ZAMBIA CLIMATE -SMART AGRICULTURE INVESTMENT PL AN Analyses to support the climate-smart development of Zambia’s agriculture sector PAGE A ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN PAGE B ZAMBIA CLIMATE -SMART AGRICULTURE INVESTMENT PL AN Analyses to support the climate-smart development of Zambia’s agriculture sector THE W ORLD B A NK  GOV ERNMENT OF Z A MBI A PAGE C © 2019 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover illustration and design by Brad Amburn Foreword In Zambia and other Sub-Saharan African countries, agriculture systems face a growing number of climate-related vulnerabilities. Climate has become increasingly variable over the past few decades, with drought, seasonal and flash flooding, and extreme temperatures occurring more frequently. In response, the World Bank is collaborating with the Government of Zambia to integrate climate change considerations into the country’s agriculture policy agenda through a Climate-Smart Agriculture Investment Plan (CSAIP). The CSAIP aims to produce evidence of climate-smart agriculture (CSA) technologies that offer the greatest potential as Zambia seeks to sustainably increase productivity, enhance household and agroecosystem resilience, and reduce or remove its greenhouse gas emissions. Going forward, it will be critical to have an understanding of how best to address the trade-offs and synergies between achieving agricultural and economic goals on one hand, and preparing for emerging climate challenges on the other. The use of evidenced-based decision making is a key part of this process. Zambia’s CSAIP is the outcome of a partnership between Zambia’s Ministry of Agriculture and the World Bank. The CSAIP represents a commitment by the Bank’s Agriculture Global Practice under the Eighteenth Replenishment of the International Development Association (IDA18) to support development of 10 country-level CSA strategies and investment plans. The CSAIP builds on existing strategy documents, including Zambia’s 7th National Development Plan and its National Agricultural Investment Plan. Through a process that combines several modeling approaches, technology foresighting, and consultations with stakeholders in the public and private sectors, civil society, and farmer groups, we hope to answer these vital questions: • Can CSA deliver on key agriculture sector indicators by 2050? • Are CSA benefits robust across a range of climate change scenarios? • Which CSA technologies should be prioritized for scale up? • Which strategies and investments will be critical to enable broad adoption of CSA technologies? In view of the risks as well as opportunities inherent in a future with changing climate and uncertainty, we consider Zambia’s CSAIP to be an integral tool in bringing about sustainable improvement in the lives and livelihoods of those that most need it—the smallholder farmers—many of whom are living below the poverty line and who make up more than half of Zambia’s total population. Julius J. Shawa Paul Noumba Um Permanent Secretary Country Director (Botswana, eSwatini, Lesotho, Ministry of Agriculture, Zambia Namibia, South Africa, Zambia, Zimbabwe) The World Bank Group PAGE I Acknowledgements The authors wish to thank the Ministry of Agriculture of the Republic of Zambia, notably Permanent Secretary Julius J. Shawa, and Dominic Namanyungu, who led the Climate-Smart Agriculture Technical Team, for their guidance, contributions and support throughout the study. The authors also wish to thank and acknowledge the continuous guidance and support of Paul Noumba Um (Country Director, Africa Region), Ina-Marlene Ruthenberg (Country Manager, Zambia), Erwin de Nys (Program Leader, Africa Region), Juergen Voegele (Senior Director, Food and Agriculture Global Practice), Martien van Nieuwkoop (Director, Food and Agriculture Global Practice), Mark Cackler (Practice Manager, Food and Agriculture Global Practice) and Lou Scura (Practice Manager, Food and Agriculture Global Practice), and Tobias Baedeker (Task Team Leader of Climate-Smart Agriculture Investment Plans—Programmatic Approach) all of the World Bank Group. The report “Zambia Climate-Smart Agriculture Investment Plan” was developed by a core team from the World Bank and Ministry of Agriculture of the Republic of Zambia. Dominic Namanyungu led the Climate-Smart Agriculture Technical Team, which consisted of Joy Sinyangwe, Douglas Mwasi, Morton Mwanza, Justin Chuunka, Reynolds Shula, and Sipawa Songiso. The World Bank Group core team consisted of Christine Heumesser and Alex Mwanakasale (both Task Team Leaders), Willem Janssen, Martin Wallner, Ngao Mubanga, all of the World Bank’s Food and Agriculture Global Practice. Both teams were greatly supported by Mick Mwala throughout the period of the study. The report is based on invaluable contributions of colleagues from collaborating institutions. These include: The International Institute for Applied Systems Analysis (IIASA), and a team consisting of Michiel van Dijk, Amanda Palazzo, Petr Havlík, Tamas Krisztin and Stefan Frank, who conducted analyses presented in section 2.5, 5, and related appendices. A team from the Food and Agriculture Organization of the United Nations (FAO), consisting of Nick Sitko, Giacomo Branca and Antonio Scognamillo, who conducted analyses presented in section 2.3, 2.4, 3, 4, 8, and related appendices, and were supported by Brent Simpson. A team from California Environmental Associates (CEA), consisting of Antonius Gagern, Michael Berger, and Helia Bidad, who led a stakeholder workshop in October 2017 and helped synthesize the technical background papers into the final report. The report was copyedited by Dina Towbin and Olufemi Terry of Dina Towbin and Associates LLC. The report’s design and layout was developed by Brad Amburn. The report benefitted greatly from peer reviewers’ comments: Ademola Braimoh and David Treguer from World Bank Group, and Hambulo Ngoma from Indaba Agricultural Policy Research Institute (IAPRI), at different stages of report development. The report benefitted from invaluable guidance and suggestions of many stakeholders who participated in two consultation workshops in October 2017 and April 2018, and separate consultation meeting throughout the process. The list of stakeholders who took part in the consultation and provided valuable feedback is provided in Appendix H. The team also wishes to thank all stakeholders who shared their views in qualitative interviews conducted during the report, specifically from the Development Partners in Zambia. PAGE III Abbreviations 7NDP Seventh National Development Plan AEZ agroecological zone AF agroforestry AFOLU Agriculture, Forestry, and Other Land Use BAU business-as-usual CA conservation agriculture CAADP Comprehensive Africa Agriculture Development Programme CBA cost-benefit analysis CE carbon equivalents CH4 methane CO2 carbon dioxide CIAT International Center for Tropical Agriculture CIMMYT International Maize and Wheat Improvement Center COMACO Community Markets for Conservation CSA climate-smart agriculture CSO Central Statistical Office CSAIP Climate-Smart Agriculture Investment Plan CPF Country Partnership Framework DiD difference in difference DIV crop diversification DTM drought-tolerant maize dM dry matter EPIC Environmental Policy Integrated Climate Model EX-ACT Ex-Ante Carbon balance Tool EIRR economic internal rate of return FAO Food and Agriculture Organization FAOSTAT Food and Agriculture Organization Corporate Statistical Database FFS farmer field school FISP Farmer Input Support Program FRA Food Reserve Agency G4M Global Forest Model GCM General Circulation Model GgCO2e Gigagram of carbon dioxide equivalent GHG greenhouse gas GLOBIOM Global Biosphere Management Model GoZ Government of the Republic of Zambia GDP gross domestic product Ha hectare IAPRI Indaba Agricultural Policy Research Institute ICTs information and communications technologies IIASA International Institute for Applied Systems Analysis IMPACT International Model for Policy Analysis of Agricultural Commodities and Trade INDC Intended Nationally Determined Contribution LCMS (Zambia) Living Conditions Monitoring Survey LPJmL Lund-Potsdam-Jena managed Land model LULUCF land use, land-use change, and forestry MFD Maximizing Finance for Development MSD minimum soil disturbance MT metric ton MtCO2e mega tons of carbon dioxide-equivalent MoA Ministry of Agriculture NAIP National Agriculture Investment Plan N2O nitrous oxide NGO Non-Governmental Organizations NDC Nationally Determined Contribution NAESS National Agricultural Extension and Advisory Services Strategy NPV net present value OLS Ordinary Least Squares PHL post-harvest loss RALS Rural Agricultural Livelihood Survey RCP Representative Concentration Pathway RR residue retention SPAM Spatial Production Allocation Model SSP Shared Socio-economic Pathways tCO2e metric tons of CO2-equivalent WFP World Food Programme WII weather index insurance ZARI Zambia Agriculture Research Institute ZMK Zambian Kwacha PAGE V Table of Contents Executive Summary 1 Section 1: Introduction 15 1.1 Agriculture Can Be Zambia’s Engine of Structural Transformation and Growth 15 1.2 Agriculture Has a High Production Potential in Zambia 17 1.3 Zambia’s Agriculture Sector is Vulnerable to Climate Change 19 1.4 Land Use Change Is the Largest Contributor to Zambia’s Carbon Footprint 19 1.5 Climate-Smart Agriculture Aims to Address Climate Challenges but Adoption Rates Remain Low 20 1.6 Climate Change Is Typically Factored into Agriculture Sector Policies and Strategies 21 1.7 Objectives of the Zambia Climate-Smart Agriculture Investment Plan 22 1.8 Report Structure 23 Section 2: Methodology 27 2.1 The CSAIP Approach: A Flexible Framework 27 2.2 Stakeholder Consultations to Determine a Vision, Goals, and Uncertainties for the Agriculture Sector 28 2.3 Econometric Analysis to Determine CSA Impacts on Welfare of Zambian Households 30 2.4 Cost-Benefit Analysis to Determine Financial and Economic Profitability of CSA 31 2.5 GLOBIOM Approach Used to Assess Agriculture Sector Goals 33 Section 3: State of CSA Adoption and Performance in Zambia 35 3.1 Minimum Soil Disturbance 35 3.2 Residue Retention 37 3.3 Legume Rotation and Intercropping 37 3.4 Commercial Horticulture 38 3.5 Agroforestry 39 3.6 Livestock Diversification 39 3.7 Timing of Planting 40 3.8 Drought-Tolerant Maize 41 3.9 Crop Diversification 41 3.10 Agricultural Liming 42 3.11 Post-Harvest Loss 43 Section 4: Analysis Results: Observed Impacts of CSA on Household Welfare 45 4.1 Findings: Analysis of Minimum Soil Disturbance 46 4.2 Findings: Analysis of Residue Retention 47 4.3 Findings: Analysis of Legume Rotation and Intercropping 48 4.4 Findings: Analysis of Small-scale Commercial Horticulture 48 4.5 Findings: Analysis of Agroforestry 49 4.6 Findings: Analysis of Livestock Diversification 49 4.7 Findings: Analysis of Timing of Planting 50 4.8 Findings: Analysis of Adoption of Drought- and Heat-tolerant Seeds 51 4.9 Findings: Analysis of Crop Diversification 51 (continued) PAGE VII ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Section 5: Analysis Results: Impacts of CSA on Agriculture Sector Performance Under Climate Change 53 5.1 Crop Yield 54 5.2 Crop Production 56 5.3 Livestock Production and Head of Livestock 59 5.4 Land Use Change 59 5.5 Food availability 62 5.6 Agricultural Trade 62 5.7 GHG Emissions 64 5.8 Policy Scenario: Impact of Carbon Tax to Curb Land Use Change 66 5.9 Economic Implications of Scaling-up CSA Adoption 70 Section 6: Prioritizing CSA Practices 73 Section 7: Enabling Environment and Mechanisms to Promote CSA 75 7.1. Incentivizing CSA Adoption by Improving the Institutional Enabling Environment in Zambia 75 7.2 Summary of Innovative Delivery Mechanisms to Support CSA Adoption 79 7.3 Investment Requirements to Scale-up CSA 84 Section 8: Conclusion 87 References 93 Appendices 99 Appendix A: Policy Frameworks 99 Appendix B: Results of First Stakeholder Workshop 102 Appendix C: Household-level Econometric Assessment 106 Appendix D: Cost-benefit Analyses 112 Appendix E: GLOBIOM 115 Appendix F: Public Expenditures on CSA Projects in Zambia 127 Appendix G: Detailed Description of Delivery Mechanisms 129 Appendix H: Stakeholders Consulted During the Preparation of this Report 139 Endnotes 142 PAGE VIII ES Executive Summary Key messages • Most climate-smart agriculture (CSA) practices have positive welfare effects on Zambian households in the long-term; in the short-term benefits seem uncertain and may be hindered by high upfront and labor cost. Tailoring CSA to site-specific, agroecological conditions is critical. • Projections until 2050 show that Zambia’s agriculture sector could achieve or surpass sectoral development goals such as increasing crop and livestock production, food availability, and net trade. CSA practices can further increase these positive effects, reduce Zambia’s trade deficit in certain commodities, and, in addition, contribute to climate mitigation as public good. • Climate change projections until 2050 show that yield of key crops could decrease by -25 percent, depending on agroecological zone. While CSA can increase crop yields up to 23 percent, these productivity increases are insufficient to avoid further expansion of agricultural land into forest land. Trends in deforestation put Zambia at risk of failing to achieve its climate commitments. A carbon tax on emissions from land use change could be an effective measures to halt deforestation for agriculture. • Among the range of CSA practices, crop diversification into legumes, commercial horticulture, agroforestry, and strategies of reducing post-harvest losses seem most promising in achieving welfare and sectoral development goals. However, adoption of CSA seems constrained by inadequate access to finance, input and output markets, and capacity building. • Business partnerships with rural communities, farmer field schools, and participatory integrated landscape management approaches seem promising and profitable mechanisms to support the development of a productive, resilient, and low-emission agriculture sector. The estimated annual investment requirements to roll out CSA to 50 percent of farmers is less than current funding for the agriculture sector, indicating that implementation quality and effectiveness are critical. • The positive private benefits as well as public good benefits of CSA provide a strong rational to leverage public support through: support agricultural research, testing and dissemination of CSA across agroecological zones; improve access to inputs and agricultural finance; introduce policy measures for increased agribusiness participation; promote cross-ministerial collaboration to promote landscape approaches; continue policy reforms to support agricultural diversification and adoption of short-duration varieties; support the development of market infrastructure such as rural storage facilities. Zambia’s agricultural sector represents the backbone of its rural economy and holds great potential for the entire country. About 60 percent of the population depends on agriculture for PAGE 1 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN a livelihood; of this subset of the population, 80 percent are classified as poor. While employment in the sector remains high, agriculture’s contribution to GDP declined from 17.3 percent in 2004 to 8.2 percent in 2017, and its labor productivity from US$702 in 2004 to US$584 in 2015. Agriculture’s relatively low contribution to GDP and its large share of the total Zambian labor force suggests that the sector is characterized by low-productivity and unskilled subsistence agriculture lacking access to productive assets and market opportunities. There is also very little crop diversity—maize production accounts for 70 percent of the country’s cropland—a narrow focus on a few crops makes farmers vulnerable to both environmental and market shocks (World Bank 2018e, 2018a). Zambia’s agriculture sector faces challenges and is likely to grow more vulnerable as a result of climate change and risk. The country has a highly variable climate, and in the past few decades has experienced climatic extremes in the form of droughts, seasonal and flash floods, and extreme temperatures. Many of these events occurred with increased frequency, intensity, and magnitude. It has been estimated that droughts caused US$438 million in agricultural sector losses; excessive rainfalls and floods had an US$172 million impact between 1982 and 2016. Over the next 10 to 20 years, climate change-related losses in agriculture are expected to amount to US$2.2–3.1 billion (Braimoh et al. 2018, World Bank 2018a). At the same time, land use, land-use change and forestry (LULUCF), and agriculture sector account for approximately 93 percent of the country’s carbon footprint. Zambia contributes only three quarters of a percentage point to global emissions. In the national emissions profile, the agriculture and LULUCF sector represent 7 percent and 86 percent of emissions, respectively. Most emissions arising from LULUCF sector are a result of biomass burning, including charcoal production. Agriculture’s direct contribution to greenhouse gas (GHG) emissions may seem small but doubles to 13 percent if its indirect effects—through the expansion of cropland and grassland—are considered (CIAT and World Bank 2017). The Government of the Republic of Zambia (GoZ) is integrating climate change concerns into its agriculture policy agenda. Zambia has developed several climate change-related policies and strategies, and the mainstreaming of climate change into sectoral policies is expected to continue. The National Climate Change Response Strategy (2010) emphasizes the role of sustainable land use systems in enhancing food security. Zambia’s Nationally Determined Contribution (NDC) sets ambitious goals for climate mitigation and adaptation that include the agriculture sector, and which aim to reduce GHG emissions by 25 percent up to as much as 47 percent, depending on the level of international support and financing. Under its Zambia Climate-Smart Agriculture (CSA) Strategy Framework, the GoZ is promoting the rollout of climate-smart agriculture practices that will sustainably increase productivity, enhance resilience, and reduce or remove GHG emissions. The CSA concept reflects an ambition to improve the integration of agricultural development and climate responsiveness. However, CSA implementation will require planning to minimize trade-offs and enhance synergies (co-benefits and “triple-wins”) among the three CSA pillars: productivity, adaptation and resilience building, and mitigation. The Zambia Climate-Smart Agriculture Strategy Framework identifies opportunities and constraints for CSA implementation. However, to achieve the goal of broad CSA adoption, it suggests that more evidence will be needed of CSA’s potential benefits at the local and national levels, specifically in response to climate change, as well as improved communication about CSA benefits. The Climate-Smart Agriculture Investment Plan (CSAIP) aims to identify and fill knowledge gaps about CSA’s local- and national-level benefits, specifically under climate change, inform policy PAGE 2 development, and prioritize investment opportunities. The World Bank collaborated with the GoZ to develop a CSAIP intended to support the operationalization of the country’s climate commitments toward development of a productive, resilient, and low-emission agriculture sector. The CSAIP draws on the goals of established policy and strategy documents and analyzes whether and how CSA can help Zambia achieve its agriculture sector goals and climate mitigation goals. Thus, the CSAIP provides quantitative assessments of CSA’s potential benefits in the event of climate change, and provides evidence how CSA can benefit rural households. The analysis concludes with the prioritization of CSA practices and suggestions for future investments in CSA. The CSAIP is expected to inform forthcoming strategies and programs, such as the Second National Agriculture Investment Plan (NAIP), the implementation of the Seventh National Development Plan (7NDP), the development of the NDC implementation plan, as well as existing and future donor projects. The CSAIP development began with a participatory process that identified the agriculture sector’s policy goals. This report takes the next step by assessing the impacts of a suite of CSA practices on achieving the sector goals and on household welfare. The report concludes with recommendations and proposals for future CSA investments. Drawing on existing policy frameworks, and informed by two in-country stakeholder workshops, a vision for the agriculture sector was established that focuses on increased productivity, climate resilience, and mitigation co-benefits. Specific, measurable targets were identified for the year 2050 that align with the vision and which can be used to evaluate alternative development pathways (see Table ES.1). The CSAIP assessment was guided by following questions: (i) What is the impact of CSA adoption on agriculture sector indicators under climate change? Can CSA practices enhance the feasibility of achieving the agriculture sector vision compared to conventional practices? (ii) What is the impact of CSA adoption on household welfare? (iii) Which CSA strategies should be prioritized for scale up according to quantitative impact indictors (iv) Which delivery mechanisms and investment opportunities can be considered to support a broad rollout of CSA in Zambia and what are the related costs? Table ES.1 Quantified targets that align to the agriculture sector vision and alignment to CSA pillars Source/Policy Target CSA pillars Normative vision as developed in stakeholder workshops By 2050, double yields and profits by means of diversification (beyond Productivity Stakeholder workshop maize), while ensuring household food and nutrition security By 2050, have an agriculture sector that is: (i) diversified in crop Resilience production; (ii) diversified in age and gender of its workforce; and (iii) Stakeholder workshop able to cope with economic and climatic shocks through enhanced capacity and policy By 2050, increase agricultural productivity while maintaining a low Mitigation Stakeholder workshop ecological footprint Quantifiable targets to achieve the long-term vision Increase the share of agricultural exports as a percentage of non- Productivity National Agriculture traditional exports from 41% in 2011 to 55% by 2018 Investment Plan (2014–2018) Increase production of cereals from 3.2 million metric tons (MT) to 6.0 Productivity million MT by 2018 National Long Term Increase land under cultivation by 900,000 hectares (ha) by 2030 Productivity Vision 2030 Increase livestock population to 6,000,000 by 2030 Productivity, Resilience National Policy on Sustainably intensify land use without converting additional land area Resilience, Mitigation Environment (2007) into agricultural land Reduce GHG emissions by 25% with limited international support Mitigation Nationally Determined or 47% conditional on the receipt of US$35 billion of international Contribution (2015) assistance PAGE 3 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN The CSAIP’s analytical approach is innovative and combines quantitative models and qualitative assessments to expand the current body of research. The quantitative modeling, this is, the partial equilibrium agriculture sector model Global Biosphere Management Model (GLOBIOM) which was adjusted to the Zambian context, an econometric approach, and a cost-benefit analysis, was allied to stakeholder engagement and a literature review . This allows to assess the impact of a suite of CSA practices on household-level welfare, achievement of sector goals, as well as the economic returns of scaling up CSA through varied investment models. GLOBIOM allows to assess CSA’s performance under alternative future scenarios, including a business-as-usual (BAU) scenario without climate change and alternative climate change scenarios, through 2050. Table ES.2 provides an overview of which CSA practices were considered most important by stakeholders and which quantitative methods are employed for the assessment of their impacts. The qualitative analyses were useful in assessing the state of Zambia’s enabling environment, innovative delivery mechanisms to enhance CSA adoption, and funding requirements. Table ES.2 Evaluation of CSA practices using key modeling approaches Household level Sector level Cost-Benefit Analysis: Econometric Financial Economic GLOBIOM analysis CSA Practice analysis analysis Conservation agriculture’s 3 pillars: Minimum soil disturbance Residue retention Legume rotation / intercropping Agroforestry Commercial horticulture Crop diversification Livestock diversification Drought-tolerant seeds Timing of planting Agricultural liming Reducing post-harvest loss The analysis shows that CSA’s impact on household income and welfare indicators is mostly positive in the long term. In the short term, the incentive to adopt CSA is constrained by high upfront and production costs. Positive impacts on household income, increase in food availability, reduction of poverty were found in the long term. Higher production costs in the short term (that is, the need for mechanization or increased labor time) could impede adoption specifically for minimum soil disturbance, residue retention, and agroforestry. Some CSA practices do not increase income but do reduce income variability. Two exceptions are minimum soil disturbance and delayed timing of planting, both of which appear ineffective in reducing income variability. The promotion of CSA must be customized to suit Zambia’s specific agroecological conditions. The analysis shows that several CSA practices perform better under dry than wet conditions, which points toward their potentially favorable impact in the event of a drier future climate. Minimum soil disturbance, residue retention, and small-scale horticulture practices did not show good results under wetter conditions. Drought-tolerant seeds, agroforestry, and crop diversification show good results under both extreme dry and wet conditions, and seem particularly suitable for climate adaptation and building household resilience. Under dry conditions, the adoption of nearly all CSA practices has a significant positive impact on food security and crop production. PAGE 4 Figure ES.1 Trends in crop production under conventional practices until 2050, for projections without and with climate change (error bars) 9,000 8,000 Production (1,000 tons dry matter) 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0.00 1961 1970 1980 1990 2000 2010 2020 2030 2040 2050 Barley Cotton Maize Rice Sugarcane Cassava Dry beans Millet Sorghum Sweet potatoes Chick peas Groundnuts Potatoes Soybeans Wheat Total production (FAOSTAT) Sources: Historical values for total production are from FAOSTAT and are indicated until the year 2011; the GLOBIOM model provides projections from 2000-2050 for scenarios without and with climate change. Note: Error bars indicate the range in total crop production as a consequence of climate change. The adoption of CSA enhances the likelihood of achieving, and even surpassing, the agriculture sector vision for crop production and food availability by 2050. Projections are driven by increased population, GDP, and food demand, which leads to a doubling of crop production and increase in food availability, even with conventional agricultural practices. Under climate change projections, total crop production could vary from an -11 percent decrease to an increase of up to 20 percent under conventional practices (see Figure ES.1). Under climate change projections, total crop production could vary from an -11 percent decrease to an increase of up to 20 percent under conventional practices (see error bars in Figure ES.1). CSA adoption is projected to increase total production by, on average, 6 percent compare to conventional practices, and further enhance food availability by 5–10 percent. The effect on crop production varies by crop and CSA practice. For instance, promoting crop diversification could increase the production of soybeans and groundnuts by up to 200 percent. These positive effects appear to be more pronounced under climate change projections. With respect to trade, the target of doubling net exports by 2050 is partially feasible. CSA, such as crop diversification, reducing post-harvest loss, or minimum soil disturbance, can potentially enhance international competitiveness and transform Zambia into a net exporter of additional commodities. The aim to enhance net exports by 2050 is reached for maize and millet, and—under extreme climate change scenarios—for cassava and cotton, even with conventional practices. When adopting CSA practices (see Figure ES.2), such as crop diversification, Zambia has the potential to become a net exporter of groundnuts, increase net exports of maize, and decrease net imports of soybeans by 2050, compared to a situation with conventional agricultural practices. PAGE 5 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure ES.2 Effect of conventional practices and CSA on net trade in 2050, in 1,000 tons Cassava Cotton Groundnuts 600 250 100 400 0 50 200 −250 0 0 −500 −200 1,000 tons no af ca rr msd dtm phl div no af ca rr msd dtm phl div no af ca rr msd dtm phl div CSA CSA CSA Maize Millet Soybeans 200 100 400 150 0 100 200 −100 50 −200 0 0 −300 −50 no af ca rr msd dtm phl div no af ca rr msd dtm phl div no af ca rr msd dtm phl div CSA CSA CSA Source: GLOBIOM model results. Note: Error bars indicate the range of net trade as a consequence of climate change; no CSA refers to conventional practices; af to agroforestry; ca to conservation agriculture; rr to residue retention; msd to minimum soil disturbance; dtm to drought-tolerant maize; phl to post-harvest loss; div refers to crop diversification. In contrast, the doubling of crop yields will not be achieved through conventional practices until 2050. Although CSA has the potential to narrow the distance to the target of doubling crop yields, additional strategies to enhance agricultural productivity will be necessary. Under a scenario of no climate change, yields for most crops are expected to increase, but not sufficiently under conventional practices to double productivity per hectare. CSA practices with the greatest potential to reach this target are reducing post-harvest loss, conservation agriculture, and agroforestry, which are expected to increase production per hectare by 12–23 percent compared to conventional practices. Climate change is projected to have a negative impact on yields of most crops. With the exception of cassava, yields of most crops may increase less than without climate change (see Figure ES.3). In the most extreme climate change scenario, the projected maize yield is 3.3 MT/ha, 15 percent lower than without climate change. However, certain CSA practices, for instance, adoption of drought-tolerant maize, conservation agriculture, and crop diversification, are found to enhance an already positive yield effect. Since projected crop yield increases are insufficient to reach the sector’s long-term targets, there is a risk that increased production will have to come from an expansion of Zambia’s cropland. CSA strategies are expected to curb the trend of land conversion, but only marginally. To meet domestic and foreign food demand, almost 2.2 million ha of mainly forested area is projected to be converted to agricultural land between 2010 and 2050, thereof converting 917,000 ha into cropland (see Figure ES.4). This scenario is just short of the target to convert no more than 0.9 million ha into cropland, as stated in Zambia’s Vision 2030 (2006), but would not comply with the goal of the National Policy PAGE 6 on Environment (2007) of completely avoiding land conversions. Individual CSA practices have the potential to reduce conversion to cropland by around 32,000 ha, in the case of conservation agriculture, and 113,000 ha in the case of reducing post-harvest losses. However, averting the conversion of forest land into cropland will not necessarily halt deforestation, as a similar land area is expected to be converted to grassland to meet increases in livestock production. Figure ES.3 Trends in crop yields under conventional practices until 2050, for projections without and with climate change (error bars) Cassava Groundnuts 16 1.6 1.4 14 1.2 12 1.0 10 0.8 8 0.6 6 0.4 tons/ha 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Maize Millet 5.0 4.5 1.8 4.0 1.6 3.5 1.4 3.0 1.2 2.5 2.0 1.0 1.5 0.8 1.0 0.6 0.5 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 BAU Historical Source: Historical values are from FAOSTAT; GLOBIOM provides projections from 2000-2050 for scenarios without and with climate change; “Vision” was determined through stakeholder consultation and represents the targets for agriculture sector development in 2050. Note: Error bars indicate the range in total crop production as a consequence of climate change. The land use change trends are reflected in Zambia’s carbon footprint and put the country at risk of failure to meet its NDC GHG emissions reduction goals of 25 percent to 47 percent. Further agricultural land expansion can be expected to continue in the future. CSA adoption can reduce emissions from land use change, but only by -0.32 percent compared to conventional practices. Minimum soil disturbance and post-harvest loss can reduce fertilizer use, while residue retention rather increases its use. The average impact on emission reduction from synthetic fertilizer is negligible compared to conventional practices. On the positive side, CSA practices are expected to enhance soil carbon sequestration and biomass growth (for instance agroforestry), which is expected to be many times more effective in reducing tons of carbon dioxide equivalent (tCO2e) emissions than reducing consumption of nitrate fertilizer. Introducing a low carbon tax on land conversion could effectively halt deforestation for agriculture even at very low carbon prices. But reducing biomass burning seems most promising strategy. A carbon tax of US$10/tCO2e emissions could reduce emissions from deforestation and other land use change by 99 percent in 2050 compared to a scenario without policies. At a very low price of US$5/tCO2e, land conversion still appears to be profitable. As a consequence, production PAGE 7 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure ES.4 Trends in land use under conventional practices, 1960-2050, for projections without climate change 45,000 40,000 35,000 30,000 Area (1,000 ha) 25,000 20,000 15,000 10,000 5,000 0 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Cropland Forest Grassland Natural land GLOBIOM Historical Sources: Historical values are from FAOSTAT, where available; GLOBIOM provides projections from 2000-2050 and food availability would slight decrease by approximately 2 percent, crop yield slightly increases, and imports increase compared to a scenario without policies. While these policies seem promising, enforcement at this point in time is unlikely. Emissions from agriculture and land conversion related to agriculture constitute only 13 percent of total emissions in Zambia today. The largest share stems from biomass burning which is not modelled in our analysis. To get serious about reducing emissions, reducing biomass burning and charcoal use must become a policy priority. While CSA’s impact on land use change may be small in terms of hectarage and tCO2e emissions, an economic analysis shows that the economic value of reducing GHG emissions to society is significant. If mitigation co-benefits are valued at a shadow price of carbon, that is, the carbon price which is consistent with achieving the core objective of the Paris Agreement on climate change (2015) of keeping temperature rise below 2 degrees, the net present value of sector-wide CSA intervention is increased by between 171 percent and 341 percent, or from US$170 million—if private benefits are considered—to between US$460 million and US$750 million. Even with a carbon market price of US$11/tCO2e emission, the net present value of the investment could rise by 44 percent to US$245 million. This considerable value of a global public good justifies exploring carbon finance to support CSA uptake and to compensate farmers for initial losses from CSA adoption. Four CSA strategies—crop diversification, commercial horticulture, agroforestry and reducing post-harvest loss—show the greatest promise, perform well while reducing trade-offs between indicators, and should receive funding and institutional support. Despite potentially promising effects, CSA adoption rates remain low. In addition to factors associated with CSA practices, such as deferred benefits or high initial costs of adoption, the enabling environment in Zambia—specifically inadequate access to rural finance, capacity building and markets, but also insecure land tenure PAGE 8 and distortionary public policies which tend to focus on maize production—present significant bottlenecks (see Table ES.3 for an overview of the enabling environment’s impacts on the adoption of CSA practices). Considering household and sectoral indicators, as well as conditions in Zambia’s enabling environment that support the adoption of CSA, the practices were ranked—showing that crop diversification, commercial horticulture, agroforestry, and reducing post-harvest loss seem most promising (see Table ES.4). Public resources should be allocated to crowd-in private sector finance and maximize finance for achieving agriculture sectoral goals and realize agriculture’s potential to contribute to Zambia’s structural transformation. The analysis explores eight proven and innovative mechanisms and investment opportunities that can enhance CSA adoption. The mechanisms are ranked by their potential to address key constraints to CSA adoption, that is, inadequate access to rural finance, capacity building and markets: (i) Business partnerships with rural communities, in which agribusinesses promote the adoption of sustainable land management practices, provide environmental certification and leverage carbon finance; thereby providing smallholders with capacity building, access to markets, and finance. (ii) Outgrower schemes with small-scale irrigation to promote commercial horticulture production. and thereby provide farmers with improved access to markets and finance, training and services. (iii) Participatory, integrated landscape management approaches to address multiple objectives of crop and livestock production, forest management, and environmental sustainability. Participatory elements favor knowledge exchange among rural communities and integrated land and water management practices enhance eligibility to participate in carbon finance projects. (iv) Farmer field schools to enhance community-based learning and timely knowledge exchange, as well as facilitated market access through strengthening of farmer groups and associations. (v) Pluralistic participatory extension approaches to enhance adoption of agricultural research and innovation, and spur private sector (agri-preneurs and agro-dealers) involvement in service delivery, to improve farmers’ business skills and facilitate market linkages. (vi) Weather index insurance schemes, which are combined with a mandatory adoption of CSA practices to enhance resilience to climate change and shocks. This scheme is expected to provide training and can be combined with savings schemes which facilitate access to finance. (vii) Cash transfer programs can be aligned with agriculture sector programs and planting cycles, to provide farmers with access to capital to start-up climate-smart agriculture operation and enhance livelihood resilience. (viii) Principles of gender-sensitive supply chains are applicable to each mechanism; gender- sensitive interventions may help overcome enabling environment constraints to support women to adopt CSA. Adopting these delivery mechanisms (i)-(vi) and implementing them for approximately 800,000 Zambian farmers, that is, more than 50 percent of the country’s smallholders, would require an annual public sector investment of US$32 million over five years. Delivery mechanisms have an indicative cost of US$15 million (weather index insurance) and up to US$59 million for participatory pluralistic extension approach which combines agricultural research and dissemination. If rolled out to 118,000 beneficiaries each, which reflects the current adoption rate of 8 percent of minimum soil disturbance in Zambia, expected rates of return on investment range from 11 percent (weather index insurance) to 39 percent (farmer field school). Only the participatory pluralistic extension approach PAGE 9 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table ES.3 Barriers to adoption of CSA practices in Zambia Land tenure Capacity Access to Access to rural Access to Distortionary systems building finance infrastructure markets public policies Conservation agriculture (minimum soil Lack of secure Deferred Upfront costs Opportunities Fertilizer disturbance, tenure is a benefits pose a (e.g., inputs, to generate subsidies residue disincentive to risk to farmers labor, income reduce may decrease retention, adopt practices but this can mechanization) risk of adopting economic rotation) with deferred be mitigated are found new practices rational for benefits by adequate to impede adoption; Agroforestry capacity adoption subsidy building schemes could be paired with advisory services to enhance adoption. Commercial horticulture Horticulture Diversifying Access to Irrigation, roads, and access Can lead has short into new crops finance critical to markets crucial for stable to lower cultivation requires access as new inputs, production and links to agri- competition cycle, and can to advisory infrastructure businesses among input be cultivated services and transport suppliers and on small are needed traders as well plots; insecure to set up as higher input land titles operations costs could hinder investment in irrigation systems. Livestock diversification Mobile asset; Diversifying To start Access to new Current focus grazing often into new operation, species, and on maize occurs on species access to health services impedes the communal land requires access finance critical; is a challenge production to advisory livestock serves of crops to services incl. as “savings develop animal health bank” livestock feed industry Timing of No specific planting impact Lack of advisory Lack of inputs Access to Lack of input Late delivery services affects negatively infrastructure markets often of subsidies planting affects timing facilitates causes planting was found to decisions of planting delivery of delays affect timing of inputs and planting reduces involuntary planting delays Drought- and heat-tolerant seeds Benefits accrue No specific Improved No specific FISP has shortly after capacities seed usually difference from facilitated planting compared to costly, access conventional access to conventional needed to start practices, improved practices operation except supply maize seeds of seeds. Crop diversification Research shows Access to Access to Access to markets and irrigation Disincentivizes small land sizes advisory finance critical supports diversification diversification; disincentivize services a new inputs but ongoing diversification crucial driver of needed to set reforms of FISP diversification up operation could reduce barriers Reducing post-harvest loss Access to Access to Access to rural Access to Disincentivizes advisory finance critical infrastructure markets private sector services can to get access includes provides involvement help reduce to adequate storage incentives to in storage and post-harvest technologies facilities which reduce post- warehouse losses on farm are critical harvest loss facilities Source: World Bank, own elaboration Low barriers to adoption Medium barriers to adoption High barriers to adoption PAGE 10 Table ES.4 Impacts of CSA practices on key decision variables Rural Past household-level impact Projected sector-level impact Develop- Feasibility ment Income Reduce Food Crop yield GHG Food Job creation Available Enabling income security mitigation availability & market funds environment variability linkages Minimum soil disturbance Wet conditions Crop yields Climate Climate Climate Requires (–) increase, change (–) change (++) change (++) establishment Long run strong spatial Strong increase Enhances Contingent of input positive; short variation under BAU; soil carbon on adoption markets run negative; decrease under sequestration taking place (implements, strong climate change in favorable inputs, mech- geographic areas anization) variation Residue retention Wet conditions Dry conditions Climate Climate Climate No notable (–) (+) change (–) change (–) change (–) impact Slight increase Increased under BAU; fertilizer negative under emissions; but climate change soil carbon sequestration occurs Crop rotation/ n/a n/a n/a intercropping Wet conditions Strong Dry conditions Investment (–) geographic (+) in legume Long run variation Strong value chain positive; short geographic development run neutral variation and Crop n/a processing; diversification access to new markets (largely Dry conditions Dry conditions Climate Climate (traders, Ongoing legumes) (+) (+) change (–) change processors) reforms Increases for Increased creating job expected to Wet conditions Wet conditions most crops, fertilizer lower barriers opportunities (+) (+) though less emissions; but under climate soil carbon change sequestration occurs Commercial n/a n/a n/a horticulture Dry conditions Access to (+) new markets, processing and value addition Agroforestry No effect in the Dry conditions Climate Climate Climate Low potential short term (+) change (–) change (–) change (++) of value chain Increase is less High levels development Wet conditions under climate of soil carbon (maybe (+) change sequestration nursery) and biomass growth Livestock n/a n/a diversification Wet conditions Dry conditions Dry conditions Increase in Opportunities Ongoing (–) (+) (+) livestock herds for veterinary policy reforms No significant increases services, expected to impact; methane trading, lower barriers geographical emissions processing variation Delayed n/a n/a n/a planting Dry conditions Development (–) of input markets and Wet conditions advisory (–) services Drought- and heat-tolerant seeds Dry conditions Dry conditions Dry conditions No notable No change No change Development Moderate (+) (+) (+) increase under under BAU; under BAU, of input constraint BAU; positive climate change positive effect markets can Wet conditions Wet conditions Wet conditions under climate could equally under climate create jobs (+) (+) (+) change increase/ change decrease the effect on mitigation Reducing post- n/a n/a n/a n/a harvest loss Source: World Bank, own elaboration Strong positive effect Low-medium positive effect Negative effect Note. (i) Colors (green, yellow, red) indicate the impact of CSA practices on key indicators; (ii) Dry conditions/Wet conditions +/- indicates whether increased or decreased rain have a positive or negative effect on CSA’s impact on the key indicator; (iii) Climate change ++/+/- corresponds to the error bars in GLOBIOM figures and indicates the magnitude and directionality of the range of effects that a CSA practice can have on indicators. It indicates whether the effect is strongly positive, or rather positive or negative under climate change projections. PAGE 11 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN would be below the opportunity cost of capital, with a 4.4 percent rate of return (see Figure ES.5). Overall, the rates of return of these delivery mechanisms are sizeable and should be considered when designing investment projects. A joint rollout of delivery mechanisms to more than 50 percent of farmers would yield a 20 percent rate of return on investment. The estimated annual investment requirements of US$32.6 million over 6 years is less than recent GoZ and donor funding for the agriculture sector. Thus, while investment quantity is necessary, the quality of delivery mechanism and targeting of investments to feasible and effective CSA practices is critical. Figure ES.5 Economic internal rate of return, in %, and net present value, in US$ million, of 7 delivery mechanisms, as well as aggregated 50% $71 $63.7 $61 39% 40% 36% $51 31% 30% $41 22% 21% $30.2 $31 19% 20% 20% $21 11% $13.3 $12.7 10% $9.6 $11 $7.9 4% $4.1 $0.6 $1 0% $9 -$10.6 -10% $19 Pluralistic Weather Weather Outgrower Integrated Business Integrated Farmers Aggregated paricipatory index index scheme landscape partnershiips landscape field schools (all delivery extension insurance insurance approach II with rural approach I mechanims) approach (with (without (focus on communities subsidy) subsidy) livestock) Source: World Bank, own elaboration Note: For each delivery mechanism economic indicators are calculated for approximatly 118,000 beneficiaries; aggregated this results in approximatly 826,000 beneficiaries To support the scale up of climate-smart agriculture practices in Zambia, a combination of investment approaches and changes in the policy and enabling environment are needed. The CSAIP concludes with recommendations of climate-smart agriculture practices which appear most robust under climate change projections and are largely effective in achieving household welfare and agriculture sector goals through 2050: crop diversification, commercial horticulture, agroforestry and reducing post-harvest loss. All of which are constrained by, among others, inadequate access to finance and markets, and lack of capacity building and skills development. A range of investment bundles and delivery mechanisms can address these constraints. Business partnerships with rural communities, farmer field schools, and participatory integrated land management approach seem most favorable and have highest economic rates of return on investment. The findings of the quantitative analyses and assessment of investment options lead to eight policy recommendations (Box ES.1), ranked by feasibility and relevance, which can further support the effectiveness of proposed delivery mechanisms and thus a scale up of climate-smart agriculture practices in Zambia. PAGE 12 Box ES.1 Policy recommendations The analyses of CSA practices under various scenarios and assessment of mechanisms to support CSA adoption show that business partnerships with rural communities which build on environmental sustainability as business strategy, farmer field schools to enhance community-based learning and technology dissemination, and participatory integrated landscape management approaches seem promising and profitable mechanisms to support the development of a productive, resilient, and low-emission agriculture sector. These can be supported by several policy actions: 1 Support farmers with improved access to inputs and finance, specifically in early stages of CSA adoption until benefits start to be realized. Support the development of markets for mechanization, innovative ICT-based solutions including agro-weather services for timely decision-making, and access to improved production inputs and seed varieties suitable for varied agroecological zones. 2 Support agricultural research, regional collaboration in research, participatory testing of CSA technologies to enhance their potential to increase crop yields, across and especially wet agroecological conditions. Support the advancement in the development and multiplication of seed varieties that are appropriate for rainfed production systems, in particular in the legume seed sector and short-duration maize seeds. 3 To increase agribusiness participation in the sector, a range of policy actions seem promising: (i) support feasibility, as well as risks and vulnerability assessments, to identify entry points and challenges; this should include an assessment of risk- and cost-sharing mechanisms which provide an understanding where the incentives align between the public good element of CSA and private sector motive; (ii) seek dialogue with agribusinesses about resulting investment opportunities; (iii) review the legal and regulatory framework to identify reform requirements and strengthen the business environment; (iv) provide opportunities for human capital development, capacity building and extension services for agriculture sector actors. 4 Support operationalization of holistic landscape management approaches by harmonizing policies and supporting cross-ministerial collaboration across the agriculture, environment, water and energy sector, and across administrative boundaries. Landscape approaches include climate-smart crop, livestock and forest management and have the potential to reduce alarming rates of land conversion for agriculture as well as addressing issues of biomass burning and charcoal production, which are key for Zambia’s carbon footprint. 5 Pursue the positive agriculture sector reform path of converting FISP into an e-voucher program, which supports agricultural diversification and adoption of short-duration varieties targeted to Zambia’s agroecological zones. The analysis shows the benefits of crop diversification into legumes, but also the need to improve the timing of planting or make short-duration varieties available to enhance crop yields. 6 Further support development of market infrastructure such as rural storage facilities and enable greater private investment in storage. Private investment may occur where production levels are high and stable and where access to border markets is feasible. In places where these conditions are not met, public investment may be required. In addition, training and collective action on improved post-harvest management bears a high potential to achieve CSA triple-benefits. PAGE 13 Section 1 Introduction 1.1 Agriculture Can Be Zambia’s Engine of Structural Transformation and Growth Zambia is a resource-rich, lower-middle-income economy that has successfully raised its average annual gross domestic product (GDP) growth rate since the early 2000s. Economic growth is heavily dependent on the minerals sector, and in particular on and changes in the international price of copper. Between 1985 and the 1990s, low levels of investment and production led to sluggish GDP growth, which rebounded between 2000 and 2014, when Zambia recorded economic growth of 7.4 percent per year; in the same period growth across sub-Saharan Africa averaged 5.8 percent. The country returned to middle-income status in 2011 as a result of improved economic management in the 1990s, a rebound in copper production, the expansion of the construction and services industries, and investment in the social sector by the government and cooperating partners (World Bank 2018e). Despite increases in per-capita economic growth, poverty and inequality in Zambia have remained high, particularly in rural areas. The proportion of Zambians living on less than US$1.90 per day increased from 49.4 percent in 2002 to 64.4 percent in 2010 before declining once more to 57.5 percent in 2015. Over the same period, inequality as measured by the Gini coefficient increased from 0.42 in 2002 to 0.57 in 2015. Inequality is particularly apparent when comparing between rural and urban areas. As rural poverty rose from 73.6 percent in 2010 to 76.7 percent in 2015, urban poverty fell from 25.7 percent to 23.7 percent during the same period (CSO 2011 and 2016; World Bank 2018a). The agricultural sector is the backbone of Zambia’s rural economy and arguably the country’s most important sector overall although it has been underperforming. Agriculture employs for 56 percent of the Zambian population, serves the critical function of buffering employment volatility in other sectors, and remains key to ensuring food and nutrition security particularly for financially vulnerable communities. While employment in the sector remains high, agriculture’s contribution to GDP declined from 17.3 percent in 2004 to 5.3 percent in 2015, evidence that most Zambians remain locked into low-productivity subsistence agriculture characterized by a lack of access to key productive assets and market opportunities (World Bank 2018e). In fact, Zambia has among the world’s highest rates of hunger and malnutrition, and although the incidence of stunting in Zambian children under five has declined from 45 percent in 2007 to 40 percent in 2015 it remains high (World Bank 2018e). PAGE 15 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Declining agricultural labor productivity is an impediment to an effective structural transformation and poverty reduction in Zambia. If the decline in agriculture’s contribution to Zambia’s GDP were matched by an increase in agricultural labor productivity and a reallocation of labor toward the more productive manufacturing, specifically agro-processing, sectors, this could be taken as a sign of structural transformation and economic modernization. However, labor productivity as measured by value added per worker in the agriculture sector decreased from US$702 in 2004 to US$584 in 2015 (in constant 2010 US dollars; see Figure 1.1) and has not been accompanied by a significant increase in the labor participation in highly productive sectors. While segments of the rural labor force are moving out of agriculture toward wage employment in Lusaka, most workers move out of agriculture into other low-productivity sectors including the informal services sector or else remain smallholder farmers. This indicates that structural transformation is not being driven by the agricultural sector and is thus unlikely to be effective in reducing rural and urban poverty (World Bank 2018a; Merotto 2017). Figure 1.1 Agriculture sector’s contribution to GDP and value added per worker 20 1,000 900 16.5 Percent contribution (bars) 15.0 800 US$ per person (line) 15 13.5 700 12.5 11.5 600 10.0 10.0 10.0 10 9.0 9.0 500 8.5 400 300 5 200 100 0 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Source: Chapoto and Chisanga (2016). Figure 1.2 Agriculture sector spending and budget allocation as a share of total budget 15% 15 15 13 12 10 11 10% 10 10 10 CAADP 8 target 8 6 7 5% 0% 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 allocated Actual (with additional funding to FRA and FISP) Source: Chapoto and Chisanga (2016). Note: Bar charts include funding for FRA and FISP. Green bars indicate years in which CAADP targets were exceeded. Low agricultural productivity is not necessarily the result of insufficient budget allocation and may instead reflect an ineffective use of funds. Poor agricultural productivity in Zambia can also be attributed to: limited access to land, water, and machinery; a lack of crop diversification; and low technology uptake (World Bank 2018e). Although the government subsidizes the agriculture sector, PAGE 16 it has not notably succeeded in overcoming these constraints and reducing rural poverty. Since 2003, Zambia’s budgetary allocation for agriculture has exceeded the 10 percent target set by the Comprehensive Africa Agriculture Development Programme (CAADP) six times (see Figure 1.2). Public expenditures include marketing support provided through the Food Reserve Agency (FRA) and input subsidies via the Farmer Input Support Program (FISP), which together accounted for an average of 79 percent of agricultural budgetary allocations between 2008 and 2016. Distribution of the agriculture budget has not sufficiently prioritized the investments needed to boost agricultural productivity growth and transform the sector (World Bank 2017a). 