AUSTRALIA-WORLD BANK GROUP STRATEGIC PARTNERSHIP IN VIETNAM Mekong Delta Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: Alternative Scenarios and Policy Implications Study on Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: Alternative Scenarios and Policy Implications Disclaimer This work is a product of the staff of The World Bank with external contributions. All omissions and inaccuracies in this document are the responsibility of the authors. The findings, interpretations, and views expressed in this guide do not necessarily represent those of the institutions involved, nor do they 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. © 2021 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org 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. Third-Party Content The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to reuse a component of the work, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. Photo Credit: Cover page: Cao Thanh Long © The World Bank; Inner pages: Thai Bao Huynh/ Pixabay, Thai Nguyen/Pixabay, Jason Kucherawy/Pixabay, Ming Kin/Pixabay, Thomas G/ Pixabay, Falco/Pixabay. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: ii Alternative Scenarios and Policy Implications Preface and Acknowledgments The study was undertaken by a World Bank team comprising Binh Thang Cao (Senior Agricultural Specialist, SEAAG and Task Team Leader), Hardwick Tchale (Senior Agricultural Economist, SEAAG), Diji Chandrasekharan Behr (Senior Natural Resources Management Specialist, SLCEN), Phuong Hoang Ai Nguyen (Environmental Specialist, SEAE2), and Steven Jaffee (Lead Economist, Consultant). The primary research and the analysis underpinning this report were conducted by an interdisciplinary consulting team (the research team), including Dang Kim Khoi, Dang Kim Son, Tran Cong Thang, Do Huy Thiep, and Truong Thi Thu Trang (Senior Researchers from the Institute of Policy and Strategy for Agricultural and Rural Development, IPSARD); Nguyen Hong Quan and Tran Duc Dung (Senior Researchers from the Center of Water Management and Climate Change [WACC]); Tran Thi Phung Ha and Van Pham Dang Tri (Senior Researchers from Can Tho University [CTU]); Tran Thai Binh (Senior Researcher from Ho Chi Minh City Space Technology Application Center); Andrew Wyaat (Senior Researcher from International Union for Conservation of Nature [IUCN]); and Nguyen Huu Thien (Senior Ecologist). The report benefited from comments from the following peer and other reviewers: Sergiy Zorya (Lead Agriculture Economist, SCAAG), Manivannan Pathy (Senior Agriculture Specialist, SSAA2), William Sutton (Lead Agriculture Economist, SAGGL), Cuong Hung Pham (Senior Water Resources Management Specialist, SEAW1), Halla Maher Qaddumi (Senior Water Economist, SEAE2), and Dinesh Aryal (Senior Environmental Specialist, SEAE2). The study received assistance from Quyen Thuy Dinh (Program Assistant, EACVF). The report was prepared with guidance from Dina Umali- Deininger (Practice Manager, SEAAG). The team wishes to thank the Government of Australia via the Australia–World Bank Group Strategic Partnership in Vietnam - Phase 2 (ABP2) for the financial support for the study. The team is also grateful to Carolyn Turk (Country Director, EACVF), Steffi Stallmeister (Portfolio and Operations Manager, EACVF), Mona Sur (Practice Manager, SEAE2) and Ahmed A. R. Eiweida (Program Leader for Sustainable Development, SEADR), for their support. The authors would like to thank the many organizations and individual experts who met the research team and provided data, information, and views. Special thanks are expressed to Nguyen Do Anh Tuan (Director of International Cooperation Department of the Ministry of Agriculture and Rural Development [MARD]); Nguyen Minh Chau, Le Anh Tuan, Ho Quang Cua, Phan Thi Hong Thuy, Le Thanh Liem, Nguyen Thanh Son, La Van Kinh, Tran Che Linh, Pham Thi Be Tu, Dinh Xuan Lap, and Tuong Phi Lai (local technical experts); colleagues from the Center for Agricultural Policy (CAP) of IPSARD, CTU, Institute of Agricultural Market and Institution Research (of Vietnam National University of Agriculture), WACC of Vietnam National University; and local governments and farmers from Mekong Delta (MKD) provinces for their support and participation in the surveys and fieldwork. The authors also highly appreciate the valuable comments received from Directorate of Water Resources, Directorate of Fisheries, Vietnam Disaster Management Authority, MARD Departments of Planning, Cooperatives and Rural Development, Agricultural Products Processing and Development, and International Cooperation, and provincial Departments of Agriculture and Rural Development and Natural Resources and Environment of all 13 MKD provinces. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: iv Alternative Scenarios and Policy Implications Acronyms 1M5R 1 Must do-5 Reductions Program 3R3G 3 Reductions-3 Gains Program ACIAR Australian Centre for International Agricultural Research ACP Agriculture Competitiveness Project AFOLU Agriculture, Forestry, and Other Land Use AgroCensus National Agricultural, Forestry, and Fisheries Census AHP Analytic Hierarchy Process ARP Agricultural Restructuring Plan ASC Aquaculture Stewardship Council AWD Alternate Wetting and Drying BAU Business as Usual CBA Cost-Benefit Analysis CCVI Climate Change Vulnerability Index CGE Computable General Equilibrium CPTPP Comprehensive and Progressive Agreement for Trans-Pacific Partnership CRSD Coastal Resources for Sustainable Development Project CSA Climate-Smart Agriculture CTU Can Tho University DARD Department of Agricultural and Rural Development DoC Department of Construction DoIT Department of Industry and Trade DOLISA Department of Labour, Invalids, and Social Affairs DONRE Department of Natural Resources and Environment EI Environmental Impact ENSO El Niño–Southern Oscillation EU European Union EVFTA EU-Vietnam Free Trade Agreement EX-ACT Ex-Ante Carbon-balance Tool FAO Food and Agriculture Organization (of the United Nations) GAP Good Agriculture Practice GDP Gross Domestic Product GHG Greenhouse Gas GIS Geographic Information System GSO General Statistical Office HCMC Ho Chi Minh City ICEM International Center for Environmental Management ICR Implementation Completion and Results Report ILO International Labour Organization IPM Integrated Pest Management Program IPSARD Institute of Policy and Strategy for Agricultural and Rural Development IRI Interregional Impact IRRI International Rice Research Institute IT Information Technology IUCN International Union for Conservation of Nature Acknowledgment v LXQ Long Xuyen Quadrangle MARD Ministry of Agricultural and Rural Development MCA Multi-Criteria Analysis MDP Mekong Delta Plan MKD Mekong Delta MKDICRSL Mekong Delta Integrated Climate Resilience and Sustainable Livelihoods Project MODIS Moderate Resolution Imaging Spectroradiometer MONRE Ministry of Natural Resources and Environment MOTA Motivation and Ability Assessment MPI Ministry of Planning and Investment NDC Nationally Determined Contribution NGO Nongovernmental Organization NMHC Non-Methane Hydrocarbon NPV Net Present Value NRU Natural Resource Use OECD Organisation for Economic Co-operation and Development PAT Proactive Agricultural Transformation PCI Provincial Competitiveness Index POR Plain of Reeds PPP Public-Private Partnership PSIA Poverty and Social Impact Analysis RCC Regional Coordination Council RCP Representative Concentration Pathway RIA2 Research Institute of Aquaculture 2 SMEs Small and Medium Sized Enterprises SOE State-Owned Enterprise SPSS Statistical Package for the Social Sciences TA Technical Assistance USDA U.S. Department of Agriculture VCCI Vietnam Chamber of Commerce and Industry VHLSS Vietnam Household Living Standard Survey VnSAT Vietnam Sustainable Agriculture Transformation Project VOC Volatile Organic Compound WACC Center for Water Management and Climate Change Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: vi Alternative Scenarios and Policy Implications Contents Executive summary  xvi Chapter I. Introduction  1 1. Rationale  2 2. Objectives  3 3. Study scope and methods  4 Chapter II. The Mekong Delta and its subregions  8 1. The development of Mekong Delta agriculture  9 1.1. The Mekong Delta’s development history  9 2. Rural household livelihood strategies  10 2.1. Commercial farming  12 2.1.1. Upper subregion  12 2.1.2. Middle subregion  12 2.1.3. Coastal subregion  13 2.2. Nonfarm livelihood  14 2.3. Rural labor and migration  15 2.4. Livelihood strategies of the poor  16 2.5. Livelihood strategies from a gender perspective  17 3. Key internal socioeconomic and biophysical dynamics - progress and problems  18 3.1. Economic dynamics  18 3.2. Social dynamics  21 3.3. Biophysical dynamics  23 3.4. Institutions  26 3.4.1. Delta management  26 3.4.2. Regional governance  27 3.4.3. Community value  27 3.4.4. Producer organizations and value chain links  28 4. Key external driving force - opportunities and challenges for future development  30 4.1. Commodities market  30 Contents vii 4.1.1. Rice  30 4.1.2. Vegetables and fruits  31 4.1.3. Aquatic products  32 4.2. Science and technology  32 4.3. Natural hazards and climate change  33 Chapter III. Assessment of main and potential livelihood models of the three Mekong Delta subregions  42 1. Upper subregion  43 1.1. General situation  43 1.2. Key agricultural livelihood models  44 1.3. Assessment of key and potential livelihood models  45 1.3.1. Natural condition suitability and climate resilience  45 1.3.2. Economic and financial aspects  47 1.3.3. Employment generation  51 1.3.4. Environmental aspects  52 1.3.5. Gender  54 1.3.6. Conclusion  55 2. Middle subregion  56 2.1. General situation  56 2.2. Key agricultural livelihood models  57 2.3. Assessment of key and potential livelihood models  59 2.3.1. Natural condition suitability and climate change resilience  59 2.3.2. Economic and financial aspects  60 2.3.3. Employment generation  64 2.3.4. Environmental impacts  64 2.3.5. Gender impacts  65 2.3.6. Conclusion  67 3. Coastal subregion  69 3.1. General situation  69 3.2. Key agricultural livelihood models  70 3.3. Assessment of the most common livelihood models  71 3.3.1. Natural condition suitability and climate change resilience  71 3.3.2. Economic and financial performance  73 3.3.3. Employment generation  76 3.3.4. Environmental aspects  77 3.3.5. Gender aspects  78 3.3.6. Conclusion  78 Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: viii Alternative Scenarios and Policy Implications Chapter IV. Sustainable agricultural livelihood transformation  81 1. Different development scenarios  82 1.1. Introducing two scenarios  82 1.2. Climate change and land-use changes  83 1.2.1. Climate change  83 1.2.2. Land-use changes  84 2. Outcomes of agricultural transformation scenarios  89 2.1. Financial and economic outcomes  89 2.1.1. Gross agricultural production value  89 3.2.1. Gross agricultural production value  89 2.1.2. Agricultural added value  90 2.2. Social outcomes  91 2.2.1. Employment  91 2.2.2. Food security  92 2.2.3. Poverty reduction  92 2.3. Environment  93 2.4. Agricultural transformation in relation to GHG emission reduction  93 2.4.1. Promoting replacement of 3-rice and 2-rice with climate-smart cropping systems  95 2.4.2. Promoting CSA and GAPs  96 2.4.3. Reducing GHG emissions from other sources in the rice value chain  99 3. Policy options and imperatives  101 3.1. Land policy  101 3.2. Fostering private investment and private provision of agricultural services  102 3.3. Stimulating innovation  103 3.4. Strengthening human resources  104 3.5. Setting targets and promoting low-carbon agricultural value chains  104 3.6. Market vision and strategy  105 3.7. Food safety and biosecurity  105 3.8. Institutional strengthening  106 3.9. International cooperation  106 Bibliography  107 Appendices  112 Appendix 1. Study Scope and Methods  112 Annex 1.1. Zoning the Mekong River Delta  112 Annex 1.2. Desk-review methodology  115 Annex 1.3. Stock-taking method of livelihood models in the three MKD subregions 116 Contents ix Annex 1.4. Soil and water suitability assessment of key livelihood models in the three MKD subregions at present (2017) and in the future (2030) 118 Annex 1.5. Sampling and survey  126 Annex 1.6. Cost and benefit analysis  127 Annex 1.7. Multi-criteria analysis  123 Annex 1.8. Gender analysis  126 Annex 1.9. MOTA analysis  132 Appendix 2. Mekong Delta and its subregions  135 Appendix 3. Assessment of Main and Potential Livelihood Models of Three Mekong Delta Subregions  139 Appendix 4. Suggested Transformation Monitoring and Assessment Framework 145 List of tables Table 1. MCA of livelihood models for suitability and sustainability, Upper subregion  xxii Table 2. Ranking of livelihood models in the three subregions  xxii Table II 1. Income sources of rural households in the MKD  14 Table III-1. Coverage of the most common livelihood models in the Upper subregion  39 Table III-2. Ranking of models: suitability to future conditions, Upper subregion  40 Table III-3. Ranking of models for climate change resilience, Upper subregion  41 Table III-4. Ranking of livelihood models for economic performance, Upper subregion  44 Table III-5. Ranking of livelihood models for risk-coping ability, Upper subregion  45 Table III-6. Ranking of livelihood models for employment impact, Upper subregion  46 Table III-7. Ranking of livelihood model suitability based on environmental impact, Upper subregion  47 Table III-8. Ranking of livelihood models for gender aspects, Upper subregion  48 Table III-9. MCA ranking of livelihood models for suitability and sustainability, Upper subregion  49 Table III-10. Potential development directions of livelihood models, Upper subregion  50 Table III-11. Coverage of the most common agricultural livelihood models, Middle subregion 52 Table III-12. Ranking of livelihood models for suitability to future conditions, Middle subregion 53 Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: x Alternative Scenarios and Policy Implications Table III-13. Ranking of livelihood models for climate change resilience, Middle subregion  54 Table III-14. Ranking of livelihood models for economic performance, Middle subregion  56 Table III-15. Ranking of livelihood models for risk-coping ability, Middle subregion  57 Table III-16. Ranking of livelihood models for social impact, Middle subregion  58 Table III-17. Ranking of livelihood model suitability based on environmental impact, Middle subregion  59 Table III-18. Ranking of livelihood models for gender aspects, Middle subregion  60 Table III-19. MCA ranking of models by suitability and sustainability index, Middle subregion 61 Table III-20. Tentative development directions of livelihood models, Middle subregion 62 Table III-21. Coverage of key livelihood models in the Coastal subregion 65 Table III-22. Ranking of livelihood models for suitability to future natural conditions, Coastal subregion  66 Table III-23. Ranking of livelihood models for climate change resilience, Coastal subregion  67 Table III-24. Ranking of livelihood models for economic performance, Coastal subregion  69 Table III-25. Ranking of livelihood models for risk-coping ability, Coastal subregion  70 Table III-26. Ranking of livelihood models for social impact, Coastal subregion  70 Table III-27. Ranking of livelihood model suitability based on environmental impact, Coastal subregion  71 Table III-28. Ranking of livelihood models for gender aspects, Coastal subregion  72 Table III-29. MCA ranking of livelihood models for suitability and sustainability, Coastal subregion  73 Table III-30. Tentative development directions of livelihood models, Coastal subregion  74 Table IV-1. Area changes of key farming models in BAU and PAT scenarios  82 Table IV-2. GHG emissions/removals of the AFOLU sector in 2016 (thousands of tons of CO2e) 88 Table IV-3. GHG emissions in different diversification scenarios  90 Table IV-4. Summer–autumn rice crop in Tien Giang in 2013  92 Table IV-5. Benefits of 1M5R adoption under the VnSAT  92 Table IV-6. Comparing post-harvest loss in rice between Vietnam and Thailand  94 Table A1-1. The area of three subregions and six subregions of the MKD (ha, thousands)  