ETHIOPIA CLIMATE ACTION THROUGH LANDSCAPE MANAGEMENT (CALM) (P170384) TECHNICAL ASSESSMENT Document of The World Bank FOR OFFICIAL USE ONLY 1 COUNTRY CONTEXT ....................................................................................................................1 2 SECTORAL AND INSTITUTIONAL CONTEXT ...................................................................................2 2.1 Strategic Relevance ................................................................................................................... 5 2.1.1 Development Issues in Sustainable Land Management that the Program will Address 5 2.1.2 Relevance to Natural Resource Management and National Development Strategies 6 2.1.3 Alignment with CPF and WB Corporate Priorities 7 2.2 Technical Soundness ................................................................................................................. 7 2.2.1 Participatory Watershed Management 8 2.2.2 Land Administration Interventions for Security of Rural Land Tenure 10 2.3 Institutional Arrangements ..................................................................................................... 11 2.3.1 Federal and Regional Institutions for SLM 11 2.3.2 Local Institutions for SLM 12 2.4 Program Expenditure Framework ........................................................................................... 12 2.5 Expenditure Governance and Spending Structure ................................................................. 14 2.6 Financial Sustainability and Funding Predictability................................................................. 14 2.7 Effectiveness of the IGFT System in Fund Transfer................................................................. 15 3 RESULTS FRAMEWORK, MONITORING AND EVALUATION .......................................................... 15 3.1 Results, Monitoring and Evaluation for Watershed Management ......................................... 15 3.2 Results, Monitoring and Evaluation for Land Administration................................................. 16 4 ECONOMIC AND FINANCIAL ANALYSIS ...................................................................................... 16 5 TECHNICAL RISK RATING ........................................................................................................... 18 6 INPUTS TO PROGRAM ACTION PLAN ......................................................................................... 18 ATTACHMENT 1: SUSTAINABLE LAND MANAGEMENT GUIDANCE PROVIDED IN NATIONAL GUIDELINES FOR WATERSHED DEVELOPMENT ..................................................................................................... 20 ATTACHMENT 2: ACTIVITIES AND RESPONSIBILITIES FOR PARTICIPATORY WATERSHED MANAGEMENT AND RURAL LAND ADMINISTRATION ................................................................................................ 32 1 PARTICIPATORY WATERSHED MANAGEMENT ........................................................................... 32 1.1 Local level: Kebele Watershed Teams .................................................................................... 32 1.2 Watershed level: Community Watershed Teams (CWTs)....................................................... 32 1.3 Watershed User Associations (WSUAs) .................................................................................. 33 1.4 Woreda Level: SLM Steering Committees (WSC).................................................................... 33 1.5 Woreda Watershed Teams (WWT) ......................................................................................... 34 1.6 Zonal and Regional Level Responsibilities .............................................................................. 34 1.7 Federal Level Responsibilities ................................................................................................. 35 2 RURAL LAND ADMINISTRATION ................................................................................................ 36 2.1 Context .................................................................................................................................... 36 2.2 Land administration activities covered by the CALM Program............................................... 37 2.3 Institutional responsibilities for land administration.............................................................. 38 2.3.1 Roles and Responsibilities at Federal Level 39 2.3.2 Roles and Responsibilities at Regional and Zonal Level 40 2.3.3 Responsibilities at Woreda Level 40 2.3.4 Roles and Responsibilities at Kebele Level 41 ATTACHMENT 3: ECONOMIC AND FINANCIAL ANALYSIS.................................................................... 47 1 SUMMARY OF KEY POINTS ........................................................................................................ 47 2 BACKGROUND .......................................................................................................................... 48 3 METHODOLOGY........................................................................................................................ 52 3.1 Assumptions and Results ........................................................................................................ 56 3.2 Financial Analysis .................................................................................................................... 56 4 ECONOMIC ANALYSIS ............................................................................................................... 59 5 RATIONALE FOR PUBLIC PROVISION AND FINANCING ................................................................ 69 6 WORLD BANK VALUE ADDED .................................................................................................... 69 7 CLOSING COMMENTS ABOUT EFA MODEL DEVELOPMENT ......................................................... 70 2 ABBREVIATIONS AND ACRONYMS ACG Anticorruption Guidelines AGP Agricultural Growth Program ATVETC Agricultural Technical and Vocational Education Training Colleges BoA Bureau of Agriculture BLAU Bureau of Land Administration and Use CALM Climate Action through Landscape Management CBPWDG Community Based Participatory Watershed Development Guidelines CDD Community Driven Development CDP Commune Development Program CF Community Facilitator CGIAR Consultative Group on International Agricultural Research CIG Community Investment Group CO2-eq Carbon Dioxide equivalent CPF Country Partnership Framework CRGE Climate Resilient Green Economy CSA Central Statistical Agency CSRP Community Storage Receipts Program CWMU Community Watershed Management and Utilization CWT Community Watershed Team DA Development Agent DFID Department for International Development (United Kingdom) DLI Disbursement-Linked Indicator EFA Economic and Financial Analysis EBCR Economic Benefit Cost Ratio EIRR Economic Internal Rate of Return EMP Environmental Management Plan ENPV Economic Net Present Value ESAP Ethiopia Social Accountability Program ESIF Ethiopia Strategic Investment Framework ESMF Environmental and Social Management Framework ESMS Environmental and Social Management System ESPES Enhancing Shared Prosperity through Equitable Services project ESSA Environmental and Social System Assessment ETB Ethiopia Birr EX-ACT Ex-Ante Carbon-balance Tool FAO Food and Agriculture Organization FBCR Financial Benefit Cost Ratio FEACC Federal Ethics and Anticorruption Commission FHH Female-Headed Households FIRR Financial Internal Rate of Return FNPV Financial Net Present Value FTC Farmer Training Center FY Fiscal Year GCF Green Climate Fund GDP Gross Domestic Product GEF Global Environment Facility GHG Greenhouse Gas GIZ Gesellschaft für Internationale Zusammenarbeit (German Development Agency) 2 GoE Government of Ethiopia GRM Grievance Redress Mechanism GRS Grievance Redress Services GTP-II Growth and Transformation Plan II ha hectare HAC High Activity Clay soil ICR Implementation Completion and Results Report ICS Individual Consultant Selection IDA International Development Association IGA Income Generating Activity IGFT Inter-Governmental Fiscal Transfers INDC Intended Nationally Determined Contribution IPF Investment Project Financing IVA Independent Verifying Agency KLAUC Kebele Land Administration and Use Committee KWT Kebele Watershed Team LAND Land Administration to Nurture Development Project LIFT Land Investment for Transformation Project LSWI Land Surface Water Index LUC Land Use Category M&E Monitoring and Evaluation MDTF Multi-Donor Trust Fund MoA Ministry of Agriculture MOALR (MOANR) Ministry of Agriculture and Natural Resources now Ministry of Agriculture and Livestock Resources 3 MoF Ministry of Finance MoWIE Ministry of Water, Irrigation and Electricity MSE Micro and Small Enterprise MYDP Multi-Year Development Plan NDVI Normalized Difference Vegetation Index NPC National Planning Commission NPV Net Present Value NRLAIS National Rural Land Administration Information System NRMD Natural Resource Management Directorate OFAG Office of the Federal Auditor General ORAG Office of the Regional Auditor General PAD Program/Project Appraisal Document PAPWG Pastoral Area Public Works Guideline PCA Procurement Capacity Assessment PDO Program/Project Development Objective PEFA Public Expenditure and Financial Accountability PES Payment for Environmental Services PFM Public Financial Management PIM Project Implementation Manual PMO Prime Minister’s Office PPR Post Procurement Review PPSD Project Procurement Strategy Document PSNP Productive Safety Net Program PCU Project Coordination Unit RARI Regional Agricultural Research Institute 4 REACCs Regional Ethics and Anticorruption Commissions REDD+ Reducing Emissions from Deforestation and Forest Degradation and foster conservation, sustainable management of forests, and enhancement of forest carbon stocks. RED&FS Rural Economic Development and Food Security Platform REILA Responsible and Innovative Land Administration Project RIARs Regional Institutes of Agricultural Research RLAS Rural land Administration System RLAU Rural Land Administration and Use RLAUD Rural Land Administration and Use Directorate RLLP Resilient Landscapes and Livelihoods Project RPF Resettlement Policy Framework SA Social Assessment SAIP Social Accountability Implementing Partners SDP Social Development Plan SLLC Second Level Landholding Certification SLM Sustainable Land Management SLMP-I Sustainable Land Management Project Phase-I SLMP-II Sustainable Land Management Project Phase-II SLWM Sustainable Land and Water Management SORT Systematic Operations Risk-rating Tool SURED Sustainable Use of Resources for Economic Development TA Technical Assistance Tonne,t Metric tonne UAV Unmanned Aerial Vehicles UNFCCC United Nations Framework Convention on Climate Change 5 US$ United States Dollar USLE Universal Soil Loss Equation WB World Bank WDS Watershed Development Strategy WLRC Water and Land Resource Centre WMP Watershed Management Plan WO/P Without Program W/P With Program WoANR Woreda Office of Agriculture and Natural Resources WoLAU Woreda Office of Land Administration and Use WsUA Watershed User Association 6 7 1 Country Context 1. Located in the Horn of Africa, Ethiopia covers an area of 1.1 million km 2, and is the second most populous country in Sub-Saharan Africa, with an estimated population of about 100 million1, of which 80.5 percent are rural dwellers. With per capita income of US$660 in 2016, Ethiopia remains the 15th poorest country in the world, but has achieved substantial progress in economic, social, and human development over the past decade. Growth has averaged nearly 11 percent per year since 2004 and extreme poverty2 fell from 55 percent in 2000 to 26.7 percent in 2016, one of the most impressive poverty reduction results recorded globally. Low levels of inequality have largely been maintained but vulnerability to returning to poverty remains high, especially for those engaged in rain-fed agriculture. The natural resource base remains the foundation for most livelihoods and is subject to considerable environmental and climate risks. 2. The Government of Ethiopia (GoE) has embarked on a structural transformation of the economy and society, completing the first phase of the Growth and Transformation Plan (GTP-I) in 2015. GTP-II is under implementation for the period 2015–2020, and puts a strong emphasis on industrialization, urbanization, and export promotion. Public infrastructure investment has been at the center of the country’s economic strategy, financed by domestic and external public borrowing. Recent announcements indicate the GoE’s renewed commitment to improving the private sector investment climate3, and recent appointments have underscored the Government’s commitment to women benefiting from and participating in the overall economic, political, and decision-making processes in Ethiopia. 3. Land degradation affects millions of rural Ethiopians and reduces their resilience to climate change and even existing levels of climate variability. The minimum estimated annual cost of land degradation in Ethiopia is 2-3 percent of Agricultural GDP, before accounting for downstream effects such as increased flood risk. By reducing soil fertility and agricultural yields, land degradation undermines livelihood security. However, sustainable land management practices offer an opportunity to build resilience, mitigate climate change and boost local livelihoods. Ethiopia has made significant progress over the past 10 years in restoring degraded watersheds, but the challenge that remains is enormous. Successful remediation has been achieved through improved natural resource management and resource rights, livelihood diversification, and gender outreach in targeted degraded watersheds. Restoration and improved management of land at the watershed level also yields climate co-benefits, by increasing vegetative cover and soil carbon content, and hence reversing years of carbon loss through carbon sequestration4. This also helps mitigate national carbon emissions and also build resilience, by diversifying livelihood options, increasing soil fertility, reducing soil loss and by addressing rural energy security needs by boosting biomass wood fuels supply. 4. Ethiopia is among the most vulnerable countries to climate change and variability. It is exposed to severe climate impacts, its economy is highly climate-sensitive, and its adaptive capacity is low. Against the backdrop of a changing climate, reductions in the poverty headcount since 2000 are very fragile. Over the past three decades, Ethiopia has experienced many localized drought events and seven major droughts, five of which have been associated with famines. The most prominent observed climate change trend has been a tendency towards lower rainfall during the main growing seasons (March–May and December–February). Meteorological records also show that mean annual temperatures have increased by 1.30C between 1960 and 2006, there has been a marked increase in the number of hot days and a decline in the number of cold days. 1 2015 estimate 2 Extreme poverty in measured as consuming less than US$1.90 (2011 Purchasing Power Parity) a day. 3 in 2017, Ethiopia’s ‘Doing Business’ ranking was 159 . 4 50% of all forestry-related GHG emissions are caused by deforestation on agricultural land according to the CRGE. 1 2 Sectoral and Institutional Context 5. Land degradation in Ethiopia represents a significant drag on rural growth and poverty reduction. Today, approximately 27 million hectares (ha) are considered to be significantly eroded. This comprises almost 50 percent of highland areas and affects approximately one in five people in Ethiopia.5 Of the total area degraded area, 14 million ha are seriously eroded, with over 2 million ha beyond reclamation. Approximately 80 percent of the country’s land surface is prone to moderate or very severe soil degradation. For highland areas, estimates of erosion rates as measured by soil loss in tonnes per hectare per year (t/ha/yr) vary considerably across different studies, with recent work indicating an average of 29.9 tons/ha/yr.6 This work also identified barriers to the adoption of soil and water conservation practices as including land tenure insecurity, limited technical support, poor community participation, and weak research–extension linkages. Separate evidence finds that Ethiopia’s land certification program has increased land-related investments.7,8 6. By aggravating land degradation, future climate change will complicate efforts to increase food production and improve food security. Addressing sensitivity to climate change is an important consideration for the government’s program to address land degradation, as more than 80 percent of Ethiopians are engaged in subsistence rain-fed agriculture, and farms are already under significant climate stress. Under future climate scenarios, average agronomic conditions for maize, wheat and sorghum may actually improve.9 Rainfall is likely to become increasingly variable and intense, however, so Sustainable Land Management (SLM) investments that slow run-off and improve groundwater recharge will become increasingly important to capitalize on any broader improvement in conditions, especially to meet growing demand from an expanding population. 7. Climate change will bring increasing risks and uncertainty for development. In the highlands, most global climate models project an increase in precipitation, but studies using more detailed regional climate models indicate that expected precipitation change is uncertain. Temperatures will very likely continue to increase for the next few decades at a rate similar to that seen in recent years. The projected increases in the inter-annual variability of precipitation in combination with this warming will likely lead to increases in the occurrence of droughts, although heavy rains and floods are projected to increase as well. One recent modelling study indicates that agronomic conditions might actually improve10 with increasing annual precipitation. However, seasonal variability and intensity of rainfall events are expected to increase, highlighting the importance of scaling-up SLM practices to (i) capture rainfall for groundwater recharge, in order to capitalize on potential trend improvements in agronomic conditions, and (ii) limit the potential for higher rates of soil erosion. Analysis by the Water and Land Resource Centre of Addis Ababa University indicates that soil erosion is expected to increase by 7-10% per year and, in the more extreme scenarios, could increase by as much as 40-70% per year, further underscoring the need to place a high priority on SLM as an adaptation strategy. 5 Bai, Z. G., Dent, D. L., Olsson, L., & Schaepman, M. E. (2008), “Global assessment of land degradation and improvement. 1. Identification by remote sensing�. Wageningen, The Netherlands: International Soil Reference and Information Centre (ISRIC). 6 Haregeweyn, N et al. (2015). Soil erosion and conservation in Ethiopia: A review. Progress in Physical Geography p1-25. 7 Deininger, K., Ali, D.A and Alemu, T (2011). Impacts of land certification on tenure security, investment, and land market participation.: Evidence from Ethiopia land. Economics 87(2):312-334. 8 Holden, S., Shiferaw, B and Pender, J (2004). Non-farm income, household welfare 9 Thomas, T, Dorosh, P., and Robertson, R. (2019). Climate Change Impacts on Crop Yields in Ethiopia . Strategy Support Program Working Paper 130. IFPRI Ethiopia and Ethiopian Development Research Institute (EDRI) 10 Thomas, T, Dorosh, P. and Robertson, R (2019). Climate change impacts on crop yields in Ethiopia. Strategy Support Program. Working Paper 130. IFPRI Ethiopia and Ethiopian Development Research Institute. February 2019 2 8. Low adaptive capacity also contributes to high vulnerability in the proposed project communities. Most of the watersheds targeted by GoE’s program to tackle land degradation are situated in regions that have low adaptive capacity. Communities in these areas depend on traditional energy sources - mostly woodfuels, suffer high levels of food insecurity and have high under-5 mortality rates. One study of vulnerability in the Tigray Region found that climate vulnerability was inextricably linked to social and economic development11. 9. In addition to reducing vulnerability to climate change, Sustainable Land Management (SLM) is also the principal intervention required to enable Ethiopia to meet its national targets for climate change mitigation. If current practices prevail, Greenhouse Gas (GHG) emissions in Ethiopia will more than double from 150 Mt CO2e to 400 Mt CO2e in 2030 and will exceed targets set-out in Ethiopia’s Nationally Determined Contribution (NDC) submission to the UN Framework Convention to Combat Climate Change (UNFCCC). GHG emissions from agriculture and forestry account for 50% and 37% respectively of Ethiopia’s total emissions (see Figure 1). They also account for by far the highest abatement Figure 1. Total GHG emissions for Ethiopia potential (forestry 51% and agriculure 36%). (based on 2010 estimates) 10. The Government of Ethiopia has made gender inclusion in natural resource management a priority as women in Ethiopia remain vulnerable to economic, social, and environmental risks. Understanding gender aspects of natural resources management is an entry point for reversing environmental and land degradation. In Ethiopia, women farmers produce on average 23% less than their male counterparts.12 This is mainly due to inequitable access to and control of resources such as land, technologies and information, and lower levels of female participation in decision-making from local to national levels, despite the significant recent steps take by GoE to promote women in positions of authority. Various manifestations of climate change, such as drought, exacerbate fuelwood and water scarcity and add more to the domestic burden of women as they are primarily responsible for homestead crops, collection of fuel wood and water. Female-headed households (26% of all rural HHs) remain particularly vulnerable to climate shocks and face multiple challenges such as lower ownership and weaker tenure rights. 11. The government program supported by this operation will continue a strong focus on targeting gender priorities to advance resilient livelihoods at multiple scales, in particular by ensuring (i) the participation of women in local institutions for watershed management, (ii) the provision of joint land certificates for married couples, and individual land titles for female-headed households; and (iii) 11 Gebrehiwot, T. and A. van der Veen (2013). Climate Change Vulnerability in Ethiopia: disaggregation of Tigray Region. In Journal of Eastern African Studies, Vol. 7, Issue 4: 607. 12 2014) World Bank. ONE Campaign, Levelling the Field: Improving Opportunities for Women Farmers in Africa. Washington, DC, World Bank. 3 women’s access to extension services for climate smart interventions that will improve yields, boost income, reduce labor and enhance value-added opportunities for women. These will contribute to improved gender equality and inclusive development. 12. Institutional responsibility for leading and coordinating action on SLM is the responsibility of the Ministry of Agriculture (MoA). Under the MoA, the Natural Resource Management Directorate (NRMD) and Rural Land Administration and Use Directorate (RLAUD) are instrumental. At regional and local level, support is channeled and delivered by regional Bureaus of Agriculture (BoAs) and Land Administration and Use (BLAUs), and administrations at the woreda level (equivalent to district). Other ministries and agencies also play relevant roles, for example, the Ministry of Women and Children Affairs leads in defining and monitoring the policy and regulatory framework for economic and social empowerment of women and girls; and the Environment, Forest and Climate Change Commission is responsible for coordinating and ensuring the forestry objectives and principles indicated in the forestry policy of Ethiopia are implemented. The Ministry of Water, Irrigation and Electricity is responsible for the development and management of its water and energy resources in a sustainable manner. 13. The proposed operation (hereafter referred to as ‘the Program’), will help Ethiopia address international and national policy commitments that seek to address climate resilience and mitigation goals, and promote the sustainable management of natural resources. These include:  The Sustainable Development Goals (SDGs): particularly Goals 1 (Ending poverty), Goal 5 (Gender Equality and Women’s Empowerment), Goal 13 (Urgent action to tackle climate change and its impacts) and Goal 15 (Protect, restore and promote sustainable use of terrestrial ecosystems, including forests and to address land degradation and halt biodiversity loss).  Ethiopia’s commitments included in the Nationally Determined Contribution (NDC, submitted to the UNFCCC in 2017)13: Ethiopia’s NDC intends to undertake adaptation initiatives to reduce the climate vulnerability of its population and economy by building on existing good practices including mainstreaming and scale up of large-scale sustainable land and natural resource management initiatives. On mitigation, Ethiopia has committed to an emissions reduction target of 255 MtCO2e or 64% compared to ‘business-as-usual’ (BAU) emissions by 2030, which includes 90 MtCO2e from agriculture and 130MtCO2e from forestry. This target implies that economic growth from 2010-2030 would be carbon neutral. Ethiopia’s long term vision is to attain middle-income status while becoming carbon neutral. On both mitigation and adaptation, the operation is supportive of calls for ambitious action through the use of ecosystem based adapation and restoration approaches proposed recently by the Intergovernmental Panel on Climate Change14. Other national policies. These include the Ethiopia Strategic Investment Framework (ESIF) for SLM (described in detail below)15, the Climate Resilience Strategy for Agriculture and Forestry (2015)16, the National Adaptation Plan to Address Climate Change (2017)17, the emerging National Forest Sector Strategy and National REDD+ Strategy18, and sector strategies for energy, water, and agriculture. The 13 http://www4.unfccc.int/ndcregistry/PublishedDocuments/Ethiopia%20First/INDC-Ethiopia-100615.pdf 14 IPCC (2018). Summary for Policymakers. In: Global Warming of 1.50C: An IPCC Special Report on the impacts of global warning of 1.5)C above pre-industrial levels and related global greenhouse gas emission pathways in the context of strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty. World Meteorological Organization, Geneva, Switzerland. 