Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) Report Number: ICRR0022888 1. Project Data Project ID Project Name P149965 UG-NUSAF3 Country Practice Area(Lead) Uganda Social Protection & Jobs L/C/TF Number(s) Closing Date (Original) Total Project Cost (USD) IDA-56450 31-Dec-2020 131,847,178.97 Bank Approval Date Closing Date (Actual) 27-May-2015 30-Jun-2021 IBRD/IDA (USD) Grants (USD) Original Commitment 130,000,000.00 0.00 Revised Commitment 130,000,000.00 0.00 Actual 131,847,178.97 0.00 Prepared by Reviewed by ICR Review Coordinator Group Judith Hahn Gaubatz Judyth L. Twigg Eduardo Fernandez IEGHC (Unit 2) Maldonado 2. Project Objectives and Components DEVOBJ_TBL a. Objectives According to the Financing Agreement (page 5) and the Project Appraisal Document (PAD, page 5), the project objectives were as follows:  To provide effective income support to and build the resilience of poor and vulnerable households in Northern Uganda. Page 1 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) b. Were the project objectives/key associated outcome targets revised during implementation? No c. Will a split evaluation be undertaken? No d. Components 1. Labor Intensive Public Works and Disaster Risk Financing (Approval: US$ 61.0 million; Actual: US$ 59.0 million): This component aimed to provide beneficiaries from poor and vulnerable households with seasonal financial transfers in return for their participation in labor-intensive public works, including building physical assets of value to local communities such as rural roads, tree nurseries, soil and water conservation measures, and marketplaces. Activities included: wages; equipment and semi-skilled labor for public works projects; capacity building for local community leaders; and unconditional transfers for households that do not have able-bodied members who can participate in the public works projects. There was also a pilot activity to temporarily scale up the labor-intensive public works projects following climatic disasters, in order to prevent drops in household consumption and to protect livelihoods. Activities included wages and the development of key mechanisms to trigger and operationalize the post-disaster response. 2. Livelihood Investment Support (Approval: US$ 43.5 million; Actual: US$ 45.5 million): This component aimed to provide livelihood support to poor and vulnerable households to help them improve livelihoods and further increase incomes. Activities included: comprehensive skills training to beneficiaries, livelihood grants, and mentoring. Grants of up to US$ 5,000 were provided to groups comprised of 10-15 households. There was also a pilot activity to develop community institutions to support household livelihood investments in the longer term, including revolving village funds. 3. Strengthening Transparency, Accountability and Anti-Corruption (Approval: US$ 5.0 million; Actual: US$ 5.0 million): This component aimed to support the Inspectorate of Government (IG) to expand its capacity at the local level to monitor and reduce misuse of funds. Activities included: sensitizing communities to prevent misuse of project funds, enhancing engagement of citizens and stakeholders, and monitoring of project activities at the community level, led by Community Monitoring Groups. 4. Safety Net Mechanisms and Project Management (Approval: US$ 20.5 million; Actual: US$ 20.5 million): This component aimed to develop the national social safety net program as envisioned in the draft Uganda Social Protection Policy, with particular emphasis on developing instruments to improve coordination across the government and development partners. e. Comments on Project Cost, Financing, Borrower Contribution, and Dates Project cost  The appraised project cost was US$ 130.0 million. The actual project cost was US$ 131.8 million. Financing Page 2 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965)  The project was financed by an IDA Credit of US$ 130.0 million, which disbursed in its entirety. Borrower contribution  There was no planned Borrower contribution. Dates  April 2018: The project was restructured to modify the results framework and to add a safeguard policy (OP 4.10) triggered due to the presence of a minority community in one of the project districts.  December 2020: The project closing date was extended from December 2020 to June 2021, due to COVID-related delays. 3. Relevance of Objectives Rationale The country of Uganda has been experiencing consistent economic growth and a significant reduction in poverty in the last two decades. However, a significant proportion of the population remains highly vulnerable to falling into poverty, particularly due to climatic and economic shocks. Socioeconomic indicators vary greatly by region, with the northern regions, which have been afflicted by conflict and violence, continuing to have high poverty headcounts and poor human development outcomes. In Northern Uganda, 80% of households rely on low productivity subsistence agriculture and are highly vulnerable to climatic events such as droughts, land degradation, and food insecurity; for example, in the Karamoja/ North-Eastern region, 74.2% of the population was classified as poor (2012/13). As noted in the PAD (page 2), economic growth alone thus far has been unable to bring prosperity to all, and therefore specific pro- poor interventions are needed to protect the poor and vulnerable. The government's recent National Development Plans 2 (2010-15) and 3 (2020-25) and the Peace and Recovery Development Plans all focus on increasing social protection for high-poverty regions, including by improving livelihoods. The