109 Table A1-2. Some available key databases  110 Table A1-3. List of key livelihood models selected for the research  111 Table A1-4. Characteristics for land unit classification  115 Table A1-5. Household survey sample  120 Table A1-6. Results of MCA in ranking current livelihood models in the Upper, Middle, and Coastal subregions of the Vietnamese MKD  125 Table A2-1. Comparison of current groundwater mining with water demand in the MKD, 2010 132 Table A2-2. Average number of laborers in agriculture, forestry, and fishery enterprises  132 Table A3-1. Social aspects of key livelihood models in the Upper subregion (1 ha/year)  136 Table A3-2. Social aspects of key livelihood models in the Middle subregion (1 ha/year)  136 Table A3-3. Social aspects of key livelihood models in the Coastal subregion (1 ha/year)  137 Table A3-4. Number of working days and percentage implemented by women in the Upper subregion  137 Table A3-5. Number of working days and percentage implemented by women in the Middle subregion  138 Table A3-6. Number of working days and percentage implemented by women in the Coastal subregion  138 List of tables xi List of figures Figure 1. MKD zoning  xxi Figure I-1. Three subregions of the MKD  4 Figure I-2. Analytical framework  5 Figure II-1. Land-use map of the MKD in 1976, 1996, and 2015  10 Figure II-2. Livelihood strategies of the MKD farmers  11 Figure II-3. Industrial zones in the three MKD subregions, 2019  19 Figure II-4. Public services related to agricultural innovation in the MKD  20 Figure II-5. Economic situations of farming households in the three MKD subregions  22 Figure II-6. Educational levels of the labor force in the three MKD subregions, 2016  22 Figure II-7. Migration flows from the MKD  23 Figure II-8. Water regulation engineering systems in the three MKD subregions  24 Figure II-9. Provincial Competitiveness Index (PCI) of 13 provinces in the MKD, 2010-18  27 Figure II-10. Contribution of the private sector in the MKD’s economy  29 Figure III-1. Map of key livelihood models in the Upper subregion  38 Figure III-2. Investment in the infrastructure of the key models, Upper subregion  42 Figure III-3. Total production cost per hectare of key livelihood models, Upper subregion  42 Figure III-4. Average annual profit of key models, Upper subregion, 2009-19  43 Figure III-5. Map of key livelihood models in the Middle subregion  52 Figure III-6. Investment in the infrastructure of key models, Middle subregion  55 Figure III-7. Total production cost of key livelihood models, Middle subregion  55 Figure III-8. Average profit of key models, Middle subregion  56 Figure III-9. Map of key livelihood models in the Coastal subregion  64 Figure III-10. Investment in infrastructure of key models, Coastal subregion  68 Figure III-11. Total production cost of key livelihood models, Coastal subregion  68 Figure III-12. Average profit of key models, Coastal subregion, 2009-19  69 Figure IV-1. Land-use changes in BAU and PAT, Upper subregion, 2030  81 Figure IV-2. Land-use changes in BAU and PAT, Middle subregion, 2030  81 Figure IV-3: Comparison of BAU and LUR scenarios, Coastal Region, 2030  82 Figure IV-4: Comparison of gross agricultural outputs between different scenarios  83 Figure IV-5. Comparison of agricultural added value between different scenarios  85 Figure IV-6. Comparison of total required working days/year between different scenarios  86 Figure IV-7. Comparison of required employment between different scenarios  86 Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xii Alternative Scenarios and Policy Implications Figure A1-1. Four main ways of zoning the MKD  106 Figure A1-2. Three subregions in the MKD  107 Figure A1-3. Six subregions in the MKD  108 Figure A1-4. Map of soil texture of the three subregions in the MKD  112 Figure A1-5. Active acid sulfate soil map and potential acid sulfate soil map  113 Figure A1-6. Map of salinity duration and salinity degree at present (2017)  114 Figure A1-7. Flood duration map and flood depth map, 2017  114 Figure A1-8. Map of salinity duration and salinity degree in the future (2030)  117 Figure A1-9. Flood duration and flood depth in the MKD by 2030  118 Figure A1-10. Suitability assessment of key livelihood models in the MKD in 2017 and 2030  119 Figure A1-11. MCA-AHP framework applied for the project  123 Figure 1.12: MOTA framework: from trigger to action  127 Figure A1-13. MOTA score mapping  127 Figure A2-1. Elevation map of the three subregions in the MKD  129 Figure A2-2. Land subsidence in the three subregions in the MKD  130 Figure A2-3. Map of land erosion points in the three subregions in the MKD, 2018  130 Figure A2-4. Inland fishery resources in the Upper subregion of the MKD  131 Figure A2-5. Average after-tax profit of agriculture, forestry, and fishery enterprises in the MKD, 2018  131 Figure A3-1. Average, minimum, and maximum profit of key livelihood models in the Upper subregion in 2009-19 (VND, millions/ha/year)  133 Figure A3-2. Average, minimum, and maximum profit of key livelihood models in the Middle subregion in 2009-19 (VND, millions/ha/year)  133 Figure A3-3. Average, minimum, and maximum profit of key livelihood models in the Coastal subregion in 2009-19 (VND, millions/ha/year)  134 Figure A3-4. Role of women in decision-making in agricultural production in the Upper subregion (%)  134 Figure A3-5. Role of women in decision-making in agricultural production in the Middle subregion  135 Figure A3-6. Role of women in decision-making in agricultural production in the Coastal subregion  135 List of figures xiii Executive summary Context The Mekong Delta (MKD) is Vietnam’s most productive agricultural region, and its agro- economy is well integrated into international markets. Its favorable natural resource base has been developed and utilized—via an extensive system of water control structures—to feed large numbers of people across Vietnam and throughout the world. The intensification of rice production in the MKD was instrumental in converting a once food-deficit country into one of the world’s largest exporters of rice and other foods. This intensification involved double and triple season cropping and heavy input use. Despite some diversification of agricultural land use, the MKD still accounts for more than half of Vietnam’s total rice production and 90 percent of its rice exports. It also accounts for nearly 90 percent of national aquaculture production, both supplying the domestic market and supporting a multi-billion-dollar export industry featuring shrimp and Pangasius fish. The MKD is also Vietnam’s largest fruit growing region, and this too is successfully servicing domestic and multiple export markets. Nevertheless, there are increasing threats to the MKD’s agricultural achievements, and other serious questions are emerging about the sustainability of many of the prevailing production systems. Sea level rise, caused by climate change, is increasingly threatening the viability of once protected cropping systems in the coastal areas. Excessive exploitation of groundwater contributes substantially to severe and more widespread land subsidence. The upstream development of hydro-dams has severely altered the river system’s flood regime and flow and siltation, contributing to increased riverbank erosion, land fertility depletion, salinity intrusion, and the depletion of fishery resources. The widespread dike systems built in the upper part of the MKD to develop a third rice crop have reduced the flood retention capacity of the delta’s Upper subregion in the flood season, causing more severe floods for the downstream areas and hampering the natural recharge of fertile sediments for rice fields. Intensive use of fertilizers, agrochemicals, and veterinary drugs has contaminated soils and water and generated food safety concerns. Agricultural output and export growth have not always been economically efficient and have often been associated with significant (yet rarely measured) environmental costs. That has been the picture in the parts of the MKD which have been especially suitable for rice cultivation. There are other areas, especially in the coastal subregion, which have never been especially suitable for rice and have become even less suitable in the face of sea level rise and salt water intrusion. The low and declining profitability for mono-crop rice farming, together with more favorable work opportunities in cities and neighboring regions (and abroad), is contributing to sizable out-migration of younger, better-educated residents, leaving behind a generally low-skilled Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xiv Alternative Scenarios and Policy Implications and aging workforce for agriculture. The region’s agriculture, therefore, appears to be at the top of its game, yet there are legitimate concerns that all this may topple in the coming decades, unless an appropriate and concerted set of interventions are made to accelerate a (large-scale) process of adaptive agricultural transformation. Recognizing these problems, the government introduced a national agricultural restructuring plan (ARP) in 2013, which advocated for the sector’s transformation, yet with balanced attention to growth and sustainable development objectives. The ARP anticipated a variety of reforms, including in relation to land and land-use planning. For many years, the government had ‘designated’ land to be used strictly for rice cultivation, to achieve production targets. The majority of agricultural land, especially in the major river deltas and along the central coast, was designated as such, totaling more than 4 million ha nationally. In most locations, the restrictions were heavily enforced, limiting the flexibility of farmers and ultimately capping their income potential, pushing many to leave agriculture. In recent years, greater attention has been given to adjusting land policies so that Vietnam can not only meet its national food security needs but also pursue a more vibrant, profitable, and sustainable agriculture. In 2012, at the time that the ARP was being prepared, the World Bank and the Institute of Policy and Strategy for Agricultural and Rural Development (IPSARD) undertook an analysis of Vietnam’s national rice balance and made projections for 2030 based on alternative assumptions and scenarios related to land use, productivity, domestic consumption, and other factors (Jaffee et al. 2012). At that time, the national policy was to ‘protect’ 3.8 million ha of rice land, while the actual area planted—not including double and triple cropping—was 4.1 million ha. That analysis found that the rice area could be reduced to 3.3 million ha and the country would still realize an exportable surplus of between 5 million tons and 6.5 million tons. Parallel research, applying a computable general equilibrium (CGE) model for the Vietnamese economy, simulated the potential impacts of a less-restrictive agricultural land designation policy (Giesecke et al 2012, 2013). That analysis found that even as low as a 10 percent reduction in the area dedicated to rice would have significant positive impacts on national gross domestic product (GDP) and agricultural sectoral growth and on household incomes. The most significant positive impacts would occur in the MKD. In 2018, the World Bank and IPSARD undertook an update of the earlier rice balance analysis and arrived at similar conclusions, even when giving more weight to concerns associated with climate change and upstream developments in the Mekong River (Zorya et al. 2018).1 That study recommended that Vietnam adopt a ‘conservative’ policy target of maintaining 3.4 million ha focused on rice production. Under this scenario, the national rice-growing land (before accounting for double and triple cropping) would fall from 4.15 million ha in 2017 to 3.4 million ha in 2030. Yet, the national rice surplus would change minimally, going from 5.66 million tons 1 Policy Note: “Vietnam’s Rice Outlook 2030 and Implications for the Designated Rice Area” prepared by Sergiy Zorya and Steven Jaffee from the World Bank and Nguyen Do Tuan Anh, Nguyen Ngoc Que, Truong Thi Thu Trang, and Nguyen Le Hoa from IPSARD. Executive summary xv to 5.55 million tons.2 The updated analysis also gave more attention to regional issues. Under the national 3.4 million ha target, the rice area in the MKD would fall from 2 million ha in 2017 to 1.55 million ha in 2030. Even with this reduction, the region’s surplus will be higher in 2030 than it was in 2017 due to productivity gains and the removal of the lowest-performing rice-growing areas. The projections showed that the MKD could experience a one-third reduction in rice land area between 2017 and 2030 (from 2 million to 1.33 million) and still not experience a reduction in the region’s rice surplus—again, because the areas (and seasons) with the lowest productivity would be the ones taken out of production. On the basis of the analysis, it recommended that around 450,000–550,000 ha of MKD rice land could be converted to alternative uses or rice/other crop/ aquaculture rotations without having much impact on the region’s overall rice surplus. This offered an enormous growth and transformative opportunity while being environmentally beneficial and essential for disrupting pest/disease cycles. These and other analyses—especially analyses related to climate change projections—have begun to influence government policy, albeit slowly. In 2017, the government issued Resolution 120 /NQ-CP on sustainable and climate-resilient development for the MKD. This resolution called for greater flexibility in agricultural land use, especially permitting shifts from rice mono-cropping to more diversified farming systems, plus advocating for improved support services to enhance value addition. The resolution emphasized the need for greater adaptation to the future changes of natural conditions and increased attention to the environmental sustainability of farming systems. On March 23, 2021, the government modified its target for rice land protection to 3.5 million ha. This constitutes a breakthrough and is one of the featured policy changes included in Decision 34/NQ-CP on national food security to 2030. Applying this policy will require careful assessment, planning, and facilitative services in different locations, where the scope for and constraints on more diversified farming systems or alternative forms of specialization vary. In support of the transformation process in the MKD, the Ministry of Planning and Investment (MPI) is currently preparing a master plan for the MKD in cooperation with other concerned ministries, aiming to put in place the necessary infrastructure, institutions, and overall enabling environment to promote a more sustainable economic and agricultural development trajectory for the MKD. 2 A much further reduction—to 3 million ha—was determined to be more than adequate for national food security, and that too would yield a large exportable surplus (of 4.76 million tons). Still, the report recommended that the government aim for the 3.4 million target, while monitoring developments over time. Some of the conversion of land use would be pertinent due to the investments involved, so maintaining a wide margin was still recommended. That being said, the value of the additional rice production stems more from commercial rather than food security considerations. If Vietnam faces an especially favorable international market for its rice, then having surpluses (or dedicated export supplies) well in excess of 5 million tons would be a beneficial situation. Yet, if the future market is not highly favorable, then having a large structural surplus could represent something of a liability for the sector and come at an opportunity cost as some other international commodity markets (that is, for fruit and for aquatic products) might represent even better commercial potential for Vietnamese farmers and agro-enterprises. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xvi Alternative Scenarios and Policy Implications This study seeks to contribute to the planning effort for the MKD by addressing some analytical gaps, especially around the technical feasibility and socioeconomic characteristics of alternative agricultural production systems in the context of the evolving natural conditions in the region, and more specifically in the MKD’s three subregions (that is, Upper, Middle, and Coastal). According to the Mekong Delta Plan (MDP) undertaken by the Vietnamese and Dutch governments in 2013, the MKD is broadly divided into three subregions: Upper, Middle, and Coastal subregions. However, the MDP does not elaborate on agricultural livelihood models for each subregion based on technical feasibility, environmental adaptability, economic and financial efficiency, and social aspects. The primary purpose of the study is to fill in that gap by reviewing and assessing different livelihood models and land-use scenarios in the MKD using multi-criteria of technical feasibly, climate change, environmental adaptability, economic and financial, and social aspects to inform the ongoing agricultural transformation in the MKD. A better understanding of the scope for alternative systems—and potential trade-offs which stakeholders may face—is critical for informing public policy and investment. The long-term emphasis of agricultural policy in the region has been that of protecting rice land and intensifying rice production. This policy, although very costly, was highly successful in meeting output targets, yet much less successful from an economic or environmental perspective. The approach used and the findings from this study are expected to contribute to defining the future trajectory of MKD agriculture and a rethinking of the policy, regulatory, and programmatic instruments that the government can best utilize in support of MKD agriculture development. The findings are also useful for the provincial planning exercise, which is happening now, as the regional master plan has reached its final stage. Study Methodology A variety of methodologies were used in carrying out this study. These included literature reviews, analysis of prior survey data, new surveys, and other consultations with various stakeholders. The research methods employed included geographic information system (GIS) mapping, cost-benefit analysis (CBA), multi-criteria analysis (MCA), gender analysis, scenario analysis, and stakeholders’ motivation and ability assessment. The most pathbreaking element of this study is the detailed comparisons of alternative agricultural production systems or ‘models’ according to economic, social, environmental, and risk management criteria. Main Findings 1. Key Features of the Mekong Delta and Its Subregions Based on the hydrological regimes, the MKD can be divided into three subregions: the Upper subregion (deep and shallow flood, freshwater), the Middle subregion (shallow, the interface between freshwater and brackish water), and the Coastal subregion (brackish water and seawater). In the Upper MKD subregion, agricultural livelihoods traditionally involve adaptation to the seasonal flood conditions (that is, one crop of seasonal floating rice, followed by Executive summary xvii vegetables, and harvesting wild fish). Over the past two decades, this subregion has witnessed a rapid transformation from a 1-crop rice system to 2 and 3 high-yield rice crop systems. In 2017, the total rice area in this subregion reached 734,000 ha (producing 9.6 million tons of rice), of which the 2-crop and 3-crop rice systems accounted for 60 percent and 36 percent, respectively. Fruit production has also expanded in the past five years. Despite its small area of 48,500 ha (15 percent of the total area), fruit production in the subregion had the highest growth rate, about 7 percent per year during 2012–2017. Horticulture developed rapidly in places where farmers converted the low-value 3-crop rice systems to fruit orchards. The rapid development of the high-dike system that enables rice triple cropping in the Upper subregion has taken up flood retention space. This has caused increased inundation in downstream provinces in the flood season and increased saline intrusion in the coastal provinces in the dry season. Also, the heightened dikes enable planting a third crop of rice inside diked areas during the flood season but cut off the land from floodwater that brings sediments and associated nutrients, leading to more rapid soil fertility depletion in the areas devoted to the triple rice. The profitability of rice farming has been modest, and employment has also declined due to increased mechanization. The Middle MKD subregion is Vietnam’s leading subregion in horticulture, although rice is still an important crop. Since the early 1990s, a shift occurred from the 1-crop rice system to 2-crop and 3-crop systems. Subsequently, from 2005 to 2017, a significant part of the paddy area was converted to aquaculture, horticulture, or high-quality rice varieties. Freshwater aquaculture (that is, Pangasius fish and freshwater giant prawn farming) and livestock in this subregion have also developed more strongly in comparison to the other parts of the MKD. With the recent development of river dikes in the Upper subregion and coastal dikes in the Coastal subregion and the increased climate change impact, the Middle subregion has been more vulnerable to seasonal floods, saline intrusion, and water pollution. Agricultural livelihoods in this subregion are getting more difficult as farmers struggle to adapt to seasonal floods and saline intrusion. The Coastal MKD subregion is Vietnam’s leading subregion in shrimp farming and brackish- water aquaculture. Before the 1990s, capturing wild fish and shrimp from mangrove forests was the primary livelihood strategy of local farmers in this subregion. In the past three decades, extensive and intensive shrimp farming in this subregion has developed rapidly, leading to a significant reduction of mangrove areas and depletion of wild fish resources. The overexploitation of groundwater for shrimp farming has resulted in severe depletion of groundwater and land subsidence. In the 1990s, the government invested in mega-irrigation projects to divert freshwater from the Hau River to the Coastal subregion to support rice production. With coastal dikes and sluice gates constructed to protect rice production, many small rivers and canals have become disconnected from the main rivers and have lost the tidal influence from the sea. The loss of this self-purification capacity, while continuously receiving heavy loads of agricultural chemicals from intensive agriculture and aquaculture, makes the water heavily polluted for aquaculture and domestic uses. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xviii Alternative Scenarios and Policy Implications FIGURE 1. MKD zoning MAP OF MEKONG RIVER DELTA ZONING LEGEND Regional boundaries Provincial boundaries District boundaries Upper Sub-region Middle Sub-region Disclaimer: Coastal Sub-region The boundaries, colors, denominations and other information shown on any Project: Identifying sustainable map in this work do not imply any judgement on the part of cliamate-resilient agriculture, The World Bank concerning aquaculture, and horticulture the legal status of any livelihood models and transition territory or the endorsement strategies in the Mekong Delta to or acceptance of such adapt with climate change. boundaries. Source: Prepared by authors (2019). 2. Assessment of Main and Potential Livelihood Models of MKD’s Three Subregions Based on the MCA performed, in all three subregions of the MKD, intensive mono-cropped rice-farming systems (involving two crops or three crops per year) are found to be inferior to many alternative systems and, in many locations, are unlikely to be sustainable beyond the current decade. In the Upper subregion, climate change will result in more irregular floods in the rainy season and droughts in the dry season. These changing conditions are not favorable for intensive rice systems. Together with the impact of high-level use of inorganic fertilizers and agrochemicals, land fertility and productivity in the 3-crop rice systems in this subregion are projected to decline quickly. The Middle subregion will also face similar challenges. With the increased impact of sea level rise, it will be more difficult for this subregion to discharge floodwaters to the West Sea, so a large area south of the Hau River will be more severely flooded. In areas bordering the East Sea, salinity intrusion will also intensify. In the Coastal subregion, saline intrusion will become more severe. These new natural conditions will not be suitable for mono-crop intensive rice farming and increase threats for fruit trees. Table 1 illustrates the inferior ratings of the mono-crop rice systems in the Upper subregion, while Table 2 provides the rank ordering of different production systems across all three subregions, considering the multiple criteria. Executive summary xix TABLE 1. MCA of livelihood models for suitability and sustainability, Upper subregion Suit- Climate Economic Risk Social Environ- Gender Final ability with change index coping index ment index suitability Livelihood natural resilience (weight ability index and sus- models condition index =2) index tainability (weight=2) index Rice-lotus 4.0 4.7 4.0 3.5 4.0 4.9 4.5 4.2 Rice-aquaculture 2.5m 4.3 4.0 4.0 3.5 4.6 4.5 3.8 Duck 4.0 3.7 2.8 3.3 2.5 4.8 4.5 3.6 Rice-fish capture 2.5 4.0 3.5 2.8 2.5 4.8 4.5 3.4 Fruit tree 2.5 1.7 4.0 2.8 4.0 3.2 4.5 3.2 Rice-vegetable 2.0 3.0 3.3 2.8 4.0 4.0 4.0 3.1 Pangasius fish 2.5 2.0 3.0 2.5 5.0 3.3 4.0 3.1 Specialized 2.5 2.0 3.0 2.8 5.0 3.1 3.5 3.0 vegetable 2-crop rice 2.5 3.3 2.5 4.0 1.0 3.6 4.5 2.9 3-crop rice 2.0 2.0 2.5 4.0 3.0 2.1 4.0 2.7 Source: Authors’ calculation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Final suitability and sustainability index (column 9) = average of all columns from 2 to 8 (considering the weight of 2 for columns 2 and 4). TABLE 2. Ranking of livelihood models in the three subregions Upper sub-region Middle sub-region Coastal sub-region Livelihood Final suitability Livelihood Final suitability Livelihood Final suitability models and suitability models and suitability models and suitability index index index Rice-lotus 4.2 Beef cattle farming 4.2 Rice-shrimp 3.8 Rice- 3.8 Rice-aquaculture 4.1 Mangrove- 3.8 aquaculture shrimp 3.6 Fruit tree 3.7 Super intensive 3.7 Duck shrimp 3.4 Coconut- 3.6 Extensive/ 3.4 Rice-fish cap- aqualculture improved ture extensive shrimp 3.2 Chicken farming 3.6 Semi-intensive/ 3.3 Fruit tree intensive shrimp Rice-vegetable 3.1 Pig farming 3.5 2-crop rice 2.8 Pangasius fish 3.1 Rice-vegetable 3.3 Specialized 3.0 3.2 Pangasius fish vegetable 2-crop rice 2.9 Rice-duck 3.1 3-crop rice 2.7 Industrial crops 3.0 Vegetable 3.0 Source: Authors’ evaluation from 2-crop rice 2.9 survey data and expert consultation. Note: 1 = very low; 2 = low; 3-crop rice 2.8 3 = medium; 4 = high; 5 = very high. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xx Alternative Scenarios and Policy Implications 3. Sustainable Agricultural Transformation: Alternative Scenarios Two vastly different scenarios for agricultural development in the MKD are considered to illustrate the benefits of agricultural transformation for the MKD. In the BAU scenario, legacy land-use patterns are more or less maintained, and only some changes are made at the margins. Many farmers prove reluctant to shift their land use to potentially more lucrative endeavors, perhaps because of labor and/or financial constraints or concerns about more significant production and market risks. Most farmers would continue their current farming models until they find better opportunities, decide to retire, or experience a major shock (for example, production, health, or financial) that may force them to lease or transfer their land-use rights to others. Also, the government would continue investing in more large-scale infrastructure to protect current farming systems (especially rice production) with limited attention to longer-term environmental sustainability. In contrast, in the PAT, land-use change occurs and is proactively facilitated by the government. Ineffective mono-rice farming areas will be gradually converted to more profitable and climate-adaptive farming models. Also, the government effectively enables and leverages private investment in agricultural innovation, services, and value chains in the MKD. Upper subregion. Under the PAT scenario, the area coverage of 3-crop rice systems would be reduced and, in some places, replaced entirely by freshwater aquaculture or flood-based crops. This rollback to traditional livelihood models would contribute to improved soil fertility and farmers’ incomes, and it would also help reduce the frequency and intensity of floods in downstream areas. With the government’s proactive facilitation and support, it is envisaged that nearly one-fourth of the current 3-crop rice areas in the Upper subregion would shift to rice-flood-based crop models. A small portion may be converted back to the 2-crop rice or other livelihoods. Still, nearly two-thirds of the 3-crop rice systems would be maintained, especially those protected by high dikes. For the current 2-crop rice areas, about one-third would be turned to freshwater aquaculture and some 12 percent to the rice-flood-based crop models. Just over half would be retained. About 80 percent of the fruit- growing area would be maintained, but most of the rest would be converted to a rice-freshwater aquaculture rotation. In short, in the next 10 years, some 151,000 ha of rice-freshwater aquaculture and 144,000 ha of rice-flood-based crops (rice-lotus, rice-floating vegetables, and so on) may emerge and replace some of the existing 3-crop rice and 2-crop rice areas. The areas inside the high dikes may continue their current production systems, such as 3-crop rice cultivation, vegetables, and fruit trees, for another 10 years to beyond 2030. In these areas, sluice gates will be opened once every two or three years to allow floodwater to enter rice fields to replenish sediments and purify the environment. When comparing the farming profits, rice production models in the Upper subregion currently have much lower profits than alternative farming systems. Annual profits of the 3-crop rice and 2-crop rice models are the lowest (VND 40–50 million/ha/year) compared to more than VND 500 million/ha/year of Pangasius fish farming and VND 70–80 million/ha/year of rice-vegetable and fruit tree models. Regarding employment generation, rice-farming models require the least labor (40–60 person-days/ha/year) compared to Pangasius fish farming (540 person-days/ha/year) and specialized vegetables and fruit trees (300 person-days/ha/year). These suggest that the PAT scenario would bring higher incomes and better employment for farmers in the Upper subregion. Executive summary xxi Middle subregion. Under the PAT scenario, the current areas without regular floods would face more frequent flood events in the coming decade, and some of the current 3-crop-rice, fruit trees, and rice crop areas would be replaced by flood-based livelihood models. While about 60 percent of the 3-crop rice protected by high dikes areas could be retained, much of the rest would be shifted to rice-freshwater aquaculture. The 2-crop rice areas would also be reduced to around 40 percent of their current coverage, with 40 percent being converted to rice-freshwater aquaculture and the remaining 20 percent changed to rice combined with other crops or duck production. About 90 percent of the current fruit tree areas could be retained in this subregion, with the balance shifted over to rice-freshwater aquaculture. In the areas where the current livelihood models (that is, 3-crop rice, fruit trees, specialized crops, and rice-other crops) are still efficient, the future emphasis would be on promoting more efficient and sustainable production practices and better integrating farming systems into coordinated value chains. The profits of the 2-crop rice and 3-crop rice models in the Middle subregion are lower compared to alternative farming models. Mono-rice farming profits are VND 40–50 million/ha/ year compared to more than VND 500 million/ha/year of Pangasius fish farming, VND 350 million/ ha/year of vegetables, VND 140 million/ha/year of fruit trees, and VND 100–110 million/ha/year of