1.2 Agriculture Has a High Production Potential in Zambia Zambia has three distinct agroecological zones (AEZs), which are distinguished by varying rainfall, temperatures, and soil types (see Figure 1.3). AEZ I, which covers most of the country’s Southern and Western Provinces is a drought-prone area characterized by low rainfall (< 800 mm/year) and a short, hot growing season of 60–90 days. AEZ IIa and IIb cover much of Zambia’s eastern, central, and western regions Figure 1.3 Zambia’s agroecological zones and have the country’s highest agricultural potential with growing seasons of 90–150 days. AEZ IIa has slightly higher rainfall (800–1,000 mm/year) than AEZ IIb (600– 800mm/year). AEZ III covers the northern regions of the country, with 1,000–1,500 mm of rainfall each year and the growing season lasts 140–200 days (Braimoh et al. 2018). Climate change and the effects of long- term processes of degradation are expected to alter the boundaries of the established AEZs through geographic shifting of crop suitability zones. Zambia has generous endowment of natural resources. Despite high biophysical potential to diversify agricultural production, maize remains the I IIa IIb III dominant production system, covering approximately Source: Department of Meteorology, cited in Braimoh 2.7 million ha. Zambia has an estimated 74 million ha et al. (2018). of land, of which 47 percent is classified as agricultural land, and 15 percent is currently under cultivation (GoZ 2016; GoZ 2017). In 2020, it is projected that there will be roughly 2.2 ha of arable land available for each Zambian (Samboko et al. 2017). Zambia has approximately 40 percent of the Southern Africa region’s water resources. Although the country has 523,000 ha of irrigable land, only 155,890 ha, or 29 percent, are technically equipped for irrigation (GoZ 2017). Despite the potential for high productivity, about 90 percent of smallholders grow maize as their main crop, which covers about 57 percent of all arable land in Zambia. Tubers and cash crops account for around 18 percent across different farm sizes, and legumes occupy about 12 percent. About 70 percent of smallholder cultivate around three crops, of which 20 percent cultivate only one crops (Chapoto and Zulu-Mbata 2016; Arslan et al. 2018). In addition to low agricultural diversification, there is a relatively high yield gap per hectare and by farm type and most production increases are the results of expansion of agricultural area. Farmers with small landholdings produce on average less than 2 MT of maize/ha/year, whereas the average maize yields for large-scale farmers (those with >20 ha of land) can exceed 5 MT/ha/year (CSO 2018; MoA 2016). Due to the low productivity, agricultural production increases stem mainly from land expansion (see Figure 1.4 on maize production). Agricultural expansion, together with PAGE 17 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN deforestation for charcoal production, are the leading causes of Zambia’s alarming deforestation rate—roughly 300,000 ha of forest are destroyed each year which is threatening the sustained provision of forest-related ecosystem services to the rural population (Wathum et al. 2016; Turpie et al. 2015). Nearly two-thirds of Zambia’s forest area is on customary lands, and deforestation is generally attributed to charcoal production. This land is often converted to agricultural land once it has been cleared for charcoal (Turpie et al. 2015). Another study showed that in 2002 an estimated 90 percent of deforestation could be attributed to the clearing of land for agricultural production. Unsustainable practices like slash-and-burn hamper regeneration of land, and contribute to degradation and field abandonment, which forces farmers to expand further into forests for cultivation (Vinya et al. 2011). Figure 1.4 Maize production and yields 4.0 4.0 3.6 3.5 3.4 3.5 3.0 Million hectares (area planted) 3.0 2.9 2.9 3.0 2.8 Million tons (production) 2.7 2.7 2.6 2.5 2.4 2.5 2.3 2.3 2.0 2.1 2.1 2.0 1.9 1.9 1.9 2.0 1.8 1.5 1.4 1.5 1.0 1.0 0.5 0.5 0.0 0.0 2007-08 2008-09 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17 Area planted Production (million tons) Yield (tons/hectare) Source: Central Statistical Office, cited in World Bank (2017a). As domestic and regional food demand increases and diversifies, unprecedented opportunities are emerging for agriculture to become a driver of Zambia’s economic growth. Several factors contribute to a high potential for increased agricultural growth. Firstly, domestic demand for a diversified diet is growing quickly in tandem with strong economic growth and the rise of an affluent urban population. Food demand is expected to triple over the next 15 years, and there is a visible dietary shift toward more processed products, vegetable oils and horticulture. Zambia already has a trade deficit in processed foods, dairy and fish even as imports of cereal, vegetables, and milling products are declining (World Bank, 2018a). This suggests increasing opportunities to substitute imports with domestic production. Secondly, Zambia is ideally situated to serve a growing regional market. It shares a direct border with seven neighboring countries, in which food demand is also expected to increase. Zambia, a net exporter in cereals, experienced export growth of 38 percent between 2003 and 2012, primarily of wheat (31 percent of cereal exports), maize (24 percent) and rice (15 percent) (Fessehaie et al. 2015). To take advantage of these opportunities, the agricultural sector must raise productivity, diversify production and improve linkages to the agro-processing sector. Higher productivity, and diversification into higher-value horticulture and cash crops, would benefit household resilience by enabling farmers to produce more nutrient-dense food, gain access to diverse markets, and contribute to a more diversified and resilient economy (World Bank 2018d). Linking farmers to local and regional markets, value chains, and agribusiness will be crucial to unlocking the sector’s PAGE 18 potential. Strengthening the agribusiness sector can provide important upstream linkages to the agriculture sector, and represents a vital driver of productivity growth. Strengthening agribusiness will also enable the reallocation of labor and resources, provide additional income opportunities for rural populations, and enhance the effectiveness of Zambia’s structural transformation (World Bank 2018a). 1.3 Zambia’s Agriculture Sector is Vulnerable to Climate Change While there is ample potential to strengthen Zambia’s agriculture, the sector is likely to become more vulnerable to the effects of climate change and climate-induced risks. Over the last decades, climate variability and change have emerged as a growing threat to Zambia’s sustainable development. Climate trends between indicate that the mean annual temperature increased by 1.3°C between 1960 and 2003, or an average of 0.34°C per decade. Mean rainfall has decreased by an average of 1.9 mm/month, or 2.3 percent each decade since 1960. Zambia has experienced a series of climatic extremes that include drought, seasonal and flash floods, extreme temperatures, and dry spells, many of these of increased frequency, intensity, and magnitude (GoZ 2015). Braimoh et al. (2018) suggest that between 1982 and 2016, droughts caused US$438 million in agriculture-sector losses; excessive rainfall and flood cost an additional US$172 million in losses. Between 2007 and 2016, climate change cost Zambia an estimated 0.4 percent or US$13.8 billion in annual GDP growth. Over the next 10–20 years, agricultural losses related to climate change are expected to reach US$2.2–3.1 billion, largely as results of waterlogged fields, water shortage, destruction of crops, and a higher incidence of crop and livestock diseases (World Bank 2018e). Studies assessing the potential impact of climate change and weather events on Zambian agricultural indicators have returned mixed results. Jain (2007) estimates that farm revenue is negatively affected by higher mean temperatures in November–December and lower mean rainfall in January–February, but is positively affected by higher January–February mean temperatures. Thurlow et al. (2012), using a combination of a hydro-crop/dynamic computable equilibrium model, found that average maize yields can be expected to decrease by around 1 percent, except in western Zambia, where they are expected to increase slightly. Under more extreme climate scenarios, yield change is expected to be in the range of 4 percent to -6 percent. GDP is projected to decrease by -0.29 to -1.31 percent annually, and the poverty rate is expected to increase by between 1.69 and 7.23 percent, compared to a normal rainfall scenario. Whether climate change has a net positive or negative impact on agriculture will depend on the province and the crop. Kanyanga et al. (2013) combine a crop model with the International Model for Policy Analysis of Agricultural Commodities and Trade (IMPACT) to assess the impact of climate change on future food security in Zambia.1 Crop growth simulations broadly indicate that maize yields may increase by more than 25 percent in Western Province, the eastern half of North-Western Province, Copperbelt Province, and most of Northern and Luapula Provinces. Whereas yields are likely to decrease by more than 25 percent in Southern Province and parts of Eastern Province. Overall, production is increasing while net exports of nearly all modeled crops are declining. Results from Kanyanga et al. (2013) suggest that climate change will result in a decrease in net trade for most agricultural commodities. Crop yields are projected to grow between 2020 and 2050 but at lower rates (CIAT and World Bank 2017).2 1.4 Land Use Change Is the Largest Contributor to Zambia’s Carbon Footprint While Zambian agriculture’s carbon footprint is small on a global scale, the sector’s contribution to total national emissions is sizable (see Figure 1.5). Adopting CSA practices is more important than ever because they offer potentially substantial climate change mitigation co-benefits. Zambia produces total GHG emissions of approximately 388 million tons of carbon dioxide equivalent (MtCO2e) each year, PAGE 19 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN which is about three-quarters of a percentage point of global GHG emissions. Within Zambia however, land use, land-use change and forestry (LULUCF) together with the agriculture sector accounts for 328 MtCO2e and 22 MtCO2e of emissions, respectively, or a combined 93.1 percent of Zambia’s emissions. Energy accounts for approximately 6.5 percent, and is included in the category “other.” Ninety-three percent of LULUCF emissions are caused by burning biomass; the remaining seven percent are produced during the conversion of forest to cropland or other land use. Within the agriculture sector, the burning of savanna contributes to 59 percent of emissions (including fire management, slash-and-burn cultivation, and encroachment), followed by emissions from enteric fermentation (13 percent), manure left on pasture and manure management (12 percent), cultivation of organic soils (10 percent), and the use of fertilizer (5 percent), according to CIAT and the World Bank (2017). 1.5 Climate-Smart Agriculture Aims to Address Climate Challenges but Adoption Rates Remain Low Climate-smart agriculture (CSA) can bolster Zambia’s agricultural sector against potentially detrimental impacts and increased vulnerability arising from climate change. CSA aims at increasing productivity and resilience, while decreasing the sector’s greenhouse gas (GHG) footprint. CSA aims to better integrate agricultural development and climate responsiveness, while achieving food security and broader development goals. CSA initiatives sustainably increase productivity, enhance resilience, and reduce or remove GHG emissions. This requires strategic planning to address trade-offs and synergies between three pillars: productivity, resilience, and mitigation (FAO 2013). The technologies and practices that are considered climate-smart are diverse and vary across regions and production systems, thereby reflecting context-specificity or opportunities, constraints and vulnerabilities. A recent review finds that across 33 countries, five technology clusters account for almost 50 percent of CSA technologies, they include: water management, crop tolerance to stress, inter-cropping, organic inputs and conservation agriculture (Sova et al. 2018). Figure 1.5 GHG emissions: Zambia’s agriculture sector in perspective World: 20.7% 45,261 million tCO2e 6% Sub-saharan Africa Agricultural soils 47.2% Burning— Savanna 8% Zambia’s Agriculture Sector Cultivation of organic Sub-saharan Africa: soils 2,792 million tCO2e 14% Zambia 391 million tCO2e 9.9% Enteric 9.9% fermentation Manure left on 3.7% pasture Synthetic fertilizers Zambia: 1.2% Sources: CIAT Climate Data 0.7% 86% Land use change Crop residues Manure management Explorer; agriculture and 7% Other land use data from FAOSTAT 7% Agriculture 0.3% 0.7% Emissions Database. Burning crop residues Manure applied to soils Note: “Other sectors” contributing to Zambia’s GHG emissions include energy, waste, and = Crops = Livestock industrial processes. It seems to vary between 7 and 9%. PAGE 20 Despite acknowledgement of the need to address climate change adaptation and mitigation challenges, policy implementation and overall adoption of CSA in Zambia remains low. CSA is by no means a new concept in Zambia: in fact, numerous CSA practices have been tested across the country and the region. Yet although CSA practices such as minimum soil disturbance, crop rotation, and permanent soil cover and conservation agriculture have been promoted extensively in the country by donors, the GoZ, non-governmental organizations (NGOs), and civil society, adoption rates remain low. In fact, as much as 95 percent of farmers have adopted and then abandoned one or more CSA practices in some locations (Arslan et al. 2014). Among significant obstacles to higher rates of adoption and retention are that CSA practices tend to be laborious, and farmers have poor access to critical labor-saving equipment such as jab planters, direct seeders, and rippers, as well as limited knowledge and capacity. As CSA practices, particularly those regarding soil health, rarely provide immediate benefits, there is a risk that farmers adopt and then quickly abandon them (CIAT and World Bank 2017). 1.6 Climate Change Is Typically Factored into Agriculture Sector Policies and Strategies The GoZ has demonstrated increased commitment to addressing climate change concerns through its policies and strategies. Zambia has developed various climate change-focused policies, strategies, and projects, including: the National Policy on Environment (GoZ 2007a); the National Climate Change Response Strategy (GoZ 2010); the National Strategy for Reducing Emissions from Deforestation and Forest Degradation (REDD +) (Matakala et al. 2015); the National Adaptation Programme of Action on Climate Change (GoZ 2007b); the Nationally Appropriate Mitigation Actions; and the National Climate Change Policy (GoZ 2017a). These initiatives are aligned with the Seventh National Development Plan (GoZ 2017b) and the Vision 2030 (GoZ 2006), to promote “[a] prosperous middle-income country by 2030.” Both support the advancement of a low carbon and climate-resilient development pathway. Zambia’s Nationally Determined Contribution (NDC) to the 2015 Agreement on Climate Change (The Paris Agreement) is an ambitious goal of both climate change mitigation and adaptation initiatives. The NDC sets a target of reducing emissions by between 25 and 47 percent from a 2030 baseline, contingent on the availability of international funding sources. Total emissions reductions of 38 MtCO2e have been pledged, conditional on US$35 billion in international assistance. The NDC acknowledges that climate-smart agriculture offers a path to achieving this goal through reductions in fertilizer use and turning of soil, increased soil carbon sequestration, and enhanced biogas factories. Government planning documents demonstrate that clear and consistent progress has been made in mainstreaming climate change considerations and CSA strategies into the relevant sectoral frameworks. Following revision of the National Policy on Environment (2007) and development of the National Climate Change Response Strategy (2010), specific reference to and elaboration of responsive climate change adaptation and mitigation actions have been incorporated into all of Zambia’s major planning documents (see Table 1.1). The ongoing revision of several sectoral policies, including the second National Agricultural Policy (2016), Extension Strategy (2016), and Forestry Policy (2014), suggests that the mainstreaming of climate change and CSA extends to sectoral planning efforts. Other sector policies currently under revision, such as the Irrigation Policy, Land Policy, and Livestock Policy, are also expected to reflect a climate change focus (summaries of the relevant policies are included in Appendix A). Owing to the broad-based consultative processes used to develop and update these policy documents, which involved all levels of government as well as outside stakeholders, mainstreaming efforts can be expected to extend well beyond the responsible ministries. PAGE 21 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table 1.1 The mainstreaming of climate change and CSA considerations within government policies Climate Change Adaptation Mitigation CSA Vision 2030 (2006) National Policy on Environment (2007) National Adaptation Programme of Action on Climate Change (2007) National Climate Change Response Strategy (2010) National Agriculture Investment Plan (2013) Nationally Determined Contributions on Climate Change (2015) National Disaster Management Policy (2015) Reduced Emissions from Deforestation and Forest Degradation (REDD+)(2015) National Policy on Climate Change (2016) Second National Agriculture Policy (2016) Seventh National Development Plan (2017) Climate-Smart Agriculture Strategy Framework (forthcoming) The finalization of a national-level Climate-Smart Agriculture Strategy Framework establishes priorities and mechanisms to strengthen CSA initiatives across the country. The strategy paper, which is being prepared under the leadership of the Ministry of Agriculture, identifies several steps to enable scale-up of CSA: (i) develop an institutional mechanism and strategic operational framework that permits effective coordination and communication among CSA stakeholders; (ii) identify CSA approaches and practices that are responsive to climate change and can attract the required budgetary support; and (iii) strengthen mechanisms for dissemination of information, knowledge, and skills about appropriate CSA practices. The paper underscores that support for scaling up CSA is predicated on assessment and communication of the local and national benefits of CSA investments (including food security and global environmental benefits), specifically the “Development and utilization of evidence to help identify [CSA] investments specific to the locally present climate change, agricultural production and institutional conditions, focus on barriers to implementation and means of overcoming them, and building local institutional and policy frameworks to support the needed transformation” (unpublished draft, page 55). It will be essential to monitor gaps in the implementation cycle to ensure that CSA-related actions are integrated into the appropriate planning documents and ultimately implemented in the field. Primary concerns include: (i) to ensure that the formulated implementation plans respond to observed weather-related stressors affecting different locations across the country; (ii) to allocate sufficient funding from domestic and external sources to support implementation; (iii) to establish a means of directing funds to critical activities and locations and tracking how these funds are ultimately used; and (iv) to ensure that facilities and trained personnel necessary to effectively carry out the planned activities are in place. 1.7 Objectives of the Zambia Climate-Smart Agriculture Investment Plan The overarching goals of the Climate-Smart Agriculture Investment Plans (CSAIP) are “to identify and prioritize key policy actions, investments and knowledge gaps” and build on existing policies, PAGE 22 strategies, and lessons learned through engagement with the agriculture sector to assess how CSA investment can strengthen climate change adaptation and mitigation initiatives and prioritize investments that promise cost-effective CSA approaches to achieving overall sector goals. The Zambia CSAIP is one element of the World Bank’s corporate commitments to develop ten CSAIPs made during the Eighteenth Replenishment of the International Development Association (IDA18). The CSAIP responds to the GoZ’s needs that are highlighted in the Climate-Smart Agriculture Strategy Framework, and aims to inform the development of a second National Agriculture Investment Plan (NAIP). The CSAIP provides evidence of the local and national benefits and impact of CSA, assesses and prioritize climate-smart interventions, and identifies opportunities to invest in CSA and provides an indicative costing and economic analysis of these opportunities. In addition to informing development of a second NAIP, which will cover the 2022–2026 period, the CSAIP will also inform implementation of Zambia’s Seventh National Development Plan. Unlike the first NAIP, the second one is expected to explicitly address climate change challenges and climate-smart agriculture. The CSAIP will provide quantitative data on the benefits of CSA under climate change, suggest investment opportunities and an indicative costing; however, the CSAIP will not provide a detailed investment plan to replace the NAIP. To achieve these goals, the development of the CSAIP was guided by a set of questions. See Table 1.2 for an overview of these questions and approaches for addressing them. The World Bank is demonstrating commitment to support and promote CSA strategies through implementation of analytical and operational projects as well as through the Zambia Country Partnership Framework (CPF). The CSAIP supports this agenda. Through Focus area I, the CPF promotes a more even territorial development by supporting opportunities and jobs for the rural poor. It aims to support sustainable, competitive, and diversified agro-food production as well as increased resilience to hydrological, climate and environmental shocks. Thus, the adoption of improved and climate-smart technologies, which is integral to the Zambia CSAIP, plays a critcial role. Besides the CPF’s objectives, there are ongoing or recently completed projects aimed at strengthening CSA, improving landscape management, and providing benefits for rural communities. These include: the Zambia Integrated Forest Landscape Project in Eastern Province; the Strengthening Climate Resilience Project; and two analytical knowledge products: “Productive Diversification in African Agriculture and Impacts on Resilience and Nutrition” (Kray et al. 2018) and “Increasing Agricultural Growth and Resilience through Better Risk Management in Zambia” (Braimoh et al. 2018). The Zambia CSAIP builds on existing work and, by pointing out knowledge gaps, aims to inform future analytical and operational work. 1.8 Report Structure The report is organized as follows: • Section 2—Methodology introduces the overarching methodological approach of CSAIPs in general as well as the specific quantitative and qualitative approaches that were used with respect to the Zambia CSAIP. • Section 3—State of CSA adoption and performance in Zambia provides a brief literature review for eleven CSA practices which are subsequently analyzed. • Section 4—Analysis Results: Observed impacts of CSA on households summarizes findings from econometric assessments of the effects of CSA implementation on household-level welfare indicators. The section also presents the impact sensitivities of high weather variability (a proxy for PAGE 23 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table 1.2 Summary of CSAIP guiding questions and key approaches Guiding questions Overview of approach as employed in the report Sector vision and policy relevance: How should the sector Stakeholder workshops held in Lusaka in October 2017 and look in the future? What are the policy goals? April 2018 • What are the quantifiable targets to achieve the Policy analysis and contextualization: Identification of sector vision, and which agricultural strategies are agriculture sector goals, targets (for example., as outlined recommended to support these goals? in the Country CSA Program and NDC), and challenges; selection of potential CSA practices through (a) extensive • What are the key trends and uncertainties confronting reviews of policy documents and literature and (b) the sector? stakeholder consultations at several stages of the process Potential impacts on households: Do households have Empirical assessments of observed impacts of CSA: sufficient incentives to adopt CSA strategies? • Cost-benefit analysis at the household level to assess • Which CSA strategies are promising and what are their financial viability projected impacts on key variables that are proxies for • Econometric analysis to evaluate the impact of CSA household welfare, such as household income and practices on household-level welfare indicators variability, food security, and poverty? Literature review and expert consultation to contextualize • What is the financial viability of CSA strategies for and validate findings households? • Which CSA strategies are robust in light of weather and climate variability? • How does the suitability of CSA strategies vary across Zambia’s AEZs? Potential impacts at the sectoral level: Do CSA practices Agriculture sector model GLOBIOM: Integrative have an advantage to deliver key agriculture sector indicators quantitative agriculture sector modeling to assess the compared to conventional practices, and are benefits robust robustness of selected CSA practices against climate change across all climate change scenarios? in different regions of the country • What is the CSA strategies aggregate effect on Zambia’s Cost-benefit analysis at the sector level: To assess agriculture sector and key variables, for example, economic profitability of supporting CSA adoption with production, mitigation, land use change, trade—in a public sector funds future with and without climate change? • How do CSA strategies support the achievement of the agriculture sector vision and intermediate targets, also under climate change scenarios? • Is there an economic rationale to support CSA scale up? Prioritization: To which CSA practices should financial Prioritization: Conduct a prioritization and ranking of resources and institutional capacity be earmarked? candidate strategies based on their likely impact and feasibility. • Which CSA strategies have proven promising at the household and sectoral levels and should receive particular attention from the government and donor community? Delivery: What is required with respect to public sector Literature review and expert consultation to elaborate financing and innovative delivery mechanisms to ensure the constraints in the enabling environment and key delivery adoption of CSA? mechanisms to support the CSA scale-up • Which factors in the enabling environment are critical to Expert consultation: Assessment of current and necessary achieve adoption of CSA? public sector funding to scale up CSA and achieve agriculture sector goals • What are innovative implementation arrangements that can support the adoption of prioritized CSA practices in Zambia? • Which CSA strategies have previously received financing and what are expected expenditures needed to enhance their adoption? climate change impacts) based on household survey data. The section also presents results of an analysis of the financial viability of adopting certain CSA practices at a household level. • Section 5—Analysis Results: Projected impacts of CSA on agriculture sector performance under climate change presents the results of the GLOBIOM model, which assesses the impacts PAGE 24 of adopting CSA practices through 2050 compared to conventional practices in a future with and without climate change. The impact is assessed on several agriculture sector indicators, which approximate the CSA areas of productivity, resilience and mitigation. • Section 6—Prioritizing CSA practices ranks CSA practices according to their performance on household and sectoral level. • Section 7—Enabling Environment, Investment Options, and Financing to Promote CSA Adoption discusses several dimensions of the enabling environment that facilitate and support the adoption of climate-smart technologies and practices; presents a summary of eight proven and innovative mechanisms that could support adoption of CSA; and provides the results of an economic analysis of each mechanism including an assessment of total investment needs. • Section 8—presents the report’s conclusions and recommendations. PAGE 25 Section 2 Methodology 2.1 The CSAIP Approach: A Flexible Framework The CSAIP approach proposes a flexible framework to support countries’ agriculture sector planning in the face of climate-related uncertainty. CSAIP incorporates quantitative modeling and analyses, stakeholder engagement, expert interviews, and a literature review, and comprised four key steps that can be tailored to fit a country’s context (see Figure 2.1): • Step 1: Identify agriculture sector goals, targets, and strategies (see Section 2.2). A collaborative, stakeholder-driven process took place to identify visions, goals, and strategies and to prioritize a handful of promising climate-smart technologies for Zambia’s agriculture sector. As part of this process, a target year was determined, and sector goals or a “normative vision” of the sector developed that factors in national development plans, climate commitments, and ambitions for that year. Working backward from this normative vision, intermediate sector targets and priority strategies were identified that connected the normative vision with the present. • Step 2: Identify key uncertainties and known trends (see Section 2.2). The second step of the stakeholder-driven process entailed developing key uncertainties and known trends into a set of potential alternatives, plausible futures, and scenarios. • Step 3: Perform quantitative and qualitative analyses (see Section 2.3–2.5). These analyses projected the scale of potential impact that CSA technologies could have on achievement of sectoral goals and key socioeconomic and environmental variables in a future with and without climate change. This enabled the assessment of CSA technologies’ desirability at the household and sectoral level. In Zambia, this step combined three quantitative approaches—an agriculture- sector model; cost-benefit analyses (CBA); and an econometric analysis to assess household- level impacts—with qualitative analyses of the enabling environment and of potential delivery mechanisms to support CSA adoption as well as past and future public-sector financing needs. See Table 1.2 for the guiding questions for Step 3. PAGE 27 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Step 4: Prioritize and evaluate CSA strategies according to quantifiable indicators and key uncertainties. Based on this assessment, investment opportunities were identified which are innovative or have proven successful in scaling up adoption of certain CSA practices. Figure 2.1 Integration of qualitative and quantitative approaches into CSAIP I. Sector goals and strategies II. Scenario development Stakeholder consultation and review of policy documents Stakeholder consultation to identify key uncertainties to identify agriculture sector vision, targets, and strategies which can impede the achievement of sector goals Key driver of uncertainty 1 Present Which CSA strategies are needed today? Scenario A Scenario B 2020 2030 2040 2050 Scenario C Scenario D 2030 2050 Milestone Milestone Key driver of uncertainty 2 Quantifiable “Normative” vision indicators and of the future in milestones target year III. Quantitative & qualitative analyses IV. Prioritization and evaluation Agriculture sector model GLOBIOM adjusted for Zambia CSA strategies are evaluated according to key to determine impacts of CSA on sector goals and indicators indicators, e.g. sector and household level impact or in scenarios with and without climate change feasibility, under consideration of uncertainty scenarios (See Appendix E) Indicator categories Empirical assessment: to analyze the impact of CSA on A B C D several indicators Uncertainty scenarios  Cost-benefit analysis Strategy 1  Econometric analysis Strategy 2 Literature review and expert consultation: To assess the enabling environment, innovative delivery mechanisms, and past and future public sector financing, to support Strategy 3 adoption of CSA Source: World Bank 2.2 Stakeholder Consultations to Determine a Vision, Goals, and Uncertainties for the Agriculture Sector A vision for the agriculture sector was identified through a combination of policy review and stakeholder engagement. As a first step, the team conducted a comprehensive literature and policy review, and gathered and analyzed available data on Zambia’s agriculture sector and climate change situation. Specifically, the analysis looked at whether and how sector policy and planning frameworks that were in effect between 2005 and 2017 planned to address challenges and opportunities to achieve the three CSA pillars of productivity, resilience, and mitigation. During a technical stakeholder workshop held in Lusaka in October 2017, participants, including ministerial officials, NGOs, think tanks and international organizations (see Appendix H for a full list), jointly reviewed existing national and agricultural planning reports and formulated a normative vision for the development of Zambia’s agriculture sector (see Table 2.1). As a next step, measurable targets were identified to assess whether and to what extent CSA practices might contribute to the agriculture sector vision. Participants identified and prioritized measurable sector targets across the three CSA pillars that would be critical to achieving the vision. See Table 2.2. for an overview of measurable targets and indication whether and how CSAIP models address PAGE 28 the target, and which CSA objectives are being addressed (see Appendix A for a detailed overview of relevant policy documents, as well as the sector vision and quantifiable targets where available). In addition, a number of promising CSA strategies were identified that offer the potential to achieve specific intermediate goals and, ultimately, the normative vision for the agriculture sector (see Appendix B for strategies identified during the 2017 technical stakeholder workshop). Following the workshop, a number of CSA strategies were chosen as priorities through an iterative process that involved key technical focal points from Zambia’s Ministry of Agriculture. Table 2.3. presents those practices, as well as the quantitative approaches used to analyze them. Due to data constraints, not all practices could be analyzed with each method. Table 2.1 Normative vision for Zambia’s agriculture sector developed during the stakeholder workshop Normative vision as developed in stakeholder Source CSAIP modeling targets(a) CSA pillars workshop By 2050, double yields and profits by means of Doubling of crop yields Stakeholder diversification (beyond maize), while ensuring household compared to baseline (2009- Productivity workshop food and nutrition security 2011 average) By 2050, develop an agriculture sector that is (i) diversified No target; assessment of Stakeholder in crop production, (ii) diversified in age and gender, impact of crop diversification Resilience workshop and (iii) able to cope with economic and climatic shocks scenario on key agriculture through enhanced capacity and policy sector indicators 25-47% decrease in GHG Stakeholder By 2050, increase the productivity of Zambia’s agricultural emissions in comparison to Mitigation workshop sector while maintaining a low ecological footprint BAU scenario Note: (a) For the modeling approach, the quantifiable targets as stated in policy documents and by stakeholders, were approximated based on data and information availability. For the normative vision of diversifying crop production no quantifiable target was available. Table 2.2 Quantifiable targets for achieving the long-term vision for Zambia’s agriculture sector Quantifiable targets to achieve Policy CSAIP modeling targets(a) CSA pillars the long-term vision Seventh Raise agriculture’s share of the GDP by 10% by 2021 N/A Productivity National Development Raise the value of agriculture, livestock, and fishery exports N/A Productivity Plan (2017) to US$931 million, or 7.8% of total export value, by 2021 National Increase the share of agricultural exports as a percentage of Doubling of net crop exports Productivity Agriculture non-traditional exports from 41% in 2011 to 55% by 2018 compared to base year 2010 Investment Plan (2014– Increase production of cereals from 3.2 million MT to 6.0 Achieve total production of 6 Productivity 2018) million MT million MT Cropland increase < 900,000 Increase land under cultivation by 900,000 hectares by 2030 ha (average 2000-2013 as Productivity National Long baseline) Term Vision 6 million cattle, double 2030 (2006) number of small ruminants Productivity, Increase livestock population to 6,000,000 animals by 2030 (average of 2012-2014 as Resilience baseline) National Sum of cropland and Policy on Sustainably intensify land use without converting additional grassland do not increase Resilience, Environment land area to agricultural land (average of 2000-2013 as Mitigation (2007) baseline) % decrease (25-47%) in Nationally Reduce GHG emissions by 25% with limited international GHG emissions by emission Determined support or 47% conditional on the receipt of US$35 billion category compared to Mitigation Contribution of international assistance business-as-usual scenario (2015) in 20504 Note: (a) For the modeling approach, the quantifiable targets as stated in policy documents and by stakeholders were approximated based on data and information availability. Some targets could not be modeled (n/a). PAGE 29 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table 2.3 CSA practices incorporated into the analysis and overview of quantitative assessments Household level Sector level Cost-Benefit Analysis: Econometric Financial Economic GLOBIOM analysis CSA Practice analysis analysis Conservation agriculture’s 3 pillars: Minimum soil disturbance Residue retention Legume rotation / intercropping Agroforestry Commercial horticulture Crop diversification Livestock diversification Drought-tolerant seeds Timing of planting Agricultural liming(a) Reducing post-harvest loss Note: (a) A statistical analysis is performed for agricultural liming due to data unavailability. Lastly, workshop participants identified a range of factors that could harm the agriculture sector and impede achievement of its goals. These factors were categorized as either climate change- related or socioeconomic uncertainties, and were used to design scenarios for projections with the modeling framework. The first results of the analysis were presented and discussed during a subsequent stakeholder workshop in April 2018. 2.3 Econometric Analysis to Determine CSA Impacts on Welfare of Zambian Households An econometric assessment enabled the estimation of CSA strategies’ impact on a range of welfare indicators. The analysis provided responses to a number of key questions: (i) What are the potential impacts of adopting CSA on key CSA variables and on household welfare, for example, household income and variability, food security, and poverty? (ii) Which strategies are robust in the light of weather variability? Which CSA strategies are suited to the specific AEZs in Zambia? See Table 2.4 for an explanation of the six welfare indicators. Also, see Appendix C for more details. Two econometric approaches were used to assess the impacts of adoption of CSA practices on key welfare outcomes. The first approach incorporated a difference in difference (DiD) estimator to assess the impacts CSA adoption had on several welfare variables (see Table 2.4), and provided an estimate of the direct causal impact of CSA adoption. However, one limitation of the approach is that the sample of adopters is restricted to only recent adopters, that is, those that adopted CSA between the 2012 and 2015 household surveys. This represents a significant constraint because the benefits of some CSA approaches are not immediate. Therefore, a second approach was used to complement the first: CSA adopters were classified as households that adopted the practice in the 2012 and 2015 surveys. The impacts of CSA adoption were then estimated using an Ordinary Least Squares (OLS) approach. A drawback of this second approach is that it fails to control for selection bias, which raises concerns about endogeneity. Accordingly, results from the second approach are interpreted as conditional correlations rather than causal impacts. PAGE 30 Table 2.4 Outcome indicators of the welfare analysis Indicator Definition Sum of the value of all crops produced, including crops sold and crops retained for Gross crop income consumption; crop prices are estimated based on the average farm-gate prices at a district level Crop income variability The coefficient of variation of gross crop income Sum of all income sources, including off-farm sources and the value of retained Gross income agricultural production Income variability The coefficient of variation of gross household income Poverty Per capita net income of less than US$1.25 a day in real 2010 US dollars Negative response to the survey question “did the household have sufficient food to meet Food insecurity household food security needs in agricultural season 2013/14” The econometric analysis is based on the Rural Agricultural Livelihoods Surveys (RALS) household survey. RALS is a nationally representative survey of Zambia’s smallholder households that cultivated fewer than 20 ha in 2012 and 2015 (CSO et al. 2012 and 2015). The survey contains georeferenced information on sociodemographic characteristics, farm management practices, access to credit and assets, access to markets and infrastructure, information availability, inputs used, and outputs produced. The survey comprised 7,254 households in 2012 and 7,934 households in 2015. To assess the performance of CSA practices under weather variability, the Standard Precipitation Index (SPI) was added to the econometric analyses. Extremely low and extremely high rainfall approximate climate sensitivity. The SPI was developed primarily to define and monitor drought, and can also be used for flood identification. SPI can be used to determine the probability of drought at a given point in time for any rainfall station with historic data. All econometric models were re-estimated using combinations of the SPI between a certain threshold to capture extreme dry and extreme wet periods. Thus, the estimation results provide an indication whether specific CSA practices are more effective under very dry or wet conditions, which approximates climate sensitivity. 2.4 Cost-Benefit Analysis to Determine Financial and Economic Profitability of CSA The cost-benefit analysis (CBA) includes a financial and economic analysis to determine the profitability of adopting CSA at the household and sectoral level. The analysis provides answers to two guiding questions: (i) What is the financial viability of CSA practices for a household? (ii) How do the anticipated costs needed to scale up CSA compare to the anticipated economic benefits? First, a financial analysis was undertaken to determine financial viability and incentives for an average household to adopt CSA in several AEZs. This was done by assessing the net incremental benefits accruing to a household following the adoption of CSA practices as compared to keeping conventional farming practices. Second, an economic analysis was used to assess the economic and societal benefits of adopting CSA, and to determine whether there was an economic rationale for the public sector to support widespread adoption by farmers of CSA. This served to answer the underlying question of whether public sector financing is warranted to support the adoption of CSA. See Table 2.5 for resulting indicators; see also Appendix D for details. The financial analysis is based on a unique household data set on CSA adoption as well as the RALS survey. FAO collected the CSA-specific household data, which provides information for 695 households about crop yields, income, and production costs associated with adoption of CSA practices in the 2012–13 cropping season. The households were determined according to a stratified PAGE 31 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN random sampling technique, and data about crop production was reported at field level (that is, 1,264 fields in AEZ IIa and III). Among other data, the dataset included information on cropland use, input quantities and cost, labor use in different management activities, and household demographic characteristics. This survey covers selected CSA practices in several AEZs, which was complemented by findings from the RALS household survey. Additional variables were incorporated into the economic analysis: economic benefits for households adopting CSA, public good benefits accruing from climate mitigation, and expected public sector investments. For the economic analysis, households’ financial net incremental benefits were evaluated at economic prices and aggregated across the projected number of CSA adopters. The Ex-Ante Carbon Balance Tool (EX-ACT) was used to assess public good benefits arising from CSA’s climate mitigation potential. EX-ACT enables calculation of the net carbon balance, that is, the difference between gross emissions under conventional practices and gross emissions under CSA practices (see Table 2.5). To quantify the value of climate mitigation, the net carbon balance is valued at a shadow price of carbon that was calculated based on the concept of marginal abatement cost.5 The shadow price of carbon indicates the carbon price which is consistent with achieving the core objective of the Paris Agreement on climate change (2015) of keeping temperature rise below 2 degrees. To assess the public expenditure needed to support CSA adoption at the sectoral level, information was sourced from previous Zambian investment projects, gray literature and personal interviews with government institutions, independent research entities and NGOs, and international institutions. The economic analysis was performed for two adoption scenarios. Scenario one: 8 percent of farmers adopt CSA, which, according to the RALS survey, reflects the current adoption rate for minimum soil disturbance. This is equivalent to 118,000 households, or 6.4 percent of households in AEZ I; 13.9 percent of households in AEZ IIIa; 1.1 percent in AEZ IIb households; and 4.7 percent of households in AEZ III. On average, households cultivate 1.71 ha. Scenario two: 50 percent of farmers adopt CSA. An adoption rate of 50 percent for minimum soil disturbance was discussed during the 2017 stakeholder workshop. It can be considered as aspirational target for 2050. Table 2.5 Outcome indicators of the cost-benefits analysis Indicator Definition Financial analysis Gross margins represent total revenues from crop production minus total production or Gross margins variable costs excluding labor Net margins are calculated as the gross margins, but factor labor costs into total production Net margins costs Net incremental This represents the expected financial returns for households generated by the adoption benefits of adoption of CSA activities as compared to conventional farming methods. Economic analysis The net carbon balance is the difference between GHG gross fluxes under CSA adoption and under conventional farming practices. Results are given in tons CO2 equivalent (tCO2e). Net carbon balance Positive numbers represent sources of CO2e emission while negative numbers represent a carbon sinks. This is valued at a shadow price of carbon and included in the economic analysis as public good benefits. The value is assessed with EX-ACT. The NPV is the difference between the present (discounted) value of cash inflows and the Net present value (NPV) present value of cash outflows over a period of time. Economic internal rate of The EIRR is the discount rate at which the calculated NPV equals zero. A high EIRR provides return (EIRR) confidence in the profitability of an investment. PAGE 32 The robustness of the economic benefits is determined through a sensitivity analysis incorporating variations of key variables. The sensitivity analysis considers the performance of the investment options under different scenarios, including changes in the flow of benefits and costs, changes in the adoption rate, and changes in crop yields. 2.5 GLOBIOM Approach Used to Assess Agriculture Sector Goals GLOBIOM is an integrated modeling approach developed by IIASA that enables joint analysis of the agriculture, forestry, and bioenergy sector. GLOBIOM is a spatially explicit partial equilibrium model of the global forestry and agricultural sectors (Havlík et al. 2014) used to analyze the competition for land use between the main land-based production sectors: agriculture, forestry, and bioenergy. GLOBIOM enables assessment of the aggregate effects of CSA technologies on key agricultural sector variables; agricultural and land use change and trade-offs and synergies around land use and ecosystem services in scenarios that include and exclude climate change. Thus, GLOBIOM enables benchmarking of CSA’s impacts against agriculture sector targets and the normative vision. See Appendix E for more details. GLOBIOM has been used extensively to analyze the medium- and long-run effects of climate change (Leclère et al. 2014; Havlík 2015); deforestation (Schmitz et al. 2014); food security (Valin et al. 2014; Palazzo et al. 2017); and bio-fuel policies (Valin et al. 2013; Havlík et al. 2011; Mosnier et al. 2013; Frank et al. 2013) at the global and regional level. The model has also been used for several in- depth country case-studies, including deforestation assessments for the Congo Basin (Mosnier et al. 2014) and Brazil (Soterroni et al. 2018), and analysis of Russia and Ukraine’s production potential (Deppermann et al. 2018). The model computes the market equilibrium for agricultural and forest products by allocating land use among production activities to maximize the sum of producer and consumer surplus, contingent on resource, technological and policy constraints. Agriculture and forest productivity is based on input from two models: Environment Policy Integrated Climate (EPIC), Lund-Potsdam-Jena managed Land (LPJmL) and the Global Forest Model (G4M). GLOBIOM captures production systems and land use in the base year 2000, using historical data from the Spatial Production Allocation Model (SPAM; You and Wood 2006), which provides the physical area for 17 of the 18 crops included in GLOBIOM under four crop management systems: subsistence farming, low input rainfed, high input rainfed, and high input irrigated (Balkovič et al. 2013). Production is calibrated to match FAO’s country- level statistics, and demand is modeled at the regional level using 30 economic regions which consist of single countries or bundles of countries based on food balance sheets developed by FAO (FAOSTAT 2017). GLOBIOM relies on a recursive dynamic approach combined with exogenous population and economic growth trends to create future projections for key indicators (See Table 2.6). For this study GLOBIOM has been modified and updated with country-specific information to approximate Zambia’s agriculture sector. This includes agricultural production statistics, land cover maps, and data on irrigation, costs, and livestock enabling the generation of a range of outcome indicators (see Table 2.6). GLOBIOM enables modelling of the impact of CSA practices on these indicators. Exploratory scenarios were used to capture and model the high level of uncertainty arising from the development of outcome indicators. This framework is conducive to assessing uncertainties arising from climate change and the possible trajectories of socioeconomic drivers of agricultural growth. In Zambia, drivers of agricultural development include both local and global factors set out in PAGE 33 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table 2.6 GLOBIOM outcome indicators Indicator Definition Crop production is determined by the agricultural or forestry productivity of a specific area, which is in turn dependent on the crop suitability and management practices, Crop production market prices which reflect demand, and the conditions and cost associated with land conversion to agricultural purposes to expand production. Using average crop yield provides an indication of the overall productivity of cropland and Crop yield context for examining the potential impacts of CSA technologies. Livestock is modeled using a detailed representation of the global livestock sector in which distinctions are made among dairy and other bovines, dairy and other sheep and goats, laying hens and broilers, and pigs (Havlík et al. 2014; Herrero et al. 2014). Livestock Livestock production production activities are defined by production systems (Herrero et al. 2013): Ruminants, and heads of livestock grass-based or mixed crop-livestock (arid, humid, and temperate/highlands), and other monogastrics, smallholders and industrial. For each species, production system, and region, a set of input-output parameters is calculated based on the approach set out in Herrero et al. (2013). Land use change is examined for irrigated and rainfed cropland, grassland, forest, and other Cropland and natural lands to determine the extent to which productivity gains spare land for protection land use change of terrestrial ecosystems and climate stabilization. Demand for crop products for food, livestock feed, bioenergy, and fiber is used as indicator for food security of a growing population. The analysis uses kilocalorie availability per capita Calorie availability per day to measure food security. It considers the total food products demanded by a region, and translates the quantity into calories. Net agricultural trade (exports minus imports) is provided for a number of key agricultural Net Trade commodities. The model assesses imports or exports as a share of domestic consumption, to assess relative dependence on trade partners. GLOBIOM accounts for 10 sources of GHG emissions: nitrous oxide (N2O) emissions created by fertilizer use; methane (CH4) emissions from rice cultivation; CH4 emissions from livestock; CH4 and N2O emissions from manure management; N2O from manure applied on pasture; AFOLU GHG emissions and above- and below-ground carbon dioxide (CO2) emissions from biomass removal after converting forest and natural land to cropland. GLOBIOM cannot model below-ground carbon sequestration, which might occur with no tillage and agroforestry practices, or emissions from burning of biomass and savanna. the shared socioeconomic pathways (SSPs) scenarios. SSPs capture a set of plausible potential future socio-economic developments in the absence of climate change and climate policies. For this analysis the projections of SSP2 “Middle of the Road” (Fricko et al. 2017) were adopted to represent a business- as-usual (BAU) scenario. Sensitivity analyses were conducted for SSP1 “Sustainability—taking the green road”, and SSP3 “Regional rivalry—a rocky road”. See Appendix E.1 for a description of all SSPs. A combination of climate models is used to assess the impact of climate change by comparison with a BAU scenario. Five general circulation models (GCMs) and two crop models were used. The crop models EPIC and LPJmL provide the bandwidth of the yield shock caused by climate change in comparison to the absence of climate change. Yield simulations are performed for representative concentration pathway (RCP) 8.5, which shows the most extreme climate outcomes (Riahi et al. 2011) and a scenario variant with and without a carbon dioxide (CO2) fertilization effect. See Appendix E.1 for details. PAGE 34 Section 3 State of CSA Adoption and Performance in Zambia Section summary Several CSA practices have been piloted in Zambia. Figure 3.1 provides maps of CSA adoption, which ranges between 4 percent (agroforestry) and 60 percent (residue retention). Literature presented in this section shows that CSA has a high potential for increasing yields and income, under certain agroecological conditions and after several years (for example, agroforestry). Barriers to adoption are high and relate to labor costs (for instance, minimum soil disturbance) or the lack of production inputs (for example, mechanization and herbicides in minimum soil disturbance, saplings for agroforestry, seeds for drought-tolerant seeds, planting on time). Diversification towards livestock, legumes and horticulture are promising due to high domestic demand. Maize-centric distortionary policies, lack of rural market infrastructure, and outputs markets remain major constraints for smallholders. 