prior Northern Uganda Social Action Fund projects 1 and 2, while more focused on post-reconstruction needs, included developing community infrastructure and providing livelihood support to increase incomes, including through a cash-for-work mechanism. The current third phase of the Social Action Fund was designed based on lessons learned from these previous phases, including having a longer-term goal to build resilience of households and improve quality of public works to better serve as safety nets. The project objectives are also consistent with the government's draft Social Protection Policy, which calls for the development of a sustainable system of social protection to protect the poor and vulnerable from negative effects of shocks. The Bank's most recent Country Partnership Framework (CPF, 2016-21) explicitly addresses the needs of the poor and vulnerable. One of the three Strategic Focus Areas is "Raising incomes in Rural Areas," including by expanding the use of safety nets, and one of the CPF objectives is "to enhance resilience of the poor and vulnerable." For the latter, outcome indicators include percentage of households sharing at least two basic assets. Page 3 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) Rating Relevance TBL Rating High 4. Achievement of Objectives (Efficacy) EFFICACY_TBL OBJECTIVE 1 Objective To provide effective income support to poor and vulnerable households in Northern Uganda Rationale The theory of change for this objective was clear. Labor-intensive public works projects were to be developed in local villages, providing temporary income opportunities for poor and vulnerable households. Livelihood grants and village revolving funds were also established as additional support to clusters of households to improve income opportunities. These activities were likely to lead to the intended outcome to provide income support to poor and vulnerable households by increasing the value of their household assets and investments. Targeting was a key element, as the project design implemented a three-layer targeting mechanism to ensure that support reached the poor and vulnerable; the first two layers relied on geographic targeting while the third relied on community-based targeting. The community-based targeting mechanism was used to select the poorest and most vulnerable households within those communities, with the community itself collectively selecting those households that were deemed most vulnerable according to clear criteria related to poverty and vulnerability. The labor intensive public works intervention was intended to reach the poorest households in the village, while the livelihood grants/village revolving funds were intended to reach the less poor households. The project covered 67 districts (out of 135 districts and 11 cities nationwide). Outputs  Public works projects in poor villages in 67 districts, which were intentionally labor-intensive and aimed to develop a physical asset of value to the local communities. According to the project team, at least 70% of the budget for the public works component was spent on wages for participants, and documentation at the community level was required to verify participation of and payment to community members. Also, the provision of technical inputs to the public works planning and implementation (such as semi-skilled labor) and annual reviews with assistance from the project's technical support team helped to ensure sufficient quality of the public works.  Financial transfers to 1,915,050 labor-intensive public works participants (target: 2,245,500) for 54 days of work at UGX 5500 per day (US$ 1.40). The ICR (page 13) noted that this was a significant amount of additional income given that 55.4% of participants lived on or less than US$ 1.90 per day. The transfers came with a mandatory savings requirement of 30% minimum of income Page 4 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) generated. Also, 2,423 households from the marginalized Ik population participated in the public works projects (target: 1,121).  Capacity building for participating communities, including "mindset training" to foster positive attitudes towards work, savings, and productivity; business skills development; and community leadership training.  Financial transfers to 223,565 beneficiaries (target: 207,500) from households with those unable to work, such as the elderly, disabled, chronically ill, pregnant or lactating mothers, orphans, and female- headed households. Also, 194 households from the marginalized Ik population received these unconditional transfers (target: 112).  Support for 9,449 livelihood subprojects (target: 7,700) in agriculture, aquaculture, livestock, tree nurseries, horticulture, and value addition, reaching 99,597 households. The subprojects provided skills training, livelihood grants (up to $5000), and mentoring support for existing and new market- driven enterprises. The grants came with a mandatory savings requirement of 30% minimum of profits. The skills training included construction techniques, cage fishing, cattle fattening, contour ploughing, tree planting, soil management, small-scale irrigation to produce crops 2-3 times a year, new crop production (e.g., coffee, matoke, tea), and use of new tools.  