The World Bank Group in Mozambique, Fiscal Years 2008–21 Country Program Evaluation © 2023 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington, DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org ATTRIBUTION Please cite the report as: World Bank. 2023. The World Bank Group in Mozambique, Fiscal Years 2008–21. Country Program Evaluation. Independent Evaluation Group. Washington, DC: World Bank. COVER PHOTO shutterstock / Yury Birukov EDITING AND PRODUCTION Amanda O’Brien GRAPHIC DESIGN Luísa Ulhoa This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The bound- aries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. RIGHTS AND PERMISSIONS The material in this work is subject to copyright. Because The World Bank encourages dissem- ination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. The World Bank Group in Mozambique, Fiscal Years 2008–21 Country Program Evaluation March 20, 2023 Contents Abbreviations v Acknowledgments vi Key Findings and Lessons vii Introduction and Country Context����������������������������������������������������������������������������1 1.  Country Context 2 Main Development Constraints 7 Evolution of World Bank Group Strategies and Operations in Mozambique�������13 2.  International Finance Corporation Strategic Engagement in Mozambique 19 World Bank Lending Portfolio 20 Support for Agricultural Productivity���������������������������������������������������������������������23 3.  Analytical Underpinnings 24 Relevance of World Bank Support 26 World Bank Effectiveness 31 Support for Access to Basic Services��������������������������������������������������������������������36 4.  Support for Addressing Weak Governance����������������������������������������������������������49 5.  Integrated Financial Management Information System 51 Budget Credibility 52 Internal and External Audit Function 53 Public Investment Management 54 Public Debt Management 57 State-Owned Enterprises Reform 60 Decentralization 62 Extractives 66 6. Resilience to Climate Change�������������������������������������������������������������������������������69 Institutional Framework 70 Building Climate Resilience in Sectors 73 Moving beyond Conventional Approaches 73 ii Building Sustainable Institutional Climate-Resilient Capacity 77 Working with Development Partners to Adopt Resilient Standards and Design 80 7. Findings and Lessons��������������������������������������������������������������������������������������������84 Bibliography���������������������������������������������������������������������������������������������������������������89 Boxes Box 1.1. Hidden Debt Crisis 4 Box 5.1. 2015 Public Investment Management Assessment Results 55 Box 6.1. Early Lessons from Nature-Based Flood Protection in Beira 75 Disbursement-Linked Indicators Included in the Disaster Risk Management Box 6.2.  and Resilience Program 79 Figures Figure 1.1. Economic Growth and Poverty Reduction in Mozambique 2 Figure 1.2. Fragility Indicators in Mozambique 5 Figure 1.3. Ratings of Governance Quality 6 Trend in Maize Yields in Mozambique Compared Figure 1.4.  with Malawi and Tanzania 7 Inequality of Income and Access to Basic Services across Figure 1.5.  Districts in Mozambique 9 Mapping of World Bank Group’s Support Areas to Mozambique’s 2016 Figure 2.1.  Systematic Country Diagnostic Priorities, Fiscal Years 2008–21 16 Investment Project Financing Commitments to Mozambique by Sector, Figure 2.2.  Fiscal Years 2008–21 21 Figure 2.3. Overall Outcome Ratings, Fiscal Years 2008–21 22 World Bank Support to Increasing Agricultural Productivity Figure 3.1.  in Mozambique, Fiscal Years 2008–21 28 Maize Productivity in Nampula, Zambezia, and Mozambique, 2000–1732 Figure 3.2.  Distribution of Project Locations by Provincial Gross Domestic Product Figure 4.1.  per Capita 40 iii Correlation between Levels of World Bank Group Support Figure 4.2.  and Need, by Strategy Period 41 Figure 4.3. Targeting Trend by Population Share, by Strategy Period 41 Figure 4.4. Sector Targeting across Strategic Period 42 Figure 4.5. Share of the Rural Population with Access to Basic Services 46 (Gini-Style) Inequality Coefficients for Each Service Figure 4.6.  across Strategy Periods 47 Tables Table 4.1. World Bank Portfolio in Basic Services, Fiscal Years 2008–21 38 World Bank Commitments in Basic Services, Fiscal Years 2008–21 Table 4.2.  39 Table 4.3. Project Ratings across the Evaluation Period 45 Table 5.1. World Bank Debt Management Engagements in Mozambique 59 Appendixes Appendix A. World Bank Group Strategies 100 Appendix B. World Bank Group Portfolio in Mozambique, Fiscal Years 2008–21 105 Appendix C. Methodology 110 Appendix D. Geospatial Analysis 122 Appendix E. Debt Management Performance Assessment Scores 130 Appendix F. World Bank Support for the Extractives Sector 132 Appendix G. World Bank Support for Climate Resilience 135 iv Abbreviations CPF Country Partnership Framework CPS Country Partnership Strategy DeMPA Debt Management Performance Assessment DMF disaster management fund DNGRH National Directorate of Water Resources Management DPO development policy operation EITI Extractive Industries Transparency Initiative e-SISTAFE Electronic State Financial Administration System Fundo de Apoio ao Sector de Educação (Education Sector Support FASE  Fund) FRELIMO Liberation Front of Mozambique FY fiscal year GDP gross domestic product IDA International Development Association IEG Independent Evaluation Group IFC International Finance Corporation IMF International Monetary Fund INAM Mozambique National Meteorology Institute PEFA Public Expenditure and Financial Accountability PIM public investment management PRSC Poverty Reduction Support Credit Independent Evaluation Group World Bank Group    v RENAMO Mozambican National Resistance SCD Systematic Country Diagnostic SECF small emerging commercial farmers SOE state-owned enterprise All dollar amounts are US dollars unless otherwise indicated. Acknowledgments The report was prepared by an Independent Evaluation Group team led by Eduardo Fernandez Maldonado (evaluation officer) and Giuseppe Iarossi (senior economist) under the supervision of Jeffrey Allen Chelsky (manag- er) and Oscar Calvo-Gonzalez (director) and the overall guidance of Alison Evans (Director-General Evaluation and vice president). The evaluation team included the following Independent Evaluation Group staff and consultants: Harsh Anuj (data scientist), Marisa Godinho Balas (consultant), Yumeka Hirano (economist), Clara Ivanescu (geographer), Malathi Jayawickrama (senior public sector specialist), Basil Kavalsky (con- sultant), Sengphet Lattanavong (extended-term consultant), Monica Luongo (consultant), Andres Liebenthal (consultant), Johan Lopez (consultant), J. W. van Holst Pellekaan (consultant), Mees Daniel van der Werf (economist), and Maria Virginia Ziulu (data scientist). Maximillian Ashwill provided editorial The World Bank Group in Mozambique, Fiscal Years 2008–21 Acknowledgments support. Patricia Acevedo, Carla Fabiola Coles, and Dung Thi Kim Chu pro- vided administrative support. The evaluation team is thankful to the following peer reviewers: Colin Bruce, former World Bank country and regional director; Sandra Sequeira, associate professor in development economics at the London School of Economics; and Esther Palacio, technical assistance coordinator and public financial management adviser at the International Monetary Fund. The Independent Evaluation Group is grateful to the numerous representa- tives of the government of Mozambique, private sector entities, internation- al and bilateral development partners, and nongovernmental organizations and academic institutions who provided valuable insights. The team benefit- ed from comments by and is grateful to Finn Tarp, a professor at the Univer- sity of Copenhagen. The team is thankful to World Bank Group management and staff who provided valuable information and feedback. vi Key Findings and Lessons This Country Program Evaluation reviews World Bank Group engagement in and support to Mozambique during fiscal years (FY)08–FY21. It assesses the extent to which the Bank Group’s support was relevant to the country’s main development challenges and drivers of fragility and the evolution and adapta- tion of support over time. The evaluation delves into four themes of particular relevance to Mozambique’s pursuit of the Bank Group’s twin goals of poverty reduction and shared prosperity: (i) low agricultural productivity; (ii) unequal access to basic services; (iii) weak institutions and governance; and (iv) vulner- ability to climate change and natural disasters. The evaluation presents findings related to each of the four themes and distills lessons from the Bank Group’s experience in Mozambique to inform future strategies and engagements. Average overall development outcome ratings for World Bank–supported projects that closed during the evaluation period were moderately satisfac- tory. Just more than half of projects and operations were rated moderately satisfactory, and one in five were rated satisfactory, which was similar to the Eastern and Southern Africa Region. The Independent Evaluation Group rat- ed the development outcome of the FY08–11 World Bank–supported Coun- try Partnership Strategy for Mozambique as moderately unsatisfactory and the Country Partnership Strategy FY12–16 as moderately satisfactory. Mozambique’s deteriorating fragility over the evaluation period was increas- Independent Evaluation Group World Bank Group    vii ingly acknowledged in World Bank–supported strategies but was reflected in operations with a lag. World Bank operations downplayed the country’s fragility until the outbreak of conflict in the north in 2019–20. Between FY08 and FY21, the World Bank described the prevailing atmosphere optimistical- ly, fueled by strong economic growth and the discovery of large gas reserves. At the same time, resource management, accountability, and access to basic services by poor people were deteriorating, and decentralization had stalled. With regard to the first theme of the evaluation, low agricultural productivity, World Bank assistance to increase agricultural productivity in Mozambique focused on the critical areas identified in analytical work produced both internally and by academics and practitioners. With most of the poor people living in rural areas, increasing agricultural productivity was appropriately seen as key for reducing rural poverty. Six areas were identified and supported by the World Bank during the evaluation period: (i) use of modern technologies; (ii) rural infrastructure; (iii) access to markets; (iv) natural resource management; (v) land policies and administration; and (vi) gender disparities. The World Bank’s portfolio focused initially on technology and irrigation, but results were inadequate. In the second half of the review period, support expanded to include market access, forestry, land administration, and rural roads, but it is too early to assess the effectiveness of that support. With regard to the second theme, unequal access to basic services, World Bank–supported projects helped improve access to basic services, such as education, health, transport, and electricity in rural areas. There were pos- itive outcomes in the delivery of education, health, transport, and (in part) electricity services. The latest household survey (2019–20) also shows im- The World Bank Group in Mozambique, Fiscal Years 2008–21  Key Findings and Lessons provements in access to these services among rural populations during the evaluation period. However, it was only in the latter half of the evaluation period that the World Bank began to directly target the poorest areas. With regard to the third theme of weak institutions and governance, World Bank support to Mozambique focused on five areas: (i) public financial management (central government); (ii) public debt management; (iii) state- owned enterprises (SOEs) reform; (iv) decentralization; and (v) transparent and effective management of extractives. » The World Bank contributed to improved public financial management by supporting an increase in the coverage of the financial management information systems and strengthening internal and external control functions at the central level. However, World Bank support for budget preparation and execution did not enhance budget credibility. Despite clear weaknesses in public investment management, it was only in the wake of the hidden debt crisis in 2016 that the World Bank made concerted efforts to intensify support. Despite progress “on paper,” institutionalization of public investment management reforms is lagging. » The World Bank had a modest impact on improving debt management viii and advancing SOE reform. Support was focused on building technical and institutional capacity but did not adequately take into account the context of weak governance. By and large, SOE and debt management challenges were seen as problems that could be addressed through technical and institution- al capacity building. Although this may have been a necessary condition to improve outcomes, underlying governance shortcomings also needed to be addressed. On balance, once the hidden debts were revealed, tangible prog- ress was made as the appetite for increased control of corruption increased and a compelling case for SOE reform and debt management was made. » The World Bank contributed to increased subnational capacity but was not effective at supporting the establishment of a coherent decentraliza- tion policy framework. Political economy constraints rendered World Bank support for decentralization ineffective. Implementation of public financial management reforms at the subnational level faced significant challenges, but many of these were addressed successfully using Program-for-Results financing. World Bank–supported projects contributed to tangible improve- ments in municipal revenue collection. » World Bank support helped improve governance in the extractives sector, but major challenges remain. The World Bank contributed to the estab- lishment of a regulatory framework for managing the extractives sector and complying with transparency standards. However, World Bank support for the implementation of a fiscal rule and sovereign wealth fund for managing revenues from the extractives sector did not lead to tangible outcomes. With regard to the fourth theme, vulnerability to climate change and natural Independent Evaluation Group World Bank Group    ix disasters, the World Bank contributed to the development of an institutional framework for strengthening climate resilience and improving disaster risk preparedness through strengthened hydrological and meteorological infor- mation services and increased financial protection against disasters. World Bank support contributed to increased climate resilience in the transport, social protection, water and sanitation, agriculture, education, energy, and urban sectors. This evaluation identifies the following lessons to guide future World Bank engagement in Mozambique; these lessons may also interest other countries facing similar development challenges. 1. In contexts characterized by corruption and state institutions being run for the benefit of high-status groups, technical solutions to public financial and debt management are unlikely to achieve desired results unless key underlying governance constraints (risks) are also confronted. In Mozambique, this was the case with SOE reform and debt management, where World Bank support focused mainly on technical and institutional capacity and was not sufficiently adapted to the underlying political economy and associated risks. Likewise, World Bank support for public investment management was largely technical, with less attention given to the implementation of risk-based approaches to identify and analyze corruption risks throughout the investment cycle. Although progress was achieved “on paper,” implementation of reforms often fell short in practice. 2. Core diagnostics are essential to inform reform priorities but require deliberate and coordinated operational follow-up. Although the World The World Bank Group in Mozambique, Fiscal Years 2008–21  Key Findings and Lessons Bank undertook several public financial and debt management diagnos- tics, it did not use the findings in a timely manner to set reform priorities and inform its work program. This was most noteworthy with respect to the 2008 Debt Management Performance Assessment findings, which flagged serious shortcomings in debt reporting and recording. However, little attention was given to address these shortcomings until the hidden debt crisis, including through prior actions in the subsequent program- matic series of development policy operations. Mozambique would also have benefited from an early and more systematic assessment of weak- nesses in public investment management, which—alongside the Debt Management Performance Assessment—could have identified some of the weaknesses that contributed to the hidden debt crisis. 3. The quality and impact of World Bank support for public financial and debt management can be enhanced by improving internal World Bank coordination and prioritization. This lesson is aligned with findings from the recent Independent Evaluation Group evaluation World Bank Support for Public Financial and Debt Management in IDA-Eligible Countries (World Bank 2021p), which found that synergies among different public financial and debt management pillars remain underexploited in many International Development Association countries. In the x case of Mozambique, the World Bank provided significant support for upstream aspects of debt management (for example, preparation of debt management strategies), with only late attention for downstream aspects (debt reporting and recording, cost and risk analysis, and debt processes and procedures). Moreover, for most of the evaluation period, support for debt management was not systematically accompanied by efforts to improve public investment management, despite widely recognized synergies among borrowing and the quality of public investment. As a result, opportunities to enhance the growth and development impact of development spending and debt-financed public investment were missed. 4. The effectiveness of agricultural extension services in raising agricultural productivity requires paying greater attention to the adequacy of staffing. Extension services play a critical role in increasing agricultural productivity in Mozambique, yet such support risks being undermined by staffing shortfalls given the large proportion of smallholders. In Mozambique, extension services in the Sustenta project were spread thin, with each extension worker responsible for, on average, about 3,900 farmers, compared with 3,000 in Malawi and 1,170 in Tanzania. 5. In situations where women dominate a disadvantaged group, such as in subsistence farming, sector-based support (for example, to enhance agricultural productivity) requires gender considerations to be fully integrated into strategies and projects. In Mozambique, women are Independent Evaluation Group World Bank Group    xi particularly disadvantaged in benefiting from extension services, with fewer than one-third of women being reached by such services (USAID 2018). Support for agricultural productivity can be more effective if it puts gender front and center in the approach, including by collecting sex- disaggregated data. 6. Support for climate resilience can be enhanced through the use of credible analytics to persuade policy makers about the costs of inac- tion. Persuading policy makers to pursue climate resilience policies can be challenging because the costs of implementing such policies are real, but benefits are uncertain. Before 2010, most government efforts with respect to climate change were focused on response to and reconstruction after extreme weather events. World Bank analysis was crucial to making the financial and fiscal cases for investing in increased climate resilience by demonstrating the impact of extreme weather events, pricing adaptation needs, and convincing government authorities that ex post reconstruction was not cost-effective. The World Bank Group in Mozambique, Fiscal Years 2008–21  Key Findings and Lessons xii 1 | Introduction and Country Context This Country Program Evaluation assesses the development effective- ness of the World Bank Group’s support to Mozambique during fiscal years (FY)08–FY21. It assesses the extent to which the Bank Group’s support was relevant for addressing the country’s main development challenges and drivers of fragility and the evolution and adaptation of that support over time. The evaluation distills lessons from the Bank Group’s experience in Mozambique to inform future engagement. The evaluation delves into four key themes of relevance to Mozambique’s pur- suit of the Bank Group’s twin goals of poverty reduction and shared prosper- ity: (i) low agricultural productivity; (ii) unequal access to basic services; (iii) weak institutions and governance; and (iv) vulnerability to climate change and natural disasters. These four themes were identified in the Bank Group’s 2021 Systematic Country Diagnostic (SCD) as major constraints to Mozambique’s development (World Bank 2021j). This evaluation seeks to answer the follow- ing questions (see appendix C for the evaluation methodology): 1. To what extent did the Bank Group support improvements in agricultural productivity and access to basic services across regions to foster poverty reduction and shared prosperity in Mozambique? 2. To what extent did the Bank Group support improvements in governance in Mozambique? 3. How successful has the Bank Group been in helping Mozambique build resilience to climate change? Chapter 1 provides the country context in which the Bank Group provided its support. Chapter 2 describes the Bank Group’s strategy and its evolu- tion and implementation over the evaluation period. Chapter 3 assesses the Bank Group’s support to enhance agricultural productivity. Chapter 4 looks at Bank Group support to improving access to basic services, and chapters 5 and 6, respectively, evaluate the Bank Group’s support for improving gov- 1 ernance and enhancing climate resilience. Chapter 7 presents findings and lessons to inform future Bank Group support to Mozambique. Country Context After the end of the civil war in 1992, Mozambique experienced strong econom- ic growth, which raised living standards and reduced poverty. Gross domestic product (GDP) expanded at an average annual rate of almost 8 percent between 1993 and 2013, making Mozambique one of the fastest-growing economies in Sub-Saharan Africa (figure 1.1, panel a). During that period, political and macro- economic stability provided the foundation for robust growth led by a rebound- ing agricultural sector and significant donor support. The economic expansion boosted incomes and living standards. GDP per capita grew at an annual aver- age of 4.8 percent. The poverty rate declined from 60.3 percent in 2002–03 to 48.4 percent in 2014–15 (figure 1.1, panel b). Figure 1.1. Economic Growth and Poverty Reduction in Mozambique a. GDP growth and GDP per capita, 1992–2018 b. Poverty rate 2002–03 through 2014–15 The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 1 600 15 80 70 500 GDP per capita (constant US$) 12 Annual GDP growth (%) 60 Poverty rate (%) 400 50 9 300 40 6 30 200 20 3 100 10 0 0 0 2002–03 2008–09 2014–15 2000 2006 2008 1998 2004 2002 1996 1992 2010 1994 2018 2016 2012 2014 Year Year GDP per capita (constant 2010 US$, left axis) GDP growth (annual %, right axis) Sources: a. World Development Indicators database; b. World Bank using Household Budget Survey (Inquérito ao Orçamento Familiar) 2002–03, 2008–09 and 2014–15. Latest available poverty data are from 2014–15. Note: GDP = gross domestic product. 2 Despite impressive growth, poverty remained high in rural areas, with the la- bor force concentrated in subsistence agriculture. Starting in 2008, drivers of growth gradually shifted away from agriculture to the services and industry sectors (World Bank 2018c, 22; IFC 2020). One implication of this transition was that growth became less beneficial to poor people. Between 2008 and 2014, per capita consumption grew on average by 7 percent annually for the top 20 percent of households but by only 2.6 percent for the bottom 40 per- cent. Almost half of the population (46.3 percent) continued to live in pover- ty, with most of them (84.9 percent) living in rural areas (World Bank 2018d, vi). Smallholder farming, defined as cultivating an average of 1.1 hectares in 2020, was the primary source of income for 9 out of 10 rural households (World Bank 2020a). Growth decelerated beginning in 2016 in the face of low commodity prices and a regional drought, and it was further eroded by the hidden debt crisis, natural disasters, and the COVID-19 pandemic. The discovery of large hidden debts (box 1.1) triggered a sharp currency depreciation and reduced finan- cial inflows from investors and donors. The crisis also marked the end of an extensive public investment boom that had helped fuel growth (World Bank 2021j). In 2019, Mozambique suffered a series of tropical cyclones, which also negatively impacted the economy. The next year was the start of the COVID-19 pandemic. As a result of these events, growth decelerated sharply from 2016 onward and entered negative territory in 2020 for the first time in almost three decades. Independent Evaluation Group World Bank Group    3 Although not always formally classified as such, Mozambique is a fragile country. For most of the evaluation period, the country was not classified as fragile, despite being characterized by deep internal grievances, high levels of economic exclusion, and limited provision of basic services to the popu- lation. In FY18, Mozambique was formally classified as fragile. Fragility in Mozambique manifested itself in four arenas: (i) access to political power and economic opportunities; (ii) access to natural resources, including ex- tractives; (iii) access to basic services; and (iv) access to justice and security (World Bank 2020g). Box 1.1. Hidden Debt Crisis In 2013, EMATUM, a state-owned fishing company, borrowed $850 million (equivalent to approximately 6 percent of the gross domestic product) from Credit Suisse and VTB Bank to finance a tuna fishing fleet. With the help of these and other banks, EMATUM securitized the loans, slicing them into smaller chunks and issuing them as unlist- ed securities with a sovereign guarantee (“tuna bonds”). In 2014 and 2015, EMATUM reported losses, implying that the government would need to assume the obligations. However, by 2016, Mozambique public finances were worsening rapidly because of falling commodity prices and insufficient agricultural harvests, and it was soon clear that the government was unable to assume the debt. Against this backdrop, the government restructured the tuna bonds into a sovereign bond in April 2016. At the same time, it was discovered that between 2009 and 2014, Mozambique had contracted the equivalent of 10 percent of the gross domestic product ($1.4 billion) in nonconcessional debt by issuing guarantees to state-controlled companies and borrowing directly from bilateral lenders. The discovery of this previously undisclosed debt (which increased Mozambique’s external debt burden to 127 percent of the gross domestic product) led the International Monetary Fund, the The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 1 World Bank, and bilateral donors to withhold financial support to Mozambique, a move that contributed to a budget crisis and an overall deterioration of the macroeconomic framework. Source: Independent Evaluation Group, based on World Bank 2017b. Violent conflict has characterized the political landscape. The end of hos- tilities between the Liberation Front of Mozambique (FRELIMO) and the Mozambican National Resistance (RENAMO) in 1992 began two decades of relative peace, which ended in 2013 when RENAMO took up arms again. After a series of temporary cease-fires, a new peace agreement was reached in 2019; soon after, however, a faction of RENAMO resumed attacks over al- leged electoral irregularities. Since October 2017, members of a religious ex- tremist group have carried out attacks in the coastal districts in the province of Cabo Delgado, displacing roughly 700,000 people.1 Violence has caused severe delays in gas production in Cabo Delgado, diminishing the prospects of gas as a catalyst for economic growth and increased government revenues. 4 The Bank Group classified Mozambique as a medium-intensity, conflict- affected country in FY21. Deaths due to conflict increased throughout the evaluation period (figure 1.2, panel a), especially after the start of the insurgency in Cabo Delgado in 2017. The Fragile States Index—which measures fragility by tracking various social, economic, political, and cohesion indicators and compares them with the state’s ability to manage those pressures—showed that Mozambique’s fragility worsened in 11 out of the 14 years covered by the index (figure 1.2, panel b). Figure 1.2. Fragility Indicators in Mozambique a. Conflict fatalities 250 200 Conflict fatalities (no.) 150 100 50 0 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Year UCDP ACLED b. Fragile States Index Independent Evaluation Group World Bank Group    5 90 Total country score 85 80 75 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 Year Source: Armed Conflict Location and Event Data Project (https://acleddata.com/) and Uppsala Conflict Data Program (https://ucdp.uu.se/). Note: ACLED = Armed Conflict Location and Event Data Project; UCDP = Uppsala Conflict Data Program. The quality of governance and state institutions has worsened over time as the debt crisis unfolded and conflict intensified. The resumption of hostil- ities between FRELIMO and RENAMO, the hidden debt crisis in 2016, and the insurgency in Cabo Delgado contributed to a sharp deterioration in governance. Scores for all World Bank Country Policy and Institutional As- sessment indicator clusters trended down during the evaluation period (fig- ure 1.3, panel a). Worldwide Governance Indicators for Mozambique declined in the areas of control of corruption, government effectiveness, political stability and absence of violence, regulatory quality, rule of law, and voice and accountability (figure 1.3, panel b).2 Mozambique is now at the bottom of these indicators compared with regional peers in Sub-Saharan Africa. Figure 1.3. Ratings of Governance Quality a. CPIA scores 5 4 CPIA score 3 2 The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 1 1 0 Structural Public Sector Policies for Economic Overall Policies Management Social Inclusion Management CPIA score and Institutions and Equity Indicator 2008 2019 b. Worldwide Governance Indicators 60 50 Percentile rank (%) 40 30 20 10 0 Control of Government Political Stability Regulatory Rule of Voice and Corruption Effectiveness and Absence Quality Law Accountability of Violence score Dimension 2008 2019 Source: World Development Indicators and Worldwide Governance Indicators database. 