Document of The World Bank FOR OFFICIAL USE ONLY Report No: ICR00005508 IMPLEMENTATION COMPLETION AND RESULTS REPORT CREDIT NUMBER IDA 5732-MZ ON A CREDIT IN THE AMOUNT OF SDR (28.7) MILLION (US$ 40.0 MILLION EQUIVALENT) AND AN ADDITIONAL GRANT IN THE AMOUNT OF US$ 20 MILLION AND A SECOND ADDITIONAL GRANT IN THE AMOUNT OF US$ 13 MILLIION TO THE Republic of Mozambique FOR THE Emergency Resilient Recovery Project 30 May 2022 Urban, Disaster Risk Managament, Resilience and Land Global Practice Africa Region CURRENCY EQUIVALENTS Exchange Rate Effective May 30, 2022 Currency Unit = New Mozambique Metical (MZN) MZN 63.87 = US$1 US$1 = SDR 0.740905 FISCAL YEAR January 1 - December 31 Regional Vice President: Hafez M. H. Ghanem Country Director: Idah Z. Pswarayi-Riddihough Regional Director: Catherine Signe Tovey Practice Manager: Catalina Marulanda Lizardo Narvaez Marulanda and Xavier Agostinho Task Team Leader(s): Chavana ICR Main Contributor: Xavier Agostinho Chavana ABBREVIATIONS AND ACRONYMS 3CP Cities and Climate Change Project ANE National Roads Administration ARA Regional Water Administration A-RAP Abbreviated Resettlement Action Plan ADM Administrative Decision-Making (TTL) AF Additional Financing AIAS Water and Sanitation Infrastructure Administration AIS Activity Initiation Summary ARA Regional Water Administration AURA Water Regulatory Authority CCA Climate Change Adaptation BP Bank Policy CERC Contingency Emergency Response Component CERRP Cyclone Idai and Kenneth Emergency Recovery and Resilience Project CMU Country Management Unit CPF Country Partnership Framework DFID Department of International Development DGBL Limpopo Basin Management Division DIEE Schools Infrastructure and Equipment Department DNAAS National Directorate for Water Supply and Sanitation DNGRH National Directorate for Water Resources Management DRM Disaster Risk Management DRR Disaster Risk Reduction ERRP Emergency Resilient Recovery Project ESIA Environmental and Social Impact Assessment ESMBPM Environmental and Social Management Best Practices Manual ESMF Environmental and Social Management Framework ESMP Environmental and Social Management Plan ESSP Education Sector Support Program FASE Education Sector Support Fund FIPAG Water Supply Assets Investment Fund FM Financial Management GBV/SEA Gender-based Violence/Sexual Exploitation and Abuse GDP Gross Domestic Product GEMS Geo-Enabling Initiative for Monitoring and Supervision GFDRR Global Facility for Disaster Reduction and Recovery GRM Grievance Redress Mechanism INE National Statistics Institute INGD National Institute for Disaster Management and Risk Reduction INIR National Institute for Irrigation IOF Households Expenditures Survey IPMP Integrated Pest Management Plan IRM Immediate Response Mechanism IRR Internal Rate of Return IRRIGA Smallholder Irrigated Agriculture and Market Access Project ISM Implementation Support Mission ISR Implementation Status Report M&E Monitoring and Evaluation MEF Ministry of Economy and Finance MINEDH Ministry of Education and Human Development MISAU Ministry of Health MOPHRH Ministry of Public Works, Housing and Water Resources MTR Mid-Term Review NGO Non-Governmental Organization NPV Net Present Value OHS Occupational Health and Safety OP Operational Policy PACE Small Emerging Commercial Farmer PAP Project Affected People PDO Project Development Objective PforR Program for Results PIM Project Implementation Manual PIU Project Implementation Unit PPP Purchasing Power Parity PROIRRI Mozambique Sustainable Irrigation Project PSC Project Steering Committee RAP Resettlement Action Plan RF Results Framework RPF Resettlement Policy Framework RSA Regional Safeguards Advisor S-ESIA Simplified Environmental and Social Impact Assessment SORT Systematic Operations Risk Rating Tool SUSTENTA Agriculture and Natural Resources Landscape Management Project TA Technical Assistance TTL Task Team Leader UN-Habitat United Nations Program for Human Settlements TABLE OF CONTENTS DATA SHEET ................................................................................................................................1 I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES .........................................................6 A. CONTEXT AT APPRAISAL ............................................................................................................. 6 B. SIGNIFICANT CHANGES DURING IMPLEMENTATION................................................................. 10 II. OUTCOME .........................................................................................................................11 A. RELEVANCE OF PDOs ................................................................................................................ 11 B. ACHIEVEMENT OF PDOs (EFFICACY) .......................................................................................... 13 C. EFFICIENCY ................................................................................................................................ 18 D. JUSTIFICATION OF OVERALL OUTCOME RATING....................................................................... 19 E. OTHER OUTCOMES AND IMPACTS ............................................................................................ 19 III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME ..................................20 A. KEY FACTORS DURING PREPARATION ................................................................................... 20 B. KEY FACTORS DURING IMPLEMENTATION ............................................................................ 21 IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME ...22 A. QUALITY OF MONITORING AND EVALUATION (M&E) .............................................................. 22 B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE ....................................................... 24 C. BANK PERFORMANCE ............................................................................................................... 25 D. RISK TO DEVELOPMENT OUTCOME .......................................................................................... 25 V. LESSONS AND RECOMMENDATIONS .................................................................................26 ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS ..............................................................29 ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION...........................44 ANNEX 3. PROJECT COST BY COMPONENT ...............................................................................47 ANNEX 4. EFFICIENCY ANALYSIS ...............................................................................................48 ANNEX 5. BORROWER AND OTHER PARTNER COMMENTS ......................................................52 ANNEX 6. SUPPORTING DOCUMENTS ......................................................................................54 The World Bank MZ - Emergency Resilient Recovery Project (P156559) DATA SHEET BASIC INFORMATION Product Information Project ID Project Name P156559 MZ - Emergency Resilient Recovery Project Country Financing Instrument Mozambique Investment Project Financing Original EA Category Revised EA Category Partial Assessment (B) Partial Assessment (B) Organizations Borrower Implementing Agency Ministry of Education and Human Development (MINEDH), National Directorate of Water Resources Ministry of Economy and Finance (MEF) Management (DNGRH), National Institute for Irrigation (INIR), Water and Sanitation Infrastructure Administration (AIAS) Project Development Objective (PDO) Original PDO The Project Development Objective is to restore the functionality of critical infrastructure in a resilient manner in the disaster-affected provinces; and to improve the Government of Mozambique's capacity to respond promptly and effectively to an eligible crisis or emergency. Page 1 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) FINANCING Original Amount (US$) Revised Amount (US$) Actual Disbursed (US$) World Bank Financing 40,000,000 39,753,287 39,415,321 IDA-57320 20,000,000 20,000,000 20,938,319 IDA-59510 13,000,000 12,350,911 12,557,956 IDA-D3960 Total 73,000,000 72,104,198 72,911,596 Non-World Bank Financing 0 0 0 Borrower/Recipient 0 0 0 Total 0 0 0 Total Project Cost 73,000,000 72,104,198 72,911,597 KEY DATES Approval Effectiveness MTR Review Original Closing Actual Closing 29-Sep-2015 18-Feb-2016 29-Jul-2018 30-Jun-2019 30-Nov-2021 RESTRUCTURING AND/OR ADDITIONAL FINANCING Date(s) Amount Disbursed (US$M) Key Revisions 16-Nov-2016 3.00 Change in Components and Cost Reallocation between Disbursement Categories 02-Nov-2017 23.84 26-Jun-2020 49.63 Change in Results Framework Change in Loan Closing Date(s) Change in Implementation Schedule 20-Nov-2020 60.59 Change in Loan Closing Date(s) 27-May-2021 71.01 Change in Loan Closing Date(s) KEY RATINGS Outcome Bank Performance M&E Quality Moderately Satisfactory Satisfactory Substantial Page 2 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) RATINGS OF PROJECT PERFORMANCE IN ISRs Actual No. Date ISR Archived DO Rating IP Rating Disbursements (US$M) 01 09-Feb-2016 Satisfactory Satisfactory 0 02 23-Sep-2016 Moderately Satisfactory Moderately Satisfactory 2.50 03 11-Apr-2017 Moderately Satisfactory Moderately Satisfactory 14.88 04 01-Nov-2017 Moderately Satisfactory Moderately Satisfactory 23.84 05 01-Jun-2018 Moderately Satisfactory Moderately Satisfactory 31.20 06 10-Dec-2018 Moderately Satisfactory Moderately Satisfactory 33.45 07 18-Jun-2019 Moderately Satisfactory Moderately Satisfactory 36.70 08 20-Dec-2019 Moderately Satisfactory Moderately Satisfactory 44.32 09 24-Jun-2020 Moderately Satisfactory Moderately Satisfactory 49.63 10 16-Dec-2020 Satisfactory Moderately Satisfactory 61.47 11 16-Jun-2021 Satisfactory Satisfactory 71.38 SECTORS AND THEMES Sectors Major Sector/Sector (%) Agriculture, Fishing and Forestry 19 Irrigation and Drainage 19 Public Administration 12 Other Public Administration 12 Education 24 Other Education 24 Page 3 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Transportation 4 Other Transportation 4 Water, Sanitation and Waste Management 41 Water Supply 23 Other Water Supply, Sanitation and Waste 18 Management Themes Major Theme/ Theme (Level 2)/ Theme (Level 3) (%) Finance 10 Finance for Development 10 Disaster Risk Finance 10 Urban and Rural Development 63 Rural Development 33 Rural Infrastructure and service delivery 33 Disaster Risk Management 30 Disaster Response and Recovery 10 Disaster Risk Reduction 10 Disaster Preparedness 10 Environment and Natural Resource Management 111 Climate change 84 Adaptation 84 Water Resource Management 27 Water Institutions, Policies and Reform 27 ADM STAFF Role At Approval At ICR Regional Vice President: Makhtar Diop Hafez M. H. Ghanem Zviripayi Idah Pswarayi Country Director: Mark R. Lundell Riddihough Page 4 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Director: Ede Jorge Ijjasz-Vasquez Catherine Signe Tovey Practice Manager: Sameh Naguib Wahba Tadros Catalina Marulanda Michel Matera, Jean Baptiste Lizardo Narvaez Marulanda, Task Team Leader(s): Migraine Xavier Agostinho Chavana ICR Contributing Author: Xavier Agostinho Chavana Page 5 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) I. PROJECT CONTEXT AND DEVELOPMENT OBJECTIVES A. CONTEXT AT APPRAISAL Context 1. In July 2015, when the Emergency Resilience Recovery Project (ERRP) was appraised, Mozambique was one of the poorest countries in the world, despite rapid economic growth. The average annual Gross Domestic Product (GDP) growth rate was 7.2 percent between 2000 and 2016, making Mozambique the fastest-growing economy in Sub-Saharan Africa. Growth was mostly driven by large-scale foreign investment projects in a context of political stability, and by significant donor support. GDP per capita grew by 4 percent a year, on average over the same period, climbing from US$561 to US$1,128 (2011 PPP). Poverty declined from 60.3 percent in 2003 to 48.4 percent in 2015. However, the country ranked 178 out of 187 on the 2013 Human Development Index and the lowest in Southern Africa. 2. At the time, about 70 percent of country’s population lived in rural areas, and depended on subsistence agriculture; 38 percent were malnourished and only 42 percent had access to an improved water source. Poverty was highest in rural areas, affecting 56.0 percent of rural population in 2015, compared to 32.0 percent in urban areas. Poverty was more profound in the northern and central provinces (66.7 percent in Niassa and 61.76 percent in Zambezia), while the provinces in southern Mozambique experienced moderate poverty reduction. In Gaza province, poverty had declined from 73.64 percent in 2003 to 43.62 percent in 2015, while in Maputo city, it had declined from 28.13 percent in 2003 to 3.83 percent in 2015, reflecting spillover effects from the rapid growth of urban centers in the region. 3. Mozambique experienced some of the Southern Africa’s most variable hydrological and meteorological conditions. The country was particularly exposed to cyclones and flooding and ranked the third most at risk from water and weather-related hazards in Africa. About 58 percent of the population and more than 37 percent of GDP were exposed to two or more hazards, which had caused annual losses averaging 1.1 percent of GDP by 2015. Flooding caused annual average losses of US$17.5 million to housing, US$0.7 million to roads and bridges and US$42.5 million to maize crops, while cyclones cause annual average losses of US$2.1 million to education infrastructure. Key challenges to building national and local capacity for disaster risk management (DRM) included improved dissemination of early warnings to vulnerable communities, systematic mainstreaming of DRM in government policies and programs, and strengthening of coordination and monitoring and evaluation (M&E) of recovery and reconstruction interventions. 4. The project was well aligned with national priorities. At appraisal, the Government of Mozambique (GoM) was focused on improving its legal and institutional disaster risk management DRM frameworks, with the approval in June 2014 of the first DRM Law (Law no. 15/2014, of June 20). Operationally, the country was focused on implementation of the Master Plan for Natural Disasters Prevention and Mitigation (2006- 2016), which linked disaster prevention, mitigation, and recovery to reducing vulnerability of the population, infrastructure, and the economy. At the policy level, disaster risk reduction (DRR) and climate change adaptation (CCA) were strategic objectives of the Five-Year Government Program (2015-2019), approved by Parliament in April 2015 through the Resolution no. 12/2015, of April 14. 5. The operation was well aligned with the Country Partnership Framework (CPF) for FY2012-2015. The project contributed directly to Pillar 2 of the CPF related to decreasing vulnerability and increasing resilience, and specifically to its second objective of improving resilience to natural disasters and the impacts of climate change. The CPF also identified the need to improve the competitiveness of the Mozambican economy, by Page 6 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) targeting interventions toward recovery and reconstruction in priority sectors, particularly agriculture, water and sanitation and education, in order to restore productive capacities and basic public services. 6. The original project was designed for recovery from cyclones and flooding in central and northern Mozambique. In January 2015, the central and northern provinces of Zambezia, Nampula and Niassa were hit by a tropical depression that caused strong winds and heavy rainfall and triggered a 30-year flood in the Licungo river basin, in Zambezia province. Damages in Nampula and Niassa provinces were mostly caused by strong winds and heavy rains, while flood waters reaching 11.9 meters at the Mocuba gauge station in the Licungo river were the primary cause of damages in Zambezia province. 7. The scope of the original project was selected based on prioritization of emergency needs. In March 2015, a joint damages assessment and early recovery was conducted by the Government of Mozambique, the World Bank, the United Nations and European Union. The assessment found that 326,000 people were affected and 140 people killed. Damages included: 30,000 houses, 2,362 classrooms, and 17 health units partially or totally destroyed, and more than 104,430 hectares of crops lost affecting 102,000 farmers. Damages were extensive to the dikes of Nante and Nicoadala, the irrigation schemes of Munda-Munda and Intabo and associated power line and access road; the water supply system of Mocuba, and the hydrometeorological network in the Licungo river basin. The cost of damages was estimated at US$384 million, around 2.4 percent of Gross Domestic Product (GDP), and total recovery and reconstruction needs were estimated at US$485 million. Therefore, the project was designed to focus on restoration of basic productive capacities, rehabilitation of public services and strengthening the capacity for the medium and long-term response. The Netherlands supported the emergency repair of the water supply system in Mocuba. The original project further supported the resilient rehabilitation of this system. Theory of Change (Results Chain) 8. A theory of change was not required at appraisal for the original project and the two Additional Financings. However, the theory of change has been derived based on the project design at closing. Figure 1. Theory of change Components Activities Outputs Outcomes Immediate Long-term Component A: Rehabilitation • Km of dikes rehabilitated • Number of Resilient and • Km of power line rehabilitated hectares of infrastructure construction • Number of hectares of agricultural land rehabilitation of resilient irrigated land rehabilitated protected from infrastructure • Number of classrooms flooding rehabilitated or constructed • Number of people Restored • Number of water systems protected from functionality and rehabilitated or constructed flooding resilience of • Number of domestic water • Number of people infrastructure to connections completed provided with recurrent • Km of water distribution access to cyclones, floods, network rehabilitated irrigation, and droughts • Dam and weir rehabilitated education, and water supply infrastructure Component B. Training of • Number of training sessions • Number of people, Technical government, delivered firms, NGOs, and Page 7 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Assistance for constructors, communities resilient engineering • Number of people trained equipped with Recovery and firms, and knowledge and Vulnerability NGOs staff • Number of guidelines or skills on resilient Reduction and local manuals produced construction community techniques members on resilience construction techniques Development of guidelines and manuals on resilient infrastructure rehabilitation Component D. Construction • Number of desalination plants • Number of people Improved capacity Contingent of critical built provided with for effective and Emergency infrastructure • Tons of food, seeds, and access to clean timely response to Response nutrition supplies purchased water crises or Component Provision of and distributed • Number of people emergencies (CERC) emergency provided access to support to • Number of irrigations kits food, seeds, and affected purchased and distributed irrigation kits people • Number of people treated of malnutrition Project Development Objectives (PDOs) and Outcomes Indicators 9. The Project Development Objectives were to restore the functionality of critical infrastructure in a resilient manner in the disaster-affected provinces, and to improve the Government of Mozambique’s capacity to respond promptly and effectively to an eligible crisis or emergency. No changes were made to the PDOs throughout project implementation. 10. The original PDO Results Indicators were the following: a. Direct Project beneficiaries, percent of which female beneficiaries b. Number of people protected by rehabilitated dike infrastructure c. Number of people with access to improved irrigation infrastructure d. Number of people in urban areas provided with access to improved water sources under the Project e. Number of children with access to improved education infrastructure, and f. Number of people supported by early warning and response systems. 