Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00005807 IMPLEMENTATION COMPLETION AND RESULTS REPORT ON AND A GRANT IN THE AMOUNT OF US$125 MILLION (CO-FINANCING FROM THE AFGHANISTAN RECONSTRUCTION TRUST FUND) TO THE Islamic Republic of Afghanistan FOR THE COVID-19 Relief Effort for Afghan Communities and Households (REACH) Project (P174119) October 15, 2022 Social Sustainability And Inclusion Global Practice South Asia Region CURRENCY EQUIVALENTS (Exchange Rate Effective July 31, 2022) Currency Unit = AFN AFN 89.71= US$ 1 US$ 1.3236 = SDR 1 FISCAL YEAR December 23 – December 22 Regional Vice President: Martin Raiser Country Director: Melinda Good Regional Director: John A. Roome Practice Manager: Kevin A Tomlinson Task Team Leader(s): Susan Wong, Andras Bodor ICR Main Contributor: Kosuke Anan ABBREVIATIONS AND ACRONYMS AFN Afghani (Currency of Afghanistan) AGE Anti-Government Element AIS Activity Initiation Summary ARTF Afghanistan Reconstruction Trust Fund CCAP Citizens’ Charter Afghanistan Project CDC Community Development Council CDD Community Driven Development COVID-19 Corona Virus Disease of 2019 CPF Country Partnership Framework ENETAWF Early Warning, Action, and Finance Project ESMP Environmental and Social Management Plan ESS Environmental and Social Standard EZ-Kar Eshteghal Zaiee-Karmondena Project FM Financial Management FP Facilitating Partner FSAC United Nation’s Food Security and Agriculture Cluster FY Fiscal Year GA Gozar Assembly GDP Gross Domestic Product GoIRA Government of the Islamic Republic of Afghanistan GRM Grievance Redress Mechanism HCT Humanitarian Country Team HRP Humanitarian Response Plan IA Implementing Agency ICR Implementation Completion Results Report IDA International Development Association IDLG Independent Directorate of Local Governance IDP Involuntarily Displaced People IOM International Organization for Migration IPC Integrated Food Security Phase Classification IPF Investment Project Financing ISR Implementation Status Report ITA Interim Taliban Administration KM Kabul Municipality KMDP Kabul Municipal Development Project MCCG Maintenance and Construction Cash Grant MIS Management Information System MMO Mobile Money Operator MOPH Ministry of Public Health MoU Memorandum of Understanding MRRD Ministry of Rural Rehabilitation and Development M&E Monitoring and Evaluation NATO North Atlantic Treaty Organization NGO Non-Governmental Organization NSP National Solidarity Program PCC Provincial Capital City PDO Project Development Objective PIU Project Implementation Unit PMU Project Management Unit PLR Performance and Learning Review REACH The COVID-19 Relief Effort for Afghan Communities and Households Project RVP Regional Vice President of the World Bank SCD Systematic Country Diagnostic SEP Stakeholder Engagement Plan SIG Social Inclusion Grant SSN PIN Social Safety Net Population in Need TPMA Third Party Monitoring Agent UN United Nations UNAMA United Nations Assistance Mission in Afghanistan UNDP United Nations Development Programme UNICEF United Nations Children’s Emergency Fund U. S. United States of America WFP World Food Programme TABLE OF CONTENTS DATA SHEET .......................................................................................................................... 1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES ....................................................... 5 A. CONTEXT AT APPRAISAL .........................................................................................................5 II. OUTCOME .................................................................................................................... 11 A. RELEVANCE OF PDOs ............................................................................................................11 B. ACHIEVEMENT OF PDOs (EFFICACY) ......................................................................................13 C. EFFICIENCY ...........................................................................................................................16 D. JUSTIFICATION OF OVERALL OUTCOME RATING ....................................................................17 E. OTHER OUTCOMES AND IMPACTS (IF ANY)............................................................................18 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ................................ 18 A. KEY FACTORS DURING PREPARATION ...................................................................................18 B. KEY FACTORS DURING IMPLEMENTATION .............................................................................21 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME .. 24 A. QUALITY OF MONITORING AND EVALUATION (M&E) ............................................................24 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE .....................................................25 C. BANK PERFORMANCE ...........................................................................................................26 D. RISK TO DEVELOPMENT OUTCOME .......................................................................................29 V. LESSONS AND RECOMMENDATIONS ............................................................................. 29 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ........................................................... 31 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION ......................... 38 ANNEX 3. PROJECT COST BY COMPONENT ........................................................................... 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name COVID-19 Relief Effort for Afghan Communities and P174119 Households (REACH) Country Financing Instrument Afghanistan Investment Project Financing Original EA Category Revised EA Category Organizations Borrower Implementing Agency Independent Directorate of Local Governance (IDLG), Islamic Republic of Afghanistan Kabul Municipality, Ministry of Rural Rehabilitation and Development (MRRD) Project Development Objective (PDO) Original PDO To provide emergency support to selected households through communities in project areas during the COVID-19 outbreak. Page 1 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 125,000,000 89,752,870 89,381,333 TF-B3543 155,000,000 155,000,000 91,573,793 IDA-D7110 Total 280,000,000 244,752,870 180,955,126 Non-World Bank Financing 0 0 0 Total 0 0 0 Total Project Cost 280,000,000 244,752,870 180,955,126 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 04-Aug-2020 14-Sep-2020 31-Dec-2021 31-Dec-2021 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions KEY RATINGS Outcome Bank Performance M&E Quality Moderately Satisfactory Satisfactory High RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) Moderately 01 21-Jan-2021 Moderately Unsatisfactory 175.00 Unsatisfactory 02 07-Apr-2021 Moderately Satisfactory Moderately Satisfactory 175.00 Page 2 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) 03 06-Oct-2021 Moderately Satisfactory Moderately Satisfactory 175.93 SECTORS AND THEMES Sectors Major Sector/Sector (%) Public Administration 48 Sub-National Government 48 Financial Sector 7 Banking Institutions 7 Social Protection 45 Social Protection 45 Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Social Development and Protection 100 Social Inclusion 100 Human Development and Gender 100 Gender 50 Disease Control 100 Pandemic Response 100 ADM STAFF Role At Approval At ICR Regional Vice President: Hartwig Schafer Martin Raiser Country Director: Henry G. R. Kerali Melinda Good Director: John A. Roome John A. Roome Practice Manager: David Seth Warren Kevin A Tomlinson Page 3 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Task Team Leader(s): Janamejay Singh, Andras Bodor Susan Wong, Andras Bodor ICR Contributing Author: Kosuke Anan Page 4 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL Context Country Context 1. The August 2021 takeover of the Afghanistan government by the Taliban marked the culmination of 20 years of violent conflict in Afghanistan that killed tens of thousands of civilians and combatants in a protracted civil war. Despite the optimism of the early years with economic growth averaging over 9 percent a year during 2003-2012, the gradual withdrawal of international forces and assistance starting in 2011-2012 saw a long-term decline in growth. The years 2017-2021 were marked by economic stagnation and deterioration as the country was hit by a worsening conflict and deepening internal political divisions. The contested Presidential elections in September 2019 were followed by a long hiatus until the announcement of a new government in May 2020. Average annual growth fell to 2.5 percent between 2015-2020, below the rate of population increase over most of that period. In 2020, the COVID-19 pandemic led to a temporary closing of the country's borders, an estimated 2 percent economic contraction and a sharp increase in poverty to reach an estimated 60-70 percent of the population by the end of the year. Most aid to the country was suspended following the seizure of power in August 2021, triggering an economic crisis, widespread and deepening poverty and major disruption to basic services. 2. Displacement, drought, and corruption were additional challenges. Displacement intensified during the period, with more than 3.5 million Afghans internally displaced by conflict by the end of 2020 and more than 2 million returning to Afghanistan—mostly from Pakistan and Iran—since 2015. Deterioration of rural livelihoods added to displacement: nearly 400,000 individuals were displaced by drought alone in 2018. Private sector development was heavily constrained by insecurity, political instability, weak institutions, inadequate infrastructure, and widespread corruption. Afghanistan was ranked 174 out of 180 countries in Transparency International's corruption perception index in 2021. According to the United Nations Development Programme (UNDP)'s Human Development Index, Afghanistan ranked 169 out of 187 countries in 2020. Many of the key educational gains of the past had been long under threat as school attendance levels had steadily fallen over several years, with more than half of children between 7-17 years missing from school in 2018, and girls making up over 60 percent of those affected (United Nations Children’s Emergency Fund (UNICEF), 2018). 3. Afghanistan's fragility, defined both in terms of weak state institutions, poor governance and dysfunctional societal relations leading to violent conflict between groups, affected almost every aspect of development. The risks for development partners were high in such an environment. At the same time, development partners had to contend with an increasingly difficult working environment, including managing long periods of remote working, third-party monitoring of projects and frequent restrictions on movement. During the COVID-19 crisis in 2020 many withdrew their international staff from Kabul and national staff worked from home. 4. The World Bank country strategy over the period, which was closely aligned to the government’s Afghanistan National Peace and Development Framework (2017) objectives, was organized around three pillars: (i) building strong and accountable institutions; (ii) supporting inclusive growth; and (iii) expanding and deepening social inclusion (Country Partnership Framework (CPF), 2016). Both the Bank's Systematic Country Diagnostic (SCD, 2016) and CPF pointed out the extreme fragility and high risk of operating in Afghanistan. The SCD argued that carefully Page 5 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) selecting and tailoring development interventions to the fragile circumstances in the country was imperative. The CPF noted the need for flexibility in project design and implementation, with the ability to adjust projects quickly as circumstances changed, and selectively downscale some interventions, while putting others on hold until the situation improved. 5. The COVID-19 Relief Effort for Afghan Communities and Households (REACH) project was also fully consistent with the priorities and approach established in the World Bank Group’s CPF. Building on the SCD, the project was designed to contribute directly to the CPF strategic priorities on: (a) “enhanced social protection for poor and vulnerable people”, and (b) “improved government and community capacity to manage and respond to natural disasters”, which were confirmed and extended through Fiscal Year 2022 in the 2019 CPF Performance and Learning Review (PLR). The project was fully aligned to these priorities by providing relief in the form of emergency social assistance/protection interventions. Furthermore, the project was aligned with two other key World Bank operations that were pertinent to these CPF priorities: (a) it was designed to enhance the community service delivery and governance capacity of the ongoing Citizens’ Charter Afghanistan Project (CCAP), and (b) it was intended to prepare institutional capacity for the Early Warning, Action, and Finance project (“ENETAWF” in Dari) in the pipeline. 6. Following the takeover of government in mid-August 2021, the World Bank paused its disbursements and suspended its operations in Afghanistan. The sudden cut-off of aid from the international community, combined with the impact of drought in rural areas, led to a dramatic fall in living standards and a humanitarian crisis. In response to the crisis, the World Bank, Afghanistan Reconstruction Trust Fund (ARTF) donors, and international partners found pragmatic ways to provide support for essential basic services to the Afghan people. On November 30, 2021, the World Bank’s Board of Executive Directors approved Approach Paper 1.0 for an immediate Transfer Out of $280 million of uncommitted ARTF funds to World Food Programme (WFP) and UNICEF for humanitarian gap financing, following a decision by ARTF donors. On March 1, 2022, responding to requests from the international community, the Board approved Approach Paper 2.0 which aims to protect the vulnerable, help preserve human capital and key economic and social institutions, reduce the need for future humanitarian assistance, and improve gender equality outcomes. This includes financing, analytical work, and coordination/convening opportunities. This support is through Recipient Executed grants, decided by the ARTF and made off budget and outside of the involvement of the interim Taliban administration (ITA), to United Nations (UN) agencies and potentially international and national non-governmental organization (NGOs). Approach 2.0 is designed to respond flexibly, based on experiences of early implementation, and informed by strong coordination among the development partners. 