3.1 Minimum Soil Disturbance Minimum soil disturbance, mostly the use of planting basins or ripping, is practiced by 7.8 percent of smallholder farmers in Zambia. However, the cost and benefit track record is variable and context specific, which contribute to elevated risk and uncertainty for farmers (Pannell et al. 2014). Minimum soil disturbance is one of the pillars of conservation agriculture. Its goal is to avoid mechanical soil disturbance in agricultural activity. This is made possible by machinery that has been developed to allow sowing on plant residues, although it is necessary to select the type of machine best suited to the conditions of each farm. In mechanized systems, minimum soil disturbance is associated with reduced production costs, mostly through savings in fuel and tractor use. However, weed management costs can be higher than for conventional practices (D’Emden et al. 2008; Erenstein et al. 2012). The possibility to utilize counter seasonal labor by preparing land in the dry season is another potential way to spread labor PAGE 35 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure 3.1 Spatial distribution of adoption rates for eight CSA practices Minimum soil disturbance Residue retention Minimum Soil Disturbance by districts Retention of Residues by districts 0% - 1.56% 20.59% - 30.43% 1.57% - 4.78% 30.44% - 47.78% 4.79% - 9.92% 47.79% - 56.14% 9.93% - 19.57% 56.15% - 63.69% 19.58% - 33.46% 63.7% - 82.3% Intercroping/crop rotation Commercial horticulture Rotation with legumes by districts Commercial gardens by districts 0.71% - 5.08% 3.6% - 15.25% 5.09% - 9.68% 15.26% - 23.53% 9.69% - 17.05% 23.54% - 30.51% 17.06% - 34.04% 30.52% - 39.44% 34.05% - 47.7% 39.45% - 52.94% Agroforestry Livestock diversification Agroforestry by districts Livestock Diversification by districts 0% - 1.15% 9.33% - 16.78% 1.16% - 3.16% 16.79% - 29.17% 3.17% - 5.36% 29.18% - 34.29% 5.37% - 10.87% 34.3% - 40.52% 10.88% - 24.81% 40.53% - 59.72% Planting on time Drought-tolerant maize Planting on time by districts Drought Tolerant Maize Seeds by districts 36.94% - 51.04% 0% - 7.69% 51.05% - 65.25% 7.7% - 23.4% 65.26% - 75.29% 23.41% - 36.47% 75.3% - 88.06% 36.48% - 50.99% 88.07% - 100% 51% - 67.5% Source: FAO, using RALS data PAGE 36 demand and lower costs. Labor demands associated with minimum soil disturbance may also vary with the duration of adoption. For example, the establishment of planting basins may require more labor hours relative to conventional practices, but once the basins are established, labor demand is less than for conventional practices (Haggblade and Tembo 2003). Weed management costs are typically higher when manual weeding is used in place of herbicides (Erenstein et al. 2012; Wall 2007; Giller et al. 2009). For many smallholder producers, herbicide use is not an option, due to lack of availability, resources, and knowledge. Field trial data from eastern and southern Africa shows a 50 percent reduction in labor demand for tillage with minimum soil disturbance, but a 30 percent increase in weeding demand. In addition to the potential direct costs and benefits of minimum soil disturbance, indirect costs and benefits must also be considered. One such benefit, per Lal (2004), is that the use of minimum tillage reduces carbon emissions relative to conventional tillage from 35.3 kg carbon equivalents (CE)/ha to 7.9 kg CE/ha. A notable indirect cost of minimum soil disturbance is that using herbicides to achieve labor savings raises potential health and environmental costs. Moreover, the potential increased demand for manual weeding places a pronounced burden on women. Promotion strategies for minimum soil disturbance need to consider these indirect costs and identify strategies to mitigate them. 3.2 Residue Retention Approximately 59 percent of smallholder households in Zambia leave their residues in the field, either as a form of mulch or for livestock grazing. The economic costs and benefits of residue retention, and barriers to adopting are highly context specific. Evidence from field trials and simulation models suggest considerable variability in yields (Baudron et al. 2012; Probert 2007). Residue retention has been found to support higher soil moisture levels in the following year, with beneficial effects on yields under dry conditions. However, yields have been found to decline following residue retention under high rainfall conditions (Rusinamhodzi et al. 2011). Residue retention is also found to reduce weed pressure under certain conditions, and thus can contribute to a reduction in weeding costs. Yields tend to decline in the short term and increase in the long term. When residue retention is combined with minimum soil disturbance, yields typically decline relative to conventional practices and then begin to increase, typically exceeding conventional yields after a decade (Pannell et al. 2014; Fowler and Rockstrom 2001; Baudron et al. 2012). Given the high discount rates of smallholder farmers, delayed benefits are likely a major barrier to adoption. Residue retention and management are also constrained by the common practice of communal grazing of crop stubble in many parts of Zambia, and the economic value of residues as animal feed. In areas with a high livestock density, it is unlikely that residues retained will be sufficient to meaningfully affect yields and yield stability (Mazvimavi and Twomlow 2009). High livestock density is more likely to be a challenge in AEZ I and AEZ IIa, where 7.3 percent of households have livestock (excluding poultry) in their farm system, as compared to about 2 percent of households in the other two AEZs. Burning residues is a common practice that provides a number of immediate benefits, including lowering weed pressure, facilitating the hunting of mice, and providing some soil nutrients; burning residues also contributes significantly to Zambia’s carbon footprint. 3.3 Legume Rotation and Intercropping 49. In Zambia, 8.5 percent of smallholder farmers practice legume rotations and 13 percent practice legume intercropping. Growing legumes in rotation or as an intercrop can provide several benefits, PAGE 37 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN such as increased crop yields and a reduction in input costs due to nitrogen fixation, a reduction in crop disease, and other potential advantages (Pannell et al. 2014; Pannell 1995). While legume rotations generally confer yield benefits on cereal crops, the overall benefits to a farm household of incorporating legumes depend on the price and/or yields generated by legumes. Identification of grain legume species with reasonable levels of market development is therefore essential for the overall economic attractiveness of legume rotations and intercroppings. For example, Rao and Mathuva (2000) found that annual grain legume-based cropping systems (based on cowpea and pigeon pea) were 32–49 percent more profitable than growing maize continuously, while rotation with gliricidia was less profitable than continuous maize. In terms of the legume market in Zambia, groundnuts have significant domestic demand due to their importance in the Zambian diet. While there are export and value-addition (peanut butter) opportunities for groundnuts, these are not widely exploited in Zambia. Soybeans are important for the animal-feed and oil industries. The soybean market is growing rapidly, but prices are also falling. Cowpeas are not widely consumed, but pigeon peas are an emerging crop in Zambia but depend on the Indian market. Finally, common beans are widely consumed in Zambia, with production concentrated in the north of the country. The choice of rotation versus intercropping is also important as it entails trade-offs in terms of land allocation and yields at the farm level. For example, legume rotations in the Eastern Province generate higher average maize yields than intercrops for most legume species, but intercrops are preferred by the majority of farmers due to land constraints (Thierfelder et al. 2017). 3.4 Commercial Horticulture In Zambia, approximately 23 percent of smallholders, or 370,000 households, produce fruits and vegetables for sale. Analyses of the smallholder horticulture sector in Zambia suggest considerable opportunities to link smallholder producers to the country’s urban sector, in ways that can produce beneficial opportunities for poverty reduction and economic growth (Tschirley et al. 2014; Hichaambwa et al. 2009). Based on urban consumption data, the average household in Lusaka spends roughly 21 percent of its food budget on horticulture products, second only to its expenditure on cereals (24 percent) (Hichaambwa et al. 2009; Chisanga and Zulu-Mbata 2018). In 2012, the total value of fruits and vegetables sold by smallholders in Zambia was approximately US$250 million and the prospects for continued demand growth for horticulture products remain high. Commercial horticulture can increase incomes for land-constrained farmers, many of which are women and produce gross margins 138–219 percent higher and returns that are 141–263 percent higher than for maize (Hichaambwa et al. 2015). As a result of rapid growth of supermarket outlets, there is a concern that smallholder producers will be excluded from emerging market opportunities for horticulture products in urban Zambia. There is substantial evidence that smallholders often struggle to meet the supply quantity, quality, and consistency requirements of export markets (Dolan and Humphrey 2000). While these concerns are valid, there is evidence that domestic horticulture markets in Africa and Asia, unlike those for meat and dairy, are often resistant to retail consolidation (Tschirley et al. 2014; Weatherspoon and Reardon 2003). Indeed, Tschirley and Hichaambwa (2010) estimate that over 90 percent of horticulture products consumed by urban Zambians are purchased through traditional market retailers. Despite growth opportunities, major constraints exist within the horticulture sector. First, smallholder markets are quite concentrated, with the top 20 percent of sellers accounting for over 80 percent of sales. Moreover, a minority of horticultural producers in Zambia are sellers; the likelihood of PAGE 38 selling rises steadily with landholding size (Chapoto et al. 2012). Second, remoteness is a major factor in determining market participation. Due to perishability and susceptibility to damage, farmers that are far from urban markets are unlikely to sell horticulture products. Third, barriers to entry include the comparatively high financial and human capital investment, and the need to understand supply chains. Small-scale irrigation is needed year-round to produce crops throughout the year. Finally, high levels of price uncertainty and generally chaotic conditions within urban wholesale markets place significant risks on farmers and limit market participation (Hichaambwa et al. 2015). 3.5 Agroforestry In Zambia, four percent of smallholders practice agroforestry. As compared with conventional maize cultivation, agroforestry in Zambia entails higher labor costs during the first and third years. Most frequently planted Agroforestry species include Faidherbia albida, Musekes (Piliostigma thonningii), Mpundu as well as Gliricdia sepium, Sesbania sesban, Tephrosia vogelii, Nyamundalo, Acacia. The literature on agroforestry in Zambia suggests the potential for financial gains from adoption, but raises concerns about discount rates, up-front labor costs, sapling availability, lack of skills to develop a nursery, the need to leave plots fallow until the third year, lack of long-term tenure security under customary land tenure systems, and livestock grazing on customary land. Higher costs in the first and third year are due to the need to establish and cut the trees and incorporate them into the soil. During the other years, labor costs relative to conventional agricultural production are lower, and agroforestry requires less labor overall over a five-year period than maize production systems (Ajayi et al. 2010). In terms of financial profitability, agroforestry is found to produce a discounted NPV of between US$233 and US$309/ha over a five-year period, depending on the species, through positive impacts on crop yields and income generated from agroforestry by-products such as firewood. This compares favorably with an NPV of US$130/ ha for unfertilized continuous maize production. However, continuous maize cultivation without fertilizer subsidies is found to be 13 percent more profitable than agroforestry; maize cultivation with subsidies for fertilizer is 61 percent more profitable (Ajayi et al. 2010). Access to subsidized fertilizer decreases the economic rationale to adopt agroforestry. Targeting the promotion of agroforestry to households that have limited access to fertilizer may offer the best opportunity to promote adoption (Place et al. 2003). Also, strategies to shorten the initial fallow period from two years to one may help improve agroforestry’s financial attractiveness to smallholders (Ajayi et al. 2010). 3.6 Livestock Diversification Zambia’s livestock sector contributes an estimated 42 percent of the total value of national agricultural output. As of 2015, 64 percent of smallholder farmers (primarily in the Southern, East and Central Provinces) raised “village chickens,” 24 percent owned cattle, and 27 percent owned goats. (Bwalya and Kalinda 2014). Given rapid income growth in urban Zambia, demand for livestock products has grown remarkably over the last decade, creating new market opportunities for smallholder producers. Lubungu et al. (2015) estimate that smallholder cattle and goat populations in Zambia doubled to 2.8 million and 2.4 million animals, respectively between 2008 and 2012. Poultry production has increased even more rapidly, with broiler chicken populations increasing threefold to reach 36 million birds between 2001 and 2011 (Bwalya and Kalinda 2014). Livestock ownership confers a range of economic and social advantages in Zambia, from sale income and household consumption to more indirect benefits including as a source of livelihood insurance, animal traction, and manure production (Moll 2005). PAGE 39 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Livestock is negatively affected by climate change. Investing into certain species has the potential to serve as an important tool for livelihood resilience in the context of climate change. Livestock diversification can increase tolerance to drought and heat waves and is effective against climate change-related outbreaks of diseases and pests (Rojas-Downing et al. 2017). High temperatures can affect growth, reproduction performance, production, and animal health and welfare (Arslan et al. 2018). Climate change has already increased the number of recorded animal disease outbreaks in Zambia. For example, six weeks of drought in January 2018 coupled with subsequent irregular, often heavy, precipitation caused the death of many cattle. As temperatures rise, investing in livestock such as donkeys, pigs, rabbits, poultry or sheep and goats rather than ruminants may not only increase resilience but also enable continuous meat and milk production (Arslan et al. 2018). Increasing livestock ownership and realizing the potential associated benefits for households through commercialization presents several challenges. First, animal husbandry capacity is low. Smallholders rarely utilize livestock vaccines or veterinary treatments, instead relying on traditional treatments. The use of supplemental feeding is low in Zambia, which leads to significant seasonal variations in weight and reproductive performance. Second, government support for the livestock sector, for instance in the form of dip tanks, veterinary services, and oversight of processing facilities remains low. Third, sociocultural norms related to livestock ownership, particularly cattle, may limit the potential for commercialization. Cattle are often acquired as part of marriage dowries and are thus the property of extended families and not of individuals, which complicates marketing decision- making. Finally, disease outbreak and a lack of market segmentation for improved meat quality limits investments in formal livestock market development and adoption of improved management practices (Lubungu et al. 2015). 3.7 Timing of Planting In rainfed production systems, the timing of planting is an important determinant of crop yields. In Kenya, Tittonell et al. (2008) find that early planted maize crops achieved yields of 2.1 MT/ha, nearly double the 1.2 MT/ha yield of late planted crops. Further, each day the planting date of seed cotton was delayed or the length of the rainy season was reduced, was associated with an average reduction of 16 kg/ha in attainable yields (Tittonell and Giller 2013). The timing of planting is a function of the length of the rainy season and temporal distribution of rainfall as well as of crop and seed variety choice. In Zambia, maize planted after December 15 is typically considered delayed. This date serves as a rule of thumb; appropriate planting dates vary between regions and the maturation length of the seed used. Delayed planting in smallholder systems is caused by a range of factors. For asset-poor households, availability of labor and/or animal traction is an important contributing factor to delays in planting. In many cases, poor households perform day labor, or ganyu as it is called in Zambia, during the planting season in order to meet household food security needs. This diverts scarce labor away from production on households’ own fields, and can contribute to a vicious cycle of piecework, food insecurity, and low productivity. Traditional seed varieties, which are commonly adopted by resource-poor households, typically have longer maturation lengths than those hybrid varieties. In areas where growing seasons are short, there is limited window for timely planting of these varieties. Market isolation may also be an important driver for delayed planting, as it limits the availability of improved seed varieties. Finally, under Zambia’s input subsidy program FISP, the delivery of inputs to farmers is often late. Namonje et al. (2015) estimate that in 2012, 21.5 percent of subsidy recipients received inputs late. Late delivery of inputs contributed to a 4.2 percent reduction in maize yields, or roughly 84,000 MT at a national level. PAGE 40 3.8 Drought-Tolerant Maize Drought stress is a common challenge in Sub-Saharan Africa. It is estimated that roughly 40 percent of Africa’s maize-growing area experiences occasional drought stress, with yield losses of 10– 25 percent. An additional 25 percent of the maize crop in Africa suffers frequent drought, with losses of as much as half the harvest (Cairns et al. 2013). Investment in the development and promotion of drought-tolerant seed varieties that are bred to sustain yields under dry conditions can contribute to significant benefits for producers and consumers in the region. Breeding of drought-tolerant maize varieties began in earnest in Zambia in 2007 as a result of the Drought Tolerant Maize for Africa project, led by CIMMYT and conducted in collaboration with Zambia Agricultural Research Institute (ZARI). Through this project, 160 varieties of drought- tolerant maize have been released in 13 countries. In Zambia, an estimated 23 percent of farmers utilize drought-tolerant maize varieties (Fisher et al. 2015). La Rovere et al. (2010) estimate that adoption of drought-tolerant seed varieties can improve average maize yields in Zambia by 17.3 percent and reduce the variance in yields associated with dry conditions by 10–15 percent. A full replacement of improved maize seeds with drought-tolerant maize seeds in Zambia would lead to total economic benefits of US$115 million, of which US$74 million come from yield growth and US$41 million from reduced yield variance. This would pull an estimated 360,000 Zambian households out of poverty. Key barriers to adoption of drought-tolerant maize varieties include: low availability of improved seed, inadequate information, lack of financial resources, high seed cost, and negative perception about the taste and storability of different varieties (Fisher et al. 2015). Stimulating greater private investment in the multiplication and delivery of drought-tolerant seed, as well as developing marketing arrangements to reach poor segments of the rural population, such as selling 1–2 kg micro packets of drought-tolerant seeds, are critical for addressing adoption barriers. 3.9 Crop Diversification Zambia exhibits one of the lowest levels of crop diversification in Africa as measured by the Simpson’s Diversity Index. About 48 percent of smallholder farmers cultivate three or more crops while 14 percent grow maize in monocrop (Maggio et al. 2018). As shown in Figure 3.1, provinces in the north and northeast have the highest crop diversification while south and central regions of Zambia are the least diversified. Livestock diversification is higher in provinces in the east and south of the country than other areas (Arslan et al. 2015). A descriptive analysis of the RALS 2012 and 2015 shows that 15–41 percent of households have low levels of livelihood and agricultural diversification, respectively. The impact of crop diversification on household welfare in Zambia is highest for low income households and lowest for wealthy households. However, wealthier households are the most likely to adopt more diverse crop production systems. The rural poor face significant adoption barriers to diversification (Ignaciuk et al. 2018). Given land constraints, households often prioritize the production of staple foods on their available land. The rural poor often face greater constraints in access to input and output markets, agricultural credit, and extension services, which limit their capacity to diversify. Crop diversification in Zambia is associated with a reduction in crop income volatility and, in many cases, with increased maize productivity (Maggio et al. 2018). The key drivers of crop diversification include access to more competitive private output markets, increasing land size, and exposure to adverse weather conditions. Proximity to the FRA, significantly reduces the probability of crop diversification (Ignaciuk et al. 2018; World Bank 2018d). PAGE 41 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN 3.10 Agricultural Liming Zambian soils show high levels of acidity, which can be addressed by agricultural liming. Burke et al. (2017) estimate that a soil pH of 5.4 marks a threshold below which yield responses to basal fertilizer drop off substantially. For Zambian soils an average level of acidity of soil pH 5.38 is assessed, which hinder the retention of potassium, calcium and magnesium, and other essential elements and limit the availability of phosphorous derived from both the soil and fertilizers. Soil microbial activities and root growth are negatively influenced, and crop yield is decreased. The authors estimate that 25–45 percent of basal fertilizer users in Zambia operate at a fiscal loss. Put differently, Kelly (2005) suggests that smallholder farmers, who plan to adopt a new technology but are risk averse, need an average value-cost ratio of 2 to adopt a new technology. At this threshold, virtually no farmer in Zambia breaks even using basal fertilizer. Raising soil pH with agricultural lime in Zambia yields potentially great economic benefit. However, the quantities required to have a meaningful effect on pH levels are substantial. To move soil from a pH of 5, which marks the 25th percentile of smallholder soils in Zambia, to a pH of 5.2 (marginally below the yield response threshold estimated by Burke et al. (2017)) would require 2–3 MT/ha of lime, and approximately 4 MT/ha to reach a pH of 6 (Sims 1996). Table 3.1 Comparative descriptive statistics on lime adopters and non-adopters Non-adopters Adopters Mean Mean Difference T-Test % of households that are female headed 0.258 0.206 0.053 Avg. years of education completed by the head of household 5.704 7.103 -1.399 ** Number of household members in adult equivalent 5.141 5.798 -0.656 Avg. agricultural wealth index (0 to 1) 0.078 0.173 -0.095 *** % of households with less than 5 hectares cultivated 0.783 0.655 0.128 % of households applying inorganic fertilizers 0.615 1.000 -0.385 *** % of households participating in Zambian FISP 0.371 0.423 -0.052 % of households that are members of a coop, farmer/women/ 0.507 0.577 -0.071 savings-loan group % of households that received information on conservation 0.378 0.497 -0.119 ** agriculture Commercialization index (0 to 1) 0.302 0.431 -0.129 ** % residing in AEZ I 0.085 0.081 0.005 % residing in AEZ IIa 0.367 0.448 -0.081 % residing in AEZ IIb 0.072 0.000 0.072 *** % residing in AEZ III 0.476 0.471 0.004 Source: RALS 2015. Note: Significant levels of T-test are * p<0.10, ** p<0.05, *** p<0.01. Unfortunately, adoption rates for agriculture lime among Zambian smallholders is too low to conduct a meaningful econometric analysis of its impacts. The descriptive details on lime adopters relative to non-adopters are therefore presented here in, with T-tests to assess the statistical significance of the differences between adopters and non-adopters. In 2014-2015 only 53 of the roughly 6,000 households in the RALS used lime on their fields. As shown in Table 3.1, there are a number of significant differences between lime users and non-lime users in Zambia’s smallholder sector. The heads of households that use lime have on average 1.4 more years of education. These households own significantly more agricultural assets, as measured through an agricultural asset index. They are also significantly more likely to use inorganic fertilizer and to receive information on conservation agriculture. Indeed, 100% of lime users also use inorganic fertilizer. On average lime PAGE 42 users commercialize a greater share of their agricultural output than non-users. As shown by the commercialization index, the average lime user sells 43% of the value of all agricultural products they produce, compared to 30% among non-users. Finally, there are no lime users included in the survey that reside in the semi-arid AEZ IIb. This is unsurprising, given the relative market isolation in that region and the region’s low potential for rain-fed crop production. 3.11 Post-Harvest Loss Following the food price crisis in 2007/2008, there is a renewed interest in reducing post-harvest losses. The World Bank (2011a), estimated that around US$4 billion annually can be saved in Sub- Saharan Africa by reducing the post-harvest loss in grains (mainly cereals and legumes). This is more than the value of total food aid received in the region over the last decade. There are four major reasons to address post-harvest loss (Sheahan and Barrett 2017): (1) improve food security; (2) improve food safety; (3) reduce unnecessary input use and (4) increase profits for food value actors. Food loss and waste can take place at all stages of the value chain between the farmers’ field and the consumers’ fork, which can be divided into five stages (FAO 2011): (1) harvesting due to mechanical damage and/or spillage, (2) post-harvest loss, which includes drying, winnowing, and storage (insect pests, rodents, rotting), (3) processing, (4) distribution and marketing, and (5) consumption. Most data on post-harvest loss is spotty and often of poor quality. None of the available sources provide data for Zambia. FAO (2011) estimated that total post-harvest loss in Sub-Saharan Africa amounts to one third of total food produced (in volume). A more comprehensive analysis is presented by Affognon et al. (2015), who conducted a literature survey of post-harvest loss in six African countries (but not including Zambia) and seven commodity categories. On average, post- harvest loss for groundnuts ranges up to 35 percent, for maize 25 percent and for mango even up to 55 percent. Kaminski and Christiaensen (2014) use nationally representative household surveys for Uganda, Tanzania and Malawi and show that on-farm post-harvest loss for maize is in the range of 1.4–5.9 percent. There is a consensus that most grains and cereals are lost during post-harvest handling and storage on-farm, while loss of fresh produce, meat, and seafood is concentrated in processing, packaging, and distribution (Sheahan and Barrett, 2017). In Sub-Saharan Africa, most post-harvest loss happens at the farm level (FAO 2011; Affognon et al. 2015). PAGE 43 Section 4 Analysis Results: Observed Impacts of CSA on Household Welfare Section summary The household level econometric and financial assessments of the impacts of CSA adoption corroborate findings from the literature (see Section 3), as follows: • The long-term, positive welfare effects, such as increased gross income and income from crops, increased food availability, and reduced levels of poverty can be confirmed for most CSA practices. • In the short term, high production and labor costs hamper the adoption of CSA practices, specifically minimum soil disturbance, residue retention, and agroforestry. These practices, as well as crop rotation, sequester high levels of carbon in the soil and thus offer significant climate mitigation potential (Section 5.7). • The performance of CSA practices varies across agroecological zones and shows varying degrees of climate sensitivity, as measured by rainfall variability. Minimum soil disturbance, residue retention, and small-scale horticulture are not well suited to very wet conditions due to waterlogging and weed pressure, or because lack of drainage. Drought- and heat-tolerant seeds, agroforestry, and crop diversification show good results under both extreme dry and wet conditions, and seem particularly suited to climate adaptation and resilience building practices. • At the household level, adoption of CSA practices can exert a positive impact on rural development in the form of market and job creation. The need for adequate mechanization or timely delivery of inputs creates opportunities for developing markets; also legumes, livestock, and horticulture have high value addition potential. All CSA practices—particularly those with varying climate sensitivity— benefit from targeted agricultural advisory services that can mitigate the risks of adoption. This leads to three recommendations to support CSA adoption: 1. Provide access to finance and/or subsidize farmers for high upfront cost. 2. Support development and promotion of practices customized to agroecological conditions. 3. Support market development for tools and services such as inputs (for example, seeds) and advisory services (for instance, customized, ICT-based agronomic advice to enhance timing of planting) that support CSA adoption. PAGE 45 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN This section is organized according to CSA technologies and shares findings from two different analytical approaches (both econometric and cost-benefit analyses). Owing to data constraints, CSA practices are analyzed discretely—although some farmers adopt practices in combination. Detailed results are presented in Appendix C.3. 4.1 Findings: Analysis of Minimum Soil Disturbance The primary adoption barriers that farmers face with minimum soil disturbance (MSD) is an increase in direct costs, particularly labor (see Table 4.1). Addressing direct costs is essential for achieving widespread adoption of minimum soil disturbance. Given the food security and income risks for households associated with changes in farm practice, low income may discourage autonomous adoption and retention of CSA practices over time. Table 4.1 Main findings of the econometric analysis: Minimum soil disturbance Analysis Main findings Interpretation DiD analysis • Adoption increases the probability of • Poverty effect stems from short-term household poverty perspective of the DiD approach. Thereby, high upfront cost play a large role OLS model • Correlated positively with crop income • This cannot be interpreted as causality and gross household income; there is but as long-run positive correlation, a negative relationship with income suggesting that adopting MSD for several variability and poverty years is correlated with a reduction in income variability and poverty. Climate sensitivity • Increases gross-income variability • Coefficients are too low to allow a meaningful interpretation • In areas of heavy rainfall MSD adopters experience 23% reduction in crop income • Trend likely due to high weed pressure relative to conventional tillage farmers and/or water logging of planting basins under MSD Crop yields and • Crop yields under MSD are higher than • Crop yields under MSD tend to be higher production cost under conventional practices in areas with on average than conventional practices in low or medium rainfall (AEZ I, IIa, and IIb). dry areas Maize yields under conventional practices • These findings appear to be mostly due in AEZ III are much higher than under to increased labor costs and the upfront MSD costs of adoption, which combined with • Disparity in crop yields by MSD- the high discount rates and risk aversion combination: With legume rotation, yield of most poor smallholder households, is 2.8 MT/ha for maize in AEZII; planting limits adoption among this socioeconomic basins achieves yield of 2.1 MT/ha. segment • MSD requires more labor than conventional tillage due to land clearing and planting basins Financial viability • Financial returns per hectare of MSD are • Low returns deter autonomous adoption for households: Net mostly positive but small of CSA practices incremental benefits • Highest net incremental benefits across • MSD is best suited to rather dry AEZs (I, per hectare AEZs are evident for rice (US$163/ha), IIa, IIb) and may produce low or negative followed by soybeans, groundnuts and results in AEZ III. Cassava, an exception, cowpeas (US$97/ha, US$79/ha, US$65/ha); shows highest results in AEZ III lowest for beans (US$ -3.2/ha). • However, while net incremental benefits • In contrast with drier AEZs, net incremental are higher for some other crops, benefits benefits in AEZ III are negative for maize associated with maize production will likely and beans (US$ -79/ha and US$ -24/ha) drive decisions at the household level but highest for cassava (US$74/ha). • For maize, highest benefits are found in AEZs I, IIa and II, where average incremental benefits are US$30/ha. Source: FAO, using RALS 2012 and 2015 data Note: Summary of statistically significant results of econometric analyses and financial analyses; see Appendix C.3 and D.3 for detailed results. PAGE 46 Household level cost-benefit ratios vary significantly across agroecological zones (AEZs) and crops. In high rainfall regions of Zambia, such as AEZ III, minimum soil disturbance should be adopted cautiously as net benefits are small or even negative. For instance, results for maize are mixed: in AEZ II minimum soil disturbance yields positive but relatively low incremental net benefits but could cause financial losses in AEZ III. Given the uncertainty of results, poor households may find it difficult to sustain adoption of MSD and will need significant support to cover upfront cost. Adjusting practices to varied biophysical conditions would be crucial (see Appendix D for detailed results of the financial analysis). Widespread adoption of MSD would likely support the creation of new markets for implements and inputs, including direct seeders, chisels and/or rippers, chaka hoes, and herbicides, and would bolster rural development. The need for advisory services would probably create new jobs as well. 4.2 Findings: Analysis of Residue Retention Residue retention reduces income variability and household food insecurity in the long term, particularly in dry regions. In the short term, retaining crop residues does not have a significant effect on most welfare indicators relative to conventional practices, with the exception of income variability, which it seems to reduce. In the long term, residue retention can be expected to increase income, particularly in dry regions. While input costs can be reduced, higher labor costs present an issue. Residue retention offers higher food security benefits in dry regions such as AEZ I; in wetter areas, the practice had a negative impact on yields and incomes, which suggests that appropriate residue retention strategies should be identified before promoting the practice in wetter areas such as AEZ III (see Table 4.2). Table 4.2 Main findings of econometric analyses: Residue retention Analysis Main findings Interpretation DiD analysis • Reduction of 17.4% in income variability • Residue retention has the potential to enhance resilience by decreasing variability and potential losses OLS model • In the long term, residue retention is • Lack of impact on other welfare factors correlated with an increase in crop income, might be due to low production and but no other welfare indicators livestock pressure Climate sensitivity • Household food insecurity reduced by • Food security benefits are positive in dry, 23.6% under low rainfall scenario but negative in wet areas • In high rainfall areas, residue retention • In dry areas, residue retention can be is associated with a 17.4% reduction in expected to have a positive impact on soil average income and an increase the moisture, likely linked to improved water probability of poverty retention capacity and reduced crop losses. Net incremental • For maize systems in AEZ IIa residue • Residue retention works well in benefits retention combined with MSD led to a combination with MSD and has a positive 30% and 11% increase in yield, and 182% effect on crop yields and 16% increases in net margin per ha, • While net margins are expected to increase, compared with conventional and MSD- this may be dampened by a simultaneous only systems increase in labor costs • Compared to MSD-only, fertilizer costs can be expected to decline, and labor costs to increase by 7% and 33%, respectively Source: FAO, using RALS 2012 and 2015 data Note: Summary of statistically significant results of econometric analyses and financial analyses; detailed results are in Appendix C.3 and D. Adoption of a well-managed residue management system may be constrained by two factors: the high costs associated with fencing fields in the dry season, as well as insecure land tenure. Grazing access on customary land may be a limiting factor for smallholder farmers. Where open grazing is the norm, the ability to fence fields is essential for managing residues. This entail high costs and requires PAGE 47 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN more secure land tenure. In addition, residue retention offers negligible potential for job creation and value-chain development and thus is not an effective agent for broad-based rural development. 4.3 Findings: Analysis of Legume Rotation and Intercropping Legume rotation and/or intercropping does not show a significant short term effect, but a positive correlation with crop income in the long term. Low crop yields are a concern. Adoption of drought-tolerant legumes seems to reduce food insecurity under dry conditions and enable higher crop productivity and two production cycles in wetter regions (see Table 4.3). Average legume yields are 40 percent lower in Zambia, compared to countries in the region. Addressing low yields in Zambia is likely a key priority for enhancing adoption of legumes in smallholder systems and improving the welfare impact. This may require a combination of investments in improved seed varieties, particularly promiscuous varieties, management practices for key legume varieties and/or support for legume market development. Growth in the legume sector holds significant potential in terms of job creation and value chain development, including: investment opportunities in legume processing and export; and job opportunities in legume seed multiplication, input supply, and off-taking. This could contribute to a reduction of protein deficiency in the Zambian diet and to improved nutrition outcomes in the country. Table 4.3 Main findings of econometric analyses: Legume rotation and/or intercropping Analysis Main findings Interpretation DiD analysis • No significant short-term impacts on • One explanation might be low yields and welfare indicators low farm gate prices OLS model • Significant long-term correlation between • Average yields of groundnuts, beans and adoption and crop income cowpeas were 41% higher in other African countries than in Zambia.(a) This points • Limited correlation with other household towards the need to invest in improved welfare indicators varieties. Climate sensitivity • Household food insecurity reduced by • Potential to reduce food insecurity may 10.4% in low rainfall scenario be due to adoption of drought-tolerant legumes • Crop income variability reduced by 15% in high rainfall scenario • Reductions in income variability potentially due to higher productivity levels of legumes under wet conditions Impact on alternative • Significant potential for growth in the • Investments in legume processing, exports, development overall legume sector and legume seed multiplications should be indicators • Widespread protein deficiency in the explored Zambian diet could be addressed through increased legume production Source: FAO, using RALS 2012 and 2015 data Note: Summary of statistically significant results of econometric analyses; see Appendix C.3 for detailed results. (a) Comparing RALS 2015 and FAOSTAT 2018 data. 4.4 Findings: Analysis of Small-scale Commercial Horticulture Adoption of small-scale horticulture provides a significant short- and long-term boost to gross income, and reduces gross income variability and poverty. The benefits, by comparison with non- adoption, are especially pronounced in dry climates. Small-scale horticulture also promises potential job creation and value chain benefits, although these are difficult to quantify. In addition, enhanced horticulture market participation in Zambia is likely to have a beneficial nutritive effect through direct consumption on-farm or by making horticulture products more widely available and affordable in local markets for urban and rural consumers (see Table 4.4). PAGE 48 Table 4.4 Main findings of econometric analyses: Small-scale commercial horticulture Analysis Main findings Interpretation DiD analysis • Adopters saw crop income and gross • Despite high levels of market-price income increase by 9.7% and 6.7% variability, horticulture products are typically short-duration crops that can be • 4.8% reduction in gross income variability produced throughout the year with small- and 1.7% decreased likelihood to fall below scale irrigation poverty line • This allows households to smoothen their OLS model • Results aligned with those from DiD model income. Climate sensitivity • Adoption is highly beneficial in a low rainfall • A short production duration enables scenario, with households experiencing: producers to better match production 21% increase in crop income; 25% increase cycles with rainfall cycles in gross income; 19.1% reduction in the • Feasible placement of horticulture fields probability to be poor; 16% reduction in near perennial water sources the probability of food insecurity under dry conditions Impact on alternative • Broadly positive, especially in combination • High income increases, especially for development with investments in value-adding process small farmers, if barriers to entry can be indicators such as canning and juicing overcome Source: FAO, using RALS 2012 and 2015 data Note: Summary of statistically significant results of econometric analyses; see Appendix C.3 for detailed results. 4.5 Findings: Analysis of Agroforestry In the longer term, adoption of agroforestry is correlated with beneficial impacts on resilience and risk. Agroforestry reduces by 25 percent the probability of food insecurity in both lower and higher rainfall conditions, making it a suitable CSA technology for scenarios in which climate change is significant. Upfront investment, forgone revenue, and delayed returns make agroforestry not well suited to adoption by smallholders, particularly at high cost of capital (discount rates), unless smallholders are directly and financially compensated, for example, via conditional cash transfers. With respect to rural development indicators, there is only small potential for job creation because agroforestry relies primarily on household labor; however, there is potential also for development of privately-owned nurseries (see Table 4.5). Table 4.5 Main findings of econometric analyses: Agroforestry Analysis Main findings Interpretation DiD analysis • No short-term causal effect • Multiple years are required to establish beneficial agroforestry systems OLS model • Correlates closely with reductions in crop- • Findings are not causal, but do suggest a and gross-income variability positive association Climate sensitivity • Adoption reduces food insecurity in both • One explanation is that agroforestry leads low and high rainfall scenarios by as much to improved water retention and water as 25% infiltration Impact on rural • Neutral to modestly positive results in term • High income increases, especially for development of key development indicators small farmers, if barriers to entry can be indicators overcome Source: FAO, using RALS 2012 and 2015 data Note: Summary of statistically significant results of econometric analyses; see Appendix C.3 for detailed results. 4.6 Findings: Analysis of Livestock Diversification At the household level, livestock diversification contributes to lower levels of poverty and higher levels of food security and resilience, and these impacts may be even pronounced under climate change (see Table 4.6). Livestock ensure greater resilience in droughts and provide financial PAGE 49 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN compensation for crop income lost to heavy rainfall. Diversification typically occurs in one of two ways. Households integrate livestock into their crop systems to diversify their agriculture production systems; alternatively, livestock owners diversify their livestock base to spread their production risk or exploit different market opportunities. Livestock boosts rural development by providing opportunities for job creation, including veterinary services, trading, livestock processing and cross-border trade. Livestock products are also amenable to a wide range of processing- and marketing-related value addition options. Improved access to livestock proteins would broadly benefit nutritional outcomes in Zambia. The high ownership and management costs associated with rearing livestock present an obstacle to adoption by the very poor. Table 4.6 Main findings of econometric analyses: Livestock diversification Analysis Main findings Interpretation DiD analysis • Integration of livestock into crop systems • Poverty reduction is likely achieved through reduces the probability of poverty a reduction in costs, including for land incidence but exerts no impact on gross preparation, and the use of animal manure income in place of fertilizers • No significant impact on agricultural • Lack of commercialization in the sector income, household income, and income limits the capacity of smallholders to turn variability livestock assets into cash OLS model • In the short and long term, diversification • Diversification into new species may delay into new livestock species has very small, positive impacts on welfare indicators due positive impacts on income variability and to the lack of or more volatile markets, or food insecurity higher disease pressure. Climate sensitivity • In dry conditions, the integration of • In the dry season, livestock owners benefit livestock into crop systems, shows a small from the relative drought resilience of positive effect on agricultural incomes and livestock the probability of poverty • Livestock ownership smoothens the • Under high rainfall conditions, livestock adverse effect of heavy rainfall on crop ownership is associated with a reduction income in both agricultural and gross household • During drought, livestock may provide income variability additional or alternative food sources, but • The diversification into new livestock liquidation of livestock holdings may result species under low rainfall conditions also in greater income variability is associated with a reduction in food insecurity, but also an increase in gross income variability Impact on rural • Multiple positive effects including income • Job opportunities include veterinary development stability, increased household income, job services, trading, livestock processing, and indicators creation, and value chain development cross-border trade opportunities Source: FAO, using RALS 2012 and 2015 data Note: Summary of statistically significant results of econometric analyses; see Appendix C.3 for detailed results. 4.7 Findings: Analysis of Timing of Planting Accurate timing of planting is especially important in dry conditions, such as those of Zambia’s AEZs I and IIb, where growing seasons are particularly short (see Table 4.7). Enabling farmers to plant in a timely manner, either by supporting access to seeds of a duration appropriate to specific agroecological circumstances or by addressing labor and/or mechanization constraints that delay planting, would have significant positive effects, particularly on crop and household income and income variability. Poorer households face the greatest challenge to access the seeds, labor, and equipment needed to plant in a timely fashion. Developing strategies to improve the timing of planting will likely produce moderate positive benefits to the poor. PAGE 50 Table 4.7 Main findings of econometric analyses: Timing of planting Analysis Main findings Interpretation DiD analysis • No significant short-term effects on welfare • Delayed planting did not happen in first indicators wave of the survey OLS model • In the long term, households that plant late • Delayed planting, which can occur in have on average 8.7% lower crop incomes, response to weather conditions, can hurt 46.3% greater crop income variability, and households’ welfare. Supporting farmers 8.6% lower gross household incomes than to be more timely in planting can have households that plant in a timely manner important beneficial effects Climate sensitivity • Negative effects of delayed planting • Proper timing for planting is especially are exacerbated under adverse weather important in climate extremes, which conditions, especially dry conditions points toward the need to support farmers with information on planting times and timely access to inputs Impact on rural • The value-chain and job-creation impacts • Positive outcomes are mainly achieved development of this strategy are modest through increase in production indicators Source: FAO, using RALS 2012 and 2015 data Note: Summary of results of statistically significant econometric analyses; see Appendix C.3 for detailed results 4.8 Findings: Analysis of Adoption of Drought- and Heat-tolerant Seeds Despite evidence that improved seeds offer robust and consistent benefits across welfare indicators, fewer than half of all seeds used in Zambia qualify as drought- or heat-tolerant (see Table 4.8). Since poor households typically face the greatest hurdles in accessing improved seeds, targeted interventions are needed. Table 4.8 Main findings of econometric analyses: drought- and heat-tolerant seeds Analysis Main findings Interpretation DiD analysis • No significant impact on indicators N/A OLS model • Robust and consistent results across all • These results suggest a strong positive indicators: 34% increase in crop income, benefit to supporting improved access to 18% increase in gross income, 36% drought tolerant varieties to smallholders reduction in crop income variability, and in Zambia 14.6% reduction in gross income variability (similar for heat-tolerant) • Reduces probability of poverty and food insecurity Climate sensitivity • As expected, significant positive impact • Drought- and heat-tolerant seeds also do under dry conditions; surprisingly also better than non-drought tolerant seed under wet conditions under wet conditions Source: FAO, using RALS 2012 and 2015 data Note: Summary of results of statistical analyses; detailed results are in Appendix C.3. 4.9 Findings: Analysis of Crop Diversification Crop diversification promises higher incomes and resilience under both dry and wet climate conditions. Crop diversification—as measured by the Gini-Simpson’s Index, which takes into consideration the number of different crops in a crop system as well as the share of total area dedicated to different crops—can be expected to yield strong benefits in terms of both income and— in the long term—also resilience (see Table 4.9). Crop diversification is found to have strong positive benefits under both dry and wet conditions, which suggests it is a broadly beneficial strategy across Zambia’s diverse agroecological conditions. PAGE 51 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Crop diversification is thought to have beneficial effects on nutrition. A key outcome is the diversification of on-farm diets and products available in local markets. Poor households can adopt diversification strategies if supported to access inputs, specifically seeds. In addition, new employment and investment opportunities are expected, including in assembly, processing, retailing, and export. Table 4.9 Main findings of econometric analyses: Crop diversification Analysis Main findings Interpretation DiD analysis • Positive impact on crop income and • One possibility is that farmers diversify out household income but not on income of low-value staples, such as maize, into variability higher value crops, which exerts a positive effect on incomes OLS model • Strong, positive effect on crop and • In the long term, reducing income household income, and reduced income variability also boosts resilience variability Climate sensitivity • Strong positive benefits under both dry • Helps to improve household resilience to and wet conditions weather shocks, across Zambia’s diverse agroecological conditions Impact on rural • Positive effects on job creation and value • New job and investment opportunities development chain development are created in crop assembly, processing, indicators • Expected to boost nutrition by contributing retailing, and export sectors to a more diverse diet Source: FAO, using RALS 2012 and 2015 data Note: Summary of statistically significant results of econometric analyses; see Appendix C.3 for detailed results. PAGE 52 Section 5 Analysis Results: Impacts of CSA on Agriculture Sector Performance Under Climate Change Section summary This section addresses the feasibility, with or without widespread CSA adoption, of achieving the vision and related targets for Zambia’s agriculture sector by 2050 under two scenarios: the absence of climate change, that is, business as usual (BAU); and under climate change projections. Key findings include: • Under climate change projections, crop yields of several crops are projected to decline. Even without climate change, the agriculture sector target of doubling crop yields in 2050 is not achieved using conventional farming practices. CSA practices can narrow the gap to the target yield but not enough to meet the goal. Under climate change the impact of CSA on crop yields varies: for instance, drought- tolerant maize, conservation agriculture or crop diversification can enhance an already positive yield effect; other CSA practices might cause crop yields to decrease compared to conventional practices. • Crop production is expected to double in 2050 compared to 2009-2011 average, even under climate change projections. CSA has the potential to increase total crop production by on average 3 percent, in the case of residue retention, up to as much as 13 percent with strategies to reduced post-harvest loss. Ruminant livestock production is expected to increase by as much as 229 percent due to improved feeding efficiency, rather than an expanded livestock population. • Increases in agricultural production are reflected in food availability. Under BAU, the lower target of the calorie availability range is within reach. CSA practices such as post-harvest loss and conservation agriculture increase food security further, even under climate change projections. • For maize and millet (and cassava under extreme climate change projections), the target of doubling net exports by 2050 is feasible. Reducing post-harvest losses can enhance net exports of cassava, millet and maize; promoting crop diversification could allow Zambia to become a net exporter of groundnuts. • Increased agricultural production tracks closely with yield increases but may also be a result of land conversion from forest land and other natural land to cropland (approximately 917,000 ha) and grassland (in total 2.2 million ha), just short of the target to convert no more than 0.9 million ha into cropland. Results under climate change projections are similar. CSA practices don’t have a significant impact on land conversion, and at a maximum spare 113,000 ha of forest land from clearance. Land use dynamics result in land conversion to grassland in a similar extent. (continued) PAGE 53 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN • Greenhouse gas emissions are projected to triple by 2050 and increase further under climate change projections. Zambia risks failing to meet its goals under NDC. CSA offers limited scope to reduce land conversion and only limited potential, even in the case of adoption of minimum soil disturbance and post-harvest loss, to reduce GHG emissions arising from fertilizer use. However, if soil carbon sequestration is taken into consideration, CSA’s positive climate mitigation impacts could be many times higher (see Box 5.1). • While the impact of CSA on land conversion is small in terms of hectarage, the value of GHG mitigation to society is substantial and, if valued at a shadow price of carbon, could enhance the economic net present value of an investment by between 171 and 341 percent. • The introduction of a carbon tax on emissions produced from land use change has the potential to halt deforestation for agricultural production, because land conversion is insufficiently profitable. Such a tax could reduce emissions from land use change by as much as 99 percent. Cropland would likely continue to expand but to a lesser extent, limiting the negative effects on production output and calorie availability, in the range of a 2 percent decrease compared to case without a carbon tax. This section concludes with three recommendations in support of adopting CSA practices 1. Investment in productivity-enhancing technologies will be necessary to further close the yield gap. Investments should strengthen agricultural research and development and dissemination of findings; support the optimal application of production inputs; introduce soil testing; facilitate access to improved seeds through rollout of local seed multiplication systems. 2. To moderate the pace of land use change, use a holistic approach to CSA that factors in externalities beyond the field level related to deforestation and land use change to avoid cropland expansion. 