11. Revision of PDO indicators: The PDO indicators were revised at the restructuring in 2018 to reflect the changes in Project scope following the activation of the Immediate Response Mechanism (IRM), in 2016 and the first Additional Financing, in 2017. The indicator related to “Number of people supported by early warning and response systems” was dropped in 2018. The final PDO indicators were: a. Direct Project beneficiaries, percent of which female beneficiaries b. Number of people protected by rehabilitated dike infrastructure c. Number of people with access to improved irrigation infrastructure d. Number of people in urban areas provided with access to improved water sources under the Project Page 8 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) e. Number of children with access to improved education infrastructure. 12. Changes were also made to targets of PDO Indicators and intermediate indicators at the First and Second Additional Financings and at restructuring in 2018 and 2020. Annex 1 presents the final Results Framework at project closing. Annex 6B presents the detailed changes to the Results Framework throughout project implementation. Components Original Components 13. Component A. Resilient Infrastructure Rehabilitation (US$31.0 million). This Component had four subcomponents, which supported: - Subcomponent A.1. The rehabilitation and strengthening of dikes and weirs, namely the Nante and, Nicoadala dikes and the Eribacela weir to increase flood protection of irrigation infrastructure and local population. - Subcomponent A.2. The rehabilitation of rural infrastructure in Maganja da Costa district along the Licungo River, with a focus on irrigation schemes of Intabo (300 ha) and Munda-Munda (400 ha); 18 km of power line; and 15 km of the Niquidua-Mualei road. - Subcomponent A.3. The rehabilitation and restoration of the intake and distribution network of the Mocuba drinking water supply system. - Subcomponent A.4. The rehabilitation of 433 conventional classrooms and the construction of 1,038 mixed-material (non-conventional) classrooms in the provinces of Niassa, Nampula and Zambezia. 14. Component B. Technical Assistance for Resilient Recovery and Vulnerability Reduction (US$6.0 million). This Component supported the enhancement of institutional, sector and local capacities to manage risks associated with natural hazards. - Subcomponent B.1 provided technical assistance for resilient school construction, with a focus on: (a) identification of resilient construction techniques; (b) selection of sites and orientation of classrooms; and (c) quality control. - Subcomponent B.2 strengthened the capacity of the country’s DRM and Recovery Framework, with a focus on strengthening the capacity and means for the Government of Mozambique (GoM) and communities to manage and respond to disaster risks. - Subcomponent B.3 supported the study on Licungo Watershed Management, including the development of a flood risk model to reduce the vulnerability of dikes and other hydraulic works along the basin. 15. Component C. Project Implementation, Monitoring and Evaluation (US3.0 million). This Component strengthened the capacity for project management, coordination and M&E of the Project Implementation Units (PIUs) at the Ministry of Public Works, Housing and Water Resources (MOPHRH) – National Directorate for Water Resources (DNGRH), the Ministry of Education and Human Development (MINEDH), the National Institute for Irrigation (INIR) and the Water and Sanitation Infrastructure Administration (AIAS), including for: (a) fiduciary management; (b) environmental and social management; (c) preparation of project reports; and (d) monitoring and evaluation. 16. Component D (US$0 million). Contingency Emergency Response Component (CERC). This Component provided immediate response to an eligible crisis or emergency through the Immediate Response Page 9 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Mechanism (IRM), which allows for the use of a portion of uncommitted project funds to respond to emergencies. At appraisal, MOPHRH, through DNGRH, was designated the Coordinating Authority responsible for coordinating and implementing the IRM. Activation of the IRM was guided by the IRM Operations Manual approved by the World Bank in 2016. B. SIGNIFICANT CHANGES DURING IMPLEMENTATION 17. Response to drought in southern and central Mozambique triggered the activation of IRM and approval of Additional Financings. Between 2014-2016, the southern and central Mozambique were hit by a 30- year drought induced by El Nino. By April 2016, about 875,818 ha of crops were lost, affecting 464,879 farmers. In addition, 533,692 cattle and 55,193 goats were at risk - 5,708 cattle ultimately died, mostly in Maputo and Gaza provinces; 1.5 million people experienced food insecurity and were in need of urgent food assistance, and 191,656 children were at risk of being malnourished. The IRM was activated on November 2, 2016 in the amount of US$20 million to finance the most urgent needs for food assistance, nutrition, medicines, agricultural recovery and water supply for people and livestock protection in the affected regions. On January 13, 2017, the World Bank approved the first Additional Financing (AF) in the amount of US$20 million to scale-up drought-resilience activities introduced in 2016 under the IRM and focused on immediate food security needs, water supply and water management infrastructure. On October 30, 2018, the second AF was approved in the amount of US$13 million to replenish the resources allocated to the IRM in 2016. The financing gap of US$7 million to complete the US$20 million of IRM activation was recovered at restructuring in December 2018 through cancelation of some activities under the original credit and in the first AF (see Annex 6C). 18. The scope of the project was expanded from cyclones and floods to include drought, and spatially from central and northern Mozambique to the entire country. The first AF expanded the geographic scope of the project to the southern region of Mozambique to tackle the impacts of droughts and floods. Consequently, the project become nationwide, and its name was changed from Emergency Resilient Recovery Project for the Northern and Central Regions to Emergency Resilient Recovery Project (ERRP). 19. Response to further cyclones and floods, and COVID 19 impacts led to successive project restructurings. In March 2019, cyclone Idai hit central Mozambique as category 4 cyclone that affected 1.5 million people, killing more than 650 people, partially or totally destroying 4,219 classrooms, 92 health units, 240,000 houses, 3,940 km of roads and 20 bridges, and flooding 715,378 hectares of crops. In the Licungo river basin, cyclone Idai further damaged the dikes at Nante and Nicoadala, as well as the power line and the irrigation at Intabo already rehabilitated by the Project in December 2018. Cyclone Idai caused damages estimated at US$ 1.4 billion, losses of US$1.39 million and total recovery and reconstruction costs reached US$ 2.9 billion, around 22 percent of the GDP. In March 2020, Mozambique reported the first case of COVID 19, and in April 2022, the country declared a state of emergency and introduced curfew, restrictions for international flights, domestic travels and gatherings. These restrictions affected the progress of rehabilitation works of the Mocuba water supply system, the Macarretane dam and construction on no-conventional classrooms. 20. The project was further restructured in 2020 to: (a) address the needs for repair of damages caused by cyclone Idai in 2019 on rural infrastructure rehabilitated under the Project; (b) tackle the delays on progress of civil works caused by flooding of Licungo and Limpopo River basins early 2020; and (c) compensate for delays caused in delivery of works due to COVID 19 restrictions on import of construction materials. Significant changes were also made to targets of PDO indicators and intermediate indicators. 21. Table 1 presents the summary of changes to the Project at restructurings and Additional Financings. Page 10 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) While the restructurings of 2017 and 2018 increased the ambition of the project by adding new activities or increasing targets of some indicators, the restructuring of 2020 reduced the targets of some PDO indicators and intermediate indicators associated with works. Details of changes in targets are presented in Annex 6B. Table 1. Rationale for Changes and Their Implications for the Original Theory of Change Nov 2016 Jan 2017 Oct 2018 Dec June Nov May Main Changes (IRM activation) First AF Second 2018 2020 2021 2021 AF Change in Results No Yes Yes Yes Yes No No Framework Change in Safeguard Policies No Yes No No No No No Triggered Change in Loan Closing No No Yes No Yes Yes Yes Date(s) Cancellations Proposed No No No Yes No No No Reallocation between Yes No Yes Yes No No No Disbursement Categories Change in Disbursement No No Yes No No No No Arrangements Change in Disbursement No Yes Yes Yes Yes Yes Yes Estimates Change to Components and Yes Yes Yes Yes No No No Cost 22. Drought response activities introduced to the project under the IRM strengthened project’s impact and allowed further scale-up of drought response activities through the first AF. By expanding the project scope from floods and cyclones response in central and northern Mozambique, to droughts and floods in southern Mozambique, the project strengthened its pioneering role of establishing the foundation for multi-sector resilient emergency recovery from the three most recurrent hazards in Mozambique. In addition, the water supply activities financed under subcomponent A.7 of the first AF allowed for the scale-up of desalination water supply systems successfully piloted under the IRM. 23. Cancellation of DRM capacity building activities improved the alignment and impact of Bank operations in the DRM sector in Mozambique. Those activities were placed under the Mozambique Disaster Risk Management Program (P166437), to avoid duplication with the original project which remained focused on resilient rehabilitation or construction of infrastructure. II. OUTCOME A. RELEVANCE OF PDOs Assessment of Relevance of PDOs and Rating 24. Throughout the project life, climate and disaster resilience remained a central topic in both the Government agenda and the World Bank strategy. The project was well-aligned with the FY2017-2021 Country Partnership Framework for Mozambique, a key pillar of which was sustainability and resilience. Page 11 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) The rehabilitation works implemented by the project contributed to the achievement of CPF indicator 21: Share of damaged infrastructure restored or replaced to resilient standards. 25. The PDOs remained highly relevant to the country context of recurrent post-disaster recovery efforts and the need to build climate resilience at the sector level. The recurrent impacts of climate-related disasters during the life of the project, and the increasing need for post-disaster recovery in agriculture, education, irrigation, water supply, and water resources management infrastructure underscored the fundamental importance of building resilience in these sectors to secure development outcomes. Results generated by the project sub-components were gradually and informally adopted by new projects and several development partners to inform the design of pilot studies and projects on resilient housing and safer hospitals and to further scale-up resilience interventions particularly in the rural water supply and education sectors. 26. The project was consistent with the GoM’s disaster risk management policy and strategy. In 2017, the Government of Mozambique approved the new Master Plan for Disaster Risk Reduction (2017-2030), which calls for increased investment in disaster resilience through strengthening of prospective and corrective risk-reduction activities and continued mainstreaming of DRR in sector policies and plans. A new Disaster Management and Risk Reduction Law (Law no. 10/2020, of August 24) and complimentary regulations (Decree no. 76/2020, of September 1, and Presidential Decree no. 41/2020, of December 28) promoted the integration and alignment of disaster risk management and climate change adaptation at all levels while strengthening the mandate of the DRM coordinating agency. Reducing the vulnerability of the economy, infrastructure and communities to climate and disaster risks remained a key strategic objective of the current GoM’s Five-Year Development Program 2020-2024, approved by Parliament in 2020 (Resolution no. 15/2020 of April 14). 27. The project paved the way for new emergency recovery and resilience projects, and for the rapid growth of Mozambique’s resilience portfolio. The ERRP Project was the Bank’s first emergency recovery operation in Mozambique and was instrumental for strategic development of resilience capacity at national and sector level. This includes three new DRM operations: (a) the Mozambique Disaster Risk Management and Resilience Program (P166437), which includes resilient recovery of education infrastructure; (b) the Cyclone Idai and Kenneth Emergency Recovery and Resilience Project-CERRP (P171040), which finances resilient recovery of housing, public infrastructure, private sector and resilience to flooding in the city of Beira; and (c) the Northern Crisis Recovery Project (P176157), which supports emergency recovery of livelihoods, public infrastructure and services in areas affected by the conflict in northern Mozambique. 28. The project also built the foundation for activation of IRM/CERC for rapid emergency response in Mozambique. The Project paved the way for systematic inclusion of IRM /CERC in most new Investment Project Financing (IPF) operations in the Country Management Unit (CMU). This measure has been an enabling vehicle for rapid response to crisis by the CMU and in Mozambique. It included the activation of IRM/CERC under the Mozambique Agriculture and Natural Resources Landscape Management Project- SUSTENTA (P149620) and the Integrated Feeder Road Development Project (P158231) for emergency response to cyclones Idai and Kenneth, in 2019; and further under the CERRP (P171040), the Page 12 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Mozambique Urban Sanitation Project (P161777) and the Mozambique Land Administration Project- Terra Segura Project (P164551), for response to COVID-19, in 2020 1. 29. Relevance of the PDO is rated Substantial. The resilience recovery approaches and IRM activation introduced by the ERRP Project have been upscaled in new Bank operations, demonstrating the continued relevance of the PDO. B. ACHIEVEMENT OF PDOs (EFFICACY) 30. The project achieved or exceeded the targets of all 5 final PDO indicators (Table 2) and the targets of 18 out of 21 intermediate indicators. Three intermediate indicators – all in the education sector were between 86.3 and 92.3 percent 2 achieved. Table 2: PDO Indicators – Achievement of Targets Level of PDO Indicators Baseline End target Final value achievement Direct project beneficiaries 0 1,138,500 1,442,509 Exceeded3 Of which women (Percentage) 0 50% 50% Number of people in urban areas provided with access to Improved Water Sources under the project 7,760 31,860 33,695.00 Exceeded4 Number of children with access to improved education infrastructure 0 78,000 159,771.00 Exceeded5 Number of people protected by rehabilitated dike infrastructure 0 69,000 69,000.00 Achieved Number of people with access to improved irrigation infrastructure 0 5,600 5,600.00 Achieved6 Assessment of Achievement of Each Objective/Outcome Outcome 1: “Restore the functionality of critical infrastructure in a resilient manner in the disaster-affected provinces” 31. The project supported the resilient rehabilitation of damaged dikes and weirs. It included the resilient rehabilitation of a total of 30 kilometers of dikes of Nante dike (25 km) and Nicoadala (5 kilometers). The resilience elements included heightening of dikes to reach 7 meters corresponding to protection for a 20- year flood, installation of hydraulic structures, outlets, and controlled flood gates in vulnerable locations to reduce potential cut-off and allow better water flows, compaction, and slopes vegetation to prevent erosion. The dikes of Nante and Nicoadala were rehabilitated to provide protection a total of 1,000 1 The IRM was activated five times for a combined amount of close to US$200 million to provide financial resources within days of a disaster or other emergency event. 2 The three indicators of the educator sector were only partially achieved due to cancellation of the construction of 40 non- conventional classrooms in the province of Niassa because of low demand and the low capacity of local contractors. 3 Overachievement partially due to the contribution of interventions under the IRM, and the rehabilitation of Macarretane dam and the Niquidua-Mualei road. 4Overachievement driven by increasing number of domestic water connections in Mocuba 5 Overachievement driven by high number of children per classroom in Zambezia and Nampula provinces 6 This target was achieved with the contribution of the WB-funded IRRIGA project (P149620). As per decision at restructuring of 2020, the ERRP Project financed the initial works for rehabilitation of the Munda-Munda irrigation scheme (400 ha) until December 2020 in the total amount of US$0.63 million. The IRRIGA Project continued the execution of works until completion in December 2021 with co-financing in the amount of US$2.4 million out of the contract value of US$3.03 million. The works were completed just one month after closing of the ERRP Project on November 30, 2021. Page 13 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) hectares of irrigated land at Intabo, Munda-Munda and Mziva irrigation schemes, 30,000 hectares of cropland, and 69,000 people in the administrative post of Nante. The project also financed rehabilitation of the Eribacela weir to increase regulation and water storage in local natural lakes and provide water security to enable for drip irrigation for Munda-Munda irrigation scheme, particularly in the dry season. 32. The project supported the resilient rehabilitation of essential rural infrastructure in the Maganja da Costa District. Works included the rehabilitation of a total 700 hectares of the Munda-Munda (400 hectares) and Intabo (300 ha) irrigation schemes. Resilience works included: (a) the installation of a suction pipe to allow continued water pumping during low water levels, particularly in the dry season in the Licungo river; (b) installation of drainage canals for removal of excessive water during heavy rainfall events; (c) land leveling to provide protection against erosion of top-soils, and (d) lining of main irrigation canals. The project also supported the resilient rehabilitation of the Niquidua-Malei unpaved road (15 km), including the Niquidua-Malei bridge to recover road access to the Munda-Munda and Intabo irrigation schemes and electricity supply line (18 km) from Nante administrative post to Intabo and Munda-Munda irrigation schemes. Resilience works included: (a) the installation of additional 6 small bridges and lateral drainage to prevent road cut-off and erosion during heavy rain events; and (b) replacement and increased foundation depths of electricity poles, repair and elevation of the transformer of the pumping station of Intabo irrigation scheme to 1 meter above recorded flood levels. The rehabilitated irrigation schemes and power line benefit 5,600 farmers while 52,050 people benefit from the rehabilitation of Niquidua-Mualei road 33. The project financed the rehabilitation of the drinking water supply in Mocuba. Activities financed by the project included the rehabilitation works of the water transmission main and associated infrastructure, such as access road, pumping station, reservoir, water tower, treatment plant; the execution of 2,228 domestic connections and installation of 7 public standpipes. Resilience works included the relocation and construction of new intake to prevent meandering of Lugela river; installation of groynes to prevent scouring of the intake during flood events and deep replacement of 27.5 kilometers of the distribution network to prevent erosion. The rehabilitation works have allowed the water supply system to produce 5,760 m3/day and provide access to clean water for 24,300 people, and 50,000 people will be served by 2024. In addition, the project financed the feasibility study, Environmental and Social Impact Assessment (ESIA) and a detailed design for Phase II of the Mocuba water supply rehabilitation to lay the ground for sustainable long-term upgrading and expansion of the rehabilitated system to provide water access to 170,000 people by 2040. 