7. In conjunction with this approach, existing projects that were suspended in August 2021 were to be phased out and Implementation Completion and Results Reports (ICRs) prepared for the portfolio by end-2022 to ensure timely capture of results under each project as well as lessons-learned from the World Bank's long engagement in the country. Sector Context 8. Afghanistan was not an exception to the spread of the COVID-19 pandemic. At the time of project appraisal on June 30, 2020, the Ministry of Public Health (MOPH) had reported 31,517 confirmed cases in the country, with cases reported in all 34 provinces. While the number of confirmed cases and deaths was relatively low compared to nearby countries, Afghanistan was extremely vulnerable to rapid spread of the virus due to limited access to information, high percentage of poor and vulnerable households who subsist on daily earnings, constrained access to water and sanitation, weaknesses in health systems, and ongoing violent conflict. Page 6 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) 9. The virus and related containment measures, including border closures and the lockdown of major cities, were leading to adverse social and economic impacts. These impacts included: (a) massive disruptions to productive economic activity (loss of employment and income) and consumption; (b) disruptions to imports, including of vital household items, leading to rapid inflation 1; (c) reduced exports due to disruptions at border points; (d) negative impacts on remittances; and (e) increased fiscal pressures, as government revenues were expected to decline by at least 30 percent below budgeted levels 2. As municipality-level lockdowns were put in place in the major cities in an attempt to slow down the spread of virus, disruptions in economic activities were expected to substantially diminish sources of labor income across the country, where over half of all household income and two-thirds in urban areas were originating from vulnerable employment sources such as work at small businesses, street or market sales, day labor in construction, agriculture, and personal services. As a consequence, the national poverty rate was projected to rise to at least 61 percent, and potentially to 72 percent, forcing an additional 1.9 to 6 million people into poverty. 3 Female-headed households were expected to be affected more adversely, since women were less likely to work than men, and when they do, they were more likely to be in vulnerable employment. Economic growth in the preceding years had barely exceeded the rate of population growth (2.7 percent annually), and per capita incomes were projected to decline substantially in the subsequent years as the economy was starting to contract in 2020, leading to a likely substantial deterioration in living standards. Analysis revealed that a common strategy in poor Afghan households to cope with negative shocks was to reduce the quantity and quality of food intake, which had potential long-term impacts, especially for children. There were also increasing signs of economic suffering through loss of jobs, livelihoods, and other income sources. 10. The United Nation’s Food Security and Agriculture Cluster (FSAC) in Afghanistan had estimated 12 million people (32 percent of the population) to face acute food insecurity that required urgent action over June-November 2020. Of these, around 4 million were in the Phase 4 “Emergency” food insecurity status of the Integrated Food Security Phase Classification (IPC) 4 and 8 million in IPC Phase 3 “Crisis” states. The UN had included Afghanistan in a list of ten countries at risk of famine in 2020, and updated the Humanitarian Response Plan (HRP) for Afghanistan in which 35 million people (over 93 percent of the population) were classified as “Social Safety Net Population in Need” (SSN PIN). 5 11. The country’s humanitarian community could not fully meet the needs of people in need ofemergency support. While the role of humanitarian agencies in addressing the COVID-19 crisis was crucial, and the Humanitarian Country Team (HCT) was already stepping up its efforts, their mission was more narrowly focused on providing life-saving assistance to those in dire need, rather than providing broad social protection to all those who were temporarily 1 World Food Programme (WFP) data indicated rapid food price increases occurred in Afghanistan's main city markets in April 2020. 2 Largely due to COVID-19, gross domestic product (GDP) was expected to contract by at least 5.5 percent in 2020. 3 April 2020. Based on World Bank (2020) Household Welfare Impacts of COVID-19 in Afghanistan: A Microsimulation Approach [presentation]. C,. Cancho, Poverty and Equity Global Practice (GP). 4 The Integrated Food Security Phase Classification (IPC) is an innovative multi-partner initiative for improving food security and nutrition analysis and decision-making by determining the severity and magnitude of acute and chronic food insecurity. See ipcinfo.org. 5 The HRP notes that inclusion of a figure for people in need (PIN) of a Social Safety Net underlines the reality that humanitarian action is just the first part of a more comprehensive package of urgent measures that is needed from wider development actors to ensure no one is left behind in Afghanistan in 2020 due to the COVID-19 crisis. If these development needs are not met in a timely manner, the group risks slipping into humanitarian need. Anticipatory development funding and action are the key to avoiding suffering and the requirement for more expensive humanitarian assistance. Page 7 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) impacted by the crisis. The Humanitarian Response Plan 2018-21 was revised in June 2020 with a request for $1.1 billion to reach 11.1 million people (out of a total 14 million estimated to require humanitarian assistance) who were acutely affected by the COVID-19 pandemic, the conflict, as well as natural disasters and other sources of vulnerabilities. 12. Leveraging Afghanistan’s longstanding community driven development (CDD) platform was thus critical to manage such a large-scale human security/social safety net effort. Over the preceding decade, many programs had actively involved local communities in the planning and implementation of development projects through a CDD approach. The furthest reaching and most comprehensive of these programs was the National Solidarity Program (NSP) which helped establish Community Development Councils (CDCs). Over a decade and a half, NSP provided the main participatory platform for service delivery in an estimated 90 percent of villages in rural Afghanistan. It was repeatedly called in to be the emergency response mechanism and national safety net for various shocks (displacement, earthquakes, droughts, etc.) the country faced. Approximately 35,000 CDCs were formed by NSP between 2003-2016 in all rural areas of the 34 provinces. NSP was succeeded by the Citizens’ Charter National Priority Program, supported by the International Development Association (IDA)/ARTF-financed CCAP. The project featured several interventions that could be used for COVID-19 responses, most notably the Social Inclusion Grant (SIG)/Grain Banks program that had established almost 10,000 “grain banks” 6 across the country and had become a critical response mechanism during the 2018 drought. These provided food and cash assistance to the most vulnerable households in the community. CCAP had also launched a cash-for-work community asset building program (Maintenance and Construction Cash Grant – MCCG) as part of the emergency response to the 2017 regional displacement crisis. 7 At the time of appraisal, CCAP was operating in one third of Afghanistan’s 36,000 rural communities as well as in the four major cities of Jalalabad, Mazar-i-Sharif, Herat, and Kandahar. 13. In this context, the Government of the Islamic Republic of Afghanistan (GoIRA), as part of the COVID-19 response plan, requested the World Bank and the ARTF to support a national program called Dastarkhan-e-Milli 8. The Dastarkhan-e-Milli program intended to provide household relief packages in a third of the country through CCAP and in other two thirds of the country through a new operation, which became the COVID-19 Relief Effort for Afghan Communities and Households (REACH). 6 These grain banks were essentially local reserves used for philanthropic contributions from the community to help the most vulnerable household in their villages. They were not ‘banks’ per se as they only serviced grants. Also, they often provided more than just grains to needy households. 7 MCCG was a key activity included in the expansion of CCAP services to internally displaced or former refugee Afghan returnee populations through the Eshteghal Zaiee-Karmondena (EZ-KAR) project. 8 The name “Dastarkhan-e-Milli” referred to the “national dinner cloth” used for communal dining in Afghanistan and was meant to reflect the Government’s effort to build solidarity and respond to food insecurity. Page 8 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Theory of Change (Results Chain) 14. The overarching objective of this project was to mitigate the COVID-19 induced economic hardship and food insecurity among vulnerable households, thereby helping them to avoid harmful, unsustainable coping strategies and maintain health. This was pursued by providing short-term relief to households in kind or cash and allowing them to maintain consumption. At the same time, the project was expected to sustain private sector distribution networks through purchases of food and other essential goods from local small retailers and wholesalers. The theory of change for the project is summarized in Diagram 1 below. Diagram 1. Results Chain (excerpted from the Project Appraisal Document) Preparation of Lists of eligible Eligible household Large-scale Higher-level outcomes Short-term outcomes Activities Inputs community lists households supported with economic hardship Staff training developed and short-term relief in and food insecurity agreed kind or cash mitigated Comprehensive communications Communications Beneficiary Households campaign campaign rolled out households able to maintain health Delivery staff/ maintain status during relief Purchase of food consumption during phase packages agencies hired and trained lockdowns Private sector Standard contracts Small retailers and distribution with local Transfers to CDC/ Gozar Assembly (GA) wholesalers profit networks sustained traders/retailers through local accounts Economic shock Preparation for purchase contracts lessened mobile money Household support mobilized in all project areas Critical assumptions: • Community leaders adhere to project criteria for choosing households with special attention to FHHs, can agree on households to be targeted by the project, and will ensure that transfers are distributed fairly. • Adequate staff including females can be hired and trained in a timely manner. • Food aid can be procured and delivered with acceptable quality in a short time period. • Food items will be shared among all household members (including women and children as well as elderly) Project Development Objectives (PDOs) 15. The Project Development Objective was to provide emergency support to selected households through communities in project areas during the COVID-19 outbreak. Page 9 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Key Expected Outcomes and Outcome Indicators • Households receiving in-kind support in rural or peri-urban areas (Target: 1,760,000) o Of which female-headed (Target: 10 Percent) • Households receiving in-kind or cash support in provincial capital cities (Target: 360,000) o Of which female-headed (Target: 10 Percent) • Households receiving in-kind or cash support in Kabul municipality (Target: 500,000) o Of which female-headed (Target: 10 Percent) Components 16. The project had four components based on the geographic coverage of three different implementing agencies (IAs) already managing relevant projects (i.e., CCAP-Rural, CCAP-Urban, and Kabul Municipal Development Project –KMDP). Components 1-3 were designed to distribute relief packages to over 90 percent of households in respective target areas by mobilizing community groups, such as CDCs and Gozar Assemblies (GAs) that had been established through previous interventions (e.g., NSP, EZ-Kar) and Ad-hoc COVID-19 CDCs and Ad-hoc COVID-19 GAs 9 where no CDCs or GAs existed prior to the project. These community groups were planned to be supported by Facilitating Partners (FPs) contracted by the project to: (a) identify eligible beneficiary households; (b) receive and manage Emergency Relief Grants in their bank accounts; (c) procure the goods to be included in the in-kind relief package; (d) distribute the relief packages; and (e) report to the implementing agencies (IAs). The project areas comprised approximately two-thirds of the country. The remaining areas were covered under CCAP. 17. Component 1: REACH Rural - Household support in rural and peri-urban areas (Original Amount: US$136 million; Actual Disbursement: US$105 million). This component aimed to provide relief support to an estimated 2.5 million households (approximately 93 percent of all households) in 275 districts and 19 provincial capitals cities (where the Ministry of Rural Rehabilitation and Development—MRRD, rather than the Independent Directorate of Local Governance—IDLG, was already operating). The component was planned to distribute an in-kind relief package worth AFN 4,000 (approx. US$50 equivalent at the time of project implementation) to each eligible household in one tranche. The package, consisting of essential food staples and hygiene products was selected based on the calculation of 50 percent of monthly food ration of an average Afghan household. 