32 pp. 15 Federal Democratic Repubic of Ethiopia (2010). Ethiopian Strategic Investment Framework for Sustainable Land Management . Ministry of Agriculure, March 2010. 16 http://www.agcrge.info/files/Vision/CRGE.pdf 17 https://www.slideshare.net/NAP_Global_Network/strategic-priorities-of-ethiopias-national-adaptation-plan 18 REDD+ stands for countries' efforts to reduce emissions from deforestation and forest degradation, and foster conservation, 4 Government also targets 22 million hectares for broader landscape restoration by 2030, in line with the CRGE ambition and the Bonn Challenge, the Africa 100, and the New York Declaration on forests. 2.1 Strategic Relevance 14. The Program has a high level of strategic relevance to Ethiopia’s high level development policy objectives, and in particuar for reducing land degradation, addressing climate change objectives and promoting resilient and sustainable livelihoods. The CALM Program (‘the Program’) will support the Ethiopia Strategic Investment Framework (ESIF) to address land degradation, enhance rural livelihoods and deliver substantial climate co-benefits and in so doing, contribute to broader national development and World Bank policy priorities. The development challenges to be addressed by the Program and their relevance are discussed below, complementing the description of the strategic context provided in Section I of the PAD. 2.1.1 Development Issues in Sustainable Land Management that the Program will Address 15. Land degradation affects millions of rural Ethiopians and reduces their resilience to climate change and even existing levels of climate variability. By reducing soil fertility and agricultural yields, land degradation undermines livelihood security. Ethiopia has made significant progress over the past 10 years in restoring degraded watersheds, but the challenge that remains is substantial. Approximately 27 million ha is considered to be significantly eroded and 2 million ha beyond reclamation. Climate change will bring increasing risks, with soil erosion expected to increase by 7-10% per year and, in the more extreme scenarios, could increase by as much as 40-70% per year. This strongly suggests a need to place a high priority on Sustainable Land Management (SLM) as an adaptation strategy. SLM is also the principal intervention required to enable Ethiopia to meet its national targets for climate change mitigation, as greenhouse gas emissons from agriculture and forestry account for 50% and 37% respectively of total emissions, and account for by far the highest abatement potential (forestry 51% and agriculture 36%). Development Challenges in Watershed Management 16. Since 2010, Ethiopia has built a strong track record of addressing land degradation and there is robust evidence that this contributes to agriculural productivity and poverty reduction. Support for SLM in the highlands of Ethiopia has demonstrated that investment in watershed development and management can increase production and productivity through preventing and reversing land degradation. Under development partner-financed projects more than a million hectares of degraded land have been brought under SLM practices. Under mainstream government programs, local communities have been mobilized on an annual basis to undertake SLM activities covering an area of approximately five million hectares. However, neither government nor development partner-financed programs are based on the establishment of durable local institutions to sustain this process and maintain restored landscapes in the long term. The Program is designed to address this shortcoming. In addition, the Program will incentivize MoA’s initiatives to improve the quality of research, training and extension for SLM practices, helping ensure the good practices developed under development partner-financed projects are disseminated more broadly through mainstream government programs. sustainable management of forests, and enhancement of forest carbon stocks. See https://ethiopiareddplus.gov.et/redd- readiness/redd-national-strategy/ 5 Development Challenges in Rural Land Administration 17. Evidence at global level, and from Ethiopia shows that land resource users without secure long-term rights have less incentive to invest in SLM practices. Ethiopia has made a progress in issuing land titles, but faces substantial challenges in relation to achieving national coverage and ensuring land title records are managed in a secure and sustainable manner. So far, 10.8 million Second Level Landholding Certificates (SLLCs) have been issued, which represents only about twenty percent of the total number of SLLCs estimated to be required. Development partner support can finance part of the cost of preparing SLLCs, but such assistance is limited both in scale, and by the fact that validation and authentication of SLLCs remains a function of woreda Bureaus of Land Administration and Use (BLAUs). A further challenge is that the tenure security provided by SLLCs is put at risk by the lack of a national database in which these can be digitized and securely recorded. The Program will address these challenges by incentivizing (i) the authentication of further SLLCs by woreda BLAUs, and (ii) the roll-out and operationalization of the National Rural Land Administration and Information System (NRLAIS), which are key elements of the GoE’s Rural Land Administration System (RLAS). 2.1.2 Relevance to Natural Resource Management and National Development Strategies 18. The Program is closely aligned with and relevant to national development policy and to Ethiopia’s ambitous Climate Resilient Green Economy strategy. The Program will support the third phase of the Ethiopia Strategic Investment Framework for SLM (ESIF-3)19, described in Section II.A of the PAD. The relevance of the Program to national development strategies and commitments is described in Section I.B of the PAD, and includes support for policy priorities identified in the Second Growth and Transformation Plan (GTP-II), the Climate-Resilient Green Economy (CRGE) Strategy launched by the GoE in 2011, as well as the SDGs and Ethiopia’s international climate commitments. The goals for natural resource management set out in GTP-II include an additional 19 million hectares to be treated with physical soil and water conservation structures, an increase in national forest coverage from 15 to 20 percent, and the provision of land use certificates to more than 7 million households. Longer-term, the Government targets 22 million hectares for broader landscape restoration by 2030. If implemented effectively, these policy goals will deliver a wide range of development and climate co-benefits. For example, the World Bank has estimated that Ethiopia could generate substantial numbers of new jobs and contributions to GDP with investments in forest plantations and processing facilities, consistent with the government’s landscape restoration program.20 19. The RLAS is strongly linked to the objectives of GTP-II, contributing to the objective of increasing the productive capacity and efficiency of the agricultural sector by providing landholders with security of tenure. RLAS also enables these landholders to access credit using their land titles as collateral, and will facilitate more efficient tax collection, contributing to the resource mobilization required to achieve national development objectives. 20. The Program addresses gender objectives as a central feature of design. By providing land rights jointly to husband and wife couples, female-headed households, and marginalized groups, RLAS also contributes to the GTP-II objective of supporting the position of women and other vulnerable groups. The Government of Ethiopia has made gender inclusion in natural resource management a priority as 19 Federal Democratic Repubic of Ethiopia (2010). Ethiopian Strategic Investment Framework for Sustainable Land Management. Ministry of Agriculure, March 2010. 20 World Bank (2016). Ethiopia Programmatic Forest Country Note. Deep Dive into Forest Smart Investments . November 2016. 6 women in Ethiopia remain vulnerable to economic, social, and environmental risks. In Ethiopia, women farmers produce on average 23% less than their male counterparts. This is mainly due to inequitable access to, and control of resources such as land, technologies and information, and a lower level of participation, from local to national level decision-making. Women are also primarily responsible for collection of fuel wood and water, resources that SLM initiatives are designed to make more available through reforestation and practices that promote groundwater recharge. The Program will incentivize (i) the participation of women in local institutions for watershed management, (ii) the provision of joint land certificates for married couples, and individual land titles for female-headed households; and (iii) women’s access to natural resource management extension services for climate smart interventions that improve yields, boost income, reduce labor and enhance value-added opportunities for women. 2.1.3 Alignment with CPF and WB Corporate Priorities 21. The Program is closely aligned with the Ethiopia Country Partnership Framework (CPF)21. Specifically, the Program is aligned with Focus Area 2 on ‘Building Resilience and Inclusiveness’, and the World Bank’s high level corporate priorities on climate change which aim to increase the level of ambition for both climate change adaptation and mitigation. This alignment is described in Section I of the PAD. 2.2 Technical Soundness 22. The Program for Results (PforR) instrument provides a means of strengthening the government’s ambitious program for addressing SLM at a transformative scale . The ESIF is being updated for its third phase (ESIF-3), providing an opportunity to adopt lessons derived from previous phases of support. These include the need to strengthen institutional arrangements at watershed level, adopting locally-driven participatory approaches, and improving service delivery in support of strengthening land tenure. The PforR instrument provides a mechanism to incentivize measures to ensure the longer-term sustainability of the national SLM program, and to scale-up investments to meet the ambitious national targets for SLM. Results-based financing will create strong incentives to meet national targets, will help GoE to mobilize domestic financing to address critical capacity constraints at federal, regional and woreda levels, and will also provide leverage for policy reform. 23. Over the past ten years, IDA Investment Project Financing (IPF) has helped restore productive capacity and build resilient livelihoods in 135 major watersheds in Ethiopia’s highlands through support for SLM practices. Through soil and water conservation structures, enclosures to limit free grazing, and afforestation or reforestation of more than 80,000 hectares, these activities have led to an average 9 percent increase in vegetation cover in treated watersheds. IDA financing for the SLM Program has also strengthened MoA’s support for land rights through the issuance of landholding certificates to over 300,000 households, including more than 200,000 women who have received titles either individually or jointly with their husbands, and more than 7,000 landless youth who have received titles to communal holdings in exchange for restoring land. Such IPF support will continue until 2024 under the Resilient Landscapes and Livelihoods Project (RLLP). 21World Bank (2017). Country Partnership Framework for the Federal Democratic Republic of Ethiopia for the Period FY18- FY22. May 22, 2017.Country Management Unit AFCE3, Africa Region. 7 2.2.1 Participatory Watershed Management 24. GoE is introducing substantive, evidence-based reforms that should Box 1: The Community Watershed Management and strengthen the institutional and regulatory Use (CWMU) Proclamation framework for implementing sustainable land management at national level. The The Proclamation will: (i) ensure active participation of reforms provide for the establishment of users in the management and utilization of watersheds; (ii) Watershed User Associations (WsUAs) create a conducive legal environment for the sustainable and through the approval by the Council of continuous use and management of the natural resources Ministers of the Community Watershed under private or public properties; and (iii) increase the Management and Use (CWMU) Proclamation capacity of WUAs to manage and utilize assets created by (Box 1). Approval of the CWMU themselves. Proclamation by the Council of Ministers, which constitutes a prior result DLI for the The principles supported by the CWMU Proclamation are Program, is expected in 2019. This will that: (i) the management, administration and maintenance of promote and set standards and procedures for development activities within the watershed should be the large scale creation of WsUAs as durable implemented by the beneficiaries with the active institutions for watershed management. These participation of the community; and, (ii) natural resource will replace the community watershed development in watersheds should give due emphasis to committees previously formed on an ad-hoc equitable use and sustainability. basis for watershed management activities. To achieve these goals, the CWMU Proclamation indicates The Program will support the implementation that it is the responsibility of the agency responsible for of this reform at national level by establishing natural resource management at the woreda level to organize incentives for the establishment and WsUAs based on the needs and interest of the watershed legalization of WsUAs. users, and that these WsUAs should have their own byelaws and be granted legal personality by the woreda 25. There is a strong evidence base for administration. It further establishes that each watershed this reform: the functionality of WsUAs has should have a plan supported by a map, and that watershed been piloted in Amhara Regional State, beneficiaries should participate in natural resource producing encouraging results. 22 In February development and maintenance activities based on the 2018, the World Bank assessed the pilot program developed by the WsUA. WsUAs in Amhara, noting strong community support, and their role in building long-term community commitment to sustainable watershed management. By March 2019, Amhara BoA reported that 8,281 WsUAs had been created. 26. The principal technical document guiding the design and application of SLM interventions under the ESIF is the Community-Based Participatory Watershed Development Guideline (CBPWD Guideline) prepared by MoA. This is described in Attachment 1. Based on the experience gained during ESIF Phases 1 and 2, this guideline is now undergoing a process of revision and update, with a particular focus on the processes and template for preparing WMPs and annual work plans for their implementation. The guidelines include the following practices which are consistent with guidance used by the Bank for 22 Piloting was supported by the IDA SLMP-2 IPF in collaboration with technical assistance from GIZ, 8 SLM interventions:23  Physical soil and water conservation practices;  Flood control and drainage practices;  Water harvesting and run-off management for multiple use practices;  Soil fertility management and biological soil conservation practices;  Agro-forestry, forage development and forestry practices;  Gully control practices. 27. These SLM practices are also climate-smart, a term used to describe an approach to agriculture that has three pillars: (i) sustainably increasing agricultural productivity and incomes; (ii) adapting and building resilience to climate change; and (iii) reducing and/or removing greenhouse gases emissions, where possible.24 These pillars are captured in the stated aim of the CBPWD Guideline, which is to: (i) guide the preparation and implementation of WMPs that increase land quality and restore degraded land; (ii) reduce the impact of climate-related hazards on priority resources and support adjustments to existing livelihood activities to reduce their sensitivity to climate impacts; and (iii) where possible contribute to the reduction of net greenhouse gas emissions. The updated CBPWD Guideline will include guidance and tools to incorporate a climate analysis in the watershed planning process and develop corresponding interventions. 28. The ongoing ESIF-3 update emphasizes the need to strengthen research, training and extension for SLM. This will focus on the roles of the Ethiopian Institute of Agricultural Research (EIAR) and Regional Agricultural Research Institutes (RARIs), and the Agricultural Technical and Vocational Education Training Colleges (ATVETCs). Responding to this challenge, the Program will support the research institutes in strengthening their natural resource management programs, emphasizing coordination between EIAR and RARIs, and collaboration among organizations and development partners. This arrangement will be formalized in an MoU to be agreed by the NRMD and EIAR. This effort will include a particular focus on adaptive research based on field data collection for the evaluation of the impacts of SLM practices on key biophysical parameters, such as soil erosion, surface water runoff and sediment load, groundwater recharge, vegetation cover and soil fertility. 29. Under the National TVET Strategy adopted in 2002 and revised in 2008, the federal and regional ATVETs have a core role in disseminating agricultural knowledge and skills. The main target group of the training programs for natural resource management are Development Advisors (DAs), extension workers employed by the government at the kebele level, expected to play a primary role in improving farmers’ knowledge and skills through the provision of up-to-date information, including on SLM practices. An assessment conducted by the MoA in 2019 found, however, that most DAs lack adequate training in natural resource management practices, and in response proposed a set of interventions to be supported under the Program, incorporating: (i) updates to the natural resource management curriculum, assessment tool, and teaching materials; (ii) provision of essential training materials to the four federal ATVET Colleges; (iii) related training for the natural resource management instructors at federal and regional ATVETCs; and, (iv) establishment of a natural resource management centre of excellence for the highland areas in Agarfa ATVET College. 30. In addition to drawing on experience under the first two phases of the ESIF, the Program also reflects lessons drawn from SLM operations in other countries. In Malawi, the Shire River Basin 24 FAO (2013). Climate-Smart Agriculture Sourcebook. Retrieved from http://www.fao.org/docrep/018/i3325e/i3325e.pdf 9 Management Project is piloting innovative approaches for integrated watershed management, and has demonstrated the value of watershed management planning to identify strategic areas for investments, and for the development of concrete action plans that villagers can implement. In Nigeria, the Erosion and Watershed Management Project has demonstrated the importance of addressing gully erosion as part of a larger catchment management plan, including investments in soil and water conservation both at gully sites and upstream to reduce further erosion, as well as in pioneering the use of vegetative interventions as part of the solution. The award-winning Karnataka Watershed Project in India has demonstrated the value of research and impact evaluation to inform evidence-based project investments. Finally, the CWMU Proclamation and the establishment of WsUAs provide a mechanism to initiate Payments for Environmental Services (PES) in Ethiopian watersheds. International experience, for example with local community contracts for reforestation and watershed management in Mexico, have demonstrated the potential of PES approaches to mobilize additional resources for SLM. A first example of this approach in Ethiopia was provided by the MoU signed in 2018 by Raya Brewery-BGI Ethiopia for investment in SLM activities to protect their water-source. ESIF-3 includes provision for PES in order to capitalize on further private sector interest in this approach. 2.2.2 Land Administration Interventions for Security of Rural Land Tenure 31. The Rural Land Administration System (RLAS) in Ethiopia will play a central role in strengthening land tenure security as a basis for sustainable land management practices . RLAS focuses on the establishment and maintenance of the land register, populated through SLLCs, and supported by a computerised system (NRLAIS), providing continuous updating of land records from land use rights’ transactions, as well as land information services. The success and sustainability of RLAS will depend heavily on the funtionality of its accompanying computerized information system - the NRLAIS, as well as complete registration of land transactions. This can only be achieved if landholders are confident in the system and realise its benefits. In turn, this requires that the system ensures effective recording of land data and is easy to use to provide related services. Following two years of development, NRLAIS is currently being piloted in a sample of ten woredas, prior to broader roll-out with the support of the Program. 32. The institutional sustainability of the RLAS is based on the ownership of the system by government, providing financial and political support for the maintenance of the policy and legal land frameworks. Operational sustainability will be achieved through the maintenance of the NRLAIS to capture and processs land transactions on a continuous basis, and efficiently provide land administration services. In addition to budgetary support, financial sustainability over the longer term will be enhanced by the capability of the land administration system to generate revenues from land information services, helping finance the recurrent costs of the system. 33. Although land administration has country-specific characteristics (for example, tenure system, level of technology and human resource development, resource allocation system), general principles for best practice may be derived from international experience25,26,27,28,29, including: 25 Good Governance in Land Administration; Principles and Best Practices; W. Zakout, B. Wehrmann, P. Törhönen; 2006 26 Good Governance in Land Administration; K. Bell; 2007 27 Improving Sustainability of Land Administration through Decentralized Service Provision: Evidence from Rwanda; D. Ali, K. Deininger, M. Duponchel; 2016 28 Benchmarking for Regional Best Practice -Comparative Evaluation of Land Administration Systems in Namibia, South Africa and Zimbabwe; D. Chimhamhiwa; 2006 29 Toolkit to support Transparency in Land Administration; GLTN/UN-Habitat; 2013 10  Simplified and standardized procedures: land administration procedures must be transparent, efficient, affordable and easy to understand, to ensure everyone can use the services;30  Effective service delivery: providing an efficient, transparent and affordable system to register new titles and transfer existing ones is an important step towards guaranteeing secure access to land and improving access to credit;31  Application of service standards: service standards are needed to guide efficiency, transparency, accountability, equity, security and effectiveness, to minimize corruption and satisfy customers in completing land transactions;32  Customer orientation: a culture of customer service is an important element of good service delivery.33 34. Four implementation assessments have been conducted of RLAS34, These provided country- specific lessons on organisational management and practice, service levels, and operations. These have been used to define mandates and budget allocations, functions of woreda and kebele land administration officials, and service performance indicators. 2.3 Institutional Arrangements 35. Institutional arrangements for leading and coordinating action on SLM are well developed, based on many years of experience addressing this issue and extend from Federal to local (Kebele) level. Overall leadership is the responsibility of the Ministry of Agriculture (MoA). Under the MoA, the Natural Resource Management Directorate (NRMD) and Rural Land Administration and Use Directorate (RLAUD) are instrumental. At regional and local level, SLM initiatives are delivered by regional Bureaus of Agriculture (BoAs) and Bureaus of Land Administration and Use (BLAUs), and administrations at the woreda level (equivalent to district). 