 2 of the Financing Agreement Source Of Fund Name Type IDAT Project Implementation Manual Effectiveness Description of Condition The Recipient has adopted aspects of the Project Implementation Manual pursuant to Section I.D.2(a) of Schedule 2 of the Financing Agreement Team Composition Bank Staff Name Role Title Specialization Unit Benedicte Leroy De La Team Leader Lead Economist GCGDR Briere (ADM Responsible) Melance Ndikumasabo Procurement Senior Procurement Procurement GGO07 Specialist (ADM Specialist Responsible) Christian Simbananiye Financial Consultant Financial GGO19 Management Management Specialist Specialist Alain Tribert Team Member Consultant GSP01 Nimpagaritse Alexandra C. Bezeredi Safeguards Advisor Lead Social Social safeguards GSU01 Development Specialist Alice Museri Team Member Team Assistant AFMBI Antonia T. Koleva Team Member Senior Operations GSP01 Officer Celine Ferre Team Member Consultant Poverty map GSP07 Ishanlosen Odiaua Safeguards Consultant Social safeguards GEN05 Specialist Issa Thiam Team Member Finance Officer Finance officer WFALA Laure Mercereau Team Member Consultant Institutional GSPDR Assessment Lucian Bucur Pop Peer Reviewer Senior Economist GSP06 Mariam Denise Brain Team Member Program Assistant Team Assistant GSP01 Mary C.K. Bitekerezo Safeguards Senior Social Social safeguards GSU07 Specialist Development Specialist Nneoma Veronica Counsel Senior Counsel LEGAM Nwogu Penelope Jane Aske Team Member Sr Social Protection Quality Assurance GSP07 Williams Specialist Stefanie Koettl - Peer Reviewer Senior Economist GSP03 Brodmann Suleiman Namara Peer Reviewer Senior Social GSP07 Protection Economist Extended Team Name Title Office Phone Location Windinbobibe Moussa Payment Systems +22670232650 Ouagadougou, Burkina Anselme Kabore Faso . Locations Country First Location Planned Actual Comments Administrative Division Burundi Gitega Gitega Province X Bugendana, Buraza, Gitega, Itaba Burundi Ruyigi Ruyigi Province X Butagwanza 2, Butezi, Bweru, Gisuru Burundi Karuzi Karuzi Province X Bugenyuzi, Gihogazi, Mutumba, Nyabikere Burundi Kirundo Kirundo Province X Bugabira, Busoni, Kirundo, Ntega . I. STRATEGIC CONTEXT A. Country Context 1. Burundi’s economic performance improved over the last decade but the gains are fragile and poverty and vulnerability remain widespread. After the Arusha breakthrough peace agreements in 2000 and the subsequent decline in violence by 2005, the Government of Burundi managed to stabilize the country's economy in a fragile environment. However, since early 2015, the political crisis has reversed some of these previous gains and triggered a severe economic crisis, which impacts the most vulnerable and their ability to meet basic needs 1. The economy contracted by seven percent in 2015 and prospects for recovery are still uncertain. At the same time, several donors have suspended aid to the country, decreasing the overall contribution from 13 to 10.3 percent of Gross Domestic Product (GDP) between 2014 and 2015. Furthermore, public debt is increasing. 2. The latest poverty data 2 shows that nearly two-thirds of Burundians are poor and that the 2015 Millennium Development Goals (MDG) target on poverty reduction (17.1 percent) was not met. Per capita gross national income more than doubled between 2005 (US$130) and 2013 (US$280) in nominal terms, but fell to US$270 in 2015. The country is now the second poorest in Africa after Malawi. To start reducing poverty significantly, the economic growth rate would have to reach at least seven percent; growth would need to become more inclusive, especially in rural areas; and a strong safety net would need to be put in place in order to prevent the deepening poverty, provide some redistribution, and build a foundation for poverty reduction. 3. Among the structural drivers of poverty and fragility is the growing scarcity of land, a major threat to the livelihoods of the poor. About 90 percent of Burundians live in rural areas. Population growth steadily decreases the average land ownership. Burundi has a very high population density (340 inhabitants per km2, and up to 500 in some collines 3), and close to 30 percent of farm households own less than one hectare of land, a major factor of vulnerability. The conflict in the 1990s and early 2000s had caused large displacement 4 (up to half of the population in some areas) and loss of valuable assets, such as productive land for many refugees and internally displaced. This structural vulnerability is exacerbated by the cumulative effect of successive adverse shocks like droughts, high food prices and the present political crisis. Inequalities measured by the Gini index increased at national level (from 43.5 percent in 2006 to 46.3 percent in 2012) and in rural areas, confirming that economic growth was insufficient to pull people out of poverty, thus the need to be complemented by targeted interventions. 4. While access to health and education had improved, malnutrition remains a serious threat to human development and is on the rise again. In 2010 (Demographic and Health Survey), 58 percent of children between six and 59 months of age were reported stunted (low height-for-age, an indicator of chronic malnutrition) while in 2014, the estimated rate remained 49 1 Burundi – Fragility Assessment Note – March 2016 2 67.1 percent in 2006 and 64.9 percent in 2014 3 Burundi is divided into 18 provinces, 129 communes and 2,638 collines, which are equivalent to large villages (with an average of 535 households). 4 In October 2016, there were over 315,000 Burundian refugees mostly in Tanzania and Rwanda (UNHCR, http://data.unhcr.org/burundi/regional.php) 1 percent (WFP). According to the latest available data, in 2013 (IFPRI), Burundi was the most food insecure country in the world. In 2010, stunting levels reached 71 percent in some northern districts in Ngozi and rates above 62 percent in the whole northern region. Chronic malnutrition among children reflects the overall extreme food insecurity in Burundi linked to the small landholdings, declining soil fertility and plant diseases as well as climate and man-made shocks. It is also associated with micronutrient deficiencies such as anemia. While the country has developed successful protocols for severe acute malnutrition, it has advanced less in terms of addressing chronic malnutrition through nutrition education, community-based prevention and treatment and related issues on early childhood development and parenting practices. B. Sectoral and Institutional Context 5. Burundi ranks 184th out of 188 countries on the 2014 Human Development Index. It had registered progress on some key Millennium Development Goals such as MDG2 on universal access to primary schooling and MDG3 on gender parity in primary education (Table 1). While the proportion of women in parliament has also more than tripled from 12.3 percent in 1993 to 36.4 percent in 2015, placing the country well in the sub-region, cultural norms still limit the full recognition of women’s contributions to the economy, especially in rural areas. In particular, women cannot inherit land. Progress on child and maternal mortality (from 954 to 712 per 100,000 live births between 2000 and 2015) was significant albeit slower than needed to reach MDGs 4 and 5. The progress reflected the results of the free provision of key health services from 2005 and performance-based financing from 2010. However, further progress on these indicators may have stalled since the beginning of the present crisis and depends on the improvements on other dimensions of the MDGs such as access to drinking water and sanitation, which lagged behind. Overall, the country still faces challenges to reach the Sustainable Development Goals in 2030. Table 1.1: Achievement of Millennium Development Goals in Burundi Main MDGs Actual performance Targets At a for Glance 2015 1990 2015 MDGs 1. Eradicate extreme poverty and hunger Poverty incidence (%) 35.0 64.6 (2014) 17.5 Not met Prevalence of child malnutrition (% wasting) 30.2 26.9 (2012) 15.1 2. Provide primary education for all Gross primary enrollment rate (%) 52.3 95.4 (2014) 100 Met Retention rate at the entrance of the 5th grade (%) Nearly 54 76 (2014) 100 met 3. Promote gender equality and empower women Ratio of girls to boys in primary education 0.78 0.99 1 4. Reduce child mortality Child mortality (per 1,000 live births) 142 Partially 183 61 met (94 IGME) 5. Improve maternal health Maternal mortality (per 100,000 live births) 714 (IGME 2014) 1100 275 (500 DHS 2010) 6. Combat HIV/AIDS, malaria, and other diseases HIV/AIDS infection rate (%) 2.1 1.4 (DHS 2010) 1.0 7. Ensure environmental sustainability Share of population with access to improved sanitation (%) 42 48.4 71 Source: Burundi, “Rapport sur les objectifs du millénaire pour le développement »2015 2 Poverty, Food insecurity and Malnutrition 6. An estimated 60 percent of Burundians were living below the food consumption poverty line prior to the current crisis (Vulnerability assessment, 2014, based on the PMS data from 2012-13) 5 and an estimated 40 percent faced extreme deprivation. Large families with a high dependency index (especially single parent families) are less likely to satisfy their basic food needs, particularly if the head of household works in agriculture. Regionally, the highest levels of deprivation are found in the North (despite improvements in Ngozi and Kirundo) and in the Center East (especially in Mwaro and Cankuso). In both regions, multiple deprivations affect up to 75 percent of households. On the other hand, food deprivation is lower in urban areas (48 percent) and in the capital Bujumbura (41 percent). 7. Burundi was the most food insecure country in the world in 2013 and food insecurity is increasing again as a result of the current political and climate crisis. The majority of the poor (97 percent) reside in rural areas, depend on rain-fed low-input agriculture on very small farms (87 percent of poor households cultivate less than ½ ha), and are highly vulnerable to climate shocks. Demographic pressure with decreasing soil fertility due to overexploitation, prevalent plant diseases (affecting bananas, corn and cassava, which are staples of rural diets) are all structural determinants of the widespread food insecurity. In addition, short-term shocks such as climate events (drought, floods, el Niño), increases in food prices (notably in 2015), and political crises have long-term impacts on households whose livelihood is so fragile. As a result, the area in the Northern and Eastern depressions of the country in the Integrated Food Security Phase Classification (IPC) Phase 3, which corresponds to an acute food security and livelihood crisis has more than doubled 6 and the rest of the country is in Phase 2 “under pressure” at any given moment. The Emergency Food Security Assessment of 2016 lists Bujumbura, Bubanza, Citiboke, Kirundo, Makamba, Rumonge, and Ruyigi in critical situation and up to 1.1 million individuals needing humanitarian assistance. Health and Education 8. Access to health services had greatly improved with the removal of fees for priority health services (for children under five, pregnant women and the treatment of illnesses such as malaria, HIV/AIDS and tuberculosis) since 2005. Substantial increases in investment in health infrastructure have improved physical access (85.8 percent of the population has access to a health center at less than five km distance, EBCM 2014). Performance-based financing (PBF) in the health sector, which started as a pilot in 2006 and became a national program in 2010 has also contributed to these improvements. However, quality of care, availability of specific interventions for mother and child health, especially emergency situations, and a functioning health management information system still lag behind. Medication shortages and informal payments are on the rise since the beginning of the crisis, potentially threatening these gains. While access to modern contraception has increased from four percent in 1990 to 22 percent in 2014, the fertility rate remains high at 5.9 children per woman in 2014 (Burkina Faso was at 5.5 and Rwanda at 3.9 in 5 The consumption poverty line is assessed through a Cost of Basic Needs approach: food for 2,200 kcal/day/adult equivalent and a non-food allowance based on the average non-food consumption of households whose total consumption is close to the food cutoff (Vulnerability Assessment, 2014). 6 The IPC classification uses a 5-point scale (minimal, stressed, crisis, emergency, famine) based on food consumption levels, livelihoods change, nutritional status, and mortality. http://www.fao.org/fileadmin/user_upload/emergencies/docs/FAOBurundi_sitrep_5May2016.pdf 3 the same year, World Bank Gender Data Portal). Demand for contraception is growing and unsatisfied demand has risen from 23 to 31 percent between 1990 and 2011 (DHS). 9. Access to and gender parity in primary education have also improved with the removal of primary education fees. Despite some regional disparities, school infrastructure and staffing have improved substantially. However, repetition is high (38.4 percent); completion remains a challenge (37.1 percent of boys and 29.7 percent of girls dropped out before they completed primary school in 2014); and more than 20 percent of children still never go to school. Social Safety Net Programs 10. Burundi faces both a humanitarian and emergency context in some areas in addition to dealing with the structural vulnerability of its poorest members, and safety nets interventions are needed to address both tracks. Donors still fund and implement the majority of safety net interventions in the country (approximately 80 percent) although their funding is decreasing (from 4.3 percent of GDP in 2010-11 to 3.6 percent in 2013 7). Support to public works programs represented 26 percent of this expenditure in 2013. Humanitarian assistance for the return and reinstallation of refugees and supplements for the treatment of acute malnutrition each accounted for 16 percent of the total. The other areas of concentration included agricultural inputs and livestock distribution (12 percent), support for demand-side measures on education such as school feeding, provision of school kits and support to vulnerable children (9 percent). 11. Government funding for safety nets is limited. In 2013, social assistance spending was increasing and amounted to 0.97 percent of GDP but that was reversed in 2015 as social spending was re-directed to security and military expenditures. Tertiary scholarships represented one third of this expenditure followed by agricultural inputs and livestock distribution (19.5 percent in 2013). Humanitarian assistance (8.9 percent of total social assistance spending in 2013) covered mainly the repatriation and reinstallation of returning refugees and relief items for communities affected by natural disasters. Medical assistance (8.4 percent of social assistance expenditure in 2013) consisted of two items: subsidies for the medical assistance card (CAM), through the Ministry of Health, and the medical assistance to settle the hospital and pharmacy bills of destitute patients, provided by the Ministry of Human Rights, Social Affairs and Gender. Government funding of school feeding (7.3 percent) started in 2011, still accounts for a small part of overall expenditure on school feeding and is decreasing. 12. As a whole, the safety net programs do not focus on reducing structural poverty or improving the resilience of vulnerable households, and do not constitute an efficient social safety net. Most of the interventions are short-term, dispersed, and small-scale. For example, school feeding covers only 15 percent of children in primary school. Interventions use different targeting methods with little monitoring and evaluation of impacts, which makes it difficult to assess their contribution to poverty reduction and their impacts on human development outcomes. 7 Social Safety Net Assessment Update, 2015 4 Box 1: The Terintambwe cash transfer pilot (Concern) The Terintambwe project selected the most deprived collines within the selected communes, through a social and wealth ranking, and then used community-based targeting methods and household visits to identify eligible households (among the 10 percent of the poorest with labor capacity). Participants received BIF 25,000 (approximately US$ 16.00 per month at the time) over 11 months. As part of the graduation strategy, at the end of the 11 months, they participate in credit and saving groups, literacy and skills training, and receive asset transfers of BIF 130-150,000 (US$ 80-100.00) to start income- generating activities. In addition, the program facilitates access to birth certificates, pays for the CAM health insurance premiums for participants, supplies school kits to school age children and offers the services of “case managers” who visit the beneficiary families once a week and discuss issues related to HIV/AIDS prevention, domestic violence, income-generation activities, hygiene and nutrition, and the importance of schooling. Approximately five percent of Terintambwe beneficiaries were Batwa households, which were explicitly included in the social ranking. They participated in the same activities as other beneficiaries and the program staff noted increased community participation of these households. Results from the mid-term evaluation of June 2014, show improvements in the quality of housing (24 pp. in use of iron sheets for roofs), lighting (63 pp. in use of battery power lamps and a comparable decrease of firewood), and sanitation (40 pp. increase in installing slabs in latrines). Participant household were able to buy durable goods (kitchen and bedding), farming tools, and small animals (goats and chickens). Participants also gained access to land (medium plot size from 300 to 700-800 m2) through purchases and rentals. Participants also reported increased capacity to see a formal health provider when sick (from 57 to 87 percent) and to secure the attendance of their children in school (10 pp. increase). Food security also increased with the proportion of adults eating two meals a day rising from 19 to 83 percent. Source: Devereux, S., K Roelen, R. Sabates, I. Ssenkubuge, D. Stoelinga., 2014. Concern Burundi. Graduation program midline report. 13. Burundi has little experience with cash transfer programs but some NGO and donor- run pilots are showing promising results. The World Food Program (WFP) is experimenting with cash-for-work in some refugee camps with plans to scale-up cash-based humanitarian assistance. The Belgian Technical Cooperation also piloted a cash-for-work program in poor neighborhoods of Bujumbura and Kirundo with a strong technical and life skills training component. This intensive pilot, which uses a “chantier-école” approach, shows promising results in terms of labor force participation once the nine months of training/work is completed. With funding from Irish Aid, since mid-2012, Concern has also been running a pilot cash transfer program, using the Bangladesh Rehabilitation Assistance Committee (BRAC) graduation approach. This pilot covered 2,000 households in three communes of Citiboke (the province with the highest rate of landlessness) and two communes of Kirundo (the province with the highest population density also resulting in severe land shortage) (Box 1). Even if the intensity of these interventions may not be replicable on a large scale, they provide an important basis of experience for future cash transfer operations. 14. Burundi needs both humanitarian responses and an effective system of medium- and long-term programs built around national objectives. The country has set up some key institutional elements to put in place these programs. The national objectives were set in the Cadre Stratégique de Lutte contre la Pauvreté (PRSP) and the National Social Protection Policy (PNPS) (2011). An inter-ministerial coordinating body, the National Social Protection Commission, was established in April 2013. A secretariat (SEP/CNPS) to support the 5 implementation of the PNPS is operational since early 2014 and set up provincial structures in 2015 and thematic groups in 2016. With support from the World Bank and UNICEF, through a participatory process, the SEP/CNPS has developed a National Social Protection Strategy that delineates the Government’s priorities (Box 2). Box 2: The National Social Protection Strategy (PNPS), 2015 Three key strategic and one cross-cutting objectives: • Increase access to basic social services such as health, water and sanitation, and education • Ensure food and basic income security both for those that can exit extreme poverty sustainably and for those that will remain vulnerable their whole life • Strengthen natural and social risks management: social protection’s role is to strengthen the resilience of vulnerable groups i.e. their capacity to better manage the risks they face, without resorting to damaging adaptation strategies • Cross-cutting objective: Contribute to decreasing young children’s chronic malnutrition. Three key principles for the implementation of the objectives: • Social protection covers everybody, with a particular focus on the poorest and most vulnerable. Strategic priorities apply to the entire population of Burundi without distinction. However, given the country’s situation, social protection programs need to give priority to the neediest and most vulnerable population groups. One of the first activities will be to develop a targeting system. • Social protection is a multi-sectoral agenda which requires wide partnerships. The SP sector will seek to contribute to the strategic objectives through two types of activities: (1) the definition, planning and implementation of key programs, under the responsibility of the CNPS and its committees, (2) the coordination with complementary programs in other sectors to maximize impacts on the target groups. • The strategy needs implementing structures. Based on the country’s commitments in terms of SP, the strategy emphasizes the need to strengthen implementation structures in a step-wise affordable approach. 15. The proposed project seeks to support the Government in protecting the most vulnerable while setting up the delivery mechanisms for a basic safety net addressing structural vulnerabilities and promoting livelihoods. The project would support the foundations of a targeted cash-based safety net program that could contribute to poverty alleviation in the short term and investments in human and productive capital for poverty reduction in the long-run, a twin-track contribution to SDG 1. The cash transfer program would also provide an anchor for an effective safety net system to help strengthen the resilience of the poor in rural areas. C. Higher Level Objectives to which the Project Contributes 16. The proposed project contributes to Pillars II and III of the World Bank Group’s Burundi Country Assistance Strategy for FY 13-16 (Report 72334-BI). It contributes to Pillar II (Resilience) by setting-up a cash transfer program targeted to the poorest and stimulating investments in their human capital and Pillar III (Governance) by setting-up the mechanisms for delivering cash transfers effectively, such as a transparent database of potential social assistance beneficiaries, a transparent and reliable payment system, and a basic monitoring and evaluation system. The proposed project is developed in close collaboration with the poverty and health teams on targeting and nutrition services respectively. For targeting, it uses analysis from the poverty assessment and poverty maps based on the recent 2013/14 household survey data. The 6 project will also make a direct contribution to the World Bank Group’s twin goals of reducing extreme poverty and boosting shared prosperity by providing transfers to the poor. 17. The project contributes to the second and fourth goals of the Government’s Second Poverty Reduction Strategy Paper (PRSP II), from 2012. It contributes to the second goal of making growth more inclusive and sensitive to vulnerable groups and the fourth goal of developing institutions to strengthen governance and the quality of services. It is aligned with the pillar on access to quality services and national solidarity. The PRSP II also targets social protection with short term actions (through community-based interventions and livelihoods promotion in rural and informal environments) and longer term initiatives (formal social insurance and social protection institutions and access). 