rice-vegetables/rice-aquaculture models. Concerning employment generation, rice-farming systems require the least labor (40–60 person-days/ha/year) compared to Pangasius fish farming, vegetable, and fruit tree models (around 600, 380, and 330 person-days/ha/year, respectively). Like in the Upper subregion, the PAT scenario would generate higher incomes and employment for the Middle subregion’s farmers. Coastal subregion. Under the PAT scenario, because salinity intrusion would be more severe in the coming decade with deeper and more prolonged inundation in some areas, most freshwater crops would be replaced by brackish-water aquaculture or salinity-resilient crops. It is envisaged that more than half of the 2-crop rice area would be converted to salinity-resilient crops, about one-fifth to brackish-water shrimp farming, and the remaining areas to the rice-shrimp rotation model. For fruit tree areas, only about 10 percent would be retained, and the remaining areas would be switched to shrimp farming or other types of livelihoods. It is also anticipated that about one-fifth of the current rice-shrimp area (which currently accounts for a large area in the subregion) would incrementally shift to semi-intensive and intensive shrimp farming by adopting advanced biosecurity and technologies for better management of water quality and shrimp diseases. Rice farming in the Coastal subregion is the least suitable under saline conditions; therefore, it has the lowest profit (only 30–40 million/ha/year). While super-intensive shrimp farming can generate the highest profit of VND 4.3 billion/ha/year, semi-intensive and intensive shrimp models can result in a profit of VND 500 million/ha/year. Other lower-investment models such as shrimp- mangroves and extensive shrimp farming can also generate VND 70–80 million/ha/year profits. In terms of employment generation, the semi-intensive, intensive, and super-intensive shrimp-farming models require the highest labor (540 person-days/ha/year) compared to 30–75 person-days/ha/year of the 2-crop rice, rice-shrimp, and extensive and improved extensive shrimp models. By adopting advanced biosecurity and technologies for better management of water quality and shrimp diseases, Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xxii Alternative Scenarios and Policy Implications rice farmers in the Coastal subregion could increase their income many times through shifting from the 2-crop rice to shrimp farming or brackish-water aquaculture. In comparing the BAU and the PAT scenarios at the MKD level, the latter is clearly preferable in sustainable development terms. The BAU would lead to a stagnating agricultural sector, with reduced output, further depletion of the sector’s human resources, an inability to attract private investment into the sector, and increased costs simply to fend off the impacts of upstream developments and climate change impacts. In terms of the production value of the main crops and aquaculture commodities, BAU would have a lower production value than even the current level, suggesting that BAU would result in a contraction of the MKD’s output. This is because the marginal rice land areas will become entirely unviable. The loss of output there would not be compensated for by a corresponding increase elsewhere. With an increased impact of climate change (that is, more prolonged floods and droughts, saline intrusion, and increased sea level rise) and more serious impacts from environmental damage in rice mono-cropped systems, a large number of small farmers will fail and considerable land could become idle (or spontaneously converted to nonagricultural uses) if the government does not have a comprehensive plan or enough resources to support the transformation on the ground (that is, proactive plans promoting land-use changes at local levels; supporting public infrastructure; and financing credits, value chains, and market). The PAT scenario, in contrast, would indicate a return to a former stage of regional moderate to high growth in agricultural value addition, assuming increased private sector investment being leveraged to improve the sector’s productivity, competitiveness, and contributions to rural incomes and regional employment. The production value gains under the PAT scenario would be significantly higher than those in the current situation (an increase of 158 percent). The improvements are mainly attributed to the policy changes shifting the nature and focus of public investments and creating a much more favorable environment for private investment in more productive systems, advanced technologies, and service delivery to improve farm productivity substantially. In terms of agricultural value added, the results for the PAT scenario would be nearly 2.5 times higher than under the BAU scenario. This difference mainly comes from the changing composition of the region’s output with reduced rice output and considerable increases in the output of higher-value aquatic products and other crops. The PAT scenario would also create much greater opportunities for remunerative employment. With the more diverse pattern of agricultural production, the PAT scenario would give rise to higher employment in the downstream functions of agricultural logistics and processing and food services and would require more high-skilled laborers (total working days under the PAT scenario would increase by about 300 million, equivalent to about 1.2 million laborers, an increase of 190 percent compared to the BAU case). Concerning national food security, although the rice cultivation area would be reduced by one-third under the PAT scenario, it would still meet the national food security target. The MKD’s paddy production volume still would be more than 18 million tons (or 11 million tons of milled rice). This would be enough for domestic consumption in the MKD, Central Highlands, and Southeast Region (about 3.2 million tons) and export targets (4.5 million tons). Moreover, rice production would shift to higher-quality rice and aromatic varieties under the PAT scenario, generating improved Executive summary xxiii profitability for farmers and agro-enterprises. This is consistent and aligns with Decision 34/NQ-CP of the government dated March 23, 2021, on national food security to 2030, which modified the target for rice land protection from the current level of more than 4 million ha to around 3.5 million ha. The PAT also has great potential in reducing greenhouse gas (GHG) emissions contributing to meeting the Nationally Determined Contributions (NDCs) by: (a) replacing unsuitable and unsustainable 3-rice and 2-rice systems with low-carbon climate-smart cropping systems adapted to the changing and fluctuating climate and environment; (b) promoting climate- smart agriculture (CSA) and good agriculture practices (GAPs) in the remaining core rice areas to mitigate GHG emissions; and (c) optimizing the use of agricultural by-products, improving water and energy efficiency, and reducing post-harvest losses to reduce GHG emissions. About 760,000 ha of 3-rice and 2-rice cultivation areas in the MKD are no longer suitable due to the impact of climate change and the changing environment. Assuming 50 percent of the marginal areas of 2-rice and 3-rice would shift to other lower-carbon farming systems (that is, with 50 percent or 70 percent less GHG emissions), it would help cut down GHG emissions by 1.5–4.2 million tons of CO2e each year. If the ‘1 Must do-5 Reductions’ program3 (1M5R) would be applied to 500,000 or 1,000,0000 ha of the remaining core rice zones, the improved farming practices would help mitigate about 4–8 million tons of CO2e each year. Rice farming in the MKD generates about 29 million tons of rice straw each year, and over 80 percent of the straw is openly burned in the rice fields after paddy harvest. Improving rice straw management through alternative uses or recycling for production purposes (that is, reducing GHG emissions by 30 percent or 50 percent compared to open burning) would help reduce GHG emissions by up to 6–10 million tons of CO2e each year. In addition, reducing post-harvest losses in the rice value chain from 13 percent to 8 percent could save about 1.2 million tons of milled rice (or 1.8 million tons of paddy) and reduce GHG emissions by approximately 1 million tons of CO2e each year. Realizing these gains would require a decisive role of the public sector in setting standards and improving policy, legislation, and the enabling environment and greater private sector involvement and investment in innovation, agricultural services, and marketing strategy, as well as the mobilization of international technical assistance (TA) and financial resources. Policy Options and Imperatives The recommendations in this section are mainly based on this present context and a vision until 2030. Looking beyond 2030 will need a comprehensive assessment of the growth trajectory the Vietnam would take, and what kind of land use plans this would entail. Agricultural transformation is a gradual and carefully planned process in which the government’s policies play a crucial role in creating incentives for efficient transformation in the MKD. In this process, the government’s role should become more facilitative and less 3 1M5R: Must use certified seed; Reducing seeding rate, fertilizers, pesticides, water use (through applying alternate wetting and drying [AWD] techniques), and post-harvest losses. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xxiv Alternative Scenarios and Policy Implications involved with implementation. The government should invest more selectively, focusing on core public goods and services and strengthening the enabling environment to foster and leverage investments made by farmers and the private sector. Concentrating public investments in infrastructure only will not work. Instead, strategic interventions should cover all critical aspects: land policy, fostering private investment and private provision of agricultural services, stimulating innovation, strengthening human resources, setting targets and promoting low-carbon agricultural value chains, market vision and strategy, food safety and biosecurity, institutional strengthening, and international cooperation. Following these principles, it is proposed that the government focus on the following policies for the MKD. 1. Land policy To achieve a scale of land-use change closer to the optimum for effective agricultural transformation in the MKD, the government should adopt a policy that promotes more flexible and diverse land uses and presumes that adjustments will be not only permitted but facilitated. This is because the agricultural transformation in the MKD is tied critically to matters of land-use change in the face of changes in natural conditions (including climate change), demographics, diets, and socioeconomic conditions. Restrictions should be applied only in special circumstances which will be clearly defined by the government through its land-use planning instrument. Yet, neither the national policy target of 3.5 million ha of rice land nor the scenarios provided in this report provide a blueprint for guiding a progressive and sustained process of land-use change from the current to one which is more optimal from the perspectives of households, communities, the agricultural sector, and the regional and national economies. Both despite and because of concerns about climate change, it is clear that the optimal shift in land use is not just a modest one but a substantial one. The economy of the region can be greatly improved if the total planted area devoted to rice is reduced by about 25–40 percent. This scale of adjustment, even if it were to occur over the span of a decade, cannot take place if local governments do not have decision-making authority over which ‘rice lands’ can and cannot be converted to alternative agricultural uses. Therefore, the central government should develop clear guidelines on facilitating more flexible and diverse land use to allow local governments to make decisions on land-use changes based on comprehensive studies and analysis of market demands and comparative advantages, land suitability, climate change, and socioeconomic factors. Along with the flexible land-use policy, it is crucial to support the agricultural transformation by reorganizing agricultural production systems through land consolidation and cooperative farming models. Over the past decades, the government has supported cooperative or collaborative farming via the ‘small farm large field model’. This has typically involved 30–100 farmers joining their fields and doing either synchronized or centrally managed land preparation, planting, water management, harvesting, and other functions. This cooperative farming is an important precondition to enable farmers to adopt GAP and alternate wetting and drying (AWD) irrigation techniques. Upward of 10 percent of the rice-growing areas in the MKD already follow this model. Support for the scaling-up of the cooperative and other clustering approaches would be essential in the future, especially as many older farmers look to reduce their involvement in agriculture. Also, the government should facilitate the development of a more active agricultural land lease market, which will enable individual entrepreneurial farmers, cooperatives, and agribusiness companies Executive summary xxv to expand their farm size. Various approaches for developing land lease markets, involving local government agencies, brokerage services, and financial institutions, have been adopted successfully in other countries, such as in China, Japan, and Republic of Korea, with complementary efforts also made to facilitate the entry of a new generation of younger farmers into farming. It is expected that the upcoming decree on land consolidation (under preparation) could provide the needed legal framework for these initiatives. 2. Fostering private investment and private provision of agricultural services For effective agricultural transformation, the government should aim to facilitate and better leverage private sector investments in MKD agriculture and agricultural value chains and the private delivery of agricultural technical, management, and other services. And over time, especially as networks of agricultural cooperatives and associations of agro-related industries become stronger, the government should explore opportunities for regulatory co-management with such entities, especially in areas related to food safety, biosecurity, and environmental protection. The government should provide incentives to the private sector in priority areas that will facilitate effective agricultural transformation through financing missing public goods and infrastructure. This should be considered on a case-by-case basis as there is no universal formula for all cases. The government may allow large corporations to invest in commonly shared infrastructure and then attract other enterprises to invest in their own infrastructure and services. Local small and medium sized enterprises (SMEs) are encouraged to invest in production support centers for key specialized production areas. At the same time, large companies can be encouraged to invest in regional rice clusters. The government should gradually withdraw its direct investment and market interventions in production and supply chains and promote the socialization of research and advisory services. Local government agencies can also act to promote more active and competitive markets for the provision of mechanization and other agricultural services by accrediting and incorporating private providers into local programs, encouraging links between them and farmer organizations, and ensuring that such service providers are able to participate in existing small business promotion schemes. Agencies can create an information platform to enable farmers to rate and comment on the quality of services provided to them. Then, the PPP investment scheme should continue for production support centers, processing zones, logistics centers, and so on. In this period, the government should completely phase out its direct production activities and agro-trade control (by completing the privatization of state-owned enterprises [SOEs]) and open the research and innovation systems to the private sector. The government will act more effectively as a facilitator by providing preferential policies and enabling the business environment at the highest level. Greater efforts to further support contract farming or similar links between farmer groups and agribusiness companies are needed. Useful functions which the Ministry of Agriculture and Rural Development (MARD) or local government agencies can play might include: (a) the development of ‘codes of practice’ for contract farming relationships to help guide company and farmer organization behavior; (b) the dissemination of analytical tools for assessing the feasibility of such arrangements and for monitoring/evaluating their performance and guidelines for problem/conflict resolution Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xxvi Alternative Scenarios and Policy Implications in such relationships; (c) the development of lines of credit with a focus on supporting companies and farmers involved in such partnerships; and, possibly, (d) a challenge fund, utilizing matching grants, to incentivize especially innovative schemes, perhaps involving new products or in especially challenging locations. 