18. Component 2: REACH Urban: Household Support in Provincial Capital Cities (Original Amount: US$56 million; Actual Disbursement: US$30 million). This component was aimed at providing relief support to an estimated 453,000 households (approximately 90 percent of total households) in 14 Provincial Capital Cities (PCCs). This included approximately 1.45 million returnees based on estimates by the International Organization for Migration (IOM), many living in settlement camps. Based on the underlying principle of near-universal coverage, affluent households that comprised approximately 10 percent of total households were planned to be excluded from the coverage using pre-determined standard exclusion criteria. A relief package for urban areas was worth AFN 8,000 (approx. US$100 equivalent at the time of project implementation) distributed to each eligible household in two tranches. The first tranche of AFN 4,000 (approx. US$50 equivalent) per household was planned to be distributed to 9Ad-hoc CDCs and GAs were formed solely for the purpose of this project, and were not involved in more extensive community mobilization activities of regular CDCs and GAs supported through CCAP, EZ-Kar, or NSP. Page 10 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) each eligible households either in-kind (based on a standard relief package that included essential food staples and hygiene products) or in cash, based on the preference of communities and subject to local circumstances based on pre-determined criteria. The second tranche of AFN 4,000 (approx. US$50 equivalent) per household was planned to be rolled out later (depending on the trajectory of the crisis), likely fully in cash. Application of the mobile money technology was being explored for the second tranche, where feasible, which depended on various factors, such as: (a) the number of households that owned an acceptable identification document (ID) for receiving mobile money transfers, and (b) mobile money operators (MMO)’s capacity to expand the network of cash-out points. 19. Component 3: REACH Kabul: Household Support in Kabul City (Original Amount US$80 million; Actual Disbursement: US$40.93 million). This component aimed to provide relief support to approximately 629,000 households (approximately 80 percent of total households) in Kabul Municipality (KM). Following the same modality and exclusion criteria of Component 2, nearly 20 percent of affluent households were expected to be excluded from the coverage, as Kabul had higher proportion of better-off households. A relief package worth a total of AFN 8,000 was planned to be distributed to each eligible household in two tranches. The planned contents of relief packages and tranche distribution timings were the same as those of Component 2. 20. Component 4: Project management, communications and monitoring (Original Amount: US$8 million; Actual Disbursement: US$4.91 million). This component supported the three implementing agencies with: (a) project management and institutional support; (b) dedicated strategic communication; (c) grievance management; and (d) project monitoring support, including gender and safeguards compliance monitoring. The three agencies were organized under this project to work together to ensure that communications, grievance management and monitoring were closely coordinated and reported jointly. B. SIGNIFICANT CHANGES DURING IMPLEMENTATION 21. The project financing from ARTF in an amount of US$35,247,130.01 was cancelled through the Cancellation Notice of Undisbursed Balance from Closed Projects dated April 4, 2022 and the Omnibus Notice of Cancellation dated April 8, 2022. The undisbursed balance of ARTF commitments was cancelled at the World Bank’s Afghanistan portfolio level to repurpose the funds for a new set of operations to be implemented in Afghanistan. The final project commitment amount reflects this cancellation. II. OUTCOME A. RELEVANCE OF PDOs Relevance of PDOs: High Assessment of Relevance of PDOs and Rating 22. The aforementioned PDO as well as its higher objective of mitigating the COVID-19 induced economic hardship and food insecurity among vulnerable households were highly relevant to the development context in Afghanistan. This was because the compound shocks of the pandemic were inducing vulnerable households to employ harmful and unsustainable coping strategies with irreversible adverse impacts on human development, which could also undo the development gains of the previous decades. Particularly, in the context in which 55 percent of the Afghan population Page 11 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) lived under the official poverty line and 93 percent lived on less than US$2 per day before the crisis 10, they were highly vulnerable to the economic shocks of the pandemic, such as loss of employment/income, scarcity and price volatility of consumption staples, and reduced access to basic needs. These vulnerabilities were compounded with high prevalence of food insecurity due to the country’s unique geography (e.g., droughts 11, floods, avalanches, and infestation of agricultural production). 23. The PDOs remained highly relevant as the change of the regime on August 15, 2021 created a further economic crisis, making the situation of the already vulnerable population even more precarious. Most international aid and all international security assistance have ceased, disrupting government services. Reductions in grant inflows have left Afghanistan unable to finance critical imports including electricity, food, fuel, and medical supplies. Access to international reserves and linkages to the international financial system have been disrupted. The economy is expected to contract by 30 percent over the year from August 15, 2021. 12 With half of the population already living below the poverty line, living conditions are becoming desperate, especially for the large internally displaced population (5.5 million), those impacted by drought (23 million face acute food insecurity), and those in urban areas affected by business closures and disruptions to public sector activity. 24. The PDOs also remained consistent with the priorities and approach established in the World Bank Group’s Afghanistan Approach Paper 2.0, which defines education, livelihoods, agriculture, and basic health as the four priority areas. The project’s focus on food security and rapid relief assistance to vulnerable households is particularly aligned with those of the Community Resilience and Livelihoods Project (P178760) and the Afghanistan Emergency Food Security Project (P178280) that were proposed in the Approach Paper 2.0. 25. Rating. Based on the above context, the relevance of the PDO is “High”. 10 Afghanistan Living Conditions Survey 2016-17. 11 Drought-induced displacement had reached record levels of nearly 300,000 individuals in 2018. On average three out of every four poor households were affected by at least one shock and 80 percent of them could not recover from their shocks within one year; many of them were often forced to turn to harmful coping strategies such as the sale of productive assets and taking children out of school for income generation. 12 World Bank Group’s Afghanistan Approach Paper 2.0. Page 12 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) B. ACHIEVEMENT OF PDOs (EFFICACY) 26. Despite a slow rollout due mainly to political and fiscal impediments at the onset, the project was making substantial progress towards achieving its objectives when the implementation was put on hold on August 15, 2021. Close to two-thirds of the project financing had been disbursed before August 15, 2021, and the project was on its way to achieving its PDO outcomes especially in urban areas outside of Kabul and rural areas. Concretely, 474,160 poor households or approximately 3.3 million Afghans had received immediate food and in-kind assistance, including 6,162 vulnerable female heads of household. Beneficiary households equivalent to 70 percent of the target number of households had been identified. This ICR analyzed the achievement of PDO, which was “emergency support provided”, based on the achievement of each of the three PDO indicators that represented the achievement in different project areas. The ratings of this ICR use a split-rating method that calculates the overall rating based on the proportionate weights given to the disbursed percentage (63%) pre- August 15, 2021 and the remaining percentage (37%) of the total project financing post-August 15, 2021 (as per the total commitment amount as of August 15, 2022) 13. Assessment of Achievement of Each Objective/Outcome 27. Households receiving in-kind support in rural or peri-urban areas (Ratings: Pre-August 15, 2021: Substantial; Post- August 15, 2021: Modest; Combined: Substantial). The project was aimed at reaching 1,760,000 households including 176,000 female-headed households in rural or peri-urban areas by December 31, 2021. As of August 15, 2021, a total of 294,288 households or approximately two million Afghans (17 percent of the target), including 5,090 female-headed households (3 percent of the target) 14 had received at least one in-kind relief assistance in rural and peri-urban areas, according to the data provided by the project’s management information system (MIS). According to TPMA reports, persons with disabilities had also received assistance. While the achievement was still low as of August 15, 2021, the project was making substantial progress, with a total of 3,739 communities mobilized (19 percent of the target), 689,330 beneficiary households identified (20 percent of the target), and grants disbursed to 2,583 communities (13 percent of the target). The field monitoring conducted by the Third Party Monitoring Agent (TPMA) across an impressively large field sample shows high credibility of the above figures reported by MRRD and those reported by IDLG below, with almost all households visited by TPMA door-to-door being included in a beneficiary list (97 percent, 128,538 out of 132,867) and most post-distribution respondents later reporting having received assistance (74 percent, 1,822 out of 2,467). 15 Post August 15, 2021, there are reports from MRRD, NGOs, CDCs and 13 The disbursement amount for the pre-August 15, 2021 period does not include the US$5 million that was paid directly to FPs after August 15, 2021 (which is included in the amount for the post-August 15, 2021 period). The disbursement percentage is calculated using the original total commitment amount of US$280 million (not the reduced amount after the cancellation of the remaining ARTF commitments in April 2022). 14 The number of reported female-headed households receiving the assistance was substantially lower that the expected and targeted number in both rural and urban areas, given many female-headed households were listed under the name of a male household member instead. The findings of post-distribution door-to-door monitoring by the TPMA indicates that 99% of female- headed households they visited in the target communities received assistance. 15 TPMA downloaded CDC beneficiary lists from the project’s MIS database for verification. TPMA used a mixed methods approach to triangulate results from both quantitative and qualitative data collection, including key informant interviews, direct observations of distributions, door-to-door surveys, and taking photographs on-site. Where monitoring identified major irregularities, findings were verified by making follow-up phone calls with multiple community members. Between CCAP and REACH, beneficiary lists were observed for 1,707 communities, distributions were observed in 1,795 communities, and 193,183 community members were interviewed. See TPMA 2022. COVID-19 Response: CCAP/REACH Monitoring, Q3 2021 Final Project Report. February. Page 13 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) the TPMA that some limited food distribution continued to take place but there are data gaps due to delays in data processing and reporting once NGO and government staff stopped being paid as a result of project suspension. 28. Households receiving in-kind or cash support in provincial capital cities (Ratings: Pre-August 15, 2021: Substantial; Post-August 15, 2021: Modest; Combined: Substantial). The project aimed at reaching 360,000 households, including 36,000 female-headed households in urban areas by December 31, 2021. As of August 15, 2021, a total of 179,872 households or 1.3 million Afghans (50 percent of the target), including 1,072 female-headed households (0.5 percent of the target) had received at least one tranche of relief assistance in urban areas, according to the data provided by the project’s MIS in September 2021. According to TPMA reports, persons with disabilities had also received assistance. As of the same date, 1,217 communities (51 percent of the initial estimate) had been mobilized, 664 (27 percent of the initial estimate) of which had completed beneficiary identification, listing 263,433 beneficiary households (73 percent of the target). These figures were indicating that while the number of communities was smaller in the target cities, the average number of beneficiary households in each community was much greater than initially estimated. Given the work in urban areas only started after the REACH urban budget was incorporated in the national budget on April 7, 2021, and 50 percent of the target value had been met in approximately four months with another 22 percent of households already identified and waiting for assistance, the project was on its way to achieving the target for the urban PDO as of August 15, 2021. This is supported by the updated data provided from the project’s MIS in April 2022, which indicates that the distribution of assistance was completed in 482 communities, reaching 318,472 households (88 percent of the target value), and benefiting 1,959,180 individuals, although there is an evidence gap and data cannot be verified due to the situation after August 15, 2021. The substantial increase is likely due to the time lag in data entry and processing; the limited data monitored by TPMA in September and October 2021 are vastly consistent with these updated numbers. 