410 village revolving funds (serving 1,466 self-help groups comprised of 33,432 households) to support household livelihood investments tapping into already existing community savings groups and mobilizing other poor households to form savings groups of their own. Activities also included utilization of community business agents who provided ongoing mentoring support and technical assistance. (Note: The revolving funds were established in villages that were not already receiving livelihood subproject grants). In addition,  Implementation of a targeting mechanism to ensure that poor and vulnerable households were reached. Almost 100% of the project's desk officers and community facilitators had positive feedback on the community-based targeting mechanism used to select individuals for participation. According to the ICR (page 17), the utilization of this innovative and transparent mechanism for the selection of beneficiaries helped empower beneficiaries and officials from local levels of government.  Implementation of social accountability mechanisms. 90.3% of participating villages were utilizing community score cards (target: 85%); 60.3% of participating villages had functioning community monitoring groups (target: 70%); and 75.6% of grievances registered about the project were resolved in a timely manner (target: 70%), and of these grievances, 21.5% represented non-project activities.  Development of a management information system to maintain a registry of safety net beneficiaries. Outcomes The project benefitted 3,031,690 individuals from poor and vulnerable households, of which 58% were female (target: 2,995,500; 40% female).  25,564,410 person-days of employment were created, nearly achieving the target of 26,946,000.  The average household monthly income increased by 223%. Page 5 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965)  The percentage of households with savings increased from 54.6% to 70% (no target provided). The amount of savings varied, although the share of households in the lowest saving group (less than UGX 50,000) decreased from 46% to 22% while the share in the highest saving group (above UGX 200,000) increased from 13% to 30%. As reported in the ICR (page 14), the end-line evaluation showed that these savings enabled beneficiaries to obtain additional credit for productive activities (such as expansion of an enterprise or purchase of an asset) or consumption smoothing of household expenditure (school fees, burials, etc.).  The value of household assets increased by 30% between 2015 and June 2021, surpassing the target of 20%. This included a 45.6% increase in household ownership of livestock, and a 16% increase in household ownership of equipment such as ox ploughs. The percentage of households owning six or fewer selected assets decreased from 53% to 22%; the percentage of those owning seven to ten selected assets increased from 46% to 71%; and those with more than ten assets increased from 2% to 9%. Achievement of this objective is rated Substantial due to evidence of increased income generated, increased savings, and increased value of household assets. Rating Substantial OBJECTIVE 2 Objective To build the resilience of poor and vulnerable households in Northern Uganda Rationale The theory of change was clear. The labor-intensive public works projects created community assets (such as wells, irrigation systems, market stalls) that could be used for income/livelihood support, particularly in times of economic or climatic hardships (such as droughts). The pilot activity to develop a support mechanism for disaster relief mitigation was to provide protection to households and avoid disruptions in consumption. These activities were likely to contribute to the intended outcome to build resilience of poor and vulnerable households in Northern Uganda, particularly by increasing land productivity and income, increasing water supply during drought events, and providing access to safety nets during shocks. Resilience is defined in the PAD as the capacities of households and communities to adapt to a new strategy in the face of shocks and crises. As noted in the ICR (page 13), the outputs for this objective overlapped with those for the first objective, and the intended outcomes of both objectives also overlapped. Outputs As reported above: Page 6 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965)  Financial transfers to 1,915,050 labor-Intensive public works participants (target: 2,245,500) for 54 days of work on public works projects at UGX 5500 per day (US$ 1.40). The transfers came with a mandatory savings requirement of 30% minimum of income generated.  Support for 9,449 livelihood subprojects (target: 7,700) in agriculture, aquaculture, livestock, tree nurseries, horticulture, and value addition, reaching 99,597 households. The grants came with a mandatory savings requirement of 30% minimum of profits.  410 village revolving funds (serving 1,466 self-help groups comprised of 33,432 households) to support household livelihood investments. In addition,  Piloting of disaster relief fund (DRF) activities, by providing labor-intensive public works transfers to 90,405 vulnerable households (target: 84,000) experiencing a shock such as lack of rainfall impacting their crops, thereby providing a temporary source of income and protecting consumption/ assets. According to the project team, the implementation cycle for approved activities was based on an analysis of the seasonal calendar, to ensure that participants had the opportunity to work during the lean season when there would no other farming operations; also, financial transfers to participants were made on a bi-weekly basis to ensure timeliness of the support.  Development of system to collect and analyze data for triggers for the DRF program. Outcomes As reported above:  The percentage of households with savings increased from 54.6% to 70% (no target provided).  