18. Development partners working in Burundi are supporting the proposed operation. The Bank has been working closely with UNICEF on the analytical underpinnings of the project, namely the Social Safety Net Assessment (2014) and the National Social Protection Strategy (2015). UNICEF will provide technical assistance for curriculum development of the nutrition/ECD promotion activities and will possibly be responsible for part of the implementation of these promotion activities – including additional ones on peace-building and other topics -- and ensure quality control. The African Development Bank has supported complementary key technical assistance on social insurance in the then Ministry of Civil Service, Labor and Social Security. The design has benefited from the experience of the Belgian Technical Cooperation and Concern cash transfer pilots and from the experience of the International Fund for Agricultural Development (IFAD) in community targeting. II. PROJECT DEVELOPMENT OBJECTIVES A. PDO 19. The development objectives of the proposed Project are to provide regular cash transfers to extreme poor and vulnerable households with children in selected areas while strengthening the delivery mechanisms for the development of a basic social safety net system. Vulnerable households are defined as food insecure households. 20. The key delivery mechanisms for a basic social safety net system will include a targeting process with a beneficiary database, a management information system with a payment module, and a basic monitoring and evaluation system. Project Beneficiaries 21. The direct project beneficiaries are the households registered in the beneficiary database, which receive the cash transfers. The direct beneficiaries within the household are: (i) women – as cash transfer recipients and targets of some of the behavior change activities; and (ii) children under 12 years of age – as targets of the behavior change activities and human capital investments. Other direct beneficiaries also include the two implementing agencies – the SEP/CNPS and the Social Assistance and National Solidarity Directorate of the MDPHASG – through capacity-strengthening activities. Indirect beneficiaries include: households living in the collines where the cash transfer program and its complementary interventions operate through the colline-level behavior change triggered by promotion activities and the impacts of the cash 7 transfers on the local economy. PDO Level Results Indicators 22. The indicators that will be used to monitor progress in reaching the project development objective are: • Number of beneficiaries of safety nets programs (households), of which: (i) number of female beneficiaries; and (ii) number of beneficiaries of unconditional cash transfers (SP core indicator) • Proportion of households enrolled in the beneficiary database living below the extreme poverty line (%) III. PROJECT DESCRIPTION A. Project Components 23. To reach its proposed development objectives, the project is structured around three components. Component 1 would support the development and implementation of a social safety net program based on cash transfers, the Merankabandi 8 program, Component 2 would support the development of the key delivery mechanisms for a basic social safety net system, and Component 3 would support the implementation of the project. Component 1 – Support to the cash transfer program (US$26.7 million equivalent) 24. This component will support the development and implementation of the Merankabandi cash transfer program to be implemented in phases. This would be the precursor to a national program, which would aim to reach all the poorest (first quintile) food insecure households across Burundi (estimated at about 300,000 households, or 1.9 million individuals 9). The proposed project would support the first phase of the program in reaching the extreme poor and vulnerable rural households with children under 12 years of age in the Northern and Eastern regions. The social safety net program would be built around two core elements: (1) regular cash payments, and (2) promotion of foundational human capital, broadly understood as, and limited to, the set of information, communication and education measures designed to encourage the adoption of positive behaviors towards strengthened nutrition, early childhood development, and school attendance. 25. The Merankabandi program will target extreme poor and vulnerable (in terms of food insecurity) rural households with children under 12 years of age in selected areas. The targeting process includes first a community-based targeting followed by a basic proxy-means test (see component 2). The final list of eligible households for the cash transfer program will be shared with the community, for validation. 26. The Merankabandi cash transfer program focuses on immediate alleviation of poverty and prevention of chronic malnutrition. This seeks to mitigate the potential long-term effects of short-term crises and shocks on human development and sets-up a medium-term vision for household development. A beneficiary household would remain in the program for three years 8 Merankabandi can be roughly translated as “Be equal to others” 9 Average households in the first quintile were comprised of 6.4 members (ECVMB 2013/14) 8 (two and a half years with cash and behavior change activities and six months with behavior change activities only). 27. Beneficiary households will receive BIF 20,000 per household per month (approx.US$ 145 per year). The national basic needs poverty line is at BIF 41,054 per adult per month and the food poverty line is at BIF 28,201 per month 10. The transfer corresponds to 18 percent of the household food poverty line (which is consistent with the experience of comparable African countries). Subcomponent 1.1: Cash transfers (US$19.5 million equivalent) 28. The objective of the cash transfers is to provide a regular income to extreme poor and vulnerable/food insecure households so as to enable them to increase their consumption and invest in the foundations of their children’s human capital. The regular cash payments are the first core element of the program intended to help beneficiary households meet their basic needs. The size and regularity of the transfers are important components of potential second-round economic impacts of the transfer through activation of the local economy and productive investments. The second core element is the behavior change communication to promote investments in children’s human capital through improved nutrition, early child care and education practices, and increased access to basic education. 29. The proposed Project will support the payment of cash transfers to beneficiary households, and associated financial fees as well as support for beneficiary households to acquire national IDs. For operational reasons, the payments to beneficiaries would initially be made every three months 11. Once systems are well established and reliable, payments would be made monthly so as to enhance the consumption-smoothing function of the transfers. The proposed initial quarterly payment schedule takes into account periods of high risk (soudure), significant expenditures (beginning of school year) and operational constraints (rain) in the definition of the payment calendar. 30. Given the novelty of cash transfers in Burundi and potential logistical issues, the program may use more than one payment method. The priority will be to ensure timeliness, accuracy (in terms of amounts and recipients) and low costs, both for the program and for households (in terms of accessibility (literacy), distance and time to reach, waiting, etc.). Results from a payment study undertaken in March 2016 in the four provinces and experience from the Concern and WFP pilots with cash transfers point to the potential of mobile payments, possibly in combination with financial institutions such as banks and the post office where the mobile ecosystem is limited (due to limited network and electricity coverage, or payment correspondents). Subcomponent 1.2: Behavior change for investments in human capital (US$7.2 million equivalent) 31. The second core element that the proposed Project will support is the promotion of 10 World Bank Poverty Assessment (forthcoming 2016) 11 The Concern pilot gave cell phones to all beneficiaries as a way to facilitate communication with beneficiary households, help households’ standing in the community, and as a payment mechanism (when it became possible). Based on this experience and the result of the payment system assessment, the first payment will consist of BIF 40,000 and a cell phone. 9 household foundational human capital investments, in particular in women and children. This includes information, education and communication (IEC) to: (1) explain program processes and goals to beneficiary households (one module), (2) encourage the adoption of positive behavioral changes towards maternal and child health, nutrition and early childhood development among households in the program’s intervention areas (three modules), and (3) encourage the participation and retention of school age children in school and provide basic financial literacy information (two modules). These sessions will continue for six months after the cash transfers end. 32. Initially, the program would track participation of beneficiary households in behavior change complementary activities (based on their composition). As administrative systems strengthen and services become more available, the payments could become conditional on use of basic health, nutrition and/or education services to strengthen the incentives of households to invest in the foundations of human capital. 33. This sub-component will support the elaboration of the methodology and curriculum of the IEC materials and the implementation and supervision of the promotion activities. Three modules of the IEC package will seek to promote better health and nutrition among children as well as pregnant and lactating women and better early childhood development outcomes for improved school preparedness. As contribution to the project, UNICEF would develop the nutrition and early childhood development package. They may also develop the lesson plans over the 36 months and provide additional assistance for the implementation, including quality control or adding modules on other topics. 34. The promotion activities will be implemented by local non-governmental organizations already involved in these issues. The Commune Family Development Centers (CFDC) from the Ministry of Human Rights, Social Affairs and Gender will participate in the supervision of the delivery of these activities, along with the program commune focal point. The Ministries of Health and Education will provide technical guidance as needed and ensure that the promotional activities encourage the use of health and education services. Component 2: Support to the key delivery mechanisms of a basic social safety net system (US$7.5 million equivalent) 35. The second component will support the development and implementation of four key instruments: (i) a targeting mechanism and potential beneficiary database; (ii) core modules of a basic management information system; (iii) monitoring and evaluation modules; and (iv) capacity building for the implementation of the PNPS. The instruments will be anchored around the cash transfer program described in Component 1, but with a view to eventually serve a broader set of targeted programs that would contribute to the realization of the PNPS objectives. Sub-component 2.1: Implementation of a precursor database for the registry (US$2.0 million equivalent) 36. Among these instruments, the beneficiary database and its associated targeting mechanism are a key pillar for the coordination of programs. They are also critical for ensuring transparency in the selection of beneficiaries in the Burundian context of recurrent fragility and entrenched structural poverty. Initially, the database will be developed to identify the 10 extreme poor for the purpose of the cash transfer program and its complementary activities. However, its ultimate function is to serve as the basis for a registry of extreme poor households for targeted key programs of the NSPS, and for the targeted complementary interventions of CNPS member ministries. 37. The development of the database will take place in phases, according to the availability of poverty information, and the implementation of the cash transfer program. Initially, for the purpose of identifying potential beneficiaries of the cash transfer program, four provinces were selected on the basis of their monetary poverty rate (ECBM 2013/14) and chronic malnutrition rates (DHS 2010) and included Gitega, Karuzi, Kirundo and Ruyigi. The selection at communes, collines, and household-level is described in Box 3. 38. As part of the registration process, the project will assess whether potential transfer recipients in beneficiary households have national ID cards 12. ID cards are required for mobile phone registration so if transfer recipients do not have an ID, the project will provide them with support to obtain one. The project may issue them temporary program cards for the first payment and the initial complementary activities’ sessions. If necessary, the Project will coordinate with the Ministry of Interior for the provision of national identity cards. 39. This sub-component will also support the background analysis for the implementation of the database and the targeting process – development of poverty and malnutrition maps, community-based targeting criteria and processes, registry questionnaire, proxy-means test analysis; the organization of and support to the different committees involved in the targeting and registration, the implementation of the PMT survey, storage and analysis of the data as well as the preparation of beneficiary lists for the cash transfer program. It will also support the development of the corresponding database and management information systems related to the registry and the targeting (hardware, software, back-up equipment). Finally, it will support potential additional activities to ensure that transfer recipients are aware of the requirements and supporting documents to obtain national ID cards. 12 Based on the experience of the Terintanbwe pilot, it is expected that up to 30 percent of expected transfer recipients may not have an ID. 11 Box 3: The selection at communes, collines, and household-level Based on the poverty map, the registry will start in the four communes with the highest estimated rural poverty rate in each province (Gitega: Bugendana, Buraza, Gitega, and Itaba; Karusi: Bugenyuzi, Gihogazi, Mutumba, and Nyabikere; Kirundo: Bugabira, Busoni, Kirundo, and Ntega; and Ruyigi: Butagwanza, Butezi, Bweru, and Gisuru). The 16 selected communes are sub-divided in 2 to 5 zones and include on average 26 collines (between 11 and 43) with an average of 535 households (between 130 and 1,400) per colline. Given the lack of information on poverty at the colline level and the limited resources, the program will randomly select half of the collines during a public event that would include the province governor, commune administrator and all colline chiefs as well as representatives from civil society (including Batwa representatives). The household-level selection for the registry will use community-based targeting with a listing of the 30 percent poorest households in the colline based on nationally agreed criteria for extreme poverty including access to land, a dependency index (taking into account all not able-bodied members: children, elderly, persons with a disability and chronically ill members), and welfare indicators, including food security. The composition of the committee will ensure a balance between government and civil society representation and the committee will receive technical assistance from an NGO to undertake the listing process. The list will be validated by the community. All households in the listing will be surveyed to compute a basic multi-dimensional proxy-means test (PMT) score and ranked by their poverty status. As information and resources become available, the process may be reviewed. Based on the progress on coordinating other information and additional financing for the program, at mid-term review, a decision will be taken to either continue with the initial process, apply the questionnaire to all potentially eligible households, based on the small-area estimate of extreme poverty in the commune (and the community validation may then occur afterwards), or use only the community-based targeting. The targeting questionnaire will collect basic socio-economic information about the household, as well as key information on specific vulnerabilities, access to basic social services, and livelihood strategies. This information could then be used for targeting by other programs managed by other Ministries which are members of the CNPS. The analysis of the data will provide a dynamic portrait of the extreme poor, their access to services and their livelihood strategies, which can help inform the design of poverty reduction programs. All individuals in the registry will be uniquely identified. The proposed methodology will initially be implemented in the 8 selected rural communes of the Karuzi and Ruyigi provinces and then in the additional 8 communes in Gitega and Kirundo provinces. This phased implementation will enable the implementation team to: (i) assess the operational processes for the implementation of the targeting and registration processes, in terms of organizational capacity, time and costs to inform the subsequent roll-out; and (ii) test the targeting efficiency of the proposed CBT and PMT combination, given the prevalence of poverty and the lack of reliable consumption data. Lessons from the first eight communes will inform the roll-out in the next eight communes. Sub-component 2.2. Core modules of the Management Information System (US$1.5 million equivalent) 40. This sub-component will support the development of basic core modules of the Management Information System to support the delivery mechanisms of a basic social safety net system. These modules will be developed for the cash transfer program and will use a unique individual identification number for each beneficiary. They will include key social program 12 indicators for the monitoring and tracking system of the SEP/CNPS. The use of these modules is expected to improve program design and implementation, foster future coordination between programs, and the monitoring and evaluation of the cash transfer program as a template for other social safety net interventions. 41. Building on the database of eligible households, these modules will include: (i) program beneficiary lists, and eventually a registration of complementary activities; (ii) payment modules (limited to the payroll and the reconciliation from the payment provider(s)); (iii) operational tracking of programs; and (iv) basic monitoring and evaluation, including beneficiary feedback and grievance redress system. The payment system will include the quarterly/monthly payroll based on beneficiary lists, the amounts transferred to the payment agency(ies), the beneficiary receipts and the reconciliation of accounts. The operational tracking module would provide an operational dashboard to enable program managers to plan and track activities, human and material resources and other inputs at the central, provincial and communal levels. The M&E system would track financial outlays, key program results (including those core indicators that would become common across programs), impacts and beneficiary feedback as inputs to guide program management in the implementation of the programs. 42. The focus on citizens’ engagement includes a robust grievance redress mechanism (GRM). The grievance redress mechanism would track grievances linked to targeting, receipt of transfers and implementation of the complementary activities. It will use several mechanisms: in- person complaints to program commune focal point, SMS-based system to a third-party grievance manager (conditional on finding a trusted and competent agent), boxes at the colline-level in the care of a civil society organization, possibly a hotline at central-level. Complaints received through SMS, phone or boxes will be logged in the MIS. A results indicator to track the progress of the implementation of the GRM system has been included in the results framework. Subcomponent 2.3: Monitoring and evaluation (US$2.8 million equivalent) 43. Since the project is supporting new interventions and processes in Burundi, and in order to ensure transparency, the third sub-component will support process evaluations of the key program processes and an impact evaluation including beneficiary surveys. The process evaluations will focus on the core operational processes: targeting, payment, delivery of complementary activities. The process evaluation in the first phase communes will inform the expansion in the second phase but also provide key input in the design of the operating processes. The process evaluations will continue in the second phase to provide real-time information about scale-up and implementation in different provinces. The process evaluation will be complemented by regular beneficiary surveys to help map out implementation successes and issues, externalities, and community dynamics and contribute to the establishment of the grievance redress mechanism. 44. The impact evaluation will focus on key poverty, welfare, and human development indicators at the household and community-levels for the cash transfers. The random selection at the colline-level will support a randomized control trial design based on a sample of participating and non-participating collines. Baseline data will be collected prior to the first transfer. A mid-term data collection will take place at 24 months. To evaluate the sustainability of impacts, the end-line data collection will take place six months after the end of the program activities (at 42 months). 13 Sub-component 2.4. Capacity building and analysis for the Implementation of the National Social Protection Strategy (US$1.2 million equivalent) 45. This sub-component will support capacity building for the main actors in charge of social protection in the central government and for the local implementers of the program. The capacity building will seek to bring a common understanding about social protection across government levels (center, provinces, communes, and collines) and sectors. The proposed project will support training on key issues about social safety nets implementation, cash transfers as safety nets, malnutrition, and early childhood development. Since the project will support new operational processes, it will provide capacity building to ensure the adoption, use, and maintenance of these processes and the corresponding tools at all levels of implementation. 46. At the central level, the project will seek to reinforce existing or planned coordination mechanisms, focusing on the CNPS Technical Committee -- and its thematic groups, especially the one on social assistance -- and on the Permanent Executive Secretariat (SEP), in their capacity of coordinators and regulators for social protection. Based on the needs identified in the forthcoming functional institutional assessment, training and technical support will be provided, as well as assistance with coordination costs. 47. This sub-component will also support some basic analytical services to build the evidence base which the Technical Committee and the SEP can use for monitoring of the overall SP system and the discussion on programs design. This will likely include analyses of basic service supply across different types of providers, institutional assessments of the main agencies, background studies for the policy on public works, health coverage for the poor, malnutrition dynamics and coordination with the community health activities supported in the Health project, and additional analysis as requested. Component 3: Project Management (US$5.8 million equivalent) 48. The third component will provide for a Project Implementation Unit, under the authority of the Ministry of Human Rights, Social Affairs and Gender (MDPHASG) and coordinating with the SEP/CNPS. This unit will be headed by a Project Coordinator and staffed by a small management team of experienced technicians. The unit will be responsible for the technical and financial implementation of project activities, including procurement, financial management, M&E, mitigation of potential negative social and environmental impacts, and communication about project implementation and results. The unit will liaise with the SEP/CNPS and the Directorate for Social Assistance and National Solidarity in the MDPHSAG at the central level. The unit will be represented at the provincial and commune-level with staff dedicated to the local implementation of the project in the selected provinces and communes in coordination with provincial and communal deconcentrated structures of the SEP/CNPS and the MDPHSAG. In Kirundo communes, the local implementation could involve Concern, building on their experience with the Terintwambe pilot. The implementation arrangements will be examined at the mid-term review to assess alternative arrangements, depending on the readiness of the SEP and MDPHASG and other partners. 