3. Stimulating innovation The transformation of MKD agriculture is going to require a much more robust system for agricultural innovation as well as the strengthening of the knowledge base of farmers and cooperative managers. In the coming years, the MKD Center, which is being established by the Ministry of Natural and Environment (MONRE) and located in Can Tho, is expected to serve as an information hub and integrated data system for the MKD. Unfortunately, it could not be passed onto a research consortium as per its original plan. However, establishing an innovation hub for the MKD region is essential and needs initial public support. The MKD innovation hub can be established as part of a university or a capable private sector company operating in the region. It will support research and innovation programs for the MKD and promote research, training, and the application of emerging technologies to adapt to climate change and reduce agriculture’s environmental footprint (including GHG emissions) at the farm, community, landscape, and value chain levels. Both public and private funds can be mobilized. The program could focus first on seeds, breeds, and technologies to improve input use efficiency (that is, especially for water and energy); reduce post-harvest losses; and optimize the use of agricultural by-products (that is, rice straw). It is recommended that the government foster reform in research management mechanisms by simplifying procedures and granting incentive policies for the private sector to participate in research and advisory and other technical services and technology transfer. Links among research institutes, universities, extension organizations, businesses, cooperatives, and local authorities should also be promoted. In pursuing these efforts, lessons can be learned from an array of agricultural incubation centers and models which have been applied elsewhere. Agribusiness incubation has emerged over the past 15–20 years to stimulate commercial agriculture and transform comparative advantages in commodity markets into competitive advantages in differentiated product markets. A defining characteristic of agribusiness incubators is that they directly engage with startups to help them grow, usually offering them a range of advisory and business development services geared to improving firms’ competitiveness and access to markets. Government involvement in agribusiness incubation varies significantly. Incubators enjoy different degrees of financial and political autonomy. Many are non-profits and start out with the public sector and other external sources of funding from which they wean themselves to varying degrees. Incubators are generally able to cover some, if not all, of their operating costs by charging firms for access to their services and facilities (by charging consulting, business development, marketing, franchising, rental, and other fees). Over time, certain incubators invest in the firms they incubate as well as in their intellectual property, allowing them to share in their profits and royalties. 4. Strengthening human resources The current human resources capacity in the MKD remains weak compared to those in other regions of Vietnam. This problem will become more serious in the context of MKD’s Executive summary xxvii agricultural transformation in which science and technology are expected to be key driving forces. The government should develop a comprehensive capacity-building policy to strengthen human resources for the MKD. The policy should aim at developing and implementing professional training programs for farmers and managers of their organizations and other key actors in agricultural value chains, focusing on sustainable farming practices, business management skills, food safety standards, and application of digital technologies to improve production efficiency and reduce environmental footprint (including GHG emissions). The training programs should be closely linked with the transformation pathway in localities. There should also be a policy to attract young experts and intellectuals to work in research and training centers and public services in the agricultural sector in the MKD. Mainstreamed in such efforts, there should be programs to promote gender equality in skills development and opportunities in remunerative farming and employment (in agricultural value chains and nonfarm sectors). 5. Setting targets and promoting low-carbon agricultural value chains To achieve the NDC target, GHG emission reduction targets should be set and implemented for all key agricultural value chains, and it should first start with the rice value chain in the next 10 years as it has the greatest potential in cutting the GHG emission. Comprehensive and integrated intervention programs should be launched and implemented, including: (a) replacing unsuitable and unsustainable 3-rice and 2-rice with low-carbon climate-smart cropping systems adapted to the changing and fluctuating climate and environment, (b) promoting CSA and GAP in the remaining core rice areas to mitigate GHG emission, (c) and optimizing the use of agricultural by- products, improving water and energy use efficiency, and reducing post-harvest losses to reduce GHG emissions. Such programs would require a decisive role of the public sector in setting standards and improving policy, legislation, and the enabling environment and greater private sector involvement and investment in innovation, agricultural services, and marketing strategy. 6. Market vision and strategy Market development and trade promotion for agribusinesses are an essential part of the MKD’s agricultural transformation. There should be a comprehensive set of policies targeting domestic and export markets that aim for diversification and high-value-added products and low- carbon branding. In addition to traditional export markets, rice export should gradually shift to higher- end markets. The sector should also pay closer attention to the domestic market to supply high-quality and specialty rice for medium- and high-income consumers, especially in Hanoi and Ho Chi Minh City and medium-quality rice for industrial kitchens. It is also suggested that rice companies in the MKD be linked to traders in big cities to promote MKD low-carbon rice. It is also essential to develop a system to monitor the rice supply and demand. For fruit commodities, the aim should be retaining the domestic market by quality and safe food through supermarkets and convenience stores. At the same time, thorough market research aiming at potential countries under the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) and EU-Vietnam Free Trade Agreement (EVFTA) should be conducted. For aquaculture commodities, frequent monitoring of the export market should be undertaken to respond to changing policies. For livestock products, the sector should aim to satisfy the demand of the regional market and partly the Ho Chi Minh City market by stable quantity, safe food, and traceability products. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xxviii Alternative Scenarios and Policy Implications The rice sub-sector should develop a strategy to penetrate and compete in high-end markets under Vietnam’s low-carbon brand names. This will provide signals for the change to high-quality rice varieties in the field and processing technology application. For fruits, producers are encouraged to increase the proportion of processed products and strictly control food safety to take advantage of tariff reductions in potential markets. For fishery products, the sub-sector should develop geographical indications and trademarks, especially for eco-farming products, to raise the added value. Livestock producers are encouraged to explore niche export markets for specialty and high-value products such as bird nests, honey, duck eggs, and so on. Geographical indications and trademarks should be developed to expand to the high-end markets to improve value addition. 7. Food safety and biosecurity Food safety and biosecurity should be given priority as this will help protect human and animal health, facilitate market access, raise value addition, and protect the environment. The government should renovate and improve the quarantine system to ensure biosafety and hygiene for imported and exported goods. It should also develop fully enforced anti-counterfeiting regulations to effectively control the trade of counterfeit production inputs (fertilizers and pesticides). In addition, it is recommended to build an independent management system or reorganize the existing system, thus ensuring accountability in quality management and food safety management among sectors and levels and strengthening coordination among ministries and sectors at all levels. This requires shifting from the current consignment-based and product-based inspection approaches to value chain- based traceability for quality, food safety, and hygiene management. This also requires establishing a systematic database system and strengthening inter-sector and inter-locality coordination in the management and traceability of agricultural, forestry, and fishery products. It is essential to promote the participation of various actors and society in quality certification and recognition. The government should move toward the judicious sharing of responsibilities—among government levels and private sector entities—and proactive, risk-based approaches to protecting consumer safety. The government should promote a preventive food safety management approach by better managing the whole value chain by food business operators and consumers rather than the reactive approach based on the inspection of final products by the government’s entities. This model is considered as co-regulation, which relies on greater private sector involvement to manage the risks. Technical and other forms of support from the public sector will be needed to enable private food operators (especially SME food processors and informal-sector food distributors) to improve their safe food management practices. Adopting a risk-based approach will help provide a set of clear foci for public interventions. 8. Institutional strengthening Policy innovations should go with improved institutional arrangements at all levels. It is critical for the MKD to reach an economic scale for commercial production and efficiently coordinate resource allocation and use, especially for investment capital, land, and water resources. There should be breakthrough solutions to promote a cooperative economy to improve value chain efficiencies, especially where market intermediation is highly fragmented. It is suggested that the central and local governments facilitate the setting up of professional groups to support the cooperative establishment Executive summary xxix and its access to policies and value chain link. At the regional level, the government should set up a joint regional database for decision-making at the MKD Center in line with the Prime Minister’s Decision 1619/QD-TTg dated October 20, 2020, on developing a multi-sectoral, integrated database system for the MKD and Decision 825/QD-TTg dated June 12, 2020, on establishing a multi-stakeholder Regional Coordination Council (RCC) to support and promote sustainable livelihoods and economic transformation of the region and enhance the efficiency of interprovincial investments, plans, and policies. The government should also strengthen the cooperation and coordination among existing institutions, including central and local governments and private sector representatives, to coordinate delta-level investment items, water regulation, production planning for key commodities, and research and innovations. At the regional level, the provinces should promote inter-provincial links through a coordinating board to share information and co-manage natural resources (that is, freshwater use, flood control, salinity prevention, natural disaster prevention, value chain management, and so on). 9. International cooperation There are relevant global experiences, lessons learned, and operational models that Vietnam can learn from to enhance the MKD’s transformation effectiveness. International cooperation will remain important to support the MKD’s agricultural transformation in the coming decades. In addition, the government should develop incentive policies to encourage farmers and producers to seek international TA and global funds to support the adoption of low-carbon and green development and CSA. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: xxx Alternative Scenarios and Policy Implications CHAPTER I Introduction Chapter I - Introduction 1 1. Rationale Vietnam’s Mekong Delta (MKD) Region has an abundance of fertile soil, water, and labor, providing it with a strong basis for supporting a diverse and highly competitive agriculture. Agriculture is the regional economy’s leading sector, accounting for just under one- third of the region’s gross domestic product (GDP) and about half of the region’s labor force. The MKD’s agriculture accounts for over one-third of the national agricultural GDP and more than half of its food exports. The region is the largest producer of rice (54 percent) and fruit (37 percent) and accounts for the largest (70 percent) aquaculture area in the country. The performance and significant contributions of MKD agriculture to the regional and national economy may be at risk as a result of a confluence of factors. Climate change has already contributed to significant sea level rise and the growing incidence of salt water intrusion into the interior of the region. These developments are expected to continue, increasing the vulnerability of coastal areas and the viability of traditional cropping patterns in some interior areas. According to the Average Climate Change Scenario Representative Concentration Pathway (RCP) 4.5, sea level rise is projected to increase by 22–23 cm on both the MKD’s east and west coasts by 2050. During the coming 15 years, average temperatures are expected to rise and rainfall is expected to become more concentrated, increasing during the rainy season and decreasing during the dry season (MONRE 2016). Equally concerning are the potential impacts from a series of upstream hydropower projects and other water diversion schemes. Reduced sediment flows may have an adverse impact on soil fertility and will certainly reduce the amount of nutrients available to the MKD’s fisheries industry. The entire water regime may be affected by the upstream structures and how they are managed. In addition, parts of the MKD are sinking at an alarming rate due to land subsidence caused by the overexploitation of groundwater.4 The current rate of subsidence is estimated at 1.1 cm/year, with the maximum speed at 2.5 cm/year in some areas, exceeding the rate of sea level rise. All of these developments could have a major adverse impact on MKD agricultural production and livelihoods. They call for a proactive strategy of adaptation and reform. In recent years, greater attention has been given to adjusting land policies so that Vietnam can not only meet its national food security needs but also pursue a more vibrant, profitable, and sustainable agriculture. Starting from 2013, the government introduced a national agricultural restructuring plan (ARP) to advocate for the sector’s transformation with balanced attention to growth and sustainable development objectives. In 2018, the government issued Resolution 120/ NQ-CP on sustainable and climate-resilient development of the MKD in response to climate change, which sets out a vision for the sustainable development of the region based upon: (a) the promotion of high-quality commercial agriculture, (b) the expansion of agro-industry to increase value addition, (c) the build-out of a modern and climate-resilient network of productive and social infrastructure, 4 Currently, 70-80 percent of rural people use groundwater of about 2 million m3 per day, mainly for domestic consumption. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 2 Alternative Scenarios and Policy Implications and (d) the more rational use of natural resources and improved protection of the environment. Most recently, on March 23, 2021, the government modified its target for rice land protection from the current level of more than 4 million ha to around 3.5 million ha. This constitutes a breakthrough and is one of the featured policy changes included in Decision 34/NQ-CP on national food security to 2030. However, applying this policy will require careful assessment, planning, and facilitative services in different locations, where the scope for and constraints on more diversified farming systems or alternative forms of specialization vary. The Ministry of Planning and Investment (MPI) is currently preparing a master plan for the MKD in cooperation with other concerned ministries and 13 MKD provinces, aiming to put in place the necessary infrastructure, institutions, and overall enabling environment to promote a more sustainable economic and agricultural development trajectory for the MKD. Given the importance of the MKD’s agricultural sector, the design of this regional master plan needs to consider both the risks and the opportunities facing MKD agriculture. Some guidance on this was provided in a 2013 Mekong Delta Plan (MDP) produced through the joint work undertaken by the Vietnamese and Dutch governments. The MDP provided the broad outline of an agricultural transformative process within the region in which the trajectory would be toward flood-based production systems in the Upper subregion, freshwater agriculture in the Middle subregion, and marine and brackish-water aquaculture in the Coastal subregion. However, the MDP neither provided a detailed analysis of potential alternative farming systems nor offered recommendations for how the new land-use patterns would be realized and supported. Through the use of multi-criteria analysis (MCA), the present study dives more deeply into the details, examining the characteristics and potential suitability of many alternative land-use patterns and farming models to identify the most promising alternatives to rice for farmers in the MKD and highlighting policy and investment options that would help accelerate the region’s agricultural transformation along a more sustainable pathway. 2. Objectives The present study aims to inform planning and policy making in support of a more sustainable, competitive, and climate-resilient agriculture for the MKD region. It does this by • Detailing the evolution of the region’s agricultural systems over the past 20–30 years as well as changes in the region’s natural conditions, including its climate; • Mapping current patterns of agricultural land use and denoting the most common farming systems or ‘agricultural livelihood models’; • Employing MCA to assess the characteristics and suitability of 15 alternative farming systems, including those based on rice, involving different types of seasonal rotations, or entirely different specializations. Eighteen different criteria are considered, related to economic, environmental, or social dimensions as well as climate resilience; and • Contrasting alternative scenarios—to 2030—for changes in agricultural land uses within the region and for the policy stances and investment foci of the government. Chapter I - Introduction 3 3. Study scope and methods Time frame In this study, the trends in socioeconomic and environmental conditions in the agricultural sector and farming systems of the MKD are reviewed for the past 30 years, with a particular focus on the past 15 years, using the most updated data available (generally for 2017). Meanwhile, the transformation pathways and strategies for the future are developed to 2030, with some coverage beyond that. Spatial frame This study classifies MKD into three subregions (Figure I-1): (a) the Upper subregion including the Plain of Reeds and Long Xuyen Quadrangle (LXQ), (b) the Middle subregion including Mid-Tien and Hau Rivers and West Hau River, (c) the Coastal subregion including Ca Mau Peninsula and the Eastern Coast. The classification into these subregions aligns with the Ministry of Agricultural and Rural Development (MARD) and the upcoming Agricultural Transformation Program in Resolution 120/NQ-CP. They are also relatively closely aligned to the ongoing regional link piloting projects of MKD provinces. FIGURE I-1. Three subregions of the MKD MAP OF MEKONG RIVER DELTA ZONING LEGEND Regional boundaries Provincial boundaries District boundaries Upper Sub-region Middle Sub-region Disclaimer: Coastal Sub-region The boundaries, colors, denominations and other information shown on any Project: Identifying sustainable map in this work do not imply any judgement on the part of cliamate-resilient agriculture, The World Bank concerning aquaculture, and horticulture the legal status of any livelihood models and transition territory or the endorsement strategies in the Mekong Delta to or acceptance of such adapt with climate change. boundaries. Source: Prepared by authors (2019). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 4 Alternative Scenarios and Policy Implications A comprehensive analytical framework applying multiple research methods was utilized for this study (Figure I-2). The work involved four steps. FIGURE I-2. Analytical framework ZONING OF THE MEKONG DELTA ASSESSMENT OF MAIN AND POTENTIAL LIVELIHOOD MODELS CONTENT METHODOLOGY FEATURES OF 3 MD REGIONS Selection of key Desk review, models Descriptive CONTENT METHODOLOGY • Taking stock statistics, • Most popular GIS mapping, Desk review, Current situation • Potentially Expert Descriptive (natural, economic, sustainable consultation statistics, social, institutional, GIS mapping, infrastructure, Expert climate change Model assessment Survey, soil and consultation Natural condition water suitability FUTURE situation suitability, climate assessment, GIS (natural, economic, change resilience, mapping analysis social, institutional, economic, social, Survey, CBA, infrastructure, environmental, risk, MCA, gender climate change) gender analysis AGRICULTURAL TRANSFORMATION SCENARIOS Scenario development and outcomes CONTENT METHODOLOGY • BAU: Only minor changes in land use Desk review, patterns Descriptive statistics • PAT: Land use change facilitated by GIS mapping government. Ineffective mono rice areas Expert consultation gradually converted to more profitable and climate-adaptive farming models. Government leverages private investment in agricultural innovation, services, and value chains Policy options and imperatives Source: Authors’ illustration. Note: BAU = Business as Usual; CBA = Cost-benefit analysis; GIS = Geographic information system; MCA = Multi-criteria analysis; MOTA = Motivation and Ability Assessment; PAT = Proactive agricultural transformation. Chapter I - Introduction 5 Step 1. Zoning of the MKD The research adopted the MKD zoning consistent with the MDP and the future MKD Agricultural Transformation Program prepared by MARD following Resolution 120/2018/ NQ-CP. This zoning approach is based on the hydrological regimes, especially floods and droughts, in which the MKD is divided horizontally into the Upper subregion (deep flood, freshwater), the Middle subregion (shallow, the interface between freshwater and brackish water), and the Coastal subregion (brackish water and seawater). Step 2. Describing the features of the MKD subregions Based on the above zoning approach, the research team then reviewed the key features of the three MKD subregions. Consolidated statistics from three primary sources were compiled: (a) annual provincial statistical yearbooks; (b) Vietnamese national agricultural, forestry, and fisheries censuses5 (AgroCensus); and (c) Vietnam Household Living Standard Survey6 (VHLSS). Data on policies, strategies, development plans, land-use maps, and administrative reports compiled by the Center for Agricultural Policy of the Institute of Policy and Strategy for Agricultural and Rural Development (IPSARD); data from provincial Departments of Agricultural and Rural Development (DARD), Investment and Planning), Natural Resources and Environment (DONRE), Construction (DoC), Industry and Trade (DoIT), and Labour, Invalids, and Social Affairs (DOLISA); and relevant previous studies were also collected and used in the analysis. Step 3. Assessing features of the livelihood models in three MKD subregions Desk review and expert consultations were carried out, including stock-taking of conventional livelihood models and potential climate-resilient livelihood models in the three subregions. Based on the data collected in previous studies, a total of 59 livelihood models were identified. From those, common livelihood models were identified based on the three indicators: area of production, number of households applying the model, and production value of the model. A joint consultation meeting in Can Tho City7 and separate consultation meetings with expert groups in provinces were conducted for ranking the most promising and sustainable models based on considerations of economic viability, environmental suitability, natural condition suitability, social and gender aspects, and climate change resilience. At the end of this step, the 15 most promising livelihood models were chosen based on the scoring and expert consultations representing all three subregions. These identified models were plotted onto maps using GIS mapping techniques8 to identify the changes in the adaptive capacity of these models between the current period and the future (2030). A sampling strategy and household survey (600 households), farmer focus group discussions, and local staff interviews were designed and conducted to perform a CBA, MCA, gender 5 This survey has been implemented countrywide for the whole farmer population by the General Statistical Office (GSO) every five years from 2001 to the present. 6 This survey has been implemented countrywide (with the sample of about 45,000 households) by the GSO every two years from 2002 to the present. 7 Experts include local experts coming from DARDs of 10 MKD provinces and commodity experts (institutes, universities, and enterprises) who have good knowledge on rice, fruit, aquaculture, and livestock commodities. 8 See detailed description of this method in Appendix 1.4. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 6 Alternative Scenarios and Policy Implications analysis, and farmers’ MOTA.9 The results were used to rank suitable livelihood models for future conditions and their tentative transformation directions. Step 4. Developing transformation scenarios, pathways, and policy recommendations Two scenarios were formulated in this study. One represents a BAU trajectory with relatively little (and mostly defensive) shifts in agricultural land use and little change in patterns of government spending and policy support. That trajectory would yield unfavorable outcomes, including a contraction in the region’s agricultural output and value addition. A radically different and much more positive set of outcomes would emerge from an alternative scenario, referred to as PAT. In this scenario, there would be significant land-use shifts and an especially large decline in the area under rice mono-cropped systems toward more diverse land uses and applications of rotation systems combining rice and other crops or aquaculture. Policy and investment options are provided to facilitate the realization of this latter scenario. This report is organized into four chapters. Chapter II summarizes the context and setting, depicting the main characteristics of the MKD in terms of natural conditions and socioeconomic, environmental, institutional, and infrastructure aspects and the implications of these for the agricultural transformation of the Delta. Chapter III assesses the suitability and sustainability of the most common and potential agricultural production systems (or livelihood models) in the MKD based on their (a) suitability to future natural conditions, (b) climate change resilience, (c) economic performance, (d) risk-coping properties, (e) social impact, (f) environmental and natural resource impact, and (g) gender dimensions. Chapter IV compares two different scenarios for the MKD, one featuring ‘Business as Usual’ (BAU) in terms of major policies and public investment and the other which is called a ‘Proactive Agricultural Transformation’ (PAT) scenario, which assumes significant shifts in policy, investment, and facilitative support. 9 See detailed description of these methods in Appendixes 1.5, 1.6, 1.7, 1.8, and 1.9. Chapter I - Introduction 7 CHAPTER II The Mekong Delta and its subregions Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 8 Alternative Scenarios and Policy Implications 1. The development of Mekong Delta agriculture 1.1. The Mekong Delta’s development history Vietnam’s MKD is a resource-rich land with a tropical climate, fertile soils, abundant water supply, and rich biodiversity, making it highly suitable for human habitation and productive economic activity. The average temperature varies between 27°C and 28°C across most of the delta. The average annual rainfall is approximately 1,800 mm and is unevenly distributed in space and time. The MKD is characterized by a dense network of waterways, including natural rivers, creeks, and canals, with the most fertile alluvial soil located along the main branches of the rivers. Its hydrology derives from the seasonal river flows of the Mekong River and the tidal influences from the East Sea and the West Sea. The delta ranks second globally for its fish diversity, after the Amazon River (ICEM 2010). These aquatic resources provide an essential source of nutrition and livelihoods for local inhabitants, especially for the rural poor (Baran, Jantunen, and Chong 2007). After the national unification in 1975, the government set food security as the leading priority, promoting double and triple rice cropping on an extensive scale supported by a complex irrigation infrastructure. More people were encouraged to migrate to the delta to settle and drain areas for crop production. Through the International Rice Research Institute (IRRI), high-yielding and short-duration rice varieties were introduced, and the 2-crop rice system gradually replaced the single season rice production system (Nguyen et al. 2007). During the 1980s and 1990s, investments were made to construct sea dikes and coastal sluice gates and to bring more freshwater to coastal areas for rice cultivation. Rice intensification from double to triple cropping per year took place in the late 1990s. To allow this to happen, investments were made in massive water regulation and irrigation systems, including low and high dikes in the floodplains. These dikes prevent floodwater from coming to the rice fields to enable the harvesting of the second crop (during July–August) and the planting of a third rice crop during the flooding season (September–December). The region’s rice cultivation area peaked in 2013, with the triple rice systems covering 45 percent of the total planted area. With these developments, the MKD became one of the world’s largest rice-growing areas. The intensification of rice production has been