2.3.1 Federal and Regional Institutions for SLM 36. At the federal level, existing institutional mechanisms to provide oversight, policy direction, and coordination of SLM initiatives will be continued for the third phase of ESIF. These are the Rural Economic Development and Food Security (RED&FS) Platform, the National SLM Steering Committee and the National SLM Technical Committee. The RED&FS Platform, chaired by the Minister for Agriculture, has high-level representation from concerned ministries and development partners supporting agriculture, natural resource management, food security and disaster risk management. RED&FS is serviced by a secretariat comprising MoA staff and a coordinator financed by development partners. Under the RED&FS, the National SLM Steering Committee, chaired by the State Minister for Natural Resources Management in MoA, comprises high level representation from agencies and development partners engaged in SLM. The Steering Committee is responsible for: (a) providing policy guidance, oversight and 30 Land Administration, why? Swedish National Land Survey; February 2008 31 Registering property; WB Doing Business, 2014 32 Good Governance in LA; principles and good practices. WB-FAO, 2006 33 Delivering land administration services at scale; LA Authority Lesotho, 2017 34 RLAS Implementation Assessment, LIFT; September 2017/February 2018/September 2018/January 2019 11 overall supervision for program implementation; (b) reviewing and approving annual work plans and budgets; (c) reviewing and approving annual performance reports, and overseeing the execution of any corrective actions that may be designed. The National SLM Technical Committee, also chaired by the State Minister for Natural Resource Management in MoA, is responsible for providing technical advice to MoA on SLM, and is made up of senior technical staff from relevant agencies, development partners supporting SLM initiatives, and civil society organizations engaged in SLM activities. 37. At regional and zonal levels, implementation of the ESIF is supported by regional SLM steering and technical committees, which oversee execution of annual regional work plans and achievement of results. The regional BoAs and BLAUs serve as the link between the federal, zonal and woreda institutions, reviewing and consolidating annual work plans and budgets submitted by the woredas. BoAs and BLAUs also review and approve implementation progress reports originating from the woredas. 2.3.2 Local Institutions for SLM 38. At the woreda and kebele levels, planning and execution of SLM activities are undertaken jointly by the woreda offices of agriculture and land administration, Development Agents (DAs) and communities. As set out in the CBPWD Guideline, at the woreda level an SLM core team is responsible for, among others: (i) participating in the selection and prioritization of community watersheds in the woreda; (ii) organizing orientation and training of DAs in watershed planning and implementation; and (iii) assisting DAs in the establishment of WsUAs, preparation of WMPs and annual work plans, and in the review of their implementation. At the community level, based on the CBWD Guideline, Community Watershed Teams (CWTs) have been responsible for the implementation of WMPs developed in largely by DAs and the Woreda SLM core team. Once the CWMU Proclamation is issued, it is anticipated that the CWTs will be replaced by duly constituted WsUAs, with a greater role in the design and implementation arrangements for WMPs, helping ensure that these are more participatory in nature, better reflecting local priorities, rather than the current process through which targets are established through a top-down process. 39. The legal land administration framework will be extended in 2019 with the introduction of the RLAU Proclamation, which will provide the legal basis for the components of the RLAS.35 At the local level, the effectiveness of the land administration process will be enhanced based on the principle of kebele land administration offices providing front-office functions, and the woreda land administration office fulfilling a back-office role. In addition, kebele land administration offices will continue to play a lead role in the development of local Participatory Land Use Plans (PLUPs), which provide an valuable input to WMPs. 2.4 Program Expenditure Framework 40. The Program finances a subset of GoE’s expenditures on ESIF-3, as summarized in Table 1 below. The Program Expenditure Framework for the proposed PforR operation is, therefore, based on Government expenditures on program goals at the federal, regional and woreda levels that are specific to watershed development and land administration. At federal level, the expenditure framework includes expenditures by MoA on Agricultural Extension Advisory and Training, Agricultural Technical and Vocational Training, Agricultural Technical and Vocational Training, Natural Resource Development and Protection, Rural Land Administration & Utilization Management, and EIAR Management and Administration. Subnational level expenditures, where the bulk of financing under the Program is focused, 35Ethiopia Rural Land Administration Legal and Institutional Framework; A Review on Completeness and Homogeneity; LIFT, February 2017 12 includes recurrent (both salary and O&M) and capital expenditures at regional and woreda agricultural offices, regional and woreda offices of Land Administration, and regional level spending on RARIs. Table 1. Expenditure Framework for the Program Within the subset of ESIF-3 activities supported by the Program, expenditures that are excluded from the Program are investments that relate to small-scale irrigation programs and all spending in those regions that are characterized as predominantly rangeland regions (the Afar and Somali regions). Additionally, the ESPES PforR expenditure framework currently captures all regional and woreda level recurrent expenditures on agriculture, although the results related to agriculture workers under ESPES do not require that level of expenditure. Noting the high importance of regional and woreda level recurrent expenditures on agriculture to the achievement of Program objectives, 70 percent of these expenditures are directly related to results under the Program and will therefore be included in its expenditure framework; 20 percent of the same budget line are directly related to results under ESPES and therefore will be reported under the expenditure framework for ESPES. This allocation of expenditures reflects the realistic distribution of resources needed to achieve the different results under each of the two operations and will avoid any overlap between these two operations in terms of results, DLIs and expenditures. The next 13 ESPES Implementation Status and Results report will document this revision, and subsequent financial reports of both programs will report expenditures in accordance with the agreed shares of expenditures for each of the Programs. 2.5 Expenditure Governance and Spending Structure 41. Expenditure on the Program is governed by the Intergovernmental Fiscal Transfer System (IGFT), funded through an annual budgeting process. This involves both bottom-up and top-down planning processes. In the top-down process, the federal government announces grants to the regional states annually and ensures that allocations are in accordance with the nationally agreed-upon equity formula. The regional governments in turn set the general-purpose grants for woredas and the woreda councils subsequently make sectoral allocations of the grants received from their respective regional bureau of finance. The woreda sector offices then distribute the allocations to sector programs. The bottom-up process, on the other hand, ensures that each of the woreda’s sector offices engage s communities in preparing their respective plans, at the same time ensuring that their allocations and program targets are aligned with national programs and goals. During the planning and consultation process, each woreda takes into consideration available block grant financing for the fiscal year which includes the budget ceiling provided by the woreda council and estimated in-kind transfers from the relevant sector ministry and other resources at the woreda level, including user fees. Around 80 percent of the block grants go to recurrent costs —mainly salaries and operational costs for government staff at regional and woreda levels. 42. The FY2017/18 budget indicates that recurrent spending takes up around 76 percent of ESIF expenditures while the remaining 24 percent was spent on capital costs. Moreover, the bulk of the Program eligible expenditures are spent at subnational level. The total regional level spending on Program eligible expenditures is more than 90 percent, while the remaining expenditures (less than 10 percent) are spent at federal level. 43. As an indication of the political commitment to sustain and improve on decentralized service delivery, the amounts of total block grant allocations are envisaged to increase over the medium term. In line with this increase, allocations to federal and regional levels on Program expenditures are forecast to grow substantially over the ESIF-3 five-year period. For instance, between Ethiopian Fiscal Year (EFY) 2008 and EFY 2009, GoE Program expenditures increased substantially in nominal Birr terms. Adjusting for annual Birr depreciation, these expenditures increased by 10 percent in dollar terms over the two-year period. Assuming at least a similar growth trend holds for the five-year ESIF-3 implementation period, GoE’s Program expenditures are expected to total approximately US$2.58 billion over the five-year implementation period. 2.6 Financial Sustainability and Funding Predictability 44. Predictability has been the cornerstone of the IGFTs. Apart from small deviations in the Afar and Somali regions (which are excluded from the Program), the predictability of transfers to federal programs, regional states, and from regions to woredas is 100 percent. Similarly, over the past several years, budget execution for the IGFTs equaled if not exceeded budgeted amounts, demonstrating high-quality planning and follow-through on commitments. 45. The operation is not expected to pose a risk to fiscal sustainability. Following the deceleration in FY18, growth is projected to pick up to 8.8 percent in FY19, aided by recovery in industrial activities 14 (including manufacturing) and continued strong performance of agriculture and services, stabilizing at around 8.9 percent over the medium term. Inflation is expected to continue its declining trend and reach single digit levels by the end of FY19. Fiscal policy at the general government level continued to be relatively prudent, decreasing to 3 percent of GDP in FY18 compared to 3.4 percent in FY17. Official gross international reserves increased from US$2.8 billion in June 2018 to about US$4 billion in December 2018, with import coverage of gross official reserves improving from 2.1 to 2.4 months, based on the authorities’ computations. Public debt moderated in the first quarter of FY19 declining from 57 percent of GDP in June 2018 to 54 percent. The Non-Concessional Debt Policy (NCBP) for FY19 is being observed with zero new non-concessional borrowing. 2.7 Effectiveness of the IGFT System in Fund Transfer 46. The IGFT system has put in place accountability and supervision mechanisms using independent agencies at the federal and regional levels. The IGFTs are embedded in the constitution, which commits all levels of government to disburse allocations to planned programs by transferring through the existing administrative tiers. The Office of the Federal Auditor General (OFAG) and the Offices of the Regional Auditor Generals (ORAGs) are a key part of the IGFT system that ensures the integrity of decentralized administrative bodies in performing their responsibilities. Consequently, the IGFT system is subject to strict fiduciary probity. Financial controls on salary payments, which constitute a majority of federal and regional level expenditure on the Program, are strict and difficult to bypass. In addition, the decentralized nature of this economic governance system helps ensure that woreda and kebele officials use funds to maintain and expand service delivery, including for ESIF objectives, as officials are held accountable to their constituents. 3 Results Framework, Monitoring and Evaluation 47. The Results Framework for the the Program is presented in Annex 1, constituting a set of PDO level indicators and Intermediate Results indicators. The national M&E system for ESIF-3 will be upgraded with the support of the Program and a completion date for this upgrade has been included in the Program Action Plan (Annex 6). Arrangements for monitoring and reporting on achievement of these indicators will build on the existing Monitoring and Evaluation (M&E) system set out in sub-component 6.3 of the ESIF. This sub-component establishes an M&E system to enable those responsible for implementing the ESIF to monitor and evaluate the results and update the ESIF as needed in the light of experience gained from its implementation.The M&E systems currently utilized by NRMD and the RLAUD are detailed below. 3.1 Results, Monitoring and Evaluation for Watershed Management 48. The NRMD M&E system receives reports on a weekly, quarterly, biannual and annual basis depending on type of financing, from the regional BoAs. In turn, the regional BoAs receive and compile reports from woreda or zonal agriculture offices. At the field level, data is generated by DAs, and provided to woreda agriculture offices. In addition to the reports generated through this system, NRMD conducts frequent supervision missions and assessment events. Assessment events chaired by the State Minister for NRM are conducted every four months jointly with regional BoAs to review implementation progress by institution. 49. Currently a number of initiatives are underway in MoA to upgrade M&E systems, in particular to support development partner project financing for PSNP and SLM. Both project 15 coordination units are working closely with MoA’s Planning and Programming Directorate to modernize M&E systems and operationalize a Management Information System (MIS) compatible with the larger MoA. The M&E system under ESIF-3 will be upgraded as part of this initiative to establish a system that is: i) fully compatible with the MoA MIS; ii) provides timely and accurate reporting on ESIF-3 indicators; iii) is responsive to the needs and capacities of stakeholders at different levels; and, v) where appropriate, leverage improved technologies (e.g. tablets, smartphones, SMS and web-based reporting) to improve the accuracy and quality of information while reducing the burden of reporting. In particular, the ESIF-3 M&E system upgrade for watershed management will focus on the following:  Review and correction: currently there is no mechanism to identify incorrect or false reports;  Evaluation: implementation takes place in the absence of a well-designed strategy to evaluate program outcomes (both biophysical and socio-economic);  Roles and responsibilities: there is a lack of clarity on the roles and responsibilities at different levels for reporting, and a lack of understanding on how to use this information (for example through analysis and targeted feedback to improve performance and address problem areas);  Consolidated database: the lack of a consistent, harmonized M&E database limits the capacity of NRMD to query the M&E information for both regular and ad-hoc reporting. 3.2 Results, Monitoring and Evaluation for Land Administration 50. As the NRLAIS has not yet been rolled-out, the M&E system employed by RLAUD currently focuses on the processes involved in issuing SLLCs. Further, since the majority of SLLCs issued to-date have been delivered through development partner-financed projects, the M&E systems are predominantly project-based. Strengthening of the RLAUD M&E system to track both SLLCs, and NRLAIS rollout, in a harmonized manner across development partner-financed projects and the mainstream government program will draw on the report “Draft Development of MIS, Performance of MIS, Performance Monitoring Report and Resource Planning� prepared for RLAUD to support the SLLC process.36 Noting that the reporting hierarchy cascades upwards from kebele to woreda through zone to region, and finally RLAUD, the report recommends strengthening the M&E function at the zonal level in particular, as zonal offices are sufficiently close to woreda offices to conduct effective monitoring, with sufficient capacity to consolidate results for reporting to regional offices. 51. ESIF-3 will support a transition from a project-by-project, and largely manual approach to M&E of land adminstration implementation, to a harmonized and automated national M&E system. This will become possible through systematic implementation of the NRLAIS throughout all land administration entities under MoA, digitization and integration into this system of existing land records and issuance of new SLLCs using NRLAIS capabilities. NRLAIS includes a dedicated reporting module allowing project activities to be tracked and which can generate associated statistics at woreda, zonal, regional and federal levels. The reporting hierarchy cascades upwards from kebele to woreda through zone to region, and finally RLAUD. Nonetheless, M&E functions will need to be strengthened, at the zonal level in particular to ensure they can undertake effective monitoring of woredas and to ensure they have sufficient capacity to consolidate results for reporting to regional offices. 4 Economic and Financial Analysis 36 ORGUT Consulting AB 16 52. Without the Program intervention, beneficiaries will continue to struggle to establish or maintain their livelihoods. Continued soil erosion, water insecurity and land insecurity leads to land degradation with direct losses to those that rely on local natural resources for food, livelihood and energy needs. With the IDA investment in the Program, future SLM will be built on strong institutions, participatory management, and secure tenure rights. After the 5-year Program costs of US$1.7 billion including US$500 million from IDA, based on the ESIF-3 update the analysis assumes that the GoE extends its Program into the future. In total, 10,000 micro watersheds are assumed to be implemented over a 10- year period covering 5 million ha. Program costs estimated for a 30-year period amounts to US$9.4 billion including US$729 million for community labor (an average of 14 million person days per year). 53. Economic and financial analysis. To assess the ex-ante efficiency of the Program investment, a 30-year cost benefit model is used. Annual cost and benefit flows are estimated as the difference between without-program and with-program net benefits for direct beneficiaries. The Program yields an economic IRR of 35 percent which exceeds the economic discount rate of 5 percent.37 Most of the Program benefits are captured on non-cropland (53%), through reduced GHG-emissions (19%), and cropland (14%), with the remaining benefits captured through improved livestock production (6.5%) and avoided soil erosion (6.5%). When excluding the social value of reduced GHG emissions (sometimes referred to as the social value of carbon), the Program yields an economic IRR of 27%.38 The financial analysis shows that estimated farm-level gross margins can increase by on average US$60 annually per person (including the value of production used for home consumption), which is 54 percent of the national food poverty line.39 This is a direct measure of increased resilience in the Program area. The financial IRR of 26% exceeds the 12% opportunity cost of capital.40 54. The Program investment is expected to yield considerable returns when accompanied by a strong risk management plan. As part of the risk management plan, the focus needs to be on ensuring that the main decision makers (farmers, other land/resource users, and local institutions) receive the necessary technical advice, land tenure security, and institutional support so that they are able and can afford to make long-term decisions. To avoid increased costs due to Program delays and prolonged implementation periods, close monitoring is necessary so as not lose momentum after the the Program is complete. Considerations should be made for the equitable inclusion of all beneficiaries in the WsUAs including vulnerable groups such as landless, youth, elderly, and female-headed households. For example, landless people may suffer from conservation interventions that include restricted access and use of communal lands. Also, restriction on livestock grazing in some areas may lead to increased erosion in other areas unless cut-and-carry systems, stocking rates and grazing rotation are incorporated in WMPs. 55. There is a strong rationale for public intervention as proposed by the Program because it supplies public goods, internalizes externalities, and corrects market failures. Current costs of land degradation warrant more targeted public investments to ensure that private sector entities are able to adopt sustainable land management practices for long term benefits. In addition to the direct effect on poverty reduction of the expected increase in farm-level gross margins indicated above, by building resilience to climate change the Program will reduce the risk to households of falling back into poverty as a result of climate-related shocks, and will contribute to economic stability by reducing the impact of extreme weather 37 The Program yields an economic NPV of US$12.3 billion (US$410 million per year) with a benefit cost ratio of 2.7. The payback period is 7 years. 38 Excluding reduced GHG emissions, the net economic Program return is US$8.5 billion (US$285 per year), with a benefit cost ratio of 2.2 and a payback period of 8 years. 39 The national food poverty line is US$112/person/year in 2020 terms 40 In financial terms the NPV is US$2.3 billion (US$75 million per year), and the estimated benefit cost ratio is 1.6 with a payback period of 9 years. 17 events, including both droughts and flooding. 5 Technical Risk Rating 56. The technical risk rating is assessed as substantial. Four main technical risks have been identified: (i) a risk that WsUAs will either not be formed, or will not function effectively, for lack of local administrative capacity to build adequate support in local communities; (ii) poorly prepared WMPs will not be effective in preventing and reversing land degradation; (iii) a lack of adequately trained staff, institutional capacity and facilities will hamper the provision of SLLCs and the operationalization of NRLAIS; and (iv) inadequate systems for monitoring and evaluation will fail to provide reliable data against which to measure the achievement of the disbursement-linked results. 57. After ten years of implementing the first two phases of ESIF, capacity exists at regional and woreda level to organize community labor for SLM interventions. Part of the transformation that the the Program is supporting through the third phase of ESIF, however, is the establishment of legally constituted WsUAs allowing local communities to take a leading role in the preparation of WMPs, and to participate fully in the development of annual work plans for their implementation. This change in approach will require a shift in the role and capacity of officials charged with natural resource management responsibilities at the woreda and kebele levels, to focus more on building local community ownership of the targets and initiatives agreed in the WMPs. 58. Project-based support for SLM during the first two phases of ESIF, including under the Bank-financed SLMP-I and SLMP-II IPF operations, has led to the development of considerable knowledge and expertise regarding the design and implementation of effective interventions to prevent and reverse land degradation. This project-based expertise has not always been reflected in the SLM interventions undertaken in non-project woredas, however, where achievement of annual activity targets imposed in a top-down manner has often been more of a priority. Achievement of the Program’s goals will require the transfer of skills and knowledge at kebele and woreda level, to ensure that design and implementation of the WMPs is of sufficient quality in all watersheds to effectively prevent and reverse land degradation. 59. The risks associated with rural land administration, are related to the challenges of introducing new functions, in the form of providing SLLCs and operating NRLAIS. Effective performance of these functions will depend critically on the availability of suitably trained staff and adequate facilities, particularly at the woreda level. 60. A significant challenge common to both participatory watershed management and land adminstration support is the need to monitor results at the kebele and woreda levels, and to efficiently consolidate and report these results up through zonal and regional administrations to the federal level. While significant experience has been developed with such monitoring and evaluation systems through project-based support for SLM, implementation of the Program will require the application of this experience to provide reliable monitoring and reporting of results on a significantly larger scale. 6 Inputs to Program Action Plan 61. A Program Action Plan (PAP, presented in Annex 6) has been developed identifying key actions required to mitigate each of the technical risks identified above. To help address the risks that 18 WsUAs will either not be formed, or will not function effectively, and that poorly prepared WMPs will not be effective in preventing and reversing land degradation, NRMD will: (i) prepare a guideline for WsUA formation and functioning; (ii) update the CBPWD Guideline; and, (iii) reach agreements with ATVETCs for the provision of training on an ongoing basis to relevant kebele and woreda officials for the application of these guidelines. 62. Mitigation options are available to mitigate the capacity risks assocated with NRLAIS roll- out. These are that RLAUD will: (i) reach agreements with regional BLAUs on staffing and facilities, setting conditions for RLAUD support for SLLCs and NRLAIS roll-out; and (ii) reach agreements with ATVETCs for the provision of training on an ongoing basis to relevant kebele and woreda officials for the implementation of RLAS and operation of NRLAIS. To address the risk that inadequate systems for monitoring and evaluation will fail to provide reliable data against which to measure the achievement of the disbursement-linked results, both NRMD and RLAUD will upgrade their M&E systems to facilitate the collection of necessary data at kebele and woreda level, and the consolidation and reporting of this information up through zonal and regional bureaus to the federal level. 