6 Note: CPIA = Country Policy and Institutional Assessment. Main Development Constraints During the evaluation period, Mozambique faced four major development constraints and drivers of fragility and conflict. Based on interviews with rel- evant Bank Group staff and reviews of World Bank–supported strategies and analytical work, this Country Program Evaluation has identified four main development constraints over the evaluation period, all of which are also considered drivers of fragility and conflict: (i) low agricultural productivity; (ii) unequal access to basic services; (iii) weak institutions and governance; and (iv) vulnerability to climate change and natural disasters.3 Low Agricultural Productivity  rend in Maize Yields in Mozambique Compared with Malawi Figure 1.4. T and Tanzania 3,5 3,0 2,5 Maize (mt/ha) 2,0 1,5 1,0 0,5 0,0 Independent Evaluation Group World Bank Group    7 2009 2015 2008 1999 2014 2018 2002 2013 2005 2012 1998 2001 2010 2000 2007 2016 2006 2011 2003 2004 2017 Year Tanzania Malawi Mozambique Source: FAOSTAT database, available at https://www.fao.org/faostat/en/#home. Note: mt/ha = metric tons per hectare. Agriculture in Mozambique is characterized by low and stagnant productivity and a declining contribution to national wealth. In 2008, agriculture, forestry, and fisheries contributed about 30 percent to the GDP; by 2020, this contribution had declined to about 20 percent. Yields of the dominant maize crop were chronically low. Between 1998 and 2019, the yields were 52 and 62 percent of the average maize yields in Malawi and Tanzania, respectively (figure 1.4); crop-growing conditions in those countries are comparable to those in Mozambique. Long-term yields for rice and cotton show similar trends and shortfalls compared with Malawi and Tanzania. Low agricultural productivity is a key driver of fragility in rural areas and a major driver of growing income inequalities between rural and urban popu- lations. Between the early 2000s and 2014–15, the Gini coefficient increased from 0.47 to 0.56, making Mozambique one of the most unequal countries in the region. Growing inequality has heightened social tensions and fueled renewed violent conflict in recent years (World Bank 2020g). Unequal Access to Basic Services Inadequate access to basic services in remote rural areas undermines social cohesion and prevents shared prosperity (figure 1.5). The percentage of ur- ban households with access to safe water and sanitation is 89.4 percent and 69.4 percent, respectively. The corresponding rates for rural households are 46.6 percent and 7.5 percent (World Bank 2018d, 18). Educational attainment and outcomes in the center and the north of the country are worse than in the south, with higher rates of absenteeism for both teachers and students, The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 1 lower levels of numeracy and literacy, and higher rates of dropouts (World Bank 2020g, 26). Southern and urban households also enjoy better health care coverage and health outcomes than those in the center and the north. Infant and child mortality rates are highest in northern Mozambique and in rural areas nationwide, and the highest incidence of stunted growth of children is in three northern provinces (Cabo Delgado, Nampula, and Niassa; World Bank 2016c, 90). 8 nequality of Income and Access to Basic Services across Districts in Mozambique Figure 1.5. I Source: World Bank 2020k; World Bank estimates using Inquérito sobre Orçamento Familiar (IOF) for 2014–15 Note: The Basic Infrastructure Access Index is an aggregate ranking of an average household’s access to transport, distance to market, distance to primary schools, distance to a clinic, and access to electricity and clean water. Independent Evaluation Group World Bank Group    9 Weak Governance Weak decentralization limits citizens’ voice and participation and contributes to the uneven presence of the state across the country. The transfer of power and resources from the central to subnational levels is limited and skewed in Mozambique, resulting in weak subnational capacity to perform government functions and disinterest in revenue collection (World Bank 2020g). Public financial and debt management have fallen short of what is needed to promote the effective and efficient use of public resources. The hidden debt crisis brought to the surface concrete challenges concerning public financial and debt management and insufficient oversight of state-owned enterprises (SOEs). The corrosive impact of corruption and ineffective public institutions has diminished trust in the state, had negative implications for the quality of most public services, and contributed to inadequate development outcomes. Grievances generated by these shortcomings have contributed to conflict and fragility. Governance of the extractives sectors is weak, undermining inclusion, transparency, and sustainable resource management. Mozambique has The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 1 struggled to have inclusive, transparent, and sustainable management of its extractives industry. The discovery of gas deposits and the potential for Mozambique to become a major exporter of liquefied natural gas have raised the cost of weak governance. The lack of effective governance arrangements and institutions to promote inclusive and transparent management of ex- tractives was a major driver of fragility and conflict. Vulnerability to Climate Change Mozambique’s vulnerability to climate change is high, and the severity, frequency, and cost of extreme weather events have increased. Mozambique faces multiple climate change–driven challenges, including rising sea levels, variable and uncertain rainfall, rising temperatures, and increased incidence of floods and droughts. The low-lying nature of the coastal zone has made Mozambique one of the most exposed countries to weather-related shocks in Sub-Saharan Africa. The 2021 Global Climate Risk Index ranked Mozambique first among countries affected by impacts of climate change, up from 58th 10 zel, and Shafer 2021).4 In 2019, Cyclones Idai and place in 2006 (Eckstein, Kun­ Kenneth struck the country consecutively and resulted in nearly 700 deaths and the displacement of hundreds of thousands, resulting in approximately $3 billion in damages and losses (Mozambique 2019). Extreme weather events have eroded Mozambique’s progress in pover- ty reduction (Mozambique 2019). The provinces that are most affected by disasters tend to have higher levels of poverty (World Bank 2020g). Floods, droughts, and cyclones led to crop failures that cut per capita food consump- tion of affected households by 25–30 percent and exacerbated underlying drivers of fragility such as inequality, institutional weakness, food insecurity, and mass displacement (Baez, Caruso, and Niu 2018; World Bank 2020g). Independent Evaluation Group World Bank Group    11 1  As per the latest estimates from the United Nations Office for the Coordination of Humani- tarian Affairs. 2  The Worldwide Governance Indicators (WGI) are a research data set summarizing the views on the quality of governance provided by a large number of enterprises, citizen and expert survey respondents in industrial and developing countries. These data are gathered from a number of survey institutes, think tanks, nongovernmental organizations, international organizations, and private sector firms. The WGI do not reflect the official views of the World Bank, its Executive Directors, or the countries they represent. The WGI are not used by the World Bank Group to allocate resources. 3  The 2021 update to the World Bank Group’s Systematic Country Diagnostic identified these as development constraints with a high potential for reducing fragility and conflict in the country (World Bank 2021j). 4  The Global Climate Risk Index is a composite of four indicators of human and economic loss from an insurance industry database: fatalities (both absolute and as a share of the popula- tion) and economic losses (both absolute and as a share of gross domestic product). The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 1 12 2 | Evolution of World Bank Group Strategies and Operations in Mozambique Highlights The overarching objective of the World Bank Group’s support to Mozambique during the evaluation period was sustained and inclusive economic growth. The three strategies covered in this evaluation pursued this goal through four focus areas: (i) gover- nance; (ii) human development and basic services; (iii) growth; and (iv) sustainable development and resilience. Bank Group support adapted to changing country conditions and shocks, such as the discovery of large gas deposits, the 2016 hidden debt crisis, and natural disasters. Bank Group–supported strategies recognized the drivers of fragil- ity, such as those related to resource management, accountability, decentralization, and access to basic services. But individual oper- ations initially downplayed fragility against a general atmosphere of optimism fueled mainly by decade-long growth and the discov- ery of significant gas reserves. During the evaluation period, the Bank Group committed $5.7 billion from the International Development Association and $1.3 billion from the International Finance Corporation. The country benefited from significant budget support, with most of the associated prior actions supporting reforms in public administration, the financial sector, and energy and extractives. Investment projects focused on water, sanitation, and waste management; education; agriculture; and public administration. Just more than half of World Bank projects and operations were rated moderately satisfactory, and one-fifth were rated satisfactory. 13  Over the evaluation period, Bank Group support was guided by three strategies: Country Partnership Strategy (CPS) FY08–11, CPS FY12–16, and Country Partnership Framework (CPF) FY17–21. All three strategies were in support of the overarching objective of making Mozambique’s growth more sustainable and inclusive. Appendix A (table A.1) presents the objectives of these strategies in detail. Bank Group support was consistent with the development priorities identi- fied in the 2016 SCD (figure 2.1). Throughout the evaluation period, the Bank Group pursued objectives in four focus areas: (i) governance; (ii) human development and basic services; (iii) growth; and (iv) sustainable develop- ment and resilience (appendix A, table A.2). The governance focus area was supported via interventions in public sector reform, decentralization, trans- parency and citizen engagement, public financial management, legal and judicial services, and economic management. In human development and basic services, the Bank Group supported interventions in health, education, water and sanitation, electricity, and social protection. To support growth, the Bank Group undertook interventions to improve the business environ- ment; increase access to finance; and develop enabling infrastructure for The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 2 private sector development, agriculture, and tourism. To promote sustain- able development and resilience, the Bank Group supported interventions to improve natural resource management, resilience and adaptive capacities to climate-related issues, and disaster risk management. Bank Group strategies adjusted to changes in country context and external shocks. Under the CPS FY08–11, the Bank Group doubled its original indic- ative lending volume to help the country cope with indirect fallout from the 2008 global financial crisis (World Bank 2007). In 2012, large gas deposits were discovered off the coast of Mozambique. To respond to this potential game changer in Mozambique’s development trajectory, the CPS FY12–15 placed an emphasis on improving the management, transparency, and oversight of natural resources, focusing on the burgeoning gas sector (World Bank 2012a). In 2016, the hidden debt crisis led many development partners (including the World Bank) to withdraw budget support, severely affecting Mozambique’s macroeconomic framework. Against this backdrop, the CPF FY17–21 emphasized support for macroeconomic stabilization and restoring 14 donor confidence (World Bank 2017b). Finally, Cyclone Idai in March 2019 and Cyclone Kenneth in April 2019 caused extensive human, physical, and economic losses. The CPF FY17–21 was adapted at the Performance and Learning Review stage to include an additional objective on recovery and rehabilitation (World Bank 2020a). The World Bank also gradually target- ed its support to lagging regions to reduce unequal access to services. The development constraints identified in the 2021 SCD update have expanded to cover virtually all fragility drivers identified by the 2020 Risk and Resil- ience Assessment, except for security and shortcomings in the justice sector (appendix A, figure A.1). Over time, Bank Group–supported strategies consistently and more explicit- ly addressed drivers of fragility and conflict. The CPS FY08–11 and the CPS FY12–16 included support to address weak resource management, lack of accountability, stalled decentralization, limited access to basic services, weak institutions, and other development constraints subsequently identified as drivers of fragility and conflict in the 2020 Risk and Resilience Assessment. The CPF FY17–21 identified regional disparities and disputes over natural resources as development constraints with a significant impact on conflict and fragility (World Bank 2017b). To address regional disparities, the CPF FY17–21 sought to deliver multisectoral support for regions with high pov- erty levels. To address challenges in natural resource management, the CPF FY17–21 included interventions to support land administration and user rights. In addition, to allow for the identification and diffusion of conflicts, it called for the inclusion of solid and transparent grievance redress mecha- nisms in Bank Group interventions. Independent Evaluation Group World Bank Group    15 Despite rising fragility, an atmosphere of optimism prevailed in the World Bank’s engagement in Mozambique. Since the end of the civil war, Mozambique has enjoyed strong growth and poverty reduction, making it a postconflict success story in the eyes of donors who were keen to provide support and financing. World Bank staff and development partners invariably described a “positive,” “optimistic,” and “forward-looking” atmosphere in which the country was seen as a “rising star” and “donor darling.” Although the World Bank was preparing for a postconflict era, this understanding led to an implicit assessment that conflict was “not a big problem” and “something of the past,” thus discounting the intensifying fragility and persistent conflict challenges. 16 The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 2  apping of World Bank Group’s Support Areas to Mozambique’s 2016 Systematic Country Diagnostic Figure 2.1. M Priorities, Fiscal Years 2008–21 CPS FY08–11 CPS FY12–16 SCD 2016 Development Constraints CPF FY17–21 Pillar 1: Increased Accountability and Pillar 1: Competitiveness Growth Sustainability Public Voice and Employment • Competitiveness of the private sector • Institutional framework for managing • Budget planning • Business regulatory environment • Productivity in agriculture and forestry public resources • Fiduciary systems • Agricultural productivity • Enabling infrastructure • Governance environment for private • Government transparency • Tourism development sector investment • Electricity access • Legal and judicial services • Enabling infrastructure (spatial • Empowerment of local communities to planning, roads, water and sanitation, manage natural resources electricity, telecommunications) • Disaster risk management Pillar 2: Equitable Access to Key • Education and skills Inclusiveness Services • Access to finance • Equity in health services access • Management of urbanization process Focus Area 2: Investing in Human • HIV/AIDS treatment Capital • Education quality • Education Pillar 2: Vulnerability and Resilience • Education and skills • Health services • Water and sanitation • Health services • Health services • Water and sanitation • Electricity access • Climate change adaptation • Water and sanitation • Coverage of social protection • Disaster risk management programs Pillar 3: Sustainable and Broad-Based Growth Focus Area 3: Supporting Recovery and • Improved business environment Sustainability Resilience • Access to finance for SMEs Pillar 3: Governance and Public • Institutional framework for managing • Accountability and transparency of Sector Capacity public resources government institutions • Enabling infrastructure (energy, roads, telecommunications) • Public financial management • Governance environment for private • Coverage of social protection and labor • Agricultural productivity and tourism • Subnational capacity building sector investment programs development • Social accountability • Empowerment of local communities to • Urbanization and decentralization • Water resource management • Natural resources and environmental manage natural resources • Climate change adaptation • Disaster risk management management • Disaster risk management • Natural resource management Sources: World Bank 2007, 2012a, 2016c, 2017b. Note: CPF = Country Partnership Framework; CPS = Country Partnership Strategy; FY = fiscal year; SCD = Systematic Country Diagnostic; SMEs = small and medium enter- prises. Both the government and the World Bank incentivized or sustained the atmosphere of optimism. The balance of evidence indicates that the gov- ernment showed little appetite for addressing conflict drivers or even pov- erty if it did not directly benefit regions in which the government’s support was concentrated. Flush with donor funding, the government encouraged development partners to focus on attracting private sector investment and anticipating Mozambique’s bright future rather than concentrating on the “difficulties of the past.” This perspective was particularly encouraged after the 2010 discovery of large gas reserves off of Mozambique’s northern coast. In parallel, factors internal to the World Bank also helped maintain optimism even as conditions in the field deteriorated. As a result, World Bank support often focused on the technical aspects of development challenges, with an implicit expectation that this aspect would itself overcome political econo- my barriers. The considerable donor support tied to World Bank disbursements put pressure on the World Bank to disburse, disincentivizing it from advocating more forcefully for critical reforms and implementation. Further issues cited as perpetuating optimism were the fact that Mozambique was not on the World Bank’s list of countries with fragility, conflict, and violence; the frequent rotation of Maputo-based staff; and the challenges of Mozambique’s northern regions, which were more difficult and expensive to visit. It is notable that signs of social conflict were often misinterpreted as exponents of an economic-political conflict between FRELIMO and RENAMO. As a result, there was less attention given to deeper conflict Independent Evaluation Group World Bank Group    17 drivers in society. At the same time, a number of World Bank staff members working on Mozambique were concerned about the regional divergencies and unequal benefits from the peace dividend. Opportunities to “right size” the prevailing optimism through analytical work were missed. Across the evaluation period, only a small share of advi- sory services and analytics (17 percent) directly described conflict risk, doing little to influence the general atmosphere of optimism. Even a pivotal report such as the Mind the Rural Investment Gap did not point to the link between comparative regional underinvestment and the resulting grievances and conflict risks (World Bank 2019a). Conflict-focused advisory services and analytics were not conducted until the publication of the Risk and Resilience Assessment in 2020. The 2016 SCD similarly presents a missed opportuni- ty. A member of the World Bank’s fragility, conflict, and violence unit was only brought on board late in the drafting process. Not surprisingly, the SCD lacked a structured conflict analysis that would have been expected in a postconflict country, and it underemphasized the importance of con- flict drivers while overestimating the country’s capacity to resolve conflict. Overall, the SCD characterizes fragility as an exponent of economic-political conflict between FRELIMO and RENAMO rather than as emblematic of the deeper conflict drivers present in society. According to staff, optimism per- sisted until 2020, when large-scale warfare escalated, a new country director was appointed, and Mozambique became eligible for the financing under the Prevention and Resilience Allocation. Engagement in decentralization was comprehensive at the beginning of the evaluation period, but—despite the roots of conflict in regional disparities—it stalled midway through the evaluation period and was limited to analytical support and subnational capacity building because of insufficient traction with the government. Investment in the lagging (and northern) regions did not increase significantly until the latest strategy period. Security and justice—key The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 2 drivers of fragility—remained unaddressed throughout the evaluation period. The CPS FY12–16 and the CPF FY17–21 cited youth disenfranchisement as a conflict driver, and the Fragile States Index listed demographic pressures as the second-fastest deteriorating indicator of fragility. However, between 2008 and 2020, few projects targeted youth or included explicit measures to pro- mote the inclusion of youth among their beneficiaries. Before the conflict in the north, the World Bank had made limited adap- tation to operating in a conflict-affected environment. When engaging in a conflict-affected environment, it is important to take into account the drivers of conflict and adapt operations to the context. Recent Independent Evaluation Group (IEG) work noted that almost half of World Bank projects in conflict-affected areas that operated without a conflict lens suffered from implementation challenges (World Bank 2021o). For most of the evaluation period, there was limited conflict-focused advisory services and analytics to inform investment projects or understand how to adapt to evolving risks. The setup of early-warning systems or collection of disaggregated data on 18 real and perceived grievances, for example, could have supported better risk monitoring and the identification of appropriate mitigation measures. Similarly, World Bank infrastructure reconstruction investments in the central and northern regions were largely conflict blind. Project teams did not include conflict specialists or monitoring. Although the teams helped strengthen service delivery, there was no explicit strategy to monitor or at- tempt to strengthen the social contract between citizens and the state. International Finance Corporation Strategic Engagement in Mozambique During the evaluation period, the International Finance Corporation (IFC) strategy for Mozambique evolved significantly to become fully integrated with the Bank Group process. In the first half of the evaluation period, the IFC strategy development process for Mozambique was largely informal, which led to the absence of a strategic approach and little coordination with the World Bank. The first two Bank Group strategy documents—covering the periods FY08–11 and FY12–15—had few references to IFC support, and IFC’s proposed engagement in these documents was not well integrated with the World Bank’s strategy. This led the FY08–11 CPS to state that support to pri- vate sector development in Mozambique “could have benefited from closer collaboration within the WBG [Bank Group]” (World Bank 2007, 67). Things started to change in 2015, and significantly so after 2018, when a more formal process of strategy development was put in place in IFC. Two Independent Evaluation Group World Bank Group    19 factors contributed to this shift: (i) a clear, high-level corporate mandate to work together; and (ii) a strong relationship between the IFC country manager and the World Bank country director at the time, who had worked together earlier at IFC and hence understood and appreciated IFC’s role and value added. The emergence of the hidden debt crisis also provided an impetus for greater collaboration. The hidden debt crisis made IFC realize the impact of the policy environment on private sector development and the need for closer coordination and collaboration with the World Bank on policy reforms. World Bank Lending Portfolio Over the evaluation period, the World Bank committed $5.7 billion to Mozambique. Of this commitment, $4.4 billion was investment project financing, $1.1 billion was development policy financing, and $220 milion was Program-for-Results operations. Trust funds provided financing of $312 million. Four sectors accounted for more than half of World Bank support during the evaluation period; just less than one-fifth of financing focused on public administration. The portfolio also included considerable investment in water, sanitation, and waste management (13 percent); education (13 percent); agriculture (12 percent); and energy and extractives (10 percent). Mozambique received significant budget support from the World Bank throughout the evaluation period. The World Bank approved 15 development policy operations (DPOs) during FY08–21. Most of the prior actions associ- ated with these DPOs supported reforms in the public administration sector (41 percent), followed by the financial sector (26 percent) and energy and extractives (12 percent; appendix B, table B.1). The main themes supported were public finance management (24 percent), public administration (14 per- The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 2 cent), financial stability (9 percent), and rural development (9 percent; appendix B, table B.2). The centerpiece of World Bank budget support was 8 DPOs contained in three programmatic series of Poverty Reduction Support Credits (PRSCs). The PRSCs were part of the overall programmatic budget support provided by the Group of Nineteen Donors (G-19). Coordinated budget support was provided in the context of a memorandum of understanding for the provi- sion of direct budget support dating back to 2004. Under this memorandum of understanding, prior actions and triggers with their corresponding indica- tors were drawn from the Performance Assessment Framework, a monitoring framework developed by the G-19. In interviews with IEG, World Bank staff acknowledged that the G-19 quickly lost relevance and the ability to push for meaningful results. World Bank staff noted that, for the most part, Per- formance Assessment Framework indicators lacked ambition and the World Bank became “locked” into a mechanism that was unable to achieve mean- ingful results. The rationale for participating, World Bank staff noted, was 20 that the process was owned by the government and had a framework with monitorable targets. But many of the targets lacked ambition, making it easy for the government to show a “glass half full” while allowing for key goals such as decentralization, debt management, and oversight of SOEs to fly under the radar. The World Bank approved 64 investment project financing projects, 3 Pro- gram-for-Results operations, and 27 trust-funded projects during FY08–21. Investment project financing focused on water, sanitation, and waste man- agement (17 percent), followed by education (15 percent); agriculture, fishing, and forestry (13 percent); and public administration (12 percent; figure 2.2). Program-for-Results projects supported reforms in health (43 percent), pub- lic administration (39 percent), and education (18 percent). The trust-funded projects targeted health (29 percent) and education (21 percent). nvestment Project Financing Commitments to Mozambique Figure 2.2. I by Sector, Fiscal Years 2008–21 800 18 700 16 Commitments (US$, millions) Share of commitments (%) 600 14 12 500 10 400 8 300 6 Independent Evaluation Group World Bank Group    21 200 4 100 2 0 0 n lth s n n es er r try rt t T to ce tio en tio io an ppo th IC iv ea es ec at i ra em rta rv ct O H uc r lS ist Su Se tra Fo o ag Ed sp n ia Ex al i d d m nc an an an ci Ad d na M So Tr an ng de Fi te ic gy hi ra l as b ,T s er Pu W i ,F try En d re s an tu du n, ul In tio ric ita Ag an ,S er at W Sector Source: World Bank Business Intelligence database, September 2021. Note: ICT = information and communication technology. World Bank–supported projects and operations that closed during the eval- uation period were rated as moderately satisfactory, on average, on overall development outcomes. IEG evaluated 32 (55 percent) of 58 closed projects. Of these, just more than half were rated moderately satisfactory, and one-fifth were rated satisfactory, which was similar to Eastern and Southern Africa, which had 53 percent of projects rated as moderately satisfactory and 21 per- cent rated as satisfactory during the same period. Roughly one-quarter of projects were rated moderately unsatisfactory or below (figure 2.3). At a sec- tor level, higher-rated outcomes were in the health, finance, and education sectors (appendix B, figure B.6). Social protection and energy and extractives were the sectors that performed less well. Figure 2.3. Overall Outcome Ratings, Fiscal Years 2008–21 Highly satisfactory (n=0) Satisfactory (n=7) Rating The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 2 Moderately satisfactory (n=18) Moderately unsatisfactory (n=4) Unsatisfactory (n=5) Highly unsatisfactory (n=0) 0 10 20 30 40 50 60 Share of projects with rating (%) Source: Independent Evaluation Group. During the review period, there were 19 IFC investments and 27 IFC advisory activities, totaling $1.3 billion and $32 million, respectively. Main sectors of activity were agriculture and forestry (26 percent) and oil, gas, and mining (21 percent). Weighted by value, electric power and finance and insurance were the main areas of focus (appendix B, figure B.7). 22 3 | Support for Agricultural Productivity Highlights Increasing agricultural productivity can reduce poverty, as agricul- ture dominates Mozambique’s economy and is the main source of income for most poor people in the country. World Bank analysis identified six priorities as critical to improving agricultural produc- tivity in Mozambique: (i) stimulate adoption of modern production technologies; (ii) improve rural infrastructure; (iii) facilitate access to markets; (iv) improve natural resource management; (v) reform land policies and administration; and (vi) correct gender disparities that undermine the ability of female farmers to make efficient use of resources and technologies. The World Bank’s assistance during the Country Program Evalua- tion period started with a focus on technology adoption, irrigation, and institutional support. Later, support expanded to include mar- ket access, forestry, land administration, and rural roads. Together, this support addressed the major constraints identified by analyti- cal work and various stakeholders. Gender inequality became part of World Bank support only later in the evaluation period. There is no concrete evidence so far of increased agriculture productivity in provinces supported by the World Bank. The two completed agricultural investment projects initiated in the earlier part of the evaluation period did not achieve their objectives. Pro- vincial-level productivity data do not show any significant improve- ments in provinces targeted by World Bank support. It is too early to determine the impact of the support initiated in the later part of the evaluation period, although one of the rainfed agriculture proj- ects (Sustenta) is considered promising, and the government has declared its intention to use that project’s design as the basis for a national program. 