34. The project supported the rehabilitation and reconstruction of resilient schools. The project was initially designed to support MINEDH to pilot new resilient construction techniques and standards developed between 2012-2015 by the UN-Habitat for resilient rehabilitation of 433 damaged conventional classrooms and the construction of new 1,038 non-conventional in the provinces of Niassa, Nampula and Zambezia. The classrooms were designed to withstand the locally mapped cyclone winds and ground shaking, and to include rainwater harvesting systems. Due to significant increases in the construction costs, partly reflecting the limited information availability on prices at project design and the significant time gap between project design and contracting of works, the final project targets were reduced to rehabilitation of 372 conventional classrooms and construction of 257 non-conventional classrooms 7, 118 administrative blocks, 150 rainwater harvesting systems and 408 toilets. These resilient classrooms are 7This target was partially achieved (84 percent), as 217 non-conventional classrooms were constructed by the project in the provinces of Nampula (96) and Zambezia (121). Page 14 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) benefiting 159,772 students in the provinces of Niassa, Nampula and Zambezia. 35. The project supported the rehabilitation and expansion of the water supply systems of Funhalouro and Chicualacuala. The scope of work included: (a) preparation of a feasibility study, assessment of willingness and ability to pay, detailed design, bidding documents and supervision of works; and (b) civil works, including the construction of reservoirs, water towers, treatment plants, 53 km of distribution network, execution of 1,302 domestic connections, and installation of 22 public standpipes. Resilience aspects included the replacement of the former low productive traditional shallow boreholes equipped with manual pumps, by 19 Cartesian boreholes with depths between 50 to 120 meters to ensure access to productive aquifers during drought events and production capacity between 35 m3/hour to 40 m3/hour, and installation of electric pumping stations. These measures will ensure stable water productivity even under drought conditions. The upgraded systems provide access to safer drinking water for 14,410 people. 36. The project financed the rehabilitation of the Macarretane Barrage. The dam structure was at risk of inclination and the safe operation of the dam was undermined due to scouring of the dissipation terrace. The scope of works included (a) replacement of the damaged concrete block carpet in the downstream terrace by a new dissipation system and a reno mattress across the 650 meters of dam length; and (b) installation of a protection wall in the right bank and new road access protection sheet pile. These works have increased the lifetime of the dam by 70-100 years. Resilience elements included: (a) the construction of a safer parking bay for the gantry crane aimed at preventing frequent traffic accidents with the crane and increasing safety and adequate manual operation of the dam in case of power outages during flood peak periods; and (b) construction of a new office building in a safer high-ground location 2.5 km away from the floodplain, and relocation of the Limpopo Basin Management Division (DGBL), in order to prevent loss of key information and protect staff to ensure continued safe operation of Macarretane dam during the occurrence of catastrophic floods. The rehabilitation works have improved flood management directly benefiting the populations of six districts and increased water security during drought events for agricultural production in the irrigation schemes downstream the dam including Chokwe (33,000 hectares) and Baixo Limpopo (11,000 hectares) benefiting 21,000 farmers (see Annex 6E). 37. The project was designed to support the construction of water desalination systems and multi-purpose boreholes in the districts of Chigubo and Massingir in Gaza Province in 2017. This intervention scaled up the desalinations systems piloted under the IRM activation. The initial design consisted of supply and installation of 2 desalination systems and the construction of 8 multifunctional boreholes in Chigubo and Massingir districts. The sites and targets were revised in 2018 for construction of: (a) four desalinations systems in the districts of Chókwe (1), Massingir (1) and Massagena (2), each with capacity to produce 24m3/day of treated drinking water; and (b) four multi-purposes boreholes in the districts of Mapai (2) and Chibuto (2), each with the capacity to produce 3m3/per hour of drinking water; (c) construction of a distribution network to serve up to 2,500 people for each water source; (d) installation of 32 public standpipes, 6 water dispensers for livestock, and a digital Management System for Operation and Maintenance for the desalination systems. Resilience elements included: (a) the replacement of the former low productive brackish water traditional shallow boreholes equipped with manual pumps by deeper multipurpose boreholes allowing access to productive aquifers during drought events; (b) installation of solar pumps in all new boreholes; and (c) construction of water towers with tanks, each with a capacity of 10,000 liters to provide enough time for recharge of aquifers and to ensure secure water distribution by gravity. These systems benefit up to 16,000 people. 38. The project supported technical assistance (TA) for resilient recovery and vulnerability reduction. TA focused on strengthening DRM and enhancing local capacity to manage risks associated with natural Page 15 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) hazards. Under subcomponent B1, TA was provided to the education sector by UN-Habitat for: (a) the identification of the best resilient construction techniques and use of local materials; (b) the selection of appropriate design and orientation of classrooms regarding to hazard zoning; (c) provision of on-the job training for contractors and communities; and (d) quality control of construction works. Overall, 41 local communities, 21 contractors and engineering firms, 3 NGOs were trained by UN-Habitat. These trainings benefitted a total of 1,744 people at national, provincial and community level. Under Sub-component B2, technical assistance was provided for INGD for the preparation of the medium and longer-term Investment Plan for the Master Plan for Disaster Risk Reduction (2017-2030). 39. All infrastructure rehabilitated or constructed under the project performed well during the lifetime of the project, demonstrating its functionality and resilience. Except for the Munda-Munda irrigation scheme for which the works were completed in December 2021 under the IRRIGA Project, all infrastructure rehabilitated or constructed under the project was completed, handed over and used by the beneficiaries with proven functionality and resilience as per design. This was the case conventional classrooms which withstood with no damages to all storms since 2019, the Munda-Munda irrigation and Mocuba water supply which suffered minor erosion of topsoil, and limited damages to the access road of intake at the passage of tropical storm Ana and Tropical Cyclone Gombe, respectively. The floods damaged the dikes of Nante and Nicodala signaling the need for further heightening of these dikes to reach between 9 to 10 meters as to provide protection close to a 30-year flood event 8. 40. Institutional strengthening: The project contributed to strengthening institutional capacity to integrate climate resilience at sector and community levels. It also helped to institutionalize safeguards management systems in beneficiary institutions. TA provided by UN-Habitat created the capacity within MINEDH, the construction industry and 41 local communities for resilient rehabilitation and construction of schools; the project strengthened the capacity and expertise of AIAS and the DNAAS to design, implement and manage contracts for construction, rehabilitation or expansion of resilient water supply systems and desalination plants in complex environment subject to flooding and droughts. The project also strengthened the capacity of the Hydraulic Works Department at DNGRH and its regional bodies to design, implement and supervise the rehabilitation and maintenance of dike systems, weir and dams. INIR’s capacity has been increased to adequately upscale the resilient construction and rehabilitation of irrigation schemes across the country. Finally, the project strengthened the capacity and systems of MINEDH, AIAS, INIR and DNGRH for adequate safeguards management and compliance, including preparation and implementation of project and activity-related safeguards instruments, timely reporting on grievance redress mechanism, and handling of gender-based violence/sexual exploitation and abuse (GBV/SEA) cases. Outcome 2: “Improve the Government of Mozambique’s capacity to respond promptly and effectively to an eligible crisis or emergency” 41. The project supported the prompt response to the drought crisis in southern and central Mozambique through the activation of the IRM in 2016. The IRM supported interventions in the agriculture, water and 8 The 2015 flood reached 11.9 meters at Mocuba gauge station (with local rainfall of 102.2 mm in 24 hours on January 12) and was considered a 30-year flood event. In March 2019, the flood induced the first landfall of cyclone Idai reached 7.0 meters (with local rainfall of 130.5 mm in 24 hours on March 3). More recently, the flood induced by Tropical Storm Ana in January 2022 reached 9.5 meters and caused damages (20 percent) to the Nante dike, while the Nicoadala dike remained intact. Further, the flood induced by Tropical Cyclone Gombe in March 2022, reached 8.4 meters (with local rainfall of 148.8 mm in 24 hours on March 14) and caused further damages to the Nante dike. Page 16 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) health sectors, as well as food assistance. In the agriculture sector, the IRM supported the provision of 4,200 drip irrigation kits to smallholder farmers and 30 irrigation draglines equipped with solar panels, and 762.1 tons of certified seeds benefiting 246,630 farmers. In the health sector, the IRM supported the acquisition of medicines and medical equipment as well as 1,781.05 tons of nutritional supplements that were used for treatment of 244,394 malnourished people over six months across the country. In the water supply sector, the IRM supported the construction of two desalination systems in the districts of Chigubo and Chicualacuala, which benefited 1,800 people as well as livestock. Finally, the IRM supported the acquisition and distribution of 5,693.16 tons of maize and beans to 519,138 people. Overall, the IRM provided support to 1,016,192 people across the country. 42. The support provided under the IRM helped the country to restore the food production capacity of beneficiaries and the response capacity of the health sector. Farmers provided with seeds for the agricultural campaigns of 2016-2017 and 2017-2018 were able to restore their food production capacity, and as of April 2018 no longer relied on Government food aid for food security. In addition, the distribution of agriculture irrigation kits further boosted the production and coping capacity of beneficiaries, including during localized dry spell events that affected the country during the life of the project. The treatment rates of malnourished cases by the national health system increased from 74 percent in 2016 to 84 percent in 2018, as result of replenishment of stocks of nutritional supplements across the country. 43. The activation of IRM/CERC under the original project built the country’s capacity and readiness for emergency response and recovery. The original project was the first operation with IRM/CERC component and the first to activate the IRM/CERC in the Mozambique portfolio. The use of a single PIU with proven capacity on fiduciary management, with technical support from the beneficiary institutions contributed for quick contracting and delivery of emergency support and streamlined reporting and accountability. Additionally, the prior selection IRM/CERC coordinating entity within Government at appraisal, the early updating of IRM/CERC Operations Manual each year, and the selection of the DRM team as leader of IRM/CERC activation within the World Bank has provided the country and the WB with the necessary institutional readiness that helped the country to quickly and timely respond and recover from emergencies and crises that affected the country between 2016 and 2020 and to become an example of effective activation and implementation of the IRM/CERC in Africa. 44. Overall, IRM/CERC implementation, performance and impact was Satisfactory as the IRM was activated swiftly, implemented within the expected timeframe, and delivered all intended results. IRM implementation was considered satisfactory as there was high level of ownership and immediate engagement of all agencies involved in the implementation of IRM activities. The performance was satisfactory as all resources (US$20 million) allocated to the IRM were fully disbursed and all contracts signed were fully executed within 12 to 18 months, except the distribution of irrigation kits which was completed with delays. The impact of IRM activation was considered high as satisfaction of direct beneficiaries reached 85 percent, and generated positive impacts in all sectors supported by the IRM activities, including development of institutional capacity (expertise, systems and procedures) within Government and World Bank for rapid emergency response through IRM /CERC, stabilization of food security and restoration of household food production capacity in the drought-affected areas, and restoration of capacity of the national health system for treatment of patients, including for malnutrition. Justification of Overall Efficacy Rating 45. Assessment of achievement of the PDO: Substantial. The project was successful in achieving its PDO. Rehabilitated and constructed infrastructure demonstrated functionality and increased resilience. Page 17 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Moreover, the Government increased its capacity to promptly respond to any crisis or emergency across the country. C. EFFICIENCY 46. An economic analysis conducted at project closing shows that the project had a high economic return. The cost-benefit analysis covered the rehabilitation of dikes, rural infrastructure, water supply systems, classrooms, and the Macarretane barrage, which together represent about 65 percent of total project costs. The benefits of these activities are compared to the total costs of the project to establish cost efficiency. See Annex 4 for details on the economic and financial analysis. 47. The economic analysis is based on the benefits of flood protection and irrigation, time savings associated with improved water access and the benefit of access to education for future earnings. Adding the benefits together, the project had a net present value (NPV) of US$98 million at a 12 percent discount rate and an internal rate of return (IRR) of 32 percent, when comparing measured benefits to total project costs and considering standard shadow pricing assumptions.9 Due to emergency nature of the original project and limited information available, the full economic analysis, including an NPV was not presented at appraisal, so comparison to an ex-ante cost benefit analysis is not possible. The estimates presented at project closing are highly conservative, as they do not include the benefits from the drought emergency support, which helped prevent a humanitarian crisis particularly in southern Mozambique. 48. The use of competitive bidding, flexible financial arrangements and co-financing had a positive impact on efficiency. Except for recovery of classrooms, all rehabilitation and construction works were contracted within or below the costs initially estimated during project design, therefore bringing initial savings to the project. These savings were later used to accommodate the repairs of damages caused by cyclone Idai on rural infrastructure, the construction of a new office for the Limpopo River Basin Management Division (DGBL), and the upgrading of two multipurpose boreholes into desalination plants. The project also financed additional works for construction of a parking bay for the gantry crane in the Macarretane dam through reallocation of part of resources assigned for the Munda-Munda irrigation scheme. The rehabilitation of Munda-Munda was covered by the IRRIGA project which financed 80 percent of civil works contract in the total amount of US$3.03 million and the full cost (US$290,377) of supervision. 49. New cyclones, floods, COVID 19 and delays in replenishment of IRM resources and in effectiveness of the AF increased costs for some works and negatively affected efficiency. This was the case of rehabilitation works of the Munda-Munda irrigation scheme, the construction of conventional classrooms, and the preparation of detailed design and bidding documents and the ESIA for Phase II of the Mocuba water supply system. These activities were resumed two years later after replenishment of IRM resources in 2018 through the second AF. Delivery of contracts of works was further delayed due to interruption of works caused by the occurrence of cyclone Idai in March 2019, the increase in water levels in the Limpopo River basin in early 2020, the COVID-19 pandemic in 2020 caused delays in the import of construction materials for the rehabilitation of the Mocuba water system and the desalination plants in Gaza Province. To accommodate these delays, the contracts for supervision of works were extended between 3 to 5 months. 50. Efficiency is rated as Substantial. Despite the project’s estimated high economic return, delays at various 9 Local labor is assumed at half of its price, international labor is considered at full price and a 20 percent import tax is deducted from all goods. Page 18 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) times during implementation negatively affected its overall efficiency. For that reason, the overall efficiency rating is Substantial. D. JUSTIFICATION OF OVERALL OUTCOME RATING 51. This ICR uses a split rating for the determination of the overall outcome rating. Based on the explanation provided in paragraph 21, the restructuring of 2020 was used as reference for the split rating as it introduced significant reduction of targets of some PDO indicators and intermediate indicators, particularly in the education sector. The results of split rating are presented in Annex 6D. 52. The PDO related to resilient recovery of infrastructure rated Substantial for Efficacy for most time of project implementation and reached High over the last year when works were gradually completed. The PDO related to government’s capacity to respond to eligible crisis rated High for Efficacy throughout project life confirming early country’s readiness, the timeliness of IRM activation, swift implementation of emergency response and recovery activities, and achievement of all expected results. 