19 Attachment 1: Sustainable Land Management Guidance Provided in National Guidelines for Watershed Development National guidelines for watershed development are currently undergoing a process of revision in parallel with the development of the new CWMU Proclamation to guide watershed development. These will be approved by the Council of Ministers in 2019. Existing guidelines41 have been used for over 13 years to guide practical implementation of sustainable land management actions at watershed level and have played an important contribution to the success of efforts over the ESIF 1 and 2 implementation period. The guidelines require participatory approaches to the planning, management and monitoring of watershed management interventions. These participatory approaches have demonstrated considerable success in Ethiopia. They encourage physical and biological treatments to begin from the upper part of a watershed or micro-watersheds, which averaging around 500 hectares (ha)—a size manageable for communities. The technologies have included, among others, soil bunds, stone bunds, bench terraces, cutoff dams, check dams, percolation ponds, and deep trenches, with the latter three especially useful in conserving soil moisture. These technologies have helped to reduce erosion and topsoil loss, improved surface-water availability, and replenish the groundwater that feeds spring flows. The guidelines are presented as separate volumes. Part 1 provides detailed guidance to development agents and communities on basic concepts and ‘process’ aspects covering participatory watershed development, watershed planning and development procedures. Part 2 comprises nine annexes and these focus mostly on technical implementation aspects of watershed management, including detailed technical guidance on sustainable land management interventions. The existing guidelines are summarized in tabular form below. 41 Government of Ethiopia (2005). Community Based Participatory Watershed Development. A Guideline. Part 1 . Ministry of Agriculrure and Rural Development. January 2005. First edition. 20 Theme Guidance Community-based Participatory Watershed Development: A Guideline Part 1 The Guidelines aim to build upon existing community-based participatory watershed efforts to harmonize and consolidate planning procedures at the grass-roots level. The intent is to provide DAs and rural communities with a workable and adaptable planning tool, whether located in a low rainfall or Objectives high rainfall area, in a severely degraded and food-insecure area or in a food secure and not yet seriously affected by land degradation area. Another main objective of the participatory watershed planning guideline is also to provide practical guidance on the correct selection of technologies under different conditions and their sequentially correct implementation. This term is defined in the guidelines as “any surface area from which runoff resulting from rainfall is collected and drained through a common confluence Watersheds point.� The guidelines provide added explanation on watershed definition and sub-sections discuss watershed degradation features, watershed size, biophysical and socio-economic aspects etc. This term is defined as:� Participatory watershed development can be defined as the rational and socially acceptable utilization of all the natural resources for Participatory optimum production to fulfill the present need with minimal degradation of watershed natural resources such as land, water, and environment.� The reason for development applying PWD is explained in detail and principals for its implementation are also defined. These include participatory, gender sensitive, building on local experience, realistic and manageable, the need for flexibility etc. This section describes participatory watershed management tools and approaches and sets-out 8 detailed steps: (i) Prioritization and selection of Watershed planning watersheds, (ii) Getting started at community level, (iii) Biophysical and socio- and development economic surveys, (iv) Identification and prioritization of interventions, (v) procedures. Discussion and approval of interventions by General Assembly, (vi) Mapping and action planning, (vii) Implementation strategies, (viii) Participatory monitoring and evaluation. 21 Community-based Participatory Watershed Development: A Guideline Part 2 (Annexes) Annex 1: Participatory mapping and understanding the target area. Links to SLMP Videos This annex provides detailed guidance https://www.youtube.com/watch?v=AIlyz2-cixs on problem identification within watersheds and participatory mapping approaches. Annex 2: Participatory mapping and socio-economic survey This covers approaches for ensuring people’s participation including stakeholder analysis, problem identification, baseline surveys. The problem identification guidance helps DA and communities identify and rank particular land degradation challenges facing local communities, for example lack of construction materials, grazing land and fuelwood. Annex 3: Biophysical survey and mapping Includes guidance on biophysical survey, mapping - including identification and categorization of areas of soil types and quality, land erosion and forest loss and degradation. It includes detailed guidance on mapping approaches (including topography, drainage etc) and for mapping sustainable land management interventions. Annex 4: Simple survey methods This includes technical guidance on basic survey methods, including measuring slope gradients, measuring vertical level intervals, soil texture analysis, etc. 22 Annex 5. Interventions and their suitability. This annex provides specifc technical guidance on the full range of sustainable land management interventions and on their suitability for their use in particular agro- ecological contexts. The main topics of ths guidance is summarized below: For each intervention, assessment of sutability based on agro-ecology is provided for arid (up to 500mm rainfall), semi-arid (500-900mm rainfall) and medium/high rainfall areas (above 900mm rainfall) Work norms for each approach are also included in terms of person days (PDs) per km or area (based on monitoring data derived from implementation projects). Table 5.1: Physical soil and water conservation (SWC) measures Soil bunds Stone bunds Stone faced soil bunds Fanya Juu bunds Bench terraces 23 Table 5.2: Flood control and drainage Rock catchment water harvesting- runoff farming and ponding Cutoff drains Vegetative waterway Stone paved waterway Waterway check and drop and apron structure Flood water diversion using spreading bunds Vertisols management Table 5.3: Water harvesting and run- off management for multiple uses and irrigation Hand dug shallow wells Low cost micro-ponds Underground cisterns Percolation pits Ponds Spring development Roof water harvesting Riverbed dams Stream diversion weir Farm dams Conservation bench terraces with runoff areas Tie-ridges Inter-row water harvesting Planting pit systems Half-moon structures 24 Table 5.4: Soil fertility management and biological soil conservation Contour cultivation Compost making Efficient use of fertilizers Grass strips along contours Stabilization of physical structures Vegetative fencing and stabilization (closures, gullies and farm boundaries) Strip cropping Ley cropping Cover/green manure crops Intercropping Sequential cropping using food crop Cropping using forage crops Relay cropping Mulching and crop residues management Crop rotation Choice of crops and plant population density Improved fallowing 25 Table 5.5 Agro-forestry, forage development and forestry (community/group/private) Physical Measures Trenches Micro-basins Eyebrow basins Herring bones Micro-trenches Improved pits Hillside terraces Hillside terraces+trenches Half-moon structures Restoring the Landscapes of Ethiopia’s Highlands Vegetative Measures Alley cropping and improved hedgerows Multi-story gardening Trees/shrubs/grass/ hedgerows Area closure Small soil or stone-faced bunds using run-off/run-on areas Narrow stone lines Large half-moon structures 26 Table 5.6 Gully Control Stone check dams, Brushwood checkdams, Gully cuts/reshaping and filling, Gully revegetation, Sil storage overflow dams, Spill storage overflow bunds Annex 6: Summary of National Work Norms This annex provides guidance and data on work norms - for example, person days per km, or per ha, for the interventions listed in the annexes above. These work norms will be updated as part of the ongoing, updating exercise. Annex 7 Useful plant species This annex lists plant species and their suitability for use in sustainable land management interventions in different agro-ecological zones. Sub-tables provide guidance on use of species for pasture management, fruit production, and horticulture/cash crops. Detailed information is also provided on uses and characteristics of particular species to guide species 27 selection. Annex 8 Community-based solidarity efforts This annex provides guidance for DAs and communities on establishing groups to manage and sustain participatory watershed management efforts. This section will be revised accordingly to ensure compatibility with the new CWMU Proclamation. Annex 9: Planning Formats and samples This includes a wide range of templates and formats to assist with participatory watershed management planning, implementation and reporting 28 Excerpts of the technical guidelines are shown below: 29 30 31 Attachment 2: Activities and responsibilities for participatory watershed management and rural land administration 1 Participatory watershed management Over the past 15-20 years, the GoE has established a well-developed institutional and technical framework for supporting work on sustainable land management at scale. This section provides a summary of the main institutional mechanisms that support the national SLM program at local (Kebele), watershed, woreda, regional and federal level. 1.1 Local level: Kebele Watershed Teams At the local level, Kebele Watershed Teams (KWTs) coordinate the preparation and implementation of Watershed Management Plans (WMPs) for micro- /community watersheds by bringing together and organizing members at micro-watershed level. The KWTs also coordinate with the Community Watershed Teams (CWT - see below) in cases where micro-watersheds comprise part of major watersheds. KWTs are composed of the Kebele Chairman, the Head of the Kebele Rural Development Office, three DAs; male and female representatives or leaders of each community (village); a respected and influential person from each community (village); and a representative of the youth. The roles and responsibilities of KWT include:  Watershed planning, coordination and prioritization at local level  Allocation of resources  Targeting and quality control  Dispute resolution and provision of support on specific issues (e.g. land certification)  Assist communities in monitoring and evaluation, compilation of reports, training and organization of field days and experience-sharing within and between kebeles; and  Convening regular meetings to review WMP implementation progress. 1.2 Watershed level: Community Watershed Teams (CWTs) Community Watershed Teams (CWTs) operate at the ‘major’ watershed level and their function is similar to KWTs operating at the kebele level: they bring together representatives of the constituent micro- watersheds and other relevant stakeholders, including village and community representatives from across the watershed and serve as a contact mechanism with development agents, communities and local leaders during the preparation, implementation and monitoring and evaluation of WMPs. CWTs generally include: (i) male and female headed households representing different social groups (including vulnerable) and living in different parts of the community, (ii) youth representatives, (iii) religious representatives (of the dominant religion in the community), (iv) clan leaders (for pastoral areas); and (v) Others as proposed by the community members (such as innovative farmers/ pastoralists, respected/ influential people, women's group, and others). 32 Watershed Management Plan (WMP) The CWT is responsible for the preparation of the multi-year WMPs which identifies required interventions and activities over a 1 to 5 year time frame. The multi-year WMP is prepared in a participatory manner with respective community members following the principles and steps indicated in the Community-Based Participatory Watershed Development Guidelines (CBPWD Guideline) issued by the the Ministry of Agriculture (MoA). All WMPs are subject to endorsement by the General Assembly. The CWT also works with the DAs and the Woreda Watershed Teams (WWTs) to formulate and agree annual work plans, which are also subject to endorsement by the General Assembly. The steps in the preparation of participatory WMPs include: � Undertake biophysical and socio-economic survey; � Participatory Problem Identification (PI) and Ranking; � Contextual Gender Analysis; � Detailed biophysical survey and mapping; � Identification and prioritization of interventions that bring change; � Key integration requirements and sequencing of activities; � Description of measures and specific technologies; � Climate change mainstreaming; � Mainstreaming of Gender, Social Development, Nutrition and Environmental and Social Screening of Interventions; � Getting the options and interventions discussed and approved by the General Assembly; � Organizing WMPs: This includes preparing the development map of the micro-watershed, identifying required inputs, timeframes for implementation, establishing milestones and sharing the final plan to all concerned bodies. 1.3 Watershed User Associations (WsUAs) The CWTs will be replaced by Watershed User Associations (WsUAs) following the approval and implementation of the Community Watershed Management and Utilization (CWMU) Proclamation by the Council of Ministers. The WsUAs are expected to have a greater role in the design and implementation arrangements for the Watershed Management Plans (WMPs), helping ensure more participation of respective communities, reflecting local priorities which in turn will ensure greater ownership and sustainability of positive development impacts. 1.4 Woreda Level: SLM Steering Committees (WSC) At the Woreda level, a Woreda SLM Steering Committee (WSC), composed of members of the Woreda Cabinet with heads of Woreda sectoral line offices as members, are responsible for reviewing and approving the annual work plans and budgets of each WMP, monitoring and evaluating implementation at woreda level, and reviewing the monthly, quarterly and annual implementation performance reports. They also help resolve conflicts, guide coordination with other activities ongoing at woreda level and provide a conduit for reporting to the BoAs at regional level. 33 1.5 Woreda Watershed Teams (WWT) WWTs are established by the Woreda Office of Agriculture (WoA) and perform the role of SLM ‘core teams’ and their principal function is to provide technical and advisory support for the implementation of participatory watershed management interventions/activities in the woreda, for example ensuring good coordination, capacity-building, providing technical support (for example on the use of GPS equipment) and supporting monitoring and evaluation. They are generally composed of a minimum of 10 experts, including experts in:  soil conservation,  forestry/agro-forestry  agronomy  water harvesting and irrigation  land use and administration  livestock  food security  cooperatives and markets  rural roads and engineering. In view of the responsibilities described above, the WWT must finalize the pre-planning/preparatory tasks required at the woreda level. The major preparatory activities are:  Identification and mapping of targeted major watersheds. This includes identifying watersheds with the highest level of degradation and the need for urgent conservation/restoration and preparing maps, training Development Agents (DAs) and Community Facilitators (if assigned) and making available materials and equipment for surveys and planning work.  Prioritization and coordination of micro-watersheds as the micro-watersheds may contain one or more kebeles and several communities.  Supporting DAs through training, technical and planning support for major and micro-watershed management, planning and implementation.  Reconnaissance visits at Kebele and community level and validation of mapping and discussions with Kebele/community leaders and DAs to introduce and explain participatory watershed management principles to the Kebele leaders and representatives. The WWT explain the intervention logic and provide concrete examples of typical watershed interactions, for example flood control, water-table recharge, spring development, gully control, and others. 1.6 Zonal and Regional Level Responsibilities The regional Bureaus of Agriculture (BoAs) are responsible for ensuring the establishment and operationalization of regional SLM Steering and Technical Committees in close collaboration with Bureaus of Land Administration and Use (BLAUs). The regional BoAs are supported by Regional SLM Technical Committees (RTC). Duties and responsibilities of the regional SLM Steering Committee (RSC) include:  Integration with regional development policies and ensuring implementation is consistent with Federal policies  Review and approval of the annual work plans and budget of the WMPs approved by respective woredas  Review and approve the quarterly and annual implementation performance of the Program and oversee dissemination of Program results  Ensure efficient utilization of the resources allocated to the SLM program.  Conflicts resolution between implementation stakeholders. 34  Encourage exchange of best practice among different regional Program implementing stakeholders/institutions. The regional BoAs also provide technical assistance to the respective woredas for the establishment of the WsUAs and follow-up on legalization of each WsUA by concerned regional government agency/institutions. They also undertake capacity development activities in close collaboration with concerned regional research institutions and Agricultural Technical and Vocational Education Training Colleges (ATVETCs), targeting different experts (at regional zonal and woreda levels) and DAs to be involved in the Program implementation. 1.7 Federal Level Responsibilities The National SLM Steering Committee is chaired by the State Minister for Natural Resources Management in MoA and is responsible for:  providing policy guidance, oversight and overall supervision for the Program implementation  reviewing and approving annual work plans and budgets  reviewing and approving annual performance reports  overseeing the execution of any corrective actions that may be designed. The National SLM Technical Committee, also chaired by the State Minister for Natural Resource Management in MoA, is responsible for providing technical advice to MoA on SLM, and is made up of senior technical staff from relevant agencies, development partners supporting SLM initiatives, and civil society organizations engaged in SLM activities. The Committee is playing an important role in overseeing the preparation of new Proclamation for Community Watershed Management and Utilization (CWMU) which will be submitted for the review and approval of the Council of Ministers (and included as a prior result DLI in this operation). The Committee is also playing an oversight role for the updating of national guidelines for participatory watershed development (see box). Box 2: Update of the Community-Based Participatory Watershed Development Guideline (CBPWD Guideline) Since 2005, the principal technical document guiding the design and application of SLM interventions under the national ESIF Program has been the CBPWD Guideline prepared and published by MoA. Based on the experience gained during the past years, this guideline is now undergoing a process of extensive revision and update, with a focus updating the processes and templates for preparing WMPs and annual work plans. The update process is being led by the Natural Resources Management Directorate of the MoA and the revised guidelines are likely to play an important role in shaping SLM practice throughout the ESIF-3 period. 35 2 Rural Land Administration Over the past fifteen years, Ethiopia has been implementing one of the largest, fastest, least expensive, and progressive land certification reforms in Africa. This section assesses the GoE’s vision for the transformation of land administration, assesses the ongoing reform program together with the actions to be implemented to deliver this vision and reform program at different levels of government. It describes existing approaches and steps in the land administration process, the GoE’s plans to introduce a modern land registry and policy targets for expanding land certification to strengthen land tenure security and thereby increase investments in sustainable land management. 2.1 Context The draft update of ESIF-3 and the second Growth and Transformation Plan (GTP -II) underlined the importance of a sound land policy which places securing land tenure rights as a critical element for inclusive and sustainable economic development. The program started with an initial revision of legal and policy framework, progressively adjusted based on the lessons learnt through several pilot operations. A broad- scale First Level Landholding Certification (FLLC)42 program built upon this initial set-up and was soon followed by a Second Level Landholding Certification (SLLC)43 program, which continued geographic expansion of territories covered by landholding rights. Empirical evidence shows that land certification contributes to increased investment in land productivity mechanisms and infrastructure, boosts rural land markets, provides more transparent transferability of land use rights and reduces land disputes, especially related to inheritance issues. Encouraging results were achieved by the FLLC program and early implementation of SLLCs (launched in 2013). The GoE targeted 28.7 million SLLC in the GTP II period (2015/16 to 2019/20) out of over 50 million parcels in the four highland regional states (Amhara, Oromia, SNNP and Tigray). A strong political commitment has been established over the past period of engagement at all levels ensuring better implementation of the policies and proclamations. Coordination between development partners was improved and the capacities of regional bureaus and woreda offices has been strengthened. Despite impressive progress over the last decade, the scale and quality of the achievements has not matched the ambitious targets set out in the GTP-II. As of January 2019, around 38% of GTPII targets had been met. To accelerate the progress, a number of systemic issues need to be addressed for example, the insufficient application of regulations and standards for business operations within land units, low capacity of the field level personnel, and poor operational conditions and equipment. Large numbers of landholding rights certificates have been manually registered as paper records, with insufficient provisions for secure their secure storage and thus for the security of rights they provide for. The existing Rural Lands Administration and Use (RLAU) Proclamation provides the legal foundation for the introduction and operation of the RLAS including SLLCs. However, it lacks critical provisions on the institutionalization and modernization of land records through computerized systems, their maintenance and standards and specifications for the operation of this system. As security of tenure requires reliable and up-to-date land information, the sustainability of functional land administration depends on the ability to keep the land information constantly updated and securely backed-up. Reflecting these shortcomings, the conclusions of the Land Governance Assessment completed in 201644 included recommendations that: (i) 42 FLLC-First Level Landholding Certification collects the information about de facto landholding rights through systematic adjudication campaign (information on the land holders and area of the parcel(s) in the land holding. 43 SLLC- Second Level Landholding Certification process includes additionally to the information collected during the FLLC adjudication geospatial information on the parcels within the landholding and creates a cadastral extract attached to the landholding certificate, which provides better land tenure security. In some of the woredas the SLLC adjudication process was performed using the IT tools allowing more secure storage of landholding information. 44 Land Governance Assessment Framework Implementation in Ethiopia: Final Country Report, supported by World Bank, 2016 36 vertical and horizontal organizational structures for land management should be clarified; (ii) mandates for the regulation and management of the land sector should be clarified; and (iii) the establishment of a computerized land information system should be accelerated. The Rural Land Administration System (RLAS) in Ethiopia focusses on the maintenance of the land register (populated through systematic Second Level Land Certification) via continuous updating of the land records from the subsequent land use rights’ transactions. It is supported by a computerized system, the National Rural Land Administration Information System (NRLAIS). 