23  Increasing agricultural productivity was key for reducing rural poverty, income inequality, and gender gaps in Mozambique. Low productivity kept most small-scale farmers trapped in poverty (World Bank 2020f, 8). This continues to be the case particularly for women, who engage in farming at a higher rate than men but do not receive a commensurate share of farm income or have access to improved technology (JICA 2015). Consequently, increased agricultural productivity has considerable potential to alleviate rural poverty and reduce income inequality in Mozambique. Analytical Underpinnings During the evaluation period, the World Bank produced a considerable body of analytical work to identify areas that are critical to increasing agricultur- al productivity in Mozambique. Mozambique—Beating the Odds: Sustaining Inclusion in a Growing Economy provided a comprehensive analysis of poverty and gender issues, with a heavy emphasis on the role of agriculture (World Bank 2008). This analysis was augmented in later years by other World Bank analyses on food security, land policy, public expenditures to support agri- culture, gender, and growth corridors, all of which contributed to a better un- The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 3 derstanding of the challenges facing the sector and actions to address those challenges.1 The World Bank’s analytical work, together with academic liter- ature and interviews of knowledgeable stakeholders, identified the following actions as critical to increasing agricultural productivity in Mozambique: » Introduce improved agricultural technologies. Intensify agricultural re- search and support services, such as extension services and literacy training of farmers, to increase understanding and application of improved technologies. » Improve rural infrastructure. Expand electricity services, provide reliable do- mestic water supplies, build and rehabilitate rural roads and small-scale gravi- ty irrigation systems, and enhance internet technology to reduce the isolation of small-scale farmers from social services, input, and output markets. » Facilitate agriculture product marketing. Facilitate access to local, national, and international agriculture product markets through efficient value chains. » Enhance natural resource management. Introduce sustainable land and soil management through crop rotation programs that will conserve soil 24 structure and retain soil fertility, thereby reducing, and ultimately eliminat- ing, slash-and-burn agriculture that causes deforestation. » Reform land policies and administration. Support regulatory reforms related to land use planning, land sales, and land leasing to increase average farm sizes and private sector investment. » Correct gender disparities. Given the high proportion of women engaged in agriculture, address obstacles to increasing female agricultural productivity, in the context of the many other responsibilities in rural households and the gender-based violence that is common in Mozambique. Women have less ac- cess to land, lower literacy rates (Mozambique, Ministry of Education 2010), less formal employment, and lower remuneration for their labor than their male counterparts. Indeed, 90 percent of rural women are either unpaid for their work or are reimbursed informally (USAID 2019). The World Bank produced a considerable body of quality analytical work on Mozambique’s agricultural sector during the evaluation period. Since FY08, the World Bank produced 17 major analytical pieces on agriculture and relat- ed subjects. The first report, Beating the Odds (World Bank 2008), published at the start of the evaluation period, was a comprehensive policy analysis of poverty and gender issues, with an emphasis on the role of agriculture in poverty reduction. It concluded that improved research and extension could grow the agriculture sector and reduce poverty by increasing farmers’ abili- Independent Evaluation Group World Bank Group    25 ties to use new technologies and access markets. Other World Bank reports contributed to understanding the challenges facing the sector by focusing on land policy, growth corridors, food security, agricultural risks, and agri- culture-related public expenditures. After 2015, the World Bank reviewed broader policy constraints to agricultural growth, emphasizing the role of agricultural productivity and private investment in transforming the sector while also carrying out a review of agriculture-related public expenditures, an economic memorandum focusing on rural infrastructure, and a 2020 flag- ship report on income growth and poverty reduction in agriculture. These reports identified challenges related to public budget allocations to the agri- culture sector, shortcomings in rural infrastructure, land administration, and weak off-farm income-earning opportunities for rural households. Another influential report concluded that market access was critical to ensure that agricultural productivity would lead to increased rural incomes. As a result, market access became a regular theme for Bank Group–assisted agriculture investment projects in Mozambique in 2016 and beyond and represented an important addition from World Bank management’s earlier focus on technol- ogy and increased productivity. Drawing on these analytical underpinnings, all three Bank Group–supported strategies during the evaluation period sought to increase agricultural pro- ductivity. Figure 3.1 illustrates the evolution of World Bank support for in- creasing agricultural productivity. The CPS FY08–11 emphasized the need to increase investment in technology and infrastructure (irrigation and roads). The second strategy (CPS FY12–16) added improving land administration, facilitating access to rural finances, and supporting the decentralization of the agriculture sector administration. The FY17–21 CPF expanded the strategy to include market access and com- mercial farming. “Republic of Mozambique: Agriculture and Rural Develop- ment Nonlending Technical Assistance—Synthesis Report” concluded that “increasing agricultural productivity can be effective in raising rural incomes only if farmers have access to output markets, knowledge, and modern tech- The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 3 nology” (World Bank 2016e).2 The 2020 Cultivating Opportunities for Faster Rural Income Growth and Poverty Reduction: Mozambique Rural Income Di- agnostic came to similar conclusions (World Bank 2020a), identifying three priority actions as necessary to achieve rural income growth: (i) adopt im- proved agricultural technologies; (ii) enhance access to markets for surplus smallholder production; and (iii) reconstruct or rehabilitate strategic rural roads. Other analytical work such as the Republic of Mozambique Agrarian Sector Transformation: A Strategy for Expanding the Role of the Private Sector focused on the role of the private sector in transforming the agrarian sector so it would be more productive and competitive (World Bank 2019c). That analysis called for the government to stimulate private sector investment in agriculture by improving the business climate. Relevance of World Bank Support Investment projects addressed several constraints to the adoption of im- proved technology and rural infrastructure. These projects included com- 26 munity demand–driven investments (Market-led Smallholder Development in the Zambezi Valley Project; World Bank 2016b) and small-scale irrigation development (MZ PROIRRI Sustainable Irrigation Development; World Bank 2019b). The community-driven development model (focused on Sofala, Man- ica, and Zambezia) provided financing for rural roads and bridges chosen by communities and on-farm grain silos with the objective of increasing farm incomes through higher yields and reduced losses. The irrigation project, which focused on the same provinces, engaged farmers in participatory irrigation management to improve the enabling environment and strength- en institutions. The IFC contributed to investments totaling $32 million in a private company to finance improvements in wheat flour mills and pasta and biscuit production and for warehousing and trucks in Maputo, Beira, and Nacala between 2007 and 2012. In addition, IFC invested $7 million in grain handling and storage facilities at the port of Nacala in 2008. Two DPOs contained prior actions to support policy and institutional re- forms to increase agricultural productivity. The two DPOs sought to support agricultural technology, access to productive assets, and the monitoring of sector performance. Prior actions aiming at supporting technology adoption focused on Southern African Development Community–compliant policy and institutional reforms governing seed production, trade, quality control, and certification; ratification of the regulations for private sector–led fer- tilizer production and marketing; and plant breeders’ rights and procedures Independent Evaluation Group World Bank Group    27 for the registration of fertilizers. Prior actions aiming at facilitating access to productive assets focused on regulations for irrigation, simplification of procedures for transferring rural land use rights (called Direito de Uso e Aproveitamento dos Terras; DUATs), development of rural financial services, and investment plans for agriculture. The third operation was canceled in the wake of the hidden debt crisis. 28 The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 3  orld Bank Support to Increasing Agricultural Productivity in Mozambique, Fiscal Years 2008–21 Figure 3.1. W Challenges to Increasing World Bank Support to Outputs Outcomes Impact on Agricultural Agricultural Productivity Agriculture Productivity FY08–16 Focus on technology and infrastructure (ACTUAL) Rainfed agriculture Community investments Some rural roads and On-farm investments productivity rose, but by Absence of modern bridges built household grain storage Modest results less than control group technologies Small-scale irrigation Rice and horticulture No significant impact on productivity increased, agricultural productivity Inadequate rural cropping intensity infrastructure Policy reforms through Access to improved Development of 3,000 unchanged, policy reform DPOs technology and finance hectares of irrigation stalled FY16—21 Support extended to market access Small-scale farmers’ Small-scale farmers Small-scale farmers activities supported by adopted linked to commercial (EXPECTED) Inadequate access to lead farmers improved technologies markets product markets Better access for Expected results smallholders to Improved irrigation commercial value chains Irrigation and natural Small-scale irrigation Expected increase in Inadequate natural management agriculture productivity resource management Sustained natural resource management due to improved access to resource management and deforestation Better forest land use Reduction of markets slash-and-burn Transformation of Land administration agriculture agriculture through Outdated or discriminatory Land rights clarified improved technology, land policy land sales Reconstruction and Gender-sensitive rehabilitation of rural roads Feeder roads DUATS Source: Independent Evaluation Group. Note: DPO = development policy operation; DUATS = land use rights; FY = fiscal year. Since 2016, the World Bank has approved three investment projects that support improved market access. First, the Agriculture and Natural Resources Landscape Management Project (Sustenta) aimed to link small-scale farmers to commercial lead farmers and cooperatives and to finance business plans for small emerging commercial farmers (SECFs) and public extension agents to assist smallholder cooperatives. The World Bank also financed an irrigation project (Smallholder Irrigated Agriculture and Market Access Project; World Bank 2018f) to increase productivity through more efficient irrigation and to improve market access of production through enhanced business linkages between farmers and traders. A third project, the Northern Mozambique Rural Resilience Project, sought to improve access to opportunities for vulnerable communities and management of natural resources in selected rural areas of northern Mozambique. This was partly an emergency operation to assist internally displaced people affected by a hostile insurgency, but it also promoted growth and improved agricultural productivity using public extension agents to assist smallholder cooperatives. In addition, the project supported SECFs with grants to implement business plans for investments in priority value chains that benefit small-scale farmers. The World Bank also supported increasing agricultural productivity through reforms to land rights, natural resource management, and rural roads. The Land Administration Project approved in 2018 was intended to increase land tenure security in selected districts and enhance the efficiency and accessi- Independent Evaluation Group World Bank Group    29 bility of land administration services. An important element of this project’s design was decentralization of land administration to the community level and more rapid issuance of DUATs (Direito de Uso e Aproveitamento dos Terras)—that is, land use rights—to farmers. This change was expected to increase the ability of small-scale farmers to sell land-user rights and allow them to use DUATs as collateral to obtain credit to finance the adoption of technological changes that would increase productivity. Another area of investment was by IFC in a company producing avocados, in 2017 (€3.9 mil- lion) and 2019 ($2.8 million). Several projects sought to reduce the substantial loss of forest areas each year as a result of slash-and-burn activities on small-scale farms. The Forest Investment Project aimed to reduce forest-area losses and increase agricul- tural and pastoral productivity (World Bank 2017c). The project sought to achieve these objectives by supporting geospatial capacity building for forest landscape development to ensure equitable and sustainable land use and by strengthening measures to adapt to and mitigate climate change. Until 2018, World Bank support for roads focused on the rehabilitation and maintenance of highways and their bridges along the important north-south transpor- tation network. In 2018, the World Bank shifted focus from main roads and highways to selected rural areas by approving the Integrated Feeder Road Development Project. This project was intended to provide rehabilitation and maintenance on sections of secondary, tertiary, and vicinal (between towns) roads, as well as on some unclassified roads to enhance mobility in Zambezia and Nampula provinces to reduce transportation costs and facil- itate farmers’ access to markets. The project is also expected to rehabilitate about 70 kilometers of primary roads to enhance connectivity to markets, ports, and other economic and social services. The scope of this project, however, is relatively small.3 The inclusion of gender in the World Bank’s agriculture productivity support became progressively more prominent over the years. At the beginning of the evaluation period, World Bank support included little gender targeting or The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 3 explicit gender objectives. The only references to women related to educa- tional inequality and maternal mortality. Gender targeting emerged in the CPS FY12–16. Specifically, objective 2 identified gender mainstreaming in land tenure and women’s micro enterprises as critical areas to support. In the latter part of the evaluation period, the World Bank incorporated gen- der in a more decisive and comprehensive manner. The SCD 2016 (World Bank 2016c), the CPF FY17–21 (World Bank 2017b), and the SCD update in 2021 dedicated space to gender in agriculture (World Bank 2021j), indicated priorities for addressing gender inequality to increase the impact of poverty reduction efforts, and focused on the need for attention to human capital development to increase the productivity of women in agriculture. The SCD update demonstrates the importance of women’s engagement in the design of agricultural extension programs. A change in the World Bank’s approach to gender targeting can also be observed with Gender Responsive Natural Re- source and Landscape Management: A Mozambique Pilot Program (2020i); the Sustenta (FY16) and Estrela (FY16) projects, which contained results indi- cators for female beneficiaries; and the Northern Mozambique Rural Resil- 30 ience Project (FY20), which included the involvement of women and girls as beneficiaries and stakeholders. World Bank Effectiveness Only two agricultural investment projects closed in the first half of the evaluation period, and neither performed well. Although beneficiaries in the Market-Led Smallholder Development in the Zambezi Valley Project achieved increased productivity, increases were substantially lower than those achieved by similar farmers not supported by the project (World Bank 2006). The quality of the project’s modest rural road construction and small bridges was low because of inappropriate design and insufficient mainte- nance. An impact evaluation of the extension program supporting improved land management found that there was a statistically insignificant impact on farmers’ adoption of sustainable land management techniques promot- ed by the project (Kondylis, Mueller, and Zhu 2014). A Project Performance Assessment Report for this project concluded that there were significant shortcomings in the project’s implementation and rated the overall outcome as moderately unsatisfactory (World Bank 2016b). An irrigation project saw the yield for major crops (rice and horticulture crops) increase, but cropping intensity in irrigation areas did not increase beyond one crop per year of rice, and new irrigation development fell short of the original target. IEG rated this project’s overall outcome as moderately unsatisfactory because of sig- nificant shortcomings in the achievement of objectives and modest efficien- cy (World Bank 2019b). Independent Evaluation Group World Bank Group    31 Agricultural productivity in provinces where World Bank projects were implemented since the start of the evaluation period does not show signifi- cant improvement compared with the average for Mozambique overall. The World Bank supported agricultural productivity in Nampula and Zambezia. These provinces produce almost one-quarter of all maize in Mozambique and represent almost 40 percent of the total population. Results show that the change in productivity in these supported provinces was similar to the change in the rest of the country (figure 3.2).4  aize Productivity in Nampula, Zambezia, and Mozambique, Figure 3.2. M 2000–17 1,2 Maize productivity (ratio) 1,0 0,8 0,6 2000 2005 2010 2017 Year Mozambique Zambezia Nampula Source: Independent Evaluation Group calculations based on International Food Policy Research Insti- tute data. The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 3 Note: Productivity = production/area. Early indications suggest positive outcomes for agricultural productivity due to greater attention to market access. Sustenta, whose core design is establishing and maintaining a network of 200 SECFs, supports improved technological know-how for small-scale traditional farmers. Bilateral donors piloted this SECF model in a range of provinces in Mozambique since 2009. Early indications are that Sustenta has made progress in building linkages between SECFs (lead farmers) and small-scale traditional farmers by meet- ing targets on developing value chains, maintaining rural roads, reforesting project areas, and promoting technology use by farmers. The design of this model benefited from lessons learned in other countries on comparable approaches, such as the case of nucleus estates, outgrower schemes, and lead farmer models in Ghana, Malawi, and Tanzania (Paglietti and Sabrie 2012; Stringfellow 1996). The Mozambican design also incorporated lessons on the essential role of extension services by ensuring that training and advice to small-scale farmers will be provided by SECFs and public extension services. 32 In interviews for this evaluation, national agricultural development experts praised Sustenta’s project design as a major improvement over the World Bank’s previous agricultural development projects. Beneficiaries have also voiced strong support for agroforestry enterprises as alternatives to defor- estation in the World Bank–financed Forest Investment Project. An import- ant and successful feature of this project has been a detailed arrangement, designed and approved by stakeholders, to equitably distribute Carbon Fund payments to program participants. Market development of the agriculture sector in Mozambique requires sup- port to its institutional setup. The agricultural sector is burdened by weak institutional capacity to facilitate business activities and numerous regulato- ry barriers that increase the cost of doing business. Major improvements are required in agricultural research and extension institutions to provide the basis for the adoption of new technologies in Mozambique’s wide range of agro-climatic zones. Institutions that support microcredit will be necessary to finance small-scale farmers’ purchase of inputs associated with improved technologies such as seed and fertilizer. Lastly, institutions that can facili- tate greater efficiency, such as weather forecasting and market information through radio and the internet, would benefit small-scale farmers and the employment prospects for unskilled workers from the agricultural sector. A review of the lessons identified in evaluations of agricultural projects in Mozambique shows good learning from experience. Implementation Com- pletion and Results Report Reviews and aide-mémoire in the first strategy Independent Evaluation Group World Bank Group    33 period identified several important lessons related to agriculture and fish- eries. These lessons included the need to examine (and support) local gov- ernment capacity at the outset of the project in decentralized projects, the importance of flexible arrangements and longtime horizons of projects to build strong decentralization capacity, and the need for clear targeting and appropriate linkages for community empowerment. Time is reported as more important than money in sensitizing communities to issues of sustainabil- ity. For this reason, local presence, dedicated staff, and regular agriculture extension visits were considered critical to implementing agricultural proj- ects in Mozambique. Many of these lessons were reflected in the following projects in the second strategy period. For example, the First Agriculture De- velopment Policy Operation rightly identified the need to build both admin- istrative and technical capacity of institutions that implemented reforms. Similarly, the Smallholder Irrigated Agriculture and Market Access Project incorporated lessons from the MZ PROIRRI Sustainable Irrigation Develop- ment Project (World Bank 2018f). Fisheries projects incorporated earlier les- sons such as the need for a long horizon to complete reforms, a phased and consultative approach with local communities, and a strong implementation plan. The Mozambique Land Administration Project (Terra Segura) benefited from lessons set out in a World Bank Policy Note (Community Land Delim- itation and Local Development; World Bank 2010a), such as the need for a more proactive, systematic, and clearly targeted program of community land delimitation. Finally, the design of the Sustenta project was based on rele- vant lessons drawn from, among other places, the 2016 SCD and the latest CPF, and the Northern Mozambique Rural Resilience Project drew lessons from Sustenta. The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 3 34 1  See the following analyses: Higher Fuel and Food Prices (FY09); Infrastructure Corridors, Growth, and Welfare: Comparative Study of the Corridors of Beira and Nacala, Africa (FY09); Community Land Delimitation and Local Development (FY11); Poly Note: Rural Land Taxation in Mozambique (FY11); Analysis of Public Expenditure in Agriculture (FY12); and Mozambique Agricultural Sector Risk Assessment (FY15). 2  A similar conclusion had been drawn in Mozambique—Beating the Odds: Sustaining Inclusion in a Growing Economy (World Bank 2008), but it did not influence the operational strategy until 2016. 3  According to the program document, the classified road network—primary, secondary, tertiary, and vicinal—is 30,464 kilometers, 24 percent of it paved. This is equivalent to a road density of 2.9 kilometers per 100 square kilometers of land, which is relatively low compared with neighboring countries such as Kenya (10.8 kilometers per 100 square kilometers) and Tanzania (5.5 kilometers per 100 square kilometers). 4  Nampula and Zambezia were the provinces receiving significant support from the World Bank. Although this analysis assumes that the volume of World Bank investments is suf- ficiently large to influence provincial changes in productivity, these results nevertheless confirm the micro-level results in the previous paragraph. Independent Evaluation Group World Bank Group    35 4 | Support for Access to Basic Services Highlights World Bank projects during the evaluation period have progres- sively targeted provinces that have insufficient access to education, health, transport, and electricity. Widespread disparities in access to these basic services in rural and urban areas were recognized by the World Bank’s analytical work. However, World Bank projects target- ed these poorer areas only in the later part of the Country Program Evaluation period. Across services, provinces with insufficient access to education and health have been targeted earlier than those with insufficient access to transport and electricity. Across virtually all basic services, changes in access for rural areas were positive. Project documents report positive outcomes across provinces in the delivery of health, transport, education, and (in part) electricity services throughout the period. Similarly, Indepen- dent Evaluation Group analysis and official statistics from a recent household survey confirm improved access to these services during the evaluation period. 36   Mozambique is characterized by large inequalities in access to basic services, limiting the degree to which poor people share in economic growth. Income levels and access to health, education, transport, and elec- tricity (henceforth referred to as basic services) are highly unequal across districts, with rural areas in the central and northern parts of the country lagging urban centers. At the beginning of the evaluation period, only 2 per- cent of rural inhabitants had access to electricity and less than 1 percent had access to water, compared with 40 percent and 13 percent, respectively, of urban inhabitants. Similarly, 61 percent of rural residents were illiterate, compared with 27 percent of urban residents, and almost half of the rural population lived at least one hour away from public transportation, com- pared with 10 percent of the urban population (National Institute of Statis- tics 2010). Rural areas are among the poorest in the country; thus, neglect of service provision in rural areas is equivalent to neglecting the poorest regions. This neglect exacerbates grievances, contributes to inequitable out- comes, and reinforces social exclusion (World Bank 2020g). Increasing the availability and quality of health, education, transport, and electricity is key to improving and equalizing opportunities for all citizens and enabling them to share in economic growth (World Bank 2018c). Throughout the evaluation period, and especially after 2016, World Bank analytical work acknowledged the stark contrast in access to basic services among regions and between urban and rural areas in Mozambique. Using Independent Evaluation Group World Bank Group    37 multiple data sources from domestic and development partners,1 the eval- uation team found several reports across the three strategy periods that recognized the inequitable access to education, health, transportation, and electricity between urban and rural populations. The availability of roads, landlines, internet, and electricity for lighting was a privilege of mainly ur- ban households (Adriano et al. 2012). Mozambique Poverty Assessment: Strong but Not Broadly Shared Growth reaffirmed that location continued to be a strong determinant of access to basic services and acknowledged that de- spite modest improvements, large disparities remained (World Bank 2018d). Mozambique Economic Update: Shifting to More Inclusive Growth placed an emphasis on public investment in rural areas to strengthen irrigation and access to rural roads and electricity to provide better conditions for increas- ing agriculture productivity and creating more dynamic rural markets (World Bank 2018c). However, addressing inequality in access to basic services was not a focus of World Bank interventions until the last strategy period. Inequalities between rural and urban communities were acknowledged in the first strategy, but no specific actions were identified. Only in the health sector was support to rural areas a focus. In the first strategy, the World Bank focused on the decentralization of government functions, which are seen as important to the delivery of basic services. The second World Bank–supported strategy focused mainly on urban areas, leaving rural support to the other develop- ment partners. Only in the third strategy did the World Bank put significant emphasis on support to rural areas. It noted that “new planned investments … will focus on geographic zones with weak indicators in the areas of repro- ductive, maternal, newborn, child, and youth health, areas most often locat- ed in the central and northern regions” (World Bank 2017b, 21). The World Bank supported all four basic services throughout the evaluation period. A total of 61 projects were approved during the FY08–21 period, for a total commitment of almost $5 billion (tables 4.1 and 4.2). The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 4 Table 4.1. World Bank Portfolio in Basic Services, Fiscal Years 2008–21 FY08–11 FY12–16 FY17–21 Total Sector Projects approved (no.) Education 4 6 7 17 Health 3 3 6 12 Transportation 2 9 5 16 Energy 4 7 5 16 Source: World Bank Business Intelligence database. Note: FY = fiscal year. 38  orld Bank Commitments in Basic Services, Fiscal Years Table 4.2. W 2008–21 FY08–11 FY12–16 FY17–21 Total Sector Commitments (US$, millions) Education 205 225.7 740 1,170.7 Health 114 172.7 544.2 830.9 Transportation 91 437.5 610 1,138.5 Energy 200 1,250.1 367.2 1,817.3 Source: World Bank Business Intelligence database. Note: FY = fiscal year. The World Bank improved its targeting of poorer provinces over the eval- uation period. Using provincial income per capita (as a proxy for access to services) and a portfolio of 29 projects (corresponding to about 70 percent of these sectors’ coverage) in 814 locations, the analysis mapped World Bank project sites in each province.2 Figure 4.1 presents the distribution across provinces and strategy periods between levels of per capita income and World Bank support. The color of the provinces represents the level of per capita income (with darker colors showing poorer provinces), and the size of the bubbles represents the number of project sites. The map shows that the World Bank’s support to provinces changed over time, with an increase in activities in provinces with lower income per capita. The analysis also grouped provinces into three income categories (highest, Independent Evaluation Group World Bank Group    39 medium, and lowest) and three categories of population (most populous, medium, and least populous). Figure 4.2 shows the evolution of the World Bank’s support for those income categories. The dotted red line indicates the relationship between level of income (on the y-axis, proxied by income per capita) and level of World Bank support (x-axis, number of project loca- tions). The figure shows a negative trend over time, with World Bank support becoming more concentrated in the poorest provinces. Similarly, figure 4.3 shows that the World Bank increasingly targeted the most populous provinc- es over the three strategy periods. 