3. Coordinate and harmonize policy across sectors including livestock, forestry, water, and energy which is conducive to enhance impact. This section presents results of key agriculture sector indicators for a number of CSA practices in both a future without climate change (business as usual) and for climate change projections. Indicators include: crop yield, crop production, livestock production, land use change, food availability, agricultural trade, and GHG emissions. The section examines the impact of CSA strategies and their feasibility in achieving the normative vision for Zambia’s agriculture sector that is set out in Section 2. Key indicators are presented for both a BAU scenario without climate change and for climate change scenarios. Indicators are provided also in graphic form (see Figures 5.1-5.17). Error bars indicate the range of results of climate change projections. Also see Appendix E.1 for assumptions on how CSA impacts are modelled, and for projected adoption rates of CSA practices in 2050. Smallholder adoption rates in 2050 were determined through stakeholder consultations and a literature review, and range from 30 percent for conservation agriculture to 70 percent for adoption of improved seeds. CSA strategies under assessment are: agroforestry (af), conservation agriculture (ca), residue retention (rr), minimum soil disturbance (msd), drought-tolerant maize (dtm), reduction of post-harvest losses (phl), crop diversification (div). See Appendix E.2 for detailed results. 5.1 Crop Yield While climate change is expected to contribute to negative yield shocks for several crops, crop yields of a range of crops are also projected to increase. The impact of climate change on crop yields are assessed with two crop models and 5 General Circulation Models (GCM). Figure 5.1 presents simulations of average change in crop yield for a selection of subsistence crops for the 2010-2050 period. Climate change is expected to contribute to a negative yield shock (represented by a median value lower than zero) for six crops: groundnuts, maize, millet, potatoes, sorghum and sweet potatoes, whereas the yield effect is positive for barley, dry beans, cassava, rice and soybeans. Figure 5.2 shows the geographical distribution of the yield shock for Zambia’s four most important crops. Maize is mostly produced in the Eastern Province, where yield loss is expected to be between PAGE 54 Figure 5.1 Changes in crop yields of several crops due to climate change between 2010-2050, in % 75 50 25 Yield growth (%) 0 -25 -50 -75 s t e m s s y ze ns va s ille an rle ut oe oe ic hu sa ai ea R dn M be Ba at at M rg as yb un ot t Po So ry C So tp ro D ee G Sw Sources: IIASA; Simulations presenting the average of 2 crop models and 5 General Circulation Models, which model climate change trajectories, using area harvested in 2000 as weights. Data are for subsistence farming systems. Figure 5.2 Changes in crop yields of key crops due to climate change across Zambia, between 2010-2050, in % Cassava Groundnuts yield loss 5−25% Maize Soybeans yield change within 5% yield gain 5−25% yield gain >25% Sources: IIASA; Crop models simulations presenting the average of 2 crop models and 5 GCMs, using area harvested in 2000 as weights. Data are for subsistence farming systems. No data is available for the white areas. PAGE 55 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN 5 percent and 25 percent. In the Northern Province, where maize production is limited by the high acidity of the soils, climate change offers potential yield gains. For the other three crops, potential yield change is more evenly distributed over the country. Groundnut yields are projected to decrease in all areas of the country). The goal of doubling yields by 2050 seems infeasible even in a scenario without climate change. In such a BAU scenario, crop yields are projected to increase between 14-88 percent. In a BAU scenario, maize yields are expected to increase by 88 percent between 2010 and 2050, reaching 3.9 MT/ha in 2050. This is comparable to findings from Kanyanga et al. (2013), projecting maize yields in the range of 3 MT/ha in 2050. Yields of cassava are projected to increase by 33 percent, of groundnuts by almost 35 percent, and millet by 15 percent compared to yields in 2010. Under climate change, cassava yields are likelier to increase than decrease, while the reverse is true for the other three crops (see Figure 5.3). In the most extreme climate change scenario, maize yields are projected to be 15 percent lower, at 3.3 MT/ha than under BAU. The target of doubling maize productivity from average 2009 and 2011 levels, as set out in the agriculture sector vision, is not achieved. In a scenario without climate change, the gap towards the target yield is smallest for maize, followed by cassava. Under climate change, cassava could potentially achieve the goals set out in the vision. CSA practices have the potential to narrow the gap towards target yields and toward achieving the agriculture sector vision, but cannot close the gap. For maize, strategies to reduce post-harvest loss offer the potential to boost production per hectare by 23 percent (see Figure 5.4), from 3.9 MT/ ha to 4.8 MT/ha, which approaches the vision target of 4.9 MT/ha. This is followed by: conservation agriculture, which could increase maize yields by around 14 percent over conventional practices; agroforestry (12 percent); minimum soil disturbance (9 percent); and residue retention (4 percent). For cotton, groundnuts, and soybeans, minimum soil disturbance offers the largest positive impact. Under the climate change projections, CSA practices show sensitivity to climate change but still outperform conventional practices in terms of yield gains. Drought-tolerant maize varieties show potential gains of as much as 8 percent; and adoption of conservation agriculture may potentially enhance the positive effects of climate change on soybean yields. The CSA practices that lead to the most pronounced yield reductions are minimum soil disturbance in cassava production systems, residue retention in soybean production and diversification in maize systems. This is due to the expansion of these crops into less productive areas which are expected to reduce the overall average yield.6 5.2 Crop Production Total crop production is projected to double to around 6.5 million tons of dry matter by 2050 even with conventional farming practices. Crop production in Zambia is driven by changes in domestic and foreign demand as well as supply side factors. Maize continues to be the dominant crop, with production likely to increase by 200 percent over 2010 levels (see Figure 5.5). Increased production of several other crops is projected to be more gradual and reflective of historical trends through 2010. This implies that without targeted interventions, the agriculture sector vision of increasing crop diversification may not be achieved. Considering climate change projections, total crop production could vary from an -11 percent decrease to an increase of up to 20 percent over levels under BAU. Although crop production is projected to double, crop yields are not projected to double, which suggests that a share of production increases can be attributed to the expansion of land for agricultural production. Even under climate change projections, CSA promises increased crop production over results from conventional practices. In a BAU scenario, reduction of post-harvest losses, which is modelled for most crops, offers a potential increase in total production of as much as 13 percent; next highest PAGE 56 Figure 5.3 Trends in crop yields under conventional practices until 2050, for projections without and with climate change (error bars) Cassava Groundnuts 16 1.6 1.4 14 1.2 12 1.0 10 0.8 8 0.6 6 0.4 tons/ha 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Maize Millet 5.0 4.5 1.8 4.0 1.6 3.5 1.4 3.0 1.2 2.5 2.0 1.0 1.5 0.8 1.0 0.6 0.5 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 BAU Historical Sources: Historical values are from FAOSTAT; GLOBIOM provides projections from 2000-2050 for scenarios without and with climate change; “Vision” was determined through stakeholder consultation and represents the targets for agriculture sector development in 2050. Note: Error bars indicate the range in crop yield for different climate change projections. Figure 5.4 Changes in crop yields under CSA, in %, as compared to conventional practices in 2050 Cassava Cotton Groundnuts 20 30 20 20 10 10 10 0 0 0 −10 −10 −10 af ca rr msd dtm phl div af ca rr msd dtm phl div af ca rr msd dtm phl div Maize Millet Soybeans 30 20 30 percent (%) 20 20 10 10 10 0 0 0 −10 −10 af ca rr msd dtm phl div af ca rr msd dtm phl div af ca rr msd dtm phl div Sugarcane 30 Source: GLOBIOM model results 20 Error bars indicate the Note: af refers to agroforestry; ca to range of possible crop conservation agriculture; rr to residue 10 yields that result from retention; msd to minimum soil different climate change disturbance; dtm to drought-tolerant 0 projections maize; phl to post-harvest loss; div refers to crop diversification. −10 af ca rr msd dtm phl div PAGE 57 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure 5.5 Trends in crop production under conventional practices until 2050, for projections without and with climate change (error bars) 9,000 8,000 Production (1,000 tons dry matter) 7,000 6,000 5,000 4,000 3,000 2,000 1,000 0.00 1961 1970 1980 1990 2000 2010 2020 2030 2040 2050 Barley Cotton Maize Rice Sugarcane Cassava Dry beans Millet Sorghum Sweet potatoes Chick peas Groundnuts Potatoes Soybeans Wheat Total production (FAOSTAT) Sources: Historical values are from FAOSTAT; GLOBIOM provides projections from 2000-2050 for scenarios without and with climate change. Note: Error bars indicate the range in total crop production as a consequence of climate change. Figure 5.6 Changes in crop production under CSA, in %, as compared to conventional practices, in 2050 Cassava Cotton Groundnuts 300 30 20 200 20 10 10 100 0 0 0 −10 −100 af ca rr msd dtm phl div af ca rr msd dtm phl div af ca rr msd dtm phl div Maize Millet Soybeans 30 20 300 percent (%) 20 10 200 10 100 0 0 0 −100 −10 af ca rr msd dtm phl div af ca rr msd dtm phl div af ca rr msd dtm phl div Sugarcane Total Source: GLOBIOM model results 40 Note: Error bars indicate the range in 20 crop production as a consequence of climate change. 20 Note: af refers to agroforestry; ca to 10 conservation agriculture; rr to residue retention; msd to minimum soil 0 0 disturbance; dtm to drought-tolerant maize; phl to post-harvest loss; div refers to crop diversification. af ca rr msd dtm phl div af ca rr msd dtm phl div PAGE 58 is crop diversification at 11 percent followed by: conservation agriculture (7 percent); minimum soil disturbance and agroforestry (both 4 percent); and residue retention (3 percent). There are notable variations between crops: adopting crop diversification has the potential to boost groundnut and soybean production by as much as 200 percent. Whereas conservation agriculture and minimum soil disturbance might increase maize and cotton production by around 10 percent and 5-20 percent, respectively compared to conventional practices. Climate change projections confirm that CSA has a positive impact on total production, an effect that could be enhanced by reducing post-harvest losses, and adopting agroforestry and minimum soil disturbance. Crop diversification away from maize seems to have a rather negative effect under climate, driven by declines in maize production (see Figure 5.6). 5.3 Livestock Production and Head of Livestock The goal included in the 2050 vision and agriculture sector targets to expand livestock herds to 6,000,000 head does not appear to be feasible, but improvements in feeding efficiencies which result in increased meat production offer gains nevertheless. In a business as usual scenario, cattle numbers are projected to increase 82 percent, from 2.6 million to 5 million head between 2010 and 2050, a significant rise although one that falls short of vision targets (see Figure 5.7). Measuring the development of the livestock sector by a cattle headcount provides at best a partial picture. Increases in feeding efficiency will result over time in livestock units that produce more output in the form of milk, eggs, and meat. Herrero et al. (2014) estimate that feed conversion efficiencies in sub- Saharan Africa could increase by 50 percent over the next 40 years (under a “Middle of the Road” SSP2 scenario). Figure 5.7 shows an increase in ruminant meat production between 2010 and 2050 of roughly 229 percent, stemming mainly from increased feeding efficiencies. The more than twofold projected increase in total production represents a strong improvement. Dairy production is expected to increase by more than 387 percent. 5.4 Land Use Change Through 2050, cropland area is projected to expand in the southern and southwestern regions of Zambia, at the expense of 0.9 million ha of mainly forest area since 2010. In 2010, 65 percent of the total land cover of Zambia consisted of forested area; agricultural land accounted for over 30 percent of the total land cover, of which only 2 percent was cropland. Under BAU, cropland area is projected to expand by almost 917,000 ha, with total agricultural land (cropland and grassland) growing by almost 2.2 million ha. Most of the new agricultural land will be formerly forested areas and natural land, reducing the country’s total forest area by 7 percent. Land conversion is projected to occur mostly in the Southern Province and Western Province, which fall within AEZ I and AEZ IIa, and have suitable soil and climate conditions for maize production. The projected growth in cropland is almost in line with the goal contained in the vision of converting no more than 0.9 million ha of land for agricultural purposes, but at odds with the goals set out in the National Policy on Environment, which aims at a zero increase in agricultural land. Figure 5.8 shows land use patterns over time and Figure APP 9 a-d (see Appendix E.2) provides a spatial representation. Projections under climate change scenarios hardly differ from those in the BAU scenario (see Figure 5.9). Following CSA adoption, increased crop yields tend to mean slower cropland expansion. The yield gains associated with the adoption of different CSA practices correspond with a avoided cropland conversion ranging from between 32,000 ha in the case of conservation agriculture and 43,000 ha in the case of minimum soil disturbance to 113,000 ha in the case of reducing post-harvest losses. For residue retention, cropland growth seems to increase by an additional 4,000 ha, the adoption of crop diversification and legume production results in a 700,000 ha increase in cropland, largely PAGE 59 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure 5.7 Number of livestock heads (above) and ruminant meat production (below) until 2050 6,500 6,000 5,500 5,000 4,500 1000 heads 4,000 3,500 3,000 2,500 2,000 1,500 1,000 500 0 1961 1970 1980 1990 2000 2010 2020 2030 2040 2050 BAU Historical 180 160 140 120 1000 tons 100 80 60 40 20 0 2000 2010 2020 2030 2040 2050 Sources: Historical values are from FAOSTAT; GLOBIOM provides projections from 2000-2050; vision was determined through stakeholder and policy consultation and indicates a target for agriculture sector development in 2050. Figure 5.8 Trends in land use under conventional practices, 1960-2050, for projections without climate change 45,000 40,000 35,000 30,000 1000 ha 25,000 20,000 15,000 10,000 5,000 0 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Sources: Historical values are from FAOSTAT, where Cropland Forest Grassland Natural land available; GLOBIOM provides projections from 2000-2050 GLOBIOM Historical Note: Historical data for forest and natural land are not available. PAGE 60 Figure 5.9 Changes in cropland until 2050, under conventional practices, for projections without and with climate change (error bars) 3,000 2,500 Historical (FAOSTAT) GLOBIOM 2,000 1,500 1000 ha 1,000 500 0 1961 1970 1980 1990 2000 2010 2020 2030 2040 2050 Barley Dry beans Millet Sorghum Sweet potatoes Cassava Groundnuts Potatoes Soybeans Wheat Cotton Maize Rice Sugarcane Total cropland (FAOSTAT) Sources: Historical values are from FAOSTAT; GLOBIOM provides projections from 2000-2050 for scenarios without and with climate change; the vision was determined during stakeholder workshop and indicates a target for agriculture sector development in 2050. Note: Total cropland excludes a number of crops (e.g., vegetables and fruits) for which total land use is small and cannot be modelled with GLOBIOM. Error bars indicate the range in total cropland as a consequence of climate change. Figure 5.10 Changes in land use due to CSA adoption, in 1000 ha, as compared to conventional practices, in 2050 Cropland Forest 800 50 600 0 400 −50 200 −100 0 −150 1000 ha af ca rr msd dtm phl div af ca rr msd dtm phl div Grassland Natural land 0 0.000000000002 0.000000000001 −250 0.000000000000 −500 −0.000000000001 −0.000000000002 af ca rr msd dtm phl div af ca rr msd dtm phl div Source: GLOBIOM model results Note: Error bars indicate the range in total land use category as a consequence of climate change. af refers to agroforestry; ca to conservation agriculture; rr to residue retention; msd to minimum soil disturbance; dtm to drought-tolerant maize; phl to post-harvest loss; div refers to crop diversification. PAGE 61 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN converted from grassland (see Figure 5.10). Any reductions or increases in cropland are matched by a corresponding expansion or decrease in grassland and hardly any forests or other natural lands are spared. All things considered, the impacts are small in comparison to total crop and grassland under BAU (around 1-5 percent of cropland and <1 percent for grassland). The adoption of CSA practices has limited impact reducing land conversion (see Figure 5.10). 5.5 Food availability Under BAU, caloric intake levels will rise steadily and hit the lower end of the recommended range of caloric intake in 2040 (see Figure 5.11). CSA practices are expected to increase food availability. According to FAOSTAT (2017), calorie availability in Zambia has hovered at or below 2,000 kilocalories (kcals) per capita per day since 1985.7 In 2010, this level dropped to 1,900 calories. In the BAU scenario diets are expected to evolve in line with FAO projections (Alexandratos and Bruinsma 2012).8 As a result, by 2050 calorie intake will have increased by 34 percent above 2010 levels, or by more than 700 kcals per day. FAO considers 2,500-3,000 kcals/capita/day a useful target for developing countries.9 CSA practices have a small but crucial effect on reducing crop prices and therefore increasing food availability, particularly, the practice of reducing post-harvest loss, which increases food availability by 7 percent, followed by conservation agriculture at 5 percent. The total effect is smaller than the increase in agricultural production because not all food produced is for consumption (see Figure 5.11). Figure 5.11 Trends in calorie availability, under conventional practices, projected until 2050 without and with climate change (error bars) 3000 Caloric upper norm 2750 Kcal/cap/day 2500 Caloric lower norm 2250 2000 1750 1990 2000 2010 2020 2030 2040 2050 GLOBIOM Historical (FAOSTAT) Sources: Historical values are from FAOSTAT; the GLOBIOM model provides projections from 2000-2050 Note: Error bars indicate the range in total calorie production as a consequence of climate change 5.6 Agricultural Trade Under BAU, trade trends vary across commodities. Climate change is a strong driver of both imports and exports. Figure 5.13 shows net agricultural trade (exports minus imports) for key agricultural commodities in 2010 and 2050 as well as the sector vision target of doubling net exports PAGE 62 Figure 5.12 Impact of CSA on food availability in comparison to conventional practices 9 6 percent (%) 3 0 −3 af ca rr msd dtm phl div Sources: GLOBIOM model results Note: Error bars indicate the range in total food availability as a consequence of climate change. Figure 5.13 Net agricultural trade in 2010 and 2050, in 1,000 tons Bovine meat Cassava Cotton Groundnuts 80.00 10.00 0.00 150.00 7.50 60.00 −100.00 100.00 40.00 5.00 −200.00 2.50 50.00 20.00 −300.00 0.00 0.00 0.00 −400.00 2010 2050 2010 2050 2010 2050 2010 2050 Maize Millet Pig meat Rice 200.00 100.00 0.00 0.00 150.00 75.00 1000 tons −30.00 −50.00 100.00 50.00 −60.00 −100.00 50.00 25.00 −90.00 −150.00 0.00 0.00 2010 2050 2010 2050 2010 2050 2010 2050 Soybeans Sugarcane Sweet potatoes Wheat 1,000.00 50.00 0.00 0.00 0.00 −100.00 −100.00 500.00 −50.00 −200.00 −100.00 0.00 −300.00 −200.00 −150.00 −400.00 −200.00 −500.00 −500.00 2010 2050 2010 2050 2010 2050 2010 2050 Sources: GLOBIOM model results (projections); vision was determined through policy and stakeholder consultation. Note: Error bars indicate the range in net trade as a consequence of climate change. PAGE 63 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN in 2050. The results are mixed. For cassava, cotton, maize, and millet net trade is positive in 2010 and is expected to increase by 2050. Of these, only maize and millet are expected to achieve the vision target of doubling net trade by 2050. For cassava and sugar cane, which also show positive net trade in 2010, a decrease is projected. For beef, pork, and several crops, including groundnuts, rice, soybeans, sweet potatoes, and wheat, net imports are expected to increase. The error bars indicate that climate change could have a substantial impact on the trade balance of several crops, particularly cassava, maize, cotton, millet, and sugar cane. For cassava and cotton, net trade is projected to exceed the 2010 values under an extreme climate change scenario, and to nearly hit the vision target. CSA is expected to increase Zambia’s international competitiveness and thus reduce its trade deficit in certain commodities. Increases in crop yield can impact the international competitiveness and contributes to higher exports. Post-harvest loss, conservation agriculture, crop diversification, and minimum soil disturbance are among the driving forces of increased net exports. Notably, Zambia has the potential to become a net exporter of groundnuts and reduce net imports of soybeans, if crop diversification is promoted through a policy change in subsidies, which lowers farmers’ production cost. However, these subsidies may jeopardize the net export potential of other crops (see Figure 5.14). Figure 5.14 Effect of conventional practices and CSA on net trade in 2050, in 1000 tons Cassava Cotton Groundnuts 600 250 100 400 0 50 200 −250 0 0 −500 −200 1000 tons none af ca rr msd dtm phl div none af ca rr msd dtm phl div none af ca rr msd dtm phl div Maize Millet Soybeans 200 100 400 150 0 100 200 −100 50 −200 0 0 −300 −50 none af ca rr msd dtm phl div none af ca rr msd dtm phl div none af ca rr msd dtm phl div Sources: GLOBIOM provides projections from 2000-2050 for scenarios without and with climate change. Note: Error bars indicate the range in net trade as a consequence of climate change. af refers to agroforestry; ca to conservation agriculture; rr to residue retention; msd to minimum soil disturbance; dtm to drought-tolerant maize; phl to post-harvest loss; div refers to crop diversification. 5.7 GHG Emissions Under conventional management practices, emission reductions arising from agriculture and LULUCF will not meet Zambia’s emission reduction goals as set out in the NDC. Livestock emissions PAGE 64 are expected to increase, and emissions from fertilizer are expected to contribute a significantly larger share to agricultural emissions by 2050. The NDC suggests emission reductions of 38 MtCO2e by 2030, which is stated to translate into a reduction potential of 47 percent from the 2010 baseline, conditional on substantial international support.10 This analysis uses the targets of reducing 25 percent, with only domestic funding, and 47 percent conditional on international support as targets for each emission category under analysis.11 Figure 5.15 shows six sources of agricultural emissions which are accounted for in the GLOBIOM model, excluding emissions from cultivation of organic soils, or burning of savanna or crop residues. Livestock accounted for more than 92 percent of emissions in 2010, the use of synthetic fertilizers, 7.3 percent, and rice cultivation, 1.3 percent. Emissions from synthetic fertilizer are projected to increase rather than decrease, to more than 13 percent in agriculture sector emission because of a shift toward more intensive crop production systems. Emissions from livestock are expected to increase steadily, reflecting expanding livestock herds (see Section 5.3). Under climate change, as indicated by the error bars in Figure 5.15, emissions are expected to increase by up to 3 percent (Figure 5.15). Figure 5.15 Trends in agricultural GHG emissions until 2050, under conventional practices, projected without and with climate change (error bars) 8.00 7.00 6.00 Limited support 5.00 Mt CO2e / year 4.00 Substantial support 3.00 2.00 1.00 0.00 1961 1970 1980 1990 2000 2010 2020 2030 2040 2050 Enteric fermentation Manure left on Pasture Rice Cultivation Total GHG Manure applied to Soils Manure Management Synthetic Fertilizers Emissions Source: Historical values are from FAOSTAT; the GLOBIOM model provides projections from 2000-2050 for scenarios without and with climate change. Note: Limited and substantial support indicate NDC goals. Error bars indicate the range in total emissions as a consequence of climate change. Minimum soil disturbance and reduced post-harvest losses are key abatement strategies to enhance the mitigation effect of synthetic fertilizers, but considered individually, are not sufficient to achieve the NDC emissions targets. See Box 5.1 for an overview of the impact of CSA adoption on GHG emission reductions. The GLOBIOM results focus on the impact of CSA on synthetic fertilizer. Figure 5.16 indicates that expanded use of minimum soil disturbance and reduction of post- harvest losses contributes to 11 percent and 5 percent decrease in emissions from synthetic fertilizer, PAGE 65 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure 5.16 Effect of CSA adoption on GHG emissions from synthetic fertilizer as compared to conventional practices, in %, in 2050 50 25 percent (%) 0 −25 af ca rr msd dtm phl div Sources: GLOBIOM model results Note: Error bars indicate the range in emissions as a consequence of climate change. af refers to agroforestry; ca to conservation agriculture; rr to residue retention; msd to minimum soil disturbance; dtm to drought-tolerant maize; phl to post-harvest loss; div refers to crop diversification. respectively, as compared to conventional practices. Adoption of post-harvest loss has an indirect impact: increased available production quantities leads to less expansion of cropland, which in turn cuts demand for nitrogen fertilizers. Taken individually, the impacts from adopting post-harvest loss and minimum soil disturbance are insufficient to meet the NDC target of reducing GHG emissions by between 25 and 47 percent for the synthetic fertilizer category. Expanding residue retention and implementing crop diversification policies will bring about respective increases in GHG emissions of 39 percent and 7 percent. However, GLOBIOM underestimates the GHG mitigation impact for CSA practices by failing to factor in their soil carbon sequestration potential (see Box 5.1; also Smith et al. 2008 and Corsi et al. 2012). Deforestation remains a major driver of GHG emissions through 2050, a trend that significantly hampers achievement of Zambia’s NDC emission reductions targets. GHG emissions from LULUCF– excluding biomass burning–are projected to almost triple from around 6.4 MtCO2e in 2010 to 37 MtCO2e in 2050 (see Figure 5.17), mainly as a result of land conversion from forest to cropland and grassland. This suggests that emission reduction goals of between 25 percent and 47 percent are not feasible. Under climate change projections, emissions from land use change and deforestation may increase even further. While some CSA practices are expected to slow the expansion of cropland (see Section 5.4), the positive effects of these on emission reduction may be outweighed by increased land conversion to grassland. Overall, CSA adoption can reduce emissions from land use change, but only by on average -0.32 percent compared to conventional practices. 5.8 Policy Scenario: Impact of Carbon Tax to Curb Land Use Change This section explores the impact of introducing a carbon tax for emissions resulting from deforestation and conversion of natural land for agricultural production. Land use change is a notable contributor to carbon emissions associated with Zambia’s agriculture sector, and presents an obstacle to the country meeting its NDC targets. We explore the impact of a range of carbon prices PAGE 66 Box 5.1 Potential impact of CSA on soil carbon sequestration The impact of CSA strategies on climate mitigation derives from: • Increasing the GHG emission intensity of a product, which can be calculated per hectare or per unit produced, that is, tCO2e emissions per unit of product produced for selected agricultural commodities. An increase in GHG intensity implies a more positive environmental impact per unit of product. This can be achieved by increasing application efficiency of fertilizer or enhancing the productivity of livestock by improving feeding practices and forage quality. • Increasing soil carbon sequestration, which entails increasing the amount of organic matter added to the soil from plant sources and animal waste and applying practices that reduce the decomposition of existing soil organic carbon stocks (for example, restoring degraded land; using residues to cover agricultural soil; avoiding soil disturbances) until a saturation point is reached, which typically occurs after 20 years. • Reducing the rate of conversion of forest land to agricultural land by increasing productivity per hectare while at the same time providing incentives to reduce land conversion. • Increasing biomass growth by introducing practices such as agroforestry, which supports the sequestration of carbon in tree biomass until the system is at full maturity. There have been few long-term studies of soil carbon dynamics or references to soil’s initial carbon stock, particularly in developing countries. Due to these data constraints, GLOBIOM cannot account for the soil carbon sequestration and biomass-growth impacts of specific CSA practices. We present additional evidence to emphasize CSA’s potential to enhance soil carbon sequestration: (i) The figure below shows experimental evidence of the soil carbon sequestration impacts of several CSA practices, which has a mitigation potential of between 0.2 and 1.1 tCO2e emission/ha. This is higher than the average emission levels occurring with conventional maize systems. (ii) The table below presents Tier 1 emission coefficients per hectare as provided by Intergovernmental Panel on Climate Change (IPCC) 2006 and the 4th Assessment Report of IPCC (Smith et al. 2007), for five management practices and two climate regimes in Zambia. Improved manure management seems to have the highest potential to enhance soil carbon sequestration. (iii) We use the tool EX-ACT, which uses Tier 1 emission coefficients, for a simple calculation examples: each practice in Table 1 is applied on 1,000 ha cropland; per hectare 50 kg synthetic fertilizer are applied instead of 100 kg. Over 20 years -112,619 tCO2e emission, or -1.1 tCO2e per year and per hectare, are saved. Thereof, 88 percent emission reduction are attributed to soil carbon sequestration and only 12 percent to emission reduction from reducing fertilizer. However, Tier 1 emission coefficients are highly aggregated and do not consider country-specific characteristics for soil carbon sequestration, and so need to be interpreted with caution. To improve coefficients for changes in soil organic carbon (SOC) stock and storage, disaggregated and area- specific information on previous land use, SOC stocks or SOC content and bulk densities for both the previous and current land use, soil depth, and time span since conversion are needed (Cardinael et al. 2018). Annual emission coefficients for soil Annual mean mitigation potential carbon sequestration, in tCO2e of CSA practices through soil carbon Practices/climate Warm Cool sequestration in Zambia regime temperate temperate and moist and dry 625 292 354 266 771 545 1105 Improved agronomic -10 0.88 0.29 practices Improved nutrient 0.55 0.26 Kg CO2 he-1 yr-1 -100 management practices No tillage, residue 0.7 0.15 retention -1000 Improved water 1.14 1.14 management -10000 Improved manure 2.79 1.54 application l l ns n s t nd ps l til til ro til Source: FAO (2016) op ov tio d Re ion tio or ll a o ed r c ed cr n N ta te uc rp ti ve uc a er ro re co ed ed co d es ue in c R C e du m id du e gu es si R an R Le Source: FAO (2015) re PAGE 67 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure 5.17 LULUCF GHG emissions under conventional practices, projected until 2050 without and with climate change (error bars) 40 35 Limited 30 support Mt CO2e / year 25 Substantial support 20 15 10 5 0 2010 2020 2030 2040 2050 Deforestation Other land use change Sources: GLOBIOM model results Note: Limited and substantial support indicate NDC goals; Error bars indicate the range in emissions as a consequence of climate change. on land use change using two policy scenarios: (i) imposition of a carbon tax on emissions arising from conversion of any kind of land to cropland, and from the conversion of forest land to grassland; however, conversion from natural land to grassland is still possible without an emissions tax (Crop- tax scenario); (ii) imposition of a carbon tax on emissions from conversion of natural land and forest to cropland and grassland; emissions from converting grassland to cropland are not taxed (Ag-tax scenario). Emissions are calculated as the difference between initial and final land cover carbon stock (Kindermann et al 2008; Ruesch and Gibbs 2008; Havlik et al. 2011).12 Three prices scenarios are used: (i) very low carbon prices (denoted vl1,2,3; and ranging between US$5, US$11 and US$21 per tCO2e emission) that reflect the typical market price of carbon for forestry projects; (ii) the lower and higher bound of shadow prices of carbon proposed by the World Bank (2017b) (denoted low and high respectively starting at US$40 and US$80, respectively); and (iii) a high price of US$200, tCO2e that Rogelj et al. (2018) and Frank et al. (2017) found may keep global warming below 1.5°C by 2100, as well as US$1 million per tCO2e, to show the upper extent of the effects of a carbon tax (denoted as vhigh and max). Prices in all scenarios are assumed to increase over time. Results shows that a carbon tax, even at low carbon prices, can effectively halt deforestation for agriculture. In the absence of a carbon tax, about 2.2 million ha of forest area would be converted (see Section 5.4). The Crop-tax scenario reduced emissions from land use change by 65 percent by 2050. The Ag-tax scenario reduces emissions from deforestation and other land use change by 99 percent by 2050 compared to BAU, and halts all deforestation and conversion of other natural land even at very low carbon prices (up to US$10/tCO2e by 2050). The difference in the effectiveness is due to the taxation of emissions released from converting other natural land to grassland, which are significant in Zambia. PAGE 68 In both scenarios, cropland area still increases, but less dramatically. The type of emissions taxed and the choice of carbon price have a significant impact on cropland expansion. In the Ag-tax scenario, expanding agricultural area from forest areas is unprofitable due to the carbon taxes, but grassland areas are converted into more profitable cropland. As a result, cropland area is reduced by only 3 percent at any carbon price compared to the no-policy scenario. In the Crop-tax scenario, even low carbon prices impact the profitability of conversion of grassland to cropland, and 8 percent less cropland would be converted by 2050 compared to a no-policy scenario. The expansion of cropland is still profitable in some areas. However, the number of hectares converted in 2050 under carbon price of US$200/tCO2e emission are half of what would be converted at low carbon prices. Thus, at very low prices, there are limited trade-offs on crop production (see Figures 5.18 and 5.19). Carbon tax policies, if enacted, would affect trade in livestock products and calorie availability. Under the Crop-tax scenario at a very high carbon price in 2050, imports could increase by nearly 30 percent compared to the no-policy scenario. At low carbon prices, maize exports can remain competitive but would shrink; and under high carbon prices millet could shift from being an export commodity to an imported one. Under low carbon prices the impact on trade would be less significant. Under the Ag-tax scenario, where grassland can substitute for cropland, the trade balance of crop commodities would be closer to the no-policy scenario. However, bovine imports would increase, and milk exports shrink. Depending on the mitigation scenario, carbon prices can impact food security. Under the Ag-tax scenario calorie availability falls by about 2 percent through 2050 compared to BAU, and by 1-3 percent under the Crop-tax scenario at very low carbon prices, or by as much as 4 percent if carbon prices are very high. Strict protection of forest and natural land (see the Ag-tax scenario) requires only a very low carbon price to achieve emission reductions and to reduce negative effects on food security indicators. Overall, depending on the goal of the policy (for instance, reducing emissions or protection of forests Figure 5.18 Available cropland in 2050 under BAU (no carbon tax) and carbon tax policy scenarios with varying prices, in million ha 2.5 2.0 1.5 Mha 1.0 0.5 0.0 vl1 vl2 vl3 low high vhigh max none Ag-tax Crop-tax No carbon tax (BAU) Sources: IIASA; GLOBIOM model results Note: Cropland area in 2010 is indicated by the black line. vl1, 2, 3, low, high, vhigh, and max denote different values of carbon prices at: US$5, US$11, US$21; US$40, US$80; US$200; US$1 million per tCO2e emission respectively. PAGE 69 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure 5.19 Difference in crop production in 2050 under carbon tax scenarios with varying prices as compared to BAU (no carbon tax), in % 0 Difference to BAU scenario in 2050 (%) −2 −4 −6 −8 −10 −12 −14 −16 −18 −20 −22 −24 vl1 vl2 vl3 low high vhigh max Ag-tax Crop-tax Sources: IIASA; GLOBIOM model results Note: Cropland area in 2010 is indicated by the black line. vl1, 2, 3, low, high, vhigh, and max denote different values of carbon prices at: US$5, US$11, US$21; US$40, US$80; US$200; US$1 million per tCO2e emission respectively. and natural land) even a very low carbon price applied to emissions from deforestation effectively stops deforestation for crop or livestock production. At a low carbon price that targets only emissions from land use change that results in expanded croplands reduces emissions by 65 percent in the case of BAU, while limiting trade-offs to food security (3 percent less kcals/capita/ per day than with BAU), crop production (5 percent lower than with BAU), and other land use change (27 percent less other natural land compared to 2010). Strict protection of all forests and other natural land requires only a very low carbon price to halt expansion of agricultural area and reduce LUC emissions by 99 percent compared to BAU. Substituting grassland for crop production under a strict land protection scenario would halt all deforestation and other land use change with only limited trade-offs in terms of food security (2 percent less kcals/capita/day) and crop production (2 percent less than with BAU. 5.9 Economic Implications of Scaling-up CSA Adoption CSA adoption can provide considerable economic benefits: under current CSA adoption rates a 27 percent rate of return on investment is feasible, climbing to 54 percent if the value of reduced GHG emissions is considered. Sections 5.1 to 5.7 showed the potential impact of CSA on key agriculture sector indicators. This section presents the results of an economic assessment of CSA, and its value to Zambian economy. As discussed in Section 2.4, two adoption scenarios were considered: an 8 percent adoption rate, equivalent to approximately 118,000 farmers, equal to the current adoption rate for minimum soil disturbance; and a 50 percent adoption rate, equivalent to approximately 700,000 farmers. The latter reflects the assumption of 50 percent CSA adoption in 2050, as modeled in GLOBIOM for minimum soil disturbance. Both analyses assume an economic cost for rollout of CSA of US$260 per household, resulting in total costs of US$30 million and US$200 million, respectively. Table 5.1 shows that the investment will yield positive returns and is robust against nearly all sensitivity analyses scenarios. PAGE 70 Table 5.1 Economic indicators for two CSA adoption scenarios showing private and public good benefits 8% adoption 50% adoption Private Public Public Private Public Public benefits benefits, benefits, benefits benefits, benefits, low SPC high SPC low SPC high SPC EIRR (%) 27% 35% 42% 34% 55% 70% NPV (US$) 21 million 35 million 48 million 170 million 460 million 750 million Sensitivity analyses: EIRR in %; NPV in million US$ EIRR NPV EIRR NPV EIRR NPV EIRR NPV EIRR NPV EIRR NPV Benefits -10% 23% 16 30% 28 37% 40 29% 136 49% 397 63% 653 Benefits -20% 20% 11 26% 22 31% 32 25% 101 43% 334 56% 562 Benefits -50% 9% -4 13% 2 17% 8.5 12% -0.6 26% 144 35% 286 Project cost + 10% 24% 18 31% 32 37% 45 30% 153 50% 443 63% 728 Project cost + 20% 21% 15 27% 29 33% 42 26% 135 45% 426 58% 711 Project cost + 50% 15% 6 20% 19 25% 32 19% 84 35% 375 47% 659 Delayed benefits: 21% 16 26% 28 30% 39 25% 132 38% 391 48% 640 1 year Delayed benefits: 17% 10.5 21% 21 24% 31 20% 97 31% 326 37% 546 2 years Market price $5 28% 23 37% 204 Market price $11 29% 25 41% 245 Market price $21 31% 28 46% 308 Source: World Bank, own calculation Note: EIRR stands for Economic Internal Rate of Return; NPV for Net Present Value; SPC stands for shadow price of carbon which means that climate mitigation benefits have been valued at a shadow price of carbon and included as public good benefit in the economic analysis. Sensitivity analyses are conducted for: (i) decreasing private benefits, (ii) increasing investment cost,(iii) delay in realization of benefits, all between 10 and 50%; and for varying carbon market price for GHG emission reduction. Increasing CSA adoption to 50 percent could generate an economic rate of return of 34 percent if private benefits are considered. Considering the value of emission reduction to society, the net present value of the investment could increase by between 171 percent and 341 percent. Both adoption scenarios achieve a sizable net carbon balance: Over 30 years and under current adoption rates for minimum soil disturbance, an annual net carbon balance of approximately -47,700 tCO2e emission could be achieved. In a scenario with a 50 percent adoption rate, the annual net carbon balance is approximately -1,137,500 tCO2e.13 Valued at a shadow price of carbon (see Section 2.4), the value of emission reduction to society is significant: the net present value in the 8 percent adoption scenario, US$21 million, could increase by between 67 percent and 129 percent; in the 50 percent adoption scenario, it could increase from US$170 million by between 171 percent and 341 percent (see Figure 5.20). Even at potential carbon market prices between US$5, US$11, and US$21/tCO2e emission (as presented in Section 5.8), the NPV could increase by between 10 and 14 percent in the current adoption scenario; and between 20 percent and 81 percent in the 49 percent adoption scenario. While the absolute value of emission reduction in tCO2e emissions is small compared to Zambia’s ambitious NDC goals, the potential economic value to society is large. PAGE 71 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure 5.20 Effect of climate mitigation benefits stemming from the scaling-up of CSA to 50% of smallholders on economic indicators (EIRR in %; NPV in US$ million) 80% $800 $751 70% 70% $700 80% $800 $751 60% 70% $600 70%  55% $700 103%  50% $500 60% 341% 46% 55% $461 $600  41% 40% 103%  37% $400 50% 34% $500 341% 46% $308 $461 30% 41% $300 40% 37% $245 $400 34% $204 20% $170 $308 $200 30% $300 $245 10% $204 $100 20% $170 $200 0% $0 10% Private benefits Carbon price Carbon price Carbon price Low shadow High shadow $100 (without US$5 US$11 US$21 price of carbon price of carbon emission $0 0% reduction Private benefits benefits) Carbon price Carbon price Carbon price Low shadow High shadow (without US$5 US$11 US$21 price of carbon price of carbon Source: Worldemission Bank, own elebatoration reduction benefits) Figure 45% 5.21 Effect of climate mitigation benefits stemming from adoption of CSA by 8% of smallholders $60 on economic indicators (EIRR in %; NPV in US$ million) 42% 40% 45%  $48 $50 $60 42% 35% 56% 34% 40% 31% 30%  29% $50 $40 28% $48 35% 56% 27%  34%$35 25% 129% 31% 30% 29% $40 28% $28 $30 20% 27%  $25 $35 25% 129% $23 $21 15% $28 $20 $30 20% $25 $23 10% $21 15% $20 $10 5% 10% 0% $0 $10 5% Private benefits Carbon price Carbon price Carbon price Low shadow High shadow (without US$5 US$11 US$21 price of carbon price of carbon 0% emission $0 reduction Private benefits benefits) Carbon price Carbon price Carbon price Low shadow High shadow (without US$5 US$11 US$21 price of carbon price of carbon emission reduction benefits) Source: World Bank, own elebatoration PAGE 72 Section 6 Prioritizing CSA Practices Crop diversification, commercial horticulture, agroforestry, and reduced post-harvest losses, produce some of the strongest results with relatively few trade-offs between positive outcomes on the household and sector level. Sections 4 and 5 discuss CSA impacts on household welfare and on sectoral performance, respectively. Section 7.1 describes CSA against the backdrop of Zambia’s enabling environment, and Section 7.3 elaborates which practices have received donor-funds in the past year. The aggregated results of both reviews are presented in Table 6.1 in the column “Feasibility”. Based on these findings, Table 6.1 ranks every CSA practice to enable decision-makers to prioritize investments in the sector. Reading the table from left to right suggests that CSA practices such as crop diversification, commercial horticulture, agroforestry and reducing post-harvest losses are more effective than others at delivering on key indicators. Reading the table from top to bottom suggests that some evaluation criteria, for example, food security criterion receives a boost from most CSA practices, while the effect of CSA on GHG emission reduction—excluding the potential for soil carbon sequestration due to modeling constraints—seems varied. What is needed is an enabling environment and investment strategies that will enable and accelerate adoption of these practices and maximize their impact. PAGE 73 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table 6.1 Impacts of CSA practices on key decision variables Rural Past household-level impact Projected sector-level impact Develop- Feasibility ment Income Reduce Food Crop yield GHG Food Job creation Available Enabling income security mitigation availability & market funds environment variability linkages Minimum soil disturbance Wet conditions Crop yields Climate Climate Climate Requires (–) increase, change (–) change (++) change (++) establishment Long run strong spatial Strong increase Enhances Contingent of input positive; short variation under BAU; soil carbon on adoption markets run negative; decrease under sequestration taking place (implements, strong climate change in favorable inputs, mech- geographic areas anization) variation Residue retention Wet conditions Dry conditions Climate Climate Climate No notable (–) (+) change (–) change (–) change (–) impact Slight increase Increased under BAU; fertilizer negative under emissions; but climate change soil carbon sequestration occurs Crop rotation/ n/a n/a n/a intercropping Wet conditions Strong Dry conditions Investment (–) geographic (+) in legume Long run variation Strong value chain positive; short geographic development run neutral variation and Crop n/a processing; diversification access to new markets (largely Dry conditions Dry conditions Climate Climate (traders, Ongoing legumes) (+) (+) change (–) change processors) reforms Increases for Increased creating job expected to Wet conditions Wet conditions most crops, fertilizer lower barriers opportunities (+) (+) though less emissions; but under climate soil carbon change sequestration occurs Commercial n/a n/a n/a horticulture Dry conditions Access to (+) new markets, processing and value addition Agroforestry No effect in the Dry conditions Climate Climate Climate Low potential short term (+) change (–) change (–) change (++) of value chain Increase is less High levels development Wet conditions under climate of soil carbon (maybe (+) change sequestration nursery) and biomass growth Livestock n/a n/a diversification Wet conditions Dry conditions Dry conditions Increase in Opportunities Ongoing (–) (+) (+) livestock herds for veterinary policy reforms No significant increases services, expected to impact; methane trading, lower barriers geographical emissions processing variation Delayed n/a n/a n/a planting Dry conditions Development (–) of input markets and Wet conditions advisory (–) services Drought- and heat-tolerant seeds Dry conditions Dry conditions Dry conditions No notable No change No change Development Moderate (+) (+) (+) increase under under BAU; under BAU, of input constraint BAU; positive climate change positive effect markets can Wet conditions Wet conditions Wet conditions under climate could equally under climate create jobs (+) (+) (+) change increase/ change decrease the effect on mitigation Reducing post- n/a n/a n/a n/a harvest loss Source: World Bank, own elaboration Strong positive effect Low-medium positive effect Negative effect Note: (i) Colors (green, yellow, red) indicate the impact of CSA practices on key indicators where green indicates a positive effect, yellow a low-to-medium positive effect; and red a negative effect; (ii) Dry conditions/Wet conditions +/- indicates whether increased or decreased rain have a positive or negative effect on CSA’s impact on the key indicator; (iii) Climate change ++/+/- corresponds to the error bars in GLOBIOM figures and indicates the magnitude and directionality of the range of effects that a CSA practice can have on indicators. It indicates whether the effect is strongly positive, or rather positive or negative under climate change projections. n/a indicates that due to data and methods constraints a quantitative analysis couldn’t be conducted. (a) Information on PAGE 74 Enabling Environment and Available Funds, can be found in Section 7. Section 7 Enabling Environment and Mechanisms to Promote CSA 7.1. Incentivizing CSA Adoption by Improving the Institutional Enabling Environment in Zambia Historically, adoption rates of CSA practices have been low and there has been uncertainty over whether the institutional and policy enabling environment have hindered adoption. As discussed in previous sections of this report, CSA practices, if tailored to specific biophysical conditions, can enhance household net incremental benefits, resilience, and welfare indicators and support both agriculture sector goals and aggregated economic benefits including public good benefits. Several barriers to adoption have been identified related to high upfront and labor costs. Section 7.1 summarizes key features of the institutional enabling environment which could enhance or hamper the adoption of CSA in Zambia. Section 7.2 introduced eight innovative and proven investment strategies which could help overcome these constraints and support a broad adoption of CSA. Security of land tenure is a key driver of on-farm investment. In rural Zambia, customary tenure is most common, and this likely impedes productive investment. Customary tenure is governed by traditional norms under which land is administered by a headman such as a chief. Land under customary tenure cannot be legally transferred, and traditional leaders have a large degree of autonomy in making decisions about the use of land, which causes uncertainty and insecurity regarding property rights—usage rights, management rights, and transfer rights. Customary tenure also complicates long-term investments which are critical for many CSA practices, such as agroforestry and the fencing of crop residues to avoid livestock intrusion. By contrast, commercial farmland is regulated by statutory tenure, which in Zambia is administered through the government’s Ministry of Land and Natural Resources and also applies to urban areas and district municipalities. Evidence PAGE 75 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN shows that land tenure status is a significant factor in whether or not smallholders adopt irrigation systems (Ngoma et al. 2017). There is also evidence that a shift from usufruct rights to full ownership increases incentives for soil conservation and productive investments such as tree planting (Deininger and Feder 2009). Improving tenure rights, transparency, and security would also substantially improve women’s economic prospects (Namonje-Kapembwa and Machina 2018). Capacity building and agricultural extension services are key for agricultural development and adoption of improved technologies. However in Zambia, public extension services are severely underfunded. Subsistence farmers are conservative adopters of technology, and need capacity building, extension and advisory services to make CSA adoption more effective and minimizes risks. While the Zambian government has acknowledged in several policy documents14 the need for a strong extension system, there are significant gaps in: institutional capacity; technical and training capacity; planning; reporting and feedback; initiatives pertaining to non-functional value chains; extension officer-to-farmer ratios; and coordination and communication. For examples, agriculture made up 9 percent of the government’s budget in 2015, a mere 2 percent of that, or ZMK89 million, was channeled toward agricultural research and extension services, as compared to the more than 50 percent budget allocated to the FISP and FRA programs (Kuteya et al. 2016). As a result, there are very few extension and advisory services that cater to the specific needs of women smallholders. In addition, agricultural extension services emphasize maize, a staple, rather than a more diverse set of nutrient-rich alternatives that are traditionally produced by women. Information and communications technologies (ICTs) can be valuable in providing agricultural advisory services to farmers although coverage and service delivery will need to be improved. Empirical evidence shows the tangible and positive impact of ICTs on agriculture. Increased mobile phone coverage in Uganda has helped farmers to better coordinate the storage and sale of perishable commodities and to achieve higher prices (Muto and Yamano 2009). ICTs also support the adoption of agricultural technologies, by allowing social learning and private information sharing (Goyal and González-Velosa 2012). ICTs also potentially increase opportunities for women, who face barriers in accessing inputs and information, and might also improve youth engagement in modern agricultural systems. In Zambia mobile phone usage has grown sharply (from <10 percent of the population in 2005 to 80 percent in 2018); by contrast internet penetration is around 47 percent, according to the ZICTA Statistics Portal. However, overall access to reliable ICT services is limited and often nonexistent in rural areas. Even in areas with coverage, the poor service delivery hinders effective use. Access to financial services and credit constrains adoption of improved technologies and enhancement of business opportunities in the agriculture sector. Low population density, low income, and poor financial literacy contribute to an environment in which it is too expensive to offer loans to smallholders. Only 15 percent of Zambia’s 1.5 million farmers have taken out loans, two thirds of which can be attributed to outgrower schemes, specifically for farmers with landholdings between 0.5 ha and 20 ha. Commercial finance is virtually nonexistent, except for large and medium size farmers. Households with less than 0.5 hectares of land resort to informal loans (Chapoto and Zulu- Mbata 2016). There are several feasible approaches to enhance access to finance: financial guarantee mechanisms that would allow financial institutions to provide services to emergent or commercial farmers, or alternatively agro-suppliers to provide smallholder farmers with seeds and fertilizer on credit that would be paid back once crops were sold. Also, warehouse receipt systems might offer farmers collateral that would enable them to get credit from microfinance institutions; and matching donor grants or credit lines extended to smallholders or cooperatives of smallholders could also be an option. All of these approaches dovetail with the provision of specific agricultural and financial advisory and extension services (World Bank 2018b; Chisanga and Chapoto 2018). PAGE 76 Inadequate production and market infrastructure impedes development of and access to domestic and regional agricultural markets, which reduces income opportunities and incentives to invest in new practices or diversify production. Irrigation infrastructure in Zambia is particularly poor, despite enormous potential, and the importance of irrigation in crop diversification. Empirical evidence confirms that households situated in wetlands are more likely to diversify crop and agriculture income than those situated in dry areas (World Bank 2018d). Minimizing distances to paved roads reduces transportation costs for both inputs and outputs, connecting farmers to markets and providing incentives to adopt improved practices or diversify production. Shorter distances to roads have been shown to have a significant positive effect on the establishment of small and new agro-processing firms (Norman et al. forthcoming). Improved access to markets is considered a driving factor for changes in production patterns. Farmers tend to specialize in crop production with high market demand when they live closer to urban centers (World Bank 2018d). There are examples of partnerships between rural communities and agribusinesses, where agribusinesses provide a reliable market for farmers’ produce, provided farmers are using improved agricultural practices (Lewis et al. 2011). In addition to hard market infrastructure, improved business skills, timely information about market prices, or the possibility to coordinate and aggregate produce, for instance through a farmer association or producer group, can enhance farmers’ market access. Distortionary public policies and subsidies encourage the adoption of mono-cropping under conventional management practices. The Zambian government maintains price and input support programs that consumed an