2.2 Land administration activities covered by the Program Nationwide implementation of the Rural Land Administration Information System (NRLAIS) in Ethiopia will generate various positive outcomes, among them sustainable long-term-oriented management of land resources, increase of investments in rural infrastructures and more efficient productivity tools, and, ultimately, resilient and sustainable livelihoods. Until now the issuance and maintenance of SLLCs was done in a semi-manual way in the offices with limited facilities for land records storage and maintenance, no security for records or equipment and rudimentary electricity and public service facilities. Establishing records without the necessary infrastructure and required capacity to maintain them is not sustainable. Poor data management systems at woreda level also poses a risk to landholders’ tenure rights security. Moreover, the cost of maintaining this information is increased by limited data-sharing capability and incompatible technological standards. The NRLAIS is a comprehensive rural cadaster software system for handling mass systematic land registration (SLLCs) and the maintenance of the land register. Digital land/cadastral information can be made available to higher land administration offices throughout the country and the NRLAIS system design makes it possible to use the software at different administrative levels: The functionality and user interface (UI) of the system is different at each level and will effectively have the following components:  Federal/RLAUD (CENLAIS): Supports aggregation and analysis for decision making and national policy development, monitoring of land administration and land use; can act as a portal for rural cadaster and land registration data to the Ethiopian National Spatial Data Infrastructure (ENSDI).  Regions (REGLAIS): Advanced data processing capabilities and management functions for cadastral parcel data, managing of all zones and woredas in the region, providing security for woreda level data-stores (included automated back-up) for analysis of land use and holding, providing basis for regional level policymaking with a regional spatial data infrastructure and supporting woreda in carrying out their mandated duties.  Zones (ZONLAIS): Almost identical to REGLAIS. A web-portal for viewing and carrying out administrative actions at the Zonal level.  Woredas (MASSREG & WONLAIS): As per the technical specification of the NRLAIS, the woreda version will be the most crucial, feature-rich part of the system. It's the heart of NRLAIS, the only place where land can be registered (systematic, sporadic or subsequently) and the only instance of the database with write access. The Property registration sub system (PRSS) is a Web application that has a responsibility of managing transactions in the NRLAIS. PRSS helps users to register applicants` applications to WORLAIS system and manipulate transactions between users. PRSS uses for users not only register applicant`s information in to the system, but also it contributes the replication of data in distributed network to Zone, Region to Federal or Ceneral server.  Kebele level: There will be a paper-based mechanism at the Kebele for submitting applications, issuing receipts and forwarding them to the Woreda for processing and handling the returns. The 37 Kebele will act as a window to get access to the system but will be handling paper documents only for the time being. The IT strategy for rural land administration and the road map document for NRLAIS roll-out confirm that a strong, well-organized IT support structure that works at different administrative levels will be crucial for success. Since NRLAIS will be a new system, it will require continual adjustments and improvements as this is rolled-out and appropriate levels of staffing and training. NRLAIS is a highly decentralized software system, which operates and interacts on four levels and will cover nine regions, 76 zones, around 850 woredas, and more than 13,000 kebeles. The size of the system, and its geographical and administrative span of coverage is likely to introduce considerable implementation challenges. The roll-out of the systems will also need to be supported by work on planning, monitoring, establishment of infrastructure, staff training, software system installation, data cleaning and migration, and maintenance. The NRLAIS roll-out plan also needs to be harmonized with registration projects. Since 2018, NRLAIS has been piloted in 10 woredas financed by LIFT, REILA II, and SLMP II and Government. In January 2019, Government launched the full roll out of NRLAIS in all IPF financed project woredas including in RLLP woredas The following paragraphs depict the key roles and responsibilities of implementing agencies starting from Federal to Kebele level. 2.3 Institutional responsibilities for land administration 38 The institutional hierarchy for land administration are summarized in the figure below and roles and responsibilities are described in further detail in the remainder of this section. 2.3.1 Roles and Responsibilities at Federal Level At the Federal level, the Rural Land Administration and Use Directorate (RLAUD) in the MOA will coordinate the overall roll-out and change management, supported by different projects including the REILA TA. RLAUD will be responsible for the overall planning, implementation, monitoring, reporting, and evaluation of the NRLAIS roll-out across the program regions. The Land Administration and Use Task Team (LAUTT) is one of the working groups under technical committee of the Sustainable Land Management platform (SLM-TC) which provides technical advice and oversight. The task team is chaired by RLAUD director and become instrumental to exchange experiences, coordinate requirements between the projects like LIFT, REILA RLLP, and CALM and to connect to the higher-level exchange with development partners in the SLM-TC. Apart from formal coordination there is also regular informal exchange between the implementing partners. At the Federal level, MOA receives information from the regions, provides stakeholders with consolidated national reports and monitors land governance activities across the country to ensure that all regulations are being properly implemented. RLAUD will also prepares procurement plans that should be elaborated at an early stage, covering practical solutions such as guidelines, framework contracts, to avoid and potential slow down project implementation, especially when the number of delivery items is big, and the delivery locations are distributed. The Federal level NRLAIS IT Support Team is a mixed team composed of RLAUD staff, REILA II experts, RLLP PSU land administration specialists and the MOA’s IT Unit. A contracted NRLAIS Maintenance Service Provider also planned to be hired by RLAUD. In the first 1-2 years, the roll-out will strongly depend on inputs and technical assistance by the REILA II and RLLP TA at Federal level and regional coordinators from REILA, RLLP and LIFT projects. However, towards the 3rd to the 5th year of the program, more responsibilities overtaken by RLAUD and the Regional Land Administration Bureaus. 39 Improved knowledge of system administrators, training materials for ToT, self-training and a series of end- user trainings will be planned and coordinated by RLAUD. NRLAIS manages data with legal status, hence its management from RLAUD needs a transparent process of code changes/maintenance of the system, should go through a controlled process, and management of user requests needs to be traceable. Moreover, RLAUD provides specifications for economical hardware configurations, which is important to roll-out NRLAIS in the program woredas and ensure that infrastructure plans to refurbish premises and procure computing and networking hardware. It also considers procurement, delivery and installation in a controlled manner. RLAUD also ensure that economical NRAIS roll-out in the program regions, zones and Woredas, which operate active NRLAIS installations comprising data cleaning, data migration, and kebele access to NRLAIS. Through the ToTs RLAUD ensures security awareness of stakeholders. It makes it an integral part of the NRLAIS management, identifies and eliminates the most serious risks, provides a policy for end-users and a security plan. RLAUD will also responsible for guidance in the establishment of realistic and suitable coordination structures and knowledge management between RLAUD, Regions, and Woredas. 2.3.2 Roles and Responsibilities at Regional and Zonal Level The Regions play a crucial role in implementation and are responsible to support and supervise the consistent implementation at the Woreda and Zone levels. The regional installation provides backup facilities for the Woreda databases and supports higher level and more complex processing such as monitoring of cadastral information, ensuring integrity of parcel geometry and other technical matters that require professional input and more advanced processing. The regional BRLAU and its subsidiary departments at Zonal level will anchor the overall planning, implementation, monitoring and reporting of NRLAIS roll-out in their respective Regions and Zones. The regions need to be empowered to be able to provide special technical and managerial support wherever it becomes necessary in woreda offices. To be able to do that, each region should set up and strengthen their own IT Support units in order to provide the IT-support services requested by zones and woredas. Much is expected in terms of capacity building activities through training to the Woredas on the use of NRLAIS and its system administration. The procurement of HW and network equipment and supply, configuration, installation, data cleaning, and migration to NRLAIS will be managed by the Regions and the zones. The regions will ensure the security and disaster recovery plans and policies are applied strictly by the program woredas. They maintain and assure the quality and spatial integration of the cadastral fabrics in NRLAIS. They also monitor and provide any technical back up and administrative supports to make sure that NRLAIS is operational in the program woredas. 2.3.3 Responsibilities at Woreda Level The Woreda computer system (WORLAIS) holds the definitive ‘register’ for all the kebeles within its jurisdiction. All transactions will be carried out on the WORLAIS with hard copy changes to the register, and certificates being delivered to kebeles for applicants. The woreda land administration office is responsible for processing land registration and certification, including scanning, geo-referencing, digitizing, editing the parcels, filing in the system, management of field forms and the registry book and issuing of land certificates. The MASSREG provides a standardized software environment for carrying out and supporting the digitization and authentication of SLLCs. The woreda land administration offices as custodian of the SLLCs information, will ensure the implement of the business rules for specific project conditions which may be related to the SLLCs or maintenance of subsequent land transactions in WORLAIS. The Woreda rural land administration offices are responsible in applying business rules for the process of maintenance of the land 40 register (populated through systematic second level land certification, SLLC) for continuous updating of the land records from land use rights’ transactions into WORLAIS. They also provide land information (services) for sector office at woreda and beyond for country development purposes. They strictly apply the specified system security measures and disaster recovery plans and back up system. The woreda RLA offices are also responsible in compiling and providing reports to the zones and regions. 2.3.4 Roles and Responsibilities at Kebele Level Kebele use paper-based systems for submitting applications, issuing receipts and forwarding them to the Woreda for processing and handling the returns. Issuance of Second Level Landholding Certifications (SLLCs) The methodologies for cadastral surveying and mapping for issuance of SLLCs are mainly based on photogrammetric approach supported by ground surveying and adjudication of land parcels using the field maps. The use of orthophotos created from aerial photography or satellite imagery is now a well-established national methodology for the rural cadaster and will be adopted for demarcation and mapping of parcel boundaries and as a basis for a comprehensive adjudication and authentication processes of SLLCs under the Program. The authentication of SLLCs will be powered by MASSREG software which is a subsystem of NRLAIS. MASSREG allows cleaning of SLLCs’ data to comply to a predefined technical and legal business rules in RLAS. It also allows program woredas to migrate SLLCs’ data into NRLAIS seamlessly and equip them with pertinent land administration functionalities that provide mechanisms to validate data authenticity ensure data security and confirm process validity in performing land transactions. The SLLCs procedures and activities are systematically categorized into four major groups of activities implemented in three phases. These are i) the preparatory activities, ii) the field level activities (adjudication of rights, demarcation of parcel boundaries on the field maps, public displays) and iii) office level activities (data processing and tenure documentation activities). The planning, implementation, monitoring, reporting, and evaluation of these activities requires a multi-level and multi-actor coordination both horizontally and vertically. Preparation of PLLUPs The Participatory Local Land Use Planning (PLLUP) approach is a bottom up planning method where land users and relevant stakeholders participate in the process of identifying and prioritizing problems and potentials of a specified land unit for a better use. The approach is based on simple and quick methods of natural resources and socioeconomic data inventory, analysis, evaluation, and decision making on types of land use and management options preferred mainly by rural household land users with technical assistance from Woreda land use planning team. During the SLMP2 the PLLUP was implemented in the 545 kebeles and technical assistance is programmed to further support consultation workshops for land use plans developments at kebele, woreda and higher administrative levels within RLLP45. Whereas the participatory process of their elaboration generated a substantial engagement and interest at community level, their application and usability for the moment is limited since NRLAIS was not yet in place and current tools do not provide incorporating PLLUPs data during the SLLCs preparation and issuance process. To address this systemic shortcoming the existing and newly produced PLLUPs will be incorporated into the adjudication workflows in the NRLAIS. 45 RLLP – Ethiopia Resilient Landscapes and Livelihood Project (P163383) 41 The implementation of the PLLUP activities will be done at kebele and woreda levels with a target to cover 1500 planned areas within five-years period according to the approach established and successfully applied during the SLMP2 project. The approach involves:  Establishment of Integrated Local Level Participatory Land Use Planning Teams at different levels;  Provision of a cascading adequate training from Federal to Kebele and/or micro watershed planning teams;  Provision of the required technical and material resources;  Standardizing scale to be used, and selection and determination of technologies such as topographic maps, aerial photos, ortho-photos, Global Mapper, Google Earth, and GPS, and the corresponding computer software when it is required depending on the capacity of Woreda and higher planning levels.  Identification of a planning unit area and developing a plan based on the socio-economic and biophysical findings;  Enforcing the plan by enacting a land use standard and bylaw by the community; and  Participation of different stakeholders. Experiences and results achieved from planning and implementation of LUPs need to be monitored, evaluated and documented to improve the process and to use the feedback for sectoral planning, strategy development, legal framework reforms, and policy actions. Different level government organizations responsible for Land Use Planning play the core role to bring different stakeholders together to form the planning teams at Woreda and Kebele levels for the development of the PLLUP. The roles and responsibilities of key actors that are expected to be engaged in the SLLCs activities mentioned above are presented in the table below. Active participation and leadership from each institution at all levels is essential for smooth and efficient implementation of the work plans. The table provides an indicative summary and the diagram below represents a general operational framework of activities related to Rural Land Administration supported by the Program. 42 Table 2.3 provides a summary of SLLC activities, roles and responsibilities at different levels in the government structure NAME OF ROLES AND RESPONSIBILITIES INSITITUTION Overall program coordination and management of Results Area 2: Rural Land Administration of CALM:  Coordinate the preparation, implementation, reporting and monitoring of program work plans  Ensure issuance and enforcement of regulatory framework  Conduct legal literacy enhancement workshops and seminars  Coordinate the procurement and supply of goods and services related to SLLCs  Organize and provide TOTs Federal  Harmonizing procedures and quality assurance RLAUD/MOA  Liaise with pertinent federal ministries and agencies such as orthophoto production Overall program coordination and management of Result Area 2: Rural Land Administration of CALM at region and zonal levels:  Coordinate the preparation, implementation, reporting and monitoring of program work plans  Organize and conduct trainings for Woreda staff and contractors  Production and distribution of base maps/field maps when necessary  Coordinating the procurement and supplies of goods and services at regions and below for Regions & SLLCs zonal land  Supervision and quality assurance of the implementation progress of the work administration  Oversight of land dispute resolutions and redress mechanisms Bureaus/offices  Liaison with Regional, Zonal and Woreda administrations Woreda Steer the overall program planning, implementation, monitoring and evaluation Administration  Resolving major implementation challenges & disputes beyond capacity of the woreda land and administration office  Support to Woreda land administration offices for resources its cabinet  Support the know your land right public information and awareness campaigns members Coordinate and manage the planning, implementation and reporting to Zones and regions:  Hire and train contractors for SLLCs  Lead the day to day operational management of SLLCs activities  Liaison with kebele community and conduct public awareness and information campaigns Woreda land on the benefits of land certification and procedures administration  Strengthen and/or establish KLAUCs office  Organize and facilitate training for the KLAUCs 43 NAME OF ROLES AND RESPONSIBILITIES INSITITUTION  Oversight the Kebele LA expert and contractor’s performance  Produce field maps and field registration forms  manage transportations for field work  provide access to existing land information in support of SLLCs  support and solve dispute resolutions at woreda & registration kebeles  conduct daily data quality assurance and security measures both in the field and office data processing and tenure documentation  authenticate and distribute SLLCs to rightsholders  Information dissemination about on-going land certification activities Organize and facilitate community engagement  Organize community meetings  Disseminate information about the on-going activities to the community Kebele  Facilitate the turn-out of landholders to the adjudication site  Support land administration committees and Kebele land administration expert in Administration adjudication land rights and dispute resolution Provide community coordination of the SLLCs activities in their respective Kebeles:  disseminate information about the SLLC plans & on-going land certification activities to rightsholders  Accompany field teams (one member per team)  Verification of field data entry in the field registration form  Facilitate, document, and report land dispute resolutions  Provide information on the benefits of land certification to the community Kebele LAUC  Liaise with the Kebele administration and Kebele Land Administration Expert Kebele Elders Resolve conflicts in the form of arbitrations and mediations Committee Attend Public Information and awareness (PIA) meetings Identify and confirm boundaries in agreement Physical presence during the field work Provide landholding information to contractors (para-surveyors and field recorders) Landholders Verify and provide no objection to the recorded information during public display 44 45 46 Attachment 3: Economic and Financial Analysis 1 Summary of Key Points 1. Program Intervention. Without Program intervention, beneficiaries will continue to struggle to establish or maintain their livelihoods. Continued soil erosion, water insecurity and land insecurity leads to land degradation with direct losses to those that rely on local natural resources for food, livelihood and energy needs. With the IDA investment in the Program, future Sustainable Land Management (SLM) will be built on strong institutions, participatory management, and secure tenure rights. After the 5-year Program costs of US$1.7 billion including US$500 million from IDA, the analysis assumes that the GoE extends its Program into the future. In total, 10,000 micro watersheds are assumed to be implemented over a 10-year period covering 5 million ha. Program costs estimated for a 30-year period amounts to US$9.4 billion including US$729 million for community labor (average 14 million person days per year). 2. Economic Analysis. To assess the ex-ante efficiency of the Program investment, a 30-year cost benefit model is used. Annual cost and benefit flows are estimated as the difference between without- program and with-program net benefits for direct beneficiaries. The Program yields an Economic NPV of US$12.3 billion (US$410 million per year) with a benefit cost ratio of 2.7. The payback period is 7 years and the Economic IRR is 35%, which exceeds the economic discount rate of 5%. Most of the Program benefits are captured on non-cropland (53%), through reduced GHG-emissions (19%) and cropland (14%), with the remaining benefits captured through improved livestock production (6.5%) and avoided soil erosion (6.5%). 3. Economic Analysis excluding GHG Emissions. When excluding the social value of reduced GHG emissions (sometimes referred to as the social value of carbon), the net economic Program return is US$8.5 billion (US$285 million per year) with a benefit cost ratio of 2.2 and an EIRR of 27% and a payback period of 8 years. This is 11% of Ethiopia’s GDP in 2017. 4. Financial Analysis. In financial terms the NPV is US$2.3 billion (US$75 million per year) with a Financial IRR of 26% that exceeds the 12% average cost of capital. The estimated benefit cost ratio is 1.6 with a payback period of 9 years. This estimated net return constitutes 3% of Ethiopia’s GDP in 2017. 5. Increased resilience. The National Poverty Line for Ethiopia is a measure of absolute poverty. The poverty line indicates the money required for food to provide the minimum required caloric intake (Food Poverty Line) and additional non-food items. In the financial analysis, estimated farm-level gross margins can increase by on average US$60/year/person (including the value of production used for home consumption), which is 54% of the Food Poverty Line (US$112/person/year in 2020 terms). This is a direct measure of increased resilience in the Program area. 6. Informing the Risk Management Plan. With the incremental net benefits estimated in this analysis and the additional potential net benefits that could not be quantified, the Program investment is expected to yield considerable returns when accompanied by a strong risk management plan. As part of the risk management plan, the focus needs to be on ensuring that the main decision makers (farmers, other land/resource users, and local institutions) receive the necessary technical advice, land tenure security, and institutional support so that they are able and can afford to make long-term decisions. To avoid increased costs due to Program delays and prolonged implementation periods close monitoring is necessary so as not lose momentum after the Program is complete. Considerations 47 should be made for the equitable inclusion of all beneficiaries in the Watershed User Associations (WSUAs) including vulnerable groups such as landless, youth, elderly, and female-headed households. For example, landless people may suffer from conservation interventions that include restricted access and use of communal lands. Also, restriction on livestock grazing in some areas may lead to increased erosion in other areas unless stocking rates and grazing rotation are incorporated in Watershed Management Plans (WMPs). 7. Rationale for Public Provision and Financing. There is a strong rationale for public intervention as proposed by the Program because it supplies public goods, internalizes externalities, and corrects market failures. The current land degradation issues warrant more targeted public investments to ensure that private sector entities are able to adopt sustainable land management practices for long term benefits. In addition to the direct effect on poverty reduction of the expected increase in farm- level gross margins indicated above, by building resilience to climate change the Program (hereafter ‘the Program’) will reduce the risk to households of falling back into poverty as a result of climate- related shocks, and will contribute to economic stability by reducing the impact of extreme weather events, including both droughts and flooding. 8. World Bank Value Added. Successful interventions to prevent or control land degradation require integrated and cross-sectoral approaches to sustainable land management. The World Bank is well- positioned to support the Program based on the wealth of technical, operational and institutional experiences and lessons learnt through prior and current investment projects. The World Bank contributions are further focused by the Africa Climate Business Plan, now in its fourth year of implementation, and the recently-launched Action Plan on Climate Change Adaptation and Resilience. 