40 The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 4  istribution of Project Locations by Provincial Gross Domestic Product per Capita Figure 4.1. D FY08—11 FY12—16 FY17—21 Cabo Delgado Nassa Cabo Delgado Nassa Cabo Delgado Nassa Nampula Nampula Tete Nampula Tete Zambezia Tete Zambezia Zambezia Manica Sofala Manica Manica Sofala Sofala Gaza Inhambane Gaza Inhambane Inhambane Gaza Maputo Maputo Maputo Lowest Projects' Sites (Cumulative) 77 60 40 20 Highest Source: Independent Evaluation Group calculations. Note: Darker colors indicate lower level of income per capita. The size of the bubble includes new and open projects in each strategy period.  orrelation between Levels of World Bank Group Support Figure 4.2. C and Need, by Strategy Period 2,500 GDP per capita, all sectors (US$) 2,000 1,500 1,000 500 FY17—21 FY12—16 FY08—11 0 0 50 100 150 200 250 Access (no. project sites) Highest income Medium income Lowest income Median Source: Independent Evaluation Group calculations. Note: FY = fiscal year; GDP = gross domestic product. Figure 4.3. Targeting Trend by Population Share, by Strategy Period 4,500,00 FY17—21 4,000,00 Independent Evaluation Group World Bank Group    41 Population, all sectors (no.) FY12—16 3,500,00 FY08—11 3,000,00 2,500,00 2,000,00 1,500,00 1,000,00 500,00 0 0 50 100 150 200 250 Access (no. project sites) Most populous Mid populous Least populous Median Source: Independent Evaluation Group calculations. Note: FY = fiscal year. Targeting of poorer provinces began earlier in the health and education sectors. The analysis also looked at the individual services using outcome indicators specific to each sector, such as level of illiteracy, access to health services, and access to transport and electricity. Provinces were divided into three groups based on need (highest, medium, and lowest), with the highest need defined as those provinces with the lowest levels of access. Social sec- tors (health and education) intentionally targeted poorer provinces already in the second period, while transport and electricity had a targeting of the poorest provinces in the latest period (figure 4.4; see appendix D). Figure 4.4. Sector Targeting across Strategic Period a. Education 70 FY08—11 FY17—21 FY12—16 60 Need: Illiteracy rate (%) 50 40 30 20 The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 4 10 0 0 10 20 30 40 50 60 70 Access (no. project sites) Highest need Medium need Lowest need Median b. Health 100 90 Need: Access to health (%) 80 70 60 50 FY12—16 FY17—21 40 FY07—11 30 20 10 0 0 10 20 30 40 50 60 70 Access (no. project sites) Highest need Medium need Lowest need 42 Median c. Transportation (road density) 0.06 0.05 (%) 0 0 10 20 30 40 50 60 70 Access (no. project sites) Highest need Medium need Lowest need Median c. Transportation (road density) 0.06 0.05 Need: Road density (%) 0.04 0.03 FY08—11 FY17—21 FY12—16 0.02 0.01 0 0 10 20 30 40 50 60 70 80 Access (no. project sites) Highest need Medium need Lowest need Median d. Energy 2.5 Need: Average radiance (%) 2.0 1.5 1.0 0.5 FY12—16 FY08—11 FY17—21 0 0 5 10 15 20 25 30 35 Independent Evaluation Group World Bank Group    43 Access (no. project sites) Highest need Medium need Lowest need Median Source: Independent Evaluation Group calculations. Note: FY = fiscal year. World Bank projects have progressively improved access to services across provinces. Project documents report the achievement of project objectives in most cases (table 4.3) and positive outcomes in education, health, trans- port, and (in part) electricity. In education, access to primary and secondary schooling surpassed targets, with net enrollment in grade 1 reaching 93 per- cent in 2019. At the same time, the pupil-teacher ratio improved in prima- ry and secondary education, going from 72 to 1 in 2008 to 66 to 1 in 2010. Zambezia’s pupil-teacher ratio fell from 105 to 1 in 2006 to 82 to 1 in 2010. Primary and secondary participation rates for girls slightly improved over the same period, with the primary completion rate for girls increasing from 36.1 percent in 2007 to 45.8 percent in 2010. Similarly, the graduation rate for both male and female students surpassed the target, and between 2014 and 2018, the attendance rates of teachers and students improved because of increased supervision. Projects exceeded targets for access to maternal health services, vaccina- tions, and immunization in targeted provinces. The percentage of institu- tional deliveries in Niassa, Nampula, and Cabo Delgado increased from a baseline of 51 percent in 2009 to 93 percent in 2017. The share of pregnant mothers who received a second dose of intermittent preventive treatment for malaria reached 70.3 percent in 2017, and in Cabo Delgado, Niassa, and Nampula, the percentage of children vaccinated with the pentavalent vac- cine increased from a baseline of 45 percent in 2009 to 97.3 percent in 2017. In addition, during the FY12–15 period, the availability of selected drugs and medical supplies improved at key distribution points. The number of provinces achieving minimum compliance with standards increased from The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 4 baseline zero in 2013 to 10 in 2018, compared with a target of eight provinc- es. The percentage of antiretroviral medications delivered to health facilities increased from 94.5 percent in 2011 to 98.6 percent in 2012 and to 100 per- cent in 2013 and 2014. However, a lack of routine data collection and endline surveys prevented a conclusive assessment on improvements in access to nutritional services in Cabo Delgado, Nampula, and Niassa. Partial data from two provinces suggest there were improvements. In transport, despite problems with safeguard provisions and codes of conduct, World Bank support led to improved all-season access through the rehabilitation of roads. Under the Maputo Municipal Development Program II (approved FY11), 968,963 people in urban areas received access to all-season roads, from a baseline of 60,000 and with a target of 300,000. Improvements were the result of investment in secondary paved and unpaved roads in peri-urban areas and the rehabilitation of arterial roads. Operating expenses for road maintenance increased at the municipal level. World Bank investment projects in bridges and water passages contributed 44 to better road networks in Nacala and Zambezia. In the Nacala Special Economic Zone, 41 bridges and water passages were rehabilitated and upgraded. The length of roads rehabilitated in the Zambezi Valley increased from 0 kilometers at the baseline (2013) to 207.6 kilometers by 2020 (only 7 kilometers short of the target of 215 kilometers). In electricity, the World Bank was partially successful in increasing access in peri-urban and rural areas through both grid extension and application of off-grid solar photovoltaic solutions. However, World Bank support did not directly contribute to reductions in the frequency of electricity outages. The limited scope of World Bank activities was not consistent with improving ac- cess to modern energy services in a sustainable and affordable manner. The installation of a small number of improved wood-fuel stoves and solar water heaters in hospitals was not sufficient to increase access to modern energy services in peri-urban and rural areas. Table 4.3. Project Ratings across the Evaluation Period Projects with Rating (no.) Moderately Moderately Sector Satisfactory satisfactory unsatisfactory Unsatisfactory Education 1 2 0 0 Health 1 2 0 0 Transportation 0 3 0 1 Energy  0 2 0  3 Independent Evaluation Group World Bank Group    45 Source: Independent Evaluation Group. Official statistics from a recent household survey confirm improved access to the four basic services for rural populations over the evaluation period. The latest household survey (2019–20) shows improvements in access to education, health, transportation, and electricity over the evaluation period (figure 4.5). The World Bank helped reduce regional inequalities in access to basic ser- vices in each sector, although with different levels of success. Several im- provements in the quantity and quality of education services were recorded, including improved access to primary and secondary schooling and increases in graduation rates and teacher-student ratios. Similarly, in health, World Bank projects facilitated access to vaccines and increases in institutional deliveries. In transport, World Bank–supported rehabilitation of roads and bridges eased access to transportation services; in electricity, World Bank support to grid extension and renewable energy improved access to electric- ity in underserved areas. To determine the extent to which these achieve- ments have contributed to more equal access to services across provinces, IEG calculated a Gini coefficient–style inequality index for each of the sec- tors using higher-level outcome indicators, such as province-level literacy rates and share of province population with access to health, transport, and electricity services. Figure 4.6 shows that over the evaluation period, equality improved in all four sectors, providing further evidence of the World Bank’s contribution to shared prosperity in Mozambique. In education, the World Bank’s contribution was positive but marginal because in the two provinces with the largest populations (Nampula and Zambezia), the improvement rate in literacy was almost half the improvement in other provinces. Interviews with staff and donors confirm that high population growth rates made it challenging to meet the growing need for educational services. Figure 4.5. Share of the Rural Population with Access to Basic Services The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 4 80 Share of rural population �with access (%) 70 60 50 40 30 20 10 0 Health Literacy Paved roads Electricity Basic service 2008–09 2014–15 2019–20 Source: National Statistical Office, various years. 46 Gini-Style) Inequality Coefficients for Each Service across Figure 4.6. ( Strategy Periods 0.25 0.6 0.5 Inequality coefficient (no.) Inequality coefficient (no.) 0.20 0.4 (more equal) 0.15 0.3 0.2 0.10 0.1 0.05 0.0 2008–09 2014–15 2019–20 Year Paved roads Literacy Access to health Access to electricity (right axis) Source: Independent Evaluation Group calculations using Inquérito sobre Orçamento Familiar 2019–20 data. Independent Evaluation Group World Bank Group    47 1  Data sets included the following: Mozambique Multiple Indicator Cluster Surveys, Demo- graphic and Health Surveys, National Household Surveys on Living Conditions, Population Census, Family Budget Household Surveys, National Institute for Social Security, physical in- frastructure databases, and other sources such as nighttime lights data, along with the World Bank’s Education Statistics, the International Disaster Database, and the United Nations Educational, Scientific, and Cultural Organization. 2  The analysis used only location as reported in project documents. Data on commitments by location were not available to be included in the analysis. The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 4 48 5 | Support for Addressing Weak Governance Highlights The World Bank contributed to improved public financial manage- ment by increasing coverage of financial management information systems and strengthening internal and external control func- tions at the central level. However, World Bank support for budget preparation and execution did not enhance budget credibility. Despite clear weaknesses in public investment management, it was only in the wake of the hidden debt crisis in 2016 that the World Bank made concerted efforts to intensify support. Despite progress “on paper,” the institutionalization of public investment management reform is lagging. World Bank support for improving debt management and reform- ing state-owned enterprises had a modest impact. Support was fo- cused on building technical and institutional capacity (which were legitimate constraints), but engagement did not adequately take into account the context of weak governance. Tangible progress was made only when this context was taken seriously into account. The World Bank was not effective in supporting the establishment of a coherent decentralization policy framework, with progress affected by political economy constraints. Implementation of public financial man- agement reforms at the subnational level faced significant challenges, but these were successfully addressed using Program-for-Results financing. World Bank–supported projects contributed to tangible improvements in municipal revenue collection. 49  The World Bank contributed to the establishment of a regulatory framework for managing the extractives sector and complying with transparency standards. However, World Bank support for the im- plementation of a fiscal rule and a sovereign wealth fund for man- aging revenues from the extractives sector did not lead to tangible outcomes. 50   World Bank support to improve governance in Mozambique focused on five areas: (i) public financial management (central government); (ii) public debt management; (iii) SOE reform; (iv) decentralization; and (v) transpar- ent and effective management of extractives. World Bank support for improving public financial management focused appropriately on priorities identified in key diagnostics. Using the 2005 Public Expenditure and Financial Accountability (PEFA) assessment and the government’s second Action Plan for the Reduction of Absolute Poverty as a starting point, the CPS FY08–11 identified three strategic priorities: (i) establishing an integrated financial management information system; (ii) enhancing budget credibility through support for budgeting capacity at all levels and better alignment between budget expenditures and policy priorities; and (iii) strengthening the external and internal audit functions. These priorities were identified as critical for improving the use of public resources and reducing fiduciary risks in Mozambique. At the beginning of the evaluation period, the World Bank’s public financial management support focused on improving budget credibility by building budgeting capacity at the central and subnational levels and rolling out an integrated financial management information system called Electronic State Financial Administration System (e-SISTAFE). The World Bank also emphasized downstream public financial management aspects such as strengthened internal and external controls and compliance with auditing and accounting standards. Later in the evaluation period, in the wake of the hidden debt crisis, the World Bank increased its support for public investment Independent Evaluation Group World Bank Group    51 management (PIM), which had previously received little attention. Integrated Financial Management Information System The World Bank contributed to increasing the coverage of Mozambique’s integrated financial management information system, known as e-SISTAFE. At the beginning of the evaluation period, Mozambique’s e-SISTAFE had low coverage, particularly among autonomous institutions, district governments, and line ministries at the central and provincial levels (World Bank 2007). The World Bank sought to expand coverage via prior actions in the PRSC se- ries, including by rolling out e-SISTAFE to ministries and requiring increased coverage of budgetary expenditures implemented through e-SISTAFE. The World Bank also provided technical assistance through the Public Sector Reform Project (FY03–10) and the National Decentralized Planning and Finance Program Project (FY10–15). The Public Sector Reform Project rolled out e-SISTAFE to all ministries at the central and provincial levels, in 50 out of 128 districts, and in 29 autonomous institutions, thus meeting the project targets. In 2010, 97 percent of the budget (goods and services) of ministries at the central and provincial levels was allocated through the system (World Bank 2014d). The National Decentralized Planning and Finance Program Project supported the implementation of e-SISTAFE at the district level. By the end of FY14, 91 districts had closed their financial processes in e-SIS- TAFE with project support, falling short of the target of 128 districts (World Bank 2016d). By the end of the evaluation period, e-SISTAFE covered plan- ning and budgeting, budget execution, and budget reporting at the central, provincial, and district levels (IMF 2019). In 2021, e-SISTAFE was considered by the International Monetary Fund to be comprehensive and adequate for enabling the government to produce fiscal reports in a timely manner (IMF 2021, 26). The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 5 Budget Credibility World Bank support to improve budget preparation and execution did not enhance budget credibility. When the composition of expenditure varies considerably in relation to the original budget, the budget is no longer a useful statement of intent with regard to government policies. The variation was 16.7 percent in 2007, 14.5 percent in 2008, and 16.6 percent in 2009, which resulted in a D score for the PEFA indicator measuring this variation (PI-2: Composition of expenditure out-turn compared with original approved budget). To address this challenge, the World Bank supported planning and budgeting capacity at both the central and local levels through a mix of investment and policy lending. The Decentralized Planning and Finance Project (FY04–09) supported training to increase planning and budgeting capacity at the provincial and district levels. By the end of the project, management reported that all 128 districts had formally adopted the project-supported strategic and annual planning and budgeting approaches. IEG rated the overall outcome of the project as moderately satisfactory. The 52 National Decentralized Planning and Finance Program Project (FY10–15) financed training and technical assistance at the central and provincial levels to strengthen national systems in support of decentralized planning and finance and help districts prepare high-quality budgets. IEG rated the overall outcome of the project as satisfactory. To reduce deviations between budget plans and execution for priority sectors, PRSCs 3 and 4 included prior actions that required a minimum level of actual expenditures for priority sectors, while PRSC 5 had a prior action requiring the alignment of the expenditures with the medium-term expenditure framework. By the end of PRSCs 3 through 5, the share of actual expenditures for priority sectors was 62 percent, falling short of the 65 percent target. Although the alignment of expenditures with the framework was achieved for 2008 and 2009, the result was reversed by the end of PRSCs 3 through 5 in 2010. This World Bank support did not result in improved budget credibility, with the PEFA score remaining at D and presenting an even greater deterioration, as the variation was 13 percent in 2012, 27 percent in 2013, and 28 percent in 2014. This high deviation is explained, in part, by a legal framework that allows budgetary reallocations without the need for legislative approval and by a sizable donor-funded external component of the budget. However, the root causes are more likely to be associated with persistent weaknesses in planning and budgeting processes at the institutional level (PEFA 2016). By the end of the evaluation period, this PEFA score remained at D (PEFA 2020). Independent Evaluation Group World Bank Group    53 Internal and External Audit Function The World Bank supported internal and external audit functions through a mix of analytical and advisory support and investment and policy lending. The 2006 PEFA indicated that the government’s external audit function required significant attention, its internal controls were weak, and its adherence to international accounting standards was nonexistent (World Bank 2007, 17). To address these shortcomings, the PRSC series had prior actions requiring the adoption of International Financial Reporting Standards; the establishment of internal control units and an expansion of internal audits; and increases in the coverage of external control over the state budget and financial audit reports. To reinforce these prior actions, the National Decentralized Planning and Finance Program Project (FY10–15) sought to strengthen the internal and external control functions by enhancing the capacity to carry out audits. The World Bank also provided nonlending technical assistance (Introduction of Risk-Based Internal Audit FY12) to improve internal audit methodologies, specifically for rolling out a risk-based audit approach. World Bank support contributed to progress in the internal and external controls function at the central level. World Bank support contributed to im- proving the external audit function’s (Tribunal Administrativo) adherence to International Standards for Supreme Audit Institutions and to increasing au- dit coverage of public entities and the state budget, reaching 51.1 percent of the state budget by 2018, up from a 26 percent baseline in 2007 (PEFA 2020). These results were sustained over time: by 2021, the International Monetary Fund Fiscal Transparency Evaluation deemed the Tribunal Administrativo’s external audit function as good (IMF 2021), although there were still delays in the implementation of audit recommendations by public entities and weaknesses in the institutional arrangement for ensuring the Tribunal Ad- ministrativo’s independence from the central government (PEFA 2020). With respect to the internal audit, World Bank support contributed to increasing the share of central and provincial-level bodies with internal control units The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 5 from 25 percent in 2007 to 100 percent in 2011 (World Bank 2014e), and audited public entities were implementing recommendations issued by the Inspectorate General of Finance and the Internal Control Units (World Bank 2018f). By the end of the evaluation period, the PEFA score for internal audit coverage had increased from a B (PEFA 2006) to an A (PEFA 2020). Public Investment Management PIM became a high priority for both the World Bank and the government in the wake of the hidden debt crisis, which had demonstrated shortcomings in PIM. For much of the evaluation period, PIM was not a strategic priority for the World Bank’s operational work, even though it had been flagged in World Bank and International Monetary Fund analytical work as an area that required attention, given rising public investment (Dabla-Norris et al. 2011; World Bank 2014b). Deficiencies were documented in a 2015 PIM Assess- ment (box 5.1). PIM weaknesses were recognized in the World Bank’s 2014 Public Expenditure Review and the 2016 SCD (World Bank 2016c, 126). The 54 2015 PIM Assessment was not referred to as an analytical underpinning for the design of PRSCs 9 through 11. Neither the FY08–11 CPS nor the FY11–16 CPS made any reference to shortcomings in PIM, which only came to the forefront in the aftermath of the hidden debt crisis in 2016, when serious deficiencies in Mozambique’s PIM system were exposed. Improved PIM was not included as an explicit strategic objective until the FY17–21 CPF. Box 5.1. 2015 Public Investment Management Assessment Results In 2015, before the hidden debt crisis, the International Monetary Fund carried out an assessment of Mozambique’s public investment management planning and apprais- al systems and found that their quality was low. The assessments found evidence of weak financial planning for capital projects and an overall weak process for selecting and assessing projects, meaning that project appraisal and selection procedures did not ensure that projects were vetted and selected based on policy or efficiency crite- ria. In addition, the assessment found that the investment program was being execut- ed in the absence of a strategic capital or infrastructure development plan, which left the capital budget anchored only in broad policy documents. Project implementation and monitoring systems were also found to be weak. The assessment found that contracts were not always awarded based on competitive transparent biddings and that an integrated system for parallel monitoring of financial and physical progress of projects was missing. Source: IMF 2015. Independent Evaluation Group World Bank Group    55 Despite World Bank support to improve regulations and technical conditions for PIM, few improvements were achieved before the hidden debt crisis. The World Bank supported PIM capacity building through a mix of analytical work and investment and policy lending. With an overall IEG outcome rating of moderately satisfactory, the Integrated Growth Poles Project (FY13–20) trained 510 public officials on project planning, proposal evaluation, and implementation monitoring (World Bank 2021h). PRSCs 9 through 11 includ- ed prior actions to develop an appraisal and evaluation manual for public projects (PRSC 9, FY14), approve a public investment plan (PRSC 10, FY15), and mandate a cost-benefit analysis (PRSC 10, FY15) and viability stud- ies (PRSC 11, FY16) for all public projects above $50 million. As a result of these prior actions, at the end of the program, all public investment projects included in the government’s public investment plan were to have been appraised and evaluated, but the target was not met. The Implementation Completion and Results Report for PRSCs 9 through 11 acknowledged that capacity constraints were underestimated by the World Bank, leading to the inclusion of targets that turned out to be unachievable. In hindsight, the World Bank relied too heavily on DPOs that were not well suited for address- ing longer-term capacity issues. This pattern, whereby the World Bank relied too heavily on DPOs to support PIM, has been identified as a broader issue in IEG’s recent evaluation World Bank Support for Public Financial and Debt Management in IDA-Eligible Countries (World Bank 2021p). Once the World Bank halted budget support in 2016 after the hidden debt crisis, government demand for PIM support increased. In response, the World Bank launched an analytical and advisory program funded by the Department for International Development (UK) to strengthen public investment, debt, and fiscal risk management. This program supported a comprehensive PIM reform plan to establish an integrated system and guide capacity-building activities. With support from this program, the government The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 5 continued to develop tools and methodologies needed for the functioning of a PIM system and trained public officials in their use. In addition, the PIM system started to be linked with the budget planning cycle and the medium-term fiscal framework; by 2020, 19 investment projects had been formulated and appraised using program-supported methodologies (World Bank 2019d). Finally, with the resumption of budget support, the COVID-19 Response Project included a prior action requiring the approval of a PIM regulatory framework that established the rules and main stakeholders of the PIM system. Overall, World Bank support has contributed to significant advancements on paper, but there are de facto shortcomings. By 2020, there was a detailed manual for the identification, formulation, and evaluation of investments. However, the 2020 PEFA found that individual projects lacked the financial and economic analysis recommended by the manual (PEFA 2020). Likewise, the 2020 PEFA found no evidence of adequate project monitoring and controls to ensure the fiduciary integrity of the projects. Overall, the 2020 PEFA indicator measuring PIM quality (PI-11) rated Mozambique’s system a D, down from a baseline of D+ in 2016 (PEFA 2020; PEFA 2016). 