average of 79 percent of the national agriculture budget between 2008 and 2016 (World Bank 2017a). Recent studies show that FRA’s maize purchases reduce the likelihood of farmers diversifying crop production and crop income (World Bank 2018d; Mofya-Mukuka and Shipekesa 2013). Traditional input subsidy programs impede the development of private input markets and incentivize production of one crop, rather than enabling diversification and adoption of improved practices. In the 2015/16 season, the Zambian government piloted an e-voucher system that increased transparency (in terms of the beneficiaries who receive subsidies), increased competition among agro-dealers which boosted prices and diversification, and encouraged farmers to redeem vouchers for diverse crop and livestock products. An evaluation of the pilot has shown that farmers still tend to redeem inputs vouchers for maize-related products, possibly due to lack of capacity or poor timing in the provision of subsidies, that is subsidies are provided too late to enable another crop than maize to be planted. FSIP should support crop diversification, be paired with capacity building efforts to improve adoption of new practices and technologies, and roll out training for agro-dealers (Kuteya et al. 2018). The savings achieved from these reforms could then be allocated to productive investment such as irrigation, agriculture research and extension services, or building rural infrastructure (World Bank 2017a; World Bank 2018a). Table 7.1 shows that three conditions appear to be most constraining for the adoption of CSA: the lack of capacity building, lack of access to finance, and access to markets. These constraints should be addressed in going forward, which will facilitate the promotion of priority CSA practices as identified in Section 6. For agroforestry, crop diversification, and commercial horticulture, capacity building and receiving appropriate advisory services is a constraining factor for adoption. Access to finance in order to obtain new varieties and farming equipment including on-farm storage solutions, as well as access to input and output markets which is a key incentive to improve and diversify farming practices, are constraining the adoption of prioritized CSA practices. PAGE 77 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table 7.1 Evaluation of dimensions of a conducive enabling environment for the adoption of CSA Land tenure Capacity Access to Access to rural Access to Distortionary systems building finance infrastructure markets public policies Conservation agriculture (minimum soil Lack of secure Deferred Upfront costs Opportunities Fertilizer disturbance, tenure is a benefits pose a (e.g., inputs, to generate subsidies residue disincentive to risk to farmers labor, income reduce may decrease retention, adopt practices but this can mechanization) risk of adopting economic rotation) with deferred be mitigated are found new practices rational for benefits by adequate to impede adoption; Agroforestry capacity adoption subsidy building schemes could be paired with advisory services to enhance adoption. Commercial horticulture Horticulture Diversifying Access to Irrigation, roads, and access Can lead has short into new crops finance critical to markets crucial for stable to lower cultivation requires access as new inputs, production and links to agri- competition cycle, and can to advisory infrastructure businesses among input be cultivated services and transport suppliers and on small are needed traders as well plots; insecure to set up as higher input land titles operations costs could hinder investment in irrigation systems. Livestock diversification Mobile asset; Diversifying To start Access to new Current focus grazing often into new operation, species, and on maize occurs on species access to health services impedes the communal land requires access finance critical; is a challenge production to advisory livestock serves of crops to services incl. as “savings develop animal health bank” livestock feed industry Timing of No specific planting impact Lack of advisory Lack of inputs Access to Lack of input Late delivery services affects negatively infrastructure markets often of subsidies planting affects timing facilitates causes planting was found to decisions of planting delivery of delays affect timing of inputs and planting reduces involuntary planting delays Drought- and heat-tolerant seeds Benefits accrue No specific Improved No specific FISP has shortly after capacities seed usually difference from facilitated planting compared to costly, access conventional access to conventional needed to start practices, improved practices operation except supply maize seeds of seeds. Crop diversification Research shows Access to Access to Access to markets and irrigation Disincentivizes small land sizes advisory finance critical supports diversification diversification; disincentivize services a new inputs but ongoing diversification crucial driver of needed to set reforms of FISP diversification up operation could reduce barriers Reducing post-harvest loss Access to Access to Access to rural Access to Disincentivizes advisory finance critical infrastructure markets private sector services can to get access includes provides involvement help reduce to adequate storage incentives to in storage and post-harvest technologies facilities which reduce post- warehouse losses on farm are critical harvest loss facilities Source: World Bank, own elaboration Low barriers to adoption Medium barriers to adoption High barriers to adoption PAGE 78 7.2 Summary of Innovative Delivery Mechanisms to Support CSA Adoption Innovative approaches are required to promote CSA adoption in Zambia and can help to overcome barriers in the enabling environment. Even though long-term household benefits incentivize CSA adoption at the farmer level, adoption remains low, often because information, skills and support to cover upfront costs are missing, inadequate access to finance, and lack of input and output markets (see Table 7.1). Several mechanisms have been evaluated that can overcome these barriers. This section presents eight mechanisms, which can be treated as investment options or project interventions to rollout the adoption of CSA. A more detailed description is presented in and Appendix G. Table 7.2 also indicates which of the three key enabling constraints, that is, inadequate access to finance, access to markets and capacity building, they can address. Table 7.2 Delivery mechanisms and their potential to address enabling environment constraints The delivery mechanisms have the potential to enhance: Capacity building and Access to finance Access to markets Delivery mechanisms advisory services Business partnerships with rural Contracts can serve as Farmers receive training communities—leveraging carbon collateral; Agribusiness on relevant commodities finance could facilitate access to and practices carbon finance Outgrower scheme—commercializing Contracts can serve as Farmers expected to horticulture production collateral; agribusiness receive training on may provide loans to relevant commodities and participating farmers practices Participatory integrated landscape Mitigation benefits from Participatory element management approach—achieving landscape approach facilitates knowledge No specific impact multiple objectives could enhance eligibility exchange to participate in a carbon finance project Farmer field school— “farming-as-a-business” community-based learning and No specific impact trainings and engaging technology adoption in farmer associations facilitate market access Pluralistic participatory extension Service providers connect approach—supporting linkages No specific impact farmers to markets; between research and dissemination “farming-as-a-business” trainings provided Weather index insurance —combining Can be combined with With mandatory CSA access to savings banks, adoption policyholders No specific impact risk transfer with risk management revolving funds and should receive trainings financial literacy training Cash transfers—Alignment with the Payments at correct Could be paired with time of season can demonstrations and No specific impact Harvesting Cycle to Promote CSA cover upfront cost for capacity building agricultural investment Gender sensitive supply chains Reducing barriers for Reducing barriers for Reducing barriers for women to access finance women to receive women to enter markets adequate training PAGE 79 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table 7.3 Comparing innovative delivery mechanisms for scaling up CSA Investment cost over Benefit- NPV(b) EIRR (%) Investment areas and public sector activities 6 years cost ratio(a) ($ million) ($ million) Weather index insurance (WII)—combining risk transfer and risk mitigation(c) US$15.7 1.32 US$4.14 18.8% (i) Capacity building and training of farmers, insurers and regulators, including development of guidelines and training US$20.14 1.03 US$0.5 11% modules, sensitization for formation of local saving groups (including subsidized (ii) Provide technical and operational support for agricultural premium of insurance scheme (e.g., feasibility study, product and scheme US$10-30) design, involvement of private partners); product delivery and operation; marketing, monitoring and results dissemination (iii) Ensure data availability, functionality, and integrity of weather station (iv) Reduce legal and regulatory risks to policyholders and insurers (v) Appoint an independent agency to verify, on behalf of insured parties, contract and payout accuracy (vi) Improve and maintain operationalization of the electronic voucher system Business partnerships with rural communities—leveraging carbon finance US$25.1 1.6 US$13.3 31% (i) Ensure that the project is aligned with regional and community development plans (ii) Provide capacity building and training for farmers to adopt CSA, post-harvest management, and farming as a business (collective marketing, business, marketing, quality control) (iii) Promote business linkages with producer groups, agro-dealers and storage facilities; conduct farmer trade fairs; conduct market research of target crops (iii) Support access to credit by enhancing capacity of commercial banks and strengthening the policy and regulatory environment, e.g., access to inputs, revolving funds, purchasing logistics, equipment and packaging materials (v) Factor private sector needs into investment in public goods and services Pluralistic participatory extension approach—supporting linkages between research and dissemination US$51.2 0.75 -US$10 4.4% (i) Support agricultural research and development facilities and specific research through public research centers to develop and test improved practices (ii) Coordination of pluralistic extension services (iii) Capacity building for public extension officers and lead farmers (iv) Setting up the FFS for demonstration purposes (v) Develop sectoral or local policy that supports private sector service provision (vi) Develop integrated planning skills for local governments (vii) Support development and strengthening of farmer organizations Farmer field school—community-based learning and technology adoption US$22.54 1.7 US$12.7 38.58% (i) Support sensitization of farmers to adopt new learning approaches (ii) Support the identification of farmers and formation of learning groups and strengthen farmer groups involved in the FFS (iii) Provision of initial production inputs, clean materials for establishing demonstration plots and seed multiplication schemes; and CSA learning material (iv) Support suitability and prioritization studies for crops and CSA practices (v) Conduct field days and knowledge exchange between different groups Source: World Bank, own elaboration Note: (a) Benefit-cost ratio was calculated by dividing discounted benefits over 30 years by discounted cost (investment cost and maintenance cost) over 30 years. A discount rate of 10% was used. A ratio above one indicates profitable investment. (b) NPV was calculated at a discount rate of 10 percent and for approximately 118,000 beneficiaries. (c) The assessment of insurance benefits is not straightforward. For this analysis we rely on Varadan and Kumar (2012) who found that crop insurance can absorb production risk and increased input use. Insured farmers spent more on inputs and realized higher returns from farming (+125 percent) than their non-insured counterparts. PAGE 80 Table 7.3 Continued Investment cost over Benefit- NPV(b) EIRR (%) Investment areas and public sector activities 6 years cost ratio(a) ($ million) ($ million) Outgrower scheme—commercializing horticulture production US$25.13 1.39 US$7.9 21.1% (i) Support the development of irrigation infrastructure and governance structures (ii) Support for contract design and enforcement (iii) Support smallholder capacity and coordination (iv) Promote private sector adherence to principles of responsible investment (v) Support effective public-private dialogue to build (regional) markets (vi) Ensure compliance with social and environmental standards (vii) Improve land tenure security and access to land (viii) Invest in public goods and services considering private sector need (ix) Enhance public inspections and quality assurance Participatory integrated landscape management approach—achieving multiple objectives US$28.1 2.32 US$30.1 36.4% (i) Ensure compatibility with community, landscape and public sector goals (ii) Community planning: support risk and vulnerability assessments; support community organization; formation and strengthening of groups; support land management/watershed planning with relevant tools (e.g., GIS, ICTs) (v) Capacity building: train communities or link them with service providers to address vulnerabilities in landscape (vi) Infrastructure, logistics materials: provide materials for land restoration, production inputs, watershed management, materials for forest management (iii) Review legal and regulatory set-up and whether there is need for reform (iv) Initiate dialogue and facilitate process US$28 1.41 US$9.6 22.4% Livestock: (i) Develop animal husbandry productivity assessments and implementation plans (ii) Support procurement and distribution of improved planting material (iii) Support forage demonstration and promotion activities (iv) Support animal health services and vaccination Table 7.3 provides suggested steps for the public sector to follow to help crowd-in private sector financing to support adoption of CSA. Currently, public resources are either scarce or are not being used effectively to support achievement of sector goals. The World Bank is committed to the Maximizing Finance for Development (MFD) framework, which aims to support public sector activities in a manner that crowds in and maximizes private sector investment. The MFD framework promotes a range of potential actions to crowd-in the private sector: promote responsible investment; increase the private sector space by removing restrictions; improve the policy and regulatory environment for private sector investment; reduce transaction costs and risks; and use public resources to invest in public or quasi-public goods and services (World Bank 2018c). Table 7.3. provides for each investment opportunity a range of activities that can support private sector involvement to fund the respective mechanisms. See Appendix G for a further discussion of each delivery mechanism. PAGE 81 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Weather Index Insurance—Combining Risk Transfer with Mitigation Weather index insurance (WII) is a risk transfer mechanism to enhance household’s resilience against climate change. In Zambia, the World Food Program (WFP) pairs WII with risk mitigation strategies and mandatory CSA adoption. WII compensates policyholders based on a weather index (for example, rainfall) and estimates losses in a particular location and period of time. In their Rural Resilience Initiative (R4) program, WFP paired WII with the roll out of CSA. The R4 has four components: improved resource management and CSA (risk mitigation); insurance (risk transfer); livelihood diversification; and microcredit (prudent risk taking), and savings (risk reserves). The R4 provides a high level of services to farmers and has so far reach approximately 4,000 households (WFP and OXFAM 2017). Business Partnerships with Rural Communities—Leveraging Carbon Finance Agribusinesses play a pivotal role in enhancing the profitability of agriculture-related climate change mitigation by: incentivizing farmers to adopt CSA practices; providing market opportunities for compliant farmers; and extending access to carbon finance. Under this arrangement, an agribusiness develops processed products, which are certified as environmentally and socially sustainable. The agribusiness provides farmers with premium prices if they adopt CSA and natural resources management practices that will advance climate mitigation. If the agribusiness can demonstrate emissions have been reduced, the resulting emission reduction credits can be sold to a carbon fund. In Zambia, the company Community Markets for Conservation (COMACO) is a well- known example of this business-model. Following the first successful verification of carbon credits in October 2017, BioCarbon Fund15 paid US$814,406 to participating communities and COMACO. The model is attractive for urban consumers who can contribute to climate mitigation by buying certified labelled products. However, accessing carbon funds is often hampered by high transaction costs. COMACO receives donor funding to finance supervision, monitoring, and implementation of its community activities (World Bank 2018c). Outgrower Schemes—Commercializing Horticulture Production Outgrower schemes are a form of contract farming in which smallholder farmers and agribusinesses or processing plants engage in a binding agreement. Through this coordinated commercial relationship, the company typically provides support to the farmers with production planning, input supply, and adoption of new technologies including CSA practices, extension advice, and transport. Farmers benefit by gaining access to markets and financial services, reduced price fluctuations and by receiving support in achieving quality standards and technology, skills and knowledge transfers, and opportunities for seasonal employment. There is a risk, however, of dependency, power asymmetry, limited transparency and even disputes over prices and contract conditions. Outgrower schemes are feasible with all types of crop production, but are most often used to source horticulture products and combined with access to small-scale irrigation. Pluralistic Participatory Extension Approach—Supporting Linkages Between Research and Dissemination The pluralistic participatory extension (PPE) approach harnesses different sources of advisory services to disseminate innovative technologies. Extension and agricultural advisory services are key for spreading the adoption of CSA and achieving development goals. The PPE approach recognizes the diversity of farmers and farming systems, and is characterized by the coexistence of multiple public and private sector approaches, providers, funding streams, service types, and sources of information and experiences. Service providers can include membership-based farmer organizations, private and commercial enterprises, and non-governmental organizations (NGOs). The PPE approach works well with agricultural research services, and supports the development and testing of climate-resilient solutions for smallholder farmers (Heemskerk and Davis 2012). PAGE 82 Farmer Field Schools—Strengthening Community-Based Learning and Technology Adoption Farmer field schools (FFS) provide education and extension services often with a focus on CSA practices. FFS have emerged as a complementary and reinforcing approach to traditional agricultural advisory services and provide season-long programs for farmers to regularly meet, learn, and experiment with particular topics. They leverage experiential learning and a community- learning approach (Davis et al. 2012). The FFS are based on a combination of demonstration, explanation, and evaluation of new crop technologies, and field days and exchange visits to facilitate learning and sharing of innovation (Anandajayasekeram 2007). FFS interventions are often decentralized and implemented by local authorities, NGOs, community organizations, and even the private sector. Integrated Participatory Landscape Management Approaches— Achieving Multiple Goals The landscape management approach entails planning interventions at a larger geographic level, such as a watershed, catchment or communal area, and is planned through a participatory land management method. The landscape approach incorporates a set of tools, concepts, methods, and approaches deployed in a specific landscape to achieve several economic, social and environmental objectives and provide vital ecosystem services under the management of the people using the land and producing those services (Sayer et al. 2013). The theory is that CSA strategies—but also forestry management practices, improved livestock management, and rangeland and water management—should not be analyzed and implemented in isolation, but within a framework of broad landscape management. Despite being a multi-stakeholder process, landscape management approaches typically exclude the private sector. While there are opportunities for private sector involvement, for example, through sustainable investments and market and job opportunities, there are concerns about power relations or the derailment of the approach’s objectives. Cash Transfers—Alignment with the Harvesting Cycle to Promote CSA Cash transfers seek to promote agricultural production while combating poverty and hunger and strengthening future generations of human capital for (Boone et al. 2013). Access to capital, particularly for rural populations in extreme poverty, is largely nonexistent. Cash transfers provide infusions of capital as an incentive to invest in agriculture, and can be rolled out in combination with CSA practices. In tandem cash transfers and CSA address the bottleneck caused by high upfront costs of many CSA practices and also enhance the resilience of the rural poor. Evidence suggests that cash transfers can also increase agricultural production and self-sufficiency. In this way, cash transfers are believed to help households surmount the initial barrier toward higher socioeconomic mobility (Boone et al. 2013; Tesliuc 2013). Gender-Sensitive Supply Chains—Facilitating Access to Assets and Services for Women The development of gender-sensitive supply chains offers women access to the same markets and resources as men. A number of socioeconomic barriers prevent women from enjoying equal access to land, markets, educational resources, and networks. This inequality is exacerbated by the additional pressures women face in managing a workload that includes earning a livelihood, caring for children, and performing domestic duties. Case studies in Ghana and Uganda confirm the negative impacts of gender work imbalances on farming, particularly in the context of cash crop production. In Ghana, women’s access to liquid capital for purchasing inputs was limited, which led many women to use sub-optimal production technologies (Hill and Vigneri 2014). These constraints have to be considered when implementing CSA programs. PAGE 83 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN 7.3 Investment Requirements to Scale-up CSA Cost-benefit analyses are conducted for six investment options and show that investment in nearly all mechanisms exceeds the opportunity cost of capital, even without factoring in environmental benefits. The analysis assumes that each mechanism will be rolled out to 8 percent of smallholder households, thus approximately 118,000. Table 7.3 provides results for six mechanisms, where weather index insurance and integrated landscape management approach are available in two variations, and shows that nearly all of them have a favorable economic rate of return on investment, where farmer field schools have the highest rate of return with 39 percent, followed by integrated landscape management approach, 36 percent, and business partnerships with rural communities, 31 percent. The pluralistic participatory extension approach has the lowest economic internal rate of return (EIRR) at 4.4 percent. This approach aims to strengthen agricultural research and development, and benefits may not be immediately evident. Weather index insurance entails the lowest average investment cost per household, and participatory pluralistic extension services the investment cost over 6 years of project duration with US$51 million. The NPV of investment is positive for all enterprises except pluralistic participatory extension approach. Business partnerships with communities, farmer field schools, and the participatory, integrated landscape management approach seem most promising mechanisms. Considering economic indicators and effectiveness to address constraints to CSA adoption, these three investment options seem most promising. Business partnership with communities could facilitate access to markets, finance and capacity building, and demonstrate an economic internal rate of return on investment of 31 percent. Farmer field schools and integrated landscape management approach address two constraints but have relatively high rates of return of 39 percent and 36 percent. Business partnerships with communities, which aim at participating in a climate finance project, can be combined with integrated landscape management approaches. Figure 7.1 Economic internal rate of return and net present value of 7 delivery mechanisms, individual and aggregated 50% $71 $63.7 $61 39% 40% 36% $51 31% 30% $41 22% 21% $30.2 $31 19% 20% 20% $21 11% $13.3 $12.7 10% $9.6 $11 $7.9 4% $4.1 $0.6 $1 0% $9 -$10.6 -10% $19 Pluralistic Weather Weather Outgrower Integrated Business Integrated Farmers Aggregated paricipatory index index scheme landscape partnershiips landscape field schools (all delivery extension insurance insurance approach II with rural approach I mechanims) approach (with (without (focus on communities subsidy) subsidy) livestock) Source: World Bank, own elaboration Note: For each delivery mechanism economic indicators are calculated for approximatly 118,000 beneficiaries; aggregated this results in approximatly 826,000 beneficiaries PAGE 84 Box 7.1 Public sector expenditure on agriculture in 2017 CSA-related expenditures in 2017 have been assessed based on secondary data about public (IAPRI 2018) and donor (von der Decken 2017) expenditures, in combination with primary data collected from 25 different CSA donors and project implementers. Government expenditure: In 2017 the GoZ allocated approximately US$540 million to the Ministry of Agriculture and US$64 million to the Ministry of Fisheries and Livestock (IAPRI 2018), which together accounted for 9.4 percent of the total national budget. Seventy percent was allocated to FISP and FRA. The remainder, about 20 percent, or US$108 million, was available for alternative public investments, including in extension support, research, and development. Donor support for the agriculture sector in Zambia is substantial: As of 2017, donor partners have allocated more than US$1 billion to support agriculture sector projects. Donor disbursement by category: Donor disbursement was categorized according to the categories in Zambia National Agriculture Investment Plan, in which CSA did not represent a separate category. The largest share of donor allocations went to improving crop productivity (28 percent), followed by market access and service delivery (17 percent). Crop productivity improvements relate directly to the CSA goals. Knowledge support systems and research, food and nutrition security and risk management, and sustainable natural resource management are evenly split in terms of budget allocation at 13 percent each. Livestock and aquaculture productivity improvements receive a much lower budget allocation compared to allocations to crops, totaling 14 percent together. As of 2017, major donors in Zambia disbursed a total of nearly US$173 million to 80 agriculture and natural resource projects (von der Decken 2018). Funding for CSA practices: Based on interviews with 25 major donors and project implementers, including public sector entities, an estimated US$118 million was spent in 2017 on 38 different CSA- related projects, reaching a total of over 1.6 million beneficiaries (see Appendix F). Of all CSA practices, donors are most likely to fund initiatives relating to conservation agriculture and agroforestry practices (85 percent of donors); horticulture (60 percent); improved land preparation (55 percent); smallholder livestock (40 percent); drought and heat-tolerant seeds (40 percent); and agricultural liming (20 percent). Geographical distribution of CSA projects: Over 80 percent of the donors and implementers interviewed have funded or supported CSA-related projects in Eastern Province. Only 24 percent of donors and implementers operated projects in the northern Luapula, Northern, and Muchinga Provinces. Sources: IAPRI 2018; van der Decken 2017; interviews conducted by FAO in 2017 and 2018. Assuming all mechanisms are rolled out at the same time, to achieve a CSA adoption rate of more than 50 percent of smallholder farmers, the total cost would be US$196 million over six years, the average duration for a investment project, or an average investment cost of US$32.6 million per year. An average six-year project could achieve an NPV of US$63.6 million and an average economic rate of return of 20 percent (see Figure 7.1). Thus, the investment required to roll-out CSA is less compared to recent funding allocations by the GoZ and donors for the agriculture sector, or CSA in particular. Agriculture expenditures available for investment, excluding investment in the Farmer Inputs Support Program (FISP) and Food Reserve Agency (FRA), amounted to approximately US$108 million in 2017. In the same year, donor investment in agriculture and natural resource management amounted to US$173 million (see Box 7.1 for agriculture sector expenditures by the GoZ and donors). Thus for investment to be effective in reducing rural poverty and bolster more resilient agricultural systems not only the quantity of investment plays a role, but also the implementation quality and effectiveness. PAGE 85 Section 8 Conclusion Climate change and related extreme events are considered as an acute and growing challenge. The CSAIP aims to support GoZ in the process of policy and strategy development to operationalize climate commitments toward a productive, resilient, and low-emissions agriculture sector. Agriculture is viewed as the backbone of the economy and as a sector that can support inclusive structural transformation, poverty reduction, and diversification of the economy away from copper. Climate change, however, is expected to exacerbate the vulnerability of economically marginalized farmer households and slow the growth of the sector. At the same time, the LULUCF sector and agriculture sector together account for approximately 94 percent of the country’s carbon footprint. Taken this into account, the CSAIP aims to identify and prioritize key investments and policy actions and to build capacity to operationalize country climate commitments toward a productive, resilient, and low-emissions agriculture sector. Zambia has ambitious goals and targets for its agriculture sector. At the same time, its goals are conflicting with respect to the question whether further forest land can be converted for agricultural purposes. This calls for further policy harmonization to achieve agriculture sector goals. The vision of the existing agricultural policy frameworks has, as its core, the aim of doubling productivity, while decreasing the country’s current overreliance on maize, pursuing diversification, and boosting production and trade, which are expected to bolster food and nutrition security. The assessment showed that at least two strategy documents are in conflict with respect to land use: Zambia’s Long- Term Vision for 2030 calls for land under cultivation to expand by 0.9 million ha, whereas the National Policy on Environment incorporates a target of sustainably intensifying land use without converting additional land area into agriculture. To ensure a sustainable achievement of a productive, resilient and low-emission sector, policy harmonization and coordinating activities will be important. For rural households, CSA has a mostly positive long-term impact. In the short term, farmers face obstacles in the form of high upfront and production costs, and will require financial and other support during the early stages if they are to adopt CSA practices. Of the 10 CSA practices (minimum soil disturbance, residue retention, crop rotation, agroforestry, crop diversification, livestock PAGE 87 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN diversification, commercial horticulture, drought-tolerant seeds, delayed planting, and reducing post- harvest loss), four had a positive impact on household income, food availability, reduction of poverty, and income variability, which is a proxy for household resilience. In the short term, households incur losses due to higher production costs (associated with the need to mechanize, increased labor time), particularly following the adoption of: minimum soil disturbance, residue retention and agroforestry. Interestingly, it is specifically these practices—as well as crop rotation—that can potentially sequester carbon in the soil and can be effective in mitigating climate change. Lastly, optimizing the timing of planting has tremendous potential to enhance household welfare. These findings point towards the need to support farmers in the early stages of adoption by providing: timely access to inputs and supporting the development of markets for mechanization; providing services and finance by aligning subsidies and cash transfer programs with timing of planting; and enhancing capacity building to reduce risk of failure. Agroecological conditions must be factored in when promoting specific CSA practices, some of which perform better under dry than wet conditions. Practices amenable to dry conditions offer significant potential impact in the event of a drier future climate. A few practices—minimum soil disturbance, residue retention, small-scale horticulture—performed worse under wetter conditions (for instance, AEZ III) than under dry conditions (for example, AEZ I, IIa, IIb), possibly due to increased weed pressure, waterlogging, and lack of drainage. The net incremental benefit to households in Zambia’s wetter regions of adopting minimum soil disturbance are notably lower than in drier regions, and for some crops even negative. However, practices such as drought-tolerant seeds, agroforestry and crop diversification show good results under extreme dry and wet conditions, and are suitable for climate adaptation and household resilience building. In testing and developing strategies to implement specific CSA practices, pluralistic participatory extension approaches, including enrolling farmers in participatory trials, should be considered. Other critical strategies and initiatives include: improved and timely agro-weather advisory services, and customized capacity-building support to farmers to optimize planting times and minimize the risks associated with adopting CSA. At the sectoral level, CSA adoption by between 25 percent and 80 percent of farmers would enhance the likelihood of achieving, and even surpassing, the agriculture sector vision’s 2050 targets for crop production, food availability, and trade. Projections through 2050 are driven by increased population, GDP, and increased food demand in Zambia, which are expected to lead to a doubling of crop production. The adoption of CSA practices, particularly reducing post-harvest loss, is expected to further increase production by up to 20 percent compared to conventional practices, and CSA’s benefits are expected to be more pronounced under climate change projections. With respect to food availability, conventional agricultural practices are on track to meet the lower limit of national caloric requirements by 2050. With CSA, food availability is expected to increase still further, even under climate change. With respect to trade, the target of doubling net exports by 2050 is feasible in terms of maize and millet, and under extreme climate change scenarios, for cassava. Adopting practices to reduce post-harvest loss and implement conservation agriculture and crop diversification practices would increase Zambia’s international competitiveness and act as a driving force for increased net exports in the future. Zambia has the potential to emerge as a net exporter of groundnuts if it pursues agricultural diversification policies, even under climate change projections. Doubling crop yields, which is a key policy goal of the agriculture sector, will not be achieved either through conventional agricultural practices or through CSA, although the latter will potentially narrow the gap to the desired yield level. Policies and strategies to enhance agricultural productivity are urgently needed. CSA practices with particular potential to narrow the yield gap by enhancing crop yields by 21 percent over conventional practices include reducing post-harvest loss, PAGE 88 minimum soil disturbance, and crop diversification. Under climate change, CSA practices have a varying effect depending on the crop and practice. For maize and cassava, for example, minimum soil disturbance promises smaller yield increases; whereas producing maize with drought-tolerant seeds and conservation agriculture or soybeans through crop diversification promise better yields under climate change than with conventional practices, enhancing an already positive yield effect. Productivity increases are an important component of an inclusive process of structural transformation of the agricultural sector. Efforts should be made to address the constraints of productivity growth through the following: targeted agriculture research and development in Zambia or in collaborative research partnerships across the regions; pluralistic participatory extension to support farmers in optimizing their farming practices; and linking farmers to business partnerships such as outgrower schemes; and making production inputs and irrigation and other infrastructure available in a timely manner. Since crop yield increases are unlikely to meet sector goals, there is a risk that production increases stem from the conversion of forest and savannah into cropland to achieve 2050 crop production targets. CSA practices are expected to curb the necessity of land conversion, but only marginally. Through 2050, considerable land use change is projected, with cropland expected to increase by almost 917,000 ha, and grassland by approximately 1.4 million ha, mostly at the expense of forest area in the productive southern and southwestern regions that form Zambia’s AEZ I and AEZ IIa. If so, the agriculture sector vision goal of converting 900,000 ha will be achieved, but the goals of the National Policy on Environment of avoiding land conversions will not be reached. The projected impact of climate change on land conversion is small. Individual CSA practices have the potential to reduce cropland land conversion by between 32,000 ha (through conservation agriculture) and 112,000 ha (reducing post-harvest loss). This reduction in cropland is accompanied by an expansion in grassland driven by an increase in livestock production in the same order of magnitude, so that deforestation is hardly halted. These trends illustrate Jevon’s paradox, which states that productivity increases per hectare provide further incentive to expand agricultural land (Alcott 2005). Trends in land use change put Zambia at high risk of failure to meet its NDC goals of reducing GHG emissions by 25–47 percent by 2030 or 2050, as emissions from agriculture (including crop and livestock) and LULUCF are projected to increase by 57–487 percent in 2050. The impact on emissions of CSA adoption is mixed. Minimum soil disturbance and reducing post-harvest loss can reduce emissions associated with fertilizer use, while residue retention has the opposite effect. There is a negligible average change in overall emissions compared to conventional practices. CSA adoption could reduce emission from land use changes, but only by on average -0.32 percent compared to conventional practices. On the positive side, some CSA practices are expected to enhance soil carbon sequestration, which could be according to a simple calculation example 7 times more effective in reducing tCO2e emissions than reducing use of nitrate fertilizers. However, data on soil carbon sequestration must be interpreted with caution, since they are based on highly aggregated emission coefficients, and lacks historical and site-specific land use and soil data. Introducing a low carbon tax on land conversion proves to be an effective policy strategy to achieve mitigation goals and could effectively halt agriculture-related deforestation. However, reducing biomass burning seems the most promising strategy. A carbon tax of US$10/tCO2e emissions could reduce emissions from deforestation and other land use change by 99 percent through 2050 compared to a business-as-usual scenario. At the very low price of US$5/tCO2e, land conversion would still be profitable. Production and food availability would slightly decrease, by approximately 2 percent, crop yield would slightly increase, and maize imports, for example, are projected to increase by up to 65 percent in certain carbon tax scenarios. While these policies seem promising, actual PAGE 89 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN enforcement would be difficult. In conclusion, while very strict carbon taxes can reduce emissions associated with land conversion and CSA can advance achievement of soil carbon sinks, both of which contribute to achievement of Zambia’s NDC targets, it is worth noting that emissions from agriculture and land conversion constitute only 12 percent of total emissions in Zambia today. The largest share originates from the burning of biomass, which has not been modelled in this analysis. If Zambia wishes to reduce its emission significantly, reducing biomass burning and charcoal use must become a policy priority. While the effect of CSA adoption on agricultural land conversion is small in terms of hectares and GHG emissions, an economic analysis shows its considerable value to public good provision for global society. If reductions in GHG emissions are valued at a shadow price of carbon, the NPV of a large-scale CSA intervention would increase by between 171 percent and 341 percent, or from US$170 million if private benefits are considered, to between US$460 to 750 million. Even a relatively small increase under a low market price for carbon represents an economic value of US$34 million, which would justify exploring the use of carbon finance to support CSA adoption and compensate farmers for initial losses associated with its adoption. The four most promising CSA practices are crop diversification, commercial horticulture, agroforestry, and reducing post-harvest loss. However, adoption is often constrained by poor access to finance and market access and inadequate levels of capacity building and skill development. Despite donor support and public sector involvement, adoption of these four practices remains low. Promising CSA strategies were identified based on a ranking across household and sectoral indicators. Other practices, while also promising, have larger trade-offs. Livestock diversification, for instance, offers great benefits in terms of household resilience and food security and should be explored, but may also require expansion of grassland at the expense of forests and will increase methane emission levels. Some of the obstacles that will need to be addressed to spur CSA adoption are: low knowledge and capacity to adopt CSA practices; poor access to finance to overcome upfront investment and labor hire constraints; inadequate access to market infrastructure and input and output markets. Public resources should be allocated to crowd-in private sector finance and maximize finance to achieve agriculture sector goals and the sector’s potential to contribute to Zambia’s structural transformation. CSA has significant potential but there is still room for improvement, for instance with respect to increasing yields, reducing land use change, and achieving other development goals including: inclusive employment, improved nutrition, and reducing incident and rural poverty. A complicated question arises: what type of public investment will address the adoption constraints of CSA and can promote private sector investment from farmers, traders, agribusinesses, and processors or financing institutes, all of which can take advantage of Zambia’s agriculture opportunities? Eight innovative and proven mechanisms are presented that can enhance CSA adoption, and public actions have been identified that can spur private sector involvement. Of eight innovative mechanisms, establishing business partnerships with rural communities, participatory, integrated landscape management approaches, and farmer field schools have the potential to address critical enabling environment constraints and achieve high rates of return on investment. The analysis assessed attributes of several mechanisms, their expected economic benefits, and public sector interventions needed to support their establishment. Supporting business partnerships with rural communities and outgrower schemes for commercial horticulture production seem most promising for overcoming enabling environment constraints. Considering economic indicators and effectiveness to address constraints to CSA adoption, business partnership with PAGE 90 Box 8.1 Policy recommendations The analyses of CSA practices under various scenarios and assessment of mechanisms to support CSA adoption show that business partnerships with rural communities which build on environmental sustainability as business strategy, farmer field schools to enhance community-based learning and technology dissemination, and participatory integrated landscape management approaches seem promising and profitable mechanisms to support the development of a productive, resilient, and low-emission agriculture sector. These can be supported by several policy actions: 1 Support farmers with improved access to inputs and finance, specifically in early stages of CSA adoption until benefits start to be realized. Support the development of markets for mechanization, innovative ICT-based solutions including agro-weather services for timely decision-making, and access to improved production inputs and seed varieties suitable for varied agroecological zones. 2 Support agricultural research, regional collaboration in research, participatory testing of CSA technologies to enhance their potential to increase crop yields, across and especially wet agroecological conditions. Support the advancement in the development and multiplication of seed varieties that are appropriate for rainfed production systems, in particular in the legume seed sector and short-duration maize seeds. 3 To increase agribusiness participation in the sector, a range of policy actions seem promising: (i) support feasibility, as well as risks and vulnerability assessments, to identify entry points and challenges; this should include an assessment of risk- and cost-sharing mechanisms which provide an understanding where the incentives align between the public good element of CSA and private sector motive; (ii) seek dialogue with agribusinesses about resulting investment opportunities; (iii) review the legal and regulatory framework to identify reform requirements and strengthen the business environment; (iv) provide opportunities for human capital development, capacity building and extension services for agriculture sector actors. 4 Support operationalization of holistic landscape management approaches by harmonizing policies and supporting cross-ministerial collaboration across the agriculture, environment, water and energy sector, and across administrative boundaries. Landscape approaches include climate-smart crop, livestock and forest management and have the potential to reduce alarming rates of land conversion for agriculture as well as addressing issues of biomass burning and charcoal production, which are key for Zambia’s carbon footprint. 5 Pursue the positive agriculture sector reform path of converting FISP into an e-voucher program, which supports agricultural diversification and adoption of short-duration varieties targeted to Zambia’s agroecological zones. The analysis shows the benefits of crop diversification into legumes, but also the need to improve the timing of planting or make short-duration varieties available to enhance crop yields. 6 Further support development of market infrastructure such as rural storage facilities and enable greater private investment in storage. Private investment may occur where production levels are high and stable and where access to border markets is feasible. In places where these conditions are not met, public investment may be required. In addition, training and collective action on improved post-harvest management bears a high potential to achieve CSA triple-benefits. communities could facilitate access to markets, finance and capacity building, and demonstrate an economic internal rate of return on investment of 31 percent. Farmer field schools and integrated landscape management approach address two constraints but have relatively high rates of return of 39 percent and 36 percent, respectively. Specifically, participatory integrated landscape approach can achieve high return on investment and provide incentives to work across sectoral silos between agricultural, environment and water sector and support policy harmonization. These approaches should be considered in future government and donor programs. PAGE 91 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Box 8.2 Knowledge gaps and further research In the course of developing the CSAIP following areas were identified which need further attention: Institutional capacities can be strengthened using these methods: (i) Conduct an assessment of institutional capacities to better address bottlenecks in project implementation. Agriculture economic research is conducted but implementation success seems to be varied. (ii) Develop a sector-wide monitoring and evaluation framework to document success in promoting CSA and implementation of agriculture investment plans. (iii) Produce an inventory of projects, spending, beneficiaries reached, and outputs. A variety of quantitative assessments are recommended, as follows: (i) Climate change impact assessments, which are important to understand climate dynamics, and climate-economy interactions, are inherently uncertain and must be interpreted with caution. (ii) Research the impact of CSA practices on crop yield or GHG emissions in a spatially explicit manner across Zambia. Little experimental evidence is available across Zambia. Mitigation through soil carbon sequestration and related data needs remain an area for future research. (iii) Further research on biomass burning for charcoal—a major source of GHG emissions that critically affects the achievement of NDC targets. Further study on introducing improved cooking solutions in urban and rural households and country-wide impacts is needed. (iv) Develop a model that considers long-run anthropogenic climate change and includes climate variability, such as extreme weather events—floods, droughts, and hurricanes—which are expected have a significant negative impact on future economic and agricultural development in Sub-Saharan Africa (Ahmed et al. 2011; Thurlow et al. 2012). Modeling the effects of year-by-year climate variations requires the use of a stochastic modelling approach, which is currently under development. Almost all delivery mechanisms show positive economic rates of return; rolling them out to more than 50 percent, of smallholder farmers, would require an annual public sector investment of US$32.6 million over six years. If rolled out to 118,000 beneficiaries, for example, expected rates of return on investment range from 11 percent (weather index insurance) to 39 percent (farmer field school). A roll out of all mechanisms at the same time would yield a rate of return on investment of 20 percent and an economic net present value of US$63.6 million. The proposed annual investment of US$32.6 million over 6 years is less than recent annual GoZ and donor funding for the agriculture sector, which was estimated around US$108 million and US$118 million, respectively, in 2017. While investment quantity is important, implementation quality and effectiveness have to be ensured to see desired results for a climate-resilient agriculture sector and sustained poverty reduction among smallholders in rural areas. The report concludes with eight recommendations, which are ranked by feasibility and relevance (Box 8.1), as well as knowledge gaps (Box 8.2) for which further research is needed. PAGE 92 References Affognon, H., C. Mutungi, P. Sanginga, and C. Borgemeister. 2015. “Unpacking Postharvest Losses in Sub-Saharan Africa: A Meta- Analysis.” World Development 66: 49–68. doi:10.1016/J.WORLDDEV.2014.08.002 Ahmed, S. A., N. S. Diffenbaugh, T. W. Hertel, D. B. Lobell, N. Ramankutty, A. R. Rios, and P Rowhani. 2011. “Climate volatility and poverty vulnerability in Tanzania.” Global Environmental Change 21 (1): 46–55. doi:10.1016/j.gloenvcha.2010.10.003 Arslan, A., N. McCarthy, L. Lipper, S. Asfaw, and A. Cattaneo. 2014. “Adoption and intensity of adoption of conservation farming practices in Zambia.” Agriculture, Ecosystems & Environment 187 (2014): 72-86 Arslan, A., N. McCarthy, L. Lipper, S. Asfaw, A. Cattaneo, and M. Kokwe. 2015. “Climate smart agriculture? Assessing the adaptation implications in Zambia.” Journal of Agricultural Economics 66 (3): 753-780 Arslan, A., R. Cavatassi, F. Alfani, N. Mccarthy, L. Lipper and M. Kokwe. 2018. “Diversification under climate variability as part of a CSA strategy in rural Zambia.” The Journal of Development Studies 54(3): 457-480 Ajayi, O. C., F. K. Akinnifesi, G. Sileshi, and W. Kanjipite. 2010. “Labour inputs and financial profitability of conventional and agroforestry- based soil fertility management practices in Zambia.” Agrekon 48 (3): 276–92. doi:10.1080/03031853.2009.9523827 Alcott, B.. 2005. “Jevons’ paradox.” Ecological Economics 54 (1): 9–21. doi:10.1016/J.ECOLECON.2005.03.020 Alexandratos, N. and J. Bruinsma. 2012. World agriculture: Towards 2030/2050: The 2012 Revision. ESA Working Paper No. 12-03. Rome: Food and Agriculture Organization of the United Nations (FAO). doi:10.1016/S0264-8377(03)00047-4 Alexandratos, N., J. Bruinsma, G. Bödeker, J. Schmidhuber, S. Broca, P. Shetty, and M. G. Ottaviani. 2006. World agriculture: Towards 2030/2050: Interim report. Rome: Food and Agriculture Organization of the United Nations (FAO). doi:10.1016/S0264-8377(03)00047-4 Anandajayasekeram, P., K. Davis, and S. Workneh. 2007. “Farmer Field Schools: An Alternative to Existing Extension Systems? Experience from Eastern and Southern Africa.” Journal of International Agricultural and Extension Education 14 (1): 81-93. Balkovič, J., M. van der Velde, E. Schmid, R. Skalský, N. Khabarov, M. Obersteiner, B. Stürmer, and W. Xiong. 2013. “Pan-European crop modelling with EPIC: Implementation, up-scaling and regional crop yield validation.” Agricultural Systems 120: 61–75. doi:10.1016/J. AGSY.2013.05.008 Bang, H. and J. M. Robins. 2005. “Doubly Robust Estimation in Missing Data and Causal Inference Models” Biometrics 61(4):962-73. doi:10.1111/j.1541-0420.2005.00377.x Baudron, F., P.T., M. Corbeel, P. Letourmy, and K. E. Giller. 2012. “Comparative performance of conservation agriculture and current smallholder farming practices in semi-arid Zimbabwe.” Field Crops Research 132: 117–128. doi:10.1016/j.fcr.2011.09.008. Beach, R. H., J. Creason, S. Bushey Ohrel, S. Ragnauth, S. Ogle, C. Li, P. Ingraham, and W. Salas. 2015. “Global mitigation potential and costs of reducing agricultural non-CO2 greenhouse gas emissions through 2030.” Journal of Integrative Environmental Sciences 12 (sup1): 87–105. doi:10.1080/1943815X.2015.1110183. Birner, R., Davis, K., Pender, J., Nkonya, E., Anandajayasekeram, P., Ekboir, J., ... & Cohen, M. (2009). From best practice to best fit: a framework for designing and analyzing pluralistic agricultural advisory services worldwide. Journal of agricultural education and extension, 15(4), 341-355. Boone, R., K. Covarribias, B. Davis, P. Winter. 2013. “Cash transfer programs and agricultural production: the case of Malawi.” Agricultural Economics. 44 (3): 365-378. doi:10.1111/agec.12017. Bouwman, A. F., K. W. Van Der Hoek, B. Eickhout, and I. Soenario. 2005. “Exploring changes in world ruminant production systems.” Agricultural Systems 84 (2): 121–53. doi:10.1016/j.agsy.2004.05.006 Braimoh, A., A. Mwanakasale, A. Chapoto, R. Rubaiza, B. Chisanga, N. Mubanga, P. Samboko, A. Giertz, and G. Obuya. 2018. “Increasing Agricultural Resilience through Better Risk Management in Zambia.” Washington DC: World Bank. Branca, G., A. Arslan, D. Banda, R. Cavatassi, U. Grewer and A. Paolantonio. 2015. The costs and the benefits of Climate-smart agriculture in Zambia. Final report. Economics and Policy Innovations for Climate-Smart Agriculture Programme. Rome: Food and Agriculture Organization of the United Nations (FAO) Burke, W., T. S. Jayne and R. Black. 2017. “Factors explaining the low and variable profitability of fertilizer application in Zambia.” Agricultural Economics 48(1):115-126. doi:10.1111/agec.12299 Bwalya, R. and T. Kalinda. 2014. “An Analysis of the Value Chain for Indigenous Chickens in Zambia’s Lusaka and Central Provinces.” Journal of Agricultural Studies 2(2): 32-51. doi:10.5296/jas.v2i2.5918 Cairns, J. E., Hellin, J., Sonder, K., Araus, J. L., MacRobert, J. F., Thierfelder, C., & Prasanna, B. M. (2013). Adapting maize production to climate change in sub-Saharan Africa. Food Security, 5(3), 345-360. Carter, M., A. de Janvry, E. Sadoulet, and A. Sarris. 2014. “Index-based weather insurance for developing countries: A review of evidence and a set of propositions for up-scaling.” Working Papers 111. Paris: Fondation pour lesetudes et recherches sur le développement international (FERDI) Cardinael, R., V., Umulisa, A., Toudert, A. Olivier, L. Bockel, and M. Bernoux. 