29. Households receiving in-kind or cash support in Kabul Municipality (KM) (Ratings: Pre-August 15, 2021: Modest; Post-August 15, 2021: Negligible; Combined: Modest). The project was aimed at reaching 500,000 households, including 50,000 female-headed households in KM by December 31, 2021. As of August 15, 2021, no household had received in-kind support in KM. However, as of the same date, 953 Ad-hoc GAs (108 percent of the initial estimate) had been mobilized, of which 862 (97 percent of the initial estimate) had submitted beneficiary household lists, and the FPs had verified the beneficiary lists for 854 of these Ad-hoc GAs (96 percent of the initial estimate) which consisted of 923,183 beneficiary households (184 percent of the target). This progress indicates that this PDO was still achievable despite the substantial delays and the beneficiary lists were still undergoing verification processes by the TPMA (for the third time) as of August 15, 2021. The factors that caused the delays are discussed later in the report. 30. Overall efficacy of the project objectives and outcomes are summarized in the table below: Page 14 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Table 1: Achievements of PDO Outcomes by PDO Indicator PDO Pre-August 15, 2021 Post-August 15, 2021 Overall Outcome (63 percent weight) (37 percent weight) Areas Rural and 294,288 households (17 percent of the target), 17 percent cumulative achievement. Substantial Peri-Urban including 5,090 female-headed households (3 (Modest) Areas percent of the target) 16 had received immediate relief assistance; progress was rapidly accelerating and was still expected to achieve the target by the closing date. (Substantial) Urban Areas 179,872 households (50 percent of the target), 51 percent cumulative achievement; Substantial including 1,072 female-headed households (0.5 (88 percent cumulative unverified percent of the target) had received at least one achievement). (Substantial) tranche of relief assistance; progress was on track to achieving 100 percent by the closing date. (Substantial) Kabul 0 percent achievement; was in the process of 0 percent cumulative achievement. Modest Municipality verifying the beneficiary households equivalent (Negligible) to 184 percent of the target); substantially delayed progress. (Modest) Overall 18 percent achievement (474,160 households 18 percent cumulative achievement; Substantial vis-à-vis the target of 2,620,000 households (23 percent unverified cumulative reached); was expected to achieve 100 percent achievement). (Modest) by the closing date (Substantial) Justification of Overall Efficacy Rating 31. The Overall Efficacy Rating is “Substantial”, given the efficacy rating prior to August 15, 2021, which had a larger split rating weight (63 percent) was “Substantial”, while the efficacy rating after August 15, 2021, which had a smaller split rating weight (37 percent) was “Modest”. The overall efficacy rating of “Substantial” is justified for the following reasons: 32. Prior to August 15, 2021, MRRD and IDLG together with CDCs were able to deliver immediate assistance in a war zone to over 3.3 million Afghans including 6,162 vulnerable female heads-of-household and other vulnerable households. As a point of comparison, the Dastarkhan-e-Milli program which was implemented through CCAP and REACH, reached a total of 10.3 million people in 11 months of implementation, while WFP had reached 4.9 million people through unconditional food assistance in 2020 in Afghanistan. 17 The independent TPMA reports confirm the high degree of accuracy of CDCs providing assistance to vulnerable households with 99 percent of sampled female heads-of households, 96 percent of sampled households headed by a person with disabilities, 92 percent of sampled 16 The number of reported female-headed households receiving the assistance was substantially lower that the expected and targeted number in both rural and urban areas, given many female-headed households were listed under the name of a male household member instead. The findings of post-distribution door-to-door monitoring by the TPMA indicates that 99% of female- headed households they visited in the target communities received assistance (sample size of 8,450). 17 Annual Country Report 2020 Afghanistan, WFP. Page 15 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) IDP households, and 91 percent of sampled elderly households all receiving assistance. 18 Given the escalating conflict situation in the country, along with severe COVID-19 challenges, by any international measure, these achievements are impressive. 33. Immediately following August 15, 2021, there are reports from the TPMA, NGOs, and the implementing agencies that the distribution of assistance continued in the field once the fighting ended. For example, the TPMA monitored food distribution by several CDCs in Jalalabad in late September and October 2021. Based upon field reports, there is a strong indication that some assistance was distributed post-August 15, 2021. There were also delays in data reporting and submission of receipts as staff stopped working due to project suspension. Therefore, due to evidence gaps, figures contained in this report are most likely underestimations. 34. Once Parliament had given approval of the national budget in February 2021, implementation moved quickly ahead in rural and urban areas, except in Kabul. The 18 percent progress towards the achievement of PDOs the project had made in four to six months after budget approval was consistent if not ahead of the projected workplan schedule, and the project was progressing towards mostly achieving the PDOs by the closing date. The preparatory phase of the project, which involved community mobilization, beneficiary identification, and disbursement of funds prior to the distribution of assistance had progressed substantially, particularly in urban areas (provincial capital cities and KM) where over 100 percent of the estimated number of beneficiaries had been identified. 35. One of the objectives of WBG engagement in FCV countries is to strengthen core state functions and support national systems. While it is not an explicit objective of the PDO, improved institutional resilience and capacity should be considered part of the contributing results to the PDO. REACH and CCAP demonstrated the ability of the CDC national platform to deliver services at scale, effectively, sustainably, and transparently, led by MRRD and IDLG with the assistance of Facilitating Partners. For the past 18 years, the CDC national delivery mechanism has served as the country’s safety net in times of emergency. Even after August 15, 2021, CDCs are being used by UN agencies and NGOs throughout the country to deliver humanitarian support especially to women and the most vulnerable. (See further discussion on CDCs in Section E). The first WB project re-engagement back into Afghanistan in 2022 is the Community Resilience and Livelihoods Project (P178760) which engages CDCs for the delivery of basic services and livelihoods. C. EFFICIENCY Efficiency Rating: Substantial Assessment of Efficiency and Rating 36. The efficiency of the project cannot be assessed or rated using regular methods, since the project implementation was halted abruptly before the planned closing date due to the collapse of GoIRA, and given the duration of the project, disbursement rate, and the outcomes do not progress in a linear manner 19. However, given the project had 18 The TPMA final report evaluated the entire CCAP/REACH COVID-19 response in 2020-2021 and based upon door-to-door surveys using large samples, the program reached 99 percent of female-headed households (sample size of 8,450); 96 percent of households headed by a person with disabilities (sample size 5,000); 92 percent of IDP households (sample size of 17,205) and 91 percent of elderly households (sample size of 14,397). 19 The project was into approximately 75 percent (completing approximately 12 of scheduled 16 months) of the implementation period, had a disbursement rate of 63 percent, and had achieved 18 percent of the PDO target as of August 15, Page 16 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) budgeted US$31.7 million (11 percent of the total project budget) for non-grant operational expenditures, as the project was sharing operational costs with CCAP, the projected cost-effectiveness was within the range expected in Afghanistan (for example, non-grant operational expenditures represented about 19 percent of the project budgets for both CCAP and the community grants component of EZ-Kar, although the nature and activities of the projects differed slightly). While the disbursements as a percentage of total financing significantly exceed results as a percentage of total targets, this does not indicate cost-inefficiency, because the mobilization/facilitation as well as the disbursement of funds to be distributed as assistance precede the actual distribution of the assistance. In terms of the implementation efficiency, as noted above, the actual implementation itself progressed as expected once the implementation began after the project experienced substantial delays in starting off the activities due to factors such as political and fiscal constraints that are explained in Section III.B. (Key Factors During Implementation) below. 37. Based on the above assessments, the project was generating outcomes at the efficiency expected in the operation’s sector, and is rated “Substantial”. D. JUSTIFICATION OF OVERALL OUTCOME RATING 38. The overall outcome rating of “Moderately Satisfactory” is justified based on the following: (a) Relevance. The objectives, design and implementation were relevant and are rated “High”. (b) Efficacy. The PDO, as measured by the indicators of the results framework show “Substantial” levels of achievement as of August 15, 2021, which declined to “Modest” by the project closing date. (c) Efficiency. The project was within the efficiency expected in its sector in the country and is rated “Substantial”. Table 2: Derivation of Project Ratings—4 Point and 6 Point Scales Period 1 Period 2 Overall Outcome breakdown Before August 15, 2021 After August 15, 2021 Relevance of PDO High (4) Efficacy (PDO) Substantial (3) Modest (2) Efficiency Substantial (3) Outcome Moderately Satisfactory (4) Moderately Unsatisfactory (3) Disbursement (percent) US$175.9m (63 percent) US$104.1m (37 percent) Weighted Rating 4 * .63 = 2.52 3 * .37 = 1.11 Overall Outcome Rating 3.63 = Moderately Satisfactory Note: Relevance, efficacy, and efficiency are rated on 4-point scales. Relevance and efficiency are rate once. The split rating only applies to efficacy. The overall outcome rating is on a 6-point scale. The disbursement amount for Period 1 does not include the US$5 million that was paid directly to FPs after August 15, 2021 (which is included in the amount for Period 2). The disbursement percentage is calculated using the original total commitment amount of US$280 million (not the reduced amount after the cancellation of the remaining ARTF commitments in April 2022). 2021. Page 17 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) E. OTHER OUTCOMES AND IMPACTS (IF ANY) Vulnerable Groups 39. The project had positive impacts on vulnerable households such as those headed by women and internally displaced persons (IDPs). According to the Final Report shared by the TPMA, based on door-to-door household surveys, almost all female headed households visited received assistance (99 percent, 8,348 out of 8,450 female households) 20. Based also on cross checks of the beneficiary lists and field staff observations conducted by TPMA, nearly all households headed by a person with a disability that were listed also received assistance, with only four percent (191 out of 5,000) not having received their package. Similarly, only nine percent of households headed by a person aged 65 or over listed did not receive assistance (1,244 out of 14,297). Out of 17,205 IDP households included in the beneficiary lists observed, 8 percent were observed to have not received their packages, despite many have recently moved.21 Capacity of Community Development Councils 40. The REACH project, along with CCAP, has had positive, sustainable impacts on the ability of communities to make decisions and take actions collectively through CDCs. As of February 2022 (six months after the fall of GoIRA), CDCs were still operating in the majority of provinces although there is variation across and within provinces, according to reports from humanitarian actors on the ground. Two World Bank-sponsored rapid surveys in February 2022 indicate that CDCs were still active and, in many cases, helping to coordinate assistance in their communities. 22 Responses from the surveys confirmed that CDCs were largely operational across the country and were being used to varying degrees by UN agencies and NGOs in ongoing humanitarian and development activities. In the key informant survey, respondents reported that CDCs were operational in 30 of 34 provinces (88 percent). In the survey of international and national NGOs, 16 out of 19 NGOs (84 percent) reported working with CDCs where they exist. 