56 Public Debt Management Before the hidden debt crisis, World Bank support to improve debt management was modest, despite evidence indicating that Mozambique had serious shortcomings. The 2008 Debt Management Performance Assessment (DeMPA) concluded that Mozambique did not meet the minimum requirements for any aspect of external borrowing, loan guarantees, debt strategy, debt administration, or debt reporting. Despite these findings, debt management did not become a strategic priority in Bank Group–supported strategies until the end of the evaluation period. The FY08–11 CPS indicated that Mozambique’s debt was sustainable because the country had received debt relief under the Heavily Indebted Poor Countries Initiative in 2001 and the Multilateral Debt Relief Initiative in 2006 (World Bank 2007). Debt management was also not considered a priority in the FY12–16 CPS,1 although PRSC 9 (FY14) and PRSC 10 (FY15) included prior actions requiring the approval of a Medium-Term Debt Management Strategy (MTDS) for the period 2012–15 and the implementation of an annual domestic borrowing plan based on the MTDS. PRSC 11 (FY16) included a prior action to support a new MTDS for the period 2015–18 and another prior action requiring the Ministry of Economy and Finance to create a Fiscal Risk Unit. The contribution of PRSCs 9 through 11 to improving debt management was negligible. There was no progress in the publication of debt reports or Independent Evaluation Group World Bank Group    57 compliance with annual domestic debt plans, and although the Ministry of Economy and Finance established a Fiscal Risk Unit, the unit did not produce adequate fiscal risks statements for the 2016 and 2017 budget laws (World Bank 2018f). The Fiscal Risk Unit remains in place, although some develop- ment partners interviewed by IEG were doubtful about the unit’s capacity to play a significant role in monitoring fiscal risks and influencing the govern- ment’s debt strategy. Only after the hidden debt crisis exposed the country’s debt management weaknesses did debt management and monitoring of fiscal risks become a strategic priority for the World Bank, with the FY17–21 CPF including explic- it debt management–related objectives. These objectives sought to improve debt sustainability by improving investment transparency and strengthening fiscal risk management (World Bank 2017b). To support the CPF’s objectives, the World Bank launched an analytical and advisory program designed to ad- dress important regulatory gaps for managing debt and fiscal risks stemming from SOEs. This program included an update to the 2008 DeMPA; the elab- oration of a Debt Management Reform Plan; and an analytical and advisory program to strengthen public investment, debt, and fiscal risk management (table 5.1). With this support, the government approved new regulations to strengthen the management of public debt and guarantees in December 2017 and increased its capacity to analyze debt sustainability and create a medium-term debt strategy. In 2018, the publication of fiscal risk statements resumed, with a revamped report prepared annually to inform budget prepa- ration. Although the practice was resumed, it only met minimum standards with respect to including an assessment of the overall financial performance or quasi-fiscal activities of the public corporations sector (IMF 2021). Although it has been slow, there has been tangible progress toward increased debt transparency. This slow progress was partly because the hidden debt situation made debt analysis and reporting on the full stock of debt and con- tingent liabilities a sensitive issue (World Bank 2019d). By 2019, there were The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 5 still significant shortcomings in transparency, with annual debt reports still not published regularly and only including information on central govern- ment direct and guaranteed debt (World Bank 2020b). With the resumption of budget support, the World Bank included a prior action to increase debt transparency by requiring the mandatory publication of annual reports with coverage of SOEs and liquefied natural gas debt. This prior action, which was one of three performance and policy actions under the Sustainable Devel- opment Finance Policy of the International Development Association (IDA), was found by IEG to be relevant for addressing shortcomings in transpar- ency, a key driver of debt distress in Mozambique (World Bank 2021b). This requirement was incorporated into the 2021 Public Financial Management Act, which now mandates the publication of annual debt reports covering the SOE sector. 58 Table 5.1. World Bank Debt Management Engagements in Mozambique Project Name Fiscal Year Debt Management Performance Assessment 2008 2008 Medium-Term Debt Management Strategy 2011 Medium-Term Debt Management Strategy 2011 PRSC 9 2014 Prior action: Council of Ministers has approved the Medium-Term Debt Management Strategy (2012–15). PRSC 10 2015 Prior action: Ministry of Economy and Finance has implemented the first annual domestic borrowing plan, prepared based on the Medi- um-Term Debt Management Strategy. PRSC 11 2016 Prior action: Ministry of Economy and Finance has prepared the recipi- ent’s Medium-Term Debt Management Strategy for 2015–18. Prior action: Ministry of Economy and Finance has created a fiscal risks unit to better manage fiscal risks. Debt Management Performance Assessment 2017 2017 Debt Management Reform Plan 2018 Support to Strengthen Public Investment, Risk, and Debt Management 2018 COVID-19 Response development policy operation 2021 Prior action: The government has improved debt transparency and fiscal risk management through mandatory annual publication of (i) annual debt reports with coverage of SOEs and liquefied natural gas debt published; (ii) the financial statements of the oil company (Em- presa Nacional de Hidrocarbonetos); and (iii) a credit risk assessment framework for SOEs from 2019 onward. Source: Independent Evaluation Group. Independent Evaluation Group World Bank Group    59 Note: PRSC = Poverty Reduction Support Credit; SOE = state-owned enterprise. For most of the evaluation period, World Bank support had a modest impact on improving debt management in Mozambique. Progress in improving debt management between 2008 and 2017 has been marginal and can be seen by comparing the evolution of DeMPA scores (appendix E). To some extent, this progress reflected a focus of World Bank support on technical and institu- tional capacity building. Although this focus may have been a necessary condition for improving outcomes, governance shortcomings also needed to be addressed to have impact in the field. Although these were generally identified as risks to relevant operations, mitigating these risks was not at the forefront of operational design (World Bank 2018f, 38). DPO prior actions focused on disclosure of medium-term debt strategies and debt and fiscal reports. More of a governance lens would have focused on establishing ex ante controls and borrowing limits for the issuance of debt, an area in which Mozambique remained weak throughout most of the evaluation period. Only toward the end of the evaluation period did the World Bank move to this approach, with the leveraging of a performance and policy action under IDA’s Sustainable Development Finance Policy, which required the adoption of a zero nonconcessional borrowing limit on external public and publicly guaran- teed debt for FY21. Finally, despite widely recognized synergies among bor- rowing and the quality of public investment, support for debt management was not systematically accompanied by efforts to improve PIM for most of the evaluation period. This lack of coordination in Mozambique is consistent with the findings from the recent IEG evaluation World Bank Support for Public Financial and Debt Management in IDA-Eligible Countries (World Bank 2021p), which found that synergies between different public financial and debt man- agement pillars remain underexploited in many IDA countries. State-Owned Enterprises Reform The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 5 Until 2014, support for SOE reform focused on improving business and oper- ations management at the enterprise level and facilitating compliance with the Extractive Industries Transparency Initiative (EITI) in the extractives sector. Although there was support for SOEs, this was not a priority in either the FY08–11 CPS or the FY12–16 CPS. Support was focused in the energy sector, whereby the World Bank sought to improve business and operations management through infrastructure rehabilitation and capacity strength- ening for the electricity company (Electricidade de Moçambique), the oil company (Empresa Nacional de Hidrocarbonetos), and the mining company (Empresa Moçambicana de Exploração Mineira). In addition, through a mix of investment and policy-based lending, the World Bank provided support for improving corporate governance, with the aim of facilitating compliance with the EITI. When fiscal risks from SOE operations became evident, the World Bank piv- oted to providing support for tightening the legal framework at the national level, but negligible results were achieved. Midway through the implemen- 60 tation of PRSC 9 (FY14), fiscal risks became evident when EMATUM, a gov- ernment-owned fishing company, issued publicly guaranteed bonds with a total value of $850 million. Concerns about transparency of fiscal risks were documented in the Implementation Status and Results Report for PRSC 9. In response to these concerns, PRSC 10 (FY15) sought to tighten the SOE legal framework, with a prior action requiring the approval of implementing regulations for the public enterprises law. The prior action was to provide the Ministry of Economy and Finance with the legal basis required to collect better information on SOEs, a potential source of fiscal risks. In addition, the DPO series (from PRSC 10) increased its focus on managing fiscal risk by strengthening debt and fiscal risk reporting and the scrutiny of public investment proposals. But these prior actions, which emphasized technical solutions, reflected an insufficient understanding of the political barriers to achieving full fiscal transparency. In this event, these measures were insuffi- cient, and overall, PRSCs 9 through 11 did not adequately identify or miti- gate risks to the macroeconomic framework stemming from nontransparent SOE borrowing. Moreover, the World Bank did not take strong remedial action when the $850 million state-guaranteed loan was disclosed, and it continued with the remainder of the DPO series with only a light refocus on SOEs. Overall development outcome and World Bank performance for PRSCs 9 through 11 were rated as unsatisfactory by IEG. Against the backdrop of the hidden debt crisis in 2016, SOE governance became a government and World Bank priority, and tangible progress was Independent Evaluation Group World Bank Group    61 achieved. In response to government demand, the World Bank launched an analytical and advisory program to strengthen public investment, debt, and fiscal risk management in August 2016. With support from this program, the government approved a new legal framework for SOEs in 2018 (law) and 2019 (regulations), which strengthened oversight, corporate governance, and performance management. The government also tightened control over SOE borrowing by requiring a more stringent approval process. With the resumption of budget support in 2020, the World Bank further advanced SOE transparency with the inclusion of prior actions requiring (i) the mandatory annual publication of the financial statements of the national hydrocarbons company (Empresa Nacional de Hidrocarbonetos), and (ii) the publication of a credit risk assessment framework for SOEs from 2019 onward. The latter requirement, which was also a performance and policy action under IDA’s Sustainable Development Finance Policy, has been incorporated into the 2021 Public Financial Management Act, which mandates the publication of annual fiscal risk statements that contain SOE credit risk reports. On balance, World Bank interventions were only effective when control of corruption and SOE reform became government priorities. This finding is consistent with what IEG found in State Your Business! An Evaluation of World Bank Group Support to the Reform of State-Owned Enterprises, FY08–18 (World Bank 2020l). This evaluation found that control of corruption is a country characteristic strongly associated with SOE reform success. In the case of Mozambique, there was low appetite to address corruption before the hidden debt crisis. In addition, many of the factors that the IEG evaluation found could mitigate the negative effects of weak control of corruption were not present—namely, client commitment to SOE reform and strong institutional capacity and coordination. Attention to the legal framework at the national level only emerged in earnest when fiscal risks mounted on the heels of the $850 million state-guaranteed EMATUM loan. This support at the national level achieved negligible results. Once the hidden debts were revealed, tan- gible progress was made as the appetite for increased control of corruption The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 5 increased and a compelling case for SOE reform was made. Decentralization World Bank support sought to alleviate impediments to successful decen- tralization. Mozambique’s second Action Plan for the Reduction of Abso- lute Poverty (PARPA II) 2006–09 conceived of decentralization as a means to improve service delivery and thereby contribute to poverty reduction. It was also seen as a means to improve accountability by bringing the govern- ment closer to the people. At the start of the evaluation period, the World Bank identified two challenges to decentralization (World Bank 2003, 2007). First, subnational entities lacked the capacity to perform basic government functions such as public financial management and revenue collection and administration. Second, intergovernmental fiscal transfers were inequitable across provinces, with some provinces (for example, Nampula and Zambezia) receiving lower fiscal transfers per capita than the rest of the country (World Bank 2016c, 4). To address these key challenges, the World Bank focused on 62 (i) establishing a policy framework for decentralization and intergovernmen- tal fiscal transfers, and (ii) building subnational capacity through support for basic government functions, such as public financial management and revenue collection and administration. The World Bank’s decentralization approach built on lessons learned from previous decentralization support in the country. The World Bank’s approach drew on four lessons that were reflected in the mix of investment and policy lending that supported decentralization during the evaluation period. The first lesson concerned the need to build subnational administrations’ capaci- ty as a precondition for more effective delivery of services. The second lesson involved the use of small infrastructure investments as a means for provid- ing hands-on training for subnational administrations. The third lesson was about the importance of simultaneously building upward and downward accountability approaches to strengthen the accountability of subnational administrations to citizens. The fourth lesson involved the importance of using a two-pronged approach to decentralization that could simultaneously focus on both policy progress and subnational capacity but could be flexible enough to be redirected to the latter when policy evolution was slow or came to a stop temporarily. The World Bank provided support for the articulation of a decentraliza- tion framework that, in the end, was not pursued by the government. As part of the project design, the Decentralized Planning and Finance Project (2003–09) financed an intergovernmental fiscal relations study that laid Independent Evaluation Group World Bank Group    63 out options for transferring responsibilities from the central and provincial levels to the district level. The study sought to inform the preparation of a decentralization policy framework, which would define the responsibilities of different levels of government, the intergovernmental fiscal architecture, and the relation between sectors and subnational administrations. In ad- dition, PRSCs 3 through 5 had a trigger foreshadowing the completion of a national decentralization strategy that would clarify the responsibilities of various levels of the state, the fiscal intergovernmental architecture, and the relation between the planning and execution on the part of the sectors and local administrations. When the government did not meet the timeline for the preparation of the national decentralization strategy, the World Bank dropped this trigger from PRSC 4. The World Bank justified this decision on the grounds that “political economy questions surrounding the issue [were] not yet well understood” (World Bank 2010b, 11). In the absence of government buy-in, the World Bank backed away from lending support but remained engaged through analytical and advisory services. During this period, the World Bank maintained a dialogue with the government on decentralization. This included a Policy Note on intergovern- mental fiscal transfers that documented the weaknesses in the fiscal transfer system and laid out options for improving its fairness, equity, and transpar- ency (Grandvoinnet et al. 2018). In 2018, a constitutional change brought about a new round of decentralization reforms, which were seen by the government, the World Bank, and development partners as key to the resto- ration of peace and stability in the country (World Bank 2020c, 2020g). This work led to a new government request for World Bank support. The World Bank responded with the National Urban Development and Decentralization Project (2020–25), which provides a mix of technical assistance and capaci- ty-building support to develop a legal and regulatory framework, clarify the roles and competencies of the different government levels for better local service delivery, and establish a coherent intergovernmental fiscal transfers The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 5 system through more transparent, needs-based, and predictable transfers. According to World Bank staff, the project was making “satisfactory” prog- ress toward the achievement of its objectives. Building Subnational Capacity Public Financial Management Early in the evaluation period, the World Bank provided considerable and generally effective investment lending support to strengthen subnational public financial management. Improving budget planning at the district, provincial, and municipal levels was one of the priorities in the FY08–11 CPS and was supported through three investment projects (World Bank 2007). The Public Sector Reform Project (2003–09) provided support for rolling out e-SISTAFE to 50 out of 128 districts though the achievement of the over- all project objective to help the government restructure public service for decentralized service delivery. This project was independently assessed as 64 moderately satisfactory (World Bank 2014d). The Decentralized Planning and Finance Project (2003–09) sought to improve the institutional performance of district administrations to plan and manage small infrastructure invest- ments in response to community demands. The project’s overall outcome rating was satisfactory (World Bank 2014a). The National Decentralized Planning and Finance Program (2010–15) sought to improve the capacity of local governments to manage public financial resources for district develop- ment in a participatory and transparent manner. At project closing, the over- all outcome rating was moderately satisfactory. Meanwhile, at the municipal level, the ProMaputo Program strengthened the institutional capacity and service delivery of the Maputo Municipal Council with an overall outcome rating of moderately satisfactory (World Bank 2020e). Despite generally positive project ratings, implementation of public finan- cial management reforms at the subnational level faced significant chal- lenges. In 2015, authorities acknowledged that there was a gap between subnational public financial management rules and actual practices and that, for this reason, the impact on service delivery had been minimal (World Bank 2014c). The Mozambique Public Financial Management for Results Program (2014–19) was the World Bank’s attempt to help bridge gaps in public financial management implementation and address service deliv- ery underperformance. According to interviews with World Bank staff, the Program-for-Results instrument was more effective for addressing imple- mentation gaps than for investment projects. IEG rated the project’s overall Independent Evaluation Group World Bank Group    65 outcome as satisfactory in light of evidence that the project had bridged gaps in public financial management implementation and improved service deliv- ery in the health and education sectors (World Bank 2020d). Subnational Revenue Mobilization Municipalities’ limited capacity to collect revenue affected the decentraliza- tion process in Mozambique. In 2005, taxes were paid on only 5 percent of Maputo’s properties because of the city’s lack of an up-to-date property ca- dastre (World Bank 2020e). Maputo also lacked the means to levy and collect other municipal taxes and fees. This hindered Maputo’s and other munici- palities’ ability to deliver essential public services. World Bank–supported projects contributed to tangible improvements in municipal revenue collection. Between 2008 and 2017, World Bank support for the ProMaputo Program resulted in increased municipal revenue through improved property tax management. More than 33,000 land property titles were issued, leading to a sixfold increase in the number of property taxpayers and a 281 percent increase in revenues (World Bank 2020e). Beyond Maputo, the World Bank provided support for local revenue enhancement to other mu- nicipalities through the Cities and Climate Change Project between 2012 and 2021. As a result, 20 municipalities implemented land registries that contrib- uted to increases in municipal own-source revenues and in the share of mu- nicipal income coming from own-source revenue (World Bank 2021a). Extractives As extractives rose in economic importance after the discovery of gas deposits in 2010, the World Bank responded with a program to promote transparent, inclusive, and effective management of the extractives sector. Exploration in the Rovuma Basin in 2010 confirmed the existence of large gas deposits. Given the potential game changer that these deposits repre- The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 5 sented, the government requested World Bank support to carry out gover- nance reforms of its mining and natural gas sectors to address the economic and social impacts of extractives development. In response, the FY12–16 CPS included improving transparency in the extractives sector as an objec- tive. This objective was to be achieved through compliance with the EITI and supported with a mix of analytical work and investment and policy lending (appendix F). The FY12–16 CPS envisaged support for (i) establishing a legal, policy, and contractual framework for attracting and securing large pri- vate investments; (ii) strengthening the government’s role as a supervisory authority, enforcer, and participant in the extractives sector; (iii) improving fiscal design to manage revenue volatility and managing and distributing revenues; (iv) helping to identify and implement policies to ensure that the extractives sector contributes to the country’s economic growth and socio- economic development; and (v) building government capacity and promot- ing policies to minimize the social and environmental costs associated with the development of the extractives sector (World Bank 2012a). 66 The World Bank was largely successful in supporting the establishment of a regulatory framework for managing the extractives sector and complying with transparency standards. The PRSC series and the Mozambique Mining and Gas Technical Assistance Project (2013–22) helped the government prepare and approve laws and regulations for the mining and hydrocarbons sectors (appendix F). This support contributed to the government becoming compliant with the EITI in 2012. By 2020, the mining and petroleum sectors’ legal frameworks were clear, public, and easily accessible, and they covered all stages of upstream exploration and extraction (IMF 2021). As of 2022, the Mozambique Mining and Gas Technical Assistance Project was making satis- factory progress toward the achievement of its objectives. In contrast, the World Bank’s contributions to improving governance in oth- er areas of the extractives sector have not had much success. The results of efforts to design and implement a fiscal rule for managing liquefied natural gas revenue volatility was modest to negligible; no fiscal rules or targets were officially adopted by the government. Likewise, policy dialogue propos- ing the use of a sovereign wealth fund to manage revenues from the ex- tractives sector has not produced concrete outcomes (World Bank 2020f, 40). Independent Evaluation Group World Bank Group    67 1  The fiscal years 2012–16 Country Partnership Strategy noted that the International Mone- tary Fund and the World Bank had upgraded Mozambique to a higher-capacity country with regard to nonconcessional borrowing capacity because of improvements to the country’s me- dium-term debt strategies and the government’s completion of an annual debt sustainability analysis (World Bank 2012a). The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 5 68 6 | Resilience to Climate Change Highlights The World Bank contributed to the development of an institution- al framework for strengthening climate resilience and improving disaster risk preparedness through strengthened hydrological and meteorological information services and increased financial pro- tection against disasters. World Bank support contributed to increased climate resilience in the transport, social protection, water and sanitation, agriculture, education, energy, and urban sectors. 69  The World Bank played an important role in identifying climate change as a major development challenge for Mozambique and making the case for climate-resilient policies. Before 2010, most of the government effort with respect to climate change was focused on response to and reconstruc- tion after extreme weather events. Through technical assistance, the World Bank supported a policy dialogue that identified the need for increased climate resilience. This culminated in a 2010 study called Economics of Adaptation to Climate Change, which estimated the potential impacts of climate change on aggregate economic performance and found that, in the worst-case scenario, Mozambique’s GDP could fall between 4 percent and 14 percent relative to baseline growth between 2040 and 2050 if adaptation strategies were not implemented (World Bank 2010b). The study was crucial to not only demonstrate the impact of these events but also show the gov- ernment that ex post reconstruction was not cost-effective and led to delays in response. The study successfully made the financial and fiscal case for increased climate resilience. The World Bank contributed to the development and implementation of a strategy for building climate resilience. At the beginning of the evaluation The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 6 period, the government lacked a strategy for building climate resilience. With support from the World Bank, the government prepared a Strategic Program for Climate Resilience, which established an ambitious three-pillar program of (i) policy and institutional reforms; (ii) pilot investments to enhance climate resilience in key sectors; and (iii) technical assistance for knowledge management, capacity building, and studies. To underpin the government’s approach, the World Bank delivered a mix of analytical work and investment and policy lending that contributed to (i) advancing a policy and institutional framework for climate resilience and (ii) strengthening climate resilience in key sectors (appendix G). Institutional Framework World Bank support contributed to the development of an institutional framework for addressing climate change in Mozambique. At the beginning of the evaluation period, there was no framework clarifying institutional mandates for coordinating efforts to enhance climate resilience. To ad- 70 dress this shortcoming, the World Bank deployed the climate change DPO series (2013–16),1 which was the vehicle for implementing the policy re- forms indicated in pillar I of Mozambique’s Strategic Program for Climate Resilience. More specifically, the DPO series sought to strengthen nation- al policy and institutional frameworks for climate resilience by including prior actions that required (i) the enactment of a disaster management law; (ii) the approval of a National Climate Change Adaptation and Mitigation Strategy; (iii) the establishment of a climate change coordination unit; and (iv) the elaboration of a national monitoring and evaluation framework for climate change and disaster risk management. In 2018, IEG rated the over- all outcome rating for the DPO series as satisfactory because, by the end of the series, a disaster management law had been approved. Local adaptation plans, which were part of the approved National Climate Change Adaptation and Mitigation Strategy, had been elaborated and were being monitored by the climate change coordination unit. In addition, the target of having 13 sectors reporting through the national monitoring and evaluation framework had been met. Finally, the government had integrated climate change into Mozambique’s Five-Year Plan (2015–19), which was approved by Parliament in April 2015 (World Bank 2019b). World Bank support contributed to the establishment of a framework for increasing financial protection against disasters. For most of the evaluation period, an annual budget allocation for the annual contingency plan was the only ex ante financial instrument for disaster preparedness and response. Independent Evaluation Group World Bank Group    71 This allocation was unpredictable and limited in size, adequate for respond- ing only to small- to medium-size events. For emergency response to larger events and postdisaster recovery and reconstruction, the government relied on ex post instruments such as ad hoc budget reallocations and mobilization of donor support. Both were slow to materialize and insufficient to cover postdisaster recovery needs. To address this challenge, the government ap- proved the creation of the national disaster management fund (DMF) called Fundo de Gestão de Calamidades in October 2017. With the World Bank– supported Mozambique Disaster Risk Management and Resilience Program (2019–24), the government adopted comprehensive regulations to govern the operations of the DMF. The program also supported the DMF’s recurrent capitalization. As of December 2021, the Disaster Risk Management and Resilience Program was making moderately satisfactory progress to achieve its objectives and had contributed to the DMF beyond its annual $10 million capitalization target, with $14.6 million and $9 million from the govern- ment and the World Bank, respectively (World Bank 2021e). The program had also strengthened the government’s capacity to enable the placement of a sovereign parametric catastrophe insurance product. As a first step, the World Bank supported the preparation of a National Disaster Risk Financing Strategy, which detailed the strategic priorities of the Ministry of Economy and Finance for financing disaster response. In June 2022, the strategy was approved by the Council of Ministers, and a contract of insurance coverage against cyclone winds was expected to be signed in 2022 (World Bank 2021e). World Bank support contributed to strengthening hydrological and meteoro- logical information services needed for disaster risk preparedness. In 2011, the Strategic Program for Climate Resilience identified Mozambique’s hydro- meteorological system as requiring urgent improvement. To address this need, the climate change DPO series (2013–15) included prior actions to (i) advance the approval of the National Meteorological Institute’s strategic and organizational structure; (ii) set up three regional meteorological centers to create regional forecasts; and (iii) establish procedures for the management The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 6 and exchange of weather and hydrological data between the National Mete- orology Institute (INAM) and the National Directorate of Water Resources Management (DNGRH). In parallel, the Transforming Hydro-Meteorological Services Project (2013–19), which was given an overall outcome rating of moderately satisfactory by IEG, strengthened the core hydrometeorologi- cal network by increasing the share of real-time reporting from hydrologi- cal monitoring stations. Although considerable progress has been made, a recent World Bank report scored the maturity level of INAM and DNGRH as only meeting an essential level (World Bank 