14 B. Project Financing Project Cost and Financing 49. The proposed project has a total cost of US$ 40 million, to be financed with IDA resources. Project Components Project cost IDA Financing % Financing 1. Support to a social safety net 26.7 26.7 100 program 1.1. Cash transfers 19.5 19.5 1.2. Complementary activities 7.2 7.2 2. Support to the key delivery 7.5 7.5 100 mechanisms of a basic safety net system 2.1.Implementation of a precursor 2.0 2.0 database for the registry 2.2. Core modules of the Management 1.5 1.5 Information System 2.3. M&E 2.8 2.8 2.4. Capacity building for the Implementation of the Social Protection 1.2 1.2 Strategy 3. Project management 5.8 5.8 100 Total Costs 40.00 40.00 100 Total Financing Required 40.00 40.00 100 C. Lessons Learned and Reflected in the Project Design 50. The project's design is the result of extensive economic and sector work, technical assistance, policy dialogue and analysis of existing experiences in the Burundian context. The Government of Burundi, with the support of the World Bank, UNICEF and other development partners has engaged in an intense dialogue on social protection and safety nets since 2011. This dialogue has resulted in the validation of the National Social Protection Policy in 2011, the institutionalization of a National Social Protection Commission in 2013 and its permanent executive secretariat in 2014 (SEP/CNPS), and ultimately the validation of the National Social Protection Strategy in early 2015, which the SEP/CNPS is mandated to oversee. The Bank has contributed to this process since 2013 with technical assistance resulting in the delivery of a Safety Nets Assessment and a Vulnerability Assessment in 2014 and support to the development of the NSPS in 2015. 51. One of the challenges the safety net assessment identified is increasing the coverage and efficiency of existing programs both for short-term humanitarian responses and interventions addressing long-term structural vulnerability. As a whole, the existing safety net programs do not focus on reducing structural poverty or improving the resilience of vulnerable 15 households, and do not constitute an efficient social safety net. Most of the interventions are short- term, dispersed, and small-scale. They use different targeting methods with little monitoring and evaluation of impacts, which makes it difficult to assess their contribution to poverty reduction and their impacts on human development outcomes. 52. Operationally, the proposed project also builds on the experience created in Burundi through pilot cash transfer projects implemented by Concern with support from Irish Aid and by WFP in refugee camps, as well as cash-for-work implemented by the Belgian Technical Cooperation in urban areas and IFAD in rural areas. These interventions are all at pilot stages and each of them provided lessons for the design of this program, especially in the fields of targeting and transfer delivery. In terms of the Concern project, the intensity of social work beyond the cash transfers was not scalable within the government structures. The target included households with working capacity, which may exclude some households with a high dependency ratio. The Project also benefits from the experience of the Food-for-Peace project implemented by CRS and various interventions supported by UNICEF with a focus on accompanying measures for behavior change in terms of child nutrition and development. These interventions all use in-kind transfers. 53. Lessons learned from other countries have also informed Project design. Through their participation in the French-speaking group of the Africa Community of Practice on Cash Transfer Programs, the authorities will continue to incorporate lessons from their West African colleagues. International experiences in comparable contexts such as Burkina Faso, Niger, Rwanda and Zambia have also yielded specific insights and guidance on how to design and implement the proposed project. The following are the most important lessons and recommendations incorporated in the current project: (a) Use cash transfers to build a twin-track response to crises: as both a crisis prevention/mitigation intervention and a structural response to chronic poverty. CTs can play an important mitigation role for households facing natural disasters or price crises as demonstrated in Nicaragua. They are also increasingly used in displacement/refugee contexts when markets are available. They enable households to protect the human capital of their members and help avoid using negative risk-coping strategies. (b) Strategies to support extreme-poor households require a combination of long-term, regular transfers and support to promote investment in households’ human capital. Emergency short-term CT programs show immediate positive effects on food security and expenses related to basic needs, but these effects are not sustained much beyond the life of the intervention. The proposed project introduces a medium-term approach to ensure sustainable impact, with transfers for 30 months – potentially covering the “first 1000 days” or the window of opportunity for nutrition. "Spillover" effects are also shown to occur on non-participating households within targeted communities. (c) Providing cash to women ensures that a larger share is spent on food and children’s needs and community-level interventions can help change norms about intra-household decision- making and practices related to children care and nutrition. Providing cash to women increases their agency in spending and may increase their bargaining power. Community- level interventions which foster norm changes can help consolidate positive changes with respect to children’s and women’s welfare and mitigate potential risks for increased domestic violence and conflicts between beneficiary and non-beneficiary households. 16 (d) Good targeting is necessary to provide support to those who are most in need. In an environment of pervasive poverty and fragility such as Burundi, a balance must be struck between targeting accuracy, cost, and complexity. A combination of geographical, categorical and community-based targeting, and the estimation of household poverty using proxy means testing (PMT) will enhance transparency, allow for objective comparisons between households and empower communities to voice their priorities. The use of objective criteria both for the selection of provinces and communes and for the community- based targeting will help build transparency in the process. (e) Partnering with experienced civil society organizations and NGOs can facilitate the implementation of the project and foster capacity building, particularly at the local level. They will be involved in providing technical assistance for targeting, for the delivery of the accompanying behavior change communication measures and may intervene also in social accountability mechanisms, such as the complaints and grievance system, to strengthen project governance. (f) Ensure that the institutional framework provides concrete value to social protection stakeholders. The development of a system of effective social safety nets begins with the definition of a long-term vision and a coherent strategy, but then must bring concrete value added to maintain its momentum. Because social protection is multi-sectoral, instruments that benefit all stakeholders – be they government, donors or civil society organizations – can contribute to operationalizing the institutional framework for social safety nets in Burundi. They include a credible targeting methodology and social registry. (g) Develop a management information system that informs strategic decision-making. As demonstrated in Liberia, Sierra Leone and Tanzania, an MIS should inform the design and implementation of programs, make it possible to demonstrate their impact to political decision-makers, development partners, and civil society, and enhance global knowledge about social safety nets in fragile environments. This is a potentially crucial ingredient to garner more donor support and expand the coverage of the proposed interventions. (h) Build the capacity of public administration. The Project supports the strengthening of the Government's capacity for planning, implementing, monitoring and evaluating social safety nets for all sectors involved in social protection, at the central and local level and helps set-up key staff and incentives to keep the key systems maintained. (i) Build synergies within the World Bank interventions in the country, and propose an effective platform for improved service delivery. During both the design and implementation phases, the proposed Project will develop synergies with the other operations in the country, such as the Projects supporting investments in health, nutrition and capacity strengthening in the public sector as well as explore potential complementarities with sustainable rural development interventions to support income- generation (coffee production and sustainable management of land fertility). (j) Donor Coordination. With a larger and long-term view of the entire safety net system, it is important to promote coordination among partners and to clarify areas of interest and investment - this will strengthen efficiency of resources and avoid unnecessary duplication of efforts. Regular coordination mechanisms can help to learn from ongoing programs, to avoid duplication, and to identify potential areas of collaboration. This also can contribute 17 to the long-term sustainability of the program by strengthening donors' confidence and interest in supporting an effective and well-targeted safety nets program. IV. IMPLEMENTATION A. Institutional and Implementation Arrangements 54. The overall implementation arrangements for the proposed Project include the following actors (Details on the composition and roles and responsibilities of these groups are outlined in Annex 3): (a) A Steering Committee, with high-level representatives of key sectoral and central stakeholders will be responsible for providing strategic guidance for the Project implementation, (b) The Permanent Executive Secretariat of the National Social Protection Commission (SEP/CNPS) is responsible for the technical coordination between the Project Implementation Unit and other government members of the CNPS and for the coordination and monitoring of social safety nets, (c) The Directorate for Social Assistance and National Solidarity in the Ministry of Human Rights, Social Affairs and Gender is responsible for supporting the implementation of project activities at the provincial and communal levels in coordination with the Community Family Development Centers (CDFCs), the provincial representations of the SEP/CNPS, and the Project Implementation Unit. (d) The Project Implementation Unit, under the authority of the Ministry of Human Rights, Social Affairs and Gender is responsible for the management and implementation of the project at the central and local levels. B. Results Monitoring and Evaluation 55. Monitoring and evaluation is a key element of the Project, because it provides the regular assessment of the Project’s performance to the Government, the World Bank and other development partners. In this Project, a core focus of both Component 1 and Component 2 is to build the capacity of the Government of Burundi to monitor its own programs in the Social Protection (SP) sector. The proposed Project includes a series of monitoring and evaluation activities for the cash transfer program. These include: process evaluations; regular spot checks and beneficiary surveys to evaluate the quality of implementation, the efficiency of the targeting and payment processes, and the overall satisfaction with the program; impact evaluation for the cash transfers and parts of the behavior change communication components. The key delivery mechanisms put in place for the CT program, such as the MIS, will eventually enable the implementing institutions to manage and monitor the implementation of their programs. 56. The Project Implementation Unit will organize annual financial audits for the Project, annual reviews of progress, and a mid-term review to guide the Project implementation after the first eight communes. The mid-term review will involve Project’s stakeholders and civil society in the review of performance, intermediary results, institutional arrangements and outcomes. It will confirm the plans and processes for expansion of the CT 18 program in the additional eight communes. C. Sustainability 57. The Government of Burundi has shown its commitment to the implementation of the National Social Protection strategy – by setting-up the SEP/CNPS as the main coordination body for Social Protection, and by providing the resources for its functioning. The proposed Project will support the Government in further financing social protection interventions, especially to establish mechanisms for counter-cyclical funding of social programs, as key pillars of both a crisis response and long-term poverty reduction. At mid-term review, the project will assess the possibility of including counterpart funding at the local level beyond human resources. 58. Development partners have participated in the development of the National Social Protection Strategy and are committed to support its implementation. They are providing technical support to the definition of the targeting methodology and the definition of the Information, Education, Communication (IEC) package. The cash transfer model is very new in Burundi but the technical and policy dialogue since 2014 is opening space to consider cash transfers as a humanitarian and social assistance tool. The impact evaluation will play a key role in marshalling further technical and financial support, especially in planning for a continuum of interventions from humanitarian relief to chronic poverty alleviation. Building solid systems and operational processes for delivery of cash transfers is a key first step to strengthen the capacity to absorb additional resources to the program. V. KEY RISKS A. Overall Risk Rating and Explanation of Key Risks 59. The overall risk of the proposed project is Substantial, because the implementation of the core elements of the Project (targeting, cash payments, and social safety net program), is a new approach in Burundi, in a period of crisis with a thwarted transition from humanitarian assistance to sustained social assistance and with limited capacity for coordination and implementation at the central and local levels. 60. In the political and governance area, risks are high. There is a risk of protracted repeated crises and a fragility trap if the underlying factors of the 2015 political crisis are not addressed. In terms of the project, the political crisis has slowed down preparation and further violence and potential movement of populations would disrupt program operations. The overlapping mandates of some key ministries in the social protection area (Health and Social Affairs on non-contributory health insurance) could cause some delays in the implementation of Component 2 because of competition between the different actors involved. The project will help clarify the division of responsibilities between the SEP/CNPS and the Directorate for Social Assistance and National Solidarity at the central, provincial and communal levels for project implementation. It will also foster better coordination between the Ministries of Health (MSPLS) and Social Affairs (MDPHASG) at the policy-level through the support to the CNPS and at program-level through the coordination on nutrition-related issues. 61. In terms of macroeconomic management, risks are substantial. In 2015, the country entered a macro-fiscal crisis, which is compounded by the reduction of foreign aid --especially budget support--, interruptions in trade both in terms of exports and on the domestic market, budget 19 reconciliation challenges and low reserves (standing at 1.5 months of imports in April 2016). 62. Given the tense political situation and the novelty of implementation of direct cash transfers, the stakeholders’ risks are substantial. The project will require coordination between different levels of government, with clarification about roles and responsibilities. Targeting in a pervasive poverty context such as Burundi may increase tensions at the local level. Providing cash to women may run against established gender norms, in a context of potential high gender-based violence. The project includes a strong component of communication and outreach to mitigate these effects. 63. In terms of technical design, the risk is substantial. The program design reflects the international and Africa experience in setting-up social safety net programs: guaranteeing transparency and fairness in eligibility and beneficiary selection; tracking and ensuring the reliability of payments to the right beneficiaries, at the right time and with the right amount. This is particularly challenging in an environment of low trust in the state and underlying political tensions. To mitigate this risk, the project is using a phased approach, focusing on the key processes and building on existing structures and experiences in the country. 64. Risks in terms of implementation capacity and sustainability are substantial. The SEP and the CNPS are new organizations and they are still in the process of establishing their roles and mandates. The recent re-organization of the MDPHASG has widened the mandate of the National Directorate for Social Assistance and National Solidarity. The proposed contracting between the Government and private operators/NGOs for the implementation of core processes is new for the social assistance sector, even if the health sector has built some experience with the performance- based project. The Project Implementation Unit will ensure the timely implementation of the project activities while building the capacity of the SEP/CNPS and the Directorate for Social Assistance and National Solidarity in the MDPHSAG to take over implementation in subsequent phases of the program. The proposed project focuses heavily on capacity building activities to ensure that capacity is strengthened as implementation unfolds. The proposed Project also focuses on building key delivery mechanisms, which would allow government agencies to continue implementing systems and interventions after the Project ends. In the present budget environment, continued donor support for the program will be necessary in the medium-term and the mid-term review should include a review of potential financing options and implementation partnerships, depending on the evolution of the fragility situation. 65. Fiduciary risks are substantial since both SEP/CNPS and the Direction for Social Assistance and National Solidarity lack experience with the Bank’s procurement rules and financial procedures. Mitigation measures include the implementation through the Project Implementation Unit, the development of the project’s manual of procedures in accordance with the Bank procedures, and the appointment of an internal auditor. VI. APPRAISAL SUMMARY A. Economic and Financial Analysis 66. The proposed project will have both direct and indirect impacts on extreme poverty and welfare. Component 1, which focuses on concrete support to extreme poor and food-insecure households with children, is expected to generate strong direct impacts on these households, and spillover effects on the targeted communities. Component 2, which focuses on building key 20 delivery mechanisms and capacity to operate them, will have broad if mostly indirect impacts. These impacts are summarized here and discussed in more detail in Annex 5. 67. The project will have a strong direct impact on the extreme poverty of its beneficiaries through the direct cash support. It will also have community-level impacts related to the complementary activities and through local economy impacts of the cash transfers. 68. The direct impacts of cash transfer programs on their beneficiaries are two-fold: poverty alleviation in the short-run through income and consumption support and in the longer- term through key investment in the human capital of children. Children who benefit from improved care and nutrition are better prepared for school, better learners, their income-generating capacity improves and their off-spring have better outcomes. (a) Impact on consumption and poverty in the four provinces where the Project would implement cash transfers: with perfect targeting to the first quintile, beneficiary household income would increase by 15 percent over the transfer period and the monetary poverty gap would shrink in half. In Burundi, the Terintambwe evaluation showed a dramatic net increase of 50 percentage points in the proportion of adults eating two meals a day (to 83 percent) and a net increase of 10 percentage points in the proportion of children eating two meals a day (to 61 percent) and 30 percentage points in those eating three meals a day (to 32 percent). (b) Impact on productive investments and coping strategies: Cash transfers can also lead to significant increases in productive investment. In Zambia, they led to increased investments in agricultural inputs, as well as a shift to higher-value agricultural production (from cassava to maize or rice) and more market sales (from 23 to 35 percent) and ownership of non-agricultural enterprise (from 22 to 39 percent). Cash transfers also led to increased small livestock ownership in Zambia, Kenya and Malawi. In Burundi, the evaluation of the Terintambwe pilot 13 implemented by Concern showed increased ownerships of farming tools, animals such as goats and chicken (from 0.25 to 1.25 in both cases) and access to land (both through purchases and rentals, beneficiary households doubled their access from 300 to 750 m2). Similar programs in Eastern Africa have shown important effects of cash transfers protecting consumption in lean seasons or crises (drought, cash crop price drop). In addition, beneficiary households were less likely to resort to negative coping strategies (distress sales of productive assets, removal of children from school, begging and high-risk income generation strategies). 69. The indirect impacts of cash transfer programs in the areas where they operate are similarly two-fold: sharing prosperity through local economy impacts of the cash (increased demand for goods and services) and in the longer-term through investments in the productive capacity of beneficiary communities and strengthening of their risk management capacity: (a) Impact on local economies. Evidence is accumulating on the positive spillovers of cash transfers on local economic activity. Cash transfers provide an immediate impact by raising the purchasing power of beneficiary households. As beneficiary households spend cash, impacts immediately spread outside beneficiary households to others inside and outside their communities. In Ghana, beneficiaries of the Livelihood Empowerment Against Poverty (LEAP) program spend about 80 percent of the transfers inside the local economy, 13 Concern. 2015. Concern Worldwide Graduation Program in Burundi. Adding to the Evidence. Summary April 2015 21 which in turn launches a new round of income increases in the community. In turn, periodic markets and purchases outside village will shift income effects to other communities. Cash transfers can therefore have large income multiplier effects, provided markets are functioning fluidly. If, on the other hand, production constraints limit supply response, transfers may lead to higher prices and a lower multiplier effect, which has been the case in some areas in Ethiopia. (b) Risk management capacity and social cohesion: in Ghana, Ethiopia and Zambia, cash transfers beneficiaries have reported increased capacity to participate in community solidarity networks. While the community networks have limited impacts to manage widespread shocks, they provide risk-management services for idiosyncratic shocks on food security. In Tanzania, female-headed households reported better satisfaction with services and increased participation in community projects. Community-level activities such as the proposed behavior change interventions and well-functioning committees have been shown to mitigate potential conflicts between beneficiary and non-beneficiary households. The Terintambwe evaluation in Burundi reports increased savings both among participants and non-participants as a result of the introduction of savings and loans cooperatives in the participating collines. In addition, participants increased their participation in school (from 15 to 45 percent) and health (from 9 to 43 percent) management committees as well as to a lesser extent in collines’ disaster risk reduction meetings. This is especially important for traditionally excluded groups such as the Twa. 70. The impacts of supporting key delivery mechanisms for a basic social safety net system include better and possibly higher allocation of resources. Targeting will help increase the poverty reduction impact of the resources allocated. In the medium to long-term, increased use of the beneficiary databases will enable policy-makers and program managers to tailor interventions better to the needs of the poorest and to address supply-side gaps in quantity and quality. The indirect impacts of strengthened coordination would include efficiency and articulation gains from moving from scattered, poorly targeted, dispersed interventions to a basic system of coordinated programs that address the life-cycle risks facing households and their members. In the medium-term, the database created for the cash transfer program could function as a gateway to access other services. 