23 UN agencies such as UNICEF and WFP have also reported working with CDCs in several provinces during the humanitarian crisis following the fall of GoIRA. It is important to note however, that there is heterogeneity across and within the provinces. Notably, challenges with CDCs were reported in Wardak, Ghazni, Paktika, and Faryab provinces. III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION 41. The emergency context in which the project was prepared amidst the COVID-19 pandemic affected the design of the project, given many uncertainties, unknowns, and the time pressure to respond urgently. Below are some of key factors that particularly affected project preparation, which necessitated multiple 20 All figures in this paragraph include both CCAP and REACH beneficiaries. 21 According to the field monitoring conducted by TPMA from July-October 2021, IDPs were likely excluded where they had moved to the community after the profile was complete (32 percent, 216 out of 677 excluded IDP households observed), something that was much more likely to occur in MRRD areas (87 percent, 188 out of 216 monitored) than in IDLG ones (13 percent, 28 out of 216 monitored), or because the community profile had not been updated (30 percent, 203 out of 677 excluded IDP households observed). 22 The first survey is a key informant survey of 356 Afghanistan humanitarian and development actors at the central, provincial and district levels. The second survey is an anonymous online survey of 19 international and national NGO partners engaged previously under CCAP. 23 World Bank. 2022. “Rapid Assessment of Afghanistan Community Development Councils: Results from Two Rapid Surveys” March. Page 18 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) revisions to the project’s scope, design, and delivery modality as the situation remained fluid during preparation. 42. Time constraints. At the time of project preparation, the nature of the COVID-19 pandemic was still not well understood nor easily predictable 24, and the project was prepared with an assumption that the impact would be immediate that required an urgent and relatively short-term relief response, followed by additional medium-term recovery and resilience responses in the future. Based on this assumption, the project followed a condensed project processing procedures of the Projects in Situations of Urgent Need of Assistance or Capacity Constraints (Paragraph 12 of Section III of the Investment Project Financing (IPF) Policy and Paragraph 56 of Section III of the IPF Directive), with an aim of getting the Board approval within three weeks (May 14, 2020) of the Activity Initiation Summary (AIS) sign-off and starting to deliver actual assistance to the beneficiaries within two months (June 30, 2020) of the AIS Sign-off. This meant not only the preparation had to be done in an extremely condensed timeframe while many GoIRA and World Bank staff were affected by the virus, but also the implementation arrangements and delivery mechanisms of the project had to be designed to enable rapid implementation. For example, many design options that required regular procurement procedures or establishment of new implementation modalities such as the use of mobile money technology had to be discarded, at least for the first tranche of assistance. 43. Availability of funds. Despite the project was being prepared with a very condensed timeline, the funding for the project could not be secured until funds from the existing projects in the ARTF and IDA portfolio were cancelled on June 24, 2020 to be redirected to the COVID-19 emergency operations. This delayed the timing of the appraisal and approval by more than two months (until July 1, 2020 and August 4, 2020, respectively). The amount of available funding also affected the scope of the project. Given the limitations to the availability of funding, even with the funds redirected from other projects in the portfolio, the project had to be designed to strike a balance between the impact on each household (amount of assistance to each household) and the coverage of assistance (geographical coverage of the project as well as the proportion of the population targeted in the target areas). Despite the funding constraints, GoIRA had expressed their desire to provide more generous assistance to each household (including the second tranche of assistance, at least for urban households) and throughout the country. The geographic coverage and the amount of each assistance had to be defined with a high level of unknowns and uncertainties over (a) the actual number of households in each district; (b) population movements triggered by conflicts, drought, and natural disasters; (c) changes to the status of the security situation in each district; and (d) future availability and price trends of basic food items in the market. 44. Technical constraints. The choice of implementation modalities was also limited in Afghanistan due to the absence of enabling technical conditions. For example, in the absence of: (a) formal registration of households; (b) reliable database to keep track of households’ whereabouts by communities or districts; (c) prevalent access to commercial bank accounts or mobile money accounts; or (d) reliable postal service 24It was not clear (a) how fast and widely the virus was going to spread; (b) how long, widely, and strictly the lockdown and social distancing measures were going to be enforced; and (c) in what frequency and intervals the waves of COVID-19 were going to affect Afghanistan. Page 19 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) or other means of delivery networks available throughout the country; technical options for efficient and safer beneficiary identification and assistance distribution were limited in the country, and the project had to resort to less efficient modalities like physical identification and verification of beneficiaries and purchasing of food packages with physical cash. 45. Varied views of stakeholders. The project was designed in response to the demands of GoIRA, while reflecting the views of ARTF donors as well as those of the World Bank. Reconciling the varying views of these different stakeholders required rounds of lengthy consultations and negotiations, including at the highest level involving the President and the First Vice President of Afghanistan. GoIRA did not want to engage the UN nor NGOs in any capacity for the implementation of the project, and wanted to directly procure food suppliers in order to support the private sector. Both GoIRA and ARTF donors wanted the project to cover the entire country, including 47 extremely insecure districts that were not under the GoIRA control, to avoid increasing social tension, while minimizing fiduciary risks by avoiding cash distribution (GoIRA was not interested in cash distribution unless it could be done through mobile money). At the same time, the World Bank initially had reservations for food distribution25 and for including the 47 extremely insecure districts 26. In case these areas were to be covered, the World Bank wanted to minimize fiduciary risks by possibly partnering with the UN. The World Bank also preferred to engage experienced NGOs as facilitating partners and leverage existing structures to mitigate fiduciary risks. Changes in any one of these parameters had major implications to the project design. As a result of multiple rounds of consultations, the project was designed to: (a) mainly distribute food assistance (with a possibility of using IDA grants—not ARTF grants—for cash-distribution in urban areas); (b) work with NGOs as facilitating partners; and (c) include an option of including 47 extremely insecure districts subject to stabilization of security conditions and availability of additional financing; but (d) not involve the UN as the implementing agency of the project. 46. Security risks. Besides the security risks in hard to reach areas or the 47 extremely insecure districts mentioned above, the level of violence and criminal activities were generally very high throughout the country, posing security risks to project staff, FPs, service providers, and beneficiaries. Risks included detainment by anti-government elements (AGEs) and robbery of cash and/or assistance packages during transportation, distribution, and collection. Concerns for such risks further reduced the variety of implementation options, such as those that involved physical transportation of cash over a long distance by beneficiaries or that required storing of food packages in a location overnight prior to distribution. 25 The reservations for food distribution stemmed from inter alia: (a) questions on the World Bank’s comparative advantage in food distribution vis-à-vis other agencies like the WFP, (b) potentially higher utility of cash for preventing harmful coping mechanisms of vulnerable households and for supporting their recovery, vis-à-vis food distribution, and (c) GoIRA’s preference for engaging private sector suppliers using untested operational mechanisms that posed high fiduciary risks. In addition, according to the policy of the World Bank, Regional Vice President (RVP) approval prior to the Appraisal Decision Review was required to make food expenditures eligible for the World Bank financing. 26 The World Bank had reservations to include the 47 extremely insecure districts, since (a) it increased project cost; (b) it ran the risk that these funds would not be used on time due to the inability to access them; and (c) there would be a risk of extortion/tax by anti-government elements (AGEs). Page 20 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) 47. Impediments imposed by COVID-19. The REACH project was also intended to support households to comply with social distancing and lockdown measures to slow the spread of the outbreak. This meant the implementation of the project also had to comply with the measures and some operational options that would have enabled much more efficient operation had to be discarded. For example, while in normal settings, relief could be made in the form of cash for work/labor intensive public works, COVID-19 made such a strategy infeasible because of the need to maintain social distancing. The processes of community- led beneficiary household identification as well as of the distribution and monitoring of the assistance packages also had to avoid large gatherings of people. Therefore, the beneficiary identification process had to be adjusted to be done in smaller groups, and modalities that required mass registration of beneficiaries, mass collection of cash at the commercial bank or money agents, or mass exchanges of vouchers at specific shops had to be discarded, limiting the project to less efficient operational options. B. KEY FACTORS DURING IMPLEMENTATION 48. Project implementation was severely affected by a number of factors, particularly political and security instability in lead up to the eventual collapse of the GoIRA. These extraordinary factors resulted in delaying the start of the implementation as well as in halting all project activities four months prior to the planned closing date of the project. However, the use of an existing project implementation structure and community-based service delivery platform helped to make quick and substantial implementation progress. 49. Political instability. When the project became effective on September 14, 2020 following the signing of the financing agreements for the project on August 27, 2020, the Government’s second-term cabinet had not been confirmed by the parliament. During this period, the parliament was holding greater political influence than usual, since the confirmation of the cabinet members depended on their votes. Under these circumstances, the parliament rejected the government-initiated Dastarkhan-e-Milli program in late August 2020, which the REACH project was intended to support. 27 Although the implementation of the REACH project and execution of its financing did not require parliamentary approval as it was implemented under the non-discretionary development budget, the parliamentary decision led to a delay in the rollout of the project, since the relevant ministers did not want to sign off on the execution of the project until their appointments were confirmed by the parliament. After the majority of cabinet member appointments were confirmed in late November 2020 and with an intervention of the President, the Ministry of Finance finally released the funds for the project on December 8, 2020 (almost three months after project effectiveness), and the IAs started signing contracts with the FPs. 50. Budget freeze between fiscal years. The fiscal year of Afghanistan runs from December 23 to December 22, and the accounts for projects are usually frozen from mid-December until the annual budget for the subsequent year is approved by the parliament. This implied that except for about a week after the project funds were released on December 8, 2020, the project’s IAs did not have access to the project financing 27Members of the parliament raised concerns that: (a) several projects were cancelled to provide funds for the Dastarkhan-e- Milli program; and (b) funds and resources could be misused. There were also other political issues potentially behind the opposition. Page 21 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) until the parliament approved the budget for 2021 in late February 2021, resulting in an additional delay. 28 Moreover, the REACH budgets for IDLG and KM were not included in the original annual budget for 2021 in order to facilitate the parliamentary approval of the annual budget for 2021 (the REACH proportion of the IAs’ budgets, which did not require parliamentary approval, was going to be added to the budget separately later). This eventually led to a further delay in implementation by IDLG and KM as the REACH budget was not incorporated in the annual budget until April 7, 2021. It particularly had an adverse impact on the implementation of Component 3, since KM, unlike IDLG or MRRD that were sharing many PIU staff and equipment with CCAP, did not have other means to mobilize staff or pay for basic operational costs, such as internet services or mobile phone bills for several months into 2021. 