2018 “Revisiting IPCC Tier 1 coefficients for soil organic and biomass carbon storage in agroforestry systems.” Environmental Research Letters 13(12). doi.org/10.1088/1748-9326/aaeb5f. CIAT (International Center for Tropical Agriculture) and World Bank. 2017. “Climate-Smart Agriculture in Zambia. CSA Country Profiles for Africa Series.” Washington DC: CIAT and World Bank. Chapoto, A., Haggblade, S., Hichaambwa, M., Kabwe, S., Longabaugh, S., Sitko, N., & Tschirley, D. L. (2012). Agricultural Transformation in Zambia: Alternative Institutional Models for Accelerating Agricultural Productivity Growth, and Commercialization (No. 132339). Michigan State University, Department of Agricultural, Food, and Resource Economics. Chapoto, A. and B. Chisanga. 2016. “Zambia Agriculture Status Report 2016.” Lusaka: Indaba Agricultural Policy Research Institute (IAPRI). Chapoto, A. and O. Zulu-Mbata. 2016. Rural Agriculture Livelihoods Survey (RALS) Survey Report 2015. Lusaka: Indaba Agricultural Policy Research Institute (IAPRI) Chisanga, B. and O. Zulu-Mbata. 2018. “The changing food expenditure patterns and trends in Zambia: implications for agricultural policies,” Food Security: The Science, Sociology and Economics of Food Production and Access to Food 10(3): 721-740 Chisanga, B. and A. Chapoto. 2018. “Grain Marketing Innovations and Investments in Zambia: Creating Marketing Opportunities for Smallholder Farmers.” Working Paper 135, Indaba Agricultural Policy Research Institute, Lusaka, Zambia PAGE 93 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Corsi, S., F., T., Kassam, A., Pisante, M., de Moraes Sà, J. 2012. “Soil organic carbon accumulation and greenhouse gas emission reductions from Conservation Agriculture: A literature review.” Integrated Crop Management 16:89. Rome: Food and Agriculture Organization of the United Nations (FAO). CSO (Central Statistics Office, Republic of Zambia). 2011. Zambia Living Conditions Monitoring Survey (LCMS) 2010. Lusaka. CSO (Central Statistics Office) and Ministry of Agriculture, Republic of Zambia; Michigan State University, Indaba Agricultural Policy Research Institute (IAPRI). 2015. 2015 Rural Agricultural Livelihoods Survey (RALS 2015). Lusaka, Zambia. CSO (Central Statistics Office) and Ministry of Agriculture, Republic of Zambia; Michigan State University, Indaba Agricultural Policy Research Institute (IAPRI). 2012. 2012 Rural Agricultural Livelihoods Survey (RALS 2012). Lusaka, Zambia. CSO (Central Statistics Office, Republic of Zambia). 2016. Zambia Living Conditions Monitoring Survey (LCMS) 2015. Lusaka. CSO (Central Statistics Office, Republic of Zambia). 2018. Zambia Data Portal. Crop Forecast Survey 2015. http://zambia. opendataforafrica.org [last accessed December 2018] D’Emden, F. H., R. S. Llewellyn, and M. P. Burton. 2008. “Factors influencing adoption of conservation tillage in Australian cropping regions.” Australian Journal of Agricultural and Resource Economics 52(2): 169-82. doi:10.1111/j.1467-8489.2008.00409.x Davis, K., E. Nkonya, E. Kato, D. A. Mekonnen, M. Odendo, R. Miiro, and J. Nkuba. 2010. “Impact of Farmer Field Schools on Agricultural Productivity and Poverty in East Africa.” World Development 40 (2): 402-413. doi:10.1016/j.worlddev.2011.05.019 Deininger, K. and G. Feder. 2009. “Land registration, governance, and development: Evidence and implications for policy.” World Bank Research Observer. doi:10.1093/wbro/lkp007 Dellink, R., J. Chateau, E. Lanzi, and B. Magné. 2017. “Long-term economic growth projections in the Shared Socioeconomic Pathways.” Global Environmental Change 42:200-214. doi:10.1016/j.gloenvcha.2015.06.004 Deppermann, A., J. Balkovic, S. Bundle, F. di Fulvio, P. Havlík, D. Leclère, M. Lesiv, A. Prishchepov, and D. Schepaschenko. 2018. “Increasing crop production in Russia and Ukraine – regional and global impacts from intensification and recultivation.” Environmental Research Letters 13 (2):025008. doi:10.1088/1748-9326/aaa4a4 Dercon, Stefan and Hoddinott, John and Woldehanna, Tassew, Growth, Poverty and Chronic Poverty in Rural Ethiopia: Evidence from 15 Communities 1994-2004 (May 2008). Available at SSRN: https://ssrn.com/abstract=1259333 or http://dx.doi.org/10.2139/ssrn.1259333 De Vente, J., M. S. Reed, L. C. Stringer, S. Valente, and J. Newig. 2016. “How does the context and design of participatory decision making processes affect their outcomes? Evidence from sustainable land management in global drylands.” Ecology and Society 21(2):24. doi:10.5751/ES-08053-210224 Dolan, C. and J. Humphrey. 2000. “Governance and Trade in Fresh Vegetables: The Impact of UK Supermarkets on the African Horticulture Industry.” Journal of Development Studies 37(2):147-176. doi:10.1080/713600072 Dyer, J., L. C. Stringer, A. J Dougill, J. Leventon, M. Nshimbi, F. Chama, J. I. Muledi et al. 2013. “Assessing participatory practices in community-based natural resource management: experiences in community engagement from southern Africa.” Journal of Environmental Management 137:137-145. doi:10.1016/j.jenvman.2013.11.057 Elbehri, A.. 2013. “Rebuilding West Africa’s food potential: policies and market incentives for smallholder-inclusive food value chains.” Rome: FAO/IFAD Erenstein, O., K. Sayre, P. Wall, J. Hellin, and J. Dixon. 2012. “Conservation Agriculture in Maize- and Wheat-Based Systems in the (Sub) tropics: Lessons from Adaptation Initiatives in South Asia, Mexico, and Southern Africa.” Journal of Sustainable Agriculture 36 (2): 180– 206. doi:10.1080/10440046.2011.620230 Estes, L. D., T. Searchinger, M. Spiegel, D. Tian, S. Sichinga, M. Mwale, L. Kehoe, T. Kuemmerle, A. Berv , N. Chaney, J. Sheffield, E. F. Wood, and K. K. Caylor. 2016. “Reconciling agriculture, carbon and biodiversity in a savannah transformation frontier.” Philosophical Transactions of the Royal Society of London, Series B, 371(1703). doi:1098/rstb.2015.0316 FAO (Food and Agriculture Organization of the United Nations). 2016. “Ex-Ante carbon-Balance Tool (Ex-ACT). Technical Guidelines for Version 7.” EASYPol Module 218. Rome, Italy. FAO (Food and Agriculture Organization of the United Nations). 2011. “Global food losses and food waste.” Rome, Italy: FAO. FAO (Food and Agriculture Organization of the United Nations). 2013. “Climate-Smart Agriculture Sourcebook.” Rome, Italy. FAO (Food and Agriculture Organization of the United Nations). 2014. “The Economic Impacts of Cash Transfer Programmes in Sub- Saharan Africa.” Policy Brief. Rome, Italy. FAO (Food and Agriculture Organization of the United Nations). 2015. “Climate change mitigation and smallholder agriculture in Zambia.” EPIC - Economics and Policy Innovations for Climate-Smart Agriculture POLICY BRIEF No.3, Rome, Italy. FAOSTAT 2017 and 2018, http://www.fao.org/faostat/en/#home [accessed December 2018] Felgenhauer, K. and D. Wolter. 2009. “Outgrower Schemes: Why Big Multinationals Link up with African Smallholders.” Paris: Organization for Economic Development (OECD). Fessehaie, J., R. Das Nair, P. Ncube, and S. Roberts. 2015. “Growth Promotion Through Industrial Strategies–A Study of Zambia.” International Growth Centre (IGC) Working Paper series. Fisher, M., T. Abate, R. W. Lunduka, W. Asnake, Y. Alemayehu, and R. B. Madulu. 2015. “Drought tolerant maize for farmer adaptation to drought in sub-Saharan Africa: Determinants of adoption in eastern and southern Africa.” Climatic Change 133 (2): 283–99. doi:10.1007/ s10584-015-1459-2 Fowler, R. and J. Rockstrom. 2001. “Conservation Tillage for Sustainable Agriculture—An Agrarian Revolution Gathers Momentum in Africa.” Soil and Tillage Research 61 (1-2):93-107. doi:10.1016/S0167-1987(01)00181-7 Frank, S., H. Böttcher, P. Havlík, H. Valin, A. Mosnier, M. Obersteiner, E. Schmid, and B. Elbersen. 2013. “How effective are the sustainability criteria accompanying the European Union 2020 biofuel targets?” GCB Bioenergy 5 (3): 306–14. doi:10.1111/j.1757- 1707.2012.01188.x Frank, S., P. Havlík, J.F., Soussana, A. Levesque, H. Valin, E. Wollenberg, U. Kleinwechter, et al. 2017. “Reducing greenhouse gas emissions in agriculture without compromising food security?” Environmental Research Letters 12 (10). doi:10.1088/1748-9326/aa8c83 Frank, S., R. Beach, P. Havlík, H. Valin, M. Herrero, A. Mosnier, T. Hasegawa, J. Creason, S. Ragnauth, and M. Obersteiner. 2018. “Structural change as a key component for agricultural non-CO2 mitigation efforts.” Nature Communications 9 (1): 1060. doi:10.1038/s41467-018- 03489-1 Fricko, O., P. Havlík, J. Rogelj, Z. Klimont, M. Gusti, N. Johnson, P. Kolp, et al. 2017. “The marker quantification of the Shared Socioeconomic Pathway 2: A middle-of-the-road scenario for the 21st century.” Global Environmental Change 42: 251–67. doi:10.1016/J. GLOENVCHA.2016.06.004 PAGE 94 Gibbs, H. K., M. Johnston, J. A. Foley, T. Holloway, C. Monfreda, N. Ramankutty, and D. Zaks. 2008. “Carbon payback times for crop-based biofuel expansion in the tropics: the effects of changing yield and technology.” Environmental Research Letters 3 (3). doi:10.1088/1748- 9326/3/3/034001 Giller, K. E., E. Witter, M. Corbeels, and P. Tittonell. 2009. “Conservation agriculture and smallholder farming in Africa: The heretics’ view.” Field Crops Research 114 (1): 23–34. doi:10.1016/J.FCR.2009.06.017 Government of the Republic of Zambia (GoZ). 2007a. “National Policy on the Environment (NPE).” Lusaka, Zambia Government of he Republic of Zambia (GoZ). 2007b. “National Adaptation Programme Of Action On Climate Change (NAPA).” Lusaka, Zambia Government of the Republic of Zambia (GoZ). 2010. “National Climate Change Response Strategy (NCCRS)” Lusaka, Zambia Government of the Republic of Zambia (GoZ). 2015. “Zambia’s (Intended) Nationally Determined Contribution ((I)NDC)) to the 2015 Agreement On Climate Change.” Lusaka, Zambia Government of the Republic of Zambia (GoZ). 2016. “Second National Agricultural Policy (SNAP).” Lusaka, Zambia Government of the Republic of Zambia (GoZ). 2017a. “National Climate Change Policy.” Lusaka, Zambia Government of the Republic of Zambia (GoZ). 2017b. “Seventh National Development Plan 2017-2021 (7NDP). Accelerating Development Efforts towards Vision 2030 without Leaving Anyone Behind.” Lusaka, Zambia Government of Zambia (GoZ). 2017c. Zambia Integrated Forest Landscape Project Resettlement Policy Framework. Interim Inter- Ministerial Climate Change Secretariat. Lusaka: Zambia Ministry of National Development Planning Goyal, A. and C. González-Velosa. 2012. “Improving Agricultural Productivity and Market Efficiency in Latin America and the Caribbean: How ICTs Can Make a Difference?” LCSSD occasional paper series on food prices. Washington DC: World Bank Haggblade, S., and G. Tembo. 2003. “Conservation farming in Zambia.” Environment and production technology division (EPTD) Discussion Paper. Washington DC: International Food Policy Research Institute (IFPRI) Harman, L. and A. Chapoto. 2017. “FISP and FRA Reforms: Investing savings into a package of smart social protections schemes for Zambia.” Lusaka, Zambia: Indaba Agricultural Policy Research Institute. Havlík, P., U. A. Schneider, E. Schmid, H. Böttcher, S. Fritz, R. Skalský, K. Aoki, et al. 2011. “Global land-use implications of first and second generation biofuel targets.” Energy Policy 39 (10): 5690–5702. doi:10.1016/J.ENPOL.2010.03.030. Havlík, P., H. Valin, M. Herrero, M. Obersteiner, E. Schmid, M. C Rufino, A. Mosnier, et al. 2014. “Climate change mitigation through livestock system transitions.” Proceedings of the National Academy of Sciences of the United States of America 111 (10): 3709–14. doi:10.1073/pnas.1308044111. Havlík, P., D. Leclère, H. Valin, M. Herrero, E. Schmid, J. Soussana, C. Müller, M. Obersteiner. 2015. “Global climate change, food supply and livestock production systems: A bioeconomic analysis.” In Climate change and food systems: global assessments and implications for food security and trade, edited by A. Elbehri, 176–208. Rome: Food and Agriculture Organization of the United Nations (FAO). Heemskerk, W. and K. Davis. 2012. “Pluralistic Extension System” Module 3, Thematic Note I. in: World Bank. (eds). Agricultural Innovation Systems: An Investment Sourcebook. Agricultural and Rural Development. Washington DC: World Bank Herrero, M., P. Havlík, J. M. McIntire, A. Palazzo, and H. Valin. 2014. African Livestock Futures: Realizing the Potential of Livestock for Food Security, Poverty Reduction and the Environment in Sub-Saharan Africa. Geneva: Office of the Special Representative of the UN Secretary General for Food Security and Nutrition; the United Nations System Influenza Coordination. Herrero, M., P. Havlík, H. Valin, A. Notenbaert, M. C. Rufino, P. K. Thornton, M. Blümmel, F. Weiss, D. Grace, and M. Obersteiner. 2013. “Biomass use, production, feed efficiencies, and greenhouse gas emissions from global livestock systems.” Proceedings of the National Academy of Sciences of the United States of America 110 (52). National Academy of Sciences: 20888–93. doi:10.1073/pnas.1308149110. Hichaambwa, M., M. Beaver, A. Chapoto, and M. Weber. 2009. “Patterns of Urban Food Consumption and Expenditure in Zambia: An Overview Report Based on the CSO/MACO/FSRP Food Consumption Survey in Urban Areas of Lusaka, Kitwe, Mansa, and Kasama, 2007- 2008.” Working Paper No. 56802. Michigan State University, Department of Agricultural, Food, and Resource Economics, East Lansing. Hichaambwa, M., J. Chamberlin, and S. Kabwe. 2015. Is Smallholder Horticulture the Unfunded Poverty Reduction Option in Zambia? A Comparative Assessment of Welfare Effects of Participation in Horticultural and Maize Market. Working Paper No 96. Lusaka: Indaba Agricultural Policy Research Institute (IAPRI). Hirano, K. and G. Imbens. 2005. “The Propensity Score with Continuous Treatments” Applied Bayesian Modeling and Causal Inference from Incomplete-Data Perspectives: An Essential Journey with Donald Rubin’s Statistical Family, Wiley/Blackwell(10.1002):73-84. doi:10.1002/0470090456.ch7. Hirano, K., G. Imbens, and G. Ridder. 2003. “Efficient Estimation of Average Treatment Effects Using the Estimated Propensity Score.” Econometrica 71(4):1161-1189. doi:10.1111/1468-0262.00442 Hill, R. V. and M. Vigneri. 2014. “Mainstreaming gender sensitivity in cash crop market supply chains.” In Gender in Agriculture, pp. 315- 341 Huntington H., A. Starosta, and S. Fenner. 2018. “Impact Evaluation of Community-Based Forest Management Program in Zambia.” Washington DC: United States Agency for International Development IAPRI (2018) Analysis of the 2018 Agricultural Sector Budget. Presentation at the Annual Budget Breakfast Meeting 10th October, 2017, Southern Sun Hotel, Lusaka. Online at: http://www.iapri.org.zm/images/Presentations/Analysis_of_the_2018_Agricultural_Budget_final. pdf Ignaciuk, A., Sitko, N., Scognamillo, A., and Alfani, F. and Kozlowska, K. .2018. “Is crop diversification a panacea for climate resilience in Africa? Welfare implications for heterogeneous households” FAO Agricultural Development Economics Policy Brief 2. Imbens, G. .2000. “The role of the propensity score in estimating dose-response functions” Biometrika 87 (3):706-710. doi:10.1093/ biomet/87.3.706 Indaba Agriculutre policy research Institute (IAPRI) .2018. Analysis of the 2018 Agricultural Sector Budget. Presentation at the Annual Budget Breakfast Meeting 10th October, 2017, Southern Sun Hotel, Lusaka. Online at: http://www.iapri.org.zm/images/Presentations/ Analysis_of_the_2018_Agricultural_Budget_final.pdf IPCC (Intergovernmental Panel on Climate Change). 2006. IPCC Guidelines for National Greenhouse Gas Inventories. Prepared by the National Greenhouse Gas Inventories Programme H. Eggleston, Buendia L, Miwa K, Ngara T, Tanabe K. (eds.), IGES, Japan. Jain, S.. 2007. An empirical economic assessment of impacts of climate change on agriculture in Zambia. Policy Research Working Papers. World Bank. doi:10.1596/1813-9450-4291. Kaminski, J. and L. Christiaensen. 2014. “Post-harvest loss in sub-Saharan Africa—what do farmers say?” Global Food Security 3 (3-4): 149–58. doi:10.1016/J.GFS.2014.10.002. PAGE 95 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Kanyanga, J., T. S Thomas, S. Hachigonta, and L. M. Sibanda. 2013. “Zambia.” In Southern African Agriculture and Climate Change, edited by Sepo Hachigonta, G.C. Nelson, T. S. Thomas, and L.M. Sibanda, 255–87. Washington, DC: International Food Policy Research Institute (IFPRI). Kelly, V.. 2005. “Farmers’ demand for fertilizer in Sub-Saharan Africa.” Food Security International Devleopment Policy Syntheses 11343. East Lansing: Michigan State University, Department of Agricultural, Food and Resource Economics. KC, S. and W. Lutz. 2017. “The human core of the shared socioeconomic pathways: Population scenarios by age, sex and level of education for all countries to 2100.” Global Environmental Change 42:181-192. doi:10.1016/j.gloenvcha.2014.06.004. Kindermann, G., Obersteiner, M., Sohngen, B., Sathaye, J. et al. .2008. “Global cost estimates of reducing carbon emissions through avoided deforestation.” Proceedings of the National Academy of Sciences 105 (30): 10302-10307. Kissinger, G., Moroge, M., and Noponen, M. 2015. Private sector investment in landscape approaches: the role of production standards and certification. In: Minang, P. A., M. van Noordwijk, O. E. Freeman, C. Mbow, J. de Leeuw, and D. Catacutan (eds.) Climate-Smart Landscapes: Multifunctionality in Practice. Nairobi, Kenya: World Agroforestry Centre (ICRAF) Kriegler, E., B. C. O’Neill, S. Hallegatte, T. Kram, R. J. Lempert, R. H. Moss, and T. Wilbanks. 2012. “The need for and use of socio-economic scenarios for climate change analysis: A new approach based on shared socio-economic pathways.” Global Environmental Change 22 (4): 807–22. doi:10.1016/j.gloenvcha.2012.05.005 Kuteya, A. N., C. Lukamaa and V. C. Malatab. 2018. “Review of e-FISP performance during 2017/2018 agricultural season.” Monitoring Report, May 2018. Lusaka: Indaba Agricultural Policy Research Institute (IAPRI). Kuteya, A. N., N. J. Sitko, A. Chapoto and E. Malawo. 2016. “An In-depth Analysis of Zambia’s Agricultural Budget: Distributional Effects and Opportunity Cost.” Working Paper 108, April 2016. Lusaka: Indaba Agricultural Policy Research Institute (IAPRI). La Rovere, R., G. Kostandini, T. Abdoulaye, J. Dixon, W. Mwangi, Z. Guo, and M. Banziger. 2010. “Potential impact of investments in drought tolerant maize in Africa.” Nairobi: International Maize and Wheat Improvement Center (CIMMYT) Lal, R.. 2004. “Soil Carbon Sequestration Impacts on Global Climate Change and Food Security” Science 304 (5677):1623-1627. doi:10.1126/ science.1097396 Leclère, D., P. Havlík, S. Fuss, E. Schmid, A. Mosnier, B. Walsh, H. Valin, M. Herrero, N. Khabarov, and M. Obersteiner. 2014. “Climate change induced transformations of agricultural systems: Insights from a global model.” Environmental Research Letters 9 (12). doi:10.1088/1748-9326/9/12/124018 Lewis, D., S. D. Bell, J. Fay, K. L. Bothi, L. Gatere, M. Kabila, M. Mukamba, E. Matokwani, M. Mushimbalume, C. I. Moraru, J. Lehmann, J. Lassoie, D. Wolfe, D. R. Lee, L. Buck, and A. J. Travis. 2011. “Community Markets for Conservation (COMACO) links biodiversity conservation with sustainable improvements in livelihoods and food production.” Proceedings of the National Academy of Sciences of the United States of America 108(34):13957-13962. doi:10.1073/pnas.1011538108 Lubungu, M., N. J. Sitko, and M. Hichaambwa. 2015. Analysis of Beef Value Chain in Zambia: Challenges and Opportunities of Linking Smallholders to Markets. Working Paper 103. Lusaka: Indaba Agricultural Policy Research Institute (IAPRI) Maggio, G., N. Sitko, and Ignaciuk, A. 2018. “Cropping system diversification in Eastern and Southern Africa: Identifying policy options to enhance productivity and build resilience.” FAO Agricultural Development Economics Working Paper 18-05. Rome, FAO. Licence: CC BY- NC-SA 3.0 IGO. Matakala, P.W., M. Kokwe, and J. Satz .2015. “Zambia National Strategy To Reduce Emissions From Deforestation And Forest Degradation (REDD+).” UN-REDD. Lusaka, Zambia. Matenga, C.. 2017. “Outgrowers and Livelihoods: The Case of Magobbo Smallholder Block Farming in Mazabuka District in Zambia.” Journal of South African Studies 43(3):551-566. doi:10.1080/03057070.2016.1211402 Mazvimavi, K., and S. Twomlow. 2009. “Socioeconomic and institutional factors influencing adoption of conservation farming by vulnerable households in Zimbabwe.” Agricultural Systems 101(1-2):20-29. doi:10.1016/j.agsy.2009.02.002 McCarl, B. A., and T. H. Spreen. 1980. “Price Endogenous Mathematical Programming as a Tool for Sector Analysis.” American Journal of Agricultural Economics 62 (1): 87. doi:10.2307/1239475 Merotto, Dino. 2017. “Zambia Jobs Diagnostic: Volume 1 - Analytics, Jobs Series Issue No. 7” World Bank, Washington, DC. https:// openknowledge.worldbank.org/handle/10986/27008 Mfune, O., M. N. Chisola and I. Ziba. 2016. “How Can Multifunctional Agriculture Support a Transition to a Green Economy in Afica? Lessons from the COMACO Model in Zambia.” Agriculture. 6(3):1-17 Milder, J. C., Hart, A. K., Dobie, P., Minai, J., and Zaleski, C. 2014. “Integrated Landscape Initiatives for African Agriculture, Development, and Conservation: A Region-Wide Assessment.” World Development 54: 68-80 Ministry of Agriculture (MoA), Republic of Zambia. 2016. Ministerial Statement by Hon. Given Lubinda, MP. Minister of Agriculture on “The Crop Forecasting Survey for the 2015/2016 Agricultural Season and the Food Balance Status for the 2016/2017 Marketing Season.” Lusaka, Zambia. Ministry of Agriculture (MoA), Republic of Zambia. Forthcoming. “Climate-Smart Agriculture Strategic Framework.” Unpublished third draft, October 2017. Mofya-Mukuka, R. and A. M. Shipekesa. 2013. Value Chain Analysis of the Groundnuts Sector in the Eastern Province of Zambia.  Working Papers 78, September 2013. Lusaka: Indaba Agricultural Policy Research institute (IAPRI). Moll, H. A. 2005. “Costs and benefits of livestock systems and the role of market and nonmarket relationships.” Agricultural Economics 32(2):181-193. doi:10.1111/j.0169-5150.2005.00210.x. Mosnier, A., P. Havlík, H. Valin, J. Baker, B. Murray, S. Feng, M. Obersteiner, B.A. McCarl, S. Rose, and U. A. Schneider. 2013. “Alternative U.S. biofuel mandates and global GHG emissions: The role of land use change, crop management and yield growth.” Energy Policy 57 (June). Elsevier: 602–14. doi:10.1016/J.ENPOL.2013.02.035. Mosnier, A., P. Havlík, S. Leduc, and M. Obersteiner. 2014. “Modeling Impact of Development Trajectories and a Global Agreement on Reducing Emissions from Deforestation on Congo Basin Forests by 2030” Environmental and Resource Economics 57 (4):505-525. doi:10.1007/s10640-012-9618-7. Muto, M. and T. Yamano. 2009. “The impact of mobile phone coverage expansion on market participation: Panel data evidence from Uganda.” World Development. 37 (12) :1887-1896. doi:10.1016/j.worlddev.2009.05.004. Naimi, B., N. Hamm, T. Groen, A. Skidmore, and A. Toxopeus. 2014. “Where is positional uncertainty a problem for species distribution modelling?” Ecography 37 (2):191-203. doi:10.1111/j.1600-0587.2013.00205.x. Namirembe, S. and F. Bernard. 2015. “Private sector engagement in landscape-based approaches - lessons from cases in East Africa.” In: Minang, P.A., M. van Nordwijk, O. E. Freeman, C. Mbow, J. de Leeuw, D. Catacutan (eds). Climate-Smart Landscapes: Multifunctionality in Practice. Nairobi, Kenya: World Agroforestry Centre (ICRAF) PAGE 96 Namonje-Kapembwa, T. and H. Machina. 2018. “Youth Employment in Zambia: What Opportunities Does Agriculture Offer?” Policy Brief 92, July 2018. Lusaka: Indaba Agricultural Policy Research Institute (IAPRI) Namonje-Kapembwa, T., R. Black, and T.S. Jayne. 2015. “Does Late Delivery of Subsidized Fertilizer Affect Smallholder Maize Productivity and Production?” Working Paper No.97. Lusaka: Indaba Agricultural Policy Research Institute (IAPRI) Ngoma, H., B. Hamududu, P. Hangoma, P. Samboko, M. Hichaambwa, and C. Kabaghe. 2017. “Irrigation Development for Climate Resilience in Zambia: The Known Knowns and Known Unknowns.” Working Paper 130. Lusaka: Indaba Agricultural Research Institute (IAPRI) Nkonya, E., J. Pender, E. Kato, S. Benin, and H. Makerlova. 2009. “Dynamics of Rural Services and Their Influence on Poverty and Rural Development: Evidence from Uganda,” Project Paper for the project: Making Rural Services Work for the Poor – The Role of Rural Institutions and Their Governance for Agriculture-Led Development” funded by the GTZ/BMZ, Germany Norman, T., D. L. Merotto, and B. Blankespoor. Forthcoming. “It’s all about the processing: Spatial analysis of agro-firm location and jobs potential in Zambia.” Washington DC: World Bank O’Neill, B. C., E. Kriegler, K. L. Ebi, E. Kemp-Benedict, K. Riahi, D. S. Rothman, B. J. van Ruijven, et al. 2017. “The roads ahead: Narratives for shared socioeconomic pathways describing world futures in the 21st century.” Global Environmental Change 42: 169–80. doi:10.1016/j. gloenvcha.2015.01.004 O’Neill, B. C., E. Kriegler, K. L. Ebi, E. Kemp-Benedict, K. Riahi, D. S. Rothman, B. J. van Ruijven, et al. 2017. “The roads ahead: narratives for shared socioeconomic pathways describing world futures in the 21st century.” Global Environmental Change 42: 169-180 Palazzo, A., J. M. Vervoort, D. Mason-D’Croz, L. Rutting, P. Havlík, S. Islam, J. Bayala, et al. 2017. “Linking regional stakeholder scenarios and shared socioeconomic pathways: Quantified West African food and climate futures in a global context.” Global Environmental Change 45: 227–42. doi:10.1016/J.GLOENVCHA.2016.12.002 Pannell, D..J., R. S. Llewellyn, and M. Corbeels. 2014. “The farm-level economics of conservation agriculture for resource-poor farmers” Agriculture, Ecosystems & Environment 187:52-64. doi:10.1016/j.agee.2013.10.014 Pannell, D. J. 1995. “Economic aspects of legume management and legume research in dryland farming systems of southern Australia.“ Agricultural Systems 49 (3): 217-236 Place, F., C. B. Barrett, H. A. Freeman, J. J. Ramisch and B. Vanlauwe. 2003. “Prospects for integrated soil fertility management using organic and inorganic inputs: evidence from smallholder African agricultural systems.” Food Policy 28 (4):365–378. doi:10.1016/j. foodpol.2003.08.009 Probert, M.E. 2007. “Modelling minimum residue thresholds for soil conservation benefits in tropical semiarid cropping systems.” Australian Centre for International Agricultural Research (ACIAR) Technical Report No. 66. Canberra, Australia: ACIAR. 34 pp Rao, M.R. and M.N. Mathuva. 2000. “Legumes for improving maize yields and income in semi-arid Kenya.” Agriculture, Ecosystems and Environment 78(2):123–137. doi:10.1016/S0167-8809(99)00125-5 Riahi, K., S. Rao, V. Krey, C. Cho, V. Chirkov, G. Fischer, G. Kindermann, N. Nakicenovic, and P. Rafaj. 2011. “RCP 8.5-A scenario of comparatively high greenhouse gas emissions.” Climatic Change 109 (1): 33–57. doi:10.1007/s10584-011-0149-y. Rogelj, J., Popp, A., Calvin, K.V., Luderer, G., Emmerling, J., Gernaat, D., Fujimori, S., Strefler, J., Hasegawa, T., Marangoni, G., Krey, V., Kriegler, E., Riahi, K., van Vuuren, D.P., Doelman, J., Drouet, L., Edmonds J, Fricko, O., Harmsen, M., Havlik, P., Humpenoeder, F., Stehfest, E., Tavoni, M. 2018. “Scenarios towards limiting global mean temperature increase below 1.5 degrees C” Nature Climate Change 8, ISSN:1758-678X, Pages: 325-+ Rojas-Downing, M. M., A.P. Nejadhashemi, T. Harrigan, and S. Woznicki. 2017. “Climate change and livestock: Impacts, adaptation, and mitigation.” Climate Risk Management 16: 145-163 Rosenbaum, P. and D. B. Rubin. 1983. “The central role of propensity score in observational studies for causal effects.” Biometrika 70(1):41- 55. doi:10.1093/biomet/70.1.41 Ruesch, A. and H. Gibbs. 2008. “New IPCC Tier-1 Global Biomass Carbon Map for the Year 2000.” http://cdiac.ess-dive.lbl.gov. Rusinamhodzi, L., M. Corbeels, M. van Wijk, M. Rufino, J. Nyamangara and K. E. Giller. 2011. “A meta-analysis of long-term effects of conservation agriculture on maize grain yield under rain-fed conditions.” Agronomy for Sustainable Development 31 (4):657-673. doi:10.1007/s13593-011-0040-2 Samboko, P. C., M. Kabisa, and G. Henley. 2017. “Constraints to Biofuel Feedstock Production Expansion in Zambia.” WIDER Working Paper 2917/62, UNU-WIDER, Helsinki Sayer, J., T. Sunderland, J. Ghazoul, J. Pfund, D. Sheil, E. Meijaard, M. Venter, A. K. Boedhihartono et al. 2013. “Ten principles for a landscape approach to reconciling agriculture, conservation, and other competing land uses.” Proceedings of the National Academy of Sciences of the United States of America 110(21):8349-8356. doi:10.1073/pnas.1210595110 Schmitz, C., H. van Meijl, P. Kyle, G. C. Nelson, S. Fujimori, A. Gurgel, P. Havlík, et al. 2014. “Land-use change trajectories up to 2050: insights from a global agro-economic model comparison.” Agricultural Economics 45 (1). Wiley/Blackwell (10.1111): 69–84. doi:10.1111/ agec.12090 Schneider, U. A., B. A. McCarl, and E. Schmid. 2007. “Agricultural sector analysis on greenhouse gas mitigation in US agriculture and forestry.” Agricultural Systems 94 (2): 128–40. doi:10.1016/J.AGSY.2006.08.001 Sheahan, M., and C. B. Barrett. 2017. “Review: Food loss and waste in Sub-Saharan Africa.” Food Policy 70: 1–12. doi:10.1016/J. FOODPOL.2017.03.012 Shitumbanuma, V.. 2012. Analysis of crop trails under Faidherbia Albida. Lusaka: University of Zambia, Conservation Farming Unit. Sims, T. J. 1996. “Lime requirement.” Methods of Soil Analysis. Part 2: Chemical properties (3rd ed.). 2001:491-517. Smith, P., D., Martino, Z., Cai, D., Gwary, H. H., Janzen, P., Kumar, B., McCarl, S., Ogle, F., O’Mara, C., Rice, R.J., Scholes, O., Sirotenko. 2007. “Agriculture. Chapter 8.” In Climate Change 2007: Mitigation. Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, (B. Metz, O.R. Davidson, P.R. Bosch, R. Dave, .A. Meyer, Eds), Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA. Smith, P., D. Martino, Z. Cai, D. Gwary, H. Janzen, P. Kumar, B. McCarl, and S. Ogle et al. 2008. “Greenhouse gas mitigation in agriculture.” Philosophical transactions of the Royal Society of London, Series B, 363:789-819. doi:10.1098/rstb.2007.2184. Soterroni, A., A. Mosnier, A. Carvalho, G. Camara, M. Obersteiner, P. Andrade, R. Souza, R. Brock, J. Pirker. 2018. “Future environmental and agricultural impacts of Brazil’s Forest Code.” Environmental Research Letters 13 (7). doi:10.1088/1748-9326/aaccbb Sova, C. A., G. Grosjean, T. Baedeker, T. N. Nguyen, M. Wallner, A. Nowak, et al. “Bringing the Concept of Climate-Smart Agriculture to Life: Insights from CSA Country Profiles Across Africa, Asia, and Latin America.” Washington, DC: World Bank  Spielman, D.J., C. Ragasa, and R. Rajalathi. 2012. “Designing Agricultural Research Linkages within an AIS Framework.” in: World Bank. (eds). Agricultural Innovation Systems: An Investment Sourcebook. Agricultural and Rural Development. Washington DC: World Bank. PAGE 97 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Takayama, T., and G. G. Judge. 1971. Spatial and temporal price and allocation models. Amsterdam: North-Holland Publishing Company. Tesliuc, C., W.J. Smith, M.R. Sunkutu, 2013. :Using Social Safety Nets to Accelerate Poverty Reduction and Share Prosperity in Zambia.” Washington DC: World Bank. Thierfelder C., P. Chivenge, W. Mupangwa, T.S. Rosenstock, C. Lamanna, and J.X. Eyre. 2017. “How climate-smart is conservation agriculture (CA)? –its potential to deliver on adaptation, mitigation and productivity on smallholder farms in southern Africa” Food Security 9 (3): 537-560. doi:10.1007/s12571-017-0665-3 Thierfelder, C., M. Mwila, and L. Rusinamhodzi. 2013. “Conservation agriculture in eastern and southern provinces of Zambia: Long-term effects on soil quality and maize productivity.” Soil and Tillage Research 126 (January). Elsevier: 246–58. doi:10.1016/J.STILL.2012.09.002 Thurlow, J., T. Zhu, and X. Diao. 2012. “Current Climate Variability and Future Climate Change: Estimated Growth and Poverty Impacts for Zambia.” Review of Development Economics 16 (3). Blackwell Publishing Ltd: 394–411. doi:10.1111/j.1467-9361.2012.00670.x Tittonell, P. and K. E. Giller. 2013. “When yield gaps are poverty traps: The paradigm of ecological intensification in African smallholder agriculture.” Field Crops Research 143:76-90. doi:10.1016/j.fcr.2012.10.007 Tittonell, P., K.D. Shepherd, B. Vanlauwe, and K. E. Giller. 2008. “Unravelling the effects of soil and crop management on maize productivity in smallholder agricultural systems of western Kenya—an application of classification and regression tree analysis.” Agriculture, Ecosystems & Environment 123 (1-3):137–150. Tschirley, D., T. Reardon, M. Dolislager, and J. Snyder. 2015. “The rise of a middle class in east and Southern Africa: implications for food system transformation.” Journal of International Development 27 (5):628–646. doi:10.1002/jid.3107. Tschirley, D., Reardon, T., Dolislager, M., & Snyder, J. (2014). The rise of a middle class in East and Southern Africa. Wider working paper 2014/119. World Institute for Development Economics Research, Helsinki Google Scholar. Tschirley, D. and M. Hichaambwa. 2010. “The Structure and Behaviour of Vegetable Markets Serving Lusaka: Main Report.” Working Paper No. 93006. Michigan State University, Department of Agricultural, Food, and Resource Economics, East Lansing. Turpie, J., B. Warr, and J. C. Ingram. 2015. “Benefits of Forest Ecosystems in Zambia and the Role of REDD+ in a Green Economy Transformation.” United Nations Environment Programme (UNEP), Nairobi, Kenya. United Nations Framework Convention on Climate Change (UNFCCC). National Adaptation Plan (NAP) Technical guidelines for the national adaptation plan process. Least Developed Countries Expert Group, 2012. United States Agency for International Development (USAID). 2018. Land Tenure dynamics in peri-urban Zambia. Policy Brief. USAID TGCC Program. Valin, H., P. Havlík, A. Mosnier, M. Herrero, E. Schmid, and M. Obersteiner. 2013. “Agricultural productivity and greenhouse gas emissions: Trade-offs or synergies between mitigation and food security?” Environmental Research Letters 8 (3): 1–9. doi:10.1088/1748- 9326/8/3/035019. Valin, H, R. D. Sands, D. van der Mensbrugghe, G. C. Nelson, H. Ahammad, E. Blanc, B. Bodirsky, et al. 2014. “The future of food demand: understanding differences in global economic models.” Agricultural Economics 45(1): 51–67. doi:10.1111/agec.12089. Vinya, R., S. Syampungani, E. C. Kasumu, C. Monde, and R. Kasubika. 2011. “Preliminary study on the drivers of deforestation and potential for REDD+ in Zambia.” FAO/Zambian Ministry of Lands and Natural Resources, Lusaka, Zambia. Van Vuuren, D. P., K. Riahi, K. Calvin, R. Dellink, J. Emmerling, S. Fujimori, S. KC, E. Kriegler, and B. O’Neill. 2017. “The Shared Socio- economic Pathways: Trajectories for human development and global environmental change.” Global Environmental Change 42: 148–52. doi:10.1016/j.gloenvcha.2016.10.009 von der Decken, J. 2017. “Zambia’s fight against rural poverty. The Development Partner Engagement in 2017.” Bachelor Thesis. Georg- August-Universität Goettingen und World Bank Group. Wall, P. C. (2007). Tailoring conservation agriculture to the needs of small farmers in developing countries: an analysis of issues. Journal of crop improvement, 19(1-2), 137-155. Wathum, G., M. Seebauer, and S. Carodenuto. 2016. “Drivers of deforestation and forest degradation in Eastern Province, Zambia. Zambia Integrated Forested Landscape project.” UNIQUE – Forestry and land use. http://ziflp.org.zm/wp-content/uploads/2018/09/ Deforestation_Drivers_UNIQUE-1.pdf [accessed December 2018] Weatherspoon, D. and T. Reardon. 2003. “The Rise of Supermarkets in Africa: Implications for Agrifood Systems and the Rural Poor.” Development Policy Review 21 (3):333-355. doi:10.1111/1467-7679.00214. Wirsenius, S., C. Azar, and G. Berndes. 2010. “How much land is needed for global food production under scenarios of dietary changes and livestock productivity increases in 2030?” Agricultural Systems 103 (9): 621–38. doi:10.1016/j.agsy.2010.07.005. Wooldridge, J. M. (2010). Econometric analysis of cross section and panel data. MIT press. World Bank. 2011. “Missing food: The Case of Postharvest Grain Losses in Sub-Saharan Africa.” Washington DC: World Bank. World Bank. 2011b. “Weather Index insurance for agriculture. Guidance for Development Practitioners.” Agriculture and Rural Development Discussion papers 50. 66274, Washington DC: World Bank. World Bank. 2017a. “9th Zambia Economic Brief. Reaping Richer Returns from Public Expenditures in Agriculture.” Washington DC: World Bank. World Bank. 2017b. “Guidance note on shadow price of carbon in economic analysis. Cover Note.” Washington DC: World Bank World Bank. 2018a. “11th Zambia Economic Brief: An Agro-Led Structural Transformation.” Washington DC: World Bank. World Bank. 2018b. Aide Memoire. COMACO Landscape Management project (P144254). Mid-Term Review Mission. February 26 to 28, 2018. Washington DC: World Bank. World Bank. 2018c. “Future of Food. Maximizing Finance for Development in Agricultural Value Chains.” Washington DC: World Bank World Bank 2018d. “Productive Diversification in African Agriculture and its Effects on Resilience and Nutrition.“ Washington DC: World Bank. World Bank. 2018e. “Zambia—Systematic Country Diagnostic (SCD).” Washington DC: World Bank. World Bank. Forthcoming. Zambia Agriculture Finance Diagnostic Report.” Washington DC: World Bank World Bank and UNCTAD (United Nations Conference on Trade and Development). 2004.” Responsible Agricultural Investment. Outgrower Schemes. Knowledge Into Action Notes.” http://www.worldbank.org/en/topic/agriculture/publication/responsible-agricultural-investment [last accessed December 2018] WFP (World Food Programme) and OXFAM. 2017. R4 Rural Resilience Initiative. Annual Report. Rome: World Food Programme. You, L. and S. Wood. 2006. “An entropy approach to spatial disaggregation of agricultural production.” Agricultural Systems 90 (1): 329– 47. doi:10.1016/j.agsy.2006.01.008 PAGE 98 Appendix A: Policy Frameworks Table APP 1: Overview of policy frameworks relevant to climate-smart agriculture in Zambia Policy Overview Goals for agriculture Measurable Strategies suggested framework sector indicators The National Provides an account of • Increase agricultural • Agriculture 10.08% • No specific strategies Long Term Zambia’s development productivity and land of GDP suggested for the Vision 2030 since independence and under cultivation by agriculture sector scrutinizes the driving factors 2030 • Increase cultivated (2006) land by 900,000 of each sector’s performance. It concludes that the large • Increase exports of hectares majority of farmers are agricultural and agro- processed products • Increasing land dedicated to subsistence under irrigation to farming and have not by 2030 400,000 hectares significantly benefitted from • Preserve the liberalization reforms, partly agricultural resource • Increase agricultural due to remoteness from major base by 2030 machinery markets. • Increase livestock • Increase fish and Vision for the agricultural livestock populations population to sector: An efficient, 6,000,000 competitive, sustainable and • Increase agricultural machinery • Increase fish export-led agriculture sector production to that assures food security and 300,000 MT increased income by 2030 The National Based on a comprehensive • Diversify incomes to • Establish 2 disease- • Promote irrigation and Adaptation assessment of climate-related allow for purchasing free zones by 2010 efficient use of water Programme hazards to the economy. It and trading of food recognizes droughts, floods, • Improve post-harvest of Action • Shift agricultural storage and marketing extreme heat, and shorter on Climate rain seasons as the major production to higher • Develop dams and dip Change (2007) challenges to productivity and land to reduce flood tanks resilience of the agriculture and heat risk sector and proposes pragmatic • Gather and sell more • Sustainable supply of feed counter-actions such as income wild food • Promote aquaculture diversification, increased use of irrigation systems, and improved post-harvest storage infrastructure. The National The National Policy on • Sustainable crop and • No measurable • Sustainably intensify land Policy on Environment was developed livestock production indicators proposed use without converting Environment to synchronize and work across additional land area into individual sectoral strategies • Employ ecologically agricultural land (2007) appropriate that pertain to Zambia’s environment and natural production • Develop a comprehensive resources (e.g., agriculture, techniques agricultural policy that mining). The NPE, therefore, incorporates components • Institute an of conservation agriculture aims to fill a policy void to appropriate be one holistic, overarching and does not increase institutional agricultural encroachment strategy to implement the framework for Millennium Development Goals sustainable • Educate and train for sustainable development by development extension workers and developing natural resources farmers on conservation to spur development while • Reduce GHG issues, particularly soil conserving important resources emissions quality and ecosystems. National Seeks to develop a multi- • Increase agricultural • No measurable • Support Zambia’s early Climate sectoral strategy for Zambia in production and food indicators proposed warning system. Change line with UNFCCC’s objective security under the to “stabilize greenhouse changing climate • Diversify crops, fisheries, Response and livestock. gas concentrations in the Strategy atmosphere.” It serves as a • Increase farmers’ • Improve soil quality (2010) primarily economic strategy, capacity to respond through conservation encompassing the Nationally to climatic shocks on the farm scale agriculture, further Determined Contribution and supported by the National Agriculture Policy. • Build infrastructure government-provided on small-holder fertilizers and lime. farms to increase production capacity • Government and private sector investment in • Diversify rural and on-farm infrastructure small-holder incomes including irrigation, water storage facilities, and fish hatcheries. • Drought-resistant varieties of crops and livestock. continued PAGE 99 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table APP 1 Continued Policy Overview Goals for agriculture Measurable Strategies suggested framework sector indicators National After a review of the previous • To increase the • Attain 10% economic • Promote sustainable Agricultural NAP (2004-2015), the Ministry annual growth rate of growth rate 2021- increase in agricultural Policy 2012- of Agriculture and Cooperatives the real GDP 2030 (economy wide) productivity of major crops 2030 (2011) developed a plan for the agricultural sector in line with • To increase the value • Improve agricultural input the Vision. and growth rate of and product markets so as crop exports to reduce marketing costs of agribusiness, including • To contribute to small-scale farmers and reduction of poverty farmer groups; and food insecurity • Increase agricultural exports to preferential markets National The Livestock Development • Attainment of • No measurable • Targeted subsidies Livestock Policy outlines the roles, the food security for indicators proposed for control diseases Development vision and focus of various the majority of of national economic Policy (2012) stakeholders in order to households importance contribute to increased production and productivity of • Increased Livestock • Diversification of the the livestock sub-sector thereby sector contribution livestock base increasing food security and to total foreign exchange earnings • Enhance market support income, and reducing poverty. • Increased livestock • Public-private contribution to GDP. partnerships, with the public sector focused • Increased incomes for on small-scale farmers those involved in the and the private sector agricultural sector investing in credit marketing and provision • Technology dissemination, particularly to small-scale farmers National The National Agriculture • Increase the value • Increase in • Support farmer Agriculture Investment Plan was crafted and growth rate agricultural exports cooperatives and increase Investment under the Comprehensive of crop exports to as a percentage funding for extension Plan 2014- Africa Agriculture Development preferential markets of non-traditional workers Programme and outlines exports from 41% in 2018 (2013) specific investment • Increase the annual 2011 to 55% by 2018 • Promote low-cost farm opportunities for the Zambian growth rate of inputs such as inorganic agricultural sector, with agricultural GDP and • Increase cereals fertilizer an emphasis on pro-poor agriculture’s share of production from 3.2 overall GDP million tonnes to • Diversify crops to include agricultural-led economic oil seeds (e.g., sunflower) development. 6.0 million tonnes • Increase production by 2018 and fruits/vegetables of staple foods, to provide a balanced including fruits and • Reduce chronic diet and supplemental vegetables, for own malnutrition among income, particularly consumption and the children under five focusing on crops that surplus for income from 45% to 30% increase soil quality generation by 2018 • Increase the number of • Reduce soil erosion farmers who have the title per hectare from 20 to their land tonnes to 10 tonnes by 2018 Zambia’s Zambia determined its • Reduce GHG • Reduce 38,000 Gg • Reduced fertilizer use and Nationally intended reduction in GHG emissions from the CO2e emission less turning of soil Determined emissions, in line with the 2015 agricultural sector conditional on Agreement on Climate Change international support • Soil carbon sequestration Contribution • Improve soil from the United Nations of US$35 billion; • Rural biogas plants to the 2015 Framework Convention on productivity leading reduce 20,000 Gg Agreement Climate Change (UNFCCC). to improved crop CO2e under the on Climate productivity domestic efforts with Change (2015) • Create agricultural limited international job opportunities and support. This alternative livelihoods translates into a contributing to reduction potential reduced rural poverty of 25% and 47% against 2010 as the base year, respectively PAGE 100 Table APP 1 Continued Policy Overview Goals for agriculture Measurable Strategies suggested framework sector indicators Zambia Forest covers approximately • Reduce GHG • No measurable • Develop a climate-smart National 60% of Zambia’s land area, emissions by indicators proposed agricultural practice Strategy although Zambia has one of improving forest and framework to Reduce the top 10 deforestation rates land management of any nation in the world. The • Use incentives to increase Greenhouse UNFCCC developed REDD+ • Increase adoption of adoption of climate-smart Gas Emissions as a mechanism to reduce agricultural practices technologies and practices from deforestation in developing that mitigation carbon emissions • Promote good Deforestation countries. This national agricultural practices and Forest plan represents the working • Regulated systems for related to reduced Degradation relationship between the wood fuel are in place emissions from agro- (REDD+) Government of Zambia and the processing dependent (2015) UNFCCC to reduce emissions on use of wood fuel from from deforestation and land indigenous forests use change in Zambia. Seventh Every five years, the • Build a diversified • 10%+ share of GDP is • Emphasize high-value National Government of Zambia and export-oriented agriculture by 2021 export crops such as Development undertakes a planning process agricultural sector cashews, coffee, sugar, to define the roadmap focused on crops, • Agriculture accounts and tea Plan 2017-2021 for 20% of new to development for the livestock, and timber (2017) coming years. This document jobs generated by • Build rural infrastructure builds toward the Vision • Improve water investments by 2021 to increase small-holder 2030 of having Zambia as a resources farmers’ access to markets development and • Agricultural, “prosperous, middle-income fisheries, and fishing • Invest in rural irrigation country by 2030” and outlines management export value of specific development goals and • Improve access $931M by 2021 (7.8% • Develop the ICT system strategies for each sector of the to finance for of total export value) • Develop livestock and economy. production and fisheries breeding centers exports • Diversify small-holder farmers PAGE 101 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Appendix B: Results of First Stakeholder Workshop B.1 Developing a Normative Vision for the Agriculture Sector in Zambia Goal(s) for the session: Develop a one-sentence visioning statement for each pillar of CSA (productivity, resilience, mitigation co-benefits) out to 2050. Methodology: Workshop participants were divided evenly into three groups (one per CSA pillar). Workshop facilitators asked participants to identify words that were associated with the CSA pillar. For each word, facilitators asked participants to write a sentence about their hope for the CSA pillar associated with that word. The sentences were clustered and combined by the group into one cohesive vision. Results: • Productivity: “By 2050, double profits and yields through diversification of crops (beyond maize), while ensuring household food and nutrition security.” • Resilience: “By 2050, have an agriculture sector that is diversified in crop production, age, and gender, and which is able to cope with economic and climatic shocks through enhanced capacity and policy.” • Mitigation: “By 2050, Zambia’s agricultural sector will increase productivity while maintaining a low ecological footprint.” Discussion: • Participants felt that low productivity could be overcome for farms by 2050, to the extent that the sector could become twice as productive. • There was an important distinction between commercial and small-scale farmers, and the discussion was around whether the goal should focus on one scale or another; participants ultimately decided that not differentiating between the scales allowed for the commercial sector to make up for any lack of productivity gains of small-scale farmers. • The group highlighted both profits and food security as measures of success because both were critical indicators of development—food security alone indicated a level of hand-to-mouth poverty, but coupled with profits indicated a level of economic empowerment. • Diversification is seen as a key tool both for ensuring that farmers are insulated from market and climatic shocks (resilience) and for increasing small-holder farmers’ standard of living, measured by both food security and incomes (productivity). • Although mitigation of GHGs is understood to be important, increasing productivity in the face of climatic and market uncertainties is the clear priority for the sector. PAGE 102 B.2 Prioritizing Goals to Work Toward Vision Goal(s) for the session: Identify specific, measurable goals that would allow Zambia to achieve the vision for each CSA pillar. Methodology: Workshop participants remained in assigned groups by CSA pillar. Each workshop group identified between 20 and 35 goals with end date 2050 for their CSA pillar; the goals were then clustered thematically. In addition to goals identified by workshop participants, the workshop moderators contributed measurable targets that were part of national plans for Zambia (e.g., the 7th National Development Plan). Workshop participants were then allowed to vote using stickers on their favorite goals, across all three CSA pillars. Results: Number of votes for goals, by category: Crop & livestock varieties Yield Policy Technology Wo men and youth empowerment Post-harvest loss management Ma rket access & info Wa ter Diversification Deforestation Financing 0 5 10 15 20 25 30 35 Productivity Resilience Mitigation Top three specific goals, by CSA pillar: Productivity Resilience Mitigation • Promote the development and use • 60% of arable land uses drought- • Increase investments in extension of stress-tolerant varieties among resistant crops and agricultural R&D to 15% of the 30% of small holder farmers agricultural budget • Post-harvest losses are reduced • Double yields for maize and food to 5% • Increase the adoption of CSA legumes for small holder farms practices by at least 30% • At least 60% of men, women • Make market information readily and youth involved in agriculture • 75% of Zambian women and available through ICT belong to functioning savings youth are participating in modern groups farming to achieve intensification Discussion: As with the visioning exercise, many workshop participants, particularly in the productivity group, focused heavily on the scale at which these goals would be applied. The commercial sector, for example, may try to double exports, while small holder farmers will try to double productivity of their maize plots. All three groups had goals related to women and youth participation in the sector, although participation of women in the agricultural sector is already high. PAGE 103 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN B.3 Prioritizing Goals to Work Toward Vision Goal(s) for the session: Identify a list of specific strategies that are necessary in order to achieve the sector goals. Methodology: Participants were placed into three different groups to address both variation in commodities and scale within Zambia’s agricultural sector: (1) smallholder crop production, (2) crop production of medium- to large-scale farmers, and (3) livestock and fisheries (across all scales). For each group, participants were asked to identify the extent to which each proposed strategy had already been tried and the extent to which it had been successful. After strategies were identified, the workshop participants were invited to vote on the most promising individual strategies using stickers. Results: Relative weight of strategic engagement method, by commodity/scale: Smallholder crop production Medium- to large- scale crop production Livestock and fisheries Crop & livestock varieties Farm technologies & practices Financing & subsidies Knowledge sharing Ma rket access & information Policy Post-harvest & value chain Soil fertility Wa ter management Top four specific strategies, by commodity/scale: Smallholder crop Medium- to large-scale Livestock and fisheries production crop production • Breeding for high-yield varieties • Research & development of new • Outgrower programs crop varieties • Gendered enterprise model to the • Micro-finance policy and value chain • Education of farm workers to legislation increased skilled manpower • Pluvial gardening in ponds • Train staff in animal husbandry • Specialized extension services and pasture management • “Pass on the Gift” model (i.e., Heifer International model) • Renewable energy for agricultural • Adopt a public-private partnership bioslurry approach with NGOs Discussion: Many groups highlighted that this was not the first strategy exercise they had been a part of, and many of the strategies had already been tried, but that the bottleneck was with regards to scaling. Many participants felt that NGOs have experimented on the ground, but that strategies were not being widely adopted. One key constraint identified was that actors on the ground were not aware of what funds were available for various projects and were therefore not scaling projects. Another was the limited capacity of the extension workers to help introduce and scale up new projects. Some practices, like liming, have been shown to be effective, but were too expensive to be widely adopted. B.4 Drivers of Uncertainty and Scenarios of Future Worlds Goal(s) for the session: Identify a list of exogenous drivers of uncertainty and quantify the extent to which they impact Zambia’s ability to achieve its goals. PAGE 104 Methodology: Participants were asked to identify a list of exogenous factors that could impact Zambia reaching its goals. The factors were then grouped into general trends that co-vary with each other. For each “driver of uncertainty,” workshop participants were asked to quantify the extent to which the uncertainty impacted Zambia’s capacity to achieve the goals previously identified in the workshop on a scale from 0-3. For example, the workshop participants were asked, “To what extent (0-3) does rainfall variability impact small-holder farmers’ ability to double maize and legume yield by 2050?” Results: Brainstormed drivers of uncertainty, grouped into clusters: Climate change Regional & international markets Domestic development • Rainfall variability • Global markets and commodity • Internal political stability (i.e., the prices Zambian government) • Temperature variability • External political stability (i.e., • Changing dietary patterns • Pests and diseases regional governments) • Availability of external funds Relative impact of uncertainty drivers on CSA pillars: Mitigation Resilience Productivity Climate change Regional and international markets Domestic development Discussion: During this session, participants focused heavily in conversation on the extent to which external political and market forces would impact their ability to achieve their goals. For example, one participant identified how commodity prices of sugar, for example, would hurt small-scale farmers who are just beginning to grow sugar. Similarly, a refugee crisis in neighboring Democratic Republic of the Congo could increase the number of mouths to feed and increase national malnutrition. However, according to the numbers, climate change was identified by all three groups as the largest wild card in achieving the stated goals. In particular, the resilience group strongly connected climate change impacts with reaching their goals. PAGE 105 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Appendix C: Household-level Econometric Assessment C.1 Methodological Approach This section expands on a methodological summary provided in Section 3.2. A difference in difference (DiD) approach is used to measure the causal effect of early CSA adoption on welfare indicators, while addressing self-selection bias. The DiD estimator is used to assess the impact of CSA adoption on households that did not adopt the practice in the first wave of the survey, but did adopt in the second wave. The choice of DiD is based on the assumptions that farmers “self-select” themselves into being in the group of CSA adopters, according to a number of specific characteristics. Households which have adopted CSA are expected to differ systematically from those that have not. Failing to control for the selection leads to biased results (Rosenbaum and Rubin 1983). The empirical strategy takes advantage of an inverse weighted DiD regression model. This approach is expected to remove the selection bias through a doubly robust estimation of the causal effects of the treatment (Hirano et al. 2003; Bang and Robins 2005). To the extent that the CSA investments are captured either by dichotomous or continuous variables, the inverse probability weights have been obtained using either a propensity score (Rosenbaum and Rubin 1983) or a general