2 Background 9. This Annex contains the Economic and Financial Analysis (EFA) of the the Program. The Program Development Objective (PDO) is to increase adoption of sustainable land management practices and expand access to secure land titles in non-rangeland rural areas. 10. Without Program intervention, beneficiaries both in the area and downstream will continue to struggle to establish or maintain their livelihoods. It is expected that without the Program (the counterfactual), land use will continue on its current path. Continued soil erosion, water insecurity and land insecurity leads to land degradation with direct losses to those that rely on crop and livestock production for their energy use and livelihood as well as related industries. Production yields will go down or farmers will have to increase their input costs, such as fertilizer, to maintain current yields. Non-agricultural land in the watershed will also continue to deteriorate without the Program due to soil erosion as well as overuse of common land through livestock grazing and firewood collection. This will put a further strain on the population who derive their livelihood from forests, woodlands, and surrounding areas. Downstream from the Program area, continued land degradation will also affect areas and households through increased flood risk and sediment build-up in irrigation and hydroelectric dams. 11. The Program aims to strengthen the Government of Ethiopia’s (GoE) program that addresses land degradation: the Ethiopia Strategic Investment Framework (ESIF). Since 2010 ESIF has guided efforts to address land degradation, reduce vulnerability to climate chocks, provide land tenure security, and address institutional knowledge and capacity constraints at local, regional and 48 national levels. Phase 3 of ESIF runs from 2019 to 2024 and provides the operational program for meeting the second Growth and Transformation Plan (GTP-II) targets for SLM. 12. ESIF phase 3 aims to deliver durable management of 5,000 micro watersheds by 2024. To achieve this, the program will facilitate through extension services the establishment of WSUAs, the preparation and regular update of participatory WMPs, and support to ensure these WSUAs are effective, their plans implemented, and their efforts monitored. Under the program, the government will also scale up its support for land titling, through the provision of Second-Level Landholding Certificates (SLLCs), and the establishment of a digital land registry, the National Rural Land Administration Information System (NRLAIS) to ensure security of land records and facilitate provision of land title information services. 13. ESIF phase 3 also introduces a significant and progressive change of course in the way in which labor is mobilized to achieve SLM objectives. It notes that community labor is essential for watershed development and management, and the MoA has therefore agreed that it intends to replace the current program of Mass Mobilization with participatory community planning and implementation for watershed development and management. It will ensure that community labor for SLM activities is provided on a voluntary basis in a participatory manner. 14. The Program will deliver performance-based financing for selected elements of the ESIF program. This will enable SLM support to be delivered to a higher quality and at a national scale covering: Field-based investment to scale up SLM; Improvements to land tenure security; Capacity building for SLM; The policy framework for SLM; Building the SLM knowledgebase; and Management of ESIF. As explained further in the main body of the Program Appraisal Document (PAD), the Program excludes three main elements for the GoE ESIF Program expenditures: i) support for irrigation; ii) activities in Afar and Somali regions; and iii) expenditures already financed by development partners. 15. Table 1 shows that the assumed Program costs in first 5-year period are US$1,676 million of which US$500 million is from IDA and the remainder is from GoE. An estimated financial value of community labor equal to US$83 million (47 million person days) for the same period is included in the analysis. 16. It is the intention of the GoE to extend its Program efforts into the future and achieve durable management of 5,000 more micro watersheds (10,000 total). Assuming that implementation of these target watersheds takes 10 years, an appropriate analysis period is 30 years to account for a full forest rotation period for 20 years. For this analysis, and as illustrated in Figure 1, Program costs are assumed to increase annually by 10% (including 2.5% inflation) while the number of micro watersheds under implementation increases (year 6). As more micro watersheds enter the maintenance phase, Program costs gradually decrease to 40% in year 30 as the Government’s focus shifts from SLM to other issues. Program costs estimated for the 30-year period amounts to US$9.4 billion including U$729 million for community labor (average 14 million-person days per year). 17. The analysis includes gradual implementation of 10,000 WSUA and 5 million ha by year 13, after which the watersheds are maintained for sustainable future net benefits. For the purposes of this analysis it is assumed that the program will target 500 major watersheds across all regions. 46 Each major watershed will eventually have an average of 20 micro watersheds, at an average size of 46From a list of 702 woredas, the Program team selected 500 woredas with the smallest share of grassland. This is to emulate the Program’s relative focus on non-rangeland areas. 49 500 ha, with each their WSUA and WMPs. Over an 8-year period WSUAs are started and their WMP is implemented in the following year. See the build-up in Table 2. Implementation of each micro watershed takes 5 years ramping up by 20% annually. Therefore, the last micro watersheds to start implementation and treatment can be expected to complete in year 13 when the 5 million ha target is reached. This constitutes 23% of the GoE’s target to achieve broader landscape restoration on 22 million ha by 2030. Table 3 shows the assumed distribution of 5 million ha across regions and Land Use Categories (LUCs). The overall distribution across LUCs is the same as across the 49.9 million ha which the 500 woredas cover (32% cropland, 21% woodland, 20% bush+shrub, 14% forest, 7% grassland, and 6% other). 18. The Program excludes costs and benefits related to the existing Resilient Landscapes and Livelihoods Project (RLLP). According to the RLLP PAD the target for 152 watersheds covering an area of 1.5 million ha is to achieve 972,000 ha with SLM practices (65%). The Program is expected to operate in 500 major watersheds covering 10% of 49.9 million ha. Therefore it is assumed that both the Program and RLLP can implement activities in different micro watersheds within the same macro watersheds, thus avoiding double-counting costs and benefits.47 Table 1: Program Costs Included in EFA, Financial values US$ million, after exclusions Total Total Total Total Year 1-5 Year 6-10 Year 11-30 Participatory Watershed Management 1,231 1,641 5,488 8,360 Rural Land Administration 445 588 1,968 3,001 CALM Program costs including inflation 1,676 2,229 7,456 11,361 CALM Program costs excluding inflation 1,631 1,985 5,004 8,620 Estimated community labor - Cumulative watersheds, million ha 2.1 5.0 5.0 - Cumulative # watersheds 5,040 10,000 10,000 - Million person days, development 44 105 26 176 - Million person days, maintenance 3 27 208 237 Community labor, person days 47 132 234 413 Community labor costs excluding inflation 83 233 413 729 Total Program costs in EFA excluding inflation 1,714 2,219 5,418 9,350 Notes: - Program costs increase annually by 10% (incl. 2.5% inflation) in years while number of micro watersheds is increasing. When number of watersheds under treatment starts decreasing (watersheds are then under maintenance) Program costs decrease gradually by 40% in year 30 as the Government’s focus shifts from SLM to other issues. - SLMP 2003-2006 used 1.4 billion person days on 19,807 watersheds = 17,556 person days per watershed per year. - Financial unskilled labor rate is US$1.8/person day (ETB 49.5/ person day) - Watershed maintenance costs 6% of development costs based on SLMP average development costs of 200 person days/km versus annual maintenance costs of 12 person days/km. 47 Detailed review of the EFA models for RLLP and CALM confirms that there are sufficient number of hectares in all 500 major watersheds to implement both those that are included in RLLP and the CALM Program EFAs. 50 Figure 1: Annual Program Costs Included in Analysis, Financial Values 500.0 1,400 # of watersheds under implementation 450.0 1,200 400.0 constant million USD 350.0 1,000 300.0 800 250.0 200.0 600 150.0 400 100.0 200 50.0 0.0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Community Labor Rural Land Administration (after exclusions) Participatory Watershed Management (after exclusions) Total # of WMPs under implementation by year (under maintenance afterwards) Note: Financial values including physical contingencies, excluding inflation. Assumptions made by the Program team for the 30-year EFA. Table 2: Watershed Management Plans under Implementation Year # woredas, # micro hectares # micro hectares major watersheds treatment watersheds under watersheds started started under treatment started treatment 1 0 0 0 0 0 2 63 1,260 642,992 252 128,598 3 126 2,520 1,193,419 756 367,282 4 189 3,780 1,713,288 1,512 709,940 5 252 5,040 2,118,678 2,520 1,133,675 6 315 6,300 2,745,006 3,780 1,682,676 7 378 7,560 3,343,490 5,040 2,222,776 8 439 8,780 4,576,228 6,292 2,899,338 9 500 10,000 5,000,000 7,536 3,556,680 10 500 10,000 5,000,000 8,528 4,132,945 11 500 10,000 5,000,000 9,268 4,583,944 12 500 10,000 5,000,000 9,756 4,915,246 13 500 10,000 5,000,000 10,000 5,000,000 Note: - One major watershed per woreda. 20 micro watershed per woreda. 500 ha per micro watershed. Each micro watershed is implemented over time (20% per year from start of implementation). - Because 71 woredas have less than 10,000 ha (20 WSUA x 500 ha) gradual build-up by year does not completely match the 2.5 and 5 million ha targets between year 1 and 9. The other watersheds are assumed to be greater than 500 ha to make up the difference. Table 3: Area Targeted for Treatment by Land Use Category and Region Region, # Major Forest Woodland Bush Cropland Grassland Bareland Total Share hectares water- +Shrub Treated sheds Area Tigray 39 14,635 27,779 81,795 108,450 12,946 42,949 288,554 5.8% Gambella 6 48,842 19,140 26,870 6,528 12,471 3,617 117,468 2.3% 51 Amhara 82 49,685 127,606 169,444 347,127 70,312 146,051 910,225 18.2% Oromia 217 345,694 535,838 499,442 815,961 159,705 76,068 2,432,707 48.7% Benishangul 13 53,360 153,495 101,777 38,574 41,059 1,429 389,695 7.8% Gumuz SNNPR 116 180,780 186,625 130,887 279,959 63,843 12,774 854,869 17.1% Harari 9 110 203 486 2,007 235 299 3,339 0.1% Dredewa 18 30 547 579 1,017 264 706 3,143 0.1% Total 500 693,136 1,051,232 1,011,279 1,599,624 360,835 283,894 5,000,000 100.0% Total 6,919,753 10,494,707 10,095,851 15,969,446 3,602,309 2,834,184 49,916,251 woreda Share 13.9% 21.0% 20.2% 32.0% 7.2% 5.7% 100.0% Note: Area size by LUCs extracted from remote-sensed data of the Program watersheds. LUCs Marsh, Swamp, Water, Alpine, and Settlements are included under bareland. 3 Methodology 19. A cost benefit model is used to assess the ex-ante efficiency of the Program. The included budget items of the Program are considered necessary to obtain the target impact; therefore, the entire Program cost as set out in Table 1 is included in this analysis. Over the first 5 years this includes US$1,676 million plus US$83 million as in-kind contributions from community labor less US$45 million price contingencies. Over the 30-year period total Program costs are estimated at US$9.4 billion including community labor and excluding price contingencies. Incremental net benefits are estimated as the difference between without-program (WO/P, or the counterfactual) and with- program (W/P) net benefits for direct beneficiaries. Figure 2: Illustration of Incremental Benefits 1,000 Change in Yield, kg/ha 500 0 -500 7 1 3 5 9 11 13 15 17 19 21 23 25 27 29 Year Program gains due to avoided losses Program gains due to yield increase WO/P Yield Trajectory W/P Yield Trajectory 20. Figure 2 illustrates how this analysis assumes a declining production without Program interventions due to soil erosion. With Program interventions the yield loss is avoided and, for some production systems (crops, livestock, and grassland), with-program yields increase over time. This yield increase is attributed to adoption of improved cultivars, improved seeds, better animal breeds, better animal 52 feed, land restoration, water management, and implementing climate smart agricultural techniques. The sum of the two shaded areas in Figure 2 constitute the incremental Program benefit. 21. All Program activities are associated with both costs and benefits. The following incremental net benefit flows are expected – some of which are quantified while others are discussed qualitatively. a) Direct net benefits to crop producers: The EFA quantifies the incremental improvement in gross margins for different crops and cropping patterns on farms in the targeted watersheds. The increment also includes transformation of a limited area from unproductive bareland to cropland along field margins, eroded gullies and so on. A portion of the incremental benefit is due to avoided soil erosion. This estimate is calculated separately from the impact on yield (see illustration in Figure 1). b) Direct net benefits to livestock producers: The EFA quantifies incremental improvements in gross margins for different livestock production systems and stocking rates on farms in targeted watersheds. c) Direct net benefits to forests and other non-croplands: The EFA quantifies the net improvement in gross margins for different categories of land use including forests, woodland, bush+shrub, and grassland. It is estimated that some areas of bareland is converted to forest plantations and grassland. A portion of this incremental benefit is due to avoided soil erosion. This estimate is calculated separately from the impact on yield (see illustration in Figure 1). d) Social value of impact on GHG emissions: The Program impact on GHG emissions is estimated using FAOs Ex-Ante Carbon-balance Tool (EX-ACT) model. This considers changes in afforestation, land use, land restoration and input and energy use. There are no actual payments of equivalent carbon credits to beneficiaries in this Program, so the social value of carbon is included only in the economic analysis and not in the financial analysis. In line with World Bank guidance note, this analysis refers to the “shadow price of carbon� (US$/tonne CO2-eq) being multiplied with the GHG emission reductions (tonne CO2-eq) to estimate the “social value of GHG emissions�. Other reports may refer to the social value of carbon, the social cost of carbon, impact on carbon balance, or carbon sequestration.48 22. The following incremental net benefit flows are not quantified explicitly in this analysis: e) Improved capacity and skill-level in public institutions: The net benefits estimated for Program activities cannot be successfully captured without the investment to strengthen local, regional and federal institutions. These institutions will provide research and improved extension, advisory and training services to stakeholders in support of WMP implementation. They will also build the knowledgebase and monitor Program achievements. However, the direct benefits from improved skill-level among public service employees is not quantified in this EFA. f) Improved capacity and skill-level in community institutions (WSUAs): Working through local institutions reduces Program costs and improves sustainability if costs can be funded in the 48World Bank (2017b) Guidance note: Different documents have been using different terms to refer to the price of carbon used in economic analysis (shadow price of carbon, social cost of carbon, and social value of carbon). These terms refer to different approaches to calculate the price of carbon. The guidance note uses the term “shadow price of carbon,� which is the price of carbon consistent with a given climate objective, as estimated for the High-Level Commission on Carbon Prices, led by Joseph Stiglitz and Nicholas Stern. 53 future. The Program aims to strengthen these institutions so that the captured benefit streams can continue into the future. While successful capacity building of WSUAs is necessary to capture benefits included the EFA, the direct benefit from improved skill-level among association members is not quantified. g) Equitable benefit distribution with participatory WMPs and tenure rights: With participatory WMPs benefits can be captured equitably among all watershed stakeholders, for example by: incorporating income-generating activities for landless and other vulnerable people; ensuring land titling for female headed households; and joint land certificates for married couples. Direct benefits from these activities are not quantified in this EFA. h) Potential benefits from conservation areas, biodiversity and tourism: The lack of secure tenure rights creates a dis-incentive for beneficiaries to undertake productive investments and adopt sustainable management practices. This is particularly the case when benefits accrue over a longer period of time. While some direct benefits are quantified as described earlier, other benefits are not quantified due to lack of data. These other benefits could include conservation of protected areas, biodiversity, and tourism. i) Additional benefits and costs of “cut & carry� livestock management system: While the current EFA quantifies impact on livestock management through improved gross margins, other associated costs and benefits are not included. It is expected that WMPs will restrict livestock from grazing freely in communal areas. This will lead to increased costs of tethering/fencing, collecting fodder, and purchasing fodder - while also saving costs on herding. A co-benefit of this may be increased opportunities for children to attend school as they are often charged with herding livestock. Increased adoption of this “cut & carry� feeding system may also have an impact on the local value of fodder crops. To meet the increased demand, fodder production can therefore provide new income- generating opportunities for beneficiaries. There is a risk that localized land degradation will occur in areas where livestock are tethered or fenced in. WMPs should therefore include plans for rotating grazing areas to minimize soil erosion. j) Diversified livelihoods and improved employment opportunities: It is expected that WMPs will include promotion of income-generating activities. To mention a few, this could include: input and service provision for agriculture sector; post-harvest services such as storage and processing facilities; forest and non-cropland related businesses; transport services; and tourism. These additional benefits can capitalize on those already captured in this analysis including through improved food- and income diversification for beneficiary households. k) Indirect benefits to other local areas: Several of the incremental benefits quantified as described above will likely have other indirect benefits. For example, these include the adoption of climate smart agriculture and land restoration techniques in neighboring watersheds due to informal dissemination outside the Program area. Other industries and employment opportunities may increase through a multiplier effect to other areas and related sectors. Due to lack of data these are not quantified in the EFA. l) Downstream effects: Downstream effects or externalities from reduced soil erosion are also not quantified due to lack of data. These benefits could include reduced risk of downstream flooding and reduced costs of sediment build-up in downstream irrigation and hydroelectric dams. m) Improved nutrition: Incremental benefits from improved nutrition have not been quantified other than through value of increased production yields. The value of a more varied food production 54 has not been estimated. This would be a direct benefit to Program beneficiaries and indirect benefit to people in neighboring areas. 23. Efficiency and other cost-benefit indicators. As shown in detail in the financial analysis below, the cost-benefit analysis is based on crop-, and farm-level data on yields, prices and costs in constant 2020 currency amounts for without- and with-program situations.49 Data are also used to establish cropping/livestock patterns representative for each region. Crop- and farm-level assumptions are shown in tables in Appendix 1. The Economic Net Present Value (ENPV) is calculated using the World Bank recommended discount rate of 5%.50 The financial discount rate used is 12% to reflect the opportunity cost of capital in Ethiopia. The foreign exchange rate used is 1 US$ = 28 ETB. In addition to sensitivity analyses of these discount rates, the break-even rates are calculated, i.e. the Economic and Financial Internal Rates of Return (EIRR and FIRR). Other indicators included are (units in parentheses): (a) Land area restored, reforested and afforested with sustainable management practices (hectares) (b) Target area in different land use categories (hectares) (c) Reduced soil erosion (tonnes) (d) Production and income in representative farm households (yield and US$) (e) Impact on GHG emissions (CO2-eq and US$) 24. Results are aggregated to different levels for further analysis. Using data collected and assumptions made by the Program team laid out in Appendix 1, the methodology goes further than the total Program results to enable analyses at different levels of aggregation: (a) At the base of the model are data on per hectare gross margin for annual crops and avoided yield loss from soil erosion. Gross margins for livestock production are calculated per head of animal. (b) Representative farms are defined in terms of farm size and combinations of different annual crops and livestock. This enables an analysis of estimated impact on incremental farm household income. (c) Incremental net benefits on non-cropland are estimated at the watershed level. This includes any projected per hectare changes in gross margins as well as avoided yield loss from soil erosion. (d) Total impact on the GHG emissions is estimated at the Program level and allocated proportionally between watersheds. 25. Conversion factors for economic analysis. An economic analysis is concerned with value addition to the gross domestic product and excludes all transfer payments such as taxes, subsidies, grants, loans, interest- and principal payment paid to or received from beneficiaries. Because none of the agricultural products quantified in this model are imported or exported, the farm gate prices are applied both in the financial and economic analysis. The opportunity cost of unskilled labor is set to 0.75 due to limited alternative employment opportunities. It is expected that some agricultural and construction inputs are imported and should be converted from farm gate prices using an economic 49 Details provided in Appendix 1 are based on SLMP-II baseline study (MOANR/MOALR 2016), SLMP-II gross margin study (Große-Rüschkamp, 2015), and data collected for the RLLP Project Appraisal Document (World Bank, 2018b). Ethiopia Consumer Price Index to adjust 2014 to 2020 prices = 172, and adjusting 2018 to 2020 prices = 120.7. 50 World Bank (2015a). Technical Note on Discounting Costs and Benefits in Economic Analysis of World Bank Projects. Washington, DC. 55 conversion factor (assumed to be 0.95). Much of the variable costs included in the analysis is unskilled labor. Therefore, average conversion factors are used for cropland variable costs (0.98), and non-cropland variable costs (0.88). All other cost assumptions are maintained from the financial analysis. As noted before, price contingencies for the first 5 years estimated at US$45 million or 2.8% of the Program budget is excluded from the analysis. 26. The Program’s impact on GHG emissions is estimated using the FAO’s EX-ACT model. The economic value of the Program’s impact on the GHG emissions and therefore the carbon balance is estimated from activities in the Program including: bio-physical structures on degraded land; afforestation and reforestation; promoting sustainable land management; promoting climate-smart agriculture practices; introducing improved grassland management; and enriching forest areas with different tree species and improved water management. The total reduced GHG emissions is multiplied by the assumed shadow price of carbon from US$40 per tonne CO2-eq in early years increasing to US$76 at the end of the 30 years.51 27. Sensitivity analyses identify key assumptions that should be the focus of risk management efforts. Three different approaches are used: i) switching values, when a change in an assumption leads to a break-even ENPV, are calculated for most assumptions. ii) Elasticities are also calculated for most assumptions to show how much a 1% change in an assumption changes total ENPV; and iii) Specific cases are analyzed to further highlight key risk factors and quantify the impact of variables that cannot be analyzed with switching values and elasticities. 4 Assumptions and Results 28. Before analyzing the economic value results, the underlying assumptions are discussed starting with a financial analysis of farm-level target beneficiaries. Except where noted, the assumptions used are the same as in the recent PAD for the RLLP (World Bank, 2018b). Note that the value of reduced GHG emissions is not included in the financial analysis because payments for carbon credits are not expected to be distributed directly to beneficiaries during the Program. 