51. Use of existing project implementation platform and structure. The project built on the community mobilization and service delivery approach of the CCAP delivery platform, utilizing the existing Project Implementation Units (PIUs) of CCAP and KMDP. This helped to kick start the project by saving the time to establish from scratch the new project implementation structures as well as arrangements including the operations manual, staff capacity, and the FPs. Given the support for the Dastarkhan-e-Milli program had started in 2020 through CCAP, using and piloting the same implementation structure and arrangements, the REACH project also benefited from the adjustments and improvements to the design made from the CCAP experience. In addition, the familiarity of the core PIU staff and FPs with the World Bank’s policies and procedures related to financial management, procurement, and environmental and social framework as well as the project’s implementation arrangements helped to swiftly start implementation once the funding was released. 52. High turnovers and uncertain contractual status of project staff. Project implementation in urban areas experienced an additional setback in December 2020 when the IDLG leadership decided unexpectedly not to renew the contracts of six of 14 senior PIU staff of IDLG, including the Project Director and the Human Resources Unit Director. At the suggestion of the World Bank, the decision to terminate the contracts was put on hold until it was overturned a few months later, but the uncertainties over the contractual status of the key PIU staff members had a detrimental impact upon the project by causing serious delays in the signing of contracts and making key operational decisions and instructions. KMDP also experienced a similar challenge as the KM authorities could not finalize the decisions on the retention of key PIU staff as their contracts were ending upon the closure of KMDP on June 30, 2021. 53. Lack of similar operational experience and context in KM. While the project was designed to take advantage of the existing implementation arrangements from CCAP, unlike MRRD and IDLG, KM was not part of CCAP and did not have any similar project implementation experience or arrangements in place. This constrained smooth implementation of Component 3 in various ways. First, the PIU was not structured for implementation of community-level relief activities, and did not have sufficient staff with 28Only MRRD was able to request an allocation of operational budget for the first month of 2021 in advance, but the allocation was not made until February. MRRD was also able to release community grants to the clearing account (which was not subject to the year-end freezing of accounts) for a limited number of communities that had already identified the beneficiary households in December 2020, allowing those grants to be accessible for the communities during the end of the fiscal year freeze period. Page 22 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) required skill sets for the purpose of the project. With the delay of the budget approval mentioned above, the PIU for Component 3 could not hire new staff for the REACH project until April 2021. 29 Second, there were no CDCs or GAs from previous projects present in KM. In the absence of these community structures, the ad-hoc CDCs/GAs had to be established from scratch in KM. This exercise was further complicated as the interventions by existing local elites, such as those of Gozar Councils (the structure of which were different from those of CDCs or GAs established through NSP and CCAP) made the formation and operation of ad-hoc CDCs/GAs in alignment with the REACH operations manual difficult. Often, geographic boundaries of ad-hoc CDCs/GAs could not be drawn and documented clearly nor consistently, resulting in overlapping or incontiguous boundaries. Third, unlike in other cities, the beneficiary household identification methodology was not easily adaptable in KM, which contained much larger affluent areas as well as apartment complexes, compared to other cities. The unique composition of communities in KM made the structure of community, nature of relationships among households, and the form of households different from other areas. For example, the definition of a household based on the use of a common kitchen made the identification of households living in apartment complexes complicated. As a result of these factors, the beneficiary lists in KM came out with many inconsistencies, with the number of beneficiary households greatly exceeding the initially estimated numbers. The verification of the household lists also faced a challenge and led to further delays as the households were not listed in geographical sequence and households from different neighborhoods were mixed. TPMA conducted verification exercises of the beneficiary lists multiple times, which concluded that 28 percent of listed beneficiary households were ineligible for assistance. 54. Advance of the Taliban and the Collapse of GoIRA. The negative impact of the deteriorating security situation on project implementation should not be under-estimated. REACH delivered assistance in 2020 and the first half of 2021 precisely during the period of record levels of violence and insecurity in the country. The Global Peace Index in 2021 listed Afghanistan as the most violent country in the world for the fifth year in a row.30 During the first half of 2021, the United Nations Assistance Mission in Afghanistan (UNAMA) reported a 47 percent increase in civilian casualties and injuries as compared with the first half of 2020. 31 Project implementation was affected by the rapidly deteriorating security situation in the country, particularly as the violence intensified and the Taliban forces took control of an increasing number of districts following the announcement of the United States (U.S.) and North Atlantic Treaty Organization (NATO) troops withdrawal in April 2021. In the city of Herat for example, the selected FP could not sign the contract since the Taliban had prohibited any activities associated with the Dastarkhan- e-Milli program in the city even prior to the fall of the municipality. In many other areas, implementation had to be put on hold during heavy fighting. Facilitation of beneficiary identification processes was also hindered by increased conflicts, reflected in the declining proportion of CDCs being able to update their 29 In conjunction, Component 3 PIU suffered additional capacity gaps as the contracts of existing PIU staff were expiring in June 2021 with the closing of KMDP as mentioned above. A prolonged period of uncertainties over the staff’s contract renewals by KM led to high turnovers and vacancies that affected the ability of the PIU to function effectively. The turnovers and re- appointments included the position of the Project Manager. Municipal authorities also proposed to replace key existing PIU staff involved in recruitment, procurement, and financial management, causing disruptions to these key functions of project implementation. 30 Institute for Economics & Peace. Global Peace Index 2021: Measuring Peace in a Complex World, Sydney, June 2021. 31 UNAMA, Afghanistan 2021 Midyear Update on Protection of Civilians in Armed Conflict, July 2021 Page 23 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) community profiles (which was used to prepare beneficiary lists), from 66 percent in the quarter ending in June 2021 to 46 percent since July 2021 (TPMA). The increased risks of extortions (or taxes) by AGEs also resulted in none of urban communities opting for the cash distribution, although it was permitted for the communities meeting the pre-determined minimum conditions. In addition, a large number 32 of conflict-induced internal displacements affected project implementation as the lists of beneficiary households were quickly outdated and had to be updated and verified multiple times in the host communities. In the end, the eventual collapse of GoIRA on August 15, 2021 led to a complete halt of all project activities four months before the planned project closing date, immensely affecting the overall final outcome of the project. IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 55. Considering the emergency nature of the project, the PDO indicators were formulated to reflect the intended project objectives that are attributable to the project (number of households receiving support), while keeping them sufficiently simple and realistic for data collection. The M&E design was reflected in the project’s budget, staffing, and the work plans, so that there was sufficient resources and time available to implement the M&E in a practical manner. The M&E design not only enabled regular monitoring of implementation progress, but also served to verify and increase the integrity of: (a) beneficiary lists; (b) distribution of assistance to the intended beneficiaries; and (c) adherence of project implementation activities to the applicable operational policies and procedures. The M&E arrangements were multi-layered, which consisted of: (a) the involvement of community leaders (Mullahs/imams, nahia representatives, teachers, youths), FPs, local governments, and PIUs who used pre-developed forms to validate beneficiaries and distribution; (b) the TPMA (both in-person and by phone) who was engaged in spot-checks as well as case-specific inspections; and (c) the use of periodic phone surveys, which contacted beneficiaries on a sample basis. The MIS was built on that of CCAP, equipped with the Kobo-toolbox applications that enabled collection of field-data through portable devices. M&E Implementation 56. M&E was systematically embedded in the implementation of the project through the Operations Manual as well as the forms that had to be filled, entered in the MIS, and approved by designated parties to proceed with project activities. This embedded M&E mechanism ensured not only that the status was tracked for each of the implementation status, such as community mobilization, beneficiary identification, grant disbursement, procurement of goods to be included in the assistance packages, and distribution of the packages, but also that the processes were followed correctly before proceeding to the next steps. For example, the grants were released to the communities only after the FPs and the provincial management units (PMUs) had checked the inclusive mobilization of the (ad- hoc) CDCs and GAs as well as the consistencies between the beneficiary lists and the grant request forms. In addition, the TPMA implemented monitoring activities as agreed with the World Bank’s task team, and provided immediate 32 According to UNHCR, 550,780 people had been internally displaced as of August 16, 2021 since the beginning of the year. Page 24 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) red flag alerts as well as weekly reports on overall implementation progress and the status of red flag issues. 33 The TPMA prepared weekly reports for 28 weeks from the onset of field activities, and those weekly reports were reviewed together with the WB task team and implementing agencies in order to make adjustments. M&E Utilization 57. The data and reports collected and provided by the project’s MIS and the TPMA were used by the World Bank team and the PIUs for various purposes. For example, the number of households and eligible beneficiary households reported from the initial set of communities from MRRD and IDLG helped to assess the trends of population movements and the adequacy of project funding to cover all eligible households in the target areas. The review of the beneficiary household lists by KM and TPMA helped to detect inconsistencies with the beneficiary lists and the processes followed to develop them, which led to require re-facilitation of beneficiary identification activities before approving the release of assistance grants for the communities in KM. Red-flag alerts on beneficiary inclusion/exclusion, civil unrests during distribution, and inadequate COVID-19 precautionary measures were also followed up and addressed swiftly by the PIU and followed up by the World Bank’s task teams throughout the implementation period. Regular meetings between the World Bank, the TPMA and the three implementing agencies occurred on a weekly if not daily basis during the field implementation period. Justification of Overall Rating of Quality of M&E 58. The overall rating of Quality of M&E is rated “High”. It reflects that the M&E was designed adequately to fit the purpose and context of the project, implemented as planned as embedded in the Project’s Operations Manual and forms, and the collected information was used effectively to address issues as they emerged and to inform course- correction measures as required. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE 59. The project was subject to five Environmental and Social Standards (ESS). They were: ESS 1 (Assessment and Management of Environmental and Social Risks and Impacts Relevant), ESS 2 (Labor and Working Conditions Relevant), ESS 3 (Resource Efficiency and Pollution Prevention and Management), ESS 4 (Community Health and Safety), and ESS 10 (Stakeholder Engagement and Information Disclosure). During project implementation, the Stakeholder Engagement Plan (SEP) and the Environmental and Social Management Plan (ESMP) were updated satisfactorily despite considerable delay, and the performance ratings for the five ESS remained Moderately Satisfactory throughout the duration of the project. There were breaches of COVID-19 prevention measures during the distribution of relief assistance packages as reported by the TPMA, which was addressed by re- communicating the importance of the measures with the creation of visual communications materials. 