propensity score analysis (Hirano and Imbens 2005), accordingly. The DiD estimator mimics an experimental research design using observational study data, by studying the differential effect of a treatment on a “treatment group” versus a “control group.” The observational study data are two sets of the Rural Agricultural Livelihoods Surveys (RALS) household surveys for 2012 and 2015. In the DiD, the adopter of a CSA practice is the treatment group. The adopter is defined as having adopted a CSA practice after the 2012 survey and before the 2015 survey. The causal effect has been estimated using a Weighted Least Squares procedure whereby the weight attributed to each observation is the inverse of the probability of the treatment (Imbens 2000; Bang and Robins 2005; Naimi et al. 2014). More weight is given to the treatment cases that resemble the controls, since they are more relevant. Similarly, since the control cases that look like those that have gotten the treatment are more relevant, they are weighted more. The baseline specification has been estimated using a DiD estimator, which enables controlling both for the unobserved time-fixed heterogeneity and the change in trend between treated and control. The common trend assumptions cannot be tested as only two survey waves are available: Equation 1: where Yi is a vector of selected outcomes, CSA is the treatment, D2013 is the time dummy identifying the follow-up survey round, and Xi is a vector of exogenous explanatory variables. In addition, Ø and ∈ are the constants and the random distributed error terms, respectively. In this framework, the parameter of interest, γ2, estimates the average treatment effect of the CSA investment. To exclude the confounders, the analysis considers only households not adopting any CSA investment at the baseline and the follow-up (control group) and households that have invested in CSA at the follow- up (treatment group). However, the limitation of this approach is that the sample of adopters is restricted to only recent adopters, that is, those who adopted between 2012 and 2015. PAGE 106 To capture impact of long-term adoption of CSA, an OLS regression model was employed to show the conditional correlation between long-term CSA adopters and welfare indicators. The impact estimates using the DiD model are restricted to early adopters of the practice. To address this limitation, an OLS model is used to examine associations between the welfare variables of interest and adoption of particular CSA practices. Under this second model, the adopters of a CSA practice are defined as households that adopted the practice in both waves of the survey. While this approach does not allow for an assessment of the causal impact of adoption of a CSA practice, as does the DiD approach, it provides insights into relationship between longer-term adoption and key welfare outcomes. The second specification (Equation 2) uses the same group (Wooldridge 2010) and tests the impact of CSA adoption for the households that have experienced a rainfall shock: Equation 2: In this case, the variable SHOCK identifies the exposure to a rainfall shock during the reference period. In this framework, the parameters β2 and γ2 are the stand-alone treatment effects of the shock and the investment, respectively. Further, 02 is the incremental treatment effect of a specific CSA investment on the subsample of households that experienced a shock. The limitation of this second approach is that selection bias cannot be controlled for, leading to concerns about endogeneity. Thus, the results of the second approach are treated as conditional correlations rather than casual impacts. To test the sensitivity of results relative to climate shocks, the analysis approximates the impact of rainfall shocks using a Standard Precipitation Index (SPI). SPI was developed for defining and monitoring drought. It enables determining the probability of a drought at a given time of interest (temporal resolution) for any rainfall station with historic data. The index can also be used for identifying flood risk. It is based on the cumulative probability of a given rainfall event which is calculated using historical rainfall data. The data, at resolution of a weather station, is smoothed through a moving width before being fitted to a gamma distribution through a maximum likelihood estimator. The baseline analysis sets the threshold for anomalous dry periods at -1 and 1 for anomalous wet periods. The rainfall cumulative distribution is calculated for a time span of six months, covering the entire agricultural season from November to April. The dummy variables identify events which have a low probability to occur in medium accumulation periods that may not have disruptive consequences. Sensitivity analysis was conducted for all models. The climate sensitivity analysis was carried out re-estimating the models using all possible combinations between the thresholds to identify shocks and the time span to calculate the cumulative distribution of rainfalls. By reducing the accumulation period to three months covering the planting and the flowering season (from November to January), the shock index captures infrequent rainfall events that reduce soil moisture and water flow in smaller creeks. Moreover, by setting the thresholds at -1.5 and 1.5, the index captures extreme dry or wet periods that affect streamflow and reservoir storage. Since the exact relationship between accumulation period and impact depends on the natural environment (e.g., geology, soils) and human interference (e.g., existence of irrigation schemes, infrastructure), the estimated treatment effect is expected to change with the indexes. C.2 Data Sources This section expands on a data sources summary provided in Section 2.3. The analysis used the Rural Agricultural Livelihoods Surveys (RALS) for 2012 and 2015. These PAGE 107 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN surveys were collected by the Zambia Central Statistics Office (CSO) in collaboration with the Ministry of Agriculture, Michigan State University, and the Indaba Agricultural Policy Research Institute (IAPRI). They are nationally representative of smallholder households in Zambia (cultivating <20 hectares). Among other things, the survey contains georeferenced information on sociodemographic characteristics, farm management practices, access to credit and assets, access to markets and infrastructure, information availability, input used, and output produced. The survey consisted of 7,254 households in 2012 and 7,934 in 2015. In 2015, 680 new households were added from different clusters in Lusaka, Eastern, and Muchinga Provinces. Daily rainfall data in high resolution is available on a daily basis. The RALS data permits controlling for geographical and biophysical conditions, and therefore it can be augmented with geospatial rainfall, soil quality, and agroecological information to control for the geographical and physical conditions. Rainfall data have been extracted from the Africa Rainfall Climatology version 2 (ARC2) of the National Oceanic and Atmospheric Administration’s Climate Prediction Center. This daily data is based on the latest estimation techniques and has a spatial resolution of 0.1 degrees (~10km). The dataset contains rainfall information at 10-day intervals over the 1983-2015 period and describes the historical rainfall pattern of each geographic location at the weather station resolution. This allows for the creation of georeferenced rainfall variables to identify anomalous wet and dry conditions, which are used in the analysis to assess the impact of climate events on adopters and non-adopters of particular CSA practices. Data on maize seed varieties used by smallholders during the 2012 and 2015 farming seasons was collected from major seed producers based in Zambia. The information collected included the recommended “not to exceed” planting date by AEZ level, and whether the seed was bred for drought and/or heat tolerance. In total, information on 74 varieties from five companies was collected. C.3 Detailed Results This section expands on an overview of high-level results provided in Section 4; it provides summary results of the econometric analysis of CSA-specific impacts on key welfare indicators (Table APP 2 and Table APP 3) and summary of the sensitivity analyses, for the estimated impact of CSA strategies under abnormally wet and dry conditions (Table APP 5 and APP 6). Table APP 4 is based on a combination of qualitative information gathered through stakeholder consultations and literature review, as well as empirical results where available. The DiD estimations aim to assess the average treatment effect and the estimation of the incremental impact of the treatment on the sub-sample of household farmers who have experienced a rainfall shock (obtained through a triple difference estimator) and included following control variables: female household head; household head’s highest level of education completed; number of household members in adult equivalent; dependency ration; agriculture asset wealth index normalized; total number of animals; whether household cultivates only one crop; household is part of a cooperative, farmer, women, savings, loan group; participation in the FISP; smallholder household; commercialization index; distance to roads, markets, FRA (log of distance in kilometers); medium number of traders; information availability on CSA index; household applies inorganic fertilize; agroecological zones. The estimation results of each model, that is, DiD, OLS and assessment of climate sensitivity for different rainfall shocks (as obtained by modifying the accumulation period and/or the identification thresholds) is available upon request in the form of a Background Paper. PAGE 108 Table APP 2 Estimated impact of short-term adoption of CSA strategies on a range of household welfare indicators using a DiD approach Crop income Gross Gross income Food Crop income Poverty variability income variability insecurity Minimum soil -0.011 0.012 -0.03 -0.003 0.020* 0.002 disturbance Residue retention 0.006 0.041 0.031 -0.174*** -0.023 0.039 Legume rotation or -0.011 -0.020 0.019 -0.029 -0.010 0.010 intercropping Commercial 0.096*** -0.015 0.067*** -0.048*** -0.017* 0.011 horticulture Agroforestry -0.048 0.017 -0.052 0.023 0.027 0.031 Crop/livestock 0.006 0.008 0.026 -0.020 -0.046** 0.019 integration Livestock -0.000 0.000 0.000 0.001* 0.000 0.000** diversification Delayed timing -0.000 -0.042 -0.031 -0.027 0.007 0.014 of planting Use of drought- 0.030 -0.021 0.018 -0.015 -0.016 -0.002 tolerant maize seeds Use of heat-tolerant 0.022 0.007 0.032 -0.024 -0.011 -0.006 maize seeds Crop diversification 0.003*** -0.001 0.003*** 0.002 0.001 0.001 Sources: Estimation by FAO, using RALS 2012 and 2015 data. Note: * p<0.10; ** p<0.05; ***p<0.01: the level of statistical significance in results increases with the number of (*). The complete estimation results with the full range of control variables are available upon request. Table APP 3 Estimated impact of longer-term adoption of CSA strategies on a range of household welfare indicators using an OLS approach Crop income Gross Gross income Food Crop income Poverty variability Income variability insecurity Minimum soil 0.554*** -0.474 0.630*** -0.487*** -0.228*** -0.110 disturbance Residue retention 0.062* 0.034 0.009 -0.036 -0.002 0.008 Legume rotation or 0.222*** -0.234 0.025 -0.154 0.031 -0.073 intercropping Commercial 0.300*** -0.334*** 0.043 -0.154** 0.005 0.044 horticulture Agroforestry -0.053 -0.493* 0.054 -0.459** -0.009 0.233 Livestock 0.115 -0.151 0.204 0.062 -0.071 -0.281*** diversification Delayed timing -0.087** 0.463*** -0.086*** 0.003 0.016 -0.018 of planting Use of drought- 0.343*** -0.363*** 0.181*** -0.146** -0.099*** -0.144*** tolerant maize seeds Use of heat-tolerant 0.348*** -0.612*** 0.220*** -0.184** -0.144*** -0.175*** maize seeds Crop diversification 0.006*** -0.003*** 0.002*** -0.003*** 0.001 0.000 Sources: Estimation by FAO, using RALS 2012 and 2015 data. Note: * p<0.10; ** p<0.05; ***p<0.01: the level of statistical significance in results increases with the number of (*). The complete estimation results with the full range of control variables are available upon request. PAGE 109 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table APP 4 Assessment of the impact of CSA strategies on alternative development indicators Likelihood of Household Value chain Job creation adoption by Nutrition income development poor farmers Minimum soil 0 + - 0 ++ disturbance Crop residue + 0 - + 0 retention Legume rotation 0 ++ + +++ ++ or intercropping Commercial ++ ++ 0 ++ ++ horticulture Agroforestry + 0 + + + Livestock ++ ++ - +++ +++ strategies Delayed planting ++ ++ + ++ ++ Use of drought- and heat-tolerant +++ + 0 +++ + maize seeds Crop ++ ++ + +++ +++ diversification Sources: Elaborated by FAO Note: (-)=negative, 0= neutral, +=low positive, ++=medium positive, +++=high positive Table APP 5 Estimated impact of CSA strategies under abnormally dry conditions Crop income Gross Gross income Food Crop income Poverty variability Income variability insecurity Minimum soil -0.000 -0.000 0.000 0.000* -0.000 0.000 disturbance Residue retention -0.160 -0.045 -0.295 -0.104 0.071 -0.236*** Legume rotation or -0.037 -0.089 0.082 -0.064 -0.031 -0.104* intercrop Commercial 0.211** -0.246 0.251** -0.148 -0.191*** -0.160** horticulture Agroforestry 0.000 -0.000 0.000 -0.000 0.000 -0.000*** Crop/Livestock 0.000*** -0.000 0.000 -0.000 -0.000*** 0.000 integration Livestock 0.001 0.003 -0.000 0.007* 0.000 -0.005** diversification Delayed planting -0.085** 7.826*** -0.082*** 0.005 0.014 -0.020 Use of drought 0.365*** -0.246* 0.195*** -0.140** -0.110*** -0.138*** tolerant maize seeds Use of heat tolerant 0.383*** -0.241* 0.248*** -0.169* -0.159*** -0.174*** maize seeds Crop diversification 0.006*** -0.005*** -0.002 -0.004*** 0.001 0.001 Sources: Estimation by FAO, using RALS 2012, 2015 and ARC2 data. Note: * p<0.10; ** p<0.05; ***p<0.01: the level of statistical significance in results increases with the number of (*). The complete estimation results with the full range of control variables are available upon request. PAGE 110 Table APP 6 Estimated impact of CSA strategies under abnormally wet conditions Crop income Gross Gross income Food Crop income Poverty variability Income variability insecurity Minimum soil -0.230*** 0.018 -0.274*** 0.040 0.119** 0.020 disturbance Residue retention -0.091 0.021 -0.174* 0.128 0.077* 0.025 Legume rotation or -0.031 -0.151*** -0.058 -0.081 -0.014 -0.040 intercrop Commercial -0.002 -0.082* 0.049 -0.036 -0.003 -0.025 horticulture Agroforestry -0.075 0.072 -0.006 0.117 0.056 -0.252*** Livestock crop -0.026 -0.164* -0.090 -0.328*** 0.050 0.065 integration Livestock -0.000 -0.000 0.001 -0.001 -0.000 0.000 diversification Delayed planting -0.099*** -0.191** -0.124*** -0.008 0.027 -0.021 Use of drought 0.390*** -0.351** 0.190*** -0.136* -0.092*** -0.150*** tolerant maize seeds Use of heat tolerant 0.435*** -0.426** 0.253*** -0.144 -0.155*** -0.201*** maize seeds Crop diversification 0.006*** -0.004*** -0.002 -0.004*** 0.001 0.001 Sources: Estimation by FAO, using RALS 2012, 2015 and ARC2 data. Note: * p<0.10; ** p<0.05; ***p<0.01: the level of statistical significance in results increases with the number of (*). The complete estimation results with the full range of control variables are available upon request. PAGE 111 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Appendix D: Cost-benefit Analyses D.1 Methodological Approach As discussed in Section 2.4, the CBA includes two assessments: (i) financial analysis to assess the financial viability of adopting CSA practices for households, and (ii) economic analysis to assess the impact of promoting CSA on the economy. This section expands on Section 2.4 to provide additional information. Key assumptions of the financial analysis: • All financial models conduct the analysis for 1 hectare of farmland. • It is expected that yield increases as a consequence of adopting certain CSA practices such as minimum soil disturbance start, on average, in year 3. • Production costs include cash input and family labor costs, such as the costs for purchase of seeds (improved varieties), chemical fertilizers (basal and top-dress), organic fertilizer (manure), herbicides, or expenses to hire animal draft power (ploughing or ripping with oxen). Financial prices are derived from the RALS data. Fertilizer prices are considered as full (financial) price and do not consider subsidies provided under FISP. • Labor costs are estimated using rural market wage of 5.3 ZMK/person-day (which equals 6 hours per day) as a proxy, which was derived from the RALS dataset. This observed wage rate is lower than the minimum wage rate for general workers (6.2 ZMK/person-day) as established by the GoZ and effective since July 2012. This implies that the analysis presented here is estimated in a realistic manner, considering the real conditions of the job market in rural areas. Labor is valued in the same way, regardless if the laborer is a family member or a hired worker. • No investment costs are considered for the crop models. • For each model, the gross and net margins (the latter including labor cost as part of total production cost) are provided. Key assumptions of the financial analysis: The economic analysis is conducted over a 30-year period. • Economic benefits are based on (i) the expected increased economic returns (private net benefits) of target households when adopting selected CSA practices as compared with conventional practices (i.e., the net incremental benefits); and (ii) public good benefits related to mitigating GHG emissions. Public good benefits are approximated by the net carbon balance, which is calculated using FAO’s EX-ACT tool and valued at an annual shadow price of carbon. • The net incremental benefits are valued at economic prices. • Financial prices of tradable goods are converted into economic prices using a Standard Conversion Factor.16 • For some key traded goods, specific import/export parity prices at farm gate have been computed with reference to international border prices, applying conversion factors for each category of costs, and eliminating taxes and transfers.17 • A discount rate of 5 percent is applied following World Bank (2016). • The economic cost of labor of 9.06 ZMW/person-day is used as a wage shadow rate, to factor in unemployment in rural areas. PAGE 112 • Investment cost for promoting CSA is assumed at US$41 (economic cost) per household per year. For a standard 6 years investment project, this results in US$260/household. The cost includes: Capacity building of extension service staff; demonstration trials and farmer field schools; provision of inputs (starter-packs) to farmers; research and development of improved/certified seeds; preparation of training material; sensitization and awareness meetings on CSA; and planning, monitoring, and evaluation of training activities. D.2 Data Sources As discussed in Section 2.4, data used include CSA-specific household data collected by the FAO, the RALS (see Appendix C.2 for a description), past investments in comparable programs, and the carbon accounting tool EX-ACT. This section expands on Section 2.4 in the main text. • The CSA-specific survey was conducted by FAO within the project on Climate-smart agriculture in Malawi and Zambia (see http://www.fao.org/climatechange/epic/home/en/), funded by the European Commission. Data refer to the 2012-3 cropping season (Branca et al. 2015). • While FAO and RALS data is available for all AEZs, the most detailed data is available for AEZ II. Specifically, for maize, detailed data for specific management practices is available. Minimum soil disturbance is the most frequently adopted practice and we use it as a proxy for other CSA practices. • The assessment of public investment costs focuses on the period 2013-2018, is based on gray literature, and is mainly related to agriculture development projects in Zambia, integrated with ad hoc personal interviews of government institutions, independent research entities, international institutions, NGOs, and the private sector. • FAO’s EX-ACT tool, was developed primarily using the Guidelines for National Greenhouse Gas Inventories (IPCC 2006) and Tier 1 emission coefficients, provides ex-ante measurements of the impact of agriculture (and forestry) projects and selected activities on GHG emissions and carbon sequestration. The tool provides a net carbon balance, as difference between gross emissions in a without and with project scenario, which is selected as an indicator of the mitigation potential of the project. EX-ACT measures carbon stocks and stock changes per unit of land, as well as methane (CH4) and nitrous oxide (N2O) emissions, expressing its results in tonnes of carbon dioxide equivalent per hectare (tCO2e/ha) for the lifespan of a project and per year. The annual net carbon balance is valued at an annual shadow price of carbon (US$/tCO2e) over the economic lifetime of the project (World Bank 2017b). • The shadow price of carbon is in the range of US$40-80 per tCO2e in 2020, rising to US$50-100 per tCO2e by 2030 and to US$78-US$156 per tCO2e by 2050. Thus, the price evolves at 2.25 percent each year. For the analysis we use the low value starting at US$40 per tCO2e, and high value starting at US$80 tCO2e as well as the high value, to account for the considerable uncertainty. The shadow price of carbon was developed by the High-Level Commission on Carbon Prices, led by Joseph Stiglitz and Nicholas Stern. The values are consistent with achieving the core objective of the Paris Agreement of keeping temperature rise below 2 degrees. For purposes of sensitivity analyses we also use potential carbon market price at US$5, US$11 and US$21 per tCO2e. • Data used to estimate the opportunity cost of capital was sourced from the World Bank and the Bank of Zambia. The financial discount rate is estimated at 12 percent, computed as an average between: (i) average deposit interest rate paid by commercial or similar banks in the country; (ii) lending interest rate; (iii) real interest rate; and (iv) long-term bonds rate (Table APP 7). PAGE 113 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table APP 7 Computation of discount rate to be used in the analysis Deposit interest Lending interest Long-term Indicator Real interest rate Average rate rate bond rate Rate (%) 7.9 13.6 2.8 23 11.8 Source: Bank of Zambia, http://www.boz.zm/, last accessed August 2018 D.3 Selected Results Results of the of the financial analysis are presented below. Table APP 8 and APP 9 provide crop yields for selected CSA practices and crops across AEZ. Table APP 10 provides annual net margins for conventional and CSA practices as well as the incremental net revenues. Table APP 8 Crop yields by AEZ and farming practices (Kg/ha) AEZ I AEZ IIa AEZ IIb AEZ III AEZ I AEZ IIa AEZ IIb AEZ III AEZ I AEZ IIa AEZ IIb AEZ III Crop Conventional CSA/MSD % difference Beans 586 497 601 539 821 696 842 680 40 40 40 26 Cassava 1,912 5,214 2,575 6,993 2,231 6,083 3,004 8,100 17 17 17 16 Cotton 1,037 992 999 n/a n/a 684 n/a n/a n/a -31 n/a n/a Cowpeas 490 616 383 400 930 1,170 728 700 90 90 90 75 Groundnuts 356 430 852 730 509 615 1,219 900 43 43 43 23 Maize 1,654 1,933 1,217 3,575 1,930 2,475 1,420 2,200 17 28 17 -38 Rice 1,891 1,771 1,510 1,386 3,045 2,852 2,431 2,231 61 61 61 61 Soybeans 1,820 894 196 831 2,639 1,296 285 1,204 45 45 45 45 Source: FAO-CSA survey (season 2012-3) and RALS 2015 datasets, personal interviews and focus group discussions Note: n/a indicates that information was not available. MSD stands for minimum soil disturbance. Values for MSD are used as a proxy for CSA practices. Table APP 9 Crop yields for maize for alternative MSD systems in AEZ IIa (Kg/ha) Conventional CSA Crop Conventional Ploughing Planting Ripping with MSD and MSD and MSD and hand hoe/ with oxen basins/ oxen agroforestry legume residue ridging potholes rotation retention Maize 1,933 1,618 2,139 2,229 2,644 2,893 2,134 Source: FAO-CSA survey (season 2012-3) and RALS 2015 datasets, personal interviews and focus group discussions Note: MSD stands for minimum soil disturbance. Table APP 10 Annual net margins and incremental benefits for each crop and AEZ in US$/ha AEZ I AEZ IIa AEZ IIb AEZ III AEZ I AEZ IIa AEZ IIb AEZ III AEZ I AEZ IIa AEZ IIb AEZ III Crop Net margins— Net margins— Net incremental benefits conventional practices CSA/MSD Beans -71 -100 -67 -87 -62 -101 -65 -110 10 0 2 -24 Cassava 24 392 98 590 13 443 89 664 -11 51 -9 74 Cotton 39 28 30 -21 0 -49 0 0 Cowpeas 23 39 10 12 88 94 96 68 65 55 87 56 Groundnuts 31 60 224 177 78 133 357 240 47 73 132 63 Maize 21 48 -22 132 51 86 1 53 30 38 23 -79 Rice 242 218 165 140 441 404 305 271 199 186 140 131 Soybeans 108 -11 -100 -19 238 90 -39 78 130 101 61 97 Source: FAO calculated based FAO-CSA survey (season 2012-3) and RALS 2015 Note: For maize in AEZ IIa, the net margins and net incremental benefits present the average across farming practices CSA/MSD practices which are presented in Table APP 9. PAGE 114 Appendix E: GLOBIOM E.1 Methodological Approach As discussed in Section 2.5, GLOBIOM is an integrated modeling approach that allows for joint analysis of the agriculture and forestry, and bioenergy sectors. This section expands on the overview provided in Section 2.5 and provides more detail about the following aspects of the GLOBIOM model: (i) The GLOBIOM analytical process; (ii) Description of policy scenarios and CSA-adoption scenarios; (iii) Sensitivity analysis: Climate change uncertainty; (iv) Sensitivity analysis: Shared Socioeconomic Pathways. (i) The GLOBIOM analytical process Drivers of agricultural development in Zambia include both local and global factors, thus a global- to-local modelling approach is adopted. At the local level, national policies, population growth and economic growth are key drivers, while at the global level, climate change and international trade are important factors. The interplay between these two sets of drivers will eventually determine the future development trajectory of agriculture in Zambia. For this analysis a global-to-local modelling approach is adopted which incorporates both the global and local drivers of agricultural growth. To capture the high level of uncertainty related to the future development of these drivers, a scenario framework is applied that considers uncertainties associated with climate change such as precipitation and temperature changes, and the possible trajectories of socioeconomic drivers of agricultural growth such as population growth, economic development and technical change. The methodology is similar to the one used in recent global impact assessments of climate change (O’Neill et al. 2017). GLOBIOM’s analytical process captures the multiple interrelationships between the different systems involved in provision of agricultural and forestry products, for example, population dynamics, ecosystems, technology, and climate (Figure APP 1). This allows for quantifying the impact of global and national developments on the agricultural sector in Zambia. GLOBIOM draws on results of following models to simulate future agriculutral development: • The biophysical process simulation model Environmental Policy Integrated Climate (EPIC) to model agronomic and environmental processes; • RUMINANT to capture livestock activities; and • BIOENERGY and G4M to capture energy demand and forestry management, respectively. The model computes market equilibrium for agricultural and forest products by allocating land use among production activities to maximize the sum of producer and consumer surplus, subject to resource, technological and policy constraints (McCarl and Spreen 1980). Trade flows are balanced out between specific geographical regions based on the spatial equilibrium approach (Schneider et al. 2007; Takayama and Judge 1971). Trade is based purely on cost competitiveness as goods are assumed to be homogeneous. This allows tracing of bilateral trade flows between individual regions. Furthermore, the following model specifications stand out: • The supply side of the model is based on a bottom-up approach considering: land cover, land use, and management systems and related production and markets. Agricultural and forest productivity stems from crop models and forest models and is modeled at the level of grid cells of 5x5 to 30x30 arc-minutes. For instance, globally gridded simulated crop yields and resource requirements PAGE 115 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure APP 1 Overview of GLOBIOM DEMAND Population, GDP, consumer preferences Food Fibers Energy Industry MARKETS Market and Trade: EU + World  PRICES FEEDSTOCK FEED FEED BIOMASS PRODUCTION EPIC RUMINANT BIOENERGY G4M Crop Model Digestibility Model Processing Global Forest model  Feed intake  MJ biofuel  Harvestable wood  Animal production  MJ bioelectric  Harvesting crops  GHG emissions  Coproducts LAND USE Worldwide: 7 animals Perennial crops Downscaled FAO FRA 18 crops (FAO + SPAM) (FAO + Gridded Short rotation coppice at grid level livestock) Management systems: Conversion Area low/high input & Ca le and buffalo technologies: Carbon stock irrigated Sheep and goat First generation biofuels Age Pig Second generation Tree size EU28: Poultry biofuels Species 9 additional crops, Rotation time crop rotations 8 different systems Biomass power plants Thinning Management options: fertilizer, irrigation, and tillage LAND COVER Cropland Grassland Short rotation Managed forest plantations Natural forest Other natural land Source: IIASA Gridded representation of world land use (fertilizer, water, costs) are based on the EPIC model described above (see Figure APP 11). Also crop models Lund-Potsdam-Jena managed Land model (LPJmL) is used and the forestry model G4M. • The level of production in a given area is determined by the agricultural or forestry productivity in that area (dependent on suitability and management), by market prices (reflecting the level of demand), and by the conditions and cost associated to conversion of the land, to expansion of the production and, when relevant, to international market access. Production is calibrated to match FAO statistics at the country level (FAOSTAT 2017). GLOBIOM captures production systems and land use in its base year (2000), using available historical data from SPAM (You and Wood 2006). PAGE 116 • The demand side is modeled at the regional level using 30 economic regions that consist of single countries or bundles of countries based on the food balance sheets developed by the FAO (FAOSTAT 2017; Alexandratos et al. 2006). For this version of the model, Zambia has been taken out of the region “Southern Africa” to further develop. • Trade flows are balanced out between different specific geographical regions based on the spatial equilibrium approach (Schneider et al. 2007; Takayama and Judge 1971). Trade is based purely on cost competitiveness, as goods are assumed to be homogeneous. • GLOBIOM uses a recursive dynamic approach combined with exogenous trends on population and economic growth to create future projections for key indicators, such as crop and livestock production and prices, land use change, greenhouse gas emissions and calorie availability. • GLOBIOM accounts for 10 sources of GHG emissions, including crop cultivation-related N2O emissions from fertilizer use, CH4 from rice cultivation, livestock-related CH4 emissions, CH4 and N2O emissions from manure management, N2O from manure applied on pasture, and above- and below-ground biomass CO2 emissions from biomass removal after converting forest and natural land to cropland. • Model adjustments for the Zambia context are described in Section 2.5. The Zambia region in GLOBIOM is modeled and optimized at 32 land units that consider differences in agro-ecological zones (AEZ) for a total of 57 land units. Structural model developments from Frank et al. (2018) were used to further develop the mitigation options for Zambia; assumptions of yield and mitigation potential of some of the mitigation technology options were further refined from Beach et al. (2015), and based on the stakeholder validation workshop held in April 2018. (ii) Description of BAU scenario and CSA adoption scenarios In order to evaluate the impact of CSA technologies on measurable goals, several assumptions about CSA technologies are made. The business as usual (BAU) scenario assumes that key trends from recent decades will continue in the future, including the application of conventional management practices on a large share of cropland area. As such, it presents the baseline to which other scenarios can be compared and is based on RALS data. Table APP 12 summarizes assumptions for modelling CSA. Three factors will determine CSA’s impact on agricultural and environmental indicators: (1) crop yield and GHG parameters, which reflects the direct impact of the CSA strategy per unit of land; (2) adoption rate, which reflects the size of the area on which the CSA strategy is used and (3) number of crops on which the CSA strategy is applied. PAGE 117 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table APP 11 Model assumptions for CSA technologies CSA technology 2010 adoption rate Crops included Productivity GHG impact per Adoption rate in (%) (BAU) impact (% hectare (% from 2050 (%) (CSA) increase from BAU) BAU) Minimum soil 8.9% of maize area Beans, cassava, Beans (40%), min -8% 50 disturbance chickpeas, cotton, cassava (17%), groundnuts, rice, chickpeas (90%), soya cotton (60%), groundnuts (43%), rice (61%), soya (45%) Residue retention 56.6% of crop area Maize, cotton Maize (24%), max +27% 80 cotton (3%) Conservation 0.3% of crop area Maize, cotton Maize (79%), -0.74% 30 agriculture cotton (42%) Agroforestry 5% of crop area Cotton, Cotton (18%), no GHG 25 groundnut, maize, groundnut (18%), mitigation(a) and soya maize (63%), and soya (18%) Drought-tolerant 23% of maize area Maize Mitigates negative no GHG 70 maize impacts from mitigation climate change Post-harvest Almost 0% All crops Figure APP 2(c) no GHG Not applicable losses reduction reached mitigation Crop Maize holdings of Maize, Soybean, Maize: production +7% Maize holdings diversification the Food Reserve Groundnut cost return to the are reduced and Agency (FRA) in pre-FSIP levels(b) would not exceed Zambia is about 300,000 MT after Soybean: 50% 500,000 MT 2020 reduction in production costs Groundnut: 50% reduction in production costs Source: IIASA using information from Thierfelder et al. (2013); Frank et al. (2018); Shitumbanuma (2012); Haggblade and Tembo (2003); Harman and Chapoto (2017) and stakeholder consultation Notes: (a) no GHG mitigation impact refers to 10 sources of GHG emissions (see above) which can be modelled with GLOBIOM; however, soil carbon sequestration or biomass growth or indirect effects through reduction of land use change or land under cultivation are not considered. (b) Traditionally, FISP reduces production cost for maize; to model crop diversification, it is assumed that production costs of maize return to pre-FSIP levels and costs for soybeans and groundnuts are decreased by 50%. (c) Figure APP 2 shows the level of post-harvest loss reduction, indicating increases in total production, which will results in increased production per hectare. These values are not specific for Zambia. They have been observed neighboring countries and serve as an approximation. Figure APP 2 Post-harvest loss diffusion scenario: 2010-2050 Dry beans 40 Cassava Potatoes Post-harvest loss reduction (%) Rice 30 Maize Groundnuts Sweet Potatoes 20 10 0 2010 2020 2030 2040 2050 Source: IIASA, using information from Affognon et al. (2015) PAGE 118 (iii) Sensitivity analysis: Climate change uncertainty To analyze the impact of climate change, sensitivity analyses are conducted to account for the high level of uncertainty associated with the impact of climate change on crop yields. The outputs of five general circulation models (GCMs) are combined with two crop models, EPIC and LPJmL, to obtain a bandwidth of the yield shock that is caused by climate change in comparison to a no climate change scenario. The GCMs are combined with different levels of atmospheric CO2 concentrations as prescribed by the representative concentration pathways (RCP) to show the impact of climate change on global changes in temperature and precipitation. We use RCP 8.5, which is the most extreme climate scenario (Riahi et al. 2011), for the period up to 2050. The GCMs include: • HadGEM2-ES—Hadley Global Environment Model 2—Earth System • IPSL-CM5A-LR—Earth System Model for the 5th IPCC report (Low Resolution) • GFDL-ESM2M—Geophysical Fluid Dynamics Laboratory Earth System Model • MIROC-ESM-CHEM—Model for Interdisciplinary Research on Climate (with atmospheric chemistry component) • NorESM1-M—Norwegian Earth System Model The effect of CO2-fertilization is debated. The scientific community has yet to reach an agreement on whether the potential benefits from increases in CO2 can be taken up and used by crops, especially if temperature and precipitation are expected to reduce crop yields. For this reason, a climate change scenario variant using the HadGEM2-ES model is included that assumes both with and without CO2 fertilization. Taken together crop yield results can show the potential range of the biophysical and economic impacts on crop yields from climate change. The simulated yield shocks only account for the effect of long-run changes in precipitation and temperature, but do not capture extreme climate effects or the potential increase in pests and diseases that affect plant growth. Projections of key productivity drivers vary significantly across models. Figure APP 3 depicts the change in mean annual temperature and mean annual precipitation between 1980/2000 and 2040/2060 in Zambia for all five GCMs. The figure demonstrates the high uncertainty in projecting climate change. The MIROC-ESM-CHEM and HadGEM2-ES climate scenarios project the most extreme increase in temperature of around 3.5 degrees Celsius, while the other three scenarios project that temperature will change by around 2-3 degrees Celsius. The MIROC-ESM-CHEM also projects the driest scenario, with a decrease in precipitation of more than 6 percent across Zambia, apart from the far north. In addition, GFDL-ESM2M and NorESM1-M project a decrease in rainfall but with variation across Zambia. In contrast, HadGEM2-ES and IPSL-CM5A-LR predict an increase in precipitation of more than 6 percent by 2040-2060, although the GCMs do not agree where in Zambia this will take place. As expected, different models yield highly diverging results across crops and regions. As discussed in Section 5.1 and shown in Figure 5.1, simulated crop yield changes dramatically across climate models for almost all crops in Zambia. It is important to take spatial effects into account, which is shown in Section 5.1, Figure 5.2. Although maize is grown throughout all regions in Zambia, most is produced in the Eastern Province, where yield loss is expected to be between 5 and 25 percent. PAGE 119 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure APP 3 Change of temperature (left) and precipitation (right) using five different GCMs with RCP 8.5 (1980/2000-2040/2060) GFDL−ESM2M HadGEM2−ES IPSL−CM5A−LR GFDL−ESM2M HadGEM2−ES IPSL−CM5A−LR MIROC−ESM−CHEM NorESM1−M MIROC−ESM−CHEM NorESM1−M Absolute change (K) % change 2 2.5 3 3.5 4 < −6% −6% to −4% −4% to −2% −2% to 2% 2% to 4% 4% to 6% >6% Source: IIASA using coupled Model Intercomparison Project Phase 5 (https://cmip.llnl.gov/cmip5) Note: (a) Difference in bias-corrected, average surface air temperature (absolute change K) over land and (b) relative difference in average annual rainfall (in percent) between 2040–2060 and 1980–2000 under RCP8.5 in the five climate models used in the ISI-MIP. To increase the visualization, the original 0.5° × 0.5° resolution has been smoothed to 5 x 5 arcmin. (iv) Sensitivity analysis: Shared Socioeconomic Pathways (SSPs) and results To model a business-as-usual (BAU) scenario for Zambia, we use the Shared Socio-economic Pathways (SSPs), which were developed as a backbone for climate change related assessments by a large consortium of researchers (Kriegler et al. 2012; van Vuuren et al. 2017; O’Neill et al. 2017). These scenarios provide a global context and/or template for scenarios at lower geographical levels and can be used to guide regional, national or sub-national planning (O’Neill et al. 2014). Sub-global scenarios can complement the SSPs with more regional contextualization of assumptions and results, even when using scenarios in the global setting. The SSPs are a set of plausible and alternative assumptions that describe potential future socio- economic development in the absence of climate policies or climate change. They consist of two elements: a narrative storyline and a quantification of key drivers. The scenarios are based on two dimensions of challenges, namely mitigation and adaptation (O’Neill et al. 2014) (Figure APP 4). The combination of challenges from which scenarios emerged were then constructed by identifying the drivers of the challenge outcomes such as population and urbanization (KC and Lutz 2017) and economic growth (Dellink et al. 2017) and building a rich narrative of each pathway using the quantitative and semi-quantitative drivers (O’Neill et al. 2017). Projections have global coverage and are provided at country level, including Zambia. Figure APP 4 Shared Socio-economic Pathways scenario framework Socioeconomic SSP 5: SSP 3: challenges for (mit. challenges dominate) (high challenges) mitigation Fossil-fueled development Regional rivalry Taking the highway SSP 2: A rocky road (intermediate challenges) SSP 1: Middle of the road SSP 4: (low challenges) (adapt. challenges dominate) Sustainability Inequality Taking the green road A road divided Source: O’Neill et al. (2017) Socioeconomic challenges for adaptation PAGE 120 For this business as usual scenario, the projections for SSP2: Middle of the Road (Fricko et al. 2017) is adopted. The narrative of SSP2 can be summarized as follows: “The world follows a path in which social, economic, and technological trends do not shift markedly from historical patterns. Development and income growth proceeds unevenly, with some countries making relatively good progress while others fall short of expectations. Most economies are politically stable. Globally connected markets function imperfectly. Global and national institutions work toward but make slow progress in achieving sustainable development goals, including improved living conditions and access to education, safe water, and health care. Technological development proceeds apace, but without fundamental breakthroughs. Environmental systems experience degradation, although there are some improvements and overall the intensity of resource and energy use declines. Even though fossil fuel dependency decreases slowly, there is no reluctance to use unconventional fossil resources. Global population growth is moderate and levels off in the second half of the century as a consequence of completion of the demographic transition. However, education investments are not high enough to accelerate the transition to low fertility rates in low-income countries and to rapidly slow population growth. This growth, along with income inequality that persists or improves only slowly, continuing societal stratification, and limited social cohesion, maintain challenges to reducing vulnerability to societal and environmental changes and constrain significant advances in sustainable development. These moderate development trends leave the world, on average, facing moderate challenges to mitigation and adaptation, but with significant heterogeneities across and within countries.” (O’Neill et al. 2017). The SSPs do not include projections for crop yields and livestock feed convergence efficiency, which are essential to model future agricultural productivity growth. Yield projections are used from Herrero et al. (2014) and Fricko et al. (2017) and are estimated using the historical relationship between GDP growth and crop yield increase. It is assumed that that these projections represent intrinsic productivity rates and reflect the increase in yield because of advances in knowledge and new technologies. Livestock feed conversion efficiencies are used from Herrero et al. (2014) and Fricko et al. (2017). However, there is less information available on future trajectories of livestock feed conversion efficiencies (the amount of feed required per livestock category such as dairy, ruminant meat, pork, and poultry) and how it could be affected climate change (e.g., Bouwman et al. 2005; Wirsenius et al. 2010). To investigate the uncertainties in the socioeconomic projections, the results of two additional SSP scenarios are modeled and compared: SSP1: Sustainability—Taking the green road and SSP3: Regional rivalry—A rocky road, which represent two opposing types of scenarios” to represent the bandwidth of the socioeconomic spectrum, including uncertainties related to both external (global markets and the political situation) and internal conditions “SSP1 with its central features of commitment to achieving development goals, increasing environmental awareness in societies around the world, and a gradual move toward less resource-intensive lifestyles, constitutes a break with recent history in which emerging economies have followed the resource-intensive development model of industrialized countries. To some extent, elements of this scenario can already be found in the proliferation of ‘green growth’ and ‘green economy’ strategies in industrialized and developing countries.” (O’Neill et al. 2017). And SSP3 “with its theme of international fragmentation and a world characterized by regional rivalry can already be seen in some of the current regional rivalries and conflicts, but contrasts with globalization trends in other areas. It is based on the assumption that these globalization trends can be reversed by a number of events” (O’Neill et al. 2017). Figures APP 5–APP 8 show crop yield projections, production, land use change, and emissions from synthetic fertilizers for the BAU scenario SSP2 as well as for SSP1 and SSP3. The figures show that for all indicators, the SSP1 scenario is more “positive” than the BAU, while SSP3 shows more “negative” results. As a consequence of high GDP growth in the SSP1 scenario, technological change is faster, resulting PAGE 121 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN in higher yields for all crops. The reverse is the case for SSP3. Due to relatively low population growth, which in turn results in low demand, total production is projected to be the lowest in SSP1. The pattern for SSP3 resembles that of SSP2. Cropland expansion is also expected to be the lowest in SSP1 because of the higher yields and lower demand. Again, a reverse pattern can be observed for SSP3. Interestingly, the expansion of cropland under the SSP3 scenario is so large that it will overshoot the maximum as set by the vision target. For emissions, SSP1 results in the lowest amount of emissions but this is not close to the level needed to reach targets as set by the NDC (which are in comparison with the BAU scenario). Figure APP 5 Crop yield under different socioeconomic scenarios Cassava Groundnuts 1.5 16 12 1.0 8 0.5 tons/ha 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Maize Millet 5 4 1.6 3 1.2 2 0.8 1 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Historical SSP1 SSP2 SSP3 Figure APP 6 Land use change under different socioeconomic scenarios Cropland Forest 2,500 44,000 2,000 43,000 1,500 1,000 42,000 Area (1000) ha 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Grassland Natural land 3,000 22,000 2,700 20,000 2,400 18,000 2,100 1,800 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 1960 1970 1980 1990 2000 2010 2020 2030 2040 2050 Historical SSP1 SSP2 SSP3 PAGE 122 Figure APP 7 Crop production under different Figure APP 8 Emissions from synthetic fertilizers socioeconomic scenarios under different socioeconomic scenarios 8,000 8 Production (1000 tons dry matter) 6,000 6 Mt CO2eq/yr 4,000 4 2,000 2 0.00 0 1961 1970 1980 1990 2000 2010 2020 2030 2040 2050 1961 1970 1980 1990 2000 2010 2020 2030 2040 2050 Historical SSP1 SSP2 SSP3 Historical SSP1 SSP2 SSP3 Sources (Figures APP 5–8): Historical values are from FAOSTAT; the GLOBIOM model provides projections from 2000-2050; “Vision” target represents agriculture sector goals in 2050 which were agreed on during the stakeholder workshop and/or policy documents. E.2 Detailed Results: Output Tables In this section, more detailed outputs of the GLOBIOM model are presented. Figure APP 9 provides a spatial representation of land use patterns, which are described in Section 5.4. Table APP 12–Table APP 16 provide summaries of GLOBIOM outputs as they pertain to production, land use, emissions, yield, and net trade. In each case, the 2010 values are contrasted with 2050 targets (as presented in Section 2, Table 2.2); we present the 2050 projections for the business as usual scenario, without and with climate change; the 2050 projections for average CSA adoption, without and with climate change. We calculate the average across CSA practices: agroforestry, conservation agriculture, crop diversification, drought-tolerant maize, minimum soil disturbance, residue retention and reduction of post-harvest loss. We present the maximum, average, and minimum values calculated across all 5 GCMs and 2 crop models, to show the range of projected impacts. In the last columns, the change of CSA compared to the application of conventional practices in 2050 in a without climate change and with climate change scenarios are shown; as well as compared to the agriculture sector targets. For all values, SSP2 was used as a default configuration. PAGE 123 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Figure APP 9 Projected land use changes between 2000 and 2050, in percent (a) Cropland (b) Grassland 2000 2050 2000 2050 20 20 20 20 40 40 40 40 60 60 60 60 80 80 80 80 100 100 100 100 2000−2050 difference in pp 2000−2050 difference in pp 20 40 15 20 10 5 0 0 (c) Forest (d) Other land 2000 2050 2000 2050 20 20 20 20 40 40 40 40 60 60 60 60 80 80 80 80 100 100 100 100 2000−2050 difference in pp 2000−2050 difference in pp −20 −10 −40 −20 −60 Table APP 12 Land use in 1000 hectares, by land use category CSA vs. CSA vs. 2050 Conventional practices 2050 CSA conventional target in 2050 practices in 2050 No No Climate Climate No Climate Change No Climate Change Climate Climate Change Change Climate Climate Change Change 2010 2050 Change Change % change % change Min. Avg. Max. Min. Avg. Max. Land Types Target Target across averages across averages Total 68,263 N/A 67,693 64,150 67,714 70,996 67,764 66,684 67,849 69,032 0.10% 0.20% N/A N/A Forest 44,555 N/A 42,041 41,371 42,026 42,989 41,660 41,460 41,650 41,756 -0.91% -0.89% N/A N/A Grassland 19,449 N/A 20,742 19,122 20,808 22,378 20,678 20,101 20,819 21,160 -0.31% 0.06% N/A N/A Nat. land 2,786 N/A 2,628 1,888 2,629 2,962 2,962 2,962 2,962 2,962 12.70% 12.66% N/A N/A Cropland 1,473 2,373 2,281 1,769 2,252 2,667 2,464 2,161 2,418 3,154 8.01% 7.38% 4% 2% Source: World Bank, based on GLOBIOM model outputs; values for 2010 are based on FAOSTAT data. Note: CSA refers to average values of following practices: agroforestry, conservation agriculture, crop diversification, drought-tolerant maize, minimum soil disturbance, residue retention and reduction of post-harvest loss. Colors indicate the relative change of land use as a result of CSA; dark green indicates the most positive change (across land types), dark red the least positive change. PAGE 124 Table APP 13 Production in 1000 tons, by crop CSA vs. CSA vs. 2050 Conventional practices 2050 CSA conventional target in 2050 practices in 2050 No No Climate Climate No Climate Change No Climate Change Climate Climate Change Change Climate Climate Change Change 2010 2050 Change Change % change % change Min. Avg. Max. Min. Avg. Max. Crop Target Target across averages across averages Total 2,788 5,576 6,548 5,700 6,870 7,998 6,933 5,685 7,313 10,085 6% 6% 24% 31% Maize 1,281 2,562 4,051.56 3,907.40 4,024.25 4,073.06 4,277.15 3,912.99 4,268.29 4,610.60 6% 6% 67% 67% Sugarcane 758 1,517 1,124.72 520.95 1,465.46 2,413.55 1,156.86 520.95 1,544.02 3,173.92 3% 5% -24% 2% Cassava 259 519 624.89 613.74 640.69 669.02 642.99 608.33 658.96 757.01 3% 3% 24% 27% Wheat 123 246 18.37 12.93 23.51 50.50 18.60 0.14 21.01 50.49 1% -11% -92% -91% Groundnut 84 169 205.59 168.47 188.41 210.38 272.35 168.47 249.71 631.60 32% 33% 61% 48% Soya 83 165 59.06 58.23 61.93 64.20 82.83 58.23 87.26 232.04 40% 41% -50% -47% Millet 44 89 188.64 160.02 176.83 197.69 194.00 160.13 181.89 237.28 3% 3% 118% 105% Cotton 42 85 72.10 66.46 77.82 87.89 76.90 66.45 82.97 105.76 7% 7% -9% -2% Sweet 37 75 36.39 31.17 36.75 41.45 38.26 31.17 38.51 55.27 5% 5% -49% -49% potato Rice 33 67 37.07 35.65 37.92 39.98 39.25 35.54 40.14 48.24 6% 6% -41% -40% Sorghum 18 36 74.77 76.76 80.05 86.02 77.37 74.22 82.57 106.39 3% 3% 112% 127% Sunflower 14 27 46.76 46.03 49.22 51.62 47.57 46.03 50.06 57.86 2% 2% 73% 82% Barley 7 15 0.15 0.15 1.77 4.17 1.09 0.15 2.26 8.89 646% 27% -93% -84% Potato 3 6 8.00 1.78 4.98 8.48 8.18 1.78 5.59 9.86 2% 12% 47% 0% Source: World Bank, based on GLOBIOM model outputs; values for 2010 are based on FAOSTAT data. Note: CSA refers to average values of following practices: agroforestry, conservation agriculture, crop diversification, drought-tolerant maize, minimum soil disturbance, residue retention and reduction of post-harvest loss. Colors indicate the relative change of production as a result of CSA; dark green indicates the most positive change, dark red the least positive change. Table APP 14 Emissions in MtCO2e/yr, by emission source CSA vs. CSA vs. 2050 Conventional practices 2050 CSA conventional target in 2050 practices in 2050 No Sub- 2050 Climate Limited No Climate Change No Climate Change Climate stantial 2050 Target Change support Climate Climate Change support Target (sub- 2010 (limited stantial Change Change % change % change Min. Avg. Max. Min. Avg. Max. Crop Target support) support) across averages across averages Total LULUCF 6.4 28.5 19.3 37.6 35.5 38.0 39.9 37.5 35.1 38.2 41.8 -0.32% 0.50% 25% 50% Deforestation 1.8 N/A N/A 23.9 22.1 23.3 23.9 23.8 21.9 23.3 23.9 -0.07% -0.18% N/A N/A Other land use 1.5 N/A N/A 12.7 12.7 13.7 14.9 12.7 12.7 13.9 15.9 -0.28% 15.59% N/A N/A change Sum of land 3.2 N/A N/A 1.0 0.7 1.0 1.1 1.0 0.5 1.1 2.1 -6.58% 109.59% N/A N/A use change Total 4.5 5.3 3.8 7.1 7.0 7.2 7.4 7.2 6.9 7.2 8.0 0.91% 0.91% 36% 92% agriculture Enteric 2.4 N/A N/A 3.6 3.6 3.6 3.7 3.6 3.6 3.7 3.8 0.21% 0.30% N/A N/A fermentation Manure left on 1.5 N/A N/A 0.1 2.1 2.1 2.2 0.1 2.1 2.1 2.2 0.16% 0.00% N/A N/A Pasture Synthetic 0.3 N/A N/A 2.1 0.9 0.9 1.0 2.1 0.8 1.0 1.4 0.17% 3.58% N/A N/A Fertilizers Manure 0.1 N/A N/A 0.1 0.1 0.2 0.2 0.1 0.1 0.1 0.2 0.19% -7.40% N/A N/A Mgmt. CH4 Manure applied to 0.1 N/A N/A 0.1 0.1 0.1 0.2 0.1 0.1 0.2 0.4 0.22% 21.88% N/A N/A Soils Manure Management 0.1 N/A N/A 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.1 0.00% 1.15% N/A N/A N20 Rice 0.1 N/A N/A 0.9 0.1 0.1 0.1 1.0 0.1 0.1 0.1 5.59% 0.01% N/A N/A Cultivation Source: World Bank, based on GLOBIOM model outputs; values for 2010 are based on FAOSTAT data. Note: CSA refers to average values of following practices: agroforestry, conservation agriculture, crop diversification, drought-tolerant maize, minimum soil disturbance, residue retention and reduction of post-harvest loss. Colors indicate the relative change of emissions as a result of CSA; dark green indicates the most positive change, dark red the least positive change. PAGE 125 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table APP 15 Yield in tons per hectare per year, by crop CSA vs. CSA vs. 2050 Conventional practices 2050 CSA conventional target in 2050 practices in 2050 No No Climate Climate No Climate Change No Climate Change Climate Climate Change Change Climate Climate Change Change Crop 2010 2050 Change Change % change % change Min. Avg. Max. Min. Avg. Max. (net trade) Target Target across averages across averages Total 10 20 8.8 14 15 17 Cassava 6.5 11.3 8.75 N/A N/A N/A 8.75 8.67 8.97 9.97 N/A N/A N/A -21% Maize 2.1 4.9 3.87 N/A N/A N/A 3.87 3.78 4.03 4.75 N/A N/A N/A -18% Millet 0.9 1.9 1.10 N/A N/A N/A 1.10 0.96 1.10 1.32 N/A N/A N/A -43% Groundnuts 0.5 1.7 0.76 N/A N/A N/A 0.76 0.75 0.80 0.89 N/A N/A N/A -53% Source: World Bank, based on GLOBIOM model outputs; values for 2010 are based on FAOSTAT data. Note: CSA refers to average values of following practices: agroforestry, conservation agriculture, crop diversification, drought-tolerant maize, minimum soil disturbance, residue retention and reduction of post-harvest loss. Colors indicate the relative change of yield as a result of CSA; dark green indicates the most positive change, dark red the least positive change. Table APP 16 Net trade in 1000 tons CSA vs. CSA vs. 2050 Conventional practices 2050 CSA conventional target in 2050 practices in 2050 No No Climate Climate No Climate Change No Climate Change Climate Climate Change Change Climate Climate Change Change Crop 2010 2050 Change Change % change % change Min. Avg. Max. Min. Avg. Max. (net trade) Target Target across averages across averages Total 498 996 -1,066 (1,835) (1,034) (402) (831) (2,053) (747) 2,410 -22% -28% -183% -175% Sugarcane 468 N/A (0.99) (383.31) 88.56 440.85 5.83 (468.91) 184.01 2,181.21 -689% 108% N/A N/A Cassava 54 N/A 25.45 25.14 87.53 163.26 75.10 (0.16) 135.58 404.22 195% 55% N/A N/A Cotton 37 N/A 49.40 34.61 50.90 68.59 53.36 13.76 53.29 88.22 8% 5% N/A N/A Maize 28 N/A 152.19 36.96 103.33 152.07 223.73 35.05 169.09 314.78 47% 64% N/A N/A Sweet 14 N/A (171.37) (203.34) (178.33) (151.94) (162.85) (203.34) (169.82) (96.75) -5% -5% N/A N/A potatoes Millet 2 N/A 81.75 27.70 60.46 92.62 85.82 17.04 63.77 137.05 5% 5% N/A N/A Bovine meat -1 N/A 9.72 6.64 8.85 9.98 10.02 4.22 8.70 10.09 3% -2% N/A N/A Soybeans -4 N/A (259.66) (260.59) (256.84) (253.96) (233.25) (260.59) (228.34) (67.46) -10% -11% N/A N/A Wheat -7 N/A (458.07) (464.91) (447.97) (420.27) (456.42) (486.47) (451.37) (420.29) 0% 1% N/A N/A Pig meat -10 N/A (96.56) (108.24) (99.99) (95.21) (95.70) (108.24) (98.79) (77.90) -1% -1% N/A N/A Rice -26 N/A (160.68) (163.40) (159.72) (155.19) (158.12) (164.76) (157.20) (147.54) -2% -2% N/A N/A Groundnuts -58 N/A (236.97) (382.55) (291.29) (252.67) (178.15) (430.69) (255.49) 84.67 -25% -12% N/A N/A Source: World Bank, based on GLOBIOM model outputs; values for 2010 are based on FAOSTAT data. Note: CSA refers to average values of following practices: agroforestry, conservation agriculture, crop diversification, drought-tolerant maize, minimum soil disturbance, residue retention and reduction of post-harvest loss. Colors indicate the relative change of net trade as a result of