4.1 Financial Analysis 29. Program interventions are assumed to lead to improved crop, livestock, and non-crop gross margins providing there is sufficient support and WMP maintenance (see Appendix 1 for detailed gross margin assumptions). With sufficient long-term maintenance, gross margin improvements accrue over a period of 7 years on cropland and non-cropland.52 For livestock the improvements occur over 1-3 years. For the purposes of this analysis, the first year of improvements is excluded and instead attributed to prior investment operations in the Program area. The remaining years of improvement are attributed to the Program achievements. Furthermore, it is assumed that there is a 10% post-harvest yield loss due to insufficient crop storage and handling capacity. It is assumed that gross margins do not change on most non-cropland areas due to Program interventions. The exception is that biophysical treatment of grassland will improve estimated gross margins. Finally, as a proxy for variability, benefits on both the with- and without-Program situations are 51 Current World Bank Guidelines suggest a shadow price of carbon of US$40 per tonne CO2-eq in 2020 building up 2.25% per year to US$80 per tonne in 2050. World Bank (2017b) Shadow price of carbon in economic analysis. Guidance note to the World Bank Group staff. Washington, DC, Nov 12. 52 Schmidt and Tadesse (2017) suggest that there are no measurable improvements in productivity from SLMP, although the authors also acknowledge that there are problems with the data: They found that, over the analysis period, value of production increased significantly in both treatment and non-treatment areas 56 reduced by 10% every 5 years due to an extreme weather event.53 Representative farm models are used for each region based on data for cropping pattern and number of livestock per household.54 30. Estimated farm-level gross margins increase and build resilience by an average of US$60/year/person including home consumption, which is 54% of the Food Poverty Line. Table 4 shows farm-level income increases by 16-45% on different representative farms. When assuming 5 persons per household farm gross margin can increase at least US$50 per household member per year including value of production used for home consumption. This is a direct measure of increased resilience. To associate this result with a measure of absolute poverty, we use the National Poverty Line for Ethiopia. The poverty line indicates the money required for food to provide the minimum required caloric intake (Food Poverty Line) and additional non-food items. The improvement in farm gross margin is around 44% of the Food Poverty Line in 2020 terms (US$112/person/year).55 This improvement is also about 23% of the total National Poverty Line (US$214/person/year). Other representative farms are estimated to capture higher growth in gross margins of up to US$77/person/year or 68% of the Food Poverty Line. 31. The Program activities help avoid yield losses caused by soil erosion. This avoided yield loss is valued based on the gross margin data on different land uses. To establish the linkage between reductions in soil erosion with the Program activities, a Universal Soil Loss Equation (USLE), adapted to Ethiopian conditions, was used to model soil loss associated with each of the technologies. The USLE relates soil loss from a field to local climatic conditions, soil type, topography, and land and crop management variables. Annual soil loss is given as a function of the rainfall erosivity of a given soil type, the slope length, crop cover factor, and the conservation practice on the land. Based on a review of studies linking soil loss to productivity loss, it is assumed that watersheds with between 10 and 25 tonnes/ha/year soil loss experience a 0.5% yield loss. When between 25 and 35 tonnes soil/ha/year are lost, yield loss is 1%, and increasing in 0.5% increments up to 2.5% if soil loss ranges between 55 and 65 t/ha/year (maximum average soil loss in included watersheds is 60 t/ha/year).56 Table 4: Representative Farm Models, Crop + Livestock Gross Margins, Financial Analysis Farm A Farm B Farm C Farm D Farm E Farm F Average Annual Farm Gross Tigray Gambella Amhara Oromia Benishangul SNNPR Margin Harari Gumuz Dredewa WO/P Avg. 1,090 1,197 1,257 1,026 1,064 1,370 W/P US$/ha/ 1,276 1,458 1,475 1,235 1,255 1,658 53 Over the past three decades Ethiopia has experienced many localized droughts and seven major droughts, five of which have been associated with famines. 54 In reality, cropping patterns are driven by demand and supply. However, the EFA model is deterministic and does not include a dynamic adjustment of cropping patterns between years and different farmers. The assumptions are based on data from SLMP-II baseline survey (MOANR/MOALR, 2016) and the RLLP Project team’s best judgement. 55 The 2011 National Poverty Line was 3,781 ETB/adult while the Food Poverty Line was 1,984 ETB/adult leaving Non-food Poverty Line of 1,796. Ethiopia Central Statistical Agency (2014) and World Bank (2015b). It is assumed that a household of 5 persons is 3.1 adult equivalents. These are converted to 2020 amounts using CPI factor 2.48. 56 Paganos et al. (2018) provide a review of many references on this topic showing ranges of productivity loss from less than 1% to over 20%. Gebreselassie et al. (2016) also refer to potential productivity losses of 10-30% due to soil erosion. 57 Change year 187 261 218 209 191 287 % Change 17% 22% 17% 20% 18% 21% WO/P Avg. 1,458 1,602 1,682 1,373 1,424 1,833 W/P US$/farm 1,708 1,951 1,974 1,652 1,679 2,218 Change /year 250 350 292 280 255 385 WO/P Avg. 292 320 336 275 285 367 W/P US$/perso 342 390 395 330 336 444 Change n/year 50 70 58 56 51 77 Note: - WO/P = Without Program (Baseline); W/P = With Program; - Includes value of home consumption (e.g. grains, vegetables, fodder, milk, eggs). Costs exclude farmer's own labor. - Average 5 household members per farm. Average farm size 1.338 ha. Exchange rate: 1 US$ = ETB 28 - Oromia assumptions used as representative for Harari and Dredewa regions. - Source: Analysis results based on crop and livestock assumptions. 32. Aggregation to Program level provides an estimated return on investment in financial terms. Net benefits as described above are aggregated up to represent the entire area of crop- and non- cropland in each watershed and in the Program overall. The aggregation includes the 150,000 ha (3% of Program area) of bareland transformed to forest plantation and 5,000 ha of bareland transformed to each of cropland and grassland. Because the target area of 5 million ha is 10% of the area in the selected 500 woredas, the same share is used to estimate the number of beneficiaries to be 6.1 million (based on the latest data woreda population in 2007 was 60.8 million). 33. The Program’s overall Financial NPV is US$2.3 billion with a Financial IRR of 26% and a benefit cost ratio of 1.6. The payback period is 9 years (see Table 5). In financial terms, the Program therefore meets the requirement by yielding a rate of return higher than the financial discount rate of 12%. This estimated net return constitutes 3% of Ethiopia’s GDP in 2017 and is also US$15 per year per ha when including the entire Program area of 5 million ha.57 The National Poverty Line for Ethiopia is a measure of absolute poverty. The poverty line indicates the money required for food to provide the minimum required caloric intake (Food Poverty Line) and additional non-food items. The FNPV measured per person per year is US$12/year/beneficiary which is 6% of the National Poverty Line and 11% of the Food Poverty Line (for 6.1 million beneficiaries). This is a direct measure of increased resilience in the Program area. Table 5: Economic and Financial Analysis – Key Efficiency Indicators Scenario 1 Scenario 2 Scenario 3 Base Case Economic Economic Analysis Financial Analysis excl. GHG Emissions Analysis 57 World Development Indicators database. GDP data for Ethiopia as of 2017. Accessed 13 March 2019. 58 US$ million Program Budget 1,676 1,676 1,676 US$ Budget per Program area ha 335 335 335 US$ Budget per Program beneficiary 275 275 275 Net Present Value, million US$ 12,294 8,549 2,260 Benefit Cost Ratio (BCR) 2.7 2.2 1.6 Internal Rate of Return (IRR) 35.36% 27.31% 25.98% Payback Period 7.2 Years 8.3 Years 8.6 Years NPV as share of 2017 GDP 15.3% 10.6% 2.8% NPV, US$/ha 2,458.77 1,709.72 452.00 NPV, US$/year 409,794,309 284,953,482 75,333,930 NPV, US$/year/ha 81.96 56.99 15.07 NPV, US$/year/household 336.51 233.99 61.86 NPV, US$/year/beneficiary 67.30 46.80 12.37 Share of National Poverty Line 31% 22% 6% Share of Food Poverty Line 60% 42% 11% Note: - In this table, Program budget excludes beneficiary contributions and recurring costs but includes price contingencies. The benefit cost analysis includes beneficiary contributions and recurring costs but excludes price contingencies. - Economic discount rate = 5%. Financial discount rate = 12%. Analysis period is 30 years. - Total 30-year GHG Emissions from EX-ACT model = 151.3 million tonne CO2-eq. Economic CO2- eq value = US$40/tonne to US$76/tonne by year 30 equivalent to ENPV of US$3,745 million. - Program area = 5 million ha. Number of beneficiaries (population) in Program area = 6.1 million or 6% of Ethiopia's population. Number of people per household = 5. - 2011 National Poverty Line, Year 1 amount = US$214 /person/year. 2011 Food Poverty Line, Year 1 amount = US$112 /person/year. 2017 GDP (current million US$) = 80,561. 2017 Agriculture GDP (current million US$) = 27,389. - Exchange rate: 1 US$ = ETB 28 4.2 Economic Analysis 34. As explained earlier, prices and costs used in the financial analysis are adjusted to value the economic impact of the Program. The economic net benefits also include a valuation of the Program’s impact on GHG emissions. Investment costs include: the Program budget (excluding price contingencies); community labor (beneficiary in-kind contributions); and annual costs after the Program is complete. 35. The Program interventions are expected to have a net-benefit on GHG emissions to the amount of 151 million tonnes of CO2-eq over 30 years, which constitutes a discounted value of US$3,745 million. GHG emission calculations using the EX-ACT model are done for a 10-year implementation period and a total 30-year time frame. The results (Table 6) show that the Program constitutes a net carbon sink of 151 million tonne CO2-eq emissions over a period of 30 years, resulting in 5 million tonne CO2-eq per year or 30 tonne CO2-eq/ha. Figure 3 illustrates that most of the carbon sequestration comes from afforestation and improvements to cropland, woodland, bush, and shrub. Together with increased use of fertilizer, herbicide, and diesel as well as building construction, the Program is estimated to constitute a net sink of GHG emissions. Ethiopia has committed to an emissions reduction target of 255 million tonnes CO2-eq or 64% compared to “business-as-usual� 59 emissions by 2030.58 The Program has potential to reduce GHG emissions by 35.6 million tonnes of CO2-eq by 2030 -14% of the national reduction target. 36. The ENPV is US$12.3 billion discounted at 5% over a 30-year period (US$410 million/year). This generates a benefit cost ratio (EBCR) of 2.7 and an EIRR of 35% with a payback period of 7 years (see Table 5). In economic investment analyses, the Program therefore meets the requirement by yielding a rate of return higher than the economic discount rate of 5%. When excluding the social value of GHG emissions, the net Program return is US$8.5 billion (US$285 million/year) with a benefit cost ratio of 2.2 and an EIRR of 27%. Without the impact of the GHG emissions, the payback period is 8 years. This is 11% of Ethiopia’s GDP (in 2017 terms). Without the reduced GHG emissions, the ENPV is US$57/year/ha or US$47/year/beneficiary. Relative to the measure of absolute poverty, this is 22% of the National Poverty line and 42% of the Food Poverty Line. The annual cost and benefit flow for the Program as a whole including with impact on GHG emissions are shown in Table 7 and illustrated in Figure 4. 37. As summarized in Table 7, the Program benefit streams quantified in this analysis are those captured on: non-cropland (53%), GHG-emissions (19%), cropland (15%), livestock (6.5%), and avoided soil erosion (6.5%). More analyses are required to allocate Program investment costs to each benefit flow to enable assessing their relative net benefits and EIRRs. 58 Ethiopia’s commitments included in the Nationally Determined Contribution (NDC, submi tted to the UNFCCC in 2017). https://www4.unfccc.int/sites/ndcstaging/PublishedDocuments/Ethiopia%20First/INDC-Ethiopia-100615.pdf 60 Table 6: Economic Analysis – Greenhouse Gas Mitigation Potential Program Activities, million tonnes CO2-eq Without With Program Net GHG emission Positive = source / negative = sink Program Balance Land Use Changes: Afforestation 0.0 -66.4 -66.4 Other 0.4 -0.9 -1.3 Land Use: Cropland 0.0 -34.5 -34.5 Woodland 0.0 -24.0 -24.0 Bush + shrub 0.0 -23.1 -23.1 Grassland 0.0 -14.0 -14.0 Degradation 0.0 0.0 0.0 Inputs & Investments 0.0 12.0 12.0 Total 0.4 -150.8 -151.3 Per hectare 0.1 -30.2 -30.3 Per hectare per year 0.0 -1.0 -1.0 Note: - EX-ACT model Version 5.2 – Standard Edition. Tropical Montane climate. Moist regime. Dominant soil type: HAC soils. Implementation phase 10 years. Capitalization phase 30 years. Dynamics of implementation are assumed linear over the Program period. Default Tier 1 coefficients are used. - Calculations from EX-ACT model based on the following assumed evolutions in land use/category: Land Use, 1,000 hectares Initial State Without Program With Program Forest/Plantation 693 693 843 Cropland 1,600 1,600 1,605 Woodland/Bush/Shrub 2,063 2,063 2,063 Other (marsh, swamp etc.) 54 54 54 Grassland 361 361 366 Degraded Land 230 160 0 Other Land 0 70 70 Total Area 5,000 5,000 5,000 - Woodland and Bush + shrub land use are included with improved water management. - Marsh, Swamp, Water, Alpine, and Settlements are included under “Other�. - Includes transformation of degraded bare land: 150,000 ha to Forest/Plantation, 5,000 to cropland, and 5,000 ha to grassland. - Assumed inputs include 37,862 tonnes of nitrogen from urea, 1,514 tonnes of active ingredients per year from herbicides, 6,399 m3 of gasoil/diesel per year, and 198,777 m2 of concrete agricultural buildings. Figure 3: Economic Analysis – Greenhouse Gas Mitigation Potential Estimated Impact on GHG Emissions, 30-year period 20.0 10.0 0.0 million tCO2-eq -10.0 -20.0 -30.0 -40.0 -50.0 -60.0 -70.0 -80.0 61 Table 7: Economic Analysis - Estimated Annual Flow of Benefits and Costs (US$ million) Year Crop Crop Non- Non- Livestock Social Program Variable Increment Discounted Benefit Benefit cropland cropland Benefit Value Costs Costs al Net Incremental without due to Benefit Benefit due of Benefits Net Benefits avoided avoided without to avoided Carbon soil erosion soil soil erosion soil erosion erosion 1 0 0 0 0 0 0 -282 0 -282 -268 2 2 0 36 0 4 28 -304 -6 -240 -218 3 9 1 109 1 11 53 -338 -20 -173 -150 4 22 4 219 3 22 78 -372 -40 -65 -53 5 16 8 365 5 -37 98 -398 -67 -10 -7 6 78 15 465 8 54 130 -433 -89 227 170 7 117 24 598 11 71 162 -435 -117 431 306 8 157 34 767 14 90 227 -437 -151 701 475 9 197 44 892 18 108 291 -438 -177 934 602 10 197 49 913 19 69 298 -417 -186 942 578 11 267 64 931 21 133 304 -396 -194 1,130 661 12 292 72 946 23 140 311 -376 -200 1,208 673 13 308 79 957 25 144 318 -356 -204 1,270 674 14 313 84 959 27 144 325 -336 -205 1,312 662 15 282 81 959 28 129 333 -322 -205 1,285 618 16 313 96 959 30 144 340 -309 -205 1,368 627 17 313 101 959 32 144 348 -296 -205 1,396 609 18 313 107 959 34 144 356 -284 -205 1,423 591 19 313 112 959 35 144 364 -273 -205 1,449 574 20 282 106 959 37 129 372 -262 -205 1,418 535 21 313 123 959 39 144 380 -251 -205 1,502 539 22 313 128 959 41 144 389 -241 -205 1,527 522 23 313 133 959 42 144 397 -231 -205 1,553 506 24 313 138 959 44 144 406 -222 -205 1,578 489 25 282 129 959 46 129 415 -213 -205 1,543 456 26 313 149 959 47 144 425 -205 -205 1,627 458 27 313 154 959 49 144 434 -197 -205 1,652 442 28 313 159 959 50 144 444 -189 -205 1,676 427 29 313 164 959 52 144 454 -181 -205 1,700 413 30 282 152 959 54 129 464 -174 -205 1,661 384 Total 6,854 2,508 23,509 835 3,195 8,943 -9,167 -4,932 31,745 Total (discounted) 2,843 948 10,335 322 1,337 3,745 -5,092 -2,145 12,294 12,294 Total share of benefits (discounted) 15% 5% 53% 2% 7% 19% EBCR @ 5%, 30 yrs ratio 2.70 ENPV @ 5%, 30 yrs million US$ 12,294 EIRR @ 30 yrs % 35.4% Payback Year Years 7.2 Years Note: EFA model calculations. See Figure 4. 62 Figure 4: Economic Analysis - Base Case - Estimated Annual Flow of Benefits and Costs 2,500 2,000 1,500 Other Benefits Social Value of Carbon million USD 1,000 Livestock Benefit Non-cropland Benefit due to avoided soil erosion Non-cropland Benefit without soil erosion 500 Crop Benefit due to avoided soil erosion Crop Benefit without avoided soil erosion 0 Variable Costs Program Costs -500 Incremental Net Benefits -1,000 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Year Note: EFA model calculations. See Table 7. Figure 5: Economic Analysis – Excluding IDA - Estimated Annual Flow of Benefits and Costs 2,500 2,000 1,500 Other Benefits Social Value of Carbon million USD 1,000 Livestock Benefit Non-cropland Benefit due to avoided soil erosion Non-cropland Benefit without soil erosion 500 Crop Benefit due to avoided soil erosion Crop Benefit without avoided soil erosion 0 Variable Costs Program Costs -500 Incremental Net Benefits -1,000 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 Year Note: EFA model calculations. See Table 8. 63 Table 8: Economic Analysis - Comparing Base Case with and without IDA Base Case Scenario 4 Unit Economic Economic Difference Analysis Analysis excl. IDA Program budget Year 1-5 million US$ 1,676 1,676 0 EIRR @ 30 yrs % 25% 23% ENPV @ 5%, 30 yrs million US$ 12,294 9,986 2,308 Period years 30 30 30 ENPV @ 5%, 30 yrs million US$/year 410 333 77 Program area million ha 5.0 5.0 0.0 ENPV @ 5%, 30 yrs US$/ha/year 82 67 15 Number of beneficiaries million persons 6.1 6.1 0.0 ENPV @ 5%, 30 yrs US$/person/year 67 55 13 Note: - In this table, Program budget excludes beneficiary contributions and recurring costs but includes price contingencies. The benefit cost analysis includes beneficiary contributions and recurring costs but excludes price contingencies. - Impact on GHG Emissions assumed to be the same when IDA is included/excluded. Further analyses required. - Exchange rate: 1 US$ = ETB 28 38. Based on some high-level assumptions, the estimated incremental benefit from including IDA focus on participatory management, land tenure security, and strengthening institutions is US$2.3 billion over a 30-year period – or an average of US$77 million per year (19% of the base case ENPV). Based on the assumptions listed below, Figure 5 illustrates the annual estimated costs and benefits for this scenario when excluding IDA funds. A visual comparison to Figure 4 highlights the assumed lack of sustainability in later years with required reinvestments and loss of benefits. To quantify expected improvement to the Government’s Program from IDA financing, this alternative scenario is designed with the following high-level assumptions. Estimated Program returns are shown in Table 8. (a) When excluding US$500 million of IDA the Program costs remain the same and target remains at 5 million ha. (b) The current lack of strong institutions, participatory management, and secure tenure rights lead to insufficient maintenance of WMP investments. (c) Lack of maintenance requires reinvestment in each micro watershed (re-treatment) every 7 years after which it takes 2 years for benefits to recover. (d) Lack of sustainability means that to maintain benefits in the future, Program costs do not decrease after year 10 and the GoE focus does not shift from SLM to other issues. 39. The estimated value of soil erosion varies between US$0.09 and 0.34/tonne soil depending on the gross margin value of different land use. Table 9 shows the estimated amount of soil erosion avoided due to Program activities. The value of the avoided erosion is based on gross margins, therefore erosion is valued at US$0.34/tonne soil (ETB 9.41) on cropland and US$0.09/tonne soil (ETB 2.45). This higher than SLMP-II economic analyses where it was assumed that the value of one tonne soil was ETB 0.79-2.1 in 2013 terms which converts to ETB 1.2-3.3 in 2020 terms (World 64 Bank PAD and ICR EFA analyses, 2013 and 2019b). Compared to the current analysis, the assumed gross margins in SLMP-II did not distinguish between high-value crops and livestock versus low- value non-cropland. Table 9: Value of Avoided Soil Erosion – Economic Analysis excluding GHG Emissions Description Unit Cropland Non- cropland Cumulative Soil loss due to erosion million tonnes over 30 Years 765 1,229 Average Value of Avoided Soil Erosion ETB/tonne soil 9.41 2.45 US$/tonne soil 0.34 0.09 Note: - Economic discount rate = 5%. Financial discount rate = 12%. Analysis period is 30 Years. - Exchange rate: 1 US$ = ETB 28 40. Sensitivity analysis can inform the design of a risk management plan. The projection of all assumptions 30 years into the future is very uncertain, so thorough sensitivity analyses are necessary. Identifying key assumptions that affect the estimated Program returns helps focus attention on uncertainty and data gaps. The analysis can also inform the design of a risk management plan.59 41. Switching values. A switching values analysis is reported in Table 10, where each assumption is changed until the Base Case ENPV turns zero (i.e. a break-even analysis). The Program break-even is not very sensitive to any one particular assumption. At the top of the list, an 99% decrease in non- cropland yields or prices can reduce the ENPV to zero. An increase in Program costs of 241% also reduces ENPV to zero. Switching values for over 240 other assumptions returned nonsensical results (e.g. negative yields and prices) and are not shown here. The large and unlikely changes required to turn the ENPV zero in the switching values analysis suggest what the key assumptions are but they do not reveal how sensitive results are at the margin. So an alternative sensitivity analysis is performed. Table 10: Sensitivity Analysis of Economic Efficiency - Switching Values % change Switching from Base Rank Assumptions Unit Base Case Value Case 1 W/P Yield Sensitivity Factor, Non-cropland % of yield 0.00 -0.99 99% 1 W/P Price Sensitivity Factor, Non-cropland % of price 0.00 -0.99 99% 3 Sensitivity Factor, Investment Costs % of investment costs 0.00 2.41 241% 4 W/P Variable Cost Sens. Factor, Livestock % of price 0.00 3.14 314% Note: - WO/P = Without Program (Baseline); W/P = With Program. - Switching value is the assumption value that causes the Base Case ENPV to turn zero (Break-even point). 59Because the Program’s impact on GHG emissions is calculated in the separate EX-ANTE Carbon-balance model, it was not possible to run a full sensitivity analysis inside the EFA model that changes the total GHG emission impact. 65 42. Elasticities. Instead of switching values, Table 11 shows the elasticities of key assumptions. A general 1% increase in non-cropland yields and prices can lead to a 1% increase in ENPV. A general 1% increase in crop or livestock yields increases ENPV by 0.88% or 0.72%, respectively. A 1% increase in the discount rate leads to a 0.88% decrease in estimated ENPV. Program returns could increase by 0.6% if the area converted to forest plantations increases by 1%. A 1% increase in investment costs can decrease ENPV by 0.4%. On the basis of this analysis, as part of a risk management plan, it is particularly important to ensure that farmers and other land/resource users can achieve target yields and negotiate and obtain fair input/output prices going forward. This includes providing technical advice that encourages adoption of improved production practices and establishment of forest plantations within the available Program costs. Table 11: Sensitivity Analysis of Economic Efficiency - Elasticities Rank Base Assumptions Unit Case ENPV elasticity 1 W/P Yield Sensitivity Factor, Non-cropland % of yield 0.00 1.01 1 Discount Rate % per year 0.05 -0.88 3 W/P Yield Sensitivity Factor, Crops % of yield 0.00 0.88 4 W/P Yield Sensitivity Factor, Livestock % of yield 0.00 0.72 5 Non-cropland Revenue (fin), Plantation Forest, W/P ETB/ha 141,086 0.60 6 Area converted from Bareland to Forest Plantation ha 150,000 0.57 7 Sensitivity Factor, Investment Costs % of inv. costs 0 -0.41 8 Yield net of losses, Milk production, local breed, W/P ltr/animal/year 332 0.39 Note: - WO/P = Without Program (Baseline); W/P = With Program. - Elasticity is measured as the %-change in Base Case ENPV with a 1% change in one assumption at a time. Economic Analysis 43. Some risk factors cannot be estimated well in a switching values or elasticity analysis. To analyze the impact on Program returns from selected assumptions, some specific cases are calculated. Table 12 summarizes the impact of key risk factors as discussed below. 44. Case 1: Doubling the economic discount rate from 5% to 10% reduces estimate Program returns by 57%. While the recommended economic discount rate of 5% is used in this analysis, Table 12 shows that ENPV falls by 35% with an 8% discount rate. 45. Case 2: Falling 25% short the 5 million ha target area can reduce estimated Program returns by 28%. In a case when implementation costs increase such that the Program only achieves 75% of the target area (3.75 million ha), the ENPV falls by 28% and the rate of return drops from 35% to 33%. If on the other hand, 125% of the area target is reached (6.25 million ha) Program return could increase by 25% with an EIRR of 36%. 