60. The financial management performance for the project was also rated Moderately Satisfactory throughout the duration of the project. This is mainly because the IAs experienced delays in completing agreed actions, such as the submission of the quarterly financial reports and cash flow projections. The signing of the Memorandum of Understanding (MoU) with Da Afghanistan Bank that was required to set up the bank accounts for the GAs in KM was also delayed. To address these issues, monthly financial management (FM) check-in meetings were 33Red flags were triggered for issues related to unverifiable households, procurement failures, adherence to COVID-19 precautionary measures, availability of information on grievance reporting, among others. Page 25 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) introduced to check and follow-up on the agreed actions, in addition to regular issue-specific interactions. There is an outstanding (unaccounted) amount of approximately US$165 million in the designated accounts of the project, and it is not clear as yet how much of those funds have been disbursed for assistance post-August 15. 61. The performance of procurement was rated moderately unsatisfactory in the initial Implementation Status Report (ISR) due to delayed procurement activities; however, the rating was updated to moderately satisfactory by the second ISR and remained the same until project closure. The project applied a fast-track emergency procurement approach. It included measures such as (a) direct contracting and/or limited competition with identified manufacturers and suppliers; (b) raising of the thresholds for allowing shopping and national procurement methods; (c) making all procurement activities subject to post reviews (rather than prior reviews); and (d) allowing shorter bidding time, no bid security, advance payments, direct payments, etc. which were applied on a case by case basis. Community procurement methods were also applied to the procurement of food and other essential goods conducted by communities. The findings of the TPMA suggest that the quality and quantities of the procurement items were in general aligned to the approved procurement plans 34, Only a limited number of CDCs had both the Bid Quotation and the Purchase Order forms available during distribution. However, after distribution the financial review of statements of expenditure showed that relevant documents were available at government offices in the majority of cases. C. BANK PERFORMANCE Quality at Entry 62. The Bank performance in ensuring quality at entry was “Satisfactory”, considering the following: (a) Strategic relevance and approach. The World Bank provided extensive support to the client in identifying a form of relief assistance to vulnerable households, which was strategically relevant to the emergency situation of the country facing multiple constraints. While the impacts of the COVID-19 pandemic were evolving fluidly with an extraordinary level of uncertainties, the World Bank’s Afghanistan country team across global practices and practice groups worked together to design the emergency operation with the government through consideration of many scenarios and rounds of consultations to reconcile varying views. The World Bank and Government were able to prepare the emergency operation in only 3.5 months from Activity Initiation Summary (AIS) in April 2020 to project approval in early August 2020. Project effectiveness followed one month later in September 2020. (b) Technical, financial, and economic aspects. The World Bank also helped the client to identify and adopt an operational approach and platform that enabled a rapid rollout of project activities, except in KM where the operation experienced a substantial delay due to its unique operational environment. (c) Poverty, gender, and social development aspects. These aspects were adequately considered in the design of the project, such as in the quasi-universal, “targeting from the top” methodology for identifying beneficiaries that minimized the exclusion of needy households, including those headed by women, the elderly, disabled persons, and IDPs. 34According to the Final Project Report by the TPMA, almost all beneficiaries received the same quantity of assistance throughout the programme (1,790 out of 1,795), with 76 percent of CDCs in REACH areas distributing goods with no visible defects (766 out of 1,003). Page 26 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) (d) Environmental and social aspects. The Environmental and Social Assessment adequately identified key environmental social risks of the project, and the Environmental and Social Commitment Plan addressed the key occupational health and safety risk in relation to the contamination with COVID-19. A grievance redress mechanism (GRM) and a hotline were also established to enable project-affected people to a raise concerns and grievances as well as to request clarifications on various aspects of the project and its potential impacts. (e) Fiduciary aspects. The World Bank’s technical input helped to establish financial management that enabled swift flow of funds to communities (upon release of the funds by the Ministry of Finance) and simplified community procurement procedures, with which no major misuse was reported or observed, although only a fraction of communities could present procurement documents to the third-party monitor. (f) Implementation arrangements. The project utilized the implementation arrangements of CCAP, partnering with experienced NGOs as facilitating partners and mobilizing community organizations (CDCs, GAs, and ad- hoc CDCs/GAs) which enabled swift rollout and implementation of the project, although the lack of experience with similar implementation arrangements by KM resulted in substantial delays. (g) M&E arrangements. The project had robust, multi-layered M&E arrangements that helped to closely monitor progress and results with: (a) straight forward PDO indicators appropriate for an emergency operation; (b) the involvement of community leaders (Mullahs/imams, district (nahia) representatives, teachers, youths), FPs, local governments, PMU; (c) the active engagement of TPMA; (d) a MIS to which field-data could be fed through portable devices using the Kobo-toolbox applications; and (e) the use of periodic phone surveys. (h) Risk assessment. Key risks and their mitigation measures, particularly those related to social exclusion, corruption, lack of security, and the spread of COVID-19 were identified during project preparation. Additional measures, such as COVID-19-specific measures, management of crowds attending aid distribution points, agreed protocols for security and social distancing, were part of the project design. The SEP elaborated on citizen engagement that consisted of: (a) dedicated communication and outreach component focusing on transparency and accountability; (b) Citizen / Community Engagement plan which will include outreach, consultations during implementation, and feedback on support received; (c) participatory monitoring methods; and (d) GRM. Quality of Supervision 63. The quality of supervision was Satisfactory, considering the following: (a) Focus on development impact. The task team proactively identified and resolved threats to the achievement of the development outcomes during the course of the project. These included: (a) advice to update the beneficiary lists multiple times to address the high influx of IDPs in some communities, especially in urban areas; (b) detailed guidance and monitoring provided to KM to ensure the integrity of the beneficiary identification process; and (c) programming of the allocation of operational staff and resources to efficiently make progress across the project coverage areas, such as where FPs could not be procured and MRRD had to implement activities directly. (b) Proactively responding to emerging challenges and a fluid context. Throughout the process, especially the crucial implementation period of September 2020 to August 2021, the task team was communicating and meeting virtually almost daily with the three implementing agencies as well as communicating regularly with the First Vice President’s Office and the Ministry of Finance, who were tasked with helping to coordinate the emergency response. Bottlenecks especially with budget allocations and financial flows, were quickly escalated to the WB’s Sustainable Development Director level, the Country Management Unit and in some Page 27 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) cases, the South Asia Vice President’s office to raise with senior government counterparts. While the World Bank Kabul office was closed and international staff were not allowed into the country due to COVID restrictions and growing insecurity, those constraints did not stop the task team from providing regular advice for problem solving and technical support to the implementing agencies virtually. Implementation support missions were also organized virtually in October 2020 and February 2021 to identify and address key project rollout and implementation challenges. As security deteriorated during the implementation period, the task team along with the Government implementing agencies worked together on ways to adapt the implementation workplan and schedule, for example to deliver services and food packages in peri-urban areas where IDPs were arriving, and in areas that experienced a lull in the fighting and brief windows of opportunity to deliver assistance. (c) Supervision of fiduciary and safeguard or environmental and social aspects. Given the high fiduciary and safeguards risks associated with the project, the World Bank’s task team had designed monitoring arrangements together with the TPMA to report on these key aspects on a weekly basis, with immediate red- flag alerts on individual cases issued to both the PIUs and the task-team on a daily basis 35. The task team was fast to respond to and follow-up on these alerts and reports with the implementing agencies throughout the implementation of the project. The WB also held firm with Kabul Municipality when it was clear from TPMA reports that the beneficiary rosters had significant deficiencies. Multiple meetings were held with the Kabul Mayor as well as the implementing NGOs about the weaknesses. Justification of Overall Rating of Bank Performance Rating: Satisfactory 64. The overall Bank performance was rated “Satisfactory”, considering the quality at project entry was “Satisfactory” while the quality of supervision was also rated “Satisfactory”. The Bank was quick to assemble a multi-disciplinary task team to respond to the Government’s request for emergency support and to put procedures in place to address the high fiduciary risks including strong third party monitoring verifications. During implementation, the task team worked together with the Government implementing agencies to adapt to the highly fluid context in addressing political bottlenecks and the quickly deteriorating security situation. 35Between CCAP and REACH, a total of 494 alert notices (on household exclusion or inclusion of ineligible households) and 453 red flags (on procurement failures, lack of COVID-19 protective measures, unverified households, civil unrests, and cash or good diversion) were reported. Page 28 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) D. RISK TO DEVELOPMENT OUTCOME 65. Risk to Development Outcome, at the time of completion, is “High”, since the development gains of Afghanistan are now at high risk as the country is facing a major economic crisis. The August 15, 2021 political crisis has resulted in an abrupt cessation of most international aid and all international security assistance. This has led to major disruption to core government services and a massive contraction in aggregate demand. Reductions to grant inflows have left Afghanistan without a source of hard currency to finance critical imports (grants previously financed a trade deficit of around 35 percent of gross domestic product (GDP), with aid inflows providing hard currency to pay for critical imports including electricity, food, fuel, and medical supplies). The exchange rate has depreciated by 15 percent against the U.S. dollar since August 2021. As a result of international sanctions, Afghanistan has lost access to international reserves while linkages to the international financial system have been disrupted, driving the financial sector into crisis. Unless mitigating measures are taken, fiscal contraction and disruptions to private sector activity are expected to lead to a 30 percent reduction in economic output over the year from August 15, 2021. 