The value of household assets increased by 30% between 2015 and June 2021, surpassing the target of 20%. Also,  3,459 community assets were built through labor-intensive public works projects (target: 3,170), which increased incomes, land productivity (soil and water conservation, irrigation) and water supply, and access to safety nets. The average distance in kilometers to markets for general goods decreased by 14% and for livestock decreased by 11%.  The percentage of households engaged in business enterprises increased by 29%, an indication that livelihoods have diversified away from subsistence agriculture.  The percentage of households having one meal a day decreased from 23% to 10%, while the percentage of those having two meals a day increased from 53% to 61%, and those having three meals increased from 8% to 23%.  The Progress-out-of-Poverty Index decreased from 62% to 40%, measuring poverty levels by looking at selected household factors such as household size, access to education, and literacy levels, as well as ownership of common household assets like mobile phones, radio, and shoes, and the nature of the dwelling. Page 7 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) Achievement of this objective is rated Substantial due to evidence of increased savings, increased household assets, increased community assets, and increased household resilience and consumption. Rating Substantial OVERALL EFF TBL OBJ_TBL OVERALL EFFICACY Rationale Overall Efficacy is rated Substantial due to evidence of improved incomes and increased resilience, based on effective targeting to reach poor and vulnerable households. Overall Efficacy Rating Substantial 5. Efficiency At appraisal (PAD, Annex 6), the estimated economic rate of return was 21.4% with a net present value of $41.6 million. This was further broken down by food crops (19.4%), cash crops (28.0%), agribusiness (24.0%), and non-agricultural (14.0%). The estimate was calculated from the following benefits: increased and diversified household incomes from a broad range of livelihood sources; improved community assets and market integration; sustainable business institutions owned by the rural poor; and increased opportunities for self- employment and skills transfer. Most of these benefits were expected to be generated from the labor-intensive public works and livelihoods investment support components (approximately 75% of total project costs). Although the actual public works projects could not be identified at project appraisal, due to their demand-driven nature, random stakeholder consultations and experience from the prior two phases of the Social Action Fund informed the estimates. Other likely benefits related to social, institutional, capacity building, disaster risk management, and human resource development were not included in the analysis. At completion (ICR, Annex 4), the economic analysis was updated and resulted in an economic rate of return of 25.1%. The benefits were calculated based on economic returns from productive investments from labor- intensive public works projects, DRF sub-projects, the livelihood support grants, and the village revolving funds. Other likely benefits arising from improved community assets (such as roads) were not included due to difficulty in quantifying benefits. The implementation schedule progressed largely as planned, with a six-month project extension caused by COVID delays rather than by any internal disruptions. There were no significant implementation delays, fiduciary issues, or cost overruns. Although there was a lengthy delay in finalizing the 2018 restructuring, there were no Page 8 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) significant impacts on implementation as the restructuring related mainly to revisions to the safeguard policies. The loan disbursed fully largely within the original project period. Efficiency Rating Substantial a. If available, enter the Economic Rate of Return (ERR) and/or Financial Rate of Return (FRR) at appraisal and the re-estimated value at evaluation: Rate Available? Point value (%) *Coverage/Scope (%) 75.00 Appraisal  21.40  Not Applicable 58.00 ICR Estimate  25.10  Not Applicable * Refers to percent of total project cost for which ERR/FRR was calculated. 6. Outcome Relevance is rated High due to strong alignment with country conditions, country development strategy, and Bank strategy. Efficacy is rated Substantial due to evidence of improved incomes and increased resilience, based on effective targeting to reach poor and vulnerable households. Efficiency is rated Substantial due to a reasonable economic rate of return on core activities and minimal implementation inefficiencies. The ratings indicate that there were only minor shortcomings in the project's preparation, implementation, and results, producing an Outcome rating of Satisfactory. a. Outcome Rating Satisfactory 7. Risk to Development Outcome The risk to development outcomes lies mainly in the sustainability of the economic activities that were financed under the livelihood support component. The sustainability of livelihood enhancement activities, such as small businesses and market enterprises, is unclear, given market demand uncertainties, narrow profit margins, lack of ongoing technical support and mentorship, and a lack of enforcement of group functioning practices. Also, there is some risk of sustainability of community assets. Although communities were expected to take responsibility for the maintenance of community assets built through the labor- intensive public works support, it is not clear whether this is actually taking place. A key follow-up Bank project -- Generating Livelihoods and Opportunities for Women (GLOW – P176747) Project -- was under preparation at the time of ICR writing, which also focuses on provision of training and women’s access to Page 9 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) finance and investment in social and economic infrastructure, but a gap in financial support from the Bank is likely. However, institutional development gains are likely to be sustained. The project supported the development of community-level institutions, such as self-help groups and village livelihood committees. The ICR noted that non-government organizations are actively targeting community groups established under this project as part of their own delivery structures. In addition, the project developed mechanisms and protocols that will continued to be used, including national guidelines for the implementation of labor-intensive public works, proof of concept of the DRF and triggering rules and procedures for its replication and scale-up, proof of concept of the village revolving fund and its potential for replicability and scalability, strengthening of the IG at the local level for governance and transparency, and mechanisms to support a national social safety net program. 8. Assessment of Bank Performance a. Quality-at-Entry This project was a continuation -- the third phase -- of a series of operations to improve livelihoods and incomes for vulnerable populations in the northern regions; however, there were additional elements of disaster risk mitigation, replicability, and sustainability addressed through introducing significant innovations. Similarly, the project used national and local government entities that had gained experience from the prior phases, while also aiming to build capacity at the community level to increase ownership and sustainability of project interventions. According to the ICR (page 19), local governments, districts in particular, played a central role in the development of community institutions and project implementation as a whole, and in ensuring accuracy and quality of the monitoring data produced at lower levels of government. The community-based approach helped communities mobilize and actively participate in collective decision making regarding the selection of beneficiaries, the prioritization of community needs, and the selection and implementation of labor-intensive public works sub-projects. The theory of change and the results framework were clear, but the latter was mostly comprised of key project indicators that were output-oriented (number of beneficiaries), with only one key indicator measuring the actual intended outcome to increase household assets. Several elements of the project objectives could have been captured as key indicators, such as increased income, savings, access to community assets, and consumption, but were instead reported through evaluations. Quality-at-Entry Rating Satisfactory b. Quality of supervision Supervision was regular and proactive, with ongoing technical support provided to the implementing agencies through constant communication, several technical supervision missions, and a comprehensive Mid-Term Review (MTR). Significant capacity building support was provided by the Bank team at the Page 10 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) district government level. This was facilitated by minimal task team leadership changes and local presence of a Bank team member. As noted in the ICR (page 27), "the good and harmonious working relationship between the Bank and government teams played an important role in the success of the project." The MTR in particular was used to assess progress on the two pilot activities, which were both subsequently expanded for increased impact. Implementation progress was well documented in supervision documents. Safeguards and fiduciary issues were effectively managed, including noting the identification of a minority community in one of the project areas and implementing necessary safeguard measures. M&E implementation was well-executed, with clear use of M&E data to inform decision making. Quality of Supervision Rating Satisfactory Overall Bank Performance Rating Satisfactory 9. M&E Design, Implementation, & Utilization a. M&E Design The theory of change and the results framework were clear and relevant. Although the key project indicators were linked to the project objectives, they mainly focused on outputs. While these indicators facilitated project implementation progress monitoring, they were less robust in assessing achievement of objectives. Other outcome-oriented indicators, such as increased savings, increased access to community assets, and increased consumption, could have been included. A project management information system was included in the project design, in part to facilitate project monitoring but also to inform subsequent steps to develop a national social protection system. Evaluative activities (such as impact assessments of the core interventions) were planned to provide additional data and information about the project's impact. b. M&E Implementation The project management information system was effectively implemented and used. A community-based M&E system, relying on reporting templates, was implemented and produced regular reports. The ICR (page 24) noted that continuous training of district officials and community facilitators, as well as quality assurance mechanisms, were used to help ensure quality and consistency of data. Evaluations were carried out as planned by Makerere University at baseline, midpoint, and endline, with a total of 3,689 beneficiaries randomly selected. c. M&E Utilization According to the ICR (pages 24-25), the M&E