2021n). Although progress has been made, Mozambique’s disaster preparedness still faces considerable challenges. The Transforming Hydro-Meteorological Services Project fell short of delivering flood forecasts to target populations (World Bank 2021g). At the local level, timely distribution of this informa- tion to intended beneficiaries was a major challenge, as the capacity of local committees for disaster risk management was very low. At the national level, there were collaboration gaps between the National Institute of Disaster 72 Management, INAM, and DNGRH. To address these challenges, the Disaster Risk Management and Resilience Program (2019–24) included disburse- ment-linked indicators for creating and strengthening local committees for disaster risk management and addressing coordination issues. According to the most recent monitoring report, the creation and strengthening of lo- cal committees for disaster risk management was progressing as planned, albeit with some delays while coordination issues were addressed through the elaboration of a memorandum of understanding between the National Institute of Disaster Management, INAM, and DNGRH (World Bank 2021e).2 World Bank staff assessed the project as making moderately satisfactory progress toward achieving its objectives. Building Climate Resilience in Sectors Beyond advancing the national policy and institutional framework for climate resilience, the World Bank supported strengthening climate resilience in key sectors. The FY12–16 CPS recognized climate change mitigation and adaptation as a new and important business line. It supported a range of projects in water resources development, roads and bridges management and maintenance, cities and climate change, social protection, and public works. The conservation of natural ecosystems and biodiversity was also included as a building block toward better resilience in the face of natural disasters and livelihood vulnerability. The FY17–21 CPF included support for improving water resources management and planning, investing in climate-resilient Independent Evaluation Group World Bank Group    73 measures at the local level, improving the management and protection of coastal zones, and integrating climate risk assessments into planning and infrastructure development. The World Bank’s experience with these sector-focused interventions highlights a few cross-cutting insights on (i) the importance of allowing for innovative solutions beyond conventional approaches; (ii) the challenges of building sustainable institutional capacity; and (iii) the importance of engaging with development partners. Moving beyond Conventional Approaches Building resilience in the urban sector required nature-based innovative solutions to control floods. The Cities and Climate Change Project (2012–20) piloted the implementation of innovative nature-based (green) solutions for reducing floods and erosion, which ended up costing far less than the tradi- tional (physical infrastructure–based) solutions. These are interventions that harness the ability of natural or nature-based features—such as bioswales, wetlands, and mangroves—to meet development goals such as the reduction of natural hazards, while simultaneously providing environmental, econom- ic, and social benefits. This is the case of the World Bank’s support to Beira, the city in Mozambique most threatened by climate change because of its exposed coastal location (low-lying land and high tidal range) and its vulner- able infrastructure and population. In Beira, the Cities and Climate Change Project (2012–20) developed and financed drainage plans that included both rehabilitation of drainage canals and the implementation of natural drainage improvements. As a result, the project exceeded its target of 1,140 hectares with reduced flooding or erosion and 667,000 people benefiting from im- proved urban living conditions. These investments paid off during two extreme rainfalls in early 2019. Ar- eas that benefited from drainage rehabilitation under the project suffered little to no flood damage compared with other areas. IEG rated the project’s overall outcome as satisfactory (World Bank 2021a), but its sustainability remains contingent on the continued commitment of a municipal govern- The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 6 ment that does not have a sufficient dedicated revenue stream to finance the operation and maintenance of the drainage and green infrastructure. Sup- port for Beira’s drainage rehabilitation continues through the Cyclone Idai and Kenneth Emergency Recovery and Resilience Project (2019–24), which is also addressing key gaps in Beira’s coastal protection system. World Bank staff reported that the project was making satisfactory progress toward the achievement of its objectives as of October 2021 (World Bank 2021d). Beira represents one of the first nature-based urban flood management interven- tions supported by the World Bank. This experience provided several lessons related to flexibility in design, preservation of habitat, and appropriate oper- ation and maintenance of green infrastructures (box 6.1). 74 Box 6.1. Early Lessons from Nature-Based Flood Protection in Beira The nature-based solutions piloted in Beira serve as an important source of experi- ence and lessons. Early lessons highlight the importance of the following: » Flexibility: Nature-based solutions may present a larger variety of options and nonstandard practices. In an urban context, subjective preferences such as aes- thetics are also relevant. Adaptation of project components may come up more frequently during the feasibility and design phases, requiring unconventional solutions contrary to traditional engineering measures. A certain degree of unpre- dictability may have to be accepted. All of this requires flexibility and close and regular communication between the main stakeholders involved. » Habitat preservation: Clear guidance has to be provided to the construction com- pany and supervisors to prioritize the preservation of the habitat, with flexibility in regard to some of the provided construction targets. The use of heavy equipment in particular might need to be limited in favor of manual labor. » Environmentally sound operation and maintenance: Operation and maintenance of green infrastructures should be done by a competent entity. Especially in an urban context, the management of public green spaces requires funding and ex- pertise. The operation must ensure good environmental management practices, with regular ecosystem monitoring. Source: World Bank 2020l. Independent Evaluation Group World Bank Group    75 Climate-resilient urban projects are piloting risk mapping. The ongoing Maputo Urban Transformation Project is supporting the implementation of green infrastructure with a citywide structuring plan that is piloting the use of risk mapping to address the serious challenge of improving the risk resilience of the large share of the population living in informal settlements located in flood-prone areas. As in Beira, many of Maputo’s and other coastal cities’ neighborhoods have grown in an unplanned manner and can be characterized by high population densities, inadequate residential areas and infrastructure, lack of adequate water supplies and waste- and stormwater drainage systems, and high poverty rates, which make them particularly vulnerable to extreme weather events. The World Bank, with the ongoing National Urban Develop- ment and Decentralization Project, is working with 22 municipalities—out of 53 in Mozambique—to strengthen their capacities and prepare urban de- velopment plans and guidelines, including a manual to incorporate climate resilience in urban planning. These plans and guidelines have been subject to some delays due to local capacity challenges and staff shortages, as well as the extensive review and validation process required from national authorities. The transport sector piloted innovative approaches to mitigate floods, such as risk maps and a geospatial climate resilience tool. The 2010 report Mozambique—Making Transport Climate Resilient was the World Bank’s first report that focused on building resilience at the sector level (World Bank 2010c). It provided a detailed assessment of the impact of climate change on roads infrastructure and of different adaptation options. It concluded with specific recommendations on the need to review and revise road-related design parameters, guidelines, and manuals to reflect climate change adaptation requirements, reduce the risk of total failure and consequential damage, and ensure that an appropriate maintenance strategy is implemented. The report’s recommendations were gradually piloted and implemented through the Roads and Bridges Management and Maintenance The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 6 Program—Phase II, which originally started as a traditional project to help improve the maintenance, rehabilitation, and upgrading of the road network. In response to recurring disasters, however, the additional finance project included components to prepare more resilient designs and construction standards for roads, pilots of the improved road designs, and the development of improved local arrangements for the roads’ maintenance. An important innovation was that, contrary to conventional engineering, it proved more cost-effective not to raise roads high above the flood lines but to allow the water to flow over the roads, rather than damming up behind them. This has some trade-offs in terms of short-term road connectivity, but it brings cost savings in the long term. Based on the project’s experience, the World Bank issued a brochure called “Road Water Management for Resil- ience in Mozambique” (World Bank Group et al., n.d.), which indicated that although water can be an important cause of damage to roads, roads are in turn a major cause of local flooding, waterlogging, and erosion. However, such problems can be turned around by designing roads to become instru- 76 ments for climate change resilience and water management by routing water to storage ponds or recharge areas that help retain water in dry riverbeds and ensure systematic spreading of floodwater. Similarly, the Integrated Feeder Road Development Project is piloting a geospatial climate resilience tool to guide the repair of climate-affected roads. The project is focused on road access in rural areas in support of the livelihoods of local communities and an immediate response to crises. Although progress to date has been satis- factory, the sector is facing a long-term challenge with the sustainability of maintenance funding, which has been inadequate because of the continuing budgetary shortfalls. Building Sustainable Institutional Climate- Resilient Capacity Building climate resilience in Mozambique requires strengthening the country’s hydrological and meteorological services. As highlighted in the government’s 2011 Strategic Program for Climate Resilience, developing an effective integrated hydrological, meteorological, and early-warning system was one of the key interventions needed to reduce climate risk. The World Bank supported this priority with multiple projects. The 2011 Programmatic Support to Disaster Risk Management Phase I Project focused on studies of reinforcing the hydro-climatological network and improving the weather ra- dar network. The 2013 Climate Resilience: Transforming Hydro-Meteorolog- ical Services Project aimed at strengthening hydrological and meteorological Independent Evaluation Group World Bank Group    77 information services by providing reliable and timely climate information to local communities. As a complementary intervention, the 2013 Climate Change Development Policy Operations I and II included prior actions for the strengthening of INAM, including the provision of daily forecasts to farmers and fishermen. However, the achievements were modest: while the daily meteorological forecasts were available to all (through daily web-portal updates, radio and television reports, or Short Message Service broadcasts), they were not yet effectively delivered to farmers and fishermen. A major challenge was the large share (about 35 percent) of poor, isolated farmers who do not have a radio or phone, which made them difficult for even local community leaders to reach. Notwithstanding consistent World Bank support, the hydrometeorological services and early-warning system is a work in progress. Although the World Bank has consistently supported the strengthening of hydrometeorological services and early-warning systems through multiple projects over the past decade, some improvements were achieved, but the system is still a work in progress. A 2021 stocktaking report noted that INAM’s capacity in terms of dissemination of early warnings for extreme weather has improved signifi- cantly (World Bank 2021m), but the maturity of INAM’s and DNGRH’s basic observation and monitoring networks was at only 32 percent and 53 percent, respectively. These scores suggest that the hydrometeorological system’s ability to adequately mitigate the country’s climate risks is at less than half of what it should be, mainly due to limitations in capacity and funding for operations and maintenance. To address these specific challenges and limitations, the ongoing Disaster Risk Management and Resilience Program (2019–24) included several relevant disbursement-linked indicators related to funding, local capacity, and the establishment of an early-warning system (box 6.2). The latest Implementation Status and Results Report (December 2021) rated the overall implementation progress of this program as moder- ately unsatisfactory. The experience with the hydrometeorological services and early-warning The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 6 system illustrates that the two main challenges in building sustainable insti- tutional capacity in Mozambique are implementation capacity and financial sustainability. A major challenge in Mozambique is the capacity limitations of implementing agencies, due to inadequate training and budgets, low wages, and high staff turnover. The unanimous and most keenly felt request emerging from IEG’s interviews with government counterparts relates to the need for greater attention and support for the transfer of knowledge and skills from project-funded consultants and project implementation units to the agencies’ regular staff. This was seen not only as a matter of funding but also as a need for greater attention to the involvement of regular agency staff at all stages and levels of project preparation and implementation to ensure their institutional memory and “learning by doing,” not just at the handover stage. In addition, a continuing challenge relates to the sustain- ability of funding for the operation and maintenance of infrastructure. Thus, although the quality and performance of the hydrometeorological system had been greatly improved over the past decade, it has recently deteriorat- ed because of inadequate funding, mainly due to the economic fallout from 78 the hidden debt crisis, the impact of the tropical cyclones in 2019, and the COVID-19 pandemic in 2020. The Disaster Risk Management and Resilience Program (2019–24) is addressing this challenge by supporting funding for the DMF, which has been progressing well. But the specific contribution to the hydrometeorological and early-warning systems remains to be seen, depending on the details in the memorandums of understanding between the National Institute of Disaster Management, INAM, and DNGRH that are under preparation.  isbursement-Linked Indicators Included in the Disaster Risk Box 6.2. D Management and Resilience Program Disbursement-linked indicator (DLI) 1, for the operationalization and funding of the di- saster management fund, involved the commitment of 0.1 percent of the annual state budget to the fund. As of mid-2021, this DLI had been progressing well, with more than $23 million having been funded, against a target of $10 million. However, since the disaster management fund is managed by the National Institute of Disaster Manage- ment, the extent to which these funds may be used to support the early-warning system (jointly managed by the Mozambique National Meteorology Institute and the National Directorate of Water Resources Management) is not clear. DLI 3 was for the creation and strengthening of functional local disaster risk manage- ment committees, a key instrument for managing and reaching the affected popu- Independent Evaluation Group World Bank Group    79 lation in case of disaster. As of mid-2021, progress in this activity was delayed due to unresolved and ongoing discussions about the role and structure of the committees, which a consultant was working to resolve. DLI 4 was for the establishment of integrated information and early-warning systems, working from a 2018 baseline of “no functional integrated flood and cyclone early- warning system.” As of mid-2021, the development of relevant memorandums of understanding between the National Institute of Disaster Management, Mozambique National Meteorology Institute, and National Directorate of Water Resources Management was underway and expected to be ready by early 2022. Source: World Bank 2021j. Working with Development Partners to Adopt Resilient Standards and Design In the education sector, the World Bank has been promoting a “safer schools” program with the involvement of the government and the donors. In Mozambique, improving resilience in schools is critical because their structural weaknesses and exposure to disasters resulted in approximately 550 classrooms being destroyed each year by cyclones and floods. Since about 2010, the World Bank initiated a “safer schools” dialogue and program that included risk assessments and the development of structurally resilient school building designs and standards in collaboration with UN-Habitat. The World Bank was able to promote these new standards through the Education Sector Support Fund (Fundo de Apoio ao Sector de Educação [FASE]) that pools all donors’ financial assistance, which the World Bank helped establish and where it remains closely engaged as the supervising entity to reduce donor concerns about corruption risks. The effectiveness of resilient school construction techniques was demon- strated during Cyclones Idai and Kenneth. The 2015 Emergency Resilient Re- The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 6 covery Project was initially designed to support the piloting of new resilient construction techniques and standards for the rehabilitation of 433 damaged conventional classrooms and the construction of 1,038 new, nonconvention- al (self-constructed by the communities) classrooms. The classrooms were designed to withstand the locally mapped cyclone winds and ground shaking and include rainwater-harvesting systems. Due to significant increases in the construction costs, the final project targets were reduced to the rehabil- itation of 372 conventional classrooms and the construction of 257 noncon- ventional classrooms. This approach received a major boost when the pilot “resilient schools” were not damaged by Cyclones Idai and Kenneth. The World Bank is currently supporting education infrastructure with the on- going Disaster Risk Management and Resilience Program (2019–24), which includes disbursement indicators linked to the climate resilience of educa- tion infrastructure.3 The education sector’s experience with the implementation of building resil- iency standards points to two key enabling factors for success: pooled funds 80 and work quality. The World Bank has had mixed experiences with pooled funds arrangements in Mozambique, which have been created in several sec- tors as a common platform for donor funding and project implementation. This approach worked well for the education sector, where the World Bank was a leader in the establishment of FASE with eight donor partners and remained involved as the supervising entity with oversight over fiduciary matters and safeguards—which helped reduce donor concerns about cor- ruption risks. The close relationships from working together with donors on FASE also helped expand the implementation of resiliency standards to all newly constructed schools. In the transport sector, to the contrary, a similar pooled funds arrangement had to be dismantled because many donors insist- ed on requiring their own policies (for example, procurement, environment, and social) to be followed, leading to extensive delays in project implemen- tation. In addition, the quality of the World Bank’s education program and the program’s ownership by the government enabled the World Bank to play a leadership role with donor partners, given the World Bank’s role as the supervising entity of FASE, its provision of training and support, and its fidu- ciary role, which gave trust to the other donors and reduced the perception of governance risks. A review of the extent to which lessons from the evaluated climate resilience portfolio informed World Bank support of downstream projects suggests that the Implementation Completion and Results Report and the Implementation Completion and Results Report Review processes constitute a major pillar Independent Evaluation Group World Bank Group    81 of the World Bank’s learning process. About half of the lessons identified in Implementation Completion and Results Report Reviews from upstream projects were referenced in the design of downstream projects. For example, at the national policy and institutions level, the 2014 Climate Change DPO II incorporated the three lessons from the earlier 2013 Climate Change DPO I in terms of strengthening intersectoral coordination, mainstreaming climate resilience into sector strategies, and incorporating strategic and targeted technical assistance. Another example was the Forest Investment Project, which drew on lessons reflected from the earlier Mozambique Conservation Areas for Biodiversity and Development Project, which highlighted the ap- propriateness of landscape approaches for ensuring compatibility of conser- vation and livelihood efforts and the need to engage local communities and address land tenure security. This review also found that a few important lessons had been learned from experience with upstream projects, in addition to those recorded in the Implementation Completion and Results Reports and the Implementation Completion and Results Report Reviews. For example, the Disaster Risk Management and Resilience Program specifically refers to the experience of the upstream Transforming Hydro-Meteorological Services Project when it recognizes that institutional coordination and information dissemination are more important than equipment for the effectiveness of early-warning systems. IEG had also identified this lesson through interviews with project teams. Similarly, the program documents for the Integrated Feeder Road Development Project referenced the earlier Roads and Bridges Management and Maintenance Program, finding that traditional road-planning approach- es do not account for the benefits of building climate resilience and often lead to suboptimal investment decisions. Again, based on IEG interviews, this learning was directly and effectively transmitted within and between project teams. Overall, about two-thirds of the lessons reflected in down- stream project designs were derived from the upstream projects and expe- rience, and the remainder came through other channels, such as within and between project teams and from the World Bank’s broader experience. The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 6 82 1  The climate change DPO series was implemented between 2012 and 2016 and originally included three series. DPO series 1 and 2 were disbursed between 2013 and 2015 on the com- pletion of policy and institutional reforms carried out by the government between 2012 and 2013 and between 2013 and 2015. Additional reforms were implemented between 2015 and 2016 for the DPO series 3, which was dropped by the end of 2016 because of the cancellation of budget support after the disclosure of the hidden debts. 2  The memorandum of understanding between the National Institute of Disaster Management, Mozambique National Meteorology Institute, and National Directorate of Water Resourc- es Management was elevated to a regulation through Decree n.27/2022, which adopted the regulations for the operationalization of the integrated platform for the dissemination and communication of flood and cyclones early-warning information to end users, including the local committees for disaster risk management. 3  With support from the Mozambique Disaster Risk Management and Resilience Program, the government enacted a Ministerial Diploma (n.122/2021) adopting technical norms for climate-resilient education infrastructure in October 2021. These norms are guiding the resil- ient reconstruction of 3,000 classrooms under the project. Independent Evaluation Group World Bank Group    83 7 | Findings and Lessons The World Bank supported Mozambique’s development objectives by strengthening the country’s agricultural productivity, access to basic ser- vices, governance, and climate resilience. Agriculture, the sector in which most of the country’s poor people are employed, has low productivity and pro- vides low wages. Combined with unequal access to basic services (education, health, transport, and electricity), these factors have contributed to low stan- dards of living. At the start of the evaluation period, the World Bank recognized the need to strengthen governance to sustain economic growth and reduce poverty. By the end of the evaluation period, the World Bank also explicitly acknowledged that weak governance drove fragility and required attention. These major areas of the World Bank’s support were relevant to Mozambique’s development challenges. The World Bank identified constraints to agricultural productivity and climate resilience at the beginning of the evaluation period, and its support to these areas became progressively broader. The World Bank also improved targeted support to regions to improve service access disparities. The World Bank’s support for good governance was constant through the evaluation period, though the nature of this support changed along with changes to the country context, such as the discovery of gas deposits in 2010 and the uncovering of the hidden debt crisis in 2016. The World Bank’s support to improve access to basic services and climate change resilience in Mozambique shows some positive and concrete results. The World Bank positively contributed to increasing access to basic ser- vices in rural areas during the latest strategy period. With respect to climate change resilience, the World Bank played a key role in identifying climate change as a major development challenge and supported the development of a strategic approach for building climate resilience. World Bank support has contributed to the development of an institutional framework to strengthen climate resilience and improve disaster risk preparedness through strength- ened hydrological and meteorological information services and increased 84   financial protection against disasters. In addition, the World Bank’s support has increased climate resilience in the transport, social protection, water and sanitation, education, agriculture, energy, and urban sectors. Agricultural productivity results were disappointing for the most part, de- spite considerable World Bank support. The World Bank’s earlier support to agricultural productivity did not achieve its intended purpose, even though later engagements seem promising thanks to a greater focus on market access, stronger linkages between lead farmers and small-scale traditional farmers, and more attention to technology use by farmers. World Bank support to improve governance in Mozambique focused on five areas: (i) public financial management (central government); (ii) public debt management; (iii) SOE reform; (iv) decentralization; and (v) transparent and effective management of extractives. The results achieved under these five areas are discussed in the following paragraphs. The World Bank contributed to improved public financial management by supporting an increase in the coverage of the financial management infor- mation systems and strengthening internal and external control functions at the central level. However, World Bank support for budget preparation and execution did not enhance budget credibility. Despite clear weaknesses in PIM, it was only in the wake of the hidden debt crisis in 2016 that the World Bank made concerted efforts to intensify support. Despite progress “on pa- per,” institutionalization of PIM reforms is lagging. Independent Evaluation Group World Bank Group    85 The World Bank had a modest impact on improving debt management and advancing SOE reform. Support was focused on building technical and in- stitutional capacity but did not adequately take into account the context of weak governance. By and large, debt management and SOE challenges were seen as problems that could be addressed through technical and institution- al capacity building. Although this may have been a necessary condition to improve outcomes, underlying governance shortcomings also needed to be addressed. On balance, once the hidden debts were revealed, tangible prog- ress was made as the appetite for increased control of corruption increased and a compelling case for SOE reform and debt management was made. The World Bank contributed to increased subnational capacity but was not effective at supporting the establishment of a coherent decentralization pol- icy framework. Political economy constraints rendered World Bank support for decentralization ineffective. Implementation of public financial manage- ment reforms at the subnational level faced significant challenges, but many of these were addressed successfully using Program-for-Results financing. World Bank–supported projects contributed to tangible improvements in municipal revenue collection. World Bank support helped improve governance in the extractives sector, but major challenges remain. The World Bank contributed to the establish- ment of a regulatory framework for managing the extractives sector and complying with transparency standards. However, World Bank support for the implementation of a fiscal rule and sovereign wealth fund for managing revenues from the extractives sector did not lead to tangible outcomes. This evaluation identifies the following lessons to guide future World Bank engagement in Mozambique and other countries facing similar development challenges: 1. In contexts characterized by corruption and state institutions being run for the benefit of high-status groups, technical solutions to public finan- The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 7 cial and debt management are unlikely to achieve desired results unless governance constraints are also confronted. In Mozambique, this was the case with support to improve debt management where World Bank support focused only on technical and institutional capacity and was not sufficiently well adapted to reflect the underlying political economy and associated risks. Likewise, World Bank support for PIM was largely techni- cal, with insufficient attention given to the implementation of risk-based approaches to identify and analyze corruption risks throughout the in- vestment cycle. Although progress was achieved “on paper,” reforms often fell short in practice. 