66. The crisis is having extreme impacts on households. Reduced availability of household products is driving increasing prices with annual inflation for a package of basic household goods reaching around 40 percent. More than two-thirds of households are unable to cover basic food and non-food needs, with around one-third of households unable to cover even food needs. Extreme poverty had led to the widespread adoption of harmful coping mechanisms - such as borrowing at high interest rates, consumption or sale of assets, and reducing investment in human capital. This will have long-term consequences, creating a cycle of poverty. While the assistance provided by REACH may have provided some cushion against the shock for the beneficiary households, and there is some assistance provided through the World Bank’s Approach Paper 2.0 as well as by the UN and humanitarian organizations, given the projected scale and duration of the impact, the risk to maintaining the outcome of the project is rated “High”. V. LESSONS AND RECOMMENDATIONS 67. Use of community structures facilitated rapid and accurate identification of beneficiaries and distribution of assistance. While the REACH project did not cover the geographic areas of CCAP, it built on the community mobilization and service delivery approach of CCAP. The REACH project either re-mobilized the CDCs that had been established under the NSP in rural areas, or by quickly mobilizing GAs and ad-hoc CDCs in other areas. The use of these community level structures proved to beessential. First, it helped to make the identification of beneficiary households faster, more accurate, and transparent in the context in which there was no formal system that could provide reliable, updated data on households and individuals. Mobilizing the CDCs and other community groups enabled decentralized, direct data collection on the ground, which helped to gather data much more efficiently and accurately than relying on a limited number of local government staff who did not necessarily know the communities well. 36 According to TPMA, the community driven approach helped to minimize conflict and tensions throughout the programme as community leaders were able to ensure that households were included that may have otherwise not been. Second, community-based procurement and distribution modalities also helped to expedite the distribution of assistance as communities could conduct these activities on their own (with the support of FPs) and simultaneously, rather than waiting for the local authorities or the PIU to undertake these activities in 36According to TPMA, almost all key informants interviewed (92 percent, 2,583 out of 2,799 between CCAP and REACH) reported being involved in the beneficiary selection process. Page 29 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) alignment with the regular procurement processes, community by community. CDCs’ existence over the past 18 years and their continuation even after the political crisis in August 2021 reflect the ability and resilience of this community platform to help deliver services in the most difficult times and foster inclusive development for Afghan women and men. The investments by the donor community in supporting these community-based institutions over the past several years have provided a foundation to continue delivering services to the most vulnerable. The fact that CDCs are operating now in the current political context and helping to provide assistance in 2022 is testament to their durability, popularity, and resilience. 68. Targeting from the top simplified beneficiary identification and helped to reduce operation costs and exclusion errors. The REACH project targeted all households in Afghanistan with incomes of US$2 per day or lower, covering an estimated 90 percent of households in targeted communities (with local flexibility). To efficiently identify and exclude a minimal proportion of wealthy households who do not require support, the project adopted an approach called “Targeting from the Top”. In this approach, a set of easily (mostly visibly) assessable and verifiable criteria to identify about 10 percent of households that did not require assistance was developed, instead of using eligible criteria to identify and include eligible households. The excluded households were then given an opportunity to claim their eligibility in case they believed they were eligible. This helped to: (a) minimize the generation of social tensions stemming from the grievances of excluded needy households; (b) reduce the costs and time required to identify the beneficiary households; and (c) reduce the possibilities of exclusion errors that could be caused by using eligibility criteria that did not adequately capture the characteristics of all truly needy households. 69. Door-to-door delivery enabled almost full reach to female-headed households. The project succeeded in delivering assistance to almost all female-headed households. This was possible since the FPs and communities organized door-to-door delivery for female-headed households in consideration of social norms and women’s safety. In the context in which the availability of female staff is limited, women’s mobility is restricted, and safe and convenient place and time cannot be made available for women to pick- up their assistance, the door-to-door delivery method proved to be practical and effective, resulting in 99 percent of female-headed households receiving assistance in areas monitored by TPMA. 70. Partnership with the humanitarian community helped to bridge the humanitarian-development divide. By closely communicating with the humanitarian community to coordinate the contents, modality, coverage, and the timing of assistance, the project has led to a framework for future cooperation and coordination in the country. In this regard, the involvement of the NGOs as FPs, who have always played roles in both humanitarian and development interventions, was helpful for sharing information on risks and feasibility of the operation, and coordinating implementation enhancement and risk mitigation measures with the initiatives of the humanitarian community. Further consultation with the NGOs that are familiar with the unique operational context of KM during the design phase may have helped to improve the design of the project, where the structure of community, nature of relationships among households, and the form of households differed from other areas and new community groups could not be mobilized the same way. . . Page 30 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: Support to Rural and Peri-urban Households Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Households receiving in-kind Number 0.00 1,760,000.00 294,288.00 support in rural or peri-urban areas 17-Aug-2020 17-Aug-2020 15-Aug-2021 Of which female-headed Number 0.00 176,000.00 5,090.00 17-Aug-2020 17-Aug-2020 15-Aug-2021 Comments (achievements against targets): Objective/Outcome: Support to Urban Households in Provincial Capital Cities Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Households receiving in-kind Number 0.00 360,000.00 179,872.00 Page 31 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) or cash support in provincial 17-Aug-2020 17-Aug-2020 15-Aug-2021 capital cities Of which female-headed Number 0.00 36,000.00 1,072.00 17-Aug-2020 17-Aug-2020 15-Aug-2021 Comments (achievements against targets): Objective/Outcome: Support to Households in Kabul Municipality Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of households Number 0.00 500,000.00 0.00 receiving in-kind or cash support in Kabul 17-Aug-2020 17-Aug-2020 15-Aug-2021 Of which female-headed Number 0.00 50,000.00 0.00 17-Aug-2020 17-Aug-2020 15-Aug-2021 Comments (achievements against targets): A.2 Intermediate Results Indicators Component: Project Effectiveness, Efficiency, and Operations Management Indicator Name Unit of Measure Baseline Original Target Formally Revised Actual Achieved at Page 32 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Target Completion Field staff/ Facilitators Number 0.00 2,000.00 1,936.00 trained in COVID-19- compliant and gender- 17-Aug-2020 17-Aug-2020 15-Aug-2021 sensitive distribution of transfers to households Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percent of field staff who are Percentage 0.00 25.00 16.00 female 17-Aug-2020 17-Aug-2020 15-Aug-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of communities Number 0.00 14,000.00 360.00 provided with support within 6 months of project 17-Aug-2020 17-Aug-2020 15-Aug-2021 effectiveness Comments (achievements against targets): Page 33 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Beneficiaries satisfied with Percentage 0.00 80.00 0.00 support received 17-Aug-2020 17-Aug-2020 15-Aug-2021 Comments (achievements against targets): Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Grievances Addressed Percentage 0.00 70.00 92.00 17-Aug-2020 17-Aug-2020 15-Aug-2021 Comments (achievements against targets): A total of 7,959 out of 8,664 grievances resolved. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Households reached through Number 0.00 2,800,000.00 185,000.00 the Government’s COVID-19 relief communications 17-Aug-2020 17-Aug-2020 15-Aug-2021 Page 34 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) campaign Comments (achievements against targets): Page 35 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) B. KEY OUTPUTS BY COMPONENT Objective/Outcome 1: Support to Rural and Peri-urban Households Outcome Indicators 1. Households receiving in-kind support in rural or peri-urban areas (Number) 1. Number of communities provided with support within 6 months of project Intermediate Results Indicators effectiveness (Number) 2. Beneficiaries satisfied with support received (Percentage) 1. Field staff/ Facilitators trained in COVID-19-compliant and gender-sensitive distribution of transfers to households (Number) Key Outputs by Component 2. Percent of field staff who are female (Percentage) (linked to the achievement of the Objective/Outcome 1) 3. Grievances Addressed (Percentage) 4. Households reached through the Government’s COVID-19 relief communications campaign (Number) Objective/Outcome 2: Support to Urban Households in Provincial Capital Cities 1. Households receiving in-kind or cash support in provincial capital cities Outcome Indicators (Number) 1. Number of communities provided with support within 6 months of project Intermediate Results Indicators effectiveness (Number) 2. Beneficiaries satisfied with support received (Percentage) 1. Field staff/ Facilitators trained in COVID-19-compliant and gender-sensitive distribution of transfers to households (Number) Key Outputs by Component 2. Percent of field staff who are female (Percentage) (linked to the achievement of the Objective/Outcome 2) 3. Grievances Addressed (Percentage) 4. Households reached through the Government’s COVID-19 relief communications campaign (Number) Page 36 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Objective/Outcome 3: Support to Households in Kabul Municipality Outcome Indicators 1. Households receiving in-kind or cash support in Kabul Municipality (Number) 1. Number of communities provided with support within 6 months of project Intermediate Results Indicators effectiveness (Number) 2. Beneficiaries satisfied with support received (Percentage) 1. Field staff/ Facilitators trained in COVID-19-compliant and gender-sensitive distribution of transfers to households (Number) Key Outputs by Component 2. Percent of field staff who are female (Percentage) (linked to the achievement of the Objective/Outcome 2) 3. Grievances Addressed (Percentage) 4. Households reached through the Government’s COVID-19 relief communications campaign (Number) Page 37 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Janamejay Singh, Andras Bodor Task Team Leader(s) Rahimullah Wardak, Kalesh Kumar Anandavalliamma Procurement Specialist(s) Karunakarakurup Ahmed Shah Ahmadzai Financial Management Specialist Syed Waseem Abbas Kazmi Financial Management Specialist Yasuhiko Matsuda Team Member Donald Herrings Mphande Team Member Erwin Willem Yvonnick Leon Knippenberg Team Member Jane Olga Ebinger Team Member Susan Wong Team Member Luiza A. Nora Team Member Zabihullah Ahrary Team Member Abdul Wali Ibrahimi Team Member Hyoung Gun Wang Team Member Ria Nuri Dharmawan Counsel Tahir Akbar Team Member Suhail Kassim Team Member Shubha Chakravarty Team Member Mohammad Yasin Noori Social Specialist Kosuke Anan Team Member Page 38 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Brigitta Bode Team Member Farishta Ali Team Member Sally Beth Murray Team Member Qais Agah Social Specialist Walker Turnbull Kosmidou-Bradley Team Member Ajmal Paiman Team Member Anna Kalashyan Team Member Anita Bimunka Takura Tingbani Environmental Specialist Amanullah Alamzai Team Member Edda Mwakaselo Ivan Smith Social Specialist Upamanyu Datta Team Member Aminata Ndiaye Team Member Obaidullah Hidayat Environmental Specialist Tobias Akhtar Haque Team Member Najla Sabri Social Specialist Jovitta Thomas Team Member Rasoul Rasouli Team Member Shaheer Shahriar Team Member Supervision/ICR Susan Wong, Andras Bodor Task Team Leader(s) Muhammad Abbass Rahimi, Rahimullah Wardak, Kalesh Procurement Specialist(s) Kumar Anandavalliamma Karunakarakurup Ahmed Shah Ahmadzai Financial Management Specialist Syed Waseem Abbas Kazmi Financial Management Specialist Camila Sorese Linneman Team Member Page 39 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Brigitta Bode Team Member Ghazal Shobair Procurement Team Farishta Ali Team Member Qais Agah Social Specialist Ajmal Paiman Team Member Anna Kalashyan Team Member Anita Bimunka Takura Tingbani Environmental Specialist Amanullah Alamzai Team Member Ahmad Shaheer Shahriar Team Member Ghulam Rasoul Rasouli Team Member Jovitta Thomas Team Member Aminata Ndiaye Team Member Obaidullah Hidayat Environmental Specialist Najla Sabri Social Specialist Kosuke Anan Team Member Olivier Lavinal Team Member Tahir Akbar Team Member Ria Nuri Dharmawan Counsel Abdul Wali Ibrahimi Team Member Anna Charlotte O'Donnell Team Member Erwin Willem Yvonnick Leon Knippenberg Team Member B. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY20 16.825 72,642.93 FY21 42.623 257,869.38 Page 40 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) Total 59.45 330,512.31 Supervision/ICR FY21 61.066 445,633.55 FY22 20.697 250,000.03 Total 81.76 695,633.58 Page 41 of 42 The World Bank COVID-19 Relief Effort for Afghan Communities and Households (REACH) (P174119) ANNEX 3. PROJECT COST BY COMPONENT Amount at Approval Actual at Project Percentage of Approval Components (US$M) Closing (US$M) (%) REACH Rural: Household Support in Rural and Peri- 136.00 105.00 77 urban Areas REACH Urban: Household Support in Provincial Capital 56.00 30.00 54 Cities REACH Kabul: Household 80.00 40.93 51 Support in Kabul Municipality Project Management, Communication, and 8.00 4.91 61 Monitoring Total 280.00 180.84 65 Page 42 of 42