53. Based on the rates assigned to Relevance of PDO (Substantial); Efficacy (Substantial) and Efficiency (Substantial), the Overall Outcome is Moderately Satisfactory. E. OTHER OUTCOMES AND IMPACTS 54. Gender: Investments under this project brought significant social and economic benefits to women and girls in rural areas. For instance, women served by rehabilitated irrigation schemes of Munda-Munda and Intabo or benefiting from improved water security provided by the Macarretane dam are gradually increasing agricultural productivity, 10 and are better placed to transition from smallholder to market- oriented farming, with support from the SUSTENTA project (P149620). Local education sector staff believes that the recovery of classrooms, particularly in remote areas will gradually contribute to the increase in self-esteem of students due to access to better school facilities, and is likely to help to reduce the incidence of dropouts in primary education 11, increase the retention rates of girls, and reduce their risk of early marriage. 12 Local women were hired by NGOs and contractors to provide water and food for the workers engaged in the construction of resilient classrooms, while the construction of desalination plants and water supply systems in rural areas of Gaza province reduced travel time for women and girls (see Annex 6F). 55. Informing the development of resilient infrastructure. The project had a transformational impact on the understanding of resilience, resulting in increased demand for climate and disaster resilient development at the political and sector levels. The project inspired the implementation of three resilience dialogues: one dialogue on financial resilience (in 2017) and two dialogues on social protection (in 2019 and 2021). At the Ministry of Public Works, Housing and Water Resources (MOPHRH), the project increased awareness on the need for revision of the Building Code, officially approved in 1969. In the education 10 With support provided by a service provider under IRRIGA Project (P164431) the agricultural yields from the irrigation schemes of Munda-Munda, Intabo and Mziva were expected to increase from previous 1 ton/ha for rice production, to reach 3 tons/ha in the dry season, and 4 tons/ha during the rainy season. 11 During a joint World Bank, MINEDH and UN-Habitat site visit in November 2021, the director of the Primary School of Abado under construction in Nicoadala district in Zambezia province indicated that per year on average, between 15 to 30 children abandoned that school or moved to other better schools, including to those schools that provide meals for school children. 12 Mozambique has the 10th highest rate of early marriage in the World, with almost half of adolescent girls ages 15–19 reporting that they are married. As of 2015, about 46 percent of this population group were already mothers or pregnant, and this percentage has increased over the last twenty years (Demographic Health Survey- DHS, 2015). Page 19 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) sector, the new resilient construction techniques successfully tested by the project (see Annex 6G) were widely adopted by development partners for post-disaster recovery following cyclone Dineo in 2017, and cyclone Idai in 2019, well before the official adoption of these standards in 2021. 56. Further, the positive results of scaling-up of desalination plants and piloting of multi-purposes boreholes using solar panels under the project has attracted development partners in the water supply sector to approve a £10 million investment plan for construction of 20 desalination systems and 23 multi-purpose boreholes in the provinces of Maputo, Gaza, Inhambane, Manica, Tete, Zambezia and Nampula. 57. Poverty reduction and shared prosperity through access to basic services and economic opportunities. The rehabilitation or construction of new classrooms equipped with access facilities for people with disability and new or renewed toilets tailored for boys and girls is expected to increase enrollment and attendance rates, including for children with special needs, and contribute to better education outcomes. Increased access to water provided by the project is expected to increase quality of life of beneficiaries, including reducing the distance and waiting times involved in fetching water, and reduction of incidence of waterborne diseases due to contamination of water tables by animal feces. The support provided under the IRM, particularly food aid and distribution of seeds contributed to the stabilization of food security and restored the food production capacity of beneficiary households in drought-affected areas, leading to the Government’s cancellation of US$3 million allocated for food assistance under the first AF. The project also generated 3,161 jobs for people recruited from local communities to construct resilient schools in their communities. III. KEY FACTORS THAT AFFECTED IMPLEMENTATION AND OUTCOME A. KEY FACTORS DURING PREPARATION Assessment of Project Design 58. The project was designed in less than three months and addressed a combination of short, medium and longer-term reconstruction needs. The initial project design focused on recovery and reduction of vulnerability in the high flood risk zones in the Licungo River Basin in Zambezia province and cyclone risk areas in the northern provinces of Niassa and Nampula. The rehabilitation of dikes and related infrastructure was considered an urgent need to provide minimum flood protection, particularly for the Munda-Munda, Intabo and Muziva irrigation schemes. The project was equipped with a TA component and analytical work required to build capacity for medium and longer-term resilience for schools, early warning systems, a recovery framework and watershed management. 59. To accelerate implementation, project activities were selected based on the experience of well- performing projects and housed withing existing well-functioning units. These were: (a) the Water Resources Development Project and Transforming Hydrological and Meteorological Services Project (at DNGRH); (b) the Sustainable Irrigation Development Project - PROIRRI (at INIR); (c) Cities and Climate Change Project- CCCP (at AIAS); and (d) the Education Sector Support Program - ESSP (at MINEDH). The existing PIU and sector teams were strengthened with the recruitment of additional technical, safeguards, financial management and procurement consultants, as appropriate. These arrangements were adjusted at INIR after the closing of the PROIRRI project in 2018. 60. A Project Steering Committee (PSC) was established under the leadership of the Ministry of Economy and Finance (MEF), with INGD acting as the PSC secretariat. The PSC was composed of directors of the four implementing agencies (DNGRH, INIR, AIAS and MINEDH) and chaired by a national director Page 20 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) appointed by MEF (as permanent members), with permanent invitees from INGD and DNAAS. The PSC was responsible for ensuring overall coordination and monitoring of recovery efforts, and to provide implementation oversight, including oversight of the consolidation of implementation progress reports from DNGRH, INIR, AIAS and MINEDH. The PSC assigned a Project Coordinator to manage coordination and the consolidation of progress reports. In 2018, DNGRH hired an Assistant Project Coordinator and took over the coordinating role from INGD. Assessment of Project Risk 61. The PAD discussed the risks and proposed adequate mitigation measures. The Systematic Operations Risk Rating Tool (SORT) was applied during project preparation and the overall project risk was rated Moderate. This risk rating remained unchanged throughout project implementation. The project was classified as Category B under the World Bank Safeguards Policies and triggered the Operational Policies (OPs) on Environmental Assessment (OP/BP4.01), Natural Habitats (OP/BP4.04), Pest Management (OP 4.09), Physical Cultural Resources (OP/BP 4.11) and Involuntary Resettlement (OP/BP4.12). The first AF further triggered the OPs on Safety of Dams (OP/BP 4.37) and Projects on International Waterways (OP/BP 7.50). During implementation, the social and environmental risks increased from Moderate to Substantial to reflect the safeguards compliance challenges associated with the rehabilitation of dikes and rural infrastructure, and the expansion of the project scope introduced by the first AF. 62. In conclusion, project design was and remained appropriate during implementation for achievement of project objectives. B. KEY FACTORS DURING IMPLEMENTATION Factors subject to client control 63. Government ensured the stability and smooth transition of project staff. Key PIU and sector staff assigned to the project remained unchanged at AIAS, DNGRH and MINEDH. INIR ensured a smooth transition of coordination of the ERRP project from the outgoing PIU of PROIRRI to the incoming PIU of IRRIGA project, while MINEDH ensured the adequate transition of safeguards management from the ERRP project to MINEDH staff. This relative stability positively impacted project implementation and management during changes of leadership at MINEDH and INIR. 64. The recruitment of an Assistant Project Coordinator strengthened oversight by the PSC. The MTR recommended the immediate contracting of an Assistant Project Coordinator to support the Project Coordinator assigned by DNGRH who was overwhelmed with the implementation of other two WB- financed operations. The support provided by the Assistant Project Coordinator improved the quality of project coordination and reporting by all sector teams. Overall, the PSC conducted 10 routine meetings at the end of WB supervision missions. 65. Limited procurement capacity and poor contract management, and delays in meeting safeguards requirements, caused delays in the implementation of works and the provision of irrigation kits. Delays in the procurement of conventional and non-conventional classrooms were associated with limited procurement capacity at MINEDH, aggravated by cumbersome administrative procedures for clearance and signature of contracts. The establishment of an internal procurement unit dedicated to the ERRP Project in 2018, and the engagement of MINEDH leadership improved procurement performance of works. At INIR, limited capacity for contract management contributed to the suspension of initial contracts and delayed the contracting of rehabilitation works at Munda-Munda and the delivery of irrigation kits. Page 21 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) At DNGRH, delays in meeting specific project safeguards requirements, caused significant delays in the commencement of rehabilitation works at lot 1 of Nante dike (19 km). Factors subject to World Bank control 66. The World Bank assigned experienced field-based staff to implement the project, including Task Team Leaders (TTLs). The project was implemented by two consecutive senior Administrative Decision Making (ADM) TTLs, two Co-TTLs and two Safeguards specialists (one Environmental and one Social). During the initial year of project implementation (2015-2016), one ADM TTL was based in field, supported by a co- TTL based in Washington. At closing, the project was led by an ADM TTL and a co-TTL, both based in the Country Office. The stability of the task team helped drive project implementation and maintain cohesion, unity, and collaboration among all PIUs. 67. The Bank provided adequate guidance and response to issues arising during implementation. Regular check-in meetings, Implementation Support Missions (ISM) and the MTR highlighted the relevance and timeliness of World Bank support in addressing: (a) implementation and documentation issues associated with the IRM; (b) the urgency of replenishing project resources mobilized for IRM activation, to ensure continued alignment of project interventions with the original design and objectives; (c) the Government’s limited capacity to comply with safeguards due to the increasing number of activities located in geographically distant sites; (d) the time constraints faced by the Project Coordinator at DNGRH in providing effective support to the PSC for project oversight, coordination and reporting; (e) the limited procurement capacity at MINEDH for the timely rehabilitation and construction of resilient classrooms. The team proactively engaged with the client and the CMU to timely use the Project to tackle the different weather-related emergencies and COVID 19 crisis. Factors outside the control of the Government and World Bank 68. Higher market prices, the frequency of weather-related hazards and the COVID-19 pandemic were the main factors that negatively affected implementation. The average cost of rehabilitating a conventional classroom increased from US$8,800 per classroom for the 217 classrooms under Phase I (between 2017- 2019), to US$11,000 for the 155 classrooms under Phase II (between 2019-2021). The average cost of construction of a non-conventional classroom reached US$20,400, against the initial estimate of US$6,000. Consequently, the number of classrooms was reduced from the initial 433 to 372 conventional classrooms, and from 1,038 to 217 non-conventional classrooms. Increase in prices from US$6,000 to US$8,500 per hectare were observed for the rehabilitation of the Munda-Munda irrigation scheme. The resulting financing gap was covered by the IRRIGA project, which took over the rehabilitation of the Munda-Munda scheme from April 2020 until its completion in December 2021. Delays in completion of civil works due to impacts of cyclone Idai, further flooding and COVID 19 restrictions resulted in extension of contracts of supervision firms for three to five months. IV. BANK PERFORMANCE, COMPLIANCE ISSUES, AND RISK TO DEVELOPMENT OUTCOME A. QUALITY OF MONITORING AND EVALUATION (M&E) M&E Design 69. The Project made use of monitoring and evaluation arrangements already in place within the existing PIU’s at DNGRH, INIR, AIAS, and MINEDH for specific project sub-components. These PIU’s were Page 22 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) expected to carry out community-level surveys on a periodic basis to compile and update the baseline of target indicators values and report the results to the Project Steering Committee and the Bank on a routine basis through bi-annual supervision missions and reports, covering Social and Environmental monitoring; regular quality supervision and certification; periodic physical progress monitoring; and results monitoring and evaluation. The original project included 7 PDO indicators and 12 intermediate indicators and targeted 1,000,000 beneficiaries. 70. Both the initial M&E framework and the final framework were robust and maintained a clear connection between project activities and the PDO. For DNGRH and AIAS, initial targets were defined based on estimates of projected population in target areas, while for INIR and MINEDH baselines were based on existing datasets of users of irrigations schemes and school attendance. AIAS and DNGRH conducted additional surveys to update the targets for water supply systems and desalination systems, while MINEDH used the attendance rates per school to update the target of beneficiaries of each rehabilitation or constructed infrastructure. The impact of the project at closing was strengthened with the inclusion of the final number of beneficiaries of nutrition supplies under the IRM as well as the beneficiaries of rehabilitation of the Niquidua-Maulei road and the Macarretane dam in final targets of PDO indicators. The Project could have further benefited with better integration of sex-disaggregated indicators in access to protected irrigation land, education infrastructure and water supply. 71. The M&E framework was fully integrated in the Government’s annual planning processes and in the dialogue between Government and developments partners, particularly those supporting the existing pools of donors funding in the education and water supply sectors. M&E Implementation 72. PDO indicators and intermediate indicators were adjusted as required during project implementation to reflect the impacts of several crisis. The M&E framework was adjusted at project restructurings to capture the scope, ambition, and impact of the project on the ground. The PDO indicator (Number of people supported by Early Warning and Response System) and two intermediate indicators (Communities/villages covered by a new or updated emergency response plan linked to improved early warning capacity, and Licungo watershed management study completed) associated with activities cancelled at restructuring in 2018 were dropped, new intermediate indicators were added to capture the activities introduced by IRM, first AF and new activities at restructuring of 2018 (construction of toilets, administrative blocks, and water harvesting systems) in the education sector. With these adjustments, the PDO indicators reduced to 6 and the intermediate indicators increased to 26 (see annex 6B). In 2018 and 2020, targets were increased as needed to capture the implications of additional works after the first AF or decreased to reflect the impacts of increased costs in the education sector. Overall, the Project reached 1,442,509 beneficiaries at closing. 73. All PIUs were responsible for tracking and reporting on progress in the activities under their responsibility. All PIU’s prepared regular separate semi-annual and annual reports, which were further consolidated in respective single reports and submitted to the World Bank by DNGRH. These reports included a dashboard with a list of all signed contracts and status of physical and financial execution and the RF tracking progress of indicators targets. Individually, all PIU’s also prepared and submitted to the World Bank all required quarterly reports of E&S management. DNGRH also conducted two surveys (one for IRM evaluation and one for the Government ICR targeting 650 and 168 beneficiaries and key informants, respectively) to assess the impacts of the project. Overall, M&E reporting to the Government to Project Steering Committee and the World Bank improved significantly and remained consistent after Page 23 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) the MTR because of cross-PIU’s coordination support provided by the Assistant Project Coordinator. M&E Utilization 74. The M&E framework was used by the Bank team and the implementing agencies as a management and decision-making tool, particularly during the MTR and the project restructuring in 2018 and 2020. In particular, progress towards achievement of targets of the original project was used for request of replenishment of resources mobilized for IRM. It was also used for request of extension of project closing date to ensure completion of all delayed activities conducive to achievement of targets. 75. The MTR confirmed relevance of PDO related to resilient recovery of functionality of critical infrastructure and rated it as substantial due to limited progress on civil works, therefore affecting progress towards achievement of PDO and intermediate results. This assertion informed the timely adjustment of project M&E framework at project restructuring in 2018 and 2020. Justification of Overall Rating of Quality of M&E 76. The Overall Rating of the Quality of M&E is Substantial. B. ENVIRONMENTAL, SOCIAL, AND FIDUCIARY COMPLIANCE 77. The project initially faced challenges with safeguards compliance and reporting, but gradually met the requirements of Bank’s safeguard policies and kept a Moderately Satisfactory rating until closing. The implementation of the Safeguards Policies included preparation of an Environmental and Social Management Framework (ESMF), a Resettlement Policy Framework (RPF) and an Integrated Pest Management Plan (IPMP). All construction activity was preceded by the preparation of a site-specific Environmental and Social management instruments, including Resettlement Action Plan (RAP) for resettlement of 4 Project-affected people (PAPs) and compensation of 16 others, affected by the rehabilitation of Mocuba water supply system. Regular reporting was ensured, including the reporting on functioning of the Grievance Redress Mechanism (GRM), Occupational Health and Safety (OHS), Gender Based Violence (GBV) and Sexual Exploitation and Abuse (SEA) cases (see Annex 6H). 78. Financial management. The project management was able to meet the financial management (FM) requirements for audits and financial reporting. The main challenge was the timely documentation of expenses for the purchase of medicines and nutritional supplements by the Ministry of Health (MISAU) under the IRM. There were also significant delays at MINEDH in the payment of invoices submitted by contractors and in the documentation of expenses by UN-Habitat. 