accompanied by an expansion in the area covered by aquaculture and fruit trees, and this has translated into a large increase in Vietnam’s food exports. Responding to favorable international demand, the area under aquaculture in the MKD tripled between 2000 and 2015, from 229,000 ha to 690,000 ha. This expansion occurred partly through encroachments on mangrove forest areas, which otherwise involved the conversion of some double rice cropping areas to shrimp or rice-shrimp farming in the Coastal subregion and freshwater catfish or rice-giant freshwater prawn systems in the Upper subregion. Over the past 15 years, the area under fruit production—especially dragon fruit, litchee, and mango—has also increased and spread geographically to the north of the Tien River, the middle of the Tien and Bassac rivers, and the south of the Bassac River in Vinh Long, Can Tho, and Hau Giang Provinces, and even to the upper area of An Giang and Dong Thap Provinces. After many years of robust growth, MKD agriculture experienced a series of setbacks in the late 2000s and early 2010s, which raised concerns about the sustainability of this growth. Chapter II - The Mekong Delta and its subregions 9 Over 2000–2008, the region’s agricultural gross value added increased at an impressive 4.6 percent per year. Subsequently, between 2009 and 2013, the growth rate was only 1.1 percent per year due to the combined effects of unfavorable weather, pest/disease problems, and market instability. Analytical work done in 2011 and 2012 by the World Bank and a consortium of Vietnamese policy and research institutions drew attention to major economic, environmental, and institutional weaknesses in the MKD agricultural development model and recommended adjustments in the underlying goals and modalities of government support for the sector (World Bank 2012). In the following year, Vietnam launched its ARP, embodying major principles of sustainable development and laying out an agenda of policy and institutional reforms and strengthening. Over the past decade, the performance of MKD agriculture has been mixed, with clear demonstrations both of its underlying strengths and vulnerabilities. The region’s agro-food export performance has been impressive with a shift taking place in the composition of rice exports (to higher-quality varieties), a major expansion in fruit exports (to many more countries), and a steady increase in the share of shrimp exports in value-added form. Yet, the decade has also been one in which productivity growth has slowed, aquaculture and livestock disease episodes have been more severe and costly, and salt water intrusion has affected wider areas of crop production. In recent years, efforts have been stepped up to promote more sustainable crop and aquaculture production, yet there remains a large unfinished business in this area. Agricultural GDP growth in the MKD was 2.7 percent per year over 2013–2020, largely mirroring the national average. FIGURE II-1. Land-use map of the MKD in 1976, 1996, and 2015 Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. 1-crop rice 2-crop rice Triple rice Fruits Fishery Forestry Source: Dang et al. (2019). 2. Rural household livelihood strategies Among rural households in the MKD, there are four common livelihood strategies, sometimes pursued in combination (Figure II-2). These are as follows: • Commercial farming. Smallholder households remain in farming, shifting over time from subsistence to mixed or market-oriented production, typically by intensifying production or adopting new forms of specialization. Some farmers expand their operations by leasing the land of others. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 10 Alternative Scenarios and Policy Implications • Rural services entrepreneurship. Some smallholder households shift away from primary production to focus on small businesses, providing either agricultural services (for example, inputs, veterinary services, processing, or aggregation) or nonagricultural services (for example, transportation or running a local shop). They typically start off on a micro or small scale. Most remain so, although some graduate to become medium enterprises with dozens or more employees. • Rural labor. Households remain in rural areas and do seasonal or more permanent work, on farms, in construction, at industrial park factories, and/or in services. • Urban migration. Entire households or individual members migrate to urban areas to undertake regular or irregular employment. While young people with a good education may become permanent urban residents, those with low education may end up with unstable jobs (workers in industrial zones) or informal employment (housemaids). Most of them still maintain agricultural land in rural areas as a secure asset. FIGURE II-2. Livelihood strategies of the MKD farmers Source: Shakhovskoy, Colina, and Hook (2019). Chapter II - The Mekong Delta and its subregions 11 2.1. Commercial farming Agriculture and fisheries are among the income sources for some two-thirds of the rural households in the MKD. 2.1.1. Upper subregion Agricultural livelihoods in this subregion involve adaptation to the seasonal flooding conditions (Tran and James 2017). Before the 1980s, farming households here produced only one rice crop per year. In the pre-dike period,10 farming households cultivated floating rice, and in the flood season, they usually turned to other activities, of which wild fish catching was the most common. Moreover, free-ranging ducks that take advantage of rice by-products as animal feed were also a good source of income for farming households. After August high dikes were constructed,11 farming households shifted to double or triple intensive rice crops per year.12 In some areas, smart farming technologies have been introduced, such as automatic irrigation. The use of mechanization is relatively high for land preparation (96 percent), harvesting (90 percent), and pesticide spraying (72 percent), but quite low in drying (46 percent) and seeding (21 percent). In recent years, the profitability of rice farming has been modest, and on-farm employment has declined as a result of increased mechanization. In response, a growing number of households have diversified their farm income sources beyond rice. They have increasingly practiced integrated farming models in the flood season (that is, freshwater prawn farming, Pangasius fish culture in a rice field, eel raising, fruits, rice intercropped or rotated with upland crops, concentrated duck raising, and so on). However, going back to flood-based systems appears challenging because the high dike structures prevent floodwater from coming into the rice fields. And the level and timing of water flows have become less predictable due to developments in the upstream portions of the river. To enable the transition from the 3-crop rice system back to flood-based systems, existing dike systems need to be modified to allow more floodwater to go into rice fields during the flooding season. 2.1.2. Middle subregion Livelihood strategies in the Middle subregion involve adaptations to flood and salinity conditions. With more favorable biophysical conditions, agricultural livelihoods here are more diversified and developed. In the past, the main livelihoods included single-crop rice cultivation and fish capturing. With the development of canals and dike systems, farmers have shifted to double or triple rice cropping. 10 The pre-dike and post-dike periods refer to the periods before and after the dike systems were constructed (both August and high dikes). The date varies among communes, generally ranging from the 1980s to 2000. 11 August dikes are dike systems that are relatively low in the Upper subregion of MKD, designed to partly prevent early floodwater penetration in the field during July–August to allow completing rice harvest of the second crop (Summer–Autumn Crop). 12 The technology packages provided by public extension include ‘3 Reductions: seeds, nitrogen fertilizer, and pesticides, and 3 Gains: productivity, quality, and profit’ and ‘One must use: certified seed, and 5 Reductions: seed, fertilizer, water, pesticides, and post-harvest losses’. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 12 Alternative Scenarios and Policy Implications Horticulture developed first in the Middle subregion as early as the 1990s as the land in this subregion is fertile and less affected by floods and saline water. The Middle subregion also witnessed a large-scale transformation from rice to tree farming during 2005–2017, when rice cultivation was yielding diminishing profits. As a result, fruit tree areas expanded further from the subregion’s center toward the Upper and the Coastal subregions. Many high-quality fruits are produced from this subregion, such as Lo Ren Star apples, Hoa Loc mangos, and Nam Roi pomelos, and their prices have been attractive and relatively stable. Mixed fruit intercropping is quite common to secure income year-round and prevent production and market risks (Thanh and Tacoli 2009). Some households have also raised fish and poultry in the fruit tree gardens to form integrated farming systems. They have also adopted advanced breeding, pest management, and other techniques, and many have been certified under one or another good agricultural practice (GAP) scheme. The Middle subregion has thus become the largest and most developed fruit growing area in the country. In some areas outside of flood-control dikes, farming households conduct integrated rice-giant freshwater prawn13 or fish-based systems. However, these livelihood strategies do not take place on a large scale. In some areas, farming households shift to livestock (poultry, pigs, beef cattle, milk cows, and so on). Livestock raising in this subregion is more concentrated than in the Upper and Coastal subregions. 2.1.3. Coastal subregion Agricultural livelihoods are closely adapted to the prevailing salinity conditions and controls. This subregion’s soil and water conditions are not well suited for crop production. Before the 1980s, local residents used to cultivate one crop of rain-fed rice in the rainy season. They otherwise relied on extensive fishery farming, the capture of natural fish, and the harvesting of mangrove non- timber products. With investments made in mega-irrigation projects in the 1990s to divert freshwater from the Hau River to the subregion, a rapid shift from the single rice to double rice and even triple rice crop systems followed. During the past 10 years, aquaculture has brought much higher profits than rice, so many rice areas have been switched to extensive and semi-intensive shrimp or shrimp- rice farming (Ha and Bush 2010). Forest-based livelihoods have diminished dramatically with the decreasing mangrove area. Except for the areas of mangrove encroachment, the transformation back from rice to brackish aquaculture appears more climate-resilient and suitable to the natural conditions and responsive to change in market demand.14 While most shrimp farming is still extensive or improved extensive, intensive shrimp farming recently developed more in Bac Lieu, Soc Trang, and Ca Mau. Shrimp farming is considered a high-risk business. However, with the recent development of technologies (that is, biofloc technology, polyculture with fish [Tilapia sp.], closed-recirculation farming systems, and so on), intensive farming can better manage diseases and water pollution than extensive farming systems. At present, intensive shrimp-farming accounts for around 10 percent because it requires large investments in pond construction and operations (that is, seeds, feeds, labor, energy, and so on). In some areas, farming households take advantage of the mudflats to culture mollusks such as blood cockle farming in Kien Giang and clam farming in Ben Tre, Tien Giang, and Tra Vinh Provinces. Marine 13 The Mekong giant freshwater prawn (Machrobrachium rosenbergii). 14 More discussion can be found in Chapter IV. Chapter II - The Mekong Delta and its subregions 13 fish culture is largely underdeveloped, although some households practice it in Kien Giang Province in the West Sea. 2.2. Nonfarm livelihood For farming households in the MKD, those reliant on agriculture and wage income make up the largest group (29-32 percent in 2010 and 2016). Agriculture contributes about 30 percent to the total income of this group. Another 16 percent comprises rural households who rely mainly on agriculture, contributing 90 percent to their total earnings. Over time, farming households’ income sources have tended to become more diversified (Table II-1). Families with three income sources (from agriculture, wage, and self-employment) have increased dramatically from 3 percent in 2010 to 10 percent in 2016. The Middle subregion has witnessed the most significant increase and the Coastal subregion the smallest. The Upper subregion has a different trend compared to the others. There, an emerging trend is a higher reliance on self-employment and wage employment. While the twinning of agricultural and wage incomes has become less common in the Upper subregion, this is not the case elsewhere. The outright exit from agriculture seems to be occurring faster in the Upper subregion. TABLE II 1. Income sources of rural households in the MKD Income sources (%) Do not Agriculture Salary Self- Agric + Agric + Wage Agric + work employ- wage self- + self- wage Year Region ment employ- employ- + self- ment ment employ- ment Whole 5.20 16.12 15.32 6.63 30.27 7.98 8.93 9.55 country Whole 5.12 17.48 12.38 5.91 32.19 8.57 8.14 10.21 2016 region Upper 5.31 14.16 13.86 12.09 21.53 8.55 15.34 9.14 Middle 5.58 14.30 14.21 7.53 31.56 8.21 9.05 9.56 Coastal 3.52 26.56 14.09 8.13 24.93 7.59 9.49 5.69 Whole 3.62 17.15 15.38 7.69 31.26 12.80 9.50 2.69 country Whole 3.12 16.51 15.34 10.79 29.42 10.74 10.85 3.23 2010 region Upper 3.24 14.75 15.34 10.62 28.02 11.80 12.68 3.54 Middle 3.48 13.57 15.10 10.86 31.89 11.45 10.60 3.05 Coastal 1.88 27.42 16.13 10.75 22.85 7.53 9.95 3.49 Source: Authors’ calculation from VHLSS 2010 and VHLSS 2016. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 14 Alternative Scenarios and Policy Implications 2.3. Rural labor and migration In the Upper subregion, the typical sources of jobs are seasonal agricultural work, construction work, or migration to find formal or informal employment in larger cities— one of the drivers is the push factor from the agricultural sector. There is a causal link between land-use change resulting from the construction of dike systems and the livelihood strategies in this subregion (Tran 2019). • The Upper subregion is dominated by the intensive rice-farming systems in dike rings. The protection from floods allows a widespread adoption of agricultural mechanization, resulting in surplus labor in these areas. • The profit from rice is decreasing, keeping households’ income low compared to other sectors. The rice monoculture system supported by the dike systems contributes to increasing pest infestation, soil degradation, and water pollution, resulting in greater use of pesticides and fertilizers. The production cost of triple rice farming households is 58 percent to 91 percent higher than that of a double rice system. The triple rice system’s total profit was initially 57 percent higher than that of the double rice system, but it quickly decreased. After 15 years of operations, the gap was only 6 percent (Tran et al. 2018). • As a result of floodwater control with dikes, flood-based wild fishery resources have depleted, and the open areas for collecting this resource have decreased. • The decreasing profit and employment and loss of traditional flood-based livelihoods have forced farming households to seek additional income sources. The other driver is the pull factor of urban and industrial zones. In the Upper subregion, industrialization is less well developed. Hence, people have to go to other regions to seek jobs, resulting in the Upper subregion having the highest out-migration rate in the MKD. Nevertheless, migrants often face challenges. For example, (a) they are omitted from the government’s poverty reduction programs in their original and destination places, (b) they often have to take informal and seasonal jobs with low wages and low social benefits, and (c) they cannot enjoy the same social benefits and services provided by the government as local people due to the government’s resident registration policy. They have to pay significantly higher than city dwellers for basic social services— health care, education, housing, and utilities—and their children cannot go to public schools. In addition to migrating to other cities, labor exodus to other countries has also commonly occurred in the Upper subregion. Dong Thap is a leading province for ‘exporting’ agricultural workers to Japan (which usually accounts for 80 percent of the total labor export in the MKD).15 The motto of Dong Thap’s labor export program is ‘Depart as employees but return as entrepreneurs’. 