71. Three main arguments justify public sector financing of safety nets in Burundi. First, the high levels of poverty and food insecurity in the country and the increasing inequality justify public intervention on grounds of equity and in order to contribute to the progressive realization of basic human rights such as the constitutional right to live in dignity14. Second, there are major market imperfections or failures in Burundi in the goods, services, credit, insurance, and labor markets facing poor households, which limit their investment in human and productive capital. The provision of cash and of complementary activities can help address some of these imperfections. Third, there are high externalities to fostering investments in human capital which translate into higher GDP and more cohesive societies, two major challenges for the country as it seeks to move out of a potential fragility trap into a sustained development path. 72. In addition, setting-up some of the key delivery mechanisms of a social safety net system will generate significant externalities. Most of the current safety nets are funded and implemented by development partners. A high degree of inefficiency plagues these interventions 14 Art. 33 of the 2005 Republic of Burundi constitution 22 since they all set up short-term ad hoc targeting and relatively weak monitoring and management information systems. This is a strong argument in favor of an intervention that would support improved targeting and coordination of programs and contribute to the initial set-up costs of tools to better anchor safety net interventions around effective needs and capacity. Efficiency gains could help free additional resources for the programs. 73. The proposed project focuses on technical areas where the World Bank has a strong competitive advantage. Building on its involvement with conditional cash transfer programs in Latin America since the early 2000s, the Bank has strongly supported the development of cash transfer programs in Africa in the past ten years. It has accumulated unrivaled expertise on targeting and registry, payment, operations in low-capacity and fragile environments and on system-building in the region. 74. For the past four years, the Bank has also facilitated a Community of Practice on Cash and Conditional Cash Transfers in Africa. The community includes 19 English-speaking countries, including Ethiopia, Tanzania, Uganda and Zambia whose experience is highly relevant to Burundi. In 2013, the Community started a French-speaking group which includes 15 countries, some of which, such as Burkina Faso, Mali, Mauritania and Niger, are also developing programs with complementary interventions in the areas of nutrition. Through the proposed Project, Burundi can take advantage of this network and build on their accumulated knowledge. 75. Social Protection is a new agenda in Burundi and the financial and technical engagement of the World Bank, along with UNICEF as the head of the social protection donor group, provides a strong demonstration effect and a commitment signaling device for the Government. B. Technical 76. The Project design is based on international best practices in cash transfers and targeting, and on Burundian and regional experience with transfer programs. Beneficiaries will be selected among the poorest with a targeting mechanism based on geographic targeting, community targeting and verification of welfare with a proxy-means test. This approach should maximize the accuracy of targeting, and therefore efficiently distribute resources to the poorest households. Furthermore, the setting up of a database that other social safety net programs could use or complement will result in improved targeting for multiple interventions. 77. The measures put in place to accompany households will be designed and implemented in collaboration with the sectoral actors, with a view to develop interventions that effectively promote behavioral changes in the areas of nutrition and early childhood development and stimulate access to related health and education services. This will also ensure consistency with sectoral activities and adaptation to the local context, as discussed in Annex 2. 78. The implementation of the registry and the cash transfer program, and more generally, the development of standardized instruments that other programs can use, will help address some of the issues identified in the 2014 social safety nets assessments through improved targeting, increased coordination, and more efficient payment systems. C. Financial Management 79. A financial management (FM) assessment of the SEP/CNPS and for the Ministry of Human Rights, Social Affairs and Gender was undertaken as part of project appraisal. The 23 objective of this assessment was to determine whether the implementing entities have adequate financial management arrangements to ensure that project funds will be used for purposes intended, in an efficient and economical way. This assessment was carried out in accordance with the Bank Directive: Financial Management Manual For World Bank Investment Project Financing Operations issued February 4, 2015 and effective from March 1, 2015; and the Bank Guidance: Financial Management in World Bank Investment Project Financing Operations Issued and Effective from February 24, 2015. 80. The assessment indicates that there are some weaknesses and risks mainly in the areas of staffing arrangements and capacity, accounting systems, both internal and external auditing, and reporting arrangements. Proposed mitigation measures, as outlined in Annex 3, have been incorporated into the design of the financial management arrangements with the aim of reducing these risks. Mitigation measures include the establishment of a Project Implementation Unit by the project launch; the financial management system of that unit will be supported by the following measures: (i) recruitment of a Financial Management specialist with qualification acceptable to the Bank; (ii) a computerized accounting system in the coordination unit; and (iii) the development of a Project Implementation Manual (PIM), including an administrative and financial component. The financial management residual risk rating for the Project is Substantial. 81. The report based disbursement method will be used by the project. Disbursements will be made quarterly based on a six month cash flow forecast supported by a work plan that is agreed between the PIU and the Bank (IDA) Task Team Leader. Other methods of disbursement that can be used by the project include direct payments; special commitments e.g. letters of credit; and reimbursements to Government for expenditures incurred under the project. An advance will be made into two USD Designated accounts opened at the Burundi Central Bank (BRB) immediately after grant effectiveness. The PIU should submit the initial withdrawal application request together with six months program’s cash forecast. Further details on FM and disbursement considerations, with a flow of funds are included in Annex 3. D. Procurement 82. Procurement for the project will be carried out in accordance with the following Bank guidelines (i) "Guidelines: Procurement of Goods, Works and Non-Consulting Services under IBRD Loans and IDA grants & Grants by World Bank Borrowers" dated January 2011, revised July 2014; (ii) "Guidelines: Selection and Employment of Consultants under IBRD Loans and IDA grants & Grants by World Bank Borrowers” dated January 2011, revised July 2014; (iii) “Guidelines on Preventing and Combating Fraud and Corruption in Projects Financed by IBRD Loans and IDA grants and Grants”, dated October 15, 2006, revised in January 2011; and the provisions stipulated in the Financing Agreement. 83. An assessment of the capacity of the Ministry of Human Rights, Social Affairs and Gender was conducted in May 2015 and updated in May 2016. The assessment revealed lack of inner capacity to properly handle procurement processes using World Bank guidelines. Accordingly, it was agreed that a Project Implementation Unit be established under the auspice of the line Ministry with a procurement officer hired competitively. The internal procurement unit will be associated in the project procurement implementation in order to build its capacity in the management of World Bank-financed project. The main risk factors identified and their proposed mitigation measures are summarized in Annex 3. 24 84. The draft procurement plan, dated November 14, 2016 for at least the first 18 months of the project was agreed with the Bank. During implementation, the procurement plan will be updated in agreement with the project team as required - at least annually - to reflect actual project implementation needs and improvements in institutional capacity. The client will ensure that it is available in the project’s database and a summary will be disclosed on the Bank’s external website once the project is approved by the IDA’s Board of Directors. E. Social (including Safeguards) 85. The Project is expected to have a positive social impact by providing economic support to the poorest households and their communities, as described in the economic analysis above. Given the novelty of cash transfers in Burundi, there are some concerns of increased intra- household (since the women will be the recipient of the cash, which may go against gender norms) and inter-household conflict (between beneficiaries and non-beneficiaries). The promotion activities will reinforce messages about the fact that the program seeks to benefit the welfare of the whole household and include modules about household conflict and domestic violence. Transparency and understanding of the targeting process will be key to minimize inter-household conflicts and the promotion activities will be open to all community members on a voluntary basis for non-beneficiary households. Both the process and the impact evaluations will track potential changes in conflict. 86. Approximately one percent of the Burundian population are Batwa, who are recognized as an indigenous people under OP/BP 4.10. In compliance with OP/BP 4.10, the Government of Burundi has prepared an Indigenous Peoples’ Planning Framework, which was consulted in the four provinces in September 2016, and disclosed in country and in the InfoShop on November 9, 2016. An estimated 7,000 Batwa households lived in the target provinces in 2015 (approx. 1.15 percent of the total population) 15. In part because of their past nomadic hunting- gathering livelihoods, Batwas have even less access to land in the communities where they settle. Their specific needs are seldom taken into account in local development plans both through lack of capacity and interest from sectoral service providers and lack of Batwa participation in local development councils. Batwa face severe human endowment gaps (access to reproductive health services, nutrition, access to schools), and very limited access to economic opportunities and productive assets. These gaps are particularly severe for Batwa women with high levels of gender- based violence and early marriages, little awareness of their rights, and limited political representation: 13 Batwa women are members of communal councils (out of 68 Batwa council members). These gaps are large compared to Batwa men but also when compared to other Burundian women. 87. The project expected impacts on the Batwa are mostly positive if they actually participate in the project. Through fair access to the cash transfers and their accompanying activities, the impacts on Batwa beneficiaries include: an improvement in the nutrition of beneficiary children and pregnant and lactating women, a general improvement in health, strengthened cohesion and social inclusion, increased school attendance and decreased drop-out, increase in the participation of Batwa women in the management of their household goods, and eventually increase in income through own income-generation. 15 Given these preliminary figures and the experience of the Concern pilot, the team expects around 5-6 percent of project beneficiaries (up to a maximum of 10 percent) to be Batwa. 25 88. Based on the experiences in addressing the specific vulnerabilities and discrimination facing the Batwa, the Project will take specific measures to ensure the participation of eligible Batwa households. Batwa households living in the targeted collines will be included in the first list of eligible households and surveyed in the PMT survey. The final list of beneficiaries will be publicly validated. It is expected that access to ID documents may be particularly low in these communities and the project will support beneficiaries in acquiring national IDs and birth certificates. 89. The program will also ensure they can participate in the accompanying measures by partnering with NGOs that are experienced in taking into account their specific challenges, in particular with hygiene, food preparation and feeding practices. Extra behavior change communication may be necessary, in particular for the potential use of cell phones as well as chargers and incentives to open accounts in micro-finance cooperatives to safeguard the transfer monies, given increased risks of thefts. Service providers will ensure that Batwa households can participate in regular IEC/BCC sessions with the rest of the community to foster social inclusion. However, specific sessions on sensitive topics such as reproductive health and gender-based violence may be conducted separately for the Batwa. Once collines are selected for participation in the project, provincial Indigenous People’s Plans will be developed. After validation of the IPP by the Batwa beneficiaries, the latter will nominate a provincial representative who will work closely with the program provincial and communal focal points and coordinate activities targeted to the Batwa. F. Environment (including Safeguards) 90. The Project will not support public works, nor any changes in land use. Therefore, the project is categorized as B and does not require an Environmental Assessment. G. World Bank Grievance Redress 91. Communities and individuals who believe that they are adversely affected by a World Bank (WB) supported project may submit complaints to existing project-level grievance redress mechanisms or the WB’s Grievance Redress Service (GRS). The GRS ensures that complaints received are promptly reviewed in order to address project-related concerns. Project affected communities and individuals may submit their complaint to the WB’s independent Inspection Panel which determines whether harm occurred, or could occur, as a result of WB non-compliance with its policies and procedures. Complaints may be submitted at any time after concerns have been brought directly to the World Bank's attention, and Bank Management has been given an opportunity to respond. For information on how to submit complaints to the World Bank’s corporate Grievance Redress Service (GRS), please visit http://www.worldbank.org/GRS. For information on how to submit complaints to the World Bank Inspection Panel, please visit www.inspectionpanel.org. 26 Annex 1: Results Framework and Monitoring Annex 1: Results Framework and Monitoring . Country: Burundi Project Name: BI-Social Safety Nets (Merankabandi) (P151835) . Results Framework . Project Development Objectives . PDO Statement The objectives of the proposed Project are to provide regular cash transfers to extreme poor and vulnerable households with childrenin selected areas while strengthening the delivery mechanisms for the development of a basic social safety net system. These results are at Project Level . Project Development Objective Indicators Cumulative Target Values End Indicator Name Baseline YR1 YR2 YR3 YR4 YR5 YR6 YR7 YR8 YR9 Target Beneficiaries of Safety Nets programs 0.00 0.00 24000.00 48000.00 48000.00 48000.00 48000.00 48000.00 (number) (Number) - (Core) Beneficiaries of Safety Nets programs - 0.00 0.00 12000.00 24000.00 24000.00 24000.00 24000.00 24000.00 Female (number) 27 (Number - Sub- Type: Breakdown) - (Core) Beneficiaries of Safety Nets programs - Unconditional cash transfers 0.00 0.00 24000.00 48000.00 48000.00 48000.00 48000.00 48000.00 (number) (Number - Sub- Type: Breakdown) - (Core) Proportion of households enrolled in the beneficiary database living 0.00 0.00 50.00 60.00 70.00 70.00 70.00 70.00 below the extreme poverty line (Percentage) . Intermediate Results Indicators Cumulative Target Values End Indicator Name Baseline YR1 YR2 YR3 YR4 YR5 YR6 YR7 YR8 YR9 Target Proportion of beneficiary households that 0.00 0.00 0.00 10.00 25.00 30.00 40.00 40.00 increased their food consumption by 28 more than 10% than the control group (Percentage) Proportion of beneficiary households in the beneficiary collines 0.00 0.00 50.00 50.00 70.00 80.00 80.00 80.00 participating in behavior change promotion activities (Percentage) Proportion of beneficiary mothers of children less than 2 years old Value Baseline Baseline Baseline Baseline Baseline Baseline Baseline who breast-feed TBD +10% +10% +20% +25% +25% +25% +25% according to the recommended protocol (Text) Proportion of beneficiaries who know the dates of their 0.00 0.00 50.00 70.00 75.00 75.00 payments and exact amounts (Percentage) Percentage of community 0.00 0.00 60.00 70.00 70.00 70.00 members (of 29 which women) satisfied with the targeting process (Percentage) Households with up-to-date information (less than 3-year 0.00 0.00 100.00 100.00 100.00 90.00 90.00 90.00 old) in the registry (Percentage) Beneficiaries receiving payments on the Baseline Baseline Baseline schedule set-up Baseline Baseline Baseline TBD 0 +25% (at +25% (at +25% (at in the +10% +10% +20% least 80%) least 80%) least 80%) operational manual (Text) Social Registry MIS developed (modules) and No Modules Module 4 Module 3 Reports Reports Reports Reports working module 1, 2 (reports) (Text) Public National National National expenditure in Mapping and and and social assistance Standardiz Governme Governme of social province province province programs Ad-hoc ed data nt and nt and assistance level level level tracked and report collection donors donors expenditur mapping mapping mappint available on a and report report report e in 4 of SA of SA of SA yearly basis at provinces expenditur expenditur expenditur the Permanent es es es 30 Executive Secretariat of the National SP Commission (Text) Basic knowledge of implementers about SSNs Scores and/or the CT N/A SSN TBD TBD TBD TBD TBD TBD program Module (concepts and practice) (Text) Cash transfer program Instrument Timely Timely Timely Timely Timely grievance Protocols s ready resolution resolution resolution resolution resolution redress N/A defined and actors of 20% of of 25% of of 30% of of 40% of of 40% of mechanism trained cases cases cases cases cases implemented (Text) Process Baseline At least 12 At least 12 evaluation CT Baseline N/A Follow-up Follow-up Follow-up evaluation evaluation implemented targeting CCT reports reports (Text) payment 3 3 Impact Baseline Midpoint evaluation evaluation evaluation Final data N/A for 2 data datasets datasets realized collection provinces collection and and (Text) reports reports . 31 Indicator Description . Project Development Objective Indicators Responsibility for Data Indicator Name Description (indicator definition etc.) Frequency Data Source / Methodology Collection Beneficiaries of Safety Nets This indicator measures the number of Yearly Cash Transfer Program Project Implementation programs (number) individual beneficiaries covered by safety MIS Unit nets programs supported by the Bank. Safety nets programs intend to provide social assistance (kind or cash) to poor and vulnerable individuals or families, including those to help cope with consequences of economic or other shock. Beneficiaries of Safety Nets This indicator measures female Yearly Cash Transfer Program Project Implementation programs - Female participation in SSN programs. It has the MIS Unit (number) same definition as the "Beneficiaries of Safety Nets programs" but applies only to female. This indicator will yield a measure of coverage of SSN projects disaggregated by gender (in absolute numbers) Beneficiaries of Safety Nets Follows the safety nets programs’ Yearly Cash Transfer Program Project Implementation programs - Unconditional classification used in SP Atlas. MIS Unit cash transfers (number) Proportion of households This indicator measures the proportion of Yearly Registry database Project Implementation enrolled in the registry extreme poor households per the national Unit living below the extreme poverty line, in the targeting database. poverty line This indicator will yield a measure of the quality of targeting . 32 Intermediate Results Indicators Responsibility for Data Indicator Name Description (indicator definition etc.) Frequency Data Source / Methodology Collection Proportion of beneficiary Based on results from similar programs in Bi-yearly Process evaluation Project Implementation households that increased Zambia and results from the Tuburamure (qualitative) and impact Unit their food consumption by trial in Cankuso and Ruyigi provinces evaluation (quantitative) more than 10% than the (IFPRI, 2015 and 2016) control group Proportion of beneficiary This indicator tracks the attendance of Every 6 Cash Transfer MIS from Project Implementation households in the beneficiaries in the monthly IEC sessions months implementing NGO reports Unit beneficiary collines through attendance records of the session participating in behavior provider. change promotion activities Proportion of beneficiary WHO data from 2009 reported exclusive Bi-yearly Beneficiary survey, spot Project Implementation mothers of children less breast-feeding at 6 months at 76% and checks (observations) and Unit than 2 years old who breast- continued breast-feeding at 2 years at 87% impact evaluation feed according to the in rural areas recommended protocol Proportion of beneficiaries This indicator measures the basic Bi Yearly Beneficiary survey Project Implementation who know the dates of their knowledge of beneficiaries about a key Unit payments and exact program process amounts Percentage of community This indicator measures the perception of Bi-Yearly Process evaluation and Project Implementation members (of which women) community members (with sex- beneficiary surveys Unit satisfied with the targeting disaggregated results) about the process community-based and proxy-means targeting in terms of transparency and efficiency. Households with up-to-date This indicator tracks the dates of Yearly Social registry MIS Project Implementation information (less than 3- interviews of households selected by the Unit year old) in the registry CBT, in the program database. It will provide a basic measure of the timeliness of the database information. 