2. Core diagnostics are essential to inform reform priorities but require deliberate and coordinated follow-up across instruments. Although the World Bank undertook several public financial and debt management diagnostics, it did not use the findings in a timely manner to set reform priorities and inform its work program. This was most noteworthy with respect to the 2008 DeMPA findings, which flagged serious shortcomings 86 in debt reporting and recording. However, little attention was given to address these shortcomings until the hidden debt crisis, including through prior actions in the subsequent programmatic series of DPOs. Mozambique would also have benefited from an early and more systematic assessment of weaknesses in PIM, which, alongside the DeMPA, could have identified some of the weaknesses that contributed to the hidden debt crisis. 3. The quality and impact of World Bank support for public financial and debt management can be enhanced by improving internal World Bank coordination and prioritization. This lesson aligns with the findings from the recent IEG evaluation World Bank Support for Public Financial and Debt Management in IDA-Eligible Countries (World Bank 2021p), which found that synergies between different public financial and debt manage- ment pillars remain underexploited in many IDA countries. In the case of Mozambique, the World Bank provided significant support for upstream aspects of debt management (for example, preparation of debt manage- ment strategies), with only late attention for downstream aspects (debt reporting and recording, cost and risk analysis, and debt processes and procedures). Moreover, for most of the evaluation period, support for debt management was not systematically accompanied by efforts to improve PIM, despite widely recognized synergies among borrowing and the qual- ity of public investment. As a result, there were missed opportunities to enhance the growth and development impact of development spending and debt-financed public investment. Independent Evaluation Group World Bank Group    87 4. The effectiveness of extension services in agricultural productivity proj- ects in Mozambique, where there is a large proportion of smallholders, requires greater attention to the adequacy of staffing. Extension services play a critical role in increasing agricultural productivity in Mozambique, yet such support risks being undermined by staffing shortfalls, as shown for similar approaches in Malawi and Tanzania (Ragasa 2019; Ragasa and Mazunda 2018). It is advisable to ensure that the extension services sup- ported by the World Bank projects are properly staffed. 5. In situations where women dominate a disadvantaged group, such as in subsistence farming, sector-based support (for example, to enhance agri- cultural productivity) requires gender considerations to be fully integrated into strategies and projects. In Mozambique, women are particularly dis- advantaged in benefiting from extension services, with fewer than one- third of women being reached by such services (USAID 2018). Support to agricultural productivity can be effective only if gender is front and center in the approach, including by collecting sex-disaggregated data. 6. Support for climate resilience can be effectively enhanced through the use of credible analytics to persuade policy makers about the costs of inaction. Persuading policy makers to pursue climate resilience policies can be chal- lenging because the costs of implementing such policies are real, while benefits are uncertain. Before 2010, most of the government effort with respect to climate change was focused on response to and reconstruction after extreme weather events. The World Bank analysis was crucial to making the financial and fiscal case for investing in increased climate re- silience by demonstrating the impact of extreme weather events, calculat- ing the cost of adaptation needs, and convincing government authorities that ex post reconstruction was not cost-effective. The World Bank Group in Mozambique, Fiscal Years 2008–21 Chapter 7 88 Bibliography Adriano, Antonio, Cassiano Soda Chipembe, Paulo Alberto Covele, Vasco Molini, Andrea Heloise Nieves, Antonio Nucifora, and Thomas Pave Sohnesen. 2012. Mozambique Then and Now: An Atlas of Socio-Economic Statistics—1997–2007. Washington, DC: World Bank Group. Baez, Javier E., German Caruso, and Chiyu Niu. 2018. “Extreme Weather and Poverty Risk: Evidence from Multiple Shocks in Mozambique.” Policy Research Working Paper 8667, World Bank, Washington, DC. Briere, Christophe, João de Lima Rego, Andreas Burzel, Conceição Leite, Dana Stuparu, Maaike Maarse, and Mark de Bel. 2018. Multi-hazard Risk Assessment for the Schools Sector in Mozambique. Washington, DC: World Bank Group. Dabla-Norris, Era, Jim Brumby, Annette Kyobe, Zac Mills, and Chris Papageorgiou. 2011. “Investing in Public Investment: An Index of Public Investment Efficien- cy.” IMF Working Paper 11/37, International Monetary Fund, Washington, DC. Eckstein, David, Vera Kunzel, and Laura Shafer. 2021. Global Climate Risk Index 2021. Who Suffers Most from Extreme Weather Events? Weather-Related Loss Events in 2019 and 2000–19. Berlin: Germanwatch. Grandvoinnet, Helene Marie, Bob Searle, Dionísio Nombora, Shomikho Raha, and Independent Evaluation Group World Bank Group    89 Sally Tolbert. 2018. Horizontal Funds Distribution to Subnational Government in Mozambique: Funding the Districts and Municipalities—Policy Note. Washington, DC: World Bank. IFC (International Finance Corporation). 2020. Mozambique Country Private Sector Diagnostic. Washington, DC: IFC. IMF (International Monetary Fund). 2015. Public Investment Management Assess- ment (PIMA). Washington, DC: IMF. IMF (International Monetary Fund). 2019. “Republic of Mozambique: Diagnostic Report on Transparency, Governance and Corruption.” Country Report 19/276, IMF, Washington, DC. IMF (International Monetary Fund). 2021. “Mozambique Fiscal Transparency Evalua- tion.” IMF Country Report 15/32, IMF, Washington, DC. JICA (Japan International Cooperation Agency). 2015. “Country Gender Profile: Mo- zambique.” Final Report, JICA, Tokyo. Kondylis, Florence, Valerie Mueller, and Siyao Zhu. 2014. “Seeing Is Believing? Evidence from an Extension Network Experiment.” World Bank Policy Research Working Paper 700, World Bank, Washington, DC. Mozambique, Government of. 2019. “Mozambique Cyclone Idai Post Disaster Needs Assessment.” Government of Mozambique, Maputo. Mozambique, Ministry of Education. 2010. “Regulamento De Avaliacao Do Ensino Secundario Geral.” Mozambique Ministry of Education, Maputo. Nagoya University Graduate School of International Development. 2009. Infrastruc- ture Corridors, Growth and Welfare: Comparative Study of the Corridors of Beira and Nacala, Africa. Prepared for Spatial and Local Development Team, FEU, The World Bank in collaboration with the JICA Research Institute and Oriental The World Bank Group in Mozambique, Fiscal Years 2008–21 Bibliography Consultants Co., Ltd. National Institute of Statistics (Instituto Nacional de Estatística—INE). Various years. Anuário Estatístico. Maputo: Mozambique. National Institute of Statistics (Instituto Nacional de Estatística—INE). 2021. In- quérito sobre Orçamento Familiar—IOF 2019/20 Relatório Final. Maputo: Mozambique. Nkala, Peter, Nelson Mango, Marc Corbeels, Gert Jan Veldwisch, and Jeroen Huising. 2011. “The Conundrum of Conservation Agriculture and Livelihoods in South- ern Africa.” African Journal of Agricultural Research 6 (24): 5520–5528. Norfund. 2020. “Central Solar do Mocuba Mozambique.” Development Impact Case Study, Norwegian Investment Fund for Developing Countries, Oslo, Norway. Paglietti, Lisa, and Roble Sabrie. 2012. Out-Grower Schemes: Advantages of Different Business Models for Sustainable Crop Intensification: Ghana Case Studies. Rome, Italy: Food and Agriculture Organization of the United Nations, Investment Centre Division. 90 PEFA (Public Expenditure and Financial Accountability). 2006. Public Financial Man- agement: Performance Measurement Framework. Washington, DC: PEFA. PEFA (Public Expenditure and Financial Accountability). 2008. Public Financial Man- agement: Performance Measurement Framework. Washington, DC: PEFA. PEFA (Public Expenditure and Financial Accountability). 2011. Public Financial Man- agement: Performance Measurement Framework. Washington, DC: PEFA. PEFA (Public Expenditure and Financial Accountability). 2016. Public Financial Man- agement: Performance Measurement Framework. Washington, DC: PEFA. PEFA (Public Expenditure and Financial Accountability). 2020. Public Financial Man- agement: Performance Measurement Framework. Washington, DC: PEFA. Ragasa, Catherine. 2019. Modelling the Effectiveness of the Lead Farmer Approach in Agricultural Extension Service Provision—Nationally Representative Panel Data Analysis in Malawi. Washington, DC: International Food Policy Research Insti- tute. Ragasa, Catherine, and John Mazunda. 2018. “The Impact of Agricultural Extension Services in the Context of a Heavily Subsidized Input System: The Case of Malawi.” World Development 105 (May): 25–47. Stringfellow, Rachel. 1996. “Smallholder Outgrower Schemes in Zambia.” Research Report, Project Number A0439, ODA Crops Post-Harvest Programme, Natural Independent Evaluation Group World Bank Group    91 Resources Institute, United Kingdom. USAID (United States Agency for International Development). 2018. “Mozambique: Developing Local Extension Capacity (DLEC) Project.” Maputo, Mozambique. USAID (United States Agency for International Development). 2019. “Gender As- sessment for USAID/Mozambique Country Development Cooperation Strategy.” Maputo, Mozambique. World Bank. 2003. “Mozambique—Decentralized Planning and Financing Project.” Project Appraisal Document 26365, World Bank, Washington, DC. World Bank. 2006. “Mozambique—Market-Led Smallholder Development in the Zambezi Valley.” Project Appraisal Document 35466-MZ, World Bank, Washington, DC. World Bank. 2007. Mozambique—Country Partnership Strategy FY08–11. Washington, DC: World Bank. World Bank. 2008. Mozambique—Beating the Odds: Sustaining Inclusion in a Grow- ing Economy—A Mozambique Poverty, Gender, and Social Assessment, Volume 1. Washington, DC: World Bank. World Bank. 2010a. “Community Land Delimitation and Local Development.” Policy Note, World Bank Group, Washington, DC. World Bank. 2010b. Mozambique—Making Transport Climate Resilient. Washington, DC: World Bank Group. World Bank. 2010c. “Mozambique—Third, Fourth, and Fifth Poverty Reduction Support Credit Project.” Implementation Completion and Results Report 1039, World Bank, Washington, DC. World Bank/FAO. 2011. “Rural Land Taxation in Mozambique.” Policy Note 106780, World Bank Group, Washington, DC. World Bank. 2012a. Mozambique—Country Partnership Strategy FY12–16. The World Bank Group in Mozambique, Fiscal Years 2008–21 Bibliography Washington, DC: World Bank. World Bank. 2012b. Mozambique—Social Protection Assessment: Review of Social Assistance Programs and Social Protection Expenditures. Washington, DC: World Bank. World Bank. 2014a. “Mozambique—Decentralized Planning and Finance Project.” Project Performance Assessment Report 84667, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2014b. Mozambique Public Expenditure Review: Addressing the Challenges of Today, Seizing the Opportunities of Tomorrow. Washington, DC: World Bank Group. World Bank. 2014c. “Mozambique—Public Financial Management for Results Pro- gram Project.” Project Appraisal Document 72624-MZ, World Bank Group, Washington, DC. 92 World Bank. 2014d. “Mozambique—Public Sector Reform Project.” Project Perfor- mance Assessment Report 88961, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2014e. “Mozambique—Sixth, Seventh, and Eighth Poverty Reduction Support Credit Projects.” Implementation Completion and Results Report Re- view ICRR14240, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2016a. Accelerating Poverty Reduction in Mozambique: Challenges and Opportunities. Washington, DC: World Bank. World Bank. 2016b. “Mozambique—Market-Led Smallholder Development in the Zambezi Valley Project.” Project Performance Assessment Report 104190-MZ, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2016c. Mozambique Systematic Country Diagnostic. Systematic Country Diagnostic. Washington, DC: World Bank. World Bank. 2016d. “National Decentralized Planning and Finance Project.” Imple- mentation Completion and Results Report ICRR0020110, Independent Evalua- tion Group, World Bank, Washington, DC. World Bank. 2016e. “Republic of Mozambique: Agriculture and Rural Development Non-Lending Technical Assistance—Synthesis Report.” Report AUS11869, World Bank, Washington, DC. Independent Evaluation Group World Bank Group    93 World Bank. 2017a. Debt Management Performance Assessment (DeMPA). Washington, DC: World Bank Group. World Bank. 2017b. Country Partnership Framework for the Period FY17–21 for the Republic of Mozambique. Washington, DC: World Bank. World Bank. 2017c. Mozambique—Forest Investment Project. Washington, DC: World Bank Group. World Bank. 2018a. “Climate Change DPO Series.” Implementation Completion and Results Report Review ICRR0021033, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2018b. “Mozambique—Agriculture DPO Series.” Implementation Completion and Results Report Review ICRR0021235, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2018c. Mozambique Economic Update: Shifting to More Inclusive Growth. Washington, DC: World Bank Group. World Bank. 2018d. Mozambique Poverty Assessment: Strong but Not Broadly Shared Growth. Washington, DC: World Bank. World Bank. 2018e. “Mozambique: Smallholder Irrigated Agriculture and Market Ac- cess Project.” Project Appraisal Document P164431, World Bank, Washington, DC. World Bank. 2018f. “Poverty Reduction Support Credit 9-10-11.” Implementation Completion and Results Report Review ICRR0021165, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2019a. Mind the Rural Investment Gap: Disparities in Access to Basic Infrastructure and Directions for Mozambique’s Public Investment Program. Washington, DC: World Bank. The World Bank Group in Mozambique, Fiscal Years 2008–21 Bibliography World Bank. 2019b. “Mozambique—MZ PROIRRI Sustainable Irrigation Develop- ment (P107598).” Implementation Completion and Results Report Review ICRR0021644, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2019c. Republic of Mozambique Agrarian Sector Transformation: A Strate- gy for Expanding the Role of the Private Sector. Washington, DC: World Bank. World Bank. 2019d. Strengthening Public Investment, Debt and Fiscal Risk Management in Mozambique: Completion Report. Washington, DC: World Bank Group. World Bank. 2020a. Cultivating Opportunities for Faster Rural Income Growth and Poverty Reduction: Mozambique Rural Income Diagnostic. Washington, DC: World Bank. World Bank 2020b. “Mozambique COVID-19 Response DPO.” Program Document PGD219, World Bank, Washington, DC. World Bank. 2020c. “Mozambique—National Urban Development and Decentralization Project.” Project Appraisal Document PAD2983, World Bank, Washington, DC. 94 World Bank. 2020d. “Mozambique PFM for Results Program.” Implementation Completion and Results Report Review ICRR0021993, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2020e. “Mozambique—ProMaputo, Maputo Municipal Development Program (MMDP I and II).” Project Performance Assessment Report 146784, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2020f. Mozambique—Reigniting Growth for All. Country Economic Mem- orandum. Washington, DC: World Bank. World Bank. 2020g. Mozambique Risk and Resilience Assessment. Washington, DC: World Bank. World Bank. 2020h. “Mozambique—Roads and Bridges Management and Maintenance Program—Phase II.” Implementation Completion and Results Report Review ICRR0021975, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2020i. Gender Responsive Natural Resource and Landscape Management: A Mozambique Pilot Program. Washington, DC: World Bank. World Bank. 2020j. “Mozambique Water Services Institutional Support II.” Implemen- tation and Supervision Report ISR43952. World Bank Group, Washington, DC. World Bank. 2020k. Mozambique–Performance and Learning Review for the Country Partnership Framework FY17–21 for the Republic for Mozambique. Washington, Independent Evaluation Group World Bank Group    95 DC: World Bank Group. World Bank. 2020l. State Your Business! An Evaluation of World Bank Group Support to the Reform of State-Owned Enterprises, FY08–18. Washington, DC: World Bank Group. World Bank. 2020m. Upscaling Nature-Based Flood Protection in Mozambique’s Cities: Knowledge Note. Washington, DC: World Bank Group. World Bank. 2021a. “Cities and Climate Change Project.” Implementation Comple- tion and Results Report Review ICRR0022585, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2021b. The International Development Association’s Sustainable Develop- ment Finance Policy: An Early-Stage Evaluation. Independent Evaluation Group. Washington, DC: World Bank. World Bank. 2021c. “Mozambique Agriculture and Natural Resources Landscape Management Project.” Implementation and Supervision Report ISR49757, World Bank, Washington, DC. World Bank. 2021d. “Mozambique—Cyclone Idai and Kenneth Emergency Recovery and Resilience Project.” Implementation and Supervision Report ISR48391, World Bank, Washington, DC. World Bank. 2021e. “Mozambique—Disaster Risk Management and Resilience Pro- gram.” Implementation and Supervision Report ISR48817, World Bank, Washington, DC. World Bank. 2021f. “Mozambique—Emergency Resilient Recovery Project.” Imple- mentation and Supervision Report ISR46946, World Bank, Washington, DC. World Bank. 2021g. “Mozambique—Hydro-Met PPCR.” Implementation Completion The World Bank Group in Mozambique, Fiscal Years 2008–21 Bibliography and Results Report Review ICRR0022203, Independent Evaluation Group, World Bank, Washington, DC. World Bank. 2021h. “Mozambique—Integrated Growth Poles Project.” Implementa- tion Completion and Results Report Review ICRR0022437, Independent Evalua- tion Group, World Bank, Washington, DC. World Bank. 2021i. “Mozambique—Social Protection Project.” Implementation and Supervision Report ISR48737, World Bank, Washington, DC. World Bank. 2021j. Mozambique—Coming Together for a Better Future. Systematic Country Diagnostic. Washington, DC: World Bank. World Bank. 2021k. “Mozambique—Urban Sanitation Project.” Implementation and Supervision Report ISR49393, World Bank, Washington, DC. World Bank. 2021l. “Mozambique—Water Resources Development.” Implementation Completion and Results Report Review ICRR0022601, Independent Evaluation Group, World Bank, Washington, DC. 96 World Bank. 2021m. A Regional Analysis of Climate, Weather, Water and Early Warning Systems in Southern Africa: Status Quo and Proposed Actions. Washington, DC: World Bank. World Bank. 2021n. “Stocktaking of Climate, Weather, Water, and Early Warning Systems in Southern Africa”, by Royal Haskoning DHV. Draft report for consul- tation, Washington, DC. World Bank. 2021o. World Bank Engagement in Situations of Conflict: An Evaluation of FY10–20 Experience. Washington, DC: World Bank Group. World Bank. 2021p. World Bank Support for Public Financial and Debt Management in IDA-Eligible Countries: An Independent Evaluation. Independent Evaluation Group. Washington, DC: World Bank Group. World Bank. 2022. “Mozambique—Greater Maputo Water Supply Expansion.” Imple- mentation Completion and Results Report ICR00005527, Independent Evalua- tion Group, World Bank, Washington, DC. World Bank Group, Administracao Nacional de Estradas, Roads for Water, and Meta. n.d. “Road Water Management for Resilience in Mozambique.” World Bank Group, Washington, DC. https://roadsforwater.org/wp-content/ uploads/2017/06/Roads-for-water-brochure_27_06_2017_Mozambique.pdf. Independent Evaluation Group World Bank Group    97 98   APPENDIXES Independent Evaluation Group The World Bank Group in Mozambique, 2008–21 Appendix A. World Bank Group Strategies  orld Bank Group Country Strategies and Objectives, Mozam- Table A.1. W bique, Fiscal Years 2002–21 CPS FY08–11 CPF FY12–15 CPF FY17–21 Pillar 1: Increased Account- Pillar 1: Competitiveness Focus Area 1: Promoting ability and Public Voice and Employment Diversified Growth and » Improved budget plan- » Improved regulatory Enhanced Productivity ning at central, district, environment in targeted » Improving economic and municipal levels areas management » Improved government » Improved management » Increasing agriculture in- fiduciary systems of development process comes and land tenure » Improved government through spatial planning security information and com- » Increased crop yields » Improving the business munication systems and overall productivity environment for job » Increased efficiency in target areas creation The World Bank Group in Mozambique, Fiscal Years 2008–21  Appendix A in legal and judicial » Increased employment » Expanding access to and services in selected and growth in targeted improving reliability of provinces areas of the tourism electricity sector » Improved provision and management of road infrastructure » Improved provision of water and sanitation services » Improved access to electricity » Improved access to affordable telecommu- nications » Better educated and skilled workforce (continued) 100 CPS FY08–11 CPF FY12–15 CPF FY17–21 Pillar 2: Equitable Access to Pillar 2: Vulnerability and Focus Area 2: Investing in Key Services Resilience Human Capital » Increased access to » Improved health services » Enhancing the skills base information on HIV/AIDS for the vulnerable » Improving health service and to treatment » Adaptation to climate delivery » Improved equity in health change and reduced » Improving access to services risk of natural disasters water and sanitation » Improved quality of » Strengthened social technical and vocational protection education » Increased access to potable water » Increased sustainable and affordable access to electricity for institu- tions outside the power network Pillar 3: Sustainable and Pillar 3: Governance and Focus Area 3: Supporting Broad-Based Growth Public Sector Capacity Recovery and Resilience » Simplified procedures to » Improved public financial » Increasing accountabil- start a business management ity and transparency of » Increased access to » Improved capacity of government institutions finance and support for local administration to » Extending coverage of SMEs manage public finances social protection and » Increased tele-density » Improved citizen partici- labor programs and access to ICT-based pation in public service » Promoting inclusive ur- services monitoring banization and decen- » Improved mobility » Greater contribution of tralization Independent Evaluation Group World Bank Group    101 » Increased access to wildlife conservation to » Improving management technologies and exten- the economy of climate risk and natu- sion information » Improved transparency in ral resources » Strengthened govern- extractive industries ment capacity to devel- op the tourism sector » Increased energy pro- duction for export, com- merce, and industry » Improved sustainable management of water resources » Enhanced capacity to respond to disasters Sources: World Bank 2007, 2012, 2017. Note: CPF = Country Partnership Framework; CPS = Country Partnership Strategy; FY = fiscal year; ICT = information and communication technology; SMEs = small and medium enterprises. Table A.2. World Bank Group–Supported Areas, Fiscal Years 2008–21 Focus Area CPS FY08–11 CPS FY12–15 CPF FY17–21 Governance ✓ ✓ ✓ Public sector reform/decentralization ✓ Transparency and citizen engage- ✓ ✓ ment Public financial management ✓ ✓ Legal and judicial services ✓ Economic management ✓ Human development and basic ✓ ✓ ✓ services Health ✓ ✓ ✓ Education ✓ ✓ ✓ Water and sanitation ✓ ✓ ✓ Electricity ✓ ✓ ✓ Social protection ✓ Growth ✓ ✓ ✓ Business regulations ✓ ✓ ✓ Access to finance ✓ ✓ ✓ ✓ The World Bank Group in Mozambique, Fiscal Years 2008–21  Appendix A Infrastructure Agriculture ✓ ✓ ✓ Tourism ✓ ✓ Sustainable development and resil- ✓ ✓ ✓ ience Sustainable resource management ✓ ✓ Climate change and disaster risk ✓ ✓ ✓ management Sources: World Bank 2007, 2012, 2017. Note: CPF = Country Partnership Framework; CPS = Country Partnership Strategy; FY = fiscal year. 102  lignment between Systematic Country Diagnostic Prioritized Constraints and Risk and Resilience As- Figure A.1. A sessment Fragility Drivers Sources: Independent Evaluation Group desk review based on World Bank 2016, 2017, 2020, 2021. Note: DRR = disaster risk reduction; PFM = public financial management; SCD = Systematic Country Diagnostic; WASH = water, sanitation, hygiene. Yellow = macroeco- nomic and governance issues; green = natural resources and resilience; orange = human capital and basic services; blue = security and justice; red = conflict. Independent Evaluation Group World Bank Group    103 References World Bank. 2007. Country Partnership Strategy FY08–11 for the Republic of Mozam- bique. Washington, DC: World Bank. World Bank. 2012. Country Partnership Strategy FY12–16 for the Republic of Mozam- bique. Washington, DC: World Bank. World Bank. 2016. Mozambique Systematic Country Diagnostic. Systematic Country Diagnostic. Washington, DC: World Bank. World Bank. 2017. Country Partnership Framework FY17–21 for the Republic of Mo- zambique. Washington, DC: World Bank. World Bank. 2020. Mozambique Risk and Resilience Assessment. Washington, DC: World Bank. World Bank. 2021. Mozambique 2021 Systematic Country Diagnostic: Coming Together for a Better Future. Systematic Country Diagnostic. Washington, DC: World Bank. The World Bank Group in Mozambique, Fiscal Years 2008–21  Appendix A 104 Appendix B. World Bank Group Portfolio in Mozambique, Fiscal Years 2008–21 ndependent Evaluation Group Risk to Development Out- Figure B.1. I come, Fiscal Years 2008–21 Source: Independent Evaluation Group DataMart, September 2021. Note: IEG = Independent Evaluation Group. Independent Evaluation Group World Bank Group    105 ndependent Evaluation Group Risk to Development Out- Figure B.2. I come by Instrument, Fiscal Years 2008–21 Source: Independent Evaluation Group DataMart, September 2021. Note: DPF = development policy financing; IEG = Independent Evaluation Group; IPF = investment proj- ect financing. The World Bank Group in Mozambique, Fiscal Years 2008–21  Appendix B Figure B.3. Risk to Development Outcome by Sector, Fiscal Years 2008–21 Source: Independent Evaluation Group DataMart, May 2022. 106 Note: ITS = industry, trade, and services; WSS = water, sanitation, and waste management. Other refers to a project with no specific sector. The figure includes double counting because one project can contrib- ute to more than one sector. Figure B.4. Overall Outcome Ratings, Fiscal Years 2008–21 Source: Independent Evaluation Group DataMart, September 2021. Note: IEG = Independent Evaluation Group. Figure B.5. Overall Outcome Ratings by Instrument, Fiscal Years 2008–21 Independent Evaluation Group World Bank Group    107 Source: Independent Evaluation Group DataMart, September 2021. Note: DPF = development policy financing; IEG = Independent Evaluation Group; IPF = investment proj- ect financing. Figure B.6. Outcome Ratings by Sector, Fiscal Years 2008–21 Source: Independent Evaluation Group DataMart, May 2022. Note: HS = highly satisfactory; HU = highly unsatisfactory; ITS = industry, trade, and services; MS = mod- erately satisfactory; MU = moderately unsatisfactory; S = satisfactory; U = unsatisfactory; WSS = water, sanitation, and waste management. Other refers to a project with no specific sector. The projects are double counted because one project can contribute to more than one sector.  otal Commitment of IFC Investment Projects to Mozambique Figure B.7. T The World Bank Group in Mozambique, Fiscal Years 2008–21  Appendix B by Sector, FY08–20 (US$, Millions) Source: Independent Evaluation Group based on the International Finance Corporation iPortal, July 2020. 108 Note: FY = fiscal year; IFC = International Finance Corporation.  rior Actions in Development Policy Operations by Sector, Table B.1. P Fiscal Years 2008–21 Sector Codes (no.) (%) Public administration 45 41 Financial sector 29 26 Energy and extractives 13 12 Agriculture, fishing, and industry 10 9 Industry, trade, and services 8 7 Health 4 4 Information and communication technologies 1 1 Total 110 100 Source: Operations Policy and Country Services Prior Actions data set.  rior Actions in Development Policy Operations by Theme Table B.2. P Code, Fiscal Years 2008–21 Prior Actions Theme Codes (no.) (%) Public finance management 26 24 Public administration 15 14 Financial stability 10 9 Rural development 10 9 Financial infrastructure and access 8 7 Independent Evaluation Group World Bank Group    109 Social protection 8 7 Business-enabling environment 7 6 Energy 7 6 Fiscal policy 5 5 Gender 3 3 Climate change 3 3 Data development and capacity building 2 2 Trade 1 1 Finance for development 1 1 Health systems and policies 1 1 Nutrition and food security 1 1 Disaster risk management 1 1 Water resource management 1 1 Total 110 101 Source: World Bank Operations Policy and Country Services Prior Actions data set. Note: Total for percentages adds up to 101 because of rounding. Appendix C. Methodology The evaluation followed a mixed methods approach, using a combination of portfolio analysis, document reviews, semistructured interviews, data analysis, and geospatial analysis as described in the following sections. See table C.2, Evaluation Design Matrix, for methods associated with each evaluation question. Portfolio Review and Analysis The evaluation includes a review of Mozambique’s World Bank Group–rele- vant portfolio, including regional projects, to identify the support delivered by the Bank Group during fiscal years (FY)08–21 to address (i) low agri- cultural productivity; (ii) insufficient service delivery (health, education, transport, and electricity); (iii) weak institutions and governance; and (iv) vulnerability to natural disasters and climate change. The evaluation draws on a structured document review of lending and nonlending portfolios to The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix C extract, code, and analyze relevant qualitative data. Semistructured Interviews Cognizant of COVID-19–related restrictions, the evaluation collected qual- itative information and identified lessons from experience through semi- structured virtual interviews with Bank Group staff, government officials, development partners, and relevant academics and members of civil society. For each set of interviews, a template with questions or topics was used. Issues covered in the interviews included the rationale for, and the nature and extent of, Bank Group support; the roles of different Bank Group in- stitutions; the adequacy of the Bank Group’s assessment of Mozambique’s development challenges; complementarity with other activities; coordina- tion with donors and counterpart agencies; and views on the efficacy of Bank Group support. 