46. Case 3: Increasing the implementation period for WMPs from 5 to 10 years decreases Program returns by 17%. Close monitoring and support for implementation of WMPs helps increase Program returns. This also includes ensuring that beneficiaries are able to contribute community labor and successfully adopt SLM practices. To capitalize on improved practices they also need support to 66 obtain land use certificates, apply for commercial loans, obtain the necessary quality inputs, and implement their investments. 47. Case 4: A 2-year Program delay can reduce returns by 16%. Similar to a longer implementation period, a delay in when beneficiaries are willing and able to adopt new farming practices and implement their investments can lead to reduced Program returns. Table 12 shows that a 2-year delay in benefits can reduce the EIRR from 35% to 24%. While not always avoidable, Program delays can be minimized with close monitoring and by ensuring implementation does not lose momentum. 48. Case 5: Estimated yield loss from soil erosion may be too low in the Base Case compared to some available studies. If the annual yield loss factors are trebled from maximum 2.5% to 7.5% - which is still conservative in accordance with some studies (see Paganos et al. 2018, and Gebreselassie, 2016) – ENPV can increase by 17%. Note that, this analysis does not take into account climate change which may create increased future soil erosion and yield loss. 49. Case 6: Accelerated soil and yield loss due to future climate change and poor landscape management can increase Program returns by 6-9% depending on severity. Analysis by the Water and Land Resource Centre of Addis Ababa University indicates that soil erosion is expected to increase by 7- 10% per year and, in the more extreme scenarios, could increase by as much as 40-70% per year, further underscoring the need to place a high priority on SLM as an adaptation strategy. Table 12 shows that with a 40% increase in soil loss by the end of the analysis period, the value of avoiding this soil loss increases by 6%. Even if the EIRR is not affected much, with a more accelerated soil loss of 70%, returns could increase by 9%. 50. Case 7: Results are sensitive to the estimated impact on GHG emissions because a 10% reduction in value can reduce ENPV by 3%. This also implies that it is important that the assumptions entered in the EX-ACT model reflect the Program accurately. 51. Case 8: The estimated returns could fall by 1-2% if the number of animals per farm dropped by 10- 20%. Further data are needed to determine if households respond by lowering the number of livestock units they own when the yield per animal goes up as noted in a livestock impact study for SLMP-II (see World Bank, 2017a). Table 12: Sensitivity Analysis of Economic Efficiency – Cases Case ENPV – 30 Years BCR – Economic Payback million % 30 IRR Period US$ change Years Base Case 12,294 0% 2.7 35% 7.2 1. Economic Discount Rate changed to 8% 7,998 -35% 2.4 35% 7.2 Economic Discount Rate changed to 10% 5,328 -57% 2.2 35% 7.2 2. 75% of target area achieved 8,884 -28% 2.3 33% 6.9 125% of target area achieved 15,314 25% 3.0 36% 7.2 3. Implement 15% of area each year until 100% 11,453 -7% 2.5 32% 7.7 Implement 10% of area each year until 100% 10,149 -17% 2.3 27% 8.9 4. 1 year Benefits Delay (0=no delay) 11,277 -8% 2.6 28% 8.6 2 years Benefits Delay (0=no delay) 10,293 -16% 2.5 24% 10.1 5. Yield Loss from Erosion Doubled 13,383 9% 2.8 37% 7.0 Yield Loss from Erosion Tripled 14,323 17% 3.0 38% 6.9 6. Accelerated soil loss by year 30 by 40% 13,021 6% 2.8 36% 7.1 Accelerated soil loss by year 30 by 70% 13,397 9% 2.9 36% 7.1 67 7. Shadow Price of Carbon reduced by 10% 11,919 -3% 2.6 35% 7.2 Shadow Price of Carbon increased by 10% 12,668 3% 2.8 36% 7.1 8. W/P # animals per farm = 90% of WO/P 12,187 -1% 2.7 35% 7.2 W/P # animals per farm = 80% of WO/P 12,080 -2% 2.7 35% 7.2 Note: - WO/P = Without Program (Baseline); W/P = With Program. - 5% discount rate - Economic Analysis 52. To ensure sustainability, community labor may require some level of payment such as through revolving funds. A quantitative analysis is not included regarding the implications of relying heavily on voluntary community labor. Historical problems of poor quality structures, top-down planning, lack of community awareness, and failure to follow a watershed approach may make it difficult to achieve participatory planning and sufficient amount of voluntary community labor. Whether some of the labor is paid for is up to each woreda or WSUA. Further analyses should consider the availability of government funds but also the feasibility of setting up revolving funds within WSUAs. Potentially these could receive some government start-up funds and later be replenished through WSUA membership fees as is common in irrigation-focused water user associations. 53. In summary, as part of the risk management plan, the focus needs to be on ensuring that the main decision makers (farmers, other land users, and local institutions) receive the necessary technical advice, land tenure security, and institutional support so that they are able and can afford to make long-term decisions. Keeping costs down to achieve targeted hectares requires fiscal prudence and sound technical advice both for treatment and maintenance of structures in the Program area. To avoid increased costs due to Program delays and prolonged implementation periods, close monitoring and capacity building are necessary so as not lose momentum after the Program is complete. Considerations should be made for the equitable inclusion of all beneficiaries in the WSUAs including vulnerable groups such as landless, youth, elderly, and female-headed households. For example, landless people may suffer from conservation interventions that include restricted access and use of communal lands. Also, restriction on livestock grazing in some areas may lead to increased erosion in other areas unless stocking rates and grazing rotation are incorporated in WMPs. 54. As discussed earlier, the main expected net benefits which could not be quantified due to lack of data include: (a) Improved capacity and skill-level in public institutions (b) Improved capacity and skill-level in community institutions (WSUAs) (c) Equitable benefit distribution with participatory WMPs and tenure rights (d) Potential benefits from conservation areas, biodiversity and tourism (e) Additional benefits and costs of “cut & carry� livestock management system (f) Diversified livelihoods and improved employment opportunities (g) Benefits to other sectors of the economy that will take advantage of increased productivity and resilience in the Program area. (h) Benefits captured in neighboring communities through informal dissemination of improved land and water management practices. (i) Downstream effects of reduced risk of flooding and reduced cost of sediment build-up in irrigation and hydroelectric dams. (j) Benefits from improved nutrition such as due to a more varied food production in the area. 68 55. In light of an ENPV of US$12.3 billion over 30 Years (US$410 million/year), an ERR of 35%, and the additional potential net benefits that could not be quantified yet, the Program investment is expected to yield considerable returns with a strong risk management plan in place. 5 Rationale for Public Provision and Financing 56. There is a strong rationale for public intervention as proposed by the Program because it supplies public goods, internalizes externalities, and corrects market failures – all core functions of government. 57. Program activities encourage public goods such as: protection of ecologically sensitive landscapes; securing beneficiaries’ land rights, and; reducing GHG emissions by increasing sequestration of carbon in soils and biomass. 58. Program activities help internalize some externalities. A participatory watershed approach increases the sense of ownership and helps internalize externalities, such as with soil erosion, when interventions in one part of a watershed are necessary to improve the situation for stakeholders in a different part. By including all watershed stakeholders in participatory management, benefits can be captured equitably for example by incorporating income-generating activities for landless people. Another externality of reduced soil erosion is to reduce costs to operating downstream irrigation and hydroelectric dams. 59. The current land degradation issues warrant more targeted public investments to ensure that private sector entities adopt sustainable management practices going forward. Net benefits captured through this type of intervention accrue over many years adding to the difficulty of attracting private sector investors. Incentives are also strategically different from standard productivity investments because substantial benefits are generated from avoided future yield losses. In this setting, low productivity leads to lack of cash and credit for working capital as well as for investments thereby preventing farmers from adopting new practices with such long-term benefits. The proposed Program helps focus attention and assistance on beneficiaries becoming more resilient to extreme weather events. For long term impact, this includes issuing authenticated SLLCs and supporting sustainable land management that improves both land degradation and agricultural productivity in the long term. 6 World Bank Value Added 60. Successful interventions to prevent or control land degradation require integrated and cross- sectoral approaches to sustainable land management. The World Bank is well-positioned to support the Program based on the wealth of technical, operational and institutional experiences and lessons learnt through the implementation of the first SLMP-I (2008-2013) and the subsequent SLMP-II (2013-2018) and the ongoing Resilient Landscapes and Livelihoods Project (RLLP). 61. The World Bank’s Africa Climate Business Plan, now in its fourth year of implementation, aligns with Program implementation by contributing to “Scaled-up and transformational 69 investments in key sectors and areas�. This places a strong emphasis on support for climate resilient landscape management and comprises one of four emerging strategic priorities.60 62. Implementation of the recently-launched World Bank Action Plan on Climate Change Adaptation and Resilience also align with this Program.61 The Action Plan prioritizes the mainstreaming of climate action at scale through programmatic scale operations. Amongst the six ‘high potential’ areas identified, are support for (i) Integrated landscape management approaches, (ii) “Triple-win� approaches such as climate smart agriculture and afforestation that seek to capture benefits from development, emissions reduction, and enhanced resilience, and (iii) Nature-based solutions (also referred to as Ecosystem-based adaptation approaches) that reduce risks, protect environmental services and generate income-generation co-benefits. 7 Closing Comments about EFA Model Development 63. The Program team should continue to collect more data to improve the current EFA analysis and also for evaluating the Program at mid-term and completion. Particular focus could be on: (a) Validating assumptions behind all changes in WO/P and W/P gross margins for crops, livestock, non-cropland, and co-benefits captured through new income-generating activities. (b) Validating the assumed farm sizes and cropping/livestock patterns of representative farms including whether Program incentives will lead to changes in cropping pattern and stocking rate. (c) Updating the analysis when the Program budget cost tables are finalized and also explore how to assign shares of the costs to different benefit flows. (d) Continuously ensuring that the EFA analysis is aligned with applicable target indicators in the results framework (e) Refining the estimation of impact on carbon balance using the EX-ACT Model. (f) Add model complexity that links more assumptions to the estimated GHG emissions such that the latter can feature more prominently in the sensitivity analyses. 64. The Microsoft Excel model used to perform this EFA is available from the Program team. This also includes an online step-by-step demonstration of the Excel file. This helps document the model as well as give guidance on how to change assumptions, perform analyses, and extract results. The username for the web page is provided below (case-sensitive). The password is provided on the “Read Me� sheet in the Excel file when it is shared by the Program team: (a) Excel model: Ethiopia_RLLP_CALM_EFA_Model_30Apr2019.xlsm (b) Web page: http://www.rygnestad.net/business/Ethiopia_CALM/index.php (c) Username: P4REthiopia 60 World Bank (2018a). Accelerating Climate-Resilient and Low-Carbon Development: The Africa Climate Business Plan. May 2018. 61 World Bank (2019a). The World Bank Group Action Plan on Climate Change Adaptation and Resilience. January 2019. 70 References: Ethiopia Central Statistical Agency (2014). Ethiopia Comprehensive Food Security and Vulnerability Analysis (CFSVA). March. Gebreselassie, S., O.K. Kirui, and A. Mirzabaev (2016). Economics of Land Degradation and Improvement in Ethiopia. Chapter 14. In E. Nkonya et al. (eds.), Economics of Land Degradation and Improvement – A Global Assessment for Sustainable Development, DOI 10.1007/978-3-319- 19168-3_14. Große-Rüschkamp, A. (2015). Productivity and Income Contribution of Family Farm Enterprises: A Gross Margin Study on the Sustainable Land Management Program (SLMP) – Summary Report. Addis Ababa: Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) GmbH. MOANR now MOALR (2016). Sustainable Land Management Project (SLMP-II) Draft Final Report. Socio-Economic and Biophysical Baseline Survey in 90 New and 45 Existing Watersheds. Ministry of Agriculture and Natural Resources now Ministry of Agriculture and Livestock Resources. December 2016. Nune, S., Menale, K., & Mungatana, E. (2013). Forest Resource Accounts for Ethiopia. In R. M. Hassan, & E. D. Mungatana (Eds.), Implementing Environmental Accounts (Vol. 28, pp. 103-142). Springer Science + Business Media Dordrecht 2013. Panagos P, G. Standardi, P. Borrelli, E. Lugato, L. Montanarella, and F. Bosello (2018). Cost of agricultural productivity loss due to soil erosion in the European Union: From direct cost evaluation approaches to the use of macroeconomic models. Land Degrad Dev. 2018;1–14. https://doi.org/10.1002/ldr.2879 Pistorius, T., S. Carodenuto, and G. Wathum. (2017). Implementing Forest Landscape Restoration in Ethiopia. Forests 2017, 8, 61; doi:10.3390/f8030061 Schmidt, E. and F. Tadesse (2017). The Sustainable Land Management Programs in the Ethiopian Highlands. An evaluation of its impact on crop production. ESSP Working Paper 103. UNFCCC (2017). Intended Nationally Determined Contribution (INDC) of the Federal Democratic Republic of Ethiopia. Accessed 28 Mar 2019. Web link: https://www4.unfccc.int/sites/ndcstaging/PublishedDocuments/Ethiopia%20First/INDC- Ethiopia-100615.pdf World Bank (2013). Ethiopia Sustainable Land Management Project (SLMP-II). Project Appraisal Document. PAD525. October 29, 2013. World Bank (2014). Ethiopia Sustainable Land Management Project (SLMP-I). Implementation Completion and Results Report. Report No. ICR3074. March 24, 2014. World Bank (2015a). Technical Note on Discounting Costs and Benefits in Economic Analysis of World Bank Projects. Washington, DC. World Bank (2015b). Ethiopia Poverty Assessment 2014. Report No. AUS6744. January. World Bank (2017a). Assessment of Livestock Impact on NRM Interventions of SLMP-II and Recommendations for the Design of Resilient Landscape Livelihood Project (RLLP). Ethiopian Sustainable Land Management Project II (SLMP-II). Final Report. December. World Bank (2017b) Shadow price of carbon in economic analysis. Guidance note to the World Bank Group staff. Washington, DC, Nov 12 World Bank (2018a). Accelerating Climate-Resilient and Low-Carbon Development: The Africa Climate 71 Business Plan. May 2018. World Bank (2018b). Ethiopia Resilient Landscapes and Livelihoods Project (RLLP). Project Appraisal Document. PAD2484. Draft May 25, 2018. World Bank (2019a). The World Bank Group Action Plan on Climate Change Adaptation and Resilience. January 2019. World Bank (2019b). Ethiopia Sustainable Land Management Project II (SLMP-II). Draft Efficiency Analysis for Implementation Completion and Results Report (ICR). February 12, 2019. 72 Appendix 1: Assumptions Underpinning Economic and Financial Analysis 1. This Appendix contains assumptions behind gross margins for cropland, livestock, and non-cropland production systems in the Program area. Collected data also establish cropping pattern and livestock units for 6 representative farms used in the EFA analysis. 2. Program interventions are assumed to lead to improved crop and livestock gross margins providing there is sufficient long-term maintenance. Appendix Tables 1.1 and 1.2 show the crop and livestock gross margins per hectare and per animal head. One farm model is established for each region based on cropping pattern and gross margin data from the SLMP-II baseline study (MOANR/MOALR, 2016) and gross margin study (Große-Rüschkamp, 2015). Ethiopia Consumer Price Index of 172 is used to adjust 2014 to 2020 prices, and 120.7 to adjust 2018 to 2020 prices. It is assumed that the Program has no impact on crop prices, although this is discussed further in the sensitivity analysis. For livestock production, a price increase is included because Program intervention is expected to lead to use of improved breeds and quality of produce. Yield increases are expected to be small (9- 12%) on irrigated crops, and larger on rain-fed crops (14-58%). To achieve these yield increases, variable costs will also increase. Gross margins on key crops are assumed to increase by between 8% and 61%. Gross margins from livestock production increase, particularly in dairy and sheep rearing (40-42%). Any yield increases are assumed to build up over time with long-term maintenance, which is also emphasized by Schmidt and Tadesse (2017).62 While crop gross margins are expected to increase gradually over 7 years, only 6 years are attributed to the Program while the first year is attributed to prior operations in the area. Similarly, the first year of increased livestock gross margins are attributed to prior operations. Furthermore, it is assumed that there is 10% post-harvest yield loss due to insufficient storage and handling. Finally, as a proxy for variability, revenue generated on cropland is reduced by 10% every 5 years due to an extreme weather event. 3. Appendix Table 1.3 shows the assumed cropping patterns and number of livestock on representative farm models. One representative farm is established for 6 regions based on cropping pattern and livestock data from the SLMP-II baseline study. For Harari and Dredewa the assumptions for Oromia are used. There are no available data on which to base an assumption of changed cropping patterns due to the Program. In recent impact studies of the SLMP-II project there are indications that farmers who are able to increase the yields of different livestock are tending to reduce their herd size (World Bank, 2017a). This is not assumed in the base case but included in the sensitivity analysis.63 4. Some crop gross margins are used as proxies for other crops when data are unavailable. Not all crops featured in the SLMP-II baseline study have gross margin data available. Therefore, some crops have been combined to cover 100% of farm area by region: sorghum is combined with millet. Oilseeds are combined with peas. It is assumed that 100% of potatoes are irrigated in Tigray, Amhara, Oromia, and Benishangul Gumuz. Potatoes are rain-fed in Gambella and SNNPR. Rain-fed vegetables are valued as rain-fed potatoes including 90% of vegetables in Tigray, Amhara, 62Schmidt and Tadesse (2017) suggest that there are no measurable improvements in productivity from SLMP, although the authors also acknowledge that there are problems with the data: They found that, over the analysis period, value of production increased significantly in both treatment and non-treatment areas. 63In reality, cropping patterns are driven by demand and supply. However, the EFA model is deterministic and does not include a dynamic adjustment of cropping patterns between years and different farmers. The assumptions are based on data from SLMP-II baseline survey (MOANR/MOALR, 2016) and the Program team’s best judgement. 73 Benishangul Gumuz, and SNNPR, and 50% of vegetables in Gambella and Oromia. The remaining vegetables are irrigated and valued as tomatoes. 5. Appendix Table 1.4 shows the estimated gross margins on non-cropland. Program interventions will transform 160,000 ha from bareland to forest plantations (150,000 ha), grassland (5,000 ha), and cropland (5,000 ha). It is assumed that gross margins do not change on most non-cropland areas. The exception is that biophysical treatment of grasslands will improve estimated gross margins by 72% due to near-doubling of the yield and increased maintenance costs. Appendix Table 1.1: Crop Gross Margins - Financial Analysis Wheat Barley Maize Teff Description Unit WO/P W/P WO/P W/P WO/P W/P WO/P W/P Yield net of loss kg/ha 2,283 2,855 2,099 2,392 2,693 3,562 1,653 2,174 Increase % of WO/P 25% 14% 32% 32% Revenue US$/ha 756 945 645 735 605 800 694 913 Variable Costs US$/ha 307 353 254 234 211 248 196 219 Gross Margin US$/ha 449 592 392 502 394 552 498 693 Increase % of WO/P 32% 28% 40% 39% Potato Rain-fed Millet Faba Bean Chickpea Description Unit WO/P W/P WO/P W/P WO/P W/P WO/P W/P Yield net of loss kg/ha 2,816 4,457 1,866 2,182 1,009 1,552 1,063 1,397 Increase % of WO/P 58% 17% 54% 31% Revenue US$/ha 439 694 438 512 500 768 465 611 Variable Costs US$/ha 223 347 100 131 66 95 49 67 Gross Margin US$/ha 215 348 338 381 434 673 416 544 Increase % of WO/P 61% 13% 55% 31% Potato, Cabbage, irrig. Tomato, irrigated irrigated Description Unit WO/P W/P WO/P W/P WO/P W/P Yield net of loss kg/ha 18,687 20,961 3,622 3,930 13,229 14,355 Increase % of WO/P 12% 9% 9% Revenue US$/ha 3,431 3,849 555 602 3,656 3,967 Variable Costs US$/ha 1,393 1,031 265 277 402 419 Gross Margin US$/ha 2,038 2,818 289 325 3,254 3,548 Increase % of WO/P 38% 12% 9% Note: - WO/P = Without Program (Baseline); W/P = With Program. Annual average allowing for a 7-year linear increase. Table only shows increase for the last 6 years. This is to attribute some benefits to prior operations in the area. - Assumes a 10% post-harvest yield loss. - Revenue includes value of home consumption (e.g. fodder). Costs exclude farmer's own labor. - Some adjustments made to original source data by RLLP Project team to obtain more conservative gross margin improvements on irrigated potatoes and tomatoes. - Main source: Große-Rüschkamp, A. (2015): Productivity and Income Contribution of Family Farm Enterprises: A Gross Margin Study on the Sustainable Land Management Program (SLMP) - Exchange rate: 1 US$ = ETB 28 74 Appendix Table 1.2: Average Annual Gross Margins - Livestock - Financial Analysis Milk Bull Sheep Egg Production Fattening Rearing Production Description Unit WO/P W/P WO/P W/P WO/P W/P WO/P W/P Years until full change 1 0 2 1 Yield Liter, kg, or egg/animal 290 332 1.0 1.0 2.2 2.3 142 146 Increase % of WO/P 14% 0% 3% 2% Revenue US$/ha 226 258 434 434 88 118 19 19 Variable Costs US$/ha 149 150 343 343 21 25 8 8 Gross Margin US$/ha 77 109 92 92 67 93 10 11 Increase % of WO/P 42% 0% 40% 7% Note: - WO/P = Without Program (Baseline); W/P = With Program. Assumes a linear change of yields and costs from WO/P to W/P situation over the number of years indicated in table, reduced by 1 year to attribute some benefits to prior operations in the area. - Revenue includes value of home consumption. Costs exclude farmer's own labor but includes home-grown fodder. - Main source: Große-Rüschkamp, A. (2015): Productivity and Income Contribution of Family Farm Enterprises: A Gross Margin Study on the Sustainable Land Management Program (SLMP) - Exchange rate: 1 US$ = ETB 28 Appendix Table 1.3: Cropping Pattern WO/P and W/P on Representative Farms and Land Area, by Crop and Livestock Farm A Farm B Farm C Farm D Farm E Farm F % of 1 ha farm Tigray Gambella Amhara Oromia Benishangul SNNPR Gumuz Wheat 18.8 9.6 19.9 13.3 4.3 10.9 Barley 15.3 9.2 14.8 9.7 0.3 11.5 Maize 12.8 42.4 8.6 21.9 26.2 23.8 Teff 23.9 3.6 16.6 15.8 6.7 12.5 Potato, Rain-fed 3.9 10.1 3.7 4.5 6.1 22.9 Millet 13.3 14.6 11.1 13.7 33.3 5.9 Faba Bean 3.4 1.0 7.6 3.5 2.1 4.8 Chickpea 7.3 1.1 9.6 7.5 18.4 6.4 Cabbage, irrigated 0.2 4.3 0.2 2.2 0.3 0.7 Potato, irrigated 0.8 0.0 7.6 5.7 1.7 0.0 Tomato, irrigated 0.2 4.3 0.2 2.2 0.3 0.7 Total 100 100 100 100 100 100 Livestock (heads/farm) Milk, local breed 2.0 7.5 3.9 2.7 3.9 6.9 Bull fattening 2.0 7.5 3.9 2.7 3.9 6.9 Sheep rearing 6.2 3.4 6.2 3.8 4.2 3.0 Egg production 8.2 13.9 4.8 6.2 10.3 5.6 Total TLU 3.5 11.0 6.1 4.3 6.0 10.0 % households with animals 91 50 88 82 78 87 75 Note: - WO/P = Without Program (Baseline); W/P = With Program - Tropical Livestock Units (TLU) conversion factors: Cattle = 0.7, Sheep = 0.1, Chicken = 0.01. - Oromia assumptions used as representative for Harari and Dredewa regions. - Developed from SLMP-II baseline study (MOANR/MOALR, 2016) and gross margin study (Große-Rüschkamp, 2015) Appendix Table 1.4: Gross Margins on Non-cropland - Financial Analysis Description Unit Plantation Natural Woodland Bush+ Grassland, Grassland, Forest Forest Shrub WO/P W/P Other yield m3/ha, 179 5.7 0.8 0.7 2,760 5,000 kg/ha Change % of WO/P 0% 0% 0% 0% 81% Revenue US$/ha 5,039 160 18.5 16.4 704 1,275 Variable Costs US$/ha 1,002 25 4.1 3.7 0 64 Gross Margin US$/ha 4,036 135 14.4 12.6 704 1,212 Change % of WO/P 0% 0% 0% 0% 72% Note: - WO/P = Without Program (Baseline); W/P = With Program. Annual average allowing for a 7-year linear increase for yields, and 7-year for costs. Table only shows increase for the last 6 years. This is to attribute some benefits to prior operations in the area. - Source: RLLP Project team communications with Oromia Forest and Wildlife Enterprise and Nune et al. (2013). - Exchange rate: 1 US$ = ETB 28 76