79. Procurement. As noted earlier in paragraph 65, weaknesses in existing procurement department at MINEDH caused delays in contracting and delivery of works for the education sector. Further, cumbersome internal administrative procedures introduced in early 2020 by the new leadership at MINEDH for approval of contracts, as well as poor contract management by the PIU, further delayed the completion of classrooms in Zambezia province. The completion of works was only possible with extension of project closing date for 11 months, and close monitoring by the World Bank task team. 80. Legal covenants. The project was initially designed with five legal covenants. All covenants were complied with after some delays due to delays in project effectiveness. The MTR was completed six months after the expected date due to implementation delays caused by the IRM activation and the inclusion of new activities added at the time of AF. Page 24 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) C. BANK PERFORMANCE Quality at Entry 81. The project was prepared in less than three months from Activity Initiation Summary (AIS) to approval by the World Bank Board on September 29, 2015. The project was processed under paragraph 12 of OP10.00 – Projects in Situation of Urgent Need of Assistance or Capacity Constraints, and preparation of safeguards instruments was deferred until project implementation. Priority was given to rehabilitation of dikes and irrigation schemes to ensuring the rapid resumption of agricultural productivity, leaving other project activities for medium and long-term implementation. The design of the first AF for drought mitigation was also based on an assessment of needs and was consistent with the Government’s priorities. 82. The PDOs defined at the design stage adequately addressed the entire scope of activities and objectives of the project. The implementation arrangements based on existing PIU and sectors teams were adequate to address the complexity of the project and to ensure a rapid start of project activities, including the IRM activation, if required. These arrangements were also adequate for the management of the PIUs’ initially identified capacity risks, including those related to safeguards and coordination of stakeholders. Quality of Supervision 83. Implementation support was timely, proactive, and focused on results. The World Bank team used Implementation Status and Results Reports (ISRs), Aides-Memoire (AM) and the MTR as tools to monitor and accelerate implementation, identify key challenges and assign performance ratings. Back-to-office reports and management letters were used to raise key implementation issues on the reconstruction of non-conventional classrooms by MINEDH. Supervision support missions were carried out in the form of field visits to construction work sites and participation in the meetings of the Project Steering Committee. Monthly check-in meetings were held with all PIUs during the final 18 months of project implementation. The World Bank conducted 13 Implementation Support Missions (ISMs) over the six years and five months of project implementation. 84. The project was restructured five times during implementation. The Bank was proactive and responsive in introducing adjustments to the project design to address changing circumstances and implementation challenges on the ground, including safeguard compliance requirements for commencement of civil works. Justification of Overall Rating of Bank Performance 85. The World Bank team successfully addressed all challenges faced by the project during preparation and implementation. The team adequately managed the complex multisector design, increased scope and nationwide expansion of project. Overall World Bank performance is rated Satisfactory. D. RISK TO DEVELOPMENT OUTCOME 86. Maintenance of sector capacities strengthened by the project is critical to mitigate risks to the project outcomes. The rehabilitation of dikes, weirs, dam, and water supply infrastructure was led by the existing specialized Government staff at AIAS, DNGRH and DNAAS, while existing PIUs provided fiduciary support. This approach enabled the three institutions to acquire increased knowledge and skills that will guide the design and implementation of similar operations in the future. At INIR, the experience of rehabilitation and operation of irrigation schemes under the project was transferred from the PIU of PROIRRI Project to Page 25 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) the IRRIGA project, which is now responsible for the management of the irrigation schemes of Intabo, Munda-Munda and Mziva. At MINEDH, the experience gained by the PIU staff and the knowledge generated by the TA of UN-Habitat on rehabilitation of classrooms is being transferred to MINEDH staff for the rehabilitation of 3,000 classrooms under the Mozambique DRM and Resilience Program. Having these teams in place will be critical to ensure sustainability of project outcomes on these sectors. 87. Continued investment is required for better operation and maintenance of rehabilitated or constructed infrastructure. The project established the necessary legal and institutional instruments to ensure adequate operation and maintenance of infrastructure provided by the project in all sectors. It includes the transfer of operation of water supply system from AIAS and DNAAS to the Water Supply Assets Investment Fund (FIPAG), private operators under the provisions of the Water Supply Delegated Management Framework13. At DNGRH, regular maintenance of the Macarretane dam and the dikes 14 will be carried out by existing government staff at the DGBL and ARA Centro, respectively. At MINEDH, routine maintenance of classrooms will be carried by each school administration and by provincial authorities, following the technical norms 15 approved by MINEDH and MOPHRH in 2021. For the irrigation schemes, INIR has hired a technical service provider to assist local farmers with agriculture production and access to markets under the IRRIGA project, in coordination with farmers associations and local community authorities, and the SUSTENTA project. V. LESSONS AND RECOMMENDATIONS 88. Initial damages assessment, selection of priority investments and implementation arrangements of the original project had informed the design and implementation of new recovery projects. The design of the original project was based on the outcomes and recommendations of a joint damaged assessment conducted in March 2015 by the Government, World Bank, European Union, and United Nations agencies. Only investments that would contribute to safeguard lives, economic production, livelihoods and to provided additional benefits associated for water supply and flood protection to similar return period were selected. The use of existing four PIUs with proven capacity on fiduciary management, supported by sector teams helped addressing the technical complexity of recovery works. It has also helped in the management of institutional capacity risks that could be triggered by having complex activities under one single PIU. A similar approach was used in 2019 for the design and implementation centered in two PIU’s for the Cyclones Idai and Kenneth Emergency Recovery and Resilience Project (P171040), based on outcomes of partners coordination in the preparation of the Post-Disaster Needs Assessment (PDNA). 89. To achieve acceptable levels of resilience, the recovery of infrastructure requires time, expertise and comprehensive technical studies. Due to the urgency of relaunching agricultural production in the Licungo River Basin, the initial rehabilitation of the Nante and Nicoadala dikes and the Intabo and Munda- Munda irrigation schemes under the ERRP project was done under emergency conditions, using designs 13 The Water Supply Delegated Management Framework was established in 1998 by Decree no.72/98 of December 23, the exploitation and management of water supply systems, can, by delegation, be entrusted to private operators through cessation, exploitation and/or management contracts. 14 A new regulation (Decree no. 78/2019, of September 19) on dikes for flood protection delegates the management and maintenance of dikes to Regional Water Administrations (ARAs). 15 These technical norms (approved by the Ministerial Diploma no.122/2021 of October 26) include: (a) a catalog of technical norms for specific hazard zones for cyclones, floods, droughts and earthquakes; (b) package of technical designs for resilient schools for each hazard zone; and (c) a standard design for resilient non-conventional schools. Page 26 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) and feasibility studies already prepared by Government, development partners and existing WB projects. As a result, all these infrastructures were further damaged by cyclone Idai in 2019, thus reflecting insufficient resilience design. The second rehabilitation of Munda-Munda irrigation scheme, all water supply infrastructure and classrooms were preceded by either a comprehensive vulnerability analysis or feasibility and technical studies, and each infrastructure was designed to be resilient to the prevalent flood, cyclone, and drought risks. This required significant time and effort from Government technical staff and the WB team to adequately conduct the necessary due diligence. The use of existing experienced government teams has played a critical role in providing quality assurance for the technical studies and designs prepared by several consultancies under the project. As a result, all infrastructure recovered or built following the recommendations of technical studies and designs showed adequate resilience to all cyclone and flood events that occurred between 2019 and early 2022. 90. Adequate coordination among all relevant stakeholders is critical to build synergies, alignment, and impact of recovery projects. The project established consultation, coordination and information exchange mechanisms with relevant developments partners, including with: (a) Department for International Development (DFID) and World Food Program (WFP) and other members of the Mozambique Humanitarian Country Team (HCT) for food support under the IRM; (b) Netherlands for phasing of rehabilitation of water supply system of Mocuba and the Munda-Munda irrigation system; (c) United Nations Development Program (UNDP), African Development Bank (AfDB) and European Union for preparation of cyclone Idai and Kenneth PDNA and recovery framework; (d) Germany, Finland, Canada, Ireland, Portugal, and France and United Nations Children Fund (UNICEF) on resilient rehabilitation of classrooms; and (e) DFID and the Netherlands on scale up of desalination plants and multipurpose. Through these coordination mechanisms, overlaps and duplication of efforts was avoided and complementarity and impact on the ground was increased, including gradual up taking of outcomes and lessons generated by the project. 91. The activation of IRM/CERC should be timely and based on emergency response country needs rather than on potential availability of additional financing. Reallocation of project funds for IRM/CERC activation requires candid discussions between Government, World Bank task teams and the CMU as there is significant risk of reduction of scope, activities, and targets of the original project if resources mobilized for IRM/CERC are not replenished through an additional financing. High Government awareness on this risk is a critical element that makes Government appreciate the benefits of having IRM/CERC when designing new projects and avoid conditioning the activation of IRM/CERC to the assurance of an additional financing. Proactivity is also required from Government to explore the opportunities provided by the Crises Response Window (CRW) and future IDA commitments as incentives to boost government’s interest for unconditional activation of IRM/CERC for timely response and recovery from crisis. 92. Ensuring flexibility in implementation of recovery projects is critical to provide adequate response to new disasters without changing the focus of the original design. The project was restructured five times to respond to: (a) impacts caused by droughts through the activation of IRM and the first AF; (b) cyclone Idai by accommodating the costs of repair of damages on infrastructure further affected by flooding in the Licungo river basin; (c) further flooding in the Licungo river through contract adjustments to address delays caused by temporary interruption of works; and (iv) impacts of COVID 19 across the country, through contracts adjustment to compensate the delays in delivery of works, and by providing temporary jobs to local communities. All restructurings were conducted timely in the aftermath of disasters and based on impacts assessment. This strategy helped the project to gradually deliver all outputs of the original project by the project closing date. Establishment of resilience design standards in all critical Page 27 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) sectors may help the Government to further increase country’s readiness to quickly implement recovery projects in the aftermath of large disaster events. 93. Cost management remained a serious challenge for implementation of works delayed due to IRM activation and further flooding. Due to urgent preparation of the original project, the cost for recovery of classrooms and irrigation schemes was underestimated, partially reflecting limited information availability at design in 2015. When tenders were launched in 2019, costs increased by 82 percent for conventional classrooms, 240 percent for non-conventional classrooms, and 42 percent for Munda- Munda irrigation scheme. To mitigate similar changes, costing of project activities at the design should be based on market price variation across the country, and accounts for likely price fluctuations related to occurrence of new disasters. 94. The timely delivery of works requires innovative monitoring approaches to rapidly detect and mitigate challenges. The introduction of monthly checking-in meetings with the PIU and technical staff in the water sector and weekly meetings with the PIU and UN-Habitat in the education sector to verify compliance with the works schedule, combined with images from the individual work sites and field visits, helped to unlock key implementation challenges and meet the deadlines. The use of a field-based third-party monitoring agency and remote monitoring tools such as Geo-Enabling Initiative for Monitoring and Supervision (GEMS), may help further improve monitoring and evaluation outcomes. . Page 28 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) ANNEX 1. RESULTS FRAMEWORK AND KEY OUTPUTS A. RESULTS INDICATORS A.1 PDO Indicators Objective/Outcome: Restore functionality of critical infrastructure in a resilient manner in disaster-affected Province Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Direct project beneficiaries Number 0.00 1,000,000.00 1,138,500.00 1,442,509.00 27-Jul-2015 29-Sep-2015 26-Jun-2020 31-Dec-2021 Female beneficiaries Percentage 0.00 50.00 50.00 50.00 Comments (achievements against targets): This indicator was exceeded by 26.7% to reflect increased access to education infrastructure, water supply systems, jobs and services provided by the rehabilitated Niquidua-Mualei road and the Macarretane dam. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of people in urban Number 7,760.00 24,300.00 31,860.00 33,695.00 Page 29 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) areas provided with access to 27-Jul-2015 29-Sep-2015 26-Jun-2020 30-Nov-2021 Improved Water Sources under the project Comments (achievements against targets): This indicator was exceed by 5.7% to reflect the continued increase in number of domestic water connections in the Mocuba water supply system Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of children with Number 0.00 588,400.00 78,000.00 159,772.00 access to improved education infrastructure 27-Jul-2015 29-Sep-2015 26-Jun-2020 30-Nov-2021 Comments (achievements against targets): The target was exceeded by 104% mainly due to the high number of students served per rehabilitated conventional classroom in Zambezia and Nampula provinces (85.6 students per classroom against the national average of 62 students per classroom). Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of people protected Number 0.00 111,700.00 69,000.00 69,000.00 by rehabilitated dike infrastructure 27-Jul-2015 29-Sep-2015 26-Jun-2020 22-Oct-2020 Comments (achievements against targets): Page 30 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) This indicator was achieved in October 2020 with the completion of physical works of rehabilitation of Lot 1 and Lot 2 of dikes of Nante and the dikes of Nicoadala Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of people with Number 0.00 5,600.00 5,600.00 5,600.00 access to improved irrigation infrastructure 27-Jul-2015 29-Sep-2015 26-Jun-2020 31-Dec-2021 Comments (achievements against targets): The target was reached with the completion of rehabilitation works of Munda-Munda irrigation scheme in December 2021 under the Smallholder Irrigated Agriculture and Market Access Project- IRRIGA (P164431) A.2 Intermediate Results Indicators Component: Component A - Resilient Infrastructure Rehabilitation Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Dike Infrastructure Kilometers 0.00 35.00 30.00 30.00 Rehabilitated 27-Jul-2015 29-Sep-2015 26-Jun-2020 22-Oct-2020 Comments (achievements against targets): Page 31 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) This indicator was achieved in October 2020 with the completion of physical works of the Lot 1 of Nante dike. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Rehabilitation of the Yes/No No Yes Yes Yes Macarretane Barrage downstream apron 24-Nov-2016 13-Jan-2017 26-Jun-2020 22-Oct-2020 Comments (achievements against targets): The rehabilitation of Macarretane dam was completed by end of August 2020. Two additional works were also completed: (i) the construction and relocation of office of the Limpopo River Basin Management Division (DGBL), and; (ii) the construction of a parking bay for the gantry crane used for manual operation of the 33 gates of the Macarretane dam Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Cropped land protected by Hectare(Ha) 0.00 30,000.00 30,000.00 30,000.00 rehabilitated dikes 27-Jul-2015 29-Sep-2015 26-Jun-2020 22-Oct-2020 Comments (achievements against targets): This indicator was achieved by October 2020 with the completion of lote 1 of dike of Nante, in addition to the Nicoadala dike and lote 2 of Nante dike completed in June 2020 Page 32 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Area provided with irrigation Hectare(Ha) 0.00 700.00 700.00 700.00 and drainage services (ha) 27-Jul-2015 29-Sep-2015 26-Jun-2020 31-Dec-2021 Area provided with Hectare(Ha) 0.00 700.00 700.00 700.00 irrigation and drainage services - Improved (ha) 27-Jul-2015 29-Sep-2015 26-Jun-2020 31-Dec-2021 Comments (achievements against targets): The target was achieved in 100% in December 2021 with the completion of the rehabilitation of Munda-Munda irrigation scheme (400ha) under the IRRIGA Project. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Piped household water Number 407.00 1,200.00 2,200.00 4,139.00 connections that are benefiting from 27-Jul-2015 29-Sep-2015 26-Jun-2020 30-Nov-2021 rehabilitation works undertaken by the project Comments (achievements against targets): This indicator was exceeded by 88% mainly due to increased number of domestic water connections in the Mocuba water supply system Page 33 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Improved community water Number 0.00 60.00 68.00 93.00 points constructed or rehabilitated under the 27-Jul-2015 29-Sep-2015 26-Jun-2020 30-Nov-2021 project Comments (achievements against targets): This indicator was exceed by 36.7% due to increased number of community water points constructed in desalination systems, multi-purpose boreholes in the districts of Chokwe, Chibuto, Massangena and Mapai, and in the water supply systems of Funhalouro and Chicualacuala Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Conventional classrooms Number 0.00 433.00 372.00 372.00 rehabilitated 27-Jul-2015 29-Sep-2015 26-Jun-2020 30-Nov-2021 Comments (achievements against targets): Target achieved in 100% Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Non-conventional ("mixed") Number 0.00 1,038.00 257.00 217.00 Page 34 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) classrooms built 27-Jul-2015 29-Sep-2015 26-Jun-2020 30-Nov-2021 Comments (achievements against targets): Target achieved partially (84.4%) as result of cancelation of construction of 40 non-conventional classrooms in Niassa province due to limited demand and low capacity of local contractors Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Administrative blocks built or Number 0.00 107.00 127.00 118.00 rehabilitated 01-Jan-2018 20-Jun-2018 26-Jun-2020 30-Nov-2021 Comments (achievements against targets): The target was achieved partially (92.9%) as result of cancelation of construction of 40 non-conventional classrooms in Niassa province due to limited demand and low capacity of local contractors Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Rain water harvesting and Number 0.00 50.00 111.00 150.00 storage systems built or rehabilitated 01-Jan-2018 20-Jun-2018 26-Jun-2020 30-Nov-2021 Comments (achievements against targets): Page 35 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) This indicator was exceed in 35% for compliance with a new directive at MINEDH that makes mandatory the construction of water harvesting systems in all schools to increase water availability as part of COVID 19 protection measures Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Toilets built or rehabilitated Number 0.00 202.00 364.00 408.00 01-Jan-2018 20-Jun-2018 26-Jun-2020 30-Nov-2021 Comments (achievements against targets): This indicator was exceed in 12% for compliance with a new directive at MINEDH for expansion of access and use of toilets as part of COVID 19 protection measures Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Jobs created in construction Number 0.00 1,176.00 1,176.00 3,161.00 works of non-conventional classrooms 01-Jan-2018 26-Jun-2020 26-Jun-2020 30-Nov-2021 Comments (achievements against targets): This indicator was exceed by 168.7% due to increased contracting of local labor for acceleration of construction of non-conventional classrooms Page 36 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Percentage of complaints Percentage 0.00 100.00 100.00 100.00 regarding project activities addressed per the grievance 01-Oct-2018 20-Jun-2018 26-Jun-2020 30-Nov-2021 redressal mechanism established by the project Comments (achievements against targets): The target was achieved in 100% at MINEDH Component: Component B - Technical Assistance for Resilient Recovery and Vulnerability Reduction Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Communities supported with Number 0.00 137.00 49.00 41.00 technical assistance from civil society organizations for 27-Jul-2015 29-Sep-2015 26-Jun-2020 30-Nov-2021 mixed school construction Comments (achievements against targets): The target was achieved partially (83.6%) as result of cancelation of construction of 40 non-conventional classrooms in Niassa province due to limited demand and low capacity of local contractors Page 37 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Component: Component D - Contingent Emergency Response (CERC/IRM) Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Immediate Response Yes/No No Yes Yes Yes Mechanism (IRM) established and ready to 27-Jul-2015 29-Sep-2015 31-Oct-2016 31-Dec-2019 provide access to financial resources in case of an eligible crisis or emergency Comments (achievements against targets): The closing of the IRM was delayed due to delays in the import and distribution of the irrigation kits by INIR. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Number of people provided Number 0.00 519,138.00 519,138.00 519,138.00 with food aid (Thousand) 01-Jul-2016 02-Nov-2016 29-Jun-2018 29-Jun-2018 Comments (achievements against targets): The target was achieved in 100% Indicator Name Unit of Measure Baseline Original Target Formally Revised Actual Achieved at Page 38 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Target Completion Number of people provided Number 0.00 146,248.00 146,248.00 244,394.00 with nutrition supplies 01-Jul-2016 02-Nov-2016 29-Jun-2018 31-Dec-2018 Comments (achievements against targets): The target was exceeded in 67% due to increased capacity for treatment of malnourished patients by the national health system. Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion People provided with Number 0.00 1,800.00 1,800.00 1,800.00 temporary drinking water systems 01-Jul-2016 02-Nov-2016 28-Sep-2018 28-Sep-2018 Comments (achievements against targets): The target was achieved in 100% Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion Farmers provided with small Number 0.00 4,230.00 4,230.00 4,230.00 irrigation systems 01-Jul-2016 02-Nov-2016 31-Dec-2019 31-Dec-2019 Comments (achievements against targets): Page 39 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) This target was achieved in 100% Formally Revised Actual Achieved at Indicator Name Unit of Measure Baseline Original Target Target Completion People provided with Number 0.00 246,630.00 246,630.00 246,630.00 improved seeds 01-Jul-2016 02-Nov-2016 29-Jun-2018 29-Jun-2018 Comments (achievements against targets): This target was achieved in 100% Page 40 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) B. KEY OUTPUTS BY COMPONENT Objective/Outcome 1. Restore the functionality of critical infrastructure in a resilient manner in the disaster-affected provinces 1. Direct project beneficiaries, percent of which female beneficiaries 2. Number of people protected by rehabilitated dike infrastructure Outcome Indicators 3. Number of people with access to improved irrigation infrastructure 4. Number of people in urban areas provided with access to improved water sources under the project, and 5. Number of children with access to improved education infrastructure 1. Dike Infrastructure Rehabilitated (Kilometers) 2. Rehabilitation of the Macarretane Barrage downstream apron (Yes/No) 3. Cropped land protected by rehabilitated dikes (Ha) 4. Area provided with irrigation and drainage services (Ha), 5. Area provided with irrigation and drainage services – Improved (Ha) 6. Piped household water connections that are benefiting from rehabilitation works undertaken by the project (Number) 7. Improved community water points constructed or rehabilitated under the project (Number) Intermediate Results 8. Conventional classrooms rehabilitated Indicators 9. Non-conventional (mixed) classrooms built (Number) 10. Administrative blocks built or rehabilitated (Number) 11. Rainwater harvesting and storage systems built or rehabilitated (Number) 12. Toilets built or rehabilitated (Number) 13. Jobs created in construction works of non-conventional classrooms (Number) 14. Percentage of complaints regarding project activities addressed per the grievance redressal mechanism established by the project (Percentage) 15. Communities supported with technical assistance from civil society organizations for mixed school construction (Number) Key Outputs by Component A. Resilient Infrastructure Rehabilitation Component 1. 30 km of dikes rehabilitated in Nante and Nicoadala (linked to the 2. Eribacela weir rehabilitated achievement of the 3. two irrigations schemes (Munda-Munda and Intabo) rehabilitated totaling 700 ha. Page 41 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Objective/Outcome 4. 15 km of unpaved road connecting Niquidua and Mualei rehabilitated 1) 5. 18 km of powerline rehabilitated 6. 4,139 domestic water connections and 29 public standpipes completed in Mocuba, Funhalouro and Chicualacuala 7. 80.5 km of water distribution network rehabilitated or constructed in Mocuba, Funhalouro and Chicualacuala 8. Feasibility study and detailed design for Phase II of rehabilitation of Mocuba water supply and Environmental and Social Impact Assessment (ESIA) completed 9. 4 desalination water systems and 4 multipurpose boreholes built 10. Management System for Operation and Maintenance of desalination systems installed 11. private operators hired for operation and management of desalination plants 12. 98 community water points constructed in rural areas of Gaza province 13. Macarretane dam rehabilitated 14. Building of offices of the DGBL built 15. Parking Bay of the gantry crane of Macarretane dam constructed 16. 372 conventional classrooms rehabilitated in the provinces of Niassa, Nampula and Zambezia 17. 217 non-conventional classrooms constructed in the provinces of Nampula and Zambezia 18. 118 administrative blocks, 150 rainwater harvesting systems, 408 toilets built or rehabilitated in the provinces of Niassa, Nampula and Zambezia 19. 3,161 people recruited from local communities as labor for construction and rehabilitation of classrooms 20. Grievance Redress Mechanism established in all PIU 21. 20 complaints received, of which 16 addressed. Component B: Technical Assistance for Resilient Recovery and Vulnerability Reduction TA to MINEDH 1. UN-Habitat contracted to provide technical assistance to MINEDH 2. 163 schools assessed, 8 technical designs packages, guidelines, and tender documents prepared for MINEDH 3. Training sessions provided to 39 contractors, engineering firms, NGOs and to 41 local communities on resilience construction of classrooms 4. Supervision and on-the job training to NGOs and contractors implemented TA to INGD 1. International consultant hired to provide technical assistance to INGD Page 42 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) 2. Investment Plan for the Master Plan for Disaster Risk Reduction (2017-2030) drafted 3. Written inputs for the revision of the Mozambique DRM Law prepared Objective/Outcome 2. Improve the Government of Mozambique’s capacity to respond promptly and effectively to an eligible crisis or emergency 1. Direct project beneficiaries, percent of which female beneficiaries Outcome Indicators 1. Immediate Response Mechanism (IRM) established and ready to provide access to financial resources in case of an eligible crisis or emergency (Yes/No) 2. Number of people provided with food aid (Number) Intermediate Results 3. Number of people provided with nutrition supplies (Number) Indicators 4. People provided with temporary drinking water systems (Number) 5. Farmers provided with small irrigation systems (Number) 6. People provided with improved seeds (Number) 1. IRM Operations Manual approved 2. USD20 million mobilized for IRM activation to finance drought response Key Outputs by 3. 1,016,192 people provided assistance under the IRM interventions Component 4.5,693.16 tons of food items distributed to 519,138 people (linked to the 5. 762.1 tons of certified seeds distributed to 246,630 farmers achievement of the 6. 1,781.05 tons of nutritional supplements acquired and delivered to the national health system Objective/Outcome 7. 244,394 people (children and pregnant women) treated against malnourishment 2) 8. Bills of medicines and health equipment for the national health system paid for 9. Two desalination plants installed in the districts of Chicualacuala and Chigubo to provide water to 1,800 people 10. 4,200 irrigation kits equipped with solar panel and 30 drag lines distributed to farmers and agricultural schools ] Page 43 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) ANNEX 2. BANK LENDING AND IMPLEMENTATION SUPPORT/SUPERVISION A. TASK TEAM MEMBERS Name Role Preparation Michel Matera, Jean Baptiste Migraine Task Team Leader(s) Antonio Laquene Chamuco Procurement Specialist(s) Elvis Teodoro Bernado Langa Financial Management Specialist Luiz Claudio Martins Tavares Team Member Luis M. Schwarz Team Member Adrianus Verweij Team Member Aniceto Timoteo Bila Team Member Mark A. Austin Team Member Christoph Pusch Team Member Luz Meza-Bartrina Counsel Johanna van Tilburg Safeguards Advisor/ESSA Cheikh A. T. Sagna Social Specialist Andre L. Carletto Team Member Shelley Mcmillan Team Member Maiada Mahmoud Abdel Fattah Kassem Team Member Enrique Blanco Armas Team Member Luis Macario Team Member Chalida Chararnsuk Team Member Fernando Ramirez Cortes Peer Reviewer Page 44 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Pieter Waalewijn Peer Reviewer Fadila Caillaud Team Member Saurabh Suresh Dani Peer Reviewer Arlete Quiteria Comissario Nkamate Team Member Alfredo Ricardo Zunguze Team Member David Malcolm Lord Team Member Paulo Jorge Temba Sithoe Social Specialist Daniel Baloi Team Member Gerard Joseph Mataban Jumamil Counsel Eden Gabriel Vieira Dava Social Specialist Laurence Elodie Esther Fanny Chalude Team Member Supervision/ICR Lizardo Narvaez Marulanda, Xavier Agostinho Chavana Task Team Leader(s) Antonio Laquene Chamuco, Amos Martinho Malate Procurement Specialist(s) Joao Tinga Financial Management Specialist Eden Gabriel Vieira Dava Team Member Paulo Jorge Temba Sithoe Environmental Specialist Bruno Alberto Nhancale Environmental Specialist Emerson John Vasco Siquice Team Member Nilsa Come Team Member George Ferreira Da Silva Team Member Carlos Mondlane Team Member Sheila Pene Team Member Mariana Margarita Montiel Counsel Rildo Santos Team Member Aniceto Timoteo Bila Team Member Salma Omar Social Specialist Page 45 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Eden Dava Social Especialist Pierre Boulenger-Xavier Team Member Jaime Palalane Team Member Lucia Nhampossa Team Member Maria Isabel Nhassengo-Massingue Procurement Team B. STAFF TIME AND COST Staff Time and Cost Stage of Project Cycle No. of staff weeks US$ (including travel and consultant costs) Preparation FY16 9.176 59,527.57 FY17 4.293 7,903.46 FY18 2.305 6,251.24 FY19 2.591 8,956.15 FY20 .335 - 76.40 Total 18.70 82,562.02 Supervision/ICR FY16 6.247 55,219.57 FY17 8.950 72,122.86 FY18 10.630 77,501.22 FY19 13.132 116,082.85 FY20 17.666 84,974.31 Total 56.63 405,900.81 Page 46 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) ANNEX 3. PROJECT COST BY COMPONENT After 1st 2nd AF Restructuring Actual at Percenta Original Restructuring AF (Jan (Dec (Dec 2018) Project ge of financing (2016) 2017) 2018) Closing Approval (US$M) (US$M) Component A. Resilient Infrastructure 31.00 46.68 100% Rehabilitation 16.00 32.5 44.68 46.68 Component B. Technical Assistance for Resilient 6.00 1.85 100% Recovery and Vulnerability Reduction 1.50 1.5 1.85 1.85 Component C. Project Implementation, 3.00 4.47 100% Monitoring and Evaluation 2.50 3.00 3.47 4.47 Component D. Contingency Emergency 0.00 20.00 100% Response (CERC) 20.0 20.00 20.00 20.00 Component E. --- --- 3.00 3.00 --- --- Total 40.00 40.00 60.00 73.0 73.0 73.00 100.00 Page 47 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) ANNEX 4. EFFICIENCY ANALYSIS Summary: An estimated 1.44 million people benefited from the project activities. The project covered the rehabilitation and construction of dikes, rural infrastructure, classrooms, water supply systems, the Macarretane barrage, water desalination plants, as well as drought response and emergency support, including food and medicine, seeds, nutritional supplements and irrigation draglines and kits. The economic analysis focused on the most expensive components of the project, including the rehabilitation of dikes, water supply systems, classrooms, rural infrastructure and the Macarretane barrage. Together these investments made up 65 percent of total project costs. Expected benefits of these activities were compared to both the cost to deliver these specific interventions and the total cost of the project. The computed internal rate of return (IRR) was 40 percent, and the corresponding net present value (NPV) was US$98 million when benefits were compared to the total project cost. When applying only the cost of the components for which benefits were quantified, the IRR was 39 percent, with a corresponding NPV of US$110 million. A 12 percent discount rate was used, and standard assumptions on the longevity of assets and the maintenance costs (here assessed at 3 percent of total capital cost) were made. Translating the financial cost to economic cost: Shadow pricing is an important aspect of converting the financial cost of a project to its economic cost. The concept involves subtracting and adding the distortions that are introduced by taxes and subsidies to economic sectors and developing an understanding of the procurement requirements to meet each component’s deliverables. For example, when procuring international goods, one would subtract any import tax (or add any import subsidy) and correct other nontrade barriers imposed by government that would make the international price diverge from the local price. In the absence of a detailed procurement plan for each component, the following conversions were made and assumed, and outlined in Table 1. The shadow price of local labor was assumed at half of its price, international labor was assumed to be competitive, and an assumed 20 percent import tax was deducted from the price of all goods. The total financial project cost of US$73 million translated to a US$51 million economic cost according to the assumptions in Table 1. Table 1. Economic costs based on allocation across local and international labor and goods Share Share Financial local international Share of Economic Subcomponents costs labor labor goods costs Nante dikes 7.5 40% 0% 60% 5.1 Nicoadala dikes 2.0 40% 0% 60% 1.4 Munda-Munda irrigation scheme Intabo irrigation scheme 5.5 40% 0% 60% 3.7 15 km road 18 km grid Comp 1 Chicualacuala water supply system 5.5 30% 10% 60% 4.0 Funhalouro water supply system Mocuba water supply system 6.5 30% 10% 60% 4.8 Classrooms 8.7 30% 10% 60% 6.3 Macarretane barrage 10.0 30% 10% 60% 7.3 Water desalination systems 1.0 30% 10% 60% 0.7 Page 48 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Comp 2 Technical assistance 1.9 40% 60% 0% 1.5 Comp 3 Project management 4.5 90% 10% 0% 2.5 Water desalination plants 0.6 40% 0% 60% 0.4 Improved irrigation 4.4 40% 0% 60% 3.0 Food aid 4.6 40% 0% 60% 3.1 Comp 4 Health (medicines) 5.0 40% 0% 60% 3.4 Nutritional supplements (Health) 4.1 30% 10% 60% 3.0 Seeds 1.3 40% 0% 60% 0.9 Total 73.0 51.1 The approach to measuring the economic benefits of the intervention relied on a combination of approaches and focused on the largest components of the project in terms of costs; namely: • Benefit: improved agriculture productivity due to irrigation and risk reduction - Rehabilitation of dikes Nante and Nicoadala (beneficiaries 69,000) - Rehabilitation of irrigation systems Munda-Munda and Intabo (beneficiaries 29,000) - Rehabilitation of Macarretane Barrage (beneficiaries 21,000) - 18 km electric grid providing electricity to the irrigation systems Munda-Munda and Intabo • Benefit: time savings - Rehabilitation and construction of water supply systems in Chicualacuala, Funhalouro and Mocuba (beneficiaries 47,500) - Construction of 15 km road (52,050) • Benefit: education - Rehabilitation and construction of classrooms (beneficiaries 159,771) Risk reduction and improved agricultural productivity Maganja da Costa and Nicoadala in Zambezia are densely populated areas in which agriculture is the main livelihood. The area suffered significant losses during the 2015 floods, amounting to US$30 million. These areas are now protected by the Nante and Nicoadala dikes, which were rehabilitated under the project. The protected area also contains the rehabilitated irrigation systems of Munda-Munda (400 ha) and Intabo (300 ha), which help farmers regulate water levels on farmland. The systems lower the risk for both flooding and drought and make it possible for farmers to grow horticulture in the off season. The project also rehabilitated the 18 km grid that provides electricity for the irrigation schemes. The benefits of the combined investments in Maganja da Costa and Nicoadala include both avoided losses and improved agricultural productivity. Avoided annual losses were quantified using the value of the losses from the 2015 flood. 16 In addition to avoided losses, the benefit of an additional growing season in the irrigated area (700 ha) is added by using an estimate for net return of tomato revenue per ha in Mozambique. 