110 Review of Analytical Work, Academic Reports, and Evaluative Analysis The evaluation reviewed relevant Bank Group analytical work, self-evalua- tions, independent evaluations, and published papers. Databases and Indicators The evaluation team searched for and collated relevant indicators of out- come and Bank Group performance from project documents national, re- gional, and international databases. These indicators were used to identify patterns and relate findings to Bank Group support over time. Geospatial Analysis The geospatial analysis aimed at ascertaining whether the World Bank’s (and other development partners’) projects between FY08 and FY21 were focused on the areas of greater need and whether there have been changes in target- ing efforts across three subperiods (FY08–11, FY12–16, and FY17–21). The analysis was conducted at the province level, which was the smallest level of disaggregation for which needed data were available. The relative need across provinces was proxied by disaggregating macro variables (such as population and gross domestic product per capita) and Independent Evaluation Group World Bank Group    111 sector-specific variables to ascertain the level of need for access to basic services (for example, education, health, transportation, and energy). The analysis included the determination of the number of the World Bank’s (and other development partners’) project sites across the same spatial and tem- poral disaggregation. The analysis relied on a customized spatially disaggregated data set at the province level, temporally disaggregated in relation to the three subpe- riods previously listed. To the extent possible, official statistics from the government of Mozambique and other reliable institutions were collected (table C.1). Additionally—and in consideration of frequent data gaps in geographically disaggregated data—these data were complemented with remote-sensing data and novel gridded raster data sets derived from satel- lite imagery using machine-learning techniques. Remote-sensing data are particularly suitable for this type of analysis given their high temporal and spatial resolution. Table C.1. Data Sources Used in the Geospatial Analysis Years of Data Data Spatial Used in Data Variable Description Source Resolution Analysis Processing Popula- Estimated WorldPop 30 arc-sec- 2008, Sum of grid tion count total number onds (~1 km 2012, 2017 cells within pro- of people per at the equa- vincial bound- grid cell tor) aries GDP This glob- Dryad 5 arc-min- 2008, Average of grid al data set utes (~8.3 km 2012, 2017 cells within pro- represents the at the equa- vincial bound- GDP of each tor) aries grid cell. GDP is given in 2011 international US dollars. The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix C GDP per GDP divided by IEG calcula- 2008, IEG calculation capita population tion 2012, 2017 based on pop- ulation count and GDP data at the province level 112 Years of Data Data Spatial Used in Data Variable Description Source Resolution Analysis Processing Nighttime V.2 annual Earth Obser- 15 arc-sec- 2012, Average of grid lights composite of vation Group onds 2015, 2017 cells within pro- data/ Visible Infra- (~500 m at vincial bound- average red Imaging the equator) aries. Given that radiance Radiometer Maputo’s aver- Suite nighttime age radiance lights data. This is substantially product is the higher than that result of pro- of the rest of cessing nightly provinces, yet observations the evaluation for each year portfolio only and apply- includes one ing an initial project site in filter to remove this area, it has cloudy, sunlit, been exclud- and moonlit, ed from the followed by a analysis. subsequent filter to remove extraneous features (such as biomass burning and aurora). The result is a sta- Independent Evaluation Group World Bank Group    113 ble measure of brightness as seen from space. Years of Data Data Spatial Used in Data Variable Description Source Resolution Analysis Processing MPI The global MPI Oxford Province 2010, n.a. is a measure of Poverty and level 2014–18 acute multi- Human De- (harmo- dimensional velopment nized poverty. It Initiative time-se- complements ries data) traditional monetary pov- erty measures by capturing the acute deprivations in health, educa- tion, and living standards that a person experiences si- multaneously. Poverty The average Oxford Province 2010, n.a. The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix C intensity proportion of Poverty and level 2014–18 deprivations Human De- (harmo- experienced velopment nized by poor people Initiative time-se- (within a given ries data) population) or the average deprivation score among poor people. The intensity is the sum of the deprivation scores, divided by the number of poor people. Road Road density Government Province 2010, Road density density is a simple of Mo� level 2012, 2018 was derived indicator of the zambique, by dividing the concentration Estatísticas length of roads of roads in an dos Trans- (in km) indicat- area. portes e Co- ed in the report municações by the area of each province 114 (in km2). Years of Data Data Spatial Used in Data Variable Description Source Resolution Analysis Processing Paved Kilometers of Government Province 2010, n.a. roads paved roads of Mo- level 2012, zambique, 2018 Estatísticas dos Trans- portes e Co- municações Illiteracy Illiteracy rate Government Province 2008–09, n.a. is considered of Mo- level 2014–15 to be the zambique, proportion of Relatório the population Final do ages 15 and Inquérito ao older who Orçamento cannot read Familiar or write in any (IOF) language. Access to Population has Government Province 2008–09, n.a. health easy access of Mo- level 2014–15 to a health zambique, unit (that is, a Relatório person resides Final do less than 30 Inquérito ao minutes away Orçamento Independent Evaluation Group World Bank Group    115 from one). Familiar (IOF) Years of Data Data Spatial Used in Data Variable Description Source Resolution Analysis Processing World Project geo-lo- World Bank Point data July 1, Projects not in- Bank proj- cations with Enterprise (latitude and 2007– cluded in these ect sites geographic Data Cata- longitude) June 30, data sets were and adminis- log—Project 2021 coded manual- tration details Geographic ly by the eval- for a location Location; uation team that is asso- last ac- by extracting ciated with a cessed geographic project September information 24, 2021 from project documents. Development policy financing was excluded from the anal- ysis. Data were reverse-geo- coded and linked to the The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix C centroid of the correspond- ing province. The number of cumulative project sites per period was calculated with consideration of the approval and exit dates for each proj- ect. 116 Years of Data Data Spatial Used in Data Variable Description Source Resolution Analysis Processing Devel- Project geo-lo- International Point data July 1, Raw data were opment cations with Aid Trans- (latitude and 2007– preprocessed partners’ geographic parency longitude) June 30, by the eval- projects and adminis- Initiative 2021 uation team. sites tration details Data were for a location reverse-geo- associated with coded and a project; does mapped to the not include centroid of the World Bank correspond- Group projects ing province. The number of cumulative project sites per period was calculated with consideration of the approval and exit dates for each proj- ect. Mozam- Administrative DIVA-GIS Adm0 n.a. n.a. bique boundaries (country) Shapefiles and Adm1 (province) Source: Independent Evaluation Group elaboration. Independent Evaluation Group World Bank Group    117 Note: GDP = gross domestic product; MPI = Multidimensional Poverty Index. To integrate and harmonize multiple independent data sources, several steps were performed to ensure all data were at the same level of geographic dis- aggregation (province). This was particularly applicable to remote-sensing imagery sources that are global data sets, each with a different spatial reso- lution. Different raster operations were performed by using specialized geo- spatial software to extract a meaningful statistic for each data set for each of Mozambique’s provinces. Details of specific data-processing steps completed for each data source are included in table C.1. Once data were processed and harmonized, all data were tabulated and mapped. As shown in figure C.1, choropleth maps were produced by super- imposing two layers. The first layer displays the variation in the level of needs across all provinces and was generated by applying a five-class equal count (quantile) classification on the underlying data. The second layer represents the number of project sites per province (either from the World Bank or from other donors). This layer has been calculated by aggregating the cumulative number of projects sites per province and linking that to the coordinates of each province’s centroid so they could be overlaid as propor- tional circles (where the size of the circle is proportional to the number of project sites per province).  opulation Count by Province (Bottom Layer) and Cumulative Figure C.1. P Number of World Bank Project Sites per Province (Top Layer)a The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix C Source: Independent Evaluation Group calculations. Note: The bottom layer refers to the shaded regions in the map. The top layer refers to the overlaid bubbles. 118 Table C.2. Evaluation Design Matrix Specific Questions Methods Data Sources Evaluation Question 1: To what extent did the World Bank Group support improvements in agricultural productivity and access to basic services across regions to foster poverty reduction and shared prosperity in Mozambique? » How relevant was the Bank Group’s » Desk review of government plans and strategies, Bank Group » Bank Group strategies, ASA, and relevant support for increasing productivity in strategies, and project and ASA documents (including self-eval- academic and gray literature; if feasible, agriculture? How was such support uation and independent validation and evaluation documents); the team will review minutes from review conducted in coordination with other review of key documents from other donors meetings and other deliberative docu- donors? » Portfolio review of Bank Group projects ments » To what extent has Bank Group support » Geospatial analysis to understand the regional distribution of » Key project documents: PADs, ICRs, focused on Mozambique’s lagging Bank Group support and the extent of success of such support; ICRRs, and PPARs regions, and how has this changed if feasible, the team will leverage data collected by AidData and » IEG CLRRs and relevant thematic evalua- over time? conduct geospatial targeting analysis similar to the one carried tions (for example, the shared prosperity » To what extent has Bank Group support out for the IEG shared prosperity evaluation evaluation and 2011 Mozambique CPE) been effective in helping to increase » Expert opinion and desk review of documents to track evidence » Bank Group staff and external stakehold- agricultural productivity and improve of learning and adaptation over time; interviews with World ers and experts (including government, access to basic services across re- Bank and government officials to elicit their views on learning IMF, donors, private sector, civil society, gions? and adaptation. The three sources will be triangulated (expert and academia) » To what extent did the Bank Group’s judgment, desk review, interviews) » International, national, and regional sta- assistance for increasing agricultural » Desk review of strategic document and thematic analysis from tistics; firm-level surveys and household productivity recognize the high share other key development partners; comparison of donors’ work surveys of women working in the sector? against Bank Group work to understand implicit and explicit » Aid data and geospatial data division of labor; interviews with key officials from the Bank » Donors’ analytical reports and strategic Group, other donor organizations, and government officials to documents understand their views with respect to the division of labor and coordination » Theory-based evaluation approach to establish the Bank Group contribution to increased agricultural productivity and improved rural livelihoods; subject to data availability, the team will con- sider the use of quasi-experimental designs Independent Evaluation Group World Bank Group    119 120 The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix C Specific Questions Methods Data Sources Evaluation Question 2: To what extent did the Bank Group support improvements in governance in Mozambique? » How relevant and effective has the » Desk review of government plans and strategies, Bank » Bank Group strategies, ASA, and relevant Bank Group support been in address- Group strategies, and project and ASA documents academic and gray literature; if feasible, ing weak governance? (including self-evaluation and independent validation the team will review minutes from review » What factors explain the degree of and evaluation documents); review of key documents meetings and other deliberative docu- success of such support? from other donors ments » How has Bank Group support for gov- » Semistructured interviews with World Bank and gov- » Key project documents: PADs, ICRs, ernance evolved over time? To what ernment officials, particularly to answer questions on ICRRs, and PPARs extent did this support adapt to expe- “quality and pertinence of ASA and convening activ- » IEG CLRRs and relevant thematic evalua- rience and lessons learned? To what ities,” “coherence and synergies across Bank Group tions (for example, the shared prosperity extent was the support coordinated institutions,” and learning and adaptation evaluation and 2011 Mozambique CPE) with other donors? » Expert opinion and desk review of documents to track » Bank Group staff and external stakehold- evidence of learning and adaptation over time ers and experts (including government, » Portfolio review of Bank Group projects IMF, donors, private sector, civil society, » Theory-based evaluation approach to answer whether it and academia) is plausible to argue that the Bank Group contributed to » National statistics improved capacity to improved governance » Donors’ analytical reports and strategic documents Specific Questions Methods Data Sources Evaluation Question 3: How successful has the Bank Group been at helping Mozambique build resilience to climate change through disaster risk man- agement capacity building? » How relevant and effective has Bank » Desk review of government plans and strategies, Bank » Bank Group strategies, ASA, and relevant Group support been at building Group strategies, and project and ASA documents academic and gray literature; if feasible, disaster risk management capacity (including self-evaluation and independent validation the team will review minutes from review and resilience to the threat of climate and evaluation documents); review of key documents meetings and other deliberative docu- change? from other donors ments » How has Bank Group support for » Semistructured interviews with World Bank and gov- » Key project documents: PADs, ICRs, disaster risk management and climate ernment officials, particularly to answer questions on ICRRs, and PPARs change resilience evolved over time? “quality and pertinence of ASA and convening activ- » IEG CLRRs and relevant thematic evalua- To what extent was this support adapt- ities,” “coherence and synergies across Bank Group tions (for example, the shared prosperity ed to changing country conditions and institutions,” and learning and adaptation evaluation and 2011 Mozambique CPE) lessons learned? » Expert opinion and desk review of documents to track » Bank Group staff and external stakehold- evidence of learning and adaptation over time ers and experts (including government, » Portfolio review of Bank Group projects IMF, donors, private sector, civil society, » Theory-based evaluation approach to assess whether and academia) the Bank Group contributed to improved capacity for » National statistics managing natural disasters and climate change adap- tation Source: Independent Evaluation Group. Note: ASA = advisory services and analytics; CLRR = Completion and Learning Review Review; CPE = Country Program Evaluation; ICR = Implementation Completion and Results Report; ICRR = Implementation Completion and Results Report Review; IEG = Independent Evaluation Group; IMF = International Monetary Fund; PAD = Project Appraisal Document; PPAR = Project Performance Assessment Report. Independent Evaluation Group World Bank Group    121 Appendix D. Geospatial Analysis Energy Table D.1. Access to Energy: Average Radiance (Nighttime Lights) (aver- age radiance, nW/cm2/sr) Province 2012 2015 2017 Cabo Delgado 0.006 0.008 0.011 Gaza 0.011 0.014 0.016 Inhambane 0.01 0.015 0.012 Manica 0.01 0.013 0.013 Maputo City 7.155 8.364 9.208 Maputo 0.067 0.102 0.128 Nampula 0.017 0.025 0.031 Niassa 0.002 0.003 0.004 Sofala 0.023 0.031 0.031 Tete 0.012 0.021 0.023 Zambezia 0.006 0.007 0.008 The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix D Source: Independent Evaluation Group calculations. Table D.2. Energy Project Sites (Cumulative) (number) Province FY08–11 FY12–16 FY17–21 Cabo Delgado 0 0 5 Gaza 0 0 6 Inhambane 0 0 5 Manica 1 1 6 Maputo City 0 0 1 Maputo 2 2 14 Nampula 1 1 7 Niassa 0 0 3 Sofala 1 1 7 Tete 1 1 7 Zambezia 0 0 10 Total 6 6 71 122 Source: Independent Evaluation Group calculations. Note: FY = fiscal year.  rovincial Distribution of Energy Project Sites across Strategic Figure D.1. P Periods Source: Independent Evaluation Group calculations. Note: FY = fiscal year. Transport Table D.3. Access to Transportation, as Indicated by Road Density (km/km2) Province 2010 2012 2018 Independent Evaluation Group World Bank Group    123 Cabo Delgado 0.0378 0.0381 0.0375 Gaza 0.0359 0.0360 0.0361 Inhambane 0.0415 0.0415 0.0419 Manica 0.0391 0.0392 0.0395 Maputo City 0.0000 0.0000 0.0000 Maputo 0.0700 0.0700 0.0698 Nampula 0.0520 0.0526 0.0514 Niassa 0.0306 0.0319 0.0309 Sofala 0.0344 0.0344 0.0358 Tete 0.0295 0.0295 0.0295 Zambezia 0.0436 0.0441 0.0440 Source: Independent Evaluation Group calculations. Note: Road density is the ratio derived by dividing the length of roads (in km) by the area of each prov- ince (in km2). Table D.4. Transport Project Sites (Cumulative) (number) Province FY08–11 FY12–16 FY17–21 Cabo Delgado 0 0 9 Gaza 0 0 0 Inhambane 0 0 0 Manica 0 0 24 Maputo City 4 4 4 Maputo 4 4 4 Nampula 0 0 25 Niassa 0 1 8 Sofala 0 3 39 Tete 0 0 27 Zambezia 0 0 22 Total 8 12 162 Source: Independent Evaluation Group calculations. Note: FY = fiscal year.  rovincial Distribution of Transport Project Sites across Stra- Figure D.2. P The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix D tegic Periods Source: Independent Evaluation Group calculations. Note: FY = fiscal year. 124  ccess to Transportation, as Indicated by Percentage of Paved Table D.5. A Roads (percent) Province 2010 2012 2018 Cabo Delgado 26.2 22.3 26.4 Gaza 21.2 22.3 33.2 Inhambane 24.0 23.5 24.5 Manica 21.0 21.0 30.3 Maputo City NA NA NA Maputo 30.4 30.4 43.4 Nampula 14.3 13.9 21.1 Niassa 13.3 12.8 17.4 Sofala 24.9 24.9 28.0 Tete 27.8 32.4 32.4 Zambezia 17.0 17.3 23.9 Source: Independent Evaluation Group calculations.  rovincial Distribution of Transport Project Sites across Stra- Figure D.3. P tegic Periods Independent Evaluation Group World Bank Group    125 Source: Independent Evaluation Group calculations. Note: FY = fiscal year. Education  ccess to Education, as Indicated by Illiteracy Rates (percent) Table D.6. A Province 2008–09 2014–15 Cabo Delgado 70.3 60.7 Gaza 46.3 32.4 Inhambane 41.4 32 Manica 44.9 34.3 Maputo City 10.9 9.5 Maputo 26 19.3 Nampula 58.8 56 Niassa 60.8 58 Sofala 45.9 43.6 Tete 50.3 55.2 Zambezia 58.4 53.9 Source: Independent Evaluation Group calculations. Table D.7. Education Project Sites (Cumulative) (number) The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix D Province FY08–11 FY12–16 FY17–21 Cabo Delgado 2 12 13 Gaza 2 18 19 Inhambane 1 9 9 Manica 2 8 8 Maputo City 0 1 1 Maputo 3 10 11 Nampula 2 14 15 Niassa 1 12 12 Sofala 2 11 12 Tete 2 16 17 Zambezia 2 20 20 Total 19 131 137 Source: Independent Evaluation Group calculations. Note: FY = fiscal year. 126  rovincial Distribution of Education Project Sites across Stra- Figure D.4. P tegic Periods Source: Independent Evaluation Group calculations. Note: FY = fiscal year. Health  ccess to Health (percentage rate) Table D.8. A Independent Evaluation Group World Bank Group    127 Province FY08–09 FY14–15 Cabo Delgado 32.6 38.7 Gaza 76.1 83.4 Inhambane 56.4 64.5 Manica 67.9 65.4 Maputo City 99.7 96.4 Maputo 82.3 86.3 Nampula 52.3 66.1 Niassa 63.5 64.6 Sofala 68 68.2 Tete 44.1 74.4 Zambezia 43.1 64.8 Source: Independent Evaluation Group calculations. Note: A person is considered to have access to health facilities when they can reach the nearest health facility by foot in less than 30 minutes. FY = fiscal year. Table D.9. Number of Health Project Sites (Cumulative) (number) Province FY08–11 FY12–16 FY17–21 Cabo Delgado 2 7 33 Gaza 1 9 9 Inhambane 1 8 8 Manica 1 5 5 Maputo City 0 1 1 Maputo 2 8 7 Nampula 2 10 12 Niassa 2 8 10 Sofala 1 6 10 Tete 1 11 12 Zambezia 1 11 12 Total 14 84 119 Source: Independent Evaluation Group calculations. Note: FY = fiscal year. The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix D 128  rovincial Distribution of Health Project Sites across Strategic Figure D.5. P Periods Source: Independent Evaluation Group calculations. Note: FY = fiscal year. Independent Evaluation Group World Bank Group    129 Appendix E. Debt Management Performance Assessment Scores  omparison between 2008 and 2017 Debt Management Per- Table E.1. C formance Assessment Scores Performance Indicator 2008 2017 DPI-1 1. Legal Framework C D DPI-2 1. Managerial Structure: Borrowing and Debt-Related Transac- C C tions 2. Managerial Structure: Loan Guarantees B D DPI-3 1. DMS: Quality of Content D D 2. DMS: Decision-Making Process N/R D DPI-4 1. Debt Reporting and Evaluation: Debt Statistical Bulletin D D 2. Debt Reporting and Evaluation: Reporting to Parliament or D C Congress DPI-5 1. Audit: Frequency and Comprehensiveness D C The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix E 2. Audit: Appropriate Response N/R D DPI-6 1. Fiscal Policy: Provision and Quality of Debt Service Forecasts D D 2. Fiscal Policy: Availability and Quality of Information on Key C D Macro Variables and DSA DPI-7 1. Monetary Policy: Clarity of Separation between DeM and D D Monetary Policy Operations 2. Monetary Policy: Regularity of Information Sharing D B 3. Monetary Policy: Limited Access to Central Bank Financing C D DPI-8 1. Domestic Borrowing: Market-Based Mechanisms and Prepa- D D ration and Publication of a Borrowing Plan 2. Domestic Borrowing: Availability and Quality of Documented N/R D Procedures DPI-9 1. External Borrowing: Borrowing Plan and Assessment of Costs D D and Terms 2. External Borrowing: Availability of Documented Procedures D D 3. External Borrowing: Involvement of Legal Advisers D D DPI-10 1. Loan Guarantees: Availability and Quality of Documented D D Policies and Procedures 2. Onlending: Availability and Quality of Documented Policies D D and Procedures 3. Derivatives: Availability and Quality of Documented Policies N/R n.a. 130 and Procedures Performance Indicator 2008 2017 DPI-11 1. Effective Cash Flow Forecasting D D   2. Effective Cash Balance Management D D DPI-12 1. Debt Administration: Availability and Quality of Documented D D Procedures for Debt Service 2. Debt Administration: Availability and Quality of Documented D D Procedures for Data Recording and Storage 3. Data Security: Availability and Quality of Documented Proce- D D dures for Data Recording and System and Access Control 4. Data Security: Frequency of Back-Ups and Security of Stor- D D age DPI-13 1. Segregation of Key Staff Duties D C 2. Staff Capacity and Human Resource Management D C 3. Operational Risk Management, Business Continuity, and D D Disaster Recovery Plans DPI-14 1. Debt Records: Completeness and Timeliness D D 2. Debt Records: Registry System C D Source: World Bank 2017. Note: Scale is D (low) to A (high). DeM = Debt Management; DMS = Debt Management Strategy; DPI = Debt Performance Indicator; DSA = Debt Sustainability Analysis; n.a. = not applicable; N/R = not rated. Reference World Bank. 2017. Mozambique— Debt Management Performance Assessment (DeM- PA). Washington, DC: World Bank Group. Independent Evaluation Group World Bank Group    131 Appendix F. World Bank Support for the Extractives Sector Table F.1. World Bank Portfolio Project/Activity Fiscal Year Instrument PRSC-7 2011 DPF PRSC-8 2012 DPF PRSC-9 2014 DPF PRSC-10 2015 DPF PRSC-11 2016 DPF Mozambique EITI implementation 2010 IPF Extractive Industries Technical Advisory Facility 2011 IPF Mozambique Phase II: EITI Implementation 2012 IPF Mozambique Mining and Gas Technical 2013 IPF Assistance Project Extractive Industries Transparency Initiative 2014 IPF Post Compliance I The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix F Mining and Gas Technical Assistance Project additional 2018 IPF financing Economic linkages for diversification 2021 IPF Mozambique EITI implementation 2009 ASA Environmental policy dialogue 2010 ASA Mozambique EI Value Chain TA 2012 ASA Mozambique gas sector policy workshop 2012 ASA Mozambique gas master plan and policies 2013 ASA Institutional review for mining and gas 2014 ASA Mozambique mining sector governance 2015 ASA Civil society organizations’ capacity building for EITI 2015 ASA Policy Notes for new government 2015 ASA Economic policy in resource-rich setting 2015 ASA Environmental and social capacity for 2016 ASA extractives industries Source: Independent Evaluation Group portfolio review. Note: ASA = advisory services and analytics; DPF = development policy financing; EI = extractives industries; EITI = Extractive Industries Transparency Initiative; IPF = investment project financing; PRSC = Poverty Reduction Support Credit; TA = technical assistance. 132 Table F.2. Development Policy Operation Prior Actions Prior Actions on Extractive Industries Transparency Initiative PRSC-7 The recipient’s Ministry of Mineral Resources and Energy has appointed a permanent national coordinator to the EITI. The recipient’s Ministry of Mineral Resources and Energy has adequately staffed its EITI secretariat. The recipient’s Ministry of Mineral Resources and Energy has held the first meeting of the EITI multistakeholders. PRSC-8 The recipient’s Ministry of Mineral Resources and Energy has produced the first report under the EITI, as evidenced by the report published by EITI and available at www.eiti.org. PRSC-9 The recipient has achieved compliance with the standards of the EITI, as evidenced by EITI’s press release dated October 26, 2012, and available at www.eiti.org. Source: Independent Evaluation Group portfolio review. Note: EITI = Extractive Industries Transparency Initiative; PRSC = Poverty Reduction Support Credit. Independent Evaluation Group World Bank Group    133 Table F.3. Prior Actions in PRSCs Prior Actions on Extractives Regulations PRSC-9 The Council of Ministers has approved the draft petroleum law and sent it to its National Assembly for approval, as evidenced by the letter issued by the director of cabinet of the recipient’s prime minister on May 24, 2013. The Council of Ministers has approved the draft mining law and sent it to its National Assembly for approval, as evidenced by the letter issued by the director of cabinet of the recipient’s prime minister on May 24, 2013. The recipient’s state budget for 2013 has allocated 2.75 percent of revenues generated by extractive industries to districts for infrastructure development of communities in which the extractive industries operate. PRSC-10 The Council of Ministers has approved the bill defining the fiscal regime for the hydrocar- bon sector and submitted the bill to its National Assembly for approval. PRSC-11 The Council of Ministers has approved the implementing regulations for Law No. 20/2014, dated August 18, 2014, published in the Boletim da Republica No. 66 Serie I on August 18, 2014 (the mining law), as evidenced by the communication of the secretariat of the Coun- cil of Ministers dated October 13, 2015. The Council of Ministers has approved the implementing regulations for Law No. 21/2014, The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix F dated August 18, 2014, published in the Boletim da Republica No. 66 Serie I on August 18, 2014 (the hydrocarbon law), as evidenced by the communication of the secretariat of the Council of Ministers dated November 12, 2015. The Ministry of Economy and Finance has revised the system by which it transfers a share of the production taxes generated by mining and petroleum projects to communities in affected areas by budgeting a share of the royalties collected during calendar year 2014, as evidenced by letter No. 106/DNAPO/GAB/15. Source: Independent Evaluation Group portfolio review. Note: PRSC = Poverty Reduction Support Credit. 134 Appendix G. World Bank Support for Climate Resilience Progress to Overall Overall PDO Outcome Outcome Financing (ISR) (ICR) (ICRR) Pillar 1: National policy and institutional framework for climate resilience Climate change technical assistance MS Climate Change DPO I & II S Emergency Resilient Recovery Project S Disaster Risk Management and Resilience MS Program Cyclone Idai and Kenneth Emergency MS Recovery and Resilience Project Pillar 2: Climate resilience in sectors Hydrometeorology Transforming Hydro-Meteorological Ser- MS vices Project Zambezi River Basin Management Project S Urban sector Cities and Climate Change Project S S National Urban Development and Decen- S tralization Project Independent Evaluation Group World Bank Group    135 Maputo Urban Transformation Project S Water and sanitation Water Resources Development and MS Flood Response Water Services and Institutional Support MS NA II Project Greater Maputo Water Supply Expansion MS Project Mozambique Urban Sanitation Project S Transport Roads and Bridges Management and MS Maintenance Project—Phase 2 Integrated Feeder Road Development S Project Agriculture Agriculture DPO I & II S Progress to Overall Overall PDO Outcome Outcome Financing (ISR) (ICR) (ICRR) Agriculture and Natural Resources Land- MS scape Management Project Education Disaster Risk Management and Resilience MS Program Emergency Resilient Recovery Project S Education Sector Support Program S Social protection Social Protection Project and Support to S Cyclone and Flood Emergencies Source: Independent Evaluation Group portfolio review. Note: Outcome ratings are from most recent available ISR, ICR, or ICRR. DPO = development policy operation; ICR = Implementation Completion and Results Report; ICRR = Implementation Completion and Results Report Review; ISR = Implementation Status and Results Report; MS = moderately satisfactory; MU = moderately unsatisfactory; PDO = project development objective; S = satisfactory; U = unsatisfactory. The World Bank Group in Mozambique, Fiscal Years 2008–21 Appendix G 136 The World Bank 1818 H Street NW Washington, DC 20433