CSA; dark green indicates the most positive change, dark red the least positive change. PAGE 126 Appendix F: Public Expenditures on CSA Projects in Zambia Table APP 17 provides an overview of all projects which promoted climate-smart agriculture in Zambia. The results stem from a survey conducted in the course of this project and may not be complete. As of 2017, major donors in Zambia dispersed a total of US$172,817,651 in Zambia through 80 agriculture and natural resource related projects (von der Decken, 2017). Based on interviews with 25 major donors and project implementers, including public sector entities, it is estimated that US$118 million was expended in 2017 on 38 different CSA related projects, reaching a total of over 1.6 million beneficiaries (Table APP 17). While this table is not a complete inventory of all CSA projects in Zambia, it provides an indication of the level of expenditure and beneficiary numbers. Table APP 17 Summary of CSA donor projects, beneficiary numbers, and 2017 expenditures Expenditure 2017 Donor/Implementer Project Name Beneficiaries (US$) World Food Programme Rural Resilience Initiative (R4) 17, 835 5,300,000 Grassroots trust Shaping a Viable Future 250,000 5,000 Grassroots trust Legume Intercropping 2,500 5,000 Grassroots trust Farmer Managed Natural Regeneration 25,000 3,000 Grassroots trust Holistic Planned Grazing 20,000 2,000 ICRAF Sustainable Agriculture Intensification N/A 350,000 ICRAF Sustainable Land Management N/A 300,000 CFU Climate-Smart Agriculture Zambia (CSAZ) 600,000 25,000,000 CFU Regional Conservation Agriculture Program N/A N/A GEF-strengthening Management Effectiveness and Generating Multiple CFU Environmental Benefits Within and Around 40,000 800,000 the Greater Kafue National Park and West Lunga National Park in Zambia SNV Sustainable Land Management SIDA 15,000 4,000,000 Systems (SLMS) Project Community-based Seed Enterprises and Self Help Africa 2,500 4,000 Participatory Crop Improvements, in all AEZs Integrated Livelihoods Programme in the Self Help Africa 17,000 20,000,000 Northern Part of Zambia Strengthening Climate Resilience in the Self Help Africa Kafue Sub Basin (SCRika) as CRAF Under 164,000 N/A Government Kaoma Farm Development Project in Kaoma Self Help Africa 150 1,500 Western Province USAID COMACO 40,000 N/A Partnership for Innovations: Legume Seed USAID 50,000 3,400,000 Development for Drought Areas IFAD Smallholder Productivity Promotion (S3P) 73,000 33,400,000 Enhancing Smallholder Livestock IFAD 180,000 15,100,000 Investment Project (E-SLIP) Empowering Local Communities to Adapt Finland 400 145,000 Zambezi Sustaining Environmental and Natural Finland 400 145,000 Resources in Kawambwa continued PAGE 127 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Table APP 17 continued Deepening Community Based Natural Finland 4,000 145,000 Resources Management in Zambia Improved Environment and Natural Resources Management through Promotion Finland 2,000 137,000 of Sustainable Organic and Other Practices/ Technologies Strengthening integrated adaptation planning and implementation in Southern FAO 6,000 400,000 Africa smallholder agriculture systems to support food security Conservation Agriculture Scaling Up Project European Union N/A 2,000,000 (CASU) Agriculture Productivity and Enhancement AFDB 75,000 1,500,000 Project (APMEP) AFDB Smallholder Irrigation Project (SIP) 5,000 2,000,000 AFDB Livestock Infrastructure Support Project 35,000 1,200,000 Strengthening Climate Resilience in the AFDB 70,000 1,500,000 Kafue Sub Basin (SCRika) Scaling up Production Food Security and CRS 5,600 350,000 Climate Adaptation CRS Farmers Access to Market Engagement 600 400,000 Enhanced Livestock Trade and Enterprise Heifer International 3,000 N/A Project Phase III (ELITE III) Kasisi Sustainable Organic Agriculture Support 100 550,000 Kasisi Seed Knowledge Initiative 150 10,000 Push-pull project-involves pest Kasisi 47 20,000 management- using biological methods Kasisi FIBL-concerned with soil health 100 120,000 Total 1,686,547 118,292,500 Source: FAO, based on interviews with development partners conducted between July-December 2017 PAGE 128 Appendix G: Detailed Description of Delivery Mechanisms This appendix expands on Section 8.2. and provides a more detailed description of the mechanisms. Weather Index Insurance—Combining Risk Transfer with Risk Mitigation Weather index insurance (WII) is a risk transfer mechanism that can enhance livelihood resilience. Short-term shocks, such as seasonal reductions in rainfall or extreme heat, can have long-term consequences (Carter et al. 2014). To cope with resulting losses, farmers tend to reduce spending on food, which affects health and education outcomes, and exacerbates poverty. WII compensates policyholders based on a weather index that approximates losses at a particular location and period of time. As the value of the weather index surpasses a threshold and approaches the maximum limit, payment increases at an established rate (for instance, US$15 is paid for each millimeter of rainfall between threshold of 100 millimeters and 50 millimeters). WII faces several challenges: (i) Basis risk is the greatest challenge and refers to the difference between the payout as measured by the index and the actual loss incurred by farmers. Since losses are approximated by the weather index, policyholders may not be paid when they suffer losses or they may receive a payment when they have suffered no loss. (ii) Sufficient historical weather data is necessary to establish an index. The integrity of weather stations and accurate recording of data must be ensured. (iii) Capacity building and sensitization for farmers, insurers, and regulators about the functioning of WII and the associated costs and benefits are crucial. (iv) A strict regulatory and legal framework is needed to ensure policyholders’ interests are protected by law (World Bank 2011b). In Zambia, the World Food Program (WFP) pairs WII with risk mitigation strategies and mandatory CSA adoption. This is the Rural Resilience Initiative (R4) program, which has four components: improved resource management and adoption of CSA (risk mitigation), insurance (risk transfer), livelihood diversification and microcredit (prudent risk taking), and savings (risk reserves) (WFP and OXFAM 2017). In 2017, 3,835 farmers, 50 percent of whom were women, were insured through the R4 initiative. The program offered financial education and collaborated with Zambia’s Meteorology Department to strengthen product design and data management. An impact evaluation of R4 in Ethiopia demonstrated that the negative consequences of climate shocks on food security decreased. In Zambia, participants managed to increase their savings six-fold compared to 2016. The program and resulting network of farmer provides an opportunity to roll out training for sustainable management practices and crop diversification. Similar WII initiatives are underway. For instance, alongside the electronic FISP which plans to scale up the scheme on a nearly national level and has the potential to incentvize farmers’ adoption of CSA practices. WII was piloted as part of the Farmer Input Support Programme. The number of policies sold and the sum insured increased from less than 20,000 and US$2 million in 2016/17 to over 1 million and US$176 million respectively in 2017/18. This rapid scale-up was achieved by adding an index insurance cover to E-FISP, the digitized version of the government’s agricultural subsidies program (World Bank, forthcoming). At the same time the scheme faces challenges with respect to transparency in payments, lack of capacity of farmers and implementors and questions related to the actual consumer demand for the insurance. Box APP 1 presents several activties the PAGE 129 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN public sector can undertake to promote a transparent and effective WII scheme which incentivizes the adoption of CSA as risk mitigation strategies, and support private sector particpation. Box APP 1 Weather Index Insurance. To maximize finance for development and crowd-in private resources to implement a WII scheme, the public sector could consider to support: • Capacity building and training for farmers, insurers, and regulators. Index insurance is a new concept for many farmers, and a rollout of the product requires intense education programs to help them understand the principle. Capacity building on risk transfer strategies should be combined with training on risk mitigation stratgies, such as adopting of climate-smart agriculture practices. Insurers may require technical assistance and extensive capacity building to enable them to undertake product development on a sustainable basis. Public sector support could include the development of guidelines and training modules for regulators and farmers, sensitization for involvement of local saving groups, and technical support to insurers to improve the contract design and meet the local realities. • Ensure data availability and functioning, and integrity of weather station. Availability of historical, accurate, and complete data is crucial to design the index. The network of weather stations should be functional and automatic to ensure security and integrity, and avoid data tampering. The degree of integrity affects the cost of the uncertainty that would be reflected in the insurance premium. An improved weather data network could enhance the existing synergies with remote sensing data. • Provide technical and operational support with technical reviews of existing schemes to build on lessons learned, feasibility studies that enhance targeting and product and scheme design, as well as monitoring and evaluation of the scheme. • Reduce legal and regulatory risks to the policyholder and insurer. Before introducing the scheme, comprehensive legal and regulatory reviews should be undertaken to identify risks to the policyholders and insurers and identify risk mitigation options. A sound legal and regulatory framework ensures that contracts are legal and enforceable. Otherwise, there is a risk that WII development costs will be wasted or that policyholders will find themselves without enforceable rights. The review may also reveal that the regulator should impose special conditions on insurers to ensure that policyholders are adequately protected. • Appoint an independent agency that, on behalf of the insured parties, can verify the accuracy of the payouts triggered. In schemes with many beneficiaries or an index that is not easily verifiable by the parties (e.g., based on satellite data), an independent agency could be appointed to verify that contracts have been carried out correctly. Source: World Bank (2011b) Business Partnerships with Rural Communities—Leveraging Carbon Finance In Zambia, COMACO is a well-known example of a business model that addresses poverty, food insecurity, and conservation goals. COMACO promotes CSA practices such as organic farming, legume food crops to enhance soil nutrition, and crop rotation with agroforestry, as well as by developing community markets and premium pricing of 10 percent for farmers who apply CSA practices. COMACO specifically targets households with high food insecurity or whose practices lead to natural resource depletion (Lewis et al. 2011). COMACO addresses constraints of capacity building to enhance adoption of CSA and provides access to output markets. At regional trading centers, the raw commodities are processed into value-added products, such as peanut butter, jams, or dried fruits, or sold in larger quantities to different markets. By providing a reliable market, COMACO also supports efforts to reduce post-harvest losses and related emissions. Farmers reported that crop yields had increased between 16 percent for soybeans and 31 percent for maize and experienced less water stress (Mfune et al. 2016). COMACO supports the adoption of CSA practices and forest conservation practices. By reducing deforestation and enhancing soil carbon sequestration, COMACO generates emission reduction credits, eligible for carbon finance. The promotion of CSA is expected to enhance soil carbon sequestration and reduce deforestation. To enhance impact on forest conservation, COMACO promotes the PAGE 130 establishment of Green Zones, where sustainably harvested non-traditional forest products (apiaries for honey and wild mushrooms) and alternatives to destructive charcoal are implemented (World Bank 2018c). The aim is a reduction of GHG emissions on approximately 200,000 ha, of which 15,000 ha should fall under CSA practices. Following the first successful verification of carbon credits in October 2017, US$ 814,406 was paid by BioCarbon Fund to participating communities and COMACO. Payments through the carbon fund are results-based. To sustain operations, COMACO receives support from donors. Monitoring, supervision, and reporting to verify GHG emission reductions demands a high level of technical capacity and support as well as human and financial resources. COMACO receives funding from different donors, which allows COMACO to set up emission reduction purchase agreements between rural communities and the BioCarbon Fund, as well as to finance related supervision activities and implement climate-smart agriculture and forest practices. Without donor funding, the financial viability of the business model in the early years may be jeopardized. Box APP 2 present strategies fo rthe public sector to support similar business-community partnerships. Box APP 2 Business Partnerships with Rural Communities. Business partnerships with rural communities, which promote CSA practices, can be supported with the following activities: • Ensure that the project is undertaken in accordance with regional and community development plans to maximize development and environmental impacts. Assist communities in engaging sustainable land management practices, which results in verifiable emission reduction. Sensitization on climate mitigation and the results-based nature of carbon finance could play a role in strengthening farmers’ commitment to the project. • Provide capacity building and training for farmers to adopt CSA and business skills. While the agribusiness is likely to provide training on CSA, the public sector can support or complement this by strengthening public extension workers’ capacity and facilities, providing training on CSA, farming as a business (marketing, transport, quality control), and post-harvest handling, supporting the establishment of revolving funds or savings groups, and organizing agricultural fairs to spread information about the business model. • Promote business linkages with producer groups, agro-dealers, and storage facilities; conduct farmer trade fairs; and conduct market research of target crops to facilitate market access and buisness linkages. • Support access to credit by enhancing the capacity of commercial banks and strengthening the policy and regulatory environment. In Zambia, financial and non-financial lending institutes have limited institutional capacity to work with smallholder farmers. Support could include providing credit guarantees, building institutional capacity, supporting business partnerships between agribusinesses and financial institutes to better serve smallholders, or support growth of warehouse receipts financing, which is still in its infancy in Zambia (World Bank forthcoming). • Invest in public goods and services considering private sector needs. Public good investments such as rural roads and access to energy reduce transaction costs and also support commercialization and the effectiveness of such business partnerships. Investment in agricultural research and development and dissemination of new varieties supports the development of new markets and further business partnerships in the long run. Outgrower Schemes—Commercializing Horticulture Production Outgrower schemes are a form of contract farming whereby binding agreements are made between producers and purchasers of agricultural products. Outgrower schemes are binding agreements, a coordinated commercial relationship. Farmers are linked with a large farm or processing plant that supports production planning, input supply, and adoption of new technologies including CSA, extension advice, and transport. The arrangement enables firms to source a product in a controlled way, obtain financial benefits (e.g., tax breaks, duty-free imports of machinery), and gain access to local markets that might not have been explored otherwise. Farmers benefit from access to markets and financial PAGE 131 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN services, reduced price fluctuations, and support in achieving quality standards and technology, skills and knowledge transfers, and opportunities for seasonal employment. For a successful relationship, mutual trust and long-term business interests are required (Felgenhauer and Wolter 2009). Outgrower schemes have been operated with varying degrees of success. An example in outgrower schemes for sugar cane in Mazabuka, southern Zambia showed that participants earned up to US$2,999 in 2012-2013 compared to US$500 in neighboring communities (Matenga 2017). At the same time, smallholders also depend on the contracting company, with asymmetries of power reflected in quality assessment and pricing of the product; there may be limited transparency and disputes over prices and contract conditions (World Bank and UNCTAD, 2004). The community in Mazabuka has experienced tension related to the significant role the scheme’s income plays in the local community, in some instances benefiting some community members at the expense of others. The study showed that men tend to capture outgrower revenues while women tend to focus on food crop production (Matenga 2017). Outgrower schemes can be combined with all types of products but often are used to source horticulture products and are combined with access to small-scale irrigation. In Zambia, small- scale irrigation often requires development and rehabilitation of small water infrastructure, permanent weirs and small-scale earth dams, and development of boreholes or solar pumps. In rural communities these are often supervised by water users associations (WUAs). Box APP 3 provides an overview of public sector activties which could support the emergence of fair and effective outgrower schemes. Box APP 3 Outgrower Schemes. Although this contracting model is driven by the private sector, there is a role for the public sector: • Promote private sector alignment with the principles of responsible investment and ensure compliance with social and environmental standards. • Review and improve investment climate and policy and regulatory frameworks. • Support effective public and private sector dialogue to build (regional) markets. • Support contract design and enforcement. The public sector can ensure contracts and agreements between parties are fair and transparent and offer access to legal services and advice. • Support smallholder capacity and coordination, to enable farmers to engage gainfully in commercial value chains. Support the development and strengthening of producer organizations that enhance knowledge transfer, bargaining power, and transaction costs for private companies. • Improve land tenure security and access to land. Only 6 percent of smallholders in Zambia have some form of land documentation. Most land is under the customary land regime (Hall et al. 2017). Enhancing land tenure security by improving capacity and performance of land institutions, clarifying land policies, and streamlining registration processes could facilitate participation in an outgrower scheme and avoid conflicts between the business and local communities. • Invest in public goods and services considering private sector needs. Public good investments such as rural roads and access to energy reduce transaction costs, support commercialization, as well as the effectiveness of such business partnerships. • Provide irrigation infrastructure and governance structure. To support horticulture, irrigation should be introduced. The public sector can provide support in the design, procurement and supervision of construction of off-farm water infrastructure design and technical support as well as procurement and supervision of construction. In addition, the public sector can improve water management, access, and governance at various levels, such as building capacity of extension staff and communities or supporting the establishment and strengthening of the WUAs that are the local-level institutional entry points for efficient management of water resources. • Enhance public inspections and quality assurance to attract agribusinesses, such as by investing in inspection services for inputs agro-dealers and in labeling programs; strengthening the capacity of public laboratories; strengthening sanitary and phytosanitary standards (SPS), regulations, and certification; and reviewing and revising trade policies and management frameworks for strategic commodities to attract private companies and support exports. PAGE 132 Pluralistic Participatory Extension Approach—Supporting Linkages Between Research and Extension The pluralistic participatory extension (PPE) approach supports different sources of advisory services to disseminate innovative CSA technologies. The results reveal that even under CSA, crop yields may not increase sufficiently to meet agriculture sector goals (Section 5). This calls for innovative approaches for agricultural extension and research. Extension and agricultural advisory services are key for spreading the adoption of CSA and achieving development goals. International evidence shows that agricultural extension visits can reduce poverty (e.g., Nkonya et al. 2009; Dercon et al. 2008). There are several extension models, such as the traditional supply-driven public extension model, private extension services including NGOs, and demand-driven, participatory, and pluralistic services. The latter recognizes the diversity of farmers and farming systems and is characterized by the coexistence of multiple public and private sector approaches, providers, funding streams, service types, and sources of information and experiences. It can include membership-based farmer organizations, private or commercial enterprises, and NGOs. Each has the potential to provide complementary services and to thereby contribute to long-term sustainability of advisory service delivery to farmers (Heemskerk and Davis 2012). The PPE approach is way to strengthen partnerships between agricultural research, extension services, and farmers. PPE provides the opportunity for farmers’ “demand articulation” to better identify the needs of different user groups and help transfer research into goods and services. This can include on-farm testing and sharing of experience with research results, development of technical options including the development and testing of climate-resilient solutions, through farmer-led trials and demonstrations. Sharing research results through PPE includes promoting farmer-to-farmer learning, farmer organization knowledge exchange networks, and access to information platforms using ICTs. In this context, farmers are organized and empowered to demand quality support services. An example is Senegal’s Agricultural Services and Producer Organizations Project (1999-2011), which established a network of producer organization, strengthened research capacity, and supported the provision of demand-driven agricultural services provided by the private sector. The project improved the quality and price of groundnuts, improved the level and quality of community seed stocks, raised agricultural income by 12 percent, and raised non-farm household income. In addition, the agricultural research Box APP 4 PPE relies on a range of service providers and a range of approaches including Farmer Field Schools. The public sector can play a role to facilitate private sector involvement: • Coordinate pluralistic extension services, thereby strengthening the market orientation of services, coordination between service providers, and interaction and learning between public and private service providers; improve targeting and differentiation of services for different farmers; and enhance downward accountability to farmers and performance-based contracts. • Develop sectoral or local policy that supports private sector service provision. • Develop integrated planning skills for the local governments, to ensure efficient coordination and management of the PPE approach and partnership with research (e.g., provide two channels for research proposals from researchers and users). • Improve capacity of extension workers, government and digitize government procurement for greater transparency; upgrade research and extension facilities where needed. • Support farmer capacity building, including the setting up of FFS: includes demonstrations, sensitization on the topic, identification of farmers and formation and production inputs and learning material for CSA, sensitization and awareness meetings on CSA; prioritization studies for crops; clean materials for establishing demonstration plots, seed multiplication schemes. • Support development and strengthening of farmer organizations to better articulate farmers’ demands, which allows for better matching of service providers with organizations and which improves service delivery. PAGE 133 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN system generated 22 technologies (Spielman et al. 2012). Box APP 4 elaborates which actions the public sector can take to support an effective PPE and enhnace private sector involvement. Farmer Field Schools—Strengthening Community-based Learning and Technology Adoption FFSs provide education and extension services, often with a focus on CSA practices. Over the years, FFSs have emerged as a complementary and reinforcing approach to traditional agricultural advisory services. FFSs leverage experiential learning and a community-learning approach (Davis et al. 2012). FFSs are season-long programs for farmers to regularly meet, learn, and experiment on relevant topics. They are based on a combination of demonstration, explanation, and evaluation of new crop technologies and include meetings on creating farmer groups, establishing demonstration plots by lead farmers, participatory evaluation of technologies, field days, and exchange visits to facilitate learning and sharing of innovations (Anandajayasekeram 2007). FFS should not be an alternative to existing extension efforts but rather should be incorporated into those efforts. FFS interventions are often decentralized and implemented by local authorities, NGOs, community organizations, or the private sector. A study in Eastern and Southern Africa found that FFSs seem most effective if mainstreamed in a traditional extension approach. To ensure effectiveness of the program, farmer facilitators and other facilitators need to be trained better and more backstopping opportunities should be provided. The presumed farmer-to-farmer multiplier effect has not proved to be completely accurate (Anandajayasekeram 2007). Increasing the diversity of trainers, e.g., female or private sector trainers, could improve the promotion of marketing activities through the FFS approach and attract more female participants (Ministry of Foreign Affairs of Denmark 2011). — Integrated Participatory Landscape Management Approach­ Achieving Multiple Objectives The landscape management approach entails the planning of interventions at a larger level, such as a watershed, a catchment area, or a communal area, as well as planning through a participatory land management method. According to FAO, landscape management approach deals with large- scale processes in an integrated and multidisciplinary manner. Thereby it combines natural resources management with environmental and livelihood considerations. It recognizes that the cause of challenges or vulnerabilities of households or landscapes may not be site-specific but may rather involve multiple scales and multiple actors.18 Therefore, the landscape approach aims to achieve multiple economic, social, and environmental objectives and to provide vital ecosystem services while being managed by the people using the land and producing those services (Sayer et al. 2013). Ten principles characterize an integrated landscape approach: (i) continual learning and adaptive management to accommodate the dynamic nature of landscape processes; (ii) solutions must result from consensus-building processes that are based on trust; (iii) take into consideration the various scales at which problems occur, from the local to national levels, and how approaches can impact each level; (iv) extensive research to acknowledge the multifunctionality of landscapes, which creates competing interests for land uses; (v) stakeholders need to be iteratively identified and engaged to ensure a just distribution of benefits and incentives; (vi) solutions must result from negotiated and transparent logic to allow for the development of trust, promote good management, avoid conflict, and allow for ease in resolving conflicts that do occur; (vii) rights and responsibilities must be clarified; (viii), participatory and user-friendly monitoring to foster shared learning; (ix) threats and vulnerabilities must be identified in order to build system-level resilience; and (x) strengthen stakeholder capacity to judge, respond, and share best practices and experiences and ensure stakeholder participation and growth (Sayer et al. 2013). PAGE 134 CSA strategies as well as practices for forestry management, improved livestock management, or rangeland and water management, should not be analyzed and implemented in isolation, but rather under consideration of broader landscape management strategies. The introduction of CSA practices to support livelihood and ecosystem outcomes is one of many strategies in an integrated landscape approach. Community forest management is another example. This can include development of participatory land and resource use planning and management—such as fire control and prevention—and sustainable production of timber and non-timber forest products including charcoal. In addition, it is critical to introduce communal agreements for livestock grazing, agreements for the use of heritage and ancestral land, and agreements concerning water governance and rights. For integrated livestock management, the development of on-farm and community forage production and improved rangeland management through over-sowing of well-adapted leguminous species can increase productivity and environmental benefits such as increased carbon sequestration and reduced emissions from enteric fermentation. Recent research in Zambia emphasizes the overall benefits of considering multiplied goals within a landscape approach and shows that conservation outcomes can be achieved at a lower cost when economic and ecological values of land are adequately considered. For instance, agricultural investment planning that prioritizes agronomic potential and existing road access are found to be correlated with solutions that seek to reduce carbon and biodiversity costs. This demonstrates the potential benefits of adopting a landscape approach reconciling multiplied goals, which can achieve economic benefit at the lowest environmental cost (Estes et al. 2016). Participatory land management is a key component of the approach and seeks to involve and empower communities in decision-making and to decentralize authority. Participatory land management is the opposite of top-down, centralized, and exclusionary approaches to agriculture and natural resource management. Top-down approaches have been criticized for not only being ineffective and inefficient but also for disempowering and harming local populations. Stakeholder engagement can help reduce and resolve conflict and tension by improving trust and learning between stakeholders and greater community buy-in for decisions and projects (Dyer et al. 2013; De Vente et al. 2016). The success of this approach is partially enabled by formal processes that are organized and fair and that allow for distribution of benefits, conflict resolution, and increased trust among participants. Government participation can have varying impacts; when government actors lead the processes, learning by participants, flexibility of solutions, and development of trust are less likely to result. When government actors are involved in a minor capacity, the outcomes of the approach are more likely to be accepted by stakeholders involved in implementation (De Vente et al. 2016). There are several examples in Zambia in which a participatory land management approach was implemented. For instance, in Zambia’s Muchinga and Eastern Provinces, USAID’s Community Forests Program, implemented participatory forest management approaches and supports the GoZ’s REDD+ strategy from 2014 to 2019 (Huntington et al. 2018). The project aims to increase the amount of forest land under improved management, improving livelihoods through the development of non-timber forest products and alternative incomes, and reducing carbon emissions from deforestation. The project is a strong indication of the support for community land management practices and its ties to climate benefits. The World Bank-funded Zambia Integrated Forest Landscape Project similarly supports participatory land management to build capacity in rural communities and support the adoption of sustainable and low-carbon land management practices, with the ultimate goal of generating emission reduction credits that are sold to the BioCarbon Fund (GoZ 2017c). PAGE 135 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Box APP 5 Participatory landscape management approaches are multi-stakeholder processes that can benefit from private sector investments. At the same time, the participatory nature and empowerment of local communities has to be ensured. The public sector can take following actions: • Support inter-ministerial collaboration and knowledge platforms and increase the planning, monitoring, and policy formulation capacities of public sector stakeholders. • Ensure compatibility with community, landscape, and public sector goals. This includes the assessment of opportunities, feasibility, goals, risks, challenges, and long-term benefits of private sector involvement, as well as monitoring and evaluation regarding whether companies stick to the commitments. • Community planning: Support risk and vulnerability assessments; support formation of community organization and strengthening of groups; support land management/watershed planning with relevant tools (e.g., GIS, ICTs). Provide capacity building: training or link communities with service providers to address vulnerabilities in landscape. Address infrastructure and logistics materials: provide materials for land restoration, production inputs, watershed management, and forest management. • Review legal and regulatory set-ups and whether there is a need for reform to define acceptable standards and set the frameworks at the jurisdictional or national levels within which businesses must operate, focus specifically on social and environmental standards, and identify new linkages and governance challenges that may develop from private sector involvement. • Initiate dialogue and facilitate processes. The public sector has a critical role to play to ease and support private sector participation in multi-sectoral partnerships. NGOs and the public sector have the capacity to develop platforms for stakeholder dialogue, negotiation, and consensus building to operationalize successful landscape approaches. This could potentially build trust and mutual understanding with community. Thereby, understanding the concerns of involved private businesses can provide a starting point for building the private sector’s trust. • Support livestock management as integral part of a landscape approach. The public sector could play a critical role in developing animal husbandry/productivity assessments and implementation plans; support procurement and distribution of improved planting material; develop forage demonstration and promotion activities to support animal health services and vaccination. Barriers to success for landscape approaches include governance issues, a lack of institutional capacity, power struggles, and lack of efficiency. Even a participatory approach is expected to experience difficulties. One critique is that participatory processes—and the inclusion of a variety of stakeholders—can lead to inefficiency, procedural failures, and difficulties in achieving success. Participatory processes can also worsen existing conflicts or result in decisions that favor particularly powerful or influential groups. Concerns also exist around the effectiveness of participatory management in dealing with issues that can be much broader than the local context (e.g., irrigation practices as they relate to water rights) (De Vente et al. 2016). However, a case study in Zambia also shows that centralized management of participatory land management can create dependencies and disempowerment in a way that runs counter to the goals of participatory land management (Dyer et al. 2013). Despite being a multi-stakeholder process, landscape approaches often do not include the private sector. Out of 104 integrated landscape initiatives documented in Latin America and 87 in Africa, only 24 and 10 respectively involved the private sector (Milder et al. 2014). While there are opportunities for private sector involvement, i.e., influence funding, sustainable investments, and market and job opportunities, there may be a lack of skills or limitations in the existing landscape management methodologies in managing power relations that could marginalize small local business initiatives, derail the multi-functional objectives of the landscape approach, or focus too narrowly on one commodity (Namirembe and Bernard 2015). However, the private sector may find landscape approaches increasingly interesting as a means to mitigate risks and to address opportunities beyond the production unit (Kissinger et al. 2015). Each landscape has its own context, stakeholders, and PAGE 136 power relationships that will affect solutions needed in a particular location. Box APP 5 provides suggestions for public sector actions to support the implementation of landscape approaches. Cash Transfers—Alignment with Agricultural Planning and Programs Cash transfers can promote agricultural production while combatting poverty and hunger and strengthening human capital for future generations. Access to capital, particularly for rural populations in extreme poverty, is largely nonexistent. Cash transfer programs are part of social protection programs, are unconditional, and designed to improve food security, health and nutritional outcomes. Evidence shows that they have economic impacts, and in particular incentivize agricultural activity. Cash transfers provide infusions of capital as an incentive to invest in agriculture. As such, they can be combined with the introduction of CSA practices. They allow participants to overcome the constraints of upfront investment cost to adopt CSA practices, which, at the same time, have the potential to enhance livelihood resilience, which is critical for the rural poor. Evidence shows that cash transfers can increase agricultural production as well as self-sufficiency due to the improved standing and economic status of participating households. In this way, the cash transfers help households surmount an initial barrier to upward socioeconomic mobility (Boone et al. 2013; Tesliuc 2013). In the context of Zambia, some argue that cash transfers are needed to lift households out of in extreme poverty (Tesliuc 2013). Some argue that the chronic nature of poverty for many small farmers makes them incapable of achieving and maintaining food security without some sort of supplementary support. It has already been recommended that revisions to the existing FISP program include cash transfers, particularly during nonproductive agricultural seasons and for households with very small land sizes (Boone et al. 2013). At the same time, cash transfers are criticized for being a “band-aid solution” and for providing a one-time support, which is not sustainable in the long term. Other critiques vary based on the guidelines associated with cash transfers. For instance, that they create dependency of poor households on the government, or that providing cash to the poor leads them to work less and live off the transfer. However, evidence shows that this is not necessarily the case, but that they promoted promoted productivity and provided an income multiplying effect at the household and local economy level (FAO 2014). Cash transfers have the potential to support the start-up and maintenance costs for agricultural technology adoption and increased agricultural productivity. A case study in Malawi shows that cash transfers can help build capacity for extremely poor farming households to expand agricultural production (Boone et al. 2013). Recipient households purchased a greater quantity of agricultural implements, such as shovels and chickens. More recipients used their own farms as their primary or secondary source of income and reduced their engagement in low-wage labor elsewhere and subsequently produced a greater variety and quantity of food on their own farms. The transfers also improved children’s nutrition and education. The study reported that cash transfers should be used as part of a larger suite of agricultural strategies; in Malawi, they were complemented by the large-scale fertilizer subsidy program. Evidence from Zambia shows that cash transfers had positive impacts on household livelihood strategies. Zambia’s Child Grant Programme led to a 34 percent increase in cultivated land and to an increase in input use, such as seeds, fertilizer and hired labor. This led to an increase in the value of overall production of 50 percent, which was largely sold and even led to an income multiplier effect. the evaluation found that the increase in per capita consumption induced by the programme was 25 percent greater than the transfer itself. In addition the programme led to a shift from agricultural PAGE 137 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN wage labor to on-farm activities for adults, and more time in family agricultural and non-agricultural businesses (FAO 2014). Thus aligning cash transfers with agricultural planning and programmes such as extension services and training on climate-smart agriculture and time it in a way that support agricultural activities most (for instance before the planting season) could have a sizable impact on rural livelihoods. Gender-Sensitive Supply Chains—Facilitating Access to Assets and Services for Women In sub-Saharan Africa, women rarely have access to the same markets and resources as men, which inhibits their potential success in cultivating crops. It is well documented that socioeconomic barriers prevent women from having equal access to land, market, education, and networks. This inequity is worsened by the additional time pressure that women face in managing their complex workloads, or complicated familial gender roles. Case studies in Ghana and Uganda confirm the negative impacts of gender imbalances on farming, particularly in the context of cash crop production. In Ghana, women’s access to liquid capital for purchasing inputs was limited, which led women to use sub-optimal production technologies. These sub-optimal technologies require more labor and land, which generate even greater costs and difficulty. In Uganda, the barriers for producing and marketing large quantities of crops and the lack of access to affordable transportation limited women to selling at markets where they received lower prices for their products. In addition, gendered power dynamics are also reported to lead to women having smaller pieces of land or land that is less fertile (Hill and Vigneri 2011). There are several gender-sensitive strategies to enhance access to the same resources and markets as men, leading to relatively equal chances at success. Improving the gender-sensitivity of supply chains can entail transportation support for female farmers to get their crops to more optimal markets; fostering fair access to capital for female farmers; developing mechanisms to ensure fairness in land allocation to women; and providing women with valuable trade knowledge and insights that might otherwise be withheld from them. If implemented, these strategies can help to improve agricultural productivity by enabling women to produce to their full potential and improve their welfare and local socioeconomic standards (Hill and Vigneri 2011; Elbehri 2013). In fact, research shows that when women have access to the same resources and markets as men, they have relatively equal chances at success (Hill and Vigneri 2011). Gender-sensitive supply chains support the adoption of CSA practices. These supply chains help to grow the farming operations of female farmers, who can obtain improved access to capital as part of supply chains that provide them equal opportunity. Access to capital can alleviate the barriers to adopting improved agricultural strategies. Gender-sensitive supply chain development can also entail increasing women’s access to education and training, for instance though participation in FFSs targeted to women. PAGE 138 Appendix H: Stakeholders Consulted During the Preparation of this Report Zambia Climate-Smart Agriculture Investment Plan Preliminary Results Stakeholder Workshop April 18 -19, 2018 Name Organization Amanda Palazzo International Institute for Applied Systems Analysis (IIASA) Antony Chapoto Indaba Agricultural Policy Research Institute (IAPRI) Arend Van Der Goes SNV Netherlands Development Organization Dailes Judge Oxfam Dominic Namanyungu Ministry of Agriculture; lead of CSA Technical Team Dorothy Mwape Solidaridad Douglas Mwasi Catholic Relief Service Etinala Tembo PELUM Zambia Eustensia Munsaka Indaba Agricultural Policy Research Institute (IAPRI) Festus Hanankani iDE Zambia Frank M Kayula National Union for Small-Scale Farmers (Nusfaz) Friedrick Mahler European Commission Gillian Phiri Namuchimba Future Seeds Hary Ngoma United States Agency for International Development (USAID) Henry Longo Care International Henry Mgomba Ministry of Agriculture Ignatius Mwale Future Seeds Julia Kirya Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ) Julius Shawa PS Agriculture; Ministry of Agriculture Kinkese Theresa Indaba Agricultural Policy Research Institute (IAPRI) Michiel Van Dijk International Institute for Applied Systems Analysis (IIASA) Misael Kokwe Food and Agriculture Organization (FAO) MM Kawamwa MF Monde S. Zulu Zambia Agricultural Research Institute (ZARI) Morton Mwanza Ministry of Agriculture Mumba Mwape Zambia Agricultural Research Institute (ZARI) Munalula Akayombokwa SNV Netherlands Development Organization Mundia Silumesi Seed Control and Certification Institute (SCCI), Ministry of Agriculture Munkombwe Graybill Zambia Agricultural Research Institute (ZARI) Musonda Kapena n/a Mwila Chikwekwe- Weijmer Agence Française de Développement (AFD) Mwimbu Ngoma Heifer Nick Sitko Food and Agriculture Organization (FAO) Norman Chisamo OXFAM Nyambe Kwalombota Cotton Board of Zambia (CBZ) Olipa Zulu Mbata World Food Program (WFP) Patricia Hlangayo World Agroforestry Centre (ICRAF) Peter Manda VUNA Phillip Siauyebe Ministry of Agriculture Richard Mumba Community Markets for Conservation continued PAGE 139 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Zambia Climate-Smart Agriculture Investment Plan Preliminary Results Stakeholder Workshop April 18 -19, 2018 Silke UHLENBROCK World Fish Simunzi Simunzi Golden Valley Research Trust (GART) Susan Chuala Future Seeds Sven Genscitick World Fish Sydney Mwamba Seed Control and Certification Institute (SCCI), Ministry of Agriculture Tamala Tonga Kambikonse University of Zambia (UNZA) Theresa Gondwe International Institute of Tropical Agriculture (IITA) Vincent Akamandisa Zambian Forum for Agricultural Extension and Advisory Services (ZAFAAS) Joseph Mwanza NWK Joseph Mbinji Oxfarm Golden Mahove Vuna Henry Mgomba MoA –Advisory Services Coillard Hamusimbi Zambia Natinal Farmers Union Tarisayi Pedzisai ICRISAT Matongo Munsaje MFL Vernacious Mulenga Musonda MFL Geoffrey Heinrich CRS Friedrich Mahler EU Henry Loongo CARE International (CARE Mike Mailloux Conservation Farming Unit(CFU) Kelly Toole DFID Africa Regional Department Andrew Chongwe Ministry of Fisheries and Livestock Antony Chapoto IAPRI Christine Yamba Yamba Ministry of Fisheries and Livestock Gilbert Kaimana Zambia Sugar Given Simalumba Ministry of Fisheries and Livestock Horeward Tembo ZARI Micheal Isimwaa Ministry of Fisheries and Livestock Mishake Nyembe Ministry of Fisheries and Livestock Stanlous M.Chisakuta MoA-TSB Swithine H. Kabilika Ministry of Fisheries and Livestock Allan Mulando World Food Programme Kundhlande, Godfrey World Agroforestry Center (ICRAF); Miguel Lizarazo CIAT Herbert M. Mwanza ACT Prof Yamba University of Zambia Gibison Simusokwe CFU Richard Mumba COMACO Bruno Mweemba Biofin-UNDP Dr Fusya Y. Goma NALEIC Lewis Bangwe African Development Bank Dr. Kalaluka Munyinda University of Zambia Dr. Elijah Phiri University of Zambia Edison Nkonde Zambia Meteorological Dr. Mick Mwala CSA- Consultant PAGE 140 Other Occasions Name Organization Wallace Mayiya Alliance for Nutrition and Reconstruction Dr. Francis M. Mulenga Department of Veterinary Services; Ministry of Fisheries and Livestock Nefuno Kabwe Ministry of Development Planning Henry Loongo Care International Dale Lewis COMACO Mick Mwala University of Zambia Zambia Climate Agriculture Investment Plan: Inception Workshop October 16 and 17, 2017; Lusaka, Zambia Name Organization Akabiwa Nyambe National Designated Authority (NDA) for the Green Climate Fund Andrew Songiso Ministry of Agriculture Antony Chipoto Indaba Agricultural Policy Research Institute (IAPRI) Arend van der Goes SNV Netherlands Development Organization Ballard A.M Zulu Indaba Agricultural Policy Research Institute (IAPRI) Beene Miyala Ministry of Agriculture - National Agriculture Information Services Brian Chisanga Indaba Agricultural Policy Research Institute (IAPRI) Cecilia Brumer Embassy of Sweden Chance Kabaghe Indaba Agricultural Policy Research Institute (IAPRI) Charles Sondashi Ministry of Agriculture (MoA) Christabel C. Chabwela Indaba Agricultural Policy Research Institute (IAPRI) Douglas Mwasi Catholic Relief Service (CRS) Frank M. Kayula, Dr. National Union for Small-Scale Farmers (Nusfaz) Friedrick Mahuer European Commission G. Chigeza International Institute of Tropical Agriculture (IITA) Gelson Tembo University of Zambia / PALM Hambulo Ngoma Michigan State University—Indaba Agricultural Policy Research Institute (IAPRI) Idriss Ali Nassah World Fish J. Mondy World Food Programme (WFP) M. Mwale Zambia Agricultural Research Institute (ZARI) Mick S. Mwala University of Zambia (UNZA) Milu Muyanga Michigan State University Misael Kokwe Food and Agriculture Organisations Morton Mwanza Ministry of Agriculture Nyambe Kwalombota Cotton Board of Zambia (CBZ) Palazzo Amanda International Institute for Applied Systems Analysis Paul Samboko Indaba Agricultural Policy Research Institute (IAPRI) Peter Manda VUNA Rhoda Mukuka Indaba Agricultural Policy Research Institute (IAPRI) Robby Mwiinga OXFAM Sebastien Grey International Centre for Tropical Agriculture (CIAT) Stanley Ndhlovu World Food Programme (WFP) Stanly Ndhlovu World Food Programme (WFP) Tamara T. Kambikambi University of Zambia (UNZA) Theresa Gondwe International Institute of Tropical Agriculture (IITA) Vincent Mofya Kasisi Agricultural Training Centre (KATC) PAGE 141 ZAMBIA CLIMATE SMART AGRICULTURE INVESTMENT PLAN Endnotes 1. IMPACT is a partial equilibrium model that is used to simulate agricultural development at the global level (but also include national representation) under a range of future scenarios. In the cited study, three socioeconomic (pessimistic, baseline, and optimistic) and four climate change scenarios were combined to account for the large uncertainty in both types of drivers. All three socioeconomic scenarios showed GDP and population increases between the base year and 2050. 2. For instance, cassava yields are expected to decline by 2.9 percent compared to a business as usual scenario, cotton by 6.5 percent, groundnuts by 5.2 percent, maize by 8.7 percent, and soybean by 3.0 percent, while rice yields are expected to increase following climate change by 3.4 percent. 3. The development objective of this task is “to identify and prioritize key policy actions, investments and knowledge gaps. The development of the plan will build on existing knowledge, foster dialogue, and build client capacity to operationalize country climate commitments towards a productive, resilient, and low-emissions agriculture sector.” 4. GHG emissions in 2050 as modeled in GLOBIOM indicate that under limited international support (reduction of 25%) (substantial international support) LULUCF sector could emit 56 MtCO2e (39 MtCO2e) emission, and save 18.8 MtCO2e (35 MtCO2e)emission compared to a business-as-usual scenario in 2050. With limited (substantial) international support the agriculture sector could emit 5.3 MtCO2e (3.7 MtCO2e) emission and has to save 1.8 MtCO2e (3.34 MtCO2e) emission compared to a business-as-usual scenario in 2050. 5. The High-Level Commission on Carbon Prices, led by Joseph Stiglitz and Nicholas Stern, concluded based on an extensive review that a range of US$40-80 per tCO2e in 2020, rising to US$50-100 per tCO2e by 2030, is consistent with achieving the core objective of the Paris Agreement of keeping temperature rise below 2 degrees, provided a supportive policy environment is in place (Carbon Pricing Leadership Coalition 2017). 6. The yield boost from adopting these practices increases the economic returns for these particular crops over other crops on land that may be less productive, therefore the expansion of cropland or switch to these crops reduces the overall yield. 7. Calorie availability is a measure of the total final demand of households or of food available for consumption, which does not include retail waste but does include household waste. 8. The FAO projects a continuation of the diet transitions for developing countries such as Zambia, where calorie availability increases over time due to the increase in GDP per capita and results in an increase in animal products. 9. Kanyanga et al. (2013) project a range of 1,600-2,400 kcal/cap/day for only a subset of crops and assume slower economic growth. Using the same modeling approach, IFPRI (2017) has more recent projections for food availability showing an increase in Zambia to 2,250 kilocalories by 2030 and to 2,640 by 2050. The projections used here are in line with IFPRI’s projections. 10. The Zambia NDC (2015) states that through 2030, estimated emission could be reduced by 38,000 GgCO2e, as compared to 20,000 GgCO2e under domestic efforts with limited international support. This translates into a reduction potential of 25 percent and 47 percent against 2010 as the base year for the domestic efforts with limited international support and domestic efforts with substantial international support, respectively (GoZ 2015) 11. This approach was chosen due to limited information in the NDC document about the baseline scenario. In addition, GLOBIOM can only present 6 emission categories in the LULUCF and agriculture sector. We assess whether the target of 25 or 47 percent is reached in each category. 12. In Zambia, the conversion of grassland to cropland releases carbon stocks (above and below ground) of about 6.8 tCO2e/ ha on average (max: 18.35; min: 5.28) based on the Global Biomass Carbon Map (Ruesch and Gibbs 2008; Gibbs et al. 2008). Additionally, conversion of natural land to cropland releases 172.51 tCO2e/ha on average (max: 220.61; min: 142.7) and conversion of natural land to grassland releases 165.7 tCO2e/ha on average (max: 212.45; min: 137.40). 13. The net carbon balance was calculated with the accounting tool EX-ACT. The analysis compared a without project scenario with conventional practices with a scenario where minimum soil disturbance is adopted on 49 percent of smallholders. It is based on data from GLOBIOM: 49 percent adoption rate for minimum soil disturbance practices resulting in 997,859 ha under cultivation; avoided deforestation for cropland on 43,000 ha cropland and conversion of forest and natural areas into grassland in the same extent. Nitrogen fertilizer use decreases by 5 percent. 14. For instance, in the 2007 National Policy on the Environment; the National Agricultural Policy 2012-2030 dated 2011; the 2012 Livestock Development Policy; the 2013 National Agriculture Investment Plan (2014-2018); the 7th National Development Plan from 2017; and the 2016 National Agricultural Extension and Advisory Services Strategy (NAESS). 15. The BioCarbon Fund Initiative for Sustainable Forest Landscapes is a multilateral fund, supported by donor governments and managed by the World Bank. It promotes reducing greenhouse gas emissions from the land sector, from deforestation and forest degradation in developing countries (REDD+), and from sustainable agriculture, as well as smarter land-use planning, policies and practices (https://www.biocarbonfund-isfl.org; accessed December 2018). 16. The Standard Conversion factors (SCF) is calculated as follows: SCF= SER/OER where OER is the Official Exchange Rate and SER is the Shadow Exchange Rate. SER= ([(M + Tm) + (X - Tx)]/ (M + X)) * OER, where M= total imports, X = total exports, Tm = import taxes, and Tx = export taxes. Average imports and exports over the 2010-15 period have been used for computing the SER, which is set equal to 9.83 Zmw/US$. Given that the OER is equal to 8.9 ZMK/US$, it is found that SCF=1,079. Since in Zambia a VAT of 16% is also applied to all tradable goods, the final SCF is 0,928. Market prices of all tradable goods are converted in economic prices by applying the SCF. 17. These conversion factors are calculated on the basis of World Bank projections for the year 2025 expressed in 2010 constant prices and adjusted to 2016 current prices using the weighted index for each category of commodity index (World Bank; commodity price forecasts; Pink sheet). Import parity prices are computed for fertilizers (Urea, Phosphate and Compound D), which are among key imported items, starting from the international FOB prices at nearest port and considering tariffs and taxes, marketing charges, and transportation costs. Export parity price is computed for soybeans, one of the major exportable crop commodities. 18. http://www.fao.org/land-water/overview/integrated-landscape-management/en/, [last accessed December 2018] PAGE 142 PAGE 143 C L I M AT E - S M A R T AG R I C U LT U R E I N V E S T M E N T PL A N ZAMBIA