system was a key tool in supporting implementation. Based on M&E progress reports, the implementing agencies were able to identify implementation gaps Page 11 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) and prepare quarterly technical support plans to provide critical support to districts. Management information system data on beneficiaries was also utilized to inform the Single Beneficiary Registry. M&E Quality Rating Substantial 10. Other Issues a. Safeguards The project was classified as an Environmental Category "B" project, with limited environmental and social impacts expected from the public works projects. Safeguards triggered included: Environmental Assessment (OP/BP 4.01), Pest Management (OP 4.09), and Physical Cultural Resources (OP/BP 4.11). Site-specific Environmental and Social Management Plans were developed and implemented, with two subprojects undergoing full Environment and Social Impact Assessments. According to the ICR (page 25), all subprojects were screened for environmental and social risks and adhered to safeguard requirements, using an integrated watershed management implementation model. During the May 2016 supervision mission, the safeguard policy OP 4.10 on Indigenous Peoples was triggered due to the identification of the Ik people, a minority community found in Kaabong District. Although there were delays in the formal approval of the project restructuring (required due to the addition of the safeguard and amendment of the Credit Agreement) due to a lengthy in-country review process by the government, preparation and implementation of the Vulnerable and Marginalized Peoples Action Plan immediately took place. No problems were reported in safeguards compliance. b. Fiduciary Compliance Financial management: According to the ICR (pages 25-26), the operation’s financial management arrangements were rated satisfactory for most of the implementation period. including staffing in compliance with agreed implementation plans. Interim unaudited financial reports and external audit reports were submitted in a timely and acceptable manner. There were no qualifications to audits. Procurement: At the central and local levels, procurement processes largely involved goods and services, and were conducted in accordance with the Bank's procurement guidelines. There were two complaints noted relating to motor vehicle procurements, which required administrative reviews and were resolved. At the community level, sub-project procurement was handled by Community Project Management Committees. Recurrent challenges at the community level were continually addressed throughout the operation's lifetime, including low literacy levels of committee members that increased susceptibility to manipulation in decision making; inadequate record keeping of procurement processes; and political interference due to the direct linkage of the implementing communities with the Page 12 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) political leadership of the areas. There were no cases of misprocurement reported. Regular procurement reviews were largely satisfactory. c. Unintended impacts (Positive or Negative) None reported. d. Other --- 11. Ratings Reason for Ratings ICR IEG Disagreements/Comment Outcome Satisfactory Satisfactory Bank Performance Satisfactory Satisfactory Quality of M&E Substantial Substantial Quality of ICR --- High 12. Lessons Lessons drawn from the ICR (pages 29-30), adapted by IEG:  The identification of a clear trigger mechanism for disaster-relief financial disbursements, agreed upon in advance, can ensure rapid response and flow of funds to beneficiaries once a shock occurs. In the case of this project, the overall strategy was in place, with readily available financing and disbursement mechanisms, in addition to the identification of independent data sources such as satellite-based observations of ground vegetation and food security indices that allowed for rapid and transparent targeting of needy participants.  Community-level revolving funds can be an effective intervention to ensure sustainability of livelihoods once project support ends. In the case of this project, the village revolving funds provided continued access to finance and sustained development of community institutions.  Operations and maintenance of community assets are more likely to be sustained by taking into account the constraints of local communities. In the case of this project, despite strong ownership by communities, robust operation and maintenance plans with intensive follow-up by the project team were needed, including understanding the constraints in leadership, financial management, workload, and other factors faced by those local communities to follow through on maintenance plans. Page 13 of 14 Independent Evaluation Group (IEG) Implementation Completion Report (ICR) Review UG-NUSAF3 (P149965) 13. Assessment Recommended? No 14. Comments on Quality of ICR The quality of the ICR was high, notable for its candor, conciseness, outcome orientation, and quality of evidence. The theory of change was clearly laid out, and the analysis of the results chain was thorough. The analysis of results went beyond the PDO indicators (which were mostly output-oriented) and drew upon additional evidence from other sources to substantiate increases in household savings, assets, consumption, resilience, etc. The quality of this additional evidence was highly satisfactory. Lessons were well drawn from project experience, including learning from incremental changes during this third phase of the series of social protection operations. a. Quality of ICR Rating High Page 14 of 14