33 Beneficiaries receiving This indicator measures the differences Semester Social Registry MIS Project Implementation payments on the schedule between the planned payment date and Unit set-up in the operational actual receipt date by the recipient. This manual will provide a measure of the payment agency(ies)’s performance. Social Registry MIS Registry MIS modules: 1 (household Yearly Social registry MIS Project Implementation developed (modules) and database), 2 (PMT), 3 (beneficiary Unit working (reports) database updating), 4 (complementary activities). Reports refer to the timely production of data for 4 quarterly reports, 1 annual report. Public expenditure in social The CNCA tracks expenditures for donors Yearly CNCA report and M&E SEP/CNPS assistance programs tracked registered with the Ministry of Finance. system and available on a yearly The report would use the format basis at the Permanent developed for the SSNA. In the third year, Executive Secretariat of the it would also include a mapping of the National SP Commission interventions from NGOs and local government levels in the provinces where the cash transfers operate. Basic knowledge of These would be modelled on the entry test Yearly Learning evaluations Project Implementation implementers about SSNs in the basic SSN course and could cover Unit and/or the CT program key operational processes relevant for (concepts and practice) different types of operators. Cash transfer program This indicator tracks the milestones in Semester Cash transfer program MIS Project Implementation grievance redress developing the system to track and resolve Unit mechanism implemented grievances about the cash transfer program. Process evaluation This indicator tracks the data collection Semester Cash Transfer Program Project Implementation implemented and analysis about the implementation of MIS Unit key program processes: targeting, payments and complementary activities. Impact evaluation realized This indicator tracks the data collection Bi-yearly Cash Transfer MIS Project Implementation and analysis of the impacts of the program Unit 34 with baseline before the transfers start, midline at 24 months and endline 6 months after the end of the program. 35 Annex 2: Detailed Project Description 1. Supporting resilience in the face of renewed fragility and addressing some structural poverty issues: The proposed Project supports the Government of Burundi in its efforts to support the extreme poor for their consumption and investments in human capital so as to increase their own income-generation capacity. The Project also supports the government in getting started on a twin-track set of interventions to address both renewed vulnerability and structural poverty and thus contribute to the consolidation of the peace-building recovery and growth. The proposed Project focuses on supporting two sets of key instruments: a) The first of these are the cash transfers and their complementary foundational human capital investment measures. By providing reliable support to the poorest and food insecure households in the medium-term, the cash transfer program support them in improving their situation in terms of strengthened resilience to shocks to avoid poverty traps; access to human development services such as education, health, and nutrition; and access to economic opportunities. While the Project provides start-up funding for sixteen communes in four provinces, it is expected that the demonstration effect stimulates increased participation of the government and other key donors; b) The second key suite of interventions are the key delivery mechanisms built around the cash transfer program but potentially adaptable for other types of social safety net interventions. The targeting processes and the ensuing eligible/beneficiary databases would provide a mechanism to target the extreme poor in an efficient and transparent way. Building such processes is complex and costly, but, when used effectively by targeted programs, they become a shared public good which allows the overall system to improve its targeting (and hence its impact) at a lower cost. The MIS would provide real-time information about program implementation, first for the SEP/CNPS and the National Social Assistance and National Solidarity Directorate of the Ministry of Human Rights, Social Affairs and Gender and potentially for other members of the CNPS. It would enable implementers and policy-makers to track inputs, outputs and results across interventions. This is at the basis of evidence-based planning for social services and provides a key tool for coordination. 2. Providing catalytic funding to build-up capacity for a future roll-out. The proposed project is spearheading the scaling-up of cash transfers in the country building on nascent efforts on donor and government coordination and growing interest in cash transfers as a social assistance instrument across both humanitarian and development sectors. The ability of the Government to target the extreme poor can be an important factor in policy discussions around addressing the economic effects of the political crisis and promoting a return to a more inclusive growth. While the cash transfer program and its associated delivery mechanisms will need to reach national coverage to fully play their transformative roles, they are complex and will need to be rolled out progressively. This will allow for testing and fine-tuning instruments and process. The proposed Project will focus on providing support for the development of the instruments, tools, methodology and material for the cash transfer program, which can subsequently be adopted more widely both geographically and across programs. 3. Facilitating multi-sectoral collaboration and impacts. The full effectiveness of the CT program relies on the coordination of efforts from several sectors which also benefit from the 36 demand-side stimulus provided by the CTs. Indeed, the transfer program relies on the contribution of social (health, nutrition, education) and development (agriculture, land management and community development, micro-finance, etc.) actors to design its promotion activities and in the future, to ensure that the supply of basic services is in place to respond to the beneficiaries’ demand. On the other hand, the targeting mechanisms provide a tool for all sectors to improve their programs (by informing program design or guiding investment plans to meet supply gaps). Eventually, this is expected to help programs coordinate their interventions, seeking synergies and avoiding duplication 4. The proposed Project is organized around three components: Component 1 supports the building blocks for a basic social safety net system; Component 2 supports the national Cash Transfer Program; and Component 3 supports Project management, monitoring and evaluation. Component 1 – Support to the cash transfer program (US$26.7 million equivalent) 5. The first component will support the development and implementation of a cash transfer program to be implemented in phases, targeted towards extreme poor and vulnerable 16 Burundian households with children under 12 years of age in selected areas. The program combines regular cash transfers (Sub-component 1.1) with promotion activities to increase knowledge about essential care and nutrition practices and change behaviors to promote investments in human capital (Sub-component 1.2). The rationale for this combination is a twin- track approach to poverty reduction. Cash transfers will contribute to poverty alleviation in the short-term and enable households to consume more and better food and to access basic services. Behavior change promotion activities will contribute to poverty alleviation in the medium-term by potentially activating synergies between health, nutrition, learning and income-generation capacity. Extreme poverty such as the conditions facing potential program beneficiaries has negative effects both on cognitive capacity (capacity to process information and make decisions) and executive control (impulse control). The cash transfer will provide households with some “breathing space” to acquire new knowledge and practices. The government also has a strategy to improve the supply of basic health and education services and strengthen universal access. While this is not in the scope of the project, ultimately the information in the database can support some of the planning by helping identify needs. 6. Objectives and beneficiaries. The objective of the cash transfer program is to support the consumption of extreme poor and vulnerable households and promote investments in the foundations of human capital (nutrition, health, early childhood development). Ultimately, a national scale program would aim to reach all the poorest households across Burundi (estimated at about 300,000 households, with approximately two million members 17). The beneficiaries are the households, which receive the cash transfers and benefit from an increase in consumption. Within the household, the focus is on women – as cash transfer recipients and targets of some of the behavior change activities – and children – as targets of the behavior change activities and 16 The main symptom of vulnerability in Burundi is food insecurity. At the country-level, it translates into most of the country classified as “stressed” and an increasing area in the North and East classified in “crisis” in the Integrated Food Security Phase classification (FAO, 2016). At the household-level, it translates in increasing numbers of individuals in chronic or acute food insecurity (estimated to 4.6 million in October 2016 (OCHA)) 17 These estimations are based on the poverty analysis of the 2013/14 household survey (Enquête sur les Conditions de Vie des Ménages Burundais, ECVMB) 37 investments in human capital through increased health, nutrition and education. The beneficiaries will be selected from the initial community listing, based on their basic proxy-means test score (see component 2). The final list of eligible households for the cash transfer program will be shared with the community, for validation. Each household will name a recipient (and a substitute) to receive the transfers. The recipient is the main care provider for the children, mostly their mother. 7. Program duration. The program is envisaged to be the precursor of a permanent national program, catering to the extreme poor. However, for each household, the cycle of program participation is three years (30 months with cash transfers and Behavior Change Communication (BCC) activities and an additional 6 months with BCC activities only), which reflects the program’s focus on extreme poverty and on building a basis for resilience. After three years, a re- certification process would again assess households’ eligibility for continued participation or graduation out of the program. Graduation from the cash transfer program does not preclude inclusion in other social protection and livelihood support programs such as health insurance, public works, micro-credit and other “solidarity economic” activities. The re-certification will be synchronized with new rounds of registration for the registry. 8. Implementation arrangements. The program will be implemented by the Permanent Executive Secretary (SEP) of the CNPS, placed under the authority of the Ministry of Human Rights, Social Affairs and Gender, under the guidance of the Steering Committee and Technical Committee for the National Social Protection Commission. Implementation will be supported by a Project Implementation Unit with a central team to organize and supervise implementation and provincial (four provincial focal points) and commune-level (16 commune focal points) teams responsible for implementation in the field. Payment activities will be contracted to payment agencies while the behavior change activities will be contracted to NGOs. Relevant government entities (Ministry of Health (MSPLS), Ministry of Education, and Ministry of Communal Development) are engaged at the national level (at the strategic and technical levels through the CNPS, its technical committee and the relevant thematic groups), the provincial level (through the CPPS) and at the local level where present. In particular, the focus of the Health System Support project ((KIRA, P156012) on community health and nutrition provides for potential synergies. De- concentrated administrative structures will also support the program’s implementation in the collines. 9. The program will be implemented in phases. The program will prioritize the poorest areas (with a combination of monetary poverty rate and chronic malnutrition at the province level and extreme poverty rate at the commune level). The project will support a first phase of the program to establish transparent and rule-based processes and allow for the gradual fine-tuning of these processes and the instruments for program operation. For example, the involvement of communities in the targeting may evolve as better household survey data becomes available; cash payments may evolve, from cash to electronic or mobile payments; the enforcement of conditionalities may evolve with the availability of services and administrative capacity. The registry will start in 16 communes of the four provinces with the largest rates of monetary poverty and chronic malnutrition. Sub-component 1.1. Cash Transfers (US$19.5 million equivalent) 10. Payment amount and schedule. The level of transfers will be BIF 20,000 (US$ 12 equivalent) per household and per month for 30 months. This corresponds to approximately 19 38 percent of the extreme poverty line 18. It is in line with transfer amounts in comparable African contexts. For operational reasons, the payments may be originally made on a quarterly basis. Once payments systems are established, alternative payment schedules will be tested, with a goal of monthly payments. The payment schedule takes into account the lean periods (March-April and December) and periods of significant expenditure (beginning of the school year in September). Figure A. 2.1: Seasonal and payment calendar Payment calendar Source: PAM 2014. Gitega, Karuzi, Kirundo, and Ruyigi provinces are located in the Eastern and Northern depressions, the Plateaux humides, and Plateaux Secs de l’Est 11. Payment mechanisms. A payment study has assessed the payment options in the four selected provinces. So far, experiences of delivering payments have relied on informing beneficiaries through mobile phones 19 and distributing cash through the post office (Concern), mobile payments (Concern), NGOs (Red Cross) and micro-finance institutions delivering cash directly, and payments through mobile units of a bank (WFP). Financial inclusion is very low across the poor (less than 7 percent of both males and females above 15 years of age have access to an account and 2.3 percent among the bottom 40 percent (Global Findex, 2014). Mobile phone companies have started to offer electronic wallets and cash transfer programs and with the recent changes of the legal framework, they could expand rapidly, following the pattern in other 18 Extreme poverty line is BIF 28,201 per adult equivalent per month 19 Mobile phone penetration rate was 22.6 percent in Burundi, placing the country as the 7th least developed market. Based on Concern experience, if possible, the first transfer of the program would include a cell phone (as an in-kind share of the first payment). 39 neighboring countries. The payment mechanisms that will be selected, through a competitive bidding process, will demonstrate the following competencies: ability to pay, accessibility (safe payment network and good connectivity), transparency, proven reliability, acceptable transaction costs, efficiency, capacity of planning and monitoring. Payment providers may differ in each province (or other geographical grouping as some of the selected communes are contiguous) to guarantee accessibility across the program’s area of operation. The payment study has identified potential bank and mobile providers in all provinces. 12. Co-responsibilities and verification. The cash transfer program aims to provide predictable and regular income to extreme poor households so as to enable them to increase consumption and invest in the foundations of their children’s human capital. Given the limited administrative capacity, beneficiary households will initially be required to participate in the behavior change communication activities described in sub-component 1.2 but the payments will not be conditional on attendance. As the provision of basic services and the capacity to monitor improve, the program will explore the provision of making the payments conditional on the actual participation in the BCC activities and possibly on the use of basic services (verified through the health card or the primary school attendance records). Participation in the behavior change activities will be recorded in the MIS. The cash transfer recipient, her husband and other adults will be expected to participate in the main BCC activities, with absences collected and transmitted to the provincial and central program offices. In a first step, a local program official (from the CFDC) and the colline focal point will visit the household, assess the reason for missing the session (illness, accident) and encourage them to participate in the future. If and when payments become conditional, in a second step, two consecutive absences without justification may cause the suspension of half of the following payment. Decisions about conditionalities will be taken at the project mid-term based on implementation performance. 13. Specifically, subcomponent 1.1. will support the payment of cash transfers to 48,000 households of the national cash transfer program, as well as the payment of associated financial fees. This corresponds to 15 percent of the poorest 20 percent households in Burundi (and 60 percent in the four provinces). Twelve thousand households in the provinces of Karuzi and Ruyigi (total 24,000) would enter the program in the first wave. Then the remaining 12,000 households in Gitega and Kirundo province would enter the program in the second wave as described in the results framework in Annex 1. At full scale, the program targets the first quintile or roughly 300,000 households with children under 12 years of age. Subcomponent 1.2: Behavior change for investments in human capital (US$7.2 million equivalent) 14. Curriculum of promotion activities. The core curriculum for the promotion activities will consist of five themes: (1) Program objectives, activities and roles and responsibilities including the information about processes to obtain birth and marriage registration and national IDs (required for cell phone registration), (2) nutrition for children and pregnant women (Alimentation et nutrition du jeune enfant), (3) early childhood development (positive parenting, play-based learning, school preparedness based on the Pratiques Essentielles Familiales package), (4) mother and child health (pregnancy risks, childhood illness management, family planning, malaria prevention, HIV/AIDS, intimate partner violence prevention), and (5) education, skills, and income-generation (importance of schooling and training, access to micro-finance, rotating saving and credit organizations nawunawaze, basic accounting, etc). These themes are related to 40 the program’s objectives, aligned with the priority needs identified in the vulnerability analysis of 2012-13, appropriate given the delivery method and the learning can yield potential significant impacts and spillovers at the household and community levels. For example, evaluations of Tubaramure, a food-assisted integrated health and nutrition intervention trial in Ruyigi and Cankuso provinces (IFPRI, 2014 and Leroy et al., 2016) show very little knowledge and practice about complementary foods for children above 6 months of age, scant knowledge of danger signs for acute respiratory infections (the main cause of death for children below five years of age) and management of diarrhea. The 2010 DHS also revealed high unmet contraception needs (31 percent for limiting and spacing births), a significant proportion of women not deciding about the use of their own income (16.2 percent in the North and 7.5 percent in the East), a high proportion of women not involved in decisions about important purchases for the household (42 percent), and a high female acceptance of intimate partner violence (74 to 78 percent for at least one of the following reasons: burning food, arguing with the husband, neglecting children, going out without informing the husband, or refusing sex). 15. The modules will be implemented through a set of activities (defined in a behavior change facilitator manual) and support materials (posters, illustrated booklet, radio and video modules). The modules will follow international and national guidelines. UNICEF will develop the curriculum for the nutrition, health and ECD modules as their contribution to the proposed project. The development will involve close coordination with the Ministries of Health, the 2nd Vice-Presidency and the SUN coordination. The development will draw on the latest methodologies for child and adult learning (active and practice-based learning) and cognitive stimulation. Lessons from the process evaluation of the USAID-supported Food for Peace program in Cankuso and Ruyigi (IFPRI, 2014) and its on-going expansion in Muyinga emphasize the importance of the design of the delivery of the behavior change communication and continued training for the frontline staff and volunteers to ensure accurate delivery. The program implementation team will develop the curriculum on program processes based on the operation manuals. 16. The materials will be tailored to the specificities of the Burundian rural context, in particular the gender norms in rural areas, the very high levels of chronic child malnutrition, and the high levels of interpersonal violence which may affect some practices. The materials and methodologies will be adapted to ensure their appropriateness for particularly marginal groups such as the Batwa or single mothers. Social exclusion may affect their capacity to learn and their practices may differ from other groups so these factors will be taken into account when designing and implementing the modules. 17. The beneficiary household is the target of the promotional activities. This reflects the program’s focus on households and the potential contribution of all household members to adopt new practices. While mothers are the primary care providers, other household members – especially fathers in their role of household head, grand-parents, and older children as potential care providers -- play an important role in defining behaviors and norms. The BCC activities will therefore engage with the household as a whole and with tailored activities to ensure that key messages are received and understood by all household members. The monthly sessions will include household sessions open to the participation of all household members above 15 years of age, mothers’ sessions and fathers’ sessions. Children themselves may be engaged through age appropriate play-based learning. Depending on logistical capacity, sessions will be open to other community members on a voluntary basis, which would help foster changes in norms. 41 18. Specific activities will target community leaders (chef de colline, religious leaders, imboneza 20) and service providers (community health workers and teachers) since they contribute to the definition of social norms and conduct. They will receive pedagogical materials (posters, booklets) to strengthen their dialogue with the community on these issues and post in their health post or classroom. 19. Behavior change activities will take place once every month. Activities will be organized and implemented by qualified NGOs, selected on the basis of their skills and knowledge of the local conditions. Preliminary institutional assessments showed that such potential implementers are present in each province. They will be supervised by the program communal teams which will include the MDPHASG social worker when available, and the program focal point. Volunteer community members (for example the imboneza, Red Cross volunteers, head of a 10-household cell, and members of the Child Protection committees) who play a leadership role may be involved to support organization and encourage attendance. They would work with 50 household and will receive an amount equivalent to half transfer. Beneficiary households will be required to attend while community leaders and service providers will be encourage to attend their sessions but their participation will be voluntary. 