15 In 2011-2014, a total of 8,000 people (on average 2,000 people per year) in the MKD went abroad to work as exported laborers. In the first 9 months of 2019, Dong Thap exported 1,538 laborers on limited-time contracts, mainly to Japan. Chapter II - The Mekong Delta and its subregions 15 In the Middle subregion, households’ income sources are more diversified, but the agricultural income still plays an important role. Farming remains important even in the most ‘urbanized’ villages because the farming systems in the Middle subregion are more stable and profitable (Thanh and Tacoli 2009). However, there are considerable variations within the subregion. In the west of the subregion, the agricultural systems still heavily depend on rice. There, households face problems similar to those in the Upper subregion (that is, increasing production costs and decreasing profit and employment). In the southern part, fruit gardens are relatively developed, although mature plantations are less labor intensive, resulting in labor surplus in some areas. In the Middle subregion, the shift from farm to nonfarm livelihoods is attributed to the pull factor from the cities and industrial zones developed in the subregion. The Middle subregion is the most urbanized and industrialized in the MKD; it is a hub of trading, transportation, and processing, generating nonfarm employment opportunities. The subregion is also well connected to the neighboring industrial cities such as Ho Chi Minh City (HCMC), Dong Nai, and Binh Duong. Short- term migration (and commuting) is more common in this subregion than in the other two subregions. There are also some in-migration flows in some areas such as Vinh Long and Tien Giang, the only two provinces in the MKD having an in-migration rate. Owing to good education, people in the Middle subregion tend to have more opportunities to find official contract-based jobs in big cities. However, employers in industrial zones prefer to recruit workers in the age range of 20 to 40, with low salaries and benefits and requiring simple skills. So, after the age of 40, the livelihood of those workers is likely uncertain. In the Coastal subregion, rural households’ livelihoods mainly rely on brackish-water farming systems. Shrimp culture brings higher profits than other farm livelihoods, so it is an important income source for most households. However, it is also vulnerable to price fluctuations and diseases, and the initial investment and operating costs are very high. So, shrimp farmers can go bankrupt after a few seasonal losses. In the Coastal subregion, the industry sector is the least developed, so it is challenging for rural households to find additional nonfarm jobs locally. Families with less diversified income sources often send a family member for temporary migration as a risk management instrument (Betcherman, Iftekharul, and Melissa 2019). In the worst case of bankruptcy, the households may have to sell their houses and farmland and migrate permanently to other places. With relatively low levels of education and skills, migrants from the Coastal subregion share the same problems as migrants in the Upper subregion. Also, they face more difficulties in transportation while staying far from urban and industrial centers. Consequently, long-distance migrants are more prevalent in this subregion. 2.4. Livelihood strategies of the poor Across all three subregions, the poor are those with minimal or no productive land and with low education. These people mainly work as laborers on farms or construction sites or migrate to cities to find informal jobs (Tran and James 2017). In the Upper subregion, after the dike systems were built, the landless poor lost their traditional flood-based livelihood of catching wild fish. In the Middle subregion and Coastal subregion, the landless poor face growing challenges as a result of farm Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 16 Alternative Scenarios and Policy Implications mechanization. Limited and unreliable livelihood options for poor households have increased their out-migration and vulnerability, further widening gaps between the rich and the poor in the MKD. 2.5. Livelihood strategies from a gender perspective In the Upper subregion and Middle subregion, women16 are less likely to be hired for non- housework jobs than males (84 percent compared to 95 percent). In the Upper subregion, women are not involved much in agricultural activities (32 percent) compared with men in this subregion (47 percent) or women in other subregions (that is, 40 percent in the Middle subregion and 42 percent in the Coastal subregion). However, many of them engage in self-employment activities (36 percent).17 The survey conducted for this study shows that these women in the Upper subregion are mostly middle-aged (around 50 years old), having a low level of education (primary or secondary), working in low-profit activities (that is, street vending, handicrafts, traditional foods and condiments, tailoring, and so on), and earning a low income of VND 2-4 million/month. Younger women prefer to work in factories (mostly aquaculture processing companies) in their neighborhoods or large cities. The women’s occupation in the Upper subregion implies that, when switching from rice production to other more labor-intensive agricultural livelihood models, female labor in this subregion is less likely to be mobilized. In the Middle subregion, the percentage of women participating in agricultural production in the Middle subregion is 8 percent higher than that of the Upper subregion (40 percent vs. 32 percent). Self-employment in the Middle subregion is not common, with the involvement of only 25 percent of the women. The Middle subregion is also the location of the most industrial zones and factories in the MKD. Therefore, it is not surprising that many women in this subregion are working a wage job (39 percent compared to 32 percent in the Upper subregion and 25 percent in the Coastal subregion).18 The survey results in this study show a huge demand for women’s labor in the industrial zones in this subregion, especially for food processing, textiles, assembling, and so on. The Coastal subregion has the most substantial gender gap in the MKD, where 14 percent of the women at working age stay at home to do housework. Among the working women, agriculture is the most common choice, which is the occupation of 42 percent of the working women in this subregion. This subregion also has some aquaculture processing factories and commercial farms. However, not many industrial parks are located in this subregion. Accordingly, not many off- farm job opportunities exist for women in this subregion (25 percent have wage jobs and 21 percent have self-employment activities). 16 Women at the age of 15-60. 17 Authors’ calculation from VHLSS 2016. 18 Authors’ calculation from VHLSS 2016. Chapter II - The Mekong Delta and its subregions 17 3. Key internal socioeconomic and biophysical dynamics - progress and problems The evolution of the agricultural systems and the selection of farming households’ livelihood strategies have been driven by, and in turn influenced, a variety of socioeconomic and biophysical factors. Such interactions will continue in the future. 3.1. Economic dynamics The agricultural sector accounts for 29 percent, 29 percent, and 34 percent of GDP, respectively, in the Middle subregion, Upper subregion, and Coastal subregion. During 2010-18, the value of agricultural produce of the MKD contributed around 34 percent to the MKD’s GDP and 35 percent of Vietnam’s total agricultural GDP. During 2016-18, MKD’s annual agricultural GDP growth rate was 3 percent, contributing one-third to the MKD’s GDP growth. The agricultural sector accounts for 57 percent of the delta’s total export value (MARD 2019). However, the agricultural sector of the delta largely relies significantly on the performance of its rice sub-sector, one which has been characterized by high output yet low or diminishing profitability. If the investment costs in dike systems and dike maintenance after floods are included, the actual rice production cost would be much higher than the revenues from triple rice production (Tong 2015). Also, if the foregone revenue from floodplain fishery resources were factored in and one calculated the costs of multiple environmental impacts (that is, water pollution, soil contamination, and so on) from excessive use of fertilizers and agrochemicals, then the social costs would almost certainly outweigh the social benefits of intensive rice farming (especially the 3-crop system) in the MKD. The MKD has fewer industrial zones than other regions, and it is not a strong manufacturing region, especially for food processing and agriculture-supporting industries. In 2018, there were 85 industrial parks and export processing zones with a total area of 14,787.6 ha, but the occupation rate averaged only 20–30 percent. Some industrial parks are entirely unoccupied. The Middle subregion is the most industrialized, with 54 zones in more than 15,000 ha, and it can be divided into two mains centers (Figure II-3). The first cluster involves multi-sectoral industrial zones located in Long An and Tien Giang Provinces, serving the HCMC market, with a total area of approximately 10,000 ha. The second cluster is in Can Tho City, with a total area of 1,500-1,800 ha, focusing on energy and agricultural and aquatic product processing. Long An is the province with the highest fill rate (more than 80 percent). The Upper subregion and the Coastal subregion have only 14 and 16 industrial zones, respectively, primarily for food processing, aquaculture production and processing, and consumer products. The Coastal subregion also has an advantage in seaports for coal and gas transportation and potential for renewable energy (wind and solar) development. During 2015-18, only 400,000 new nonagricultural jobs were created in the MKD, representing less than 10 percent of the national figure of 4.2 million. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 18 Alternative Scenarios and Policy Implications FIGURE II-3. Industrial zones in the three MKD subregions, 2019 Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ illustration.19 The MKD’s transportation system is largely undeveloped and is considered a major bottleneck for the delta’s economic development. Investment in transportation infrastructure in the MKD accounted for only 15 percent of the total national investment during 2016-20. The primary transportation means are roads (79 percent) and maritime shipping routes (13 percent), while inland waterways only 1 percent. Railways are nonexistent. At present, there are only 40 km of highways in the MKD. Intra-provincial transportation and intra-regional transportation are not smooth. Airports are small and mainly for human transport and have little capacity for air freight. In the MKD, the Middle subregion has the most advanced transportation system, including the Trung Luong–My Thuan highway, national roads (1A, 91, 53, 54, 57, and 60), the waterway connecting HCMC and Kien Giang, the Tra Noc international airport, and the biggest Can Tho Riverport in Can Tho Province. In the other two subregions, the road density is low, underdeveloped, and not fully connected. The MKD region’s public agriculture support services are focused mainly on primary production (that is, irrigation, scientific and technological research, agricultural extension, veterinary, plant protection, licensing related to import and export of inputs [varieties, fertilizer, and plant protection products], and quarantine). Not many public services are developed to support processing and trading (that is, market information, trade promotion, and building and developing of brands), while private services are underdeveloped. Most public services show some limitations and have not kept up with the development of the sector. 19 Information was collected from websites of industrial zone management boards of all provinces in the MKD. Chapter II - The Mekong Delta and its subregions 19 The irrigation systems in the MKD mainly serve rice production. It can provide irrigation for 2 million ha and supply domestic water for 40 percent of the rural households. The irrigation management systems in the MKD include irrigation companies, irrigation departments, and water cooperatives. Their organizations and operations vary among provinces, constraining the regional cooperation in irrigation operation and even creating conflict among districts and provinces in border areas. Furthermore, limited personnel and limited financial capacity lead to improper maintenance and adjustment to suit non-rice-based farming systems. The MKD has widespread agricultural extension services (Figure II-4), and the quality of these extension services varies among subregions. The Upper subregion has an extensive network of government extension services and veterinary staff. Based on the calculation from AgroCensus data 2016, there are about 3 public extension and veterinary staff per 10,000 residents on average.20 The Middle subregion has the largest number of extension and veterinary officers, 5.3 staff per 10,000 residents on average, but the coverage is quite uneven. For instance, the network concentrates more in the western zone in Hau Giang Province but has very few staff in the middle part of the region between the Tien and Hau Rivers, where large fruit areas exist (less than 1 staff per 10,000 residents). In the Coastal subregion, the number of extension and veterinary officers is lower than in the other subregions, with only 1.5 staff per 10,000 residents on average. The limited support for involving private sector extension will hamper agricultural transformation in the MKD in the future. FIGURE II-4. Public services related to agricultural innovation in the MKD Agricultural extension service system Universities and research institutes Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ calculation from GSO 2018 (left); Authors’ illustration (right). 20 Local input dealers and private veterinary workers also provided some types of informal extension and technical advice to their clients. However, as they are not part of the formal systems, they receive little training from the government. The quality of their extension provided to farmers varies and is not monitored. They tend to advertise their products to sell them as much as possible. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 20 Alternative Scenarios and Policy Implications Science and technology are underdeveloped in the MKD. Most research institutes and universities in the MKD21 are concentrated mainly in the Middle subregion, focused on plant varietal development, feed innovation, agricultural mechanization, and agricultural processing technologies. For seeds, the Upper subregion can produce and supply sufficient rice seeds and Pangasius fingerlings while the Middle subregion and the Coastal subregion rely on imports for fruit seeds and shrimp post-larvae. Except for rice seed, the seed quality is still low and not adequately regulated. Regarding farm mechanization, the rice sector is highly mechanized and intensive, along with intensive shrimp farming,22 but this is not the case for other crops and other types of shrimp-farming schemes. The processing industry is relatively developed for rice and Pangasius fish but not for fruit, vegetable, and livestock products. Modern integrated processing systems that effectively use