17 However, it is reasonable to assume that some farmers would still attempt a second growing season even in the absence of an irrigation scheme. The analysis therefore subtracts avoided losses from the benefits of an additional growing season using a set of assumptions about rainfed, dry season production. For example, with a combination of maize and bean production, it is assumed that farmers use only 50 percent of their land for a second growing season in the absence of irrigation. This is an optimistic estimate, since studies on rain-fed agriculture in Zambezia found 16In the absence of flood protection, the risk of a similar flood event happening in any given year is 3.3%. 17Comparative analysis of tomato value chain competitiveness in selected areas of Malawi and Mozambique (Mango, et al., 2015) https://orcid.org/0000-0002-5241-3777 Page 49 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) that only 10 percent of farmers used the second growing season for maize and bean production. 18 This approach renders a conservative estimate of the real benefits of the investment, since the dikes will also protect the area from more frequent flooding. Assuming a 10-year lifespan of all the investments needed to generate productivity and flood protection benefits (dikes, grid and irrigation systems), the investment generates an annual benefit of US$1.1 million. The rehabilitation of the barrage of Macarretane also supports irrigation, since the barrage regulates the water flow and enables continuous year-round water access for two of the largest and most important irrigation systems in Mozambique – Chokwe and Xai-Xai (33,000 and 11,000 ha, respectively), located in the Lower Limpopo area. These irrigation systems enable rice production in the hot rainy season. To calculate the benefit of irrigated rice production, the analysis uses the productivity gain of irrigated compared to rainfed rice production. A cross-country study of 12 Sub-Saharan countries finds that rainfed rice yield is 0.6-2.3 tons per ha, while irrigated rice yield is between 2.5-5.6 tons per ha. 19 As a conservative estimate, this analysis assumes a productivity increase from 2.3 to 2.5 t per ha, which is a 9 percent improvement and the lowest possible improvement according to the study. In addition, the irrigation schemes enable an additional growing season, for which the assessment uses the net return of tomato revenue per ha to quantify the benefit. Similar to the assessment in Zambezia, it is realistic to assume that some farmers would still use the second growing season for production even in the absence of irrigation. Farmers located in non-irrigated areas in the Lower Limpopo use the second growing season for maize and bean production, but productivity and household income are much lower than in the irrigated areas. 20 It is assumed that only half the area is used for dry season production, and the value of this production is subtracted from the benefit of the additional growing season. A conservative 20-year lifespan of the rehabilitation of the barrage is assumed, and the annual benefit of the investment is US$9.2 million. The project also provided 4,200 small-scale drip irrigation kits and 30 draglines to farmers. The draglines can irrigate 5 ha per unit and 30 units were provided as part of the project. The small-scale irrigation kits were distributed to 4,200 farmers with 0.1-1 ha of land. To capture the benefits of this investment, it is assumed that the irrigation system can support an additional growing season, and the net return of tomato revenue per ha is used. To be conservative, a 0.1 ha of land per farmer is assumed (which was the minimum of land ownership needed to access the kits). Similar to the analysis in Zambezia and Gaza, a lower production in the dry season, using a combination of maize and beans, is subtracted from the benefits of a more productive additional growing season. With a 5-year lifespan assumed, the annual benefit of the irrigation kits and draglines is US$119,052. Water supply systems The project also included rehabilitation and construction of water supply systems in Chicualacuala, Funhalouro and Mocuba. Improved water supply systems led to health and time-saving benefits. Since there is uncertainty about health benefits, this assessment quantified the time savings. In Chicualacuala and Funhalouro, before the project, the population relied on wells to fetch water and, according to Household Expenditures Survey (IOF), travel time to a well varied from 51 minutes in Chicualacuala to 106 minutes in 18 Access to Markets for Smallholder Farmers in Alto Molocue and Molumbo, Mozambique (Erman, et al., 2016). file:///C:/Users/wb454020/Downloads/p15738coll2_130451.pdf 19 Assessing rice production sustainability performance indicators and their gaps in 12 Sub-Saharan African countries (Arouna, et al., 2021). https://www.sciencedirect.com/science/article/pii/S0378429021002094 20 Livelihood baseline profiles, Limpopo Basic, Mozambique (FEWSNET, 2011). https://fews.net/sites/default/files/documents/reports/mz_baseline_rural%20limpopo_en.pdf Page 50 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Funhalouro. In Mocuba, it is assumed that without the investment, the population would rely on local wells and water streams to fetch water, and according to IOF, the average travel time to a well is 40 minutes. Installation of 2,228 yard taps, 26 public standpipes and 1,302domestic water connections under the project substantially reduced this travel time. Due to lack of exact information on the new travel times in each town, the assessment used average travel time to yard taps and public standpipes (16 and 39 minutes, respectively) to proxy the time-saving benefit. Assuming a 10-year lifespan of the investments in the water supply systems, the annual benefit is US$2.3 million. Key assumptions of the time-saving assessment: • Consumption of water is estimated at 25 liters per day per person (including drinking, hygiene, and cooking needs). • Household size is 4.9 in Mocuba and 6.1 in Funhalouro and Chicualacuala. • One person can carry 20 liters of water in one trip. • Assumed travel times from IOF are summarized in Table 2. • Minimum wage in agriculture in Mozambique is 4,390 meticais per month, 21 equivalent to US$2 per day, and 45 percent of the population works (58 percent are of working age, and 76 percent of the working-age population is employed 22). Table 2. Water access and travel time Well (before project) In-house Yard tap* Public standpost* travel travel travel travel time No. built time No. built time No. built time No. built Chicualacuala 51 0 0 0 16 702 39 7 Funhalouro 106 0 0 0 16 600 39 4 Mocuba 40 0 0 609 16 2,228 39 12 *Average travel time is from IOF; Mozambique average used by the National Statistics Institute (INE) for yard taps and public standpipes. Rehabilitation and construction of classrooms The project rehabilitated 372 conventional classrooms and constructed 217 non-conventional classrooms, providing education for a total of 159,771 students. This assessment takes into account that communities tend to find ways to provide education even in the absence of a physical classroom. They either build their own non-conventional classrooms or educate children outdoors. As a result, the main benefit of the rehabilitation of conventional classrooms and construction of non-conventional classrooms under the project was to enable more students to receive education than would otherwise have been possible. Non- conventional classrooms constructed by local communities are smaller and provide limited protection from hazard events. Evidence suggest that parents are unwilling or hesitant to send their children to a school without proper facilities, driving up dropout rates or forcing children to travel long distances to access education. One analysis found that building or renovating a school will decrease the dropout prevalence by 1.9 percent. 23 The analysis uses the average return of an additional year of education, as presented in 21 2019 Minimum wage Mozambique (Avillez Bacar Centeio & Cambule, 2019). https://www.abcc.co.mz/xms/files/Minimum_wages_2019.pdf 22 Mozambique Jobs Diagnostic: Volume 1 - Analytics (Lachler & Walker, 2018). 23 An analysis of school dropout in Mozambique 2014-2015 (Mambo, et al., 2019). https://www.wider.unu.edu/sites/default/files/Publications/Working-paper/PDF/wp-2019-49.pdf Page 51 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Psacharopoulos and Patrinos (2018), 24 to quantify the benefit of education. The paper reviews 1,120 estimates of the rate of return of education in 139 countries and finds that the rate of return of one additional year of education in Sub-Saharan Africa is 10.5 percent. Other studies find both higher 25 and lower 26 rates of return of education for Mozambique, providing some confidence that 10.5 percent is a conservative yet insightful estimate. The average size of a non-conventional classroom is 30 square meters, and the average size of conventional classroom is 56 square meters. 27 (The non-conventional classrooms constructed under the project had the same dimensions as conventional classrooms.) It is therefore assumed that non-conventional classrooms can only provide education for about half as many students as conventional classrooms. This is a conservative estimate because, beyond the size of classrooms, other factors are also at play in limiting the ability of communities to provide education in non-conventional classrooms. Considering the above, it is assumed that without the investments made under the project, only 85,592 out of the 159,771 students would have attained education. In any given year, it is assumed that 5.0 percent of students will drop out, but using the results of the study mentioned above, it is assumed that the rehabilitation and construction of classrooms will mitigate the drop-out rate by 1.9 percent. The net number of students for which the investment has enabled education in any given year is therefore 70,630 students. The monetized value of this, considering a lifespan of 20 years of the classrooms built under the project, a minimum wage increase for jobs in agriculture 28 and a 79 percent labor force participation, 29 the annual benefit adds up to US$5.0 million. In the sensitivity analysis, five key assumptions were altered to explore the impact on results, shown in Table 3. None of the assumed parameters are highly sensitive, even when adjusted jointly. The conclusion to be drawn from here is that the results are extremely robust, despite opting for conservative assumptions. Table 3. Sensitivity analysis Assumptions NVP IRR 50% lower price of rice AND 50% drop in tomato production $25,442,390 18% 50% lower employment rates $64,261,322 26% 50% lower wage AND 50% lower return to education $53,130,468 23% 50% fewer beneficiaries of all investments $12,387,125 15% 50% shorter lifespan of all investments $56,349,678 29% ANNEX 5. BORROWER AND OTHER PARTNER COMMENTS Comments from Borrower A Project completion report was prepared by the Client and submit to the World Bank on August 30, 2021. 24 Returns to Investment in Education: A Decennial Review of the Global Literature (Psacharopoulos & Patrinos, 2018). 25 Returns to Education in Mozambique: Learning from a Quasi-Experiment (Simione, 2009). 26 The evolution of private returns to education during post-conflict transformation: Evidence from Mozambique (Jones et al, 2018) 27 Multi-Hazard Risk Assessment for the Schools Sector in Mozambique (Briere, et al. 2018) 28 2019 Minimum wage Mozambique (Avillez Bacar Centeio & Cambule, 2019) https://www.abcc.co.mz/xms/files/Minimum_wages_2019.pdf 29 Mozambique Jobs Diagnostic: Volume 1 - Analytics (Lachler & Walker, 2018) https://documents1.worldbank.org/curated/en/655951534181476346/pdf/Mozambique-Jobs-Diagnostic-Volume-1- Analytics.pdf Page 52 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) The draft ICR was shared with the Client (DNGRH, AIAS, INIR, MINEDH, DNAAS) and UN-Habitat on April 18, 2022. The ICR team received written comments from DNGRH, AIAS and INIR between April 19-22, 2022. Comments provided by DNGRH • DNGRH agreed with the ratings proposed in the ICR. The DNGRH highlighted that the indicators and targets in the report are in accordance with the design and changes agreed during implantation. • DNGRH raised concerns on lessons learned related to rehabilitation and performance of dikes of Nante and Nicoadala to provide the expected protection to infrastructure, business, and human settlements against flooding from the Licungo river basin. • DNGRH confirmed that the rehabilitation of the dikes was resilient, and the works were designed for a 20-year return period flood. For the rehabilitation works, structural measures focused in the construction of hydraulic infrastructures to allow greater water flow, reduce the impact on the dikes slopes, and prevent sliding, cut-off and downstream erosion. • DNGRH noted that to achieve better performance of the dikes there is need for construction of a dam infrastructure upstream, to store the high volumes of water during heavy rainfalls events, delay the flood wave, therefore allowing better management of response capacity of dikes. • DNGRH also pointed out the increased frequency of occurrence of tropical storms in the Nante and Nicodala region due to climate changes, notably cyclones Idai, Kenneth, Eloise, Chalane, Guambe, Ana and Gombe, which generated high water flows that reached between 8,000 m3/s to 10,000 m3/s. • DNGRH emphasized the need for additional structures measures are required for floods management, which includes the construction of a dam to help regularize the water flows and mitigate the negative impacts on downstream locations, such as Nante Administrative Post and other nearby districts. Comments provided by AIAS • AIAS agreed with the content of the ICR and provided additional information on the impacts of Tropical Cyclone Gombe on the water supply system of Mocuba. • AIAS confirmed that the water supply system of Mocuba suffered limited damages to the access road to the intake due to erosion caused by tropical cyclone Gombe, in March 2022 • AIAS made edits to Table 2 in the Annex 4 to ensure consistency between the number of yards taps and domestic connections with the numbers presented in paragraphs 30 and 32 and in the results framework. Comments provided by INIR • INIR commended the quality of the ICR and made some edits to the report to clarify the leading role provided by the IRRIGA project in the management of Intabo, Munda-Munda irrigation schemes, in coordination with SUSTENTA Project. INIR pointed the need for the lessons learned to stress of investment in institutional capacity for the development of sustainable irrigation infrastructures. INIR also pointed out the need of involvement of local authorities in supporting producers (associates) to ensure good management of irrigation schemes. • INIR also recognized the need to focus on the development of commercial agriculture through market linkages, as a critical factor to boost the motivation and engagement of local farmers to increase the use of improved technologies and intensification of production, therefore contributing to increased yields. INIR believes that these changes will increase the ownership of the irrigation schemes by the farmers, and who are expected to contribute to their longevity and sustainability. Page 53 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) ANNEX 6. SUPPORTING DOCUMENTS A. Map. Project geographic areas Page 54 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) B. Changes to Indicators and targets Original First AF Second AF Restructuring Restructuring Project closing financing (Jan 2017) (Oct 2018) (Dec 2018) (April 2020) (Nov 2021) (July 2015) PDO Indicators Direct project beneficiaries 1,000,000 1,400,000 1,400,000 1,400,000 1,138,500 1,442,509 Female beneficiaries 50 50 50 50 50 (Percentage) Number of people in urban areas 65,300 65,300 65,300 31,860 33,695 24,300 provided with access to Improved Water Sources under the project Number of children with access to 588,400 353,600 353,600 78,000 159,772 588,400 improved education infrastructure Number of people protected by 111,700 111,700 69,000 69,000 69,000 69,000 rehabilitated dike infrastructure Number of people with access to 5,600 5,600 5,600 5,600 5,600 5,600 improved irrigation infrastructure Number of people supported by early 500,000 warning and response systems. 500,000 500,000 Cancelled Cancelled Cancelled Intermediate Results indicators Dike Infrastructure Rehabilitated 35 35 35 35 30.0 30 Rehabilitation of the Macarretane --- Barrage downstream apron (Yes/No) Yes Yes Yes Yes Yes Cropped land protected by rehabilitated 30,000 30,000 30,000 30,000 30,000 30,000 dikes (Hectare(Ha) Area provided with irrigation and 700 700 700 700 700 700 drainage services (Hectare(Ha) Area provided with irrigation and 700 drainage services - Improved 700 700 700 700 700 Page 55 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) (Hectare(Ha) Piped household water connections that are benefiting from rehabilitation 1,200 2,200 2,200 2,200 2,200 4,139 works undertaken by the project Improved community water points constructed or rehabilitated under the 60 116 68 68 68 93 project Conventional classrooms rehabilitated 433 433 372 372 372 372 Non-conventional ("mixed") classrooms 1,038 512 512 257 217 1,038 built Communities supported with technical assistance from civil society organizations 137 137 68 68 49 41 for mixed school construction Administrative blocks built or 107 127 118 --- --- 107 rehabilitated Rain water harvesting and storage 50 111 150 --- --- 50 systems built or rehabilitated Toilets built or rehabilitated --- --- 202 202 364 408 Jobs created in construction works of --- --- --- non-conventional classrooms ---- 1,176 3,161 Percentage of complaints regarding project activities addressed per the --- --- 100 grievance redressal mechanism 100 100 100 established by the project Communities/villages covered by a new Cancelled Cancelled Cancelled or updated emergency response plan 150 150 150 linked to improved early warning capacity Licungo watershed management study Cancelled Cancelled Cancelled 100 100 100 completed (Percentage) Immediate Response Mechanism (IRM) established and ready to provide access Yes to financial resources in case of an Yes Yes Yes Yes Yes Page 56 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) eligible crisis or Emergency (Yes/NO) Number of people provided with food aid --- 519,138 519,138 519,138 519,138 519,138 Number of people provided with 146,248 146,248 146,248 244,394 --- nutrition supplies People provided with temporary drinking 1,800 1,800 1,800 1,800 --- water systems Farmers provided with small irrigation 4,230 4,230 4,230 4,230 ---- systems People provided with improved seeds ---- 246,630 246,630 246,630 246,630 C. Changes in Components and Resources allocations Original 1st 2nd AF (Oct Restructuring Restructuring Restructuring Project closing Components/Sub-components financing AF (Jan 2017) 2018) (Dec 2018) (April 2020) (2016) (Nov 2021) Component A. Resilient 31.0 16.0 Infrastructure Rehabilitation 16.5 12.18 46.68 46.68 46.68 Reposition Reduction of of resources Increased financing in Sub- to Sub- alocation 32.18 32.18 Sub-components A1-A4 31.0 components A2- --- components A4 A2-A4 (-15.0) (+12.18) (+2.0) 3 new --- Sub-components A5-A7 -- --- subcomponents 16.5 16.5 (+16.5) Component B. Technical Assistance for Resilient --- --- 1.85 1.85 1.85 6.0 --- Recovery and Vulnerability Reduction -- Increase of --- 1.35 1.35 Sub-component B1 1.0 --- resources (+0.35) Page 57 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Cancelled these Reduction of Subcomponent B2-B3 5.0 --- --- sub-components 0.5 0.5 financing (-4.5) (-3) Increased Reposition Increased Component C. Project Reduction of resources of resources resources 4.47 4.47 Implementation, Monitoring 3.0 financing and Evaluation (-0.5) (+0.5) (+0.47) (+1.0) Component D. Contingency New allocation --- --- --- 20.0 20.0 0 Emergency Response (CERC) (+20.0) New alocation --- Cancelled this 0.0 0.0 Component E. Food security --- -- (+3.0) component emergency interventions (-3.0) Total 40.00 40.0 20.00 13.0 73.0 73.00 73.00 Page 58 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) D. Split rating for overall outcome Original objectives and Revised objectives and/or targets with targets without restructuring restructuring of June 2020 Relevance of PDO Substantial Efficacy (PDO) Substantial High A: Restore the functionality of critical infrastructure in a Substantial High resilient manner in the disaster-affected provinces B: Improve the Government of Mozambique’s capacity to High respond promptly and effectively to an eligible crisis or emergency Efficiency Substantial 1 Outcome ratings Moderately satisfactory Satisfactory 2 Numerical value of the outcome ratings 4 5 3 Disbursement US$ 49.63 million US 22.89 million 4 Share of Disbursement 0.68 (or 68%) 0.31 (or 31%) Weighted value of the 2.72 1.55 5 outcome rating (Row 2 X Row 4) Final Outcome rating Moderately satisfactory 6 2.72+1.55=4.25 (rounding it to 4.0) E. The benefits of rehabilitation of the Macarretane dam, in Gaza Province The Macarretane dam was built in the late 1950’s with three key features: (i) a 650 meters long weir equipped with 33 gates to elevate the water levels in the Limpopo river for irrigation of 33,000 hectares of the Chokwe Irrigation scheme; a railway and a road bridge system on the top of the weir connecting the capital city of Maputo and the neighboring Zimbabwe, and also serving the Gaza province northern districts of Guija, Massingir, Mabalane, Mapai, Chicualacuala and Massangena; (iii) a gantry crane that moves on top of the roads platform used for regular manual maintenance of gates of the Macarretane dam. The dam was primarily built for drought mitigation at Chokwe irrigation scheme, but in the rainy season it is also used for flood control through adequate automated and manual operation of gates. The catastrophic floods that hit the region in 2000 damaged some gates and entirely submerged the offices of the Limpopo River Basin Management Division (DGBL) causing loss of information and equipment and endangering the operation of the dam. Rehabilitation works were implemented between 2002-2006. These works introduced hydraulic changes which triggered scouring on the dissipation terrace of the dam. New flood events in 2013 further aggravated the scouring which reached 9 meters deep, destabilized the protection banks, put the dam structure at risk of inclination and undermined the safe operation of the dam. The offices of DGBL were flooded again causing further losses of equipment and information on dam management. On the other hand, trucks crossing on the bridge at night in average cause 5 to 6 crashes against the gantry per month. The most serious of these accidents often take 20 days or more for the repair of the crane and reopening of road traffic. Besides the cost incurred by the southern Regional Water Administration (ARA Sul) for the repair of the crane, the interruption of road traffic across the bridge often forced the rerouting of Page 59 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) traffic to Chinhacanine (15 km south of Macarretane dam), at additional cost to the drivers associated with additional fuel needs and payment of toll gate fee at Chokwe. Further, the interruption of operation of the crane could put at risk the operation of the dam should occur power outages at times of large floods. The Bank financed a total investment of US$11 million which was used by the National Directorate for Water Resources Management (DNGRH) a construction of new dissipation system and a reno mattress across the 650 meters of dam length in replacement of the damaged concrete block carpet in the downstream terrace; installation of a protection wall in the right bank and new road access protection sheet pile; construction of a new office building in a safer high-ground location 2.5 km away from the floodplain; and construction of a safer parking bay for the gantry crane out of the road platform. As a result of the works supported by the project, the lifetime of the dam was increased by 70-100 years; the risk of traffic accidents with the crane was eliminated and the second layer of safety and adequate manual operation of the dam in case of power outages during flood peak period was restored. The Limpopo Basin Management Division (DGBL) was relocated to the new safer offices. This measure has provided physical protection to staff of UGBL and critical information and equipment from flooding and built a third layer of security for continued operation of Macarretane dam during the occurrence of catastrophic floods. The rehabilitation works have improved flood management directly benefiting the populations in the towns and villages in the districts of Chokwe, Guija, Chibuto, Xai-xai and Limpopo. It has also increased water security during drought events for agricultural production in the irrigation schemes downstream of the dam, including Chokwe (33,000 hectares) and Baixo Limpopo (11,000 hectares) directly benefiting 21,000 farmers. F. Summary of desalination water systems in Gaza Province Between 2014 and 2016, the province of Gaza was severely affected by the worst drought in the country in 30 years. Induced by the El Nino phenomenon, the lack of rain dried up the groundwater tables and led to increased water scarcity in the driest areas of southern and central Mozambique. Data from the Ministry of Agriculture and Food Security indicated that by April 2016, 875,818 ha of crops had been lost (18.2 percent of total cultivated area in the country), 464,879 farmers affected, 533,692 cattle and 55,193 goats put at risk and 5,708 cattle had died, mostly in Maputo and Gaza provinces. When the IRM was activated in 2016, the ERRP invested US$0.6 million to finance the construction of two pilot desalination systems in two communities in the districts of Chigubo and Chicualacuala to provide water for 1,800 people as well as cattle. The construction works were completed in six months and the systems were handed over to local communities and authorities in December 2018. An independent evaluation of IRM implementation in 2020 noted that the two systems were functioning perfectly and were being well managed by a private operator. Beneficiaries interviewed during the evaluation showed high levels of satisfaction (95 percent) with the water access, quality and service provided. However, others (5 percent) raised concerns about the productivity of the systems and the fact that water distribution ended every day at 3 pm. As an additional benefit, the desalinization systems put an end to disputes between humans and livestock over access to productive water sources. In 2017, the AF of ERRP invested another US$1.3 million to construct desalination systems in four additional communities in the districts of Massagena, Chokwe and Massingir, to provide water for 10,000 people. These systems were completed and handed over in July 2021. Testimonies at a showcase 30 conducted by the World 30A short video and report were prepared in 2021 by the World Bank with financial support from GFDRR covering the communities: Bombofo, in Chokwe district; Macavene-Banga, in Massingir district; and Chinhanguane, in Massangena district. Page 60 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Bank in Gaza province in November 2021 revealed high levels of satisfaction with the new systems. Beneficiaries said that having water sources closer to their homes had reduced the distance they had to walk to fetch water; that the systems provide high-quality water, similar to bottled mineral water; that water access has enabled them to introduce horticulture production and has facilitated the emergence of the local construction industry; that it has mitigated the incidence of water borne diseases; and that no more cattle were dying due to water scarcity. Distant communities in nearby districts were also sharing the benefits from these water systems. Beneficiaries also showed their willingness to pay for water service, as demonstrated by the increased demand for domestic water connections and by community commitment to pay the fixed daily and monthly water fee. Further, community leaders said that access to water had halted the migration of the local population to other communities with better water access. These leaders had been empowered to mediate the relationship between community members and the private operators charged with the daily operation of constructed desalination systems. The active participation and commitment of beneficiaries and community leaders in using and paying for the water service, and the expansion of benefits to include distant communities, has been the core of the sustainability model adopted by the National Directorate for Water Supply and Sanitation (DNAAS) to ensure the continued cost-effectiveness of the operation, and the management and maintenance of desalination systems by private operators under concession and exploration contracts. G. Summary of showcase on resilient schools In 2015, tropical storms and floods destroyed 2,362 conventional classrooms in the provinces of Niassa, Nampula and Zambezia. The damage revealed the high level of vulnerability of these classrooms to cyclones, strong winds and flooding due to location of schools in hazardous sites, lack of adequate building designs and standards, and poor construction materials and techniques. By December 2018, 217 conventional classrooms in the three provinces were retrofitted under the project to be able to withstand wind speeds of up to 155 km/hour in Niassa province, and up to 207 km/hour in the coastal areas of Zambezia and Nampula provinces. On March 5, 2019, Cyclone Idai made its first landfall as a tropical depression in Zambezia, with wind speed up to 50 km/hour, causing flooding and heavy rain. The winds of the second landfall of Cyclone Idai on March 13, 2019, hit Zambezia province with speeds between 60 to 90 km/ hour. In Zambezia province, most public buildings, schools, and churches – traditionally used by the national disaster management authorities as shelters – were severely damaged or totally destroyed. However, the Primary School of Licuar in the district of Nicoadala, which was retrofitted under the project withstood the impacts of Cyclone Idai with no damage, demonstrating the effectiveness of the new resilient construction standards. The local population used this school as shelter during and after the storm. In 2020, a new storm was recorded in the district of Muecate in Nampula province. This storm destroyed the roofing of classrooms of the Primary School of First and Second degrees of Muecate. However, the classrooms that were retrofitted under the ERRP Project remained intact, again clearly demonstrating the benefits of resilient rehabilitation achieved under the project. In the testimonies collected in 2021 for the showcase of UN-Habitat, local community members who witnessed the resilience the primary school of Muecate believe that all retrofitted classrooms will be able to withstand a storm of any intensity and still stand firm for 350 years. There were similar testimonies from community members in a showcase 31 conducted by the World Bank in 2022 in the provinces of Nampula and 31 A short video and written report are under preparation by the World Bank with financial support from GFDRR to covering two primary schools located in the district of Nicoadala, in Zambezia province and Muecate district in Nampula province. Both schools were subject to strong winds between 2019 and 2020 after completion of rehabilitation works under the ERRP Project. Page 61 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) Zambezia. At the provincial level, all local governments are clamoring for the scale-up of the retrofitting program to bring local schools up to resilience standards. During a recent (March 2022) visit to Mozambique by the World Bank’s Regional Safeguards Advisor (RSA) for Eastern and Southern Africa, the Minister of Education and Human Development commended the WB for supporting the retrofitting of classrooms under the ERRP. The minister noted that the retrofitted schools were the only structures that remained intact during the storms that had hit the central and northern regions of Mozambique since 2019. These results show that retrofitting is the best way to reduce the vulnerability of schools exposed to cyclones and strong winds. The current retrofitting of 3,000 vulnerable classrooms across the country under the Mozambique Disaster Risk Management and Resilience Program will help address some of the existing needs in the education sector while consolidating the results achieved under the ERRP project. H. Environmental and Social compliance The project initially faced challenges with safeguards compliance, but gradually met the requirements of Bank’s safeguard policies and kept a Moderately Satisfactory rating until closing. The implementation of the Safeguards Policies included preparation of an Environmental and Social Management Framework (ESMF), a Resettlement Policy Framework (RPF) and an Integrated Pest Management Plan (IPMP). Each construction activity was preceded by the filing of an environmental and social screening form and followed, as appropriate, by the preparation of a site-specific Environmental and Social Impact Assessment (ESIA); Environmental and Social Management Plan (ESMP) or Simplified Environmental and Social Impact Assessment (S-ESIA); Environmental and Social Management Best Practices Manual (ESMBPM); and Resettlement Action Plan (RAP) or Abbreviated Resettlement Action Plan (A-RAP). At appraisal, the assessment concluded that the project’s overall environmental and social impacts would not be significant, as most would be site specific, temporary, reversible, and easily manageable. The project was classified as Category B, triggering the following safeguards: Environmental Assessment (OP/BP4.01), Natural Habitats (OP/BP4.04), Pest Management (OP 4.09), Physical Cultural Resources (OP/BP 4.11) and Involuntary Resettlement (OP/BP4.12). In 2017, the first AF, which included the rehabilitation of Macarretane dam in the Limpopo River Basin, triggered Safety of Dams (OP/BP 4.37) and Projects on International Waterways (OP/BP 7.50). Activities under components A and B had limited environmental and social impacts but required adequate supervision and close monitoring during implementation. All rehabilitation of existing rural infrastructure, dikes, water supply systems, conventional classrooms, hydrometeorological stations, and the Macarretane dam, as well as the construction of non-conventional classrooms and the offices of the DGBL, were expected to generate limited impacts, as these works were implemented at existing sites. In particular, the rehabilitation of the intake network of the Mocuba water supply system, and rehabilitation of Macarretane dam were expected to affect some already modified habitats along the rivers banks, while the rehabilitation of irrigation schemes had the potential to indirectly trigger the resumption of pesticide use for agriculture production. Therefore, the preparation of specific ESIAs/ESMPs/RAPs was a precondition for commencements of works under Component A. All works fully complied with this requirement, and the ESIAs/S-ESIAs and corresponding ESMPs/ESMBPMs/ RAP were prepared, reviewed and cleared by the Bank and included in the respective tender documents. Notably, the timely preparation of safeguards instruments was a serious Page 62 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) challenge and caused some delays in commencement of civil works, particularly in lot 1 of Nante dike 32. Regular World Bank implementation support missions ensured compliance of all activities with these instruments. Regular reporting on safeguards compliance was challenging in the initial years, but issues were addressed with the approval of the E&S Safeguards Action Plan by the World Bank in 2018. The Action Plan set forth sequential procedures for preparation and implementation of relevant safeguard instruments, which particularly mitigated the noncompliance noted at INIR, related to the closing of borrow pits during rehabilitation of the Intabo irrigation scheme. Challenges were also noted for regular reporting on functioning of the Grievance Redress Mechanism (GRM) established at each PIU. The MTR recommended that monitoring of environmental and social safeguards be strengthened by requiring each PIU to submit regular reports on safeguards compliance in all ongoing works. The project was successful in handling Occupational Health and Safety (OHS) and Gender Based Violence (GBV) and Sexual Exploitation and Abuse (SEA) cases. All four OHS incidents 33 and one GBV/SEA case 34 that occurred under the project were reported within 48 hours and were effectively investigated, managed and adequately resolved by the respective PIUs. An Environmental, Social and Labor audit was conducted in 2020 to verify the circumstances of all reported incidents. The audit confirmed that the E&S management procedures at all PIUs, contractors’ and supervision firms were satisfactory. OP 4.12 on Involuntary Resettlement was triggered at appraisal due to the potential adverse impacts of rehabilitation works under Component A. At appraisal, civil works associated with the rehabilitation of dikes, resilient schools, irrigation schemes and water supply systems were expected to potentially cause temporary or permanent physical and/or economic displacement and involuntary resettlement and require land acquisition for temporary or permanent settlement of those displaced. However, only the rehabilitation of the Mocuba water supply system required permanent land acquisition. During implementation, RAPs were required for rehabilitation of the Mocuba system and for lot I of the Nante dike. The RAP for the Mocuba system was prepared, reviewed and implemented; while the RAP for the Nante dike was prepared, reviewed but never implemented, as the Government canceled the rehabilitation of 5 km of dikes at Mukera. Instead, the vulnerable section of the dike was strengthened with the construction of a new hydraulic structure equipped with an outlet to regulate water flows from and to the Licungo River Basin. This decision raised concerns about adequate management of expectations, as the local population had been consulted during RAP preparation. DNGRH conducted meetings with local authorities and communities at Nante to explain the reasons for cancellation of resettlement as induced by changes in technical design of the dike while maintaining the expected safety to local populations. For the Mocuba water system, the Government 32 The contract for rehabilitation of lot 1 of the Nante dike was signed in September 2016 but civil works only started in July 2019, after the rehabilitation of 5 km of the Nante dike was canceled because it would have caused displacement of the population at Mukera. 33 These included: (a) collapse of a bridge at the Administrative Post of Nante due to excessive weight of a crane brought in by the contractor in charge of repair of lot 2 of the Nante dike. The crane fell into the river, the driver was wounded but was diligently assisted at a local health unit; (b) an accident at Namacurra between a pedestrian and a vehicle belonging to the contractor in charge for rehabilitation of the Mocuba water supply system; the pedestrian died at the site; (c) physical aggression by the supervising engineer against his co-worker on site during the rehabilitation of the Mocuba water supply system; (d) a car accident at Chokwe; the director of rehabilitation of the Macarretane dam died on site and the driver received treatment at the local hospital. 34 This was a case of suspected pregnancy involving a foreign male supervisor engineer and a national female co-worker, both involved in the rehabilitation of Mocuba water supply system. Later, the female worker reported a natural miscarriage, however, a compensation was paid by the supervisor engineer and the case closed Page 63 of 64 The World Bank MZ - Emergency Resilient Recovery Project (P156559) successfully implemented the permanent resettlement of 4 project-affected people (PAPs) and compensated 16 others, in accordance with World Bank safeguards policies and national regulations. Page 64 of 64