20. Specifically, the proposed project will support activities including: (a) The proposed Project will finance mostly the development of the modules on the program, and on schooling, skills and income-generation. The development of the modules on health, nutrition and ECD will be the responsibility of UNICEF. (b) For the delivery of the BCC activities to the beneficiaries and their community, the proposed Project will support output-based contracts to qualified NGOs. The project will also finance the production of the related learning materials (posters, images, booklets, radio messages, videos). The program itself will not distribute goods to beneficiary households but may coordinate with delivery NGOs to give small materials to encourage the behavior changes such as soap, a covered bucket for water, mosquito nets, etc. (c) The proposed Project will also support activities for the programming, supervision, monitoring and evaluation of the activities by the program provincial and communal teams with the support of the commune-level structures of the MDPHASG. UNICEF may provide assistance for the implementation (and resources for additional modules on peace- building), technical supervision and quality control. Component 2: Support to the key delivery mechanisms of a basic social safety net system (US$7.5 million equivalent) 21. The second component will support the development and implementation of four key instruments for the implementation of the cash transfer program and the future coordination of other targeted social safety net interventions targeted to the poorest households in Burundi. The framework for the component comes from the National Social Protection Strategy, which the SEP/CNPS and Technical Committee of the National Social Protection 20 Volunteer focal points of the MDPHSAG in charge of mediating in cases of domestic violence and child protection issues. 42 Commission21 developed with support from the African Development Bank, UNICEF and the World Bank in 2014-15. The strategy has three key objectives: (i) increase access to basic social services, (ii) ensure food security and basic income security, (iii) strengthen social and natural risk management; and a transversal objective (iv) contribute to decrease chronic malnutrition. (Box 2). 22. The four key instruments include: a targeting mechanism and potential beneficiary database; core modules of a basic management information system; a monitoring and evaluation module; and capacity building for the implementation of the NSPS. The instruments will be initially anchored around the cash transfer program described in Component 1, but with a view to serve a broader set of targeted programs that would contribute to the realization of the NSPS objectives. Sub-component 2.1: Implementation of a precursor database for the registry (US$2.0 million equivalent) 23. The beneficiary database and its associated targeting mechanism are a key pillar for the coordination of programs. They are also critical for ensuring transparency in the selection of beneficiaries in the Burundian context of recurrent fragility and entrenched structural poverty The database will be initially be developed to identify the extreme poor households for the purpose of the cash transfer program and its complementary activities. However its function is to ultimately serve as the foundation for a registry of extreme poor households, which would be used for all targeted key programs as identified in the NSPS, and for the complementary targeted interventions of CNPS member ministries. A registry is a public good and its use will lower the overall cost of targeting. The fact that the targeting will use clear and transparent rules (explicit criteria, community involvement, and public validation) in the registration of extreme poor households will contribute to strengthening the governance of the social assistance sector. This is especially important in the Burundian context of recurrent fragility and persistent post-conflict divisions. 24. The development of the database will take place in phases, according to the availability of poverty information and the implementation of the cash transfer program. Initially, the program will start in selected provinces 22 (the provinces were selected as the ones with the highest combination of monetary poverty from ECVMB 2013-14 and chronic malnutrition from DHS 2010 and include Gitega, Karuzi, Kirundo and Ruyigi) for the purpose of identifying potential beneficiaries of the cash transfer program. Given the implementation of a US$50 million Food-for-Peace project in the province of Muyinga 23, targeting all mothers of children less than two years of age, the project will not work in that province. The project preparation team worked with the Poverty team and the National Statistical Institute (ISTEEBU) to build small-area extreme poverty maps (based on the ECVMB 2013-14 in combination with the 2010 Census data), which provided the basis for a more systematic geographical targeting at the commune-level and could inform the planned expansion of the safety net program and other programs using the registry. In each province, the four communes with the highest poverty (from 21 The CNPS includes 11 ministers under the authority of the President 22 The country is divided in 18 provinces, 129 communes and 2,638 collines. 23 The Amashiga program started in 2016 and aims to foster community based-nutrition, using the lessons of the Tubaramure pilot in Cankuso and Ruyigi. The program would distribute food to all households with pregnant women or children under two, and foster behavior change in terms of food consumption, preparation, production, water and sanitation and hygiene practices, and access to health. 43 the 2013-14 ECVMB data analysis) were initially selected for the program (see Table A.2.1). Table A.2.1: Provincial rates of monetary poverty and chronic malnutrition Province Monetary poverty Stunting (percent) (percent) Bubanza 53.7 56.1 Bujumbura mairie 17.4 27.9 Bujumbura rural 35.2 59.8 Bururi 48.9 49.6 Cankuzo 71.1 48.3 Citiboke 54.8 52.2 Gitega 72.0 59.7 Karuzi 65.8 68.1 Kayanza 71.8 55.4 Kirundo 69.2 60.3 Makamba 53.3 62.1 Muramvya 48.0 63.5 Muyinga 80.4 61.1 Mwaro 39.9 56.1 Ngozi 49.5 71.2 Rutana 60.1 56.6 Ruyigi 77.8 66.1 Sources: ECVM 2013/14 and DHS 2010 25. The first phase of the project will start in the Karuzi and Ruyigi provinces. The second phase will include the remaining eight communes in the provinces of Gitega and Kirundo (Table A.2.2). The collines within the commune or province capital will not be included in the program. In the first phase, in the Karuzi communes, the program will select randomly half of the collines with a stratification by zones to ensure that some collines of all the zones are included. In the Ruyigi communes, the program will first select randomly a zone and include all collines in the selected zone then select another zone until it reaches half of the collines in the selected communes. The process evaluation will follow the cost-benefits associated with concentrating collines vs. covering all zones in terms of operational costs and feasibility and political acceptability to inform the process of the roll-out in the second phase. The randomization at colline-level is an important element of transparency as no recent information is available on communes about welfare levels, it is also key for the identification of program impacts in the impact evaluation. The random selection of collines will take place publicly in a commune-level assembly involving program staff, the province governor, the commune administrator and all the chefs de colline in that commune as well as representatives from religious groups, traditional authorities, civil society and the Batwa. 44 Table A.2.2: Administrative divisions of the sixteen targeted communes Province Communes Number Number of of zones collines Gitega Bugendana 3 22 Buraza 3 19 Gitega 4 38 Itaba 3 21 Karuzi Bugenyuzi 3 27 Gihogazi 3 21 Mutumba 2 11 Nyabikere 3 14 Kirundo Bugabira 3 12 Busoni 5 41 Kirundo 4 29 Ntega 4 37 Ruyigi Butaganzwa 2 4 39 Butezi 3 15 Bweru 4 20 Gisuru 3 43 Averages 3.4 25.6 26. The household-level selection for the registry will use community-based targeting with listing of the 30 percent poorest households in the colline based on nationally agreed criteria for extreme poverty including an indicator of access to land, a dependency index (taking into account all not able-bodied members: children, elderly, persons with a disability and chronically ill members), and welfare indicators. Because of the extreme social exclusion they face, Batwa households will be automatically registered in this first stage list through a separate listing process. All households in the listing will be surveyed to compute a basic multi-dimensional proxy-means test (PMT) score, and rank households by their poverty status 24. The preliminary and final lists will be validated by the community. 27. The composition of the targeting committees, as well as the instructions they receive will be designed to minimize exclusion errors. The targeting committee will include representatives from the colline development committee (president and female member), representatives from religious groups in the colline (Caritas or priest, pastor), representatives from the Child Protection Committee, the imboneza if present (volunteer women in charge of addressing domestic violence and children issues, with support from MDPHSAG), community health workers and the Red Cross, the community leader (bachingonazi) or customary leader (Abagobo- 24 The poverty analysis on the household survey data (ECVMB 2014) for the poverty map will provide the PMT coefficients and inform the design of a questionnaire to collect data on the variables associated with extreme poverty at the household-level. 45 amushingonare) for a total of 8 to 10 members with at least half of the members from civil society. Following the FIDA model, an NGO will be recruited to provide technical assistance to the targeting committees to ensure that they follow the criteria and adhere to the process. The committee list will be public and the final cash transfer beneficiary list will be validated in a colline in an assembly. All households in the listing will be surveyed by a data collection firm to compute a multi-dimensional proxy-means test (PMT) score, and rank households by their poverty status. 28. As the program expands, the process may be reviewed. Based on the progress on coordinating other information and additional funding available for the program, at mid-term review, a decision will be taken to continue with the initial process; or to apply the questionnaire to all potentially eligible households, based on the small-area estimate of extreme poverty in the commune (and the community validation may then occur afterwards); or to use only the community-based targeting. 29. The targeting questionnaire will collect basic socio-economic information about the household for the PMT calculation, as well as key information on specific vulnerabilities, access to basic social services, and livelihood strategies. The exact content of the questionnaire will be discussed by an advisory group including technical staff from SEP/CNPS, MDPHASG, MSPLS, other ministries involved in core SSN interventions, as well as ISTEEBU. This information could then be used for targeting by other “core” SP programs, and other programs managed by the Ministries members of the CNPS. The analysis of the data will provide a portrait of the extreme poor, their access to services and their livelihood strategies which can help inform the design of poverty reduction programs and interventions to increase the extreme poor’s access to basic services. Information-sharing protocols will be developed so that programs that target the extreme poor can have access to the information but that the information remains secure. 30. The proposed methodology will be initially implemented in the 8 selected rural communes of the Karuzi and Ruyigi provinces and then will be reviewed for the implementation in the additional 8 communes in Gitega and Kirundo provinces to: (i) assess the operational processes for the implementation of the targeting, in terms of organizational capacity and credibility, time and costs, (ii) test the targeting efficiency of the proposed CBT and PMT combination given the prevalence of poverty and the dearth of reliable consumption data. The first phase results will inform the roll-out in the next eight communes. 31. As part of the registration process, the project will register the national ID number of the proposed cash transfer recipient and a potential substitute. An estimated 20 percent of Burundians do not have an ID card and this proportion reached 30 percent among the beneficiaries of the Terintambwe pilot). If needed, the project will coordinate with the Ministry of Interior for the provision of national identity cards to potential recipients of the cash transfer program. This will also help ensure their access to other public services and increase their citizen participation in society. 46 Box A.2.1: National ID card in Burundi The information on the present ID card includes: last and first names, names of parents, birth date, civil status, occupation, a black and white picture of the bearer and his/her signature. It also includes an ID number and the places of residence. To obtain an ID card, one must present a birth certificate and proof of residence signed by the chef de colline and the commune administrator. Cost in 2016 is BIF 2,500.00 (US$ 1.61). Burundi has piloted a biometric ID card. This card would include: first and last name, sex, birth date, province, commune, current residence, parents’ names and children’s names, CAM card number, Social Security National Institute number, Mutuelle de la Fonction Publique number, profession, place of birth, picture, fingerprint, profession. It would be machine-readable per the requirements of the EAC. 32. The sub-component will support the background analysis for the implementation of the database and its implementation. These include: (a) Design and implementation of key tools: updated poverty and malnutrition maps, development of community-based targeting criteria and processes, development of registration questionnaire, construction of proxy-means test score, the organization of the different committees involved in the registration, the implementation of the PMT survey, storing and analysis of the data as well as the preparation of beneficiary lists for the cash transfer program and its complementary activities. It will also support the acquisition of key equipment (hardware, software, back-up equipment). (b) Implementation of the targeting and registration in selected areas including the organization and support of the targeting committees, the implementation and processing of questionnaires and the preparation of the list of registered households as well as eligible households for the cash transfer program. (c) Information campaigns to explain the purposes and processes of the targeting and registration to commune and colline administration and local government staff and to households in the selected collines, potential additional activities to ensure that beneficiaries are aware of the requirements and supporting documents for national ID cards, (d) Quality controls: Spot checks, process evaluations, analysis of targeting efficiency to identify necessary adjustments. Sub-component 2.2. Core modules of the Management Information System (US$1.5 million equivalent) 33. This sub-component will support the development of basic core modules for Management Information System to support the delivery mechanisms for a basic social safety net system. All modules will use a unique individual identification number. The modules will enable the production of key social program indicators used for the monitoring and tracking system of the SEP. The use of these modules is expected to improve program design and implementation, foster future coordination between programs, improve program design and 47 implementation, and the monitoring and evaluation of the cash transfer program, as a template for other social safety net interventions. 34. Building on the database of eligible households, these modules will include: (i) program beneficiary lists with an eventual registration of complementary activities, (ii) payment modules (payroll and the reconciliation from the payment provider(s)), (iii) operational tracking of program, and (iv) basic monitoring and evaluation, including beneficiary feedback and grievance redress mechanisms when operational. The program beneficiary lists will start with the cash transfer beneficiary list and track beneficiaries’ participation in the complementary activities set-up by the program. While initially, participation will be required but payments will not be conditional on participation, the system will provide the functionality to set up conditionalities in the future. The payment system will include the quarterly/monthly payroll based on beneficiary lists, the amounts transferred to the payment agency(ies), the beneficiary receipts and the reconciliation of accounts. The operational tracking module would provide an operational dashboard to enable program managers to plan and track activities, human and material resources and other inputs at the central, provincial and communal levels. The M&E system would track financial outlays, key program results (including those core indicators that would be common across programs within the SP system), impacts and beneficiary feedback as inputs to guide program management in the implementation of the programs. The grievance redress mechanism would track grievances linked to targeting, receipt of transfers and implementation of the complementary activities. 35. The project will also finance the development and management of a grievance redress mechanisms to respond to complaints and ensure a high level of accountability across program operations. These mechanisms include: in-person complaints to program commune focal point, SMS-based system to a third-party grievance manager (conditional on finding a trusted and competent agent and for possibilities for social control of a Government program), boxes at the colline-level in the care of a civil society organization, possibly a hotline at central-level. All complaints, whether received in person, through SMS, phone or boxes will be logged in the MIS and steps for their resolution will be tracked and monthly reports sent to the operation coordinator, the coordinator for complementary activities as well as the project coordinator. 36. Typical grievances range from information requests, questions about exclusion and inclusion from the program by the community-based targeting or the proxy-means test, requests for correction of errors in registration (identification, household composition, socio- economic status), complaints about stakeholders – program officials, service providers including mobile providers and financial intermediaries and NGOs in charge of the promotion activities, local committees, authorities). The grievance redress will be built to provide resolution as close to the point of occurrence as possible and to guarantee accessibility by beneficiaries. If resolution is not satisfactory to the plaintiff, voluntary mediation via traditional conflict resolution mechanism available at the commune level could provide an additional layer of redress before escalating to the provincial and central levels. 37. This sub-component will support: (a) The development of an overall Management Information System, with modules including beneficiary lists, payment-related information, basic monitoring of operational processes including the delivery of BCC activities and the participation of beneficiary households, grievance redress records. The MIS should also produce regular reports on program 48 implementation at the colline, commune, province and central levels. The MIS will need to interface with the beneficiary database for the cash transfer program and potentially other databases constructed on the same structure for potential additional programs and with the payment agency(ies) system. (b) The design and implementation of grievance redress mechanisms with different points of entry at the colline, commune and province-levels and different channels for citizen feedback. (c) A specific MIS operational manual with a clear definition of roles in access, quality control, update, and safeguarding of personal data. (d) The implementation of the modules with the development of software and acquisition of key IT equipment and hardware (some parts of the MIS and the registry may be hosted in a cloud or on local servers, depending on relative cost-efficiency in terms of safeguarding and resilience) (e) Related communication and multi-media outreach materials as well as training for key stakeholders. Subcomponent 2.3: Monitoring and evaluation (US$2.8 million equivalent) 38. Since the project is supporting new interventions and processes in Burundi and to ensure transparency, the third sub-component will also support process evaluations of the key program processes and an impact evaluation including beneficiary surveys. The evaluations are set up to provide learning between the first and second wave of communes implementation and to provide key impact information to garner support for cash transfers as a relevant social protection intervention in Burundi. The process evaluation will focus on the core operational processes: targeting and payment processes as well as the delivery of the complementary activities. In the first phase, the process evaluation will assess the relative efficiency of concentrating collines by zones within communes. This will inform the expansion in the second wave of communes and inform the design of the operating processes for the program. The process evaluation will be complemented by regular beneficiary surveys to help map out operational bottlenecks, complaints and issues, and complement the grievance redress mechanism. The impact evaluation will focus on key poverty, welfare, and human development indicators (including women and children health and nutrition outcomes) at the household and community- levels for the cash transfers and the behavior change communication. It may also explore variations in the delivery of the modules and specific women empowerment issues related to managing cash and fostering behavior change on parenting practices and social norms. 39. The process evaluations and beneficiary surveys will provide evidence for the mid- term review stock taking of the first phase of the program, which will be used to update processes for the expansion in the other two provinces. Subsequent evaluations will feed into the program monitoring system to assess the quality of implementation. 40. The main impact evaluation will take advantage of the random selection of collines at the commune-level for a randomized cluster design. Baseline data collection in participating and control collines will take place before transfers start for beneficiary households. Mid-line data collection will take place at 24 months (at the same time of year). End line data collection will take place six months after the CT program is supposed to end (at 42 months) to assess the medium- 49 term sustainability of the changes in household welfare, a key concern of Burundian policy- makers. These evaluations will support setting-up evidence-based program management and policy design in the SEP and CNPS. 41. This sub-component will support: (a) The design of the process and impact evaluation as well as complementary analytical tools to feed into the MIS, (b) Data collection and analysis for periodic (twice a year) process evaluations, beneficiary surveys and an impact evaluation with a baseline, mid-line and endline rounds. Sub-component 2.4. Capacity building and analysis for the Implementation of the Social Protection Strategy (US$1.2 million equivalent) 42. This sub-component will support capacity building for the main actors in charge of social protection in the central government and for the local implementers of the program. The capacity building will seek to bring a common understanding about social protection across government levels (at the central, provincial, communal and colline levels) and sectors. Building on the approach of the modules and training developed for provincial stakeholders with the support of the RSR trust fund (P149639), the project will support training on key issues about social safety nets implementation, workfare as safety nets, malnutrition, and early childhood development. Since the project will support a new type of intervention (the cash transfers), new operational processes (targeting, automatized registration and payments, coordination mechanisms, grievance registration and redress and systematic M&E, contracting and partnership with CSOs), the project will support capacity building to ensure the understanding of the interventions and the adoption, use, and maintenance of these processes and the corresponding tools. 43. At the central level, the project will seek to reinforce existing or planned coordination mechanisms, focusing on the CNPS Technical Committee and Permanent Executive Secretariat, in their capacity of coordinators and regulators for Social Protection. The proposed Project would provide support to the CNPS thematic groups, including the development of their terms of references and support to their activities. Based on the needs identified in the forthcoming functional institutional assessment, the sub-component will support training and technical support, as well as assistance with coordination costs. 44. This sub-component will also support some basic analytical services to build the evidence base which the technical committee and the SEP can use for the monitoring of the overall SP system and the discussion on programs design. This will likely include an analysis of the supply of basic services (public, non-governmental and private providers), institutional analysis of the different actors involved in social protection, potential for harmonization of the government and donor information, a mapping of public works program and background analysis for the proposed high-labor intensity public works policy, complementary studies on health coverage for the poor, analysis of malnutrition dynamics and additional analysis as requested. 45. This sub-component would support: (a) Development of training materials for program implementers and other key stakeholders such as commune and province local government administrators, and training modules (face-to-face and online) 50 (b) Development of system indicators and paper-based or e-forms to collect the data, (c) Analytical services for the thematic groups, the SEP and the technical committee, (d) Activities of the steering committee (e) Participation in South-South knowledge exchanges such as the Africa Community of Practice on Cash Transfers Component 3: Project Management (US$5.8 million equivalent) 46. The third component will support a Project Implementation Unit (PIU), under the authority of MDPHASG. The unit will be headed by a Project Coordinator and staffed by a small management team of experienced technicians, who will transfer knowledge to the SEP by learning- by-doing in implementing the program. The unit will be responsible for the technical and financial implementation of project activities, including procurement, financial management, M&E, mitigation of potential negative social and environmental impacts, and communication about project implementation and results. The unit will liaise with the SEP/CNPS and the Direction for Social Assistance and National Solidarity in the Ministry of Human Rights, Social Affairs and Gender at the central level. The PIU will be represented at the provincial and commune-level with staff dedicated to the local implementation of the project in the selected provinces and communes in coordination with provincial and communal deconcentrated structures of the SEP/CNPS and the MDPHASG. At mid-term review, the project team will evaluate the possibility to revise the institutional set-up of the Project Implementation Unit with a view to transfer responsibilities to the SEP and the MDPHASG. 47. Specifically, this component will finance: (a) Core administrative and operational staff: a Project coordinator and operation manager, fiduciary specialists (audit, financial management and procurement staff), an IT system specialist, an M&E specialist, a complementary activity coordinator, a communication specialist, and support staff (senior and team assistants, drivers, guard) at the central level. The complementary activity coordinator will be in charge of monitoring social impacts, in particular as related to potential Batwa beneficiaries. In addition, the PIU will count on four provincial focal points and 16 commune-level operators in the communes where the project will operate. (b) Equipment and operating costs linked to the daily management of the proposed Project (central office rental, vehicles operation, maintenance and insurance; utilities and supplies; IT hardware and software; equipment maintenance) (c) Regular internal and external audits (focusing on financial management and procurement aspects) (d) Costs associated with project reporting including a mid-term review involving stakeholders. (e) Costs associated with development and implementation of the four provincial indigenous peoples’ plans. 51 Annex 3: Implementation Arrangements BURUNDI: Social Safety Nets project Project Institutional and Implementation Arrangements 1. A steering committee has been appointed for the project, with representatives from central and provincial government and civil society. The committee will be presided by the Permanent Secretary of the Ministry of Finance, Budget and Privatization with the Permanent Executive Secretary of the National Social Protection Commission as Vice President. The secretary will be the Project coordinator. Members will include: (a) from Government: the National Director for Social Assistance and National Solidarity at MDPHASG, high-level representatives from the Ministry of Public Health and Fight against AIDS, the Ministry of Interior and Patriotic Awareness Training, the Ministry of External Relations and International Cooperation, the Ministry of Education, Higher Education and Research, the Ministry of Agriculture and Livestock, and the Ministry for Communal Development, as well as a representative of the provincial governors; (b) from civil society: a representative from the Burundian Red Cross. 2. The steering committee is in charge of the general supervision of the project and ensures that the Project’s activities are implemented to reach the Project’s objectives and in accordance with the dispositions in the Financing Agreement and the Operations Manual. The steering committee approves annual budgets and work plans, examines annual reports and decides about main corrective measures as necessary. The steering committee meets at least twice a year on invitation of its president or vice-president (Ministerial Order 540/225.01/1986 of October 10, 2016). 3. The Technical Committee for the National Social Protection Commission (CT-CNPS) is presided by the Minister for Human Rights, Social Affairs and Gender and the UNICEF resident representative is the vice-president in his/her role as coordinator of the donor social protection group. Its members include representatives from the 11 ministries members of the Commission25 as well as representatives from religious groups, employers, unions and health insurance mutuelles. The committee is a high-level committee in charge of validating strategic plans for SP, approving technical documents and studies related to social protection issues, monitoring the key programs under the PNPS, supervising the SEP and the management of the SP support fund (Fonds d’ Appui à la Protection Sociale). The committee includes four thematic groups for social insurance, social assistance, access to income and financing for SP. The Social Assistance thematic group includes representatives from the Ministries of Civil Service, National Solidarity, Communal Development, Finance and Planning, Health, and Committee on Aid Coordination as well as representatives from the main religious groups. The technical committee acted as the Steering Committee for the 25 The committee includes members from the Presidency, the 1st and 2nd Vice-Presidencies, the Ministry of Civil Service, Labor and Employment, the Ministry of Human Rights, Social Affairs and Gender, the Ministry of Health and Fight against AIDS, the Ministry of Interior, the Ministry of Communal Development, the Ministry of Finance and Development Planning, the Ministry of Youth, Sports and Culture, the Ministry of Agriculture and Livestock, representatives from the Catholic, Protestant and Islamic religions, representatives from the Burundian Association of Employers, from the Confederation of Unions, Health Insurance “Mutuelles” , representatives from the Aid Coordination Commission and the Permanent Secretary for Economic and Social Reforms. 52 National Social Protection Strategy development. 4. The Permanent Executive Secretariat of the CNPS will facilitate the technical coordination between the Project Implementation Unit and other key members of the CNPS. The SEP was created in January 2014, from the Social Protection direction of the then Ministry for Civil Service, Labor and Social Security and is now under the authority “tutelle” of the Ministry for Human Rights, Social Affairs and Gender. The SEP coordinates activities to extend and strengthen social protection; it executes action plans to reinforce the promotion and regulation of SP, it coordinates with its provincial and communal representations, monitors and evaluates SP programs, provides secretary functions to the Technical Committee and ensures the technical implementation of the Committee and Commission decisions. The SEP has three directorates: (a) Technical Directorate to promote Social Protection, (b) Directorate for Control, Monitoring and Evaluation of SP systems, and (c) Directorate for Administration and Finance. 5. As part of its coordination mandate to extend and strengthen social protection, the SEP will facilitate technical coordination between the project activities and those of other government agents within the CNPS at central level and the CPPS at provincial level. At central level, such partners include the National Statistical Institute (ISTEEBU), the National Aid Coordination Committee (CNCA) and key ministries in the social and local development sectors 6. The National Directorate of Social Assistance and National Solidarity in the Ministry of Human Rights, Social Affairs and Gender will support the implementation of the core local activities of the cash transfer program (Component 1), based on its experience with humanitarian assistance to vulnerable groups (persons with disabilities, orphans, street children, vulnerable elderly and victims of gender-based violence). It will coordinate with the commune level Commune Family Development Centers (CFDC), which depend on the Ministry’s cabinet. In particular, social workers from the CDFC will monitor the implementation of the complementary promotion activities. 7. In a first phase, a Project Implementation Unit will be responsible for the overall implementation of the project activities. This unit will under the authority of the MDPHASG. It will be headed by a Project Coordinator, reporting to the SEP/CNPS and the General Director for Social Assistance and National Solidarity and staffed by a small management team of experienced technicians. The unit will be co-located with the SEP. The Project Coordinator will manage the Project designated account. To help accelerate the activities in the first phase, the Project Implementation Unit will initially be responsible for the technical and financial implementation of project activities, including procurement, financial management, M&E, mitigation of potential negative social and environmental impacts, and communication about project implementation and results. At mid-term review, the project team will evaluate the possibility of a transition strategy to embed the implementation unit within the SEP/CNPS, based on the results of the capacity building activities supported in Component 2. The unit will contract the development of the MIS, the survey firm(s) for household registration and impact evaluation, the financial institutions in charge of delivering the payments to the beneficiary households, and the NGOs in charge of delivering the complementary activities. In Kirundo, where Concern has been operating a pilot cash transfer for 2,000 households in 2 communes, local implementation could be delegated to Concern. 53 8. The Project Implementation Unit also includes staff dedicated to the local implementation of the project at the provincial and commune-levels in the selected provinces and communes in coordination with provincial and communal deconcentrated structures of the SEP/CNPS and the MDPHASG. At the provincial level, these include the regional representatives of the SEP and the CDFCs and at commune-level, the social workers and the community health agents. At the local level, the Project will benefit from support from the administrative structure (provincial governorates (“Gouverneur”), communes (“Administrateur Communal”) and Collines (“Chefs de colline”) and existing representatives from sectoral ministries. It will rely on existing coordination mechanisms such as the Provincial Committee for Social Protection (“Comité Provincial de Protection Sociale”) and the Colline Development Committees (“Comité de Développement Collinaire”). It will also rely on civil society; community organizations and religious groups and existing committees or monitoring/conflict-resolution structures such as (Comité de Protection de l’Enfance, imboneza (mediators for domestic violence), Red Cross volunteers). Figure A3.1. Implementation arrangements Implementation Coordination Steering CNPS Committee Technical MDPHASG Directorate Committee Social Implementati SEP-CNPS and Cabinet Assistance on Support thematic Unit groups Province Provincial CDFC focal point RP CPPS (4) Commune Social Communal worker focal point (16) Colline Service providers 54 9. A project implementation manual, covering all aspects of the project implementation will be prepared. In anticipation of any possible capacity constraints, the Project Implementation Manual (PIM) will be written and adopted and a series of training workshops will be held to familiarize management and relevant staff with their contents and use. The PIM will include two main sections: (i) the administrative, procurement, financial and accounting aspects of project implementation; and (ii) the technical implementation of the cash transfers, complementary measures and key delivery mechanisms (this part of the PIM will be referred to as the operational manual). In addition, the operational manual will include several annexed detailing field procedures once they have been developed and tested and are ready for implementation. (a) The government will prepare and submit a draft PIM including the administrative, procurement, financial and accounting section by the time of project effectiveness. It will describe clearly the procedures governing the operation of the Project Implementation Unit and its various project activities. It will also detail the responsibilities of the Project Implementation Unit and its staff at the central, provincial and commune-level. (b) The technical component of the PIM (also referred to as the operational manual) will be developed during the first six months of the project implementation (legal covenant). It will describe clearly such implementation aspects as beneficiary targeting, cash transfer procedures, complementary measures, delivery mechanisms, accounting documentation, and information flows throughout the project implementation cycle. No disbursements for cash transfers will be made before the technical component of the PIM is prepared and adopted. 10. In addition to the measures mentioned above, the Project Implementation Unit will outsource several activities to local consulting firms, NGOs to facilitate the implementation of the projects. These arrangements may include: (a) Contractual arrangements with the National Statistical Institute (Institut de Statistiques et Études Économiques du Burundi, ISTEEBU) or other data collection firms, to conduct household-level surveys for the targeting and impact evaluation and individual-level surveys for the beneficiary surveys; (b) Contracts with payment agencies to carry out transfer payments; (c) Contracts with NGOs or other entities with the required technical expertise, qualifications, and experience to facilitate the work of the targeting committees, the implementation of the complementary measures and the implementation of the grievance redress mechanisms. Financial Management 11. Financial Management Assessment. A financial management (FM) assessment of the project to be implemented by the SEP/CNPS and the Ministry of Human Rights, Social Affairs and Gender was carried out to determine: (a) whether the implementing entities have adequate financial management arrangements to ensure that project funds will be used for purposes intended, in an efficient and economical way; (b) project financial reports will be prepared in an accurate, reliable and timely manner; and (c) the entity’s assets will be safeguarded. Those assessment were carried out in accordance with the Bank Directive: Financial Management Manual For World Bank Investment Project Financing Operations issued February 4, 2015 and effective from March 1, 2015; and the Bank Guidance: Financial Management in World Bank Investment Project Financing Operations Issued and Effective February 24, 2015. 55 12. The assessment indicates that those entities do not have capacity and experience in implementing IDA-financed lending operations. There are some weaknesses and risks mainly in the areas of staffing arrangements and capacity, accounting systems, both internal and external auditing, and reporting arrangements. Proposed mitigation measures have been incorporated into the design of the financial management arrangements with the aim of reducing these risks. Mitigation measures include the establishment of a Project Implementation Unit within project launch; the financial management system of that unit will be supported by the following measures: (i) recruitment of a Financial Management specialist with qualification acceptable to the Bank; (ii) a computerized accounting system in the implementation unit; and (iii) the development of Project Implementation Manual (PIM), including an administrative and financial component. 13. Risk Assessment and Mitigation Measures. The main financial risks and mitigating measures of this project are listed in Table A3.1 below. 14. Implementing Entity. The PIU will have the overall responsibility of project financial management and procurement aspects, including budgeting, disbursement, book keeping, reporting, supervision, management of the designated accounts and auditing. It will be strengthened with the recruitment of a fiduciary team well versed in World Bank fiduciary procedures and through a competitive process. The fiduciary team will be composed of at least one financial management specialist and one accountant. The World Bank will have the right to review the recruitment process and issue a non-objection before formal appointment. The overall selection should be finalized before Project effectiveness as a condition for effectiveness. 15. Planning and Budgeting arrangements. The Annual Work Plan and Budget (AWPB) along with the disbursement forecast will be elaborated by the PIU. It will be submitted to the Project Steering Committee for approval, and thereafter to IDA for no objection no later than October 30 of the year preceding the year the work plan should be implemented. The fiduciary unit of the PIU will monitor its execution in accordance with the budgeting procedures specified in the Project Implementation Manual. The budgeting system should forecast for each fiscal year the origin and use of funds under the project. Only budgeted expenditures would be committed and incurred so as to ensure that the resources are used within the agreed upon allocations and for the intended purposes. The quarterly IFRs will be used to monitor the execution of the AWPB. 16. Accounting arrangements. The current accounting standards in use in Burundi for ongoing Bank-financed projects will be applicable. An integrated financial and accounting system will be put in place and used by the fiduciary unit. The Project code and chart of accounts will be developed to meet the specific needs of the project and documented in the Project Implementation Manual. The accounting system is expected to include a general ledger, auxiliary ledgers, general balance, cash record, and fixed assets record. These ledgers and records should be maintained with the support of financial management software that should be operational no later than three (3) months after Project effectiveness. Financial management staff at the PIU should be trained in the use of the software by the same date. 17. Internal Control and Internal Auditing arrangements. The Administrative and Accounting Procedures Manual will provide a clear description of the approval and authorization processes in respect of the rule of segregation of duties. The Bank will pay attention to the adequacy of internal control during its supervision missions. 56 Table A3.1: Risk Assessment and mitigation measures Risk Risk Risk Mitigating Measures Incorporated into Conditions for Residual Rating Project Design Effectiveness Risk (Y/N) Inherent risk S S Country level. Burundi is still a high The Government is committed to a reform risk country from the fiduciary program that includes the strengthening of the perspective. The PER, the PEFA 2008 PFM. This project will enhance Government’s N H and 2014 as well as the UCS reports H institutional capacity in adopting and using IDA outlined PFM weaknesses at both central FM procedures. and decentralized levels. The PIU finance team will be composed of a financial management specialist and an Entity level. Financial and regulation accountant with extensive experience with WB rules which govern the fiduciary S fiduciary procedures. Implementing a FM Y M arrangement will be implemented by the procedures manual (which will be part of the PIU. Project implementation manual) will also help to mitigate internal control weaknesses. The recruitment of a Project Coordinator, a Field Operation Manager, a Financial Management Specialist, a Procurement Specialist, M&E Project level. The project will be Specialist and a Communication Specialist will implemented by a PIU team to be be completed within project launch. recruited. S Training on fiduciary procedures will be N S Ensuring funds are used for purposes conducted for all FM staff throughout the life of intended, as well as a fluid and efficient the project. Three FM supervision supports will flow of funds might be challenging. be conducted the first 12 months following the effectiveness. Control Risk S S The project Financial Procedures Manual (which will be part of the Project implementation Budgeting. The AWPB will be prepared manual) will define the arrangements for by the PIU and approved by the Project budgeting, budgetary control and the Steering Committee based on the policy S requirements for budgeting revisions. Annual N M guideline. The risks will be that there are detailed disbursement forecasts and budgets will many actors involved and this may delay be required. IFR will provide information on the process. budgetary control and analysis of variances between actual and budget. The current accounting standards in use in Burundi for ongoing Bank-financed projects will be applicable. Accounting procedures will be documented in the manual of procedures (which Accounting. This project will use an will be part of the Project implementation accounting software which complies manual). The FM functions will be carried out with WB procedures. The risks will be by qualified consultants (individuals) composed the following: Poor policies and S N S of one Financial Management Specialist and one procedures, and delay in keeping reliable accountant to be recruited on competitive basis. and auditable accounting records. A software will be customized to take into consideration the need for this new project. Training will be provided to the Financial team on the use of international accounting procedures as well as the project’s software. Internal Control. Internal control system may be weak due to the fact that A project financial management Manual of some of the project activities will have Procedures (which will be part of the Project technical control by the SEP/CNPS and S Implementation Manual) and training on the use N M the Direction of National Solidarity in of the manual will help to mitigate the risk the MHRSAG. Insufficient safeguards relating to internal control and controls may result in improper use 57 of funds and impact the implementation. of the project Funds Flow. Due to the nature of the activities which involve a huge number of beneficiaries, funds may not be timely released to allow the implementation of activities. Training on the new disbursement procedures particularly e-disbursement and report based The absence of financial institution in disbursement will be provided. Also regular S N S some project implementation areas could meetings between the PIU, the SEP/CNPS and delay the payment to the beneficiaries. the Gov. technical focal points and FMS when required. The country may not be familiar with customized disbursement method to the project beneficiaries under component 2. A computerized accounting system will be used. In addition, IFR and financial statements formats and contents were agreed upon during Financial Reporting. Inaccuracy and project negotiations and IFRs template will be delays in submission of IFR to the WB S N M included in the FM manual of procedures. The due to weak capacity of the FM team. IDA team will follow-up closely on FM reporting, including training at the outset of the project in IFRs. IFR will be produced on a quarterly basis, reviewed to improve its quality. Auditing. The risk would be the delay in submission of audit report or qualified Bank FMS will provide support if required opinion and delays in the S N M implementation of audit reports The audit firms will be recruited before the end recommendations. of the first year of the project. The TOR of the external auditor will comprise a specific chapter on corruption auditing. In Governance and Accountability. addition, (i) FM procedures manual will be Possibility of circumventing the internal approved before project effectiveness; (ii) robust control, and abuse of administrative M FM arrangements will be established with N M positions are potential risks, mis- quarterly IFR including budget execution and procurement etc., is a critical issue. monitoring; and (iii) Measures will be taken to improve transparency such as providing information on the project status to the public. OVERALL FM RISK H S Disbursements 18. Disbursement arrangements. The report based disbursement method will be used by the project. Disbursements will be made on a quarterly basis based on a six month cash flow forecast supported by a work plan that is agreed between the PIU and the Bank (IDA) Task Team Leader. Other methods of disbursement that can be used by the project include direct payments to a third party upon the PIU’s request; special commitments e.g. letters of credit; and reimbursements to Government for expenditures incurred under the project. Further details about disbursements to the project will be included in the disbursement letter as well as the disbursement guidelines. The Bank will conduct disbursement training for PIU staff on a need basis based on their performance and/or when new staff are recruited. An advance will be made into a two USD Designated accounts opened at the Burundi Central Bank (BRB) immediately after credit effectiveness. To facilitate transactions with beneficiaries, two subsidiary accounts in BIF will be opened in a commercial bank acceptable to the bank. The PIU should submit the initial withdrawal application request together with six months program’s cash forecast. 58 19. Eligible Expenditures. Table A.3.2 specifies the categories of eligible expenditures that may be financed out of the proceeds of the project, the allocations of the amounts of the financing to each category, and the percentage of expenditures to be financed for eligible expenditures in each category Table A3.2. Eligible Expenditures Category Amount of the Grant Percentage of Expenditures to Allocated (expressed be Financed in SDR) (inclusive of Taxes) (1) (a) First Cash Transfer 100% of amounts paid by the installments under Part A.1 Recipient under the Cash of the Project 7,100,000 Transfer (b) Second Cash Transfer 100% of amounts paid by the installments under Part A.1 7,100,000 Recipient under the Cash of the Project Transfer (2) Goods, non-consulting services, and consultants’ services, Training and Operating 15,000,000 100% Costs for the Project TOTAL AMOUNT 29,200,000 20. Banking Arrangements. Two separate Designated Accounts (DA) for the project will be opened in the Banque de la République du Burundi (BRB) on terms and conditions acceptable to IDA. Two subsidiary accounts in BIF will be opened in Commercial Bank acceptable to the bank. Those two accounts will be under the fiduciary responsibility of the PIU. Those accounts will be used for all eligible payments financed by the credit as indicated in the specific terms and conditions of the Financing Agreement. 21. Financial Reporting arrangements. The quarterly Interim Financial Reports (IFRs) will be prepared at the end of each quarter and submitted to the Bank not later than 45 days after the end of the quarter. The format and content of the IFRs were discussed and agreed with the government. The IFRs will include Sources and Uses of Funds Statement, Uses of Funds by Project Activity/Component, Designated Account Activity Statement and Physical Progress (Output Monitoring) Report. 22. To support the continued use of report-based disbursement TPA will be required to submit: (a) Interim Financial Report (IFR). (b) Designated Account (DA) Activity Statement. (c) DA and Project bank accounts statements. (d) Bank reconciliations for both the DA and project bank accounts (e) Summary Statement of DA Expenditures for Contracts subject to Prior Review. (f) Summary Statement of DA Expenditures for contracts not subject to Prior Review. 59 Figure A3.2. Flow of funds IDA grant Designated Account Designated Account in US $ (In BRB) in US $ (In BRB) Subsidiary Account in BIF Subsidiary Account in BIF (ISU) in a commercial bank (ISU) in a commercial bank for the Cash Transfers Transfers to the payment Transactions to be financed in BIF agents (in BIF) or other currencies (US$, Euro) Cash transfer beneficiaries (in BIF) 23. The financial statements will be prepared in accordance with International Financial Reporting Standards (IFRS). The IDA financing Agreement will require the submission of audited financial statements to the Bank within six months after the financial year end. These Financial Statements will comprise of: (a) A Statement of Sources and Uses of Funds / Cash Receipts and Payments which recognizes all cash receipts, cash payments and cash balances controlled by the entity; and separately identifies payments by third parties on behalf of the entity. (b) A Statement of Affairs/ Balance Sheet as at the end of the financial year showing all the assets and liabilities of the project. (c) The Accounting Policies Adopted and Explanatory Notes. The explanatory notes should be presented in a systematic manner with items on the Statement of Cash Receipts and Payments being cross referenced to any related information in the notes. Examples of this information include a summary of fixed assets by category of assets, and a summary of SOE Withdrawal Schedule, listing individual withdrawal applications; and 60 (d) A Management Assertion that Bank funds have been expended in accordance with the intended purposes as specified in the relevant World Bank legal agreement. 24. Indicative formats of these statements will be developed in accordance with IDA requirements and agreed with the Country Financial Management Specialist. 25. External auditing arrangements. A qualified, experienced, and independent external auditor will be recruited on approved terms of reference three months after effectiveness. The external audit will be carried out according to International Standards on Auditing (ISAs) and will cover all aspects of project activities implemented and include verification of expenditures eligibility and physical verification of goods and services acquired. The report will also include specific controls such as compliance with procurement procedures and financial reporting requirements and consistency between financial statements and management reports and field visits (e.g. physical verification). The audit period will be on annual basis and the reports including the project financial statements submitted to IDA and to the auditors six months after the end of each fiscal year. The project will comply with the WB disclosure policy of audit reports (e.g. making them publicly available, promptly after receipt of all final financial audit reports (including qualified audit reports) and place the information provided on the official website within one month of the report being accepted as final by the team. 26. Action Plan. The following actions need to be taken in order to enhance the financial management arrangements for the Project: Responsible Nº Action Due Date Entity 1. Prepare a project manual of procedures (which will be part By effectiveness PIU/WB of the Project implementation manual) which will define FM and accounting procedures to run the project. 2. Recruitment of one financial management expert and one By effectiveness PIU/WB accountant 3. Purchase and installation of a new software which integrates all the functionality required to manage a World By effectiveness PIU Bank project. Train the fiduciary staff on the use of that software. 4. Recruitment of internal auditor with competence and Four months after PIU experience satisfactory to the Bank effectiveness 5. Selection an external auditor with competence and Four months after PIU experience satisfactory to the Bank effectiveness 27. Implementation Support Plan. Implementation support missions will be conducted over the project’s lifetime. The project will be supervised on a risk-based approach. It will comprise inter alia, the review of audit reports and IFRs, and advice to the task team on all FM issues. Based on the current risk assessment which is substantial, the project will be supervised at least twice a year and this may be adjusted when the need arises. However, three FM supervision supports will be conducted the first 12 months following the effectiveness. The ISR will include a FM rating of the project. An implementation support mission will be carried out before effectiveness to ensure the project readiness. To the extent possible, mixed on-site supervision missions will be undertaken with procurement monitoring and evaluation and disbursement colleagues. Based on the outcome of the FM risk assessment, the following implementation support plan is proposed with the objectives of ensuring the project maintains satisfactory financial management systems throughout the project’s life: 61 FM Activity Frequency Desk reviews Interim financial reports review Quarterly Audit report review of the program Annually Review of other relevant information such as interim internal Continuous as they become available control systems reports. On site visits Review of overall operation of the FM system Semi-Annually (Implementation Support Mission) Monitoring of actions taken on issues highlighted in audit As needed but at least during each implementation reports, auditors’ management letters, internal audit and other support mission reports Transaction reviews (if needed) As needed Capacity building support FM training sessions by World Bank FM team At project launch and thereafter as needed 28. Financial Covenants. There are no FM effectiveness conditions. The Borrower shall establish and maintain a financial management system including records, accounts and preparation of related financial statements in accordance with accounting standards acceptable to the Bank. The Financial Statements will be audited in accordance with international auditing standards. The Audited Financial Statements for each period shall be furnished to the Association not later than six (6) months after the end of the project fiscal year. The Borrower shall prepare and furnish to the Association not later than 45 days after the end of each calendar quarter, interim un-audited financial reports for the Project, in form and substance satisfactory to the Association. The Borrower will be compliant with all the rules and procedures required for withdrawals from the Designated Accounts of the project. Procurement 29. Processes and procedures. Procurement under the Project will be carried out in accordance with the World Bank’s (i) “Guidelines: Procurement of Goods, Works, and Non-Consulting Services under IBRD Loans and IDA grants & Grants by World Bank Borrowers” dated January 2011, revised in July 2014 (ii) “Guidelines: Selection and Employment of Consultants under IBRD Loans and IDA grants & Grants by World Bank Borrowers” dated January 2011, revised in July 2014 and (iii) “Guidelines on Preventing and Combating Fraud and Corruption in Projects Financed by IBRD Loans and IDA grants and Grants”, dated October 15, 2006, as revised in January 2011, as well as the provisions stipulated in the Financing Agreement. 30. The general description of various items under different expenditure categories is presented below under Procurement Arrangements. For each contract, the different procurement methods or consultant selection methods, the need for prequalification, estimated costs, prior review requirements, and time frame are agreed between the Borrower and the Bank in the Procurement Plan. The Procurement Plan will be updated at least annually, or as required, to reflect the actual project implementation needs and improvements in institutional capacity. 31. In addition to prior review of contracts by the Bank as indicated in the Procurement Plan, the procurement capacity assessment recommends at least one supervision mission each year to carry out post-review of procurement actions. The percentage of the procurements subject to post- reviews and technical reviews will be decided on case-by-case by the World Bank mission. 32. Advertising procedures. In order to get the broadest attention from eligible bidders and 62 consultants, a General Procurement Notice (GPN) will be prepared by the implementing entity and published in the United Nations Development Business online (UNDB online), on the World Bank’s external website and in at least one newspaper of national circulation in the Borrower’s country, or in the official gazette, or a widely used website or electronic portal with free national and international access. The Borrower will keep record of the responses received from potential bidders/consultants interested in the contracts and send them the Specific Procurement Notices. 33. Specific Procurement Notices for all goods and non-consulting services to be procured under International Competitive Bidding (ICB) will be published in the UNDB online, on the Bank’s external website, and in at least one newspaper of national circulation in the Borrower’s country, or in the official gazette, or a widely used website or electronic portal with free national and international access. Specific Procurement Notices (SPN) for goods and non-consulting services to be procured using National Competitive Bidding (NCB) will be published in at least one newspaper of national circulation in the Borrower’s country. 34. Procurement documents. Procurement would be carried out using the Bank’s Standard Bidding Documents (SBD) for all International Competitive Bidding (ICB) for goods and Standard Request for Proposal (RFP) for the selection of consultants through competitive procedures. For National Competitive Bidding (NCB), the Borrower could build on the Bank’s SBD and submit a sample form of bidding documents to the Bank for prior review. This type of document would be used throughout the project implementation once agreed upon. 35. Fraud and Corruption. All procurement entities as well as bidders and service providers (i.e., suppliers, service providers, and consultants) shall observe the highest standard of ethics during the procurement and execution of contracts financed under the Project in accordance with paragraphs 1.16 and 1.17 (Fraud and Corruption) of the Procurement Guidelines and paragraph 1.23 and 1.24 (Fraud and Corruption) of the Consultants Guidelines, and the “Guidelines on Preventing and Combating Fraud and Corruption in Projects Financed by IBRD Loans and IDA grants and Grants”, dated October 15, 2006 and revised in January 2011. Procurement arrangements 36. Procurement of Works. No civil works contracts are foreseen under the proposed Project. 37. Procurement of Goods. Goods to be procured under the Project will include vehicles, motorbikes, computers and accessories, printers, photocopiers, mobile phones and other IT equipment, software and licenses, furniture, and stationary for training, recording equipment for communication. Contracts with an amount equal or above US$3,000,000 equivalent shall be procured through ICB. Goods orders shall be grouped into larger contracts wherever possible to achieve greater economies of scale. Contracts with an amount lower than US$3,000,000, but equal to or above US$50,000 may be procured through NCB. Contracts with an amount below US$50,000 may be procured using shopping procedures in accordance with paragraph 3.5 of the Procurement Guidelines and based on a model request for quotations satisfactory to the Bank. Shopping consists of the comparison of at least three price quotations in response to a written request. Direct contracting may be used in exceptional circumstances only with the prior approval of the Bank regardless of the amount, in accordance with paragraph 3.7 and 3.8 of the Procurement Guidelines. 38. Procurement of non-consulting services. Procurement of non-consulting services, such as services for organizing workshops, training, data collection, transport services and maintenance 63 of office equipment, will follow procurement procedures similar to those stipulated for the procurement of goods, depending on their nature. 39. Selection of Consultants. Services of both national and international consultants will be required under the Project, as related to studies, advisory services, surveys, and process and financial auditing. Selection of consultants will be carried out in compliance with the Consultant Guidelines. The provisions vary for consulting services provided by firms and individual consultants as follows: (a) Firm. Selection of consulting firms will include launching a Request for Expressions of Interest, preparing short-lists and issuing a Request for Proposal using Banks’ standard formats, when and as required by the Bank’s Guidelines. The selection method shall be chosen among the following: Quality and Cost Based Selection (QCBS) whenever possible; Quality Based Selection (QBS); Selection under a Fixed Budget (FBS); Least Cost Selection (LCS); Single Source Selection (SSS) as appropriate; Consultant’s Qualifications (CQS) for consultancy services estimated to cost less than US$200,000 equivalent. The shortlist of firms for assignments estimated to cost less than US$200,000 equivalent per contract may be composed entirely of national firms in accordance with the provisions of paragraph 2.7 of the Consultant Guidelines, provided that a sufficient number of qualified national firms are available and no foreign consultant desiring to participate is barred. (b) Individual Consultants. Individual consultants will be selected by comparing qualifications of at least three candidates and hired in accordance with the provisions of Section V. of the Consultant Guidelines. 40. Single-Source Selection (SSS) may be used for consulting assignments that meet the requirements of paragraphs 3.8 - 3.11 of the Consultant Guidelines and will always require Bank’s prior review regardless of the amount. Procedures of Selection of Individual Consultants (IC) will apply to assignments which meet the requirements of paragraphs 5.1 and 5.6 of the Consultant Guidelines. 41. Procurement from United Nations Agencies. There may be situations in which procurement directly from agencies of the United Nations (UN), following their own procurement procedures may be the most appropriate methods. In such circumstances the Recipient shall submit to the Bank for its no objection a full justification and the draft form of agreement with UN agency. 42. All terms of reference (ToRs) for the selection of firms and individual consultants, regardless of the estimated value of the assignment, will be subject to Bank review and clearance. 43. The recruitment of civil servants as individual consultants or as part of the team of consulting firms will strictly abide by the provisions of paragraphs 1.9 to 1.13 of the Consultants Guidelines. 44. Workshops, Seminars and Conferences. Training activities would comprise workshops and training, based on individual needs, as well as group requirements, on-the-job training, and hiring consultants for developing training materials and conducting training. Selection of consultants for training services follows the requirements for selection of consultants above. All training and workshop activities (other than consulting services) would be carried out on the basis of approved Annual Work Plans / Training Plans that would identify the general framework of training activities for the year, including: (i) the type of training or workshop; (ii) the personnel to 64 be trained; (iii) the institutions which would conduct the training and reason for selection of this particular institution; (iv) the justification for the training, how it would lead to effective performance and implementation of the project and or sector; (v) the duration of the proposed training; and (vi) the cost estimate of the training. Report by the trainee(s), including completion certificate/diploma upon completion of training, shall be provided to the Project Coordinator and will be kept as parts of the records, and will be shared with the Bank if required. 45. Operating Costs. Operating Costs are incremental expenses arising under the Project and based on Annual Work Plans and Budgets approved by the Bank pursuant to the Financing Agreements. They are incurred based on eligible expenses as defined in the Financing Agreement and cannot include salaries of the Borrower’s civil and public servants. The procedures for managing these expenditures will follow the procedures agreed in the implementation manual, acceptable to the Bank. 46. Record Keeping and Filing. To ensure transparency in the procurement process, the Project will document all steps in the process for each contract. These procurement documents will be filed in a manner that will facilitate retrieval in the event of audits or reviews. Procurement record-keeping will be the responsibility of the Procurement Unit. Assessment of Procurement Capacity of the Implementing Agency and Risks 47. Procurement activities will be managed by the PIU which will have overall responsibility in carrying the following activities: (i) managing the overall procurement activities, and ensuring compliance with the procurement process described in the relevant manuals; (ii) preparing and updating the procurement plan; (iii) preparing bidding documents, draft Request For Proposals (RFPs), evaluation reports, and contracts in compliance with World Bank procedures; and (iv) seeking and obtaining approval of national entities and of IDA on procurement documents as required. 48. Procurement risks. The table below summarizes procurement risks and their proposed mitigation measures. The overall project risk for procurement is High and would be updated to Substantial with the implementation of mitigations measures. 49. Frequency of procurement supervision. In addition to the prior-review supervision conducted from Bank offices, the Bank will carry out annual supervision missions to conduct post review of procurement actions and contracts under prior review thresholds on basis of a sample of about 20 percent of contracts within review period. 65 Table A.3.3: Procurement Risk assessment and Mitigation Measures Risk Action Completion Responsible Date Entity Accountability for Procurement Ensure the procurement decision Three months Decisions in the Implementing making is fully covered in the after Agency or Agencies. Manual of the PIU and effectiveness GoB available/known to staff. Lack of procurement staff familiar Hire a procurement officer Before with WB guidelines. conversant with modern procurement effectiveness procedures in general and specifically GoB WB guidelines. Involve members of procurement unit of line Ministry in the procurement activities for their capacity building, including training. Internal manuals and clarity of the Prepare manual as part of project Three months procurement process. preparation. after effectiveness GoB Procurement planning. Develop and maintain procurement Throughout planning that ensure competition in project line with market of procured items. implementation GoB Record keeping and records Establish a system of Three months management. records/procurement documents after keeping. effectiveness GoB 50. Prior review threshold. At preparation, the prior review thresholds are as follows: a) Procurement of Goods and Non-consulting services Expenditure Procurement method Threshold (US$) Contracts subject to prior category review 1. Goods ICB ≥ US$3 million ≥1.5 million NCB < US$3 million ≥1.5 million Shopping < US$50,000 None Direct Contracting All values All contracts (≥US$100,000 to be processed through procys/step) 2. Non- NCB < US$0.5 million None consulting Services Shopping < US$50,000 None Direct Contracting All values All contracts (≥US$100,000 to be processed through procys/step) 66 b) Selection of Consultants Expenditure category Contract value Procurement method Contracts subject to (Threshold) US$ prior review (a) Firms ≥ US$200,000 QCBS, QBS, FBS, LCS ≥ US$ 500,000- All contracts < US$200,000 QCBS, CQS, LCS, QBS, None (Selected FBS, contracts as indicated in the procurement plan) All values SSS All contracts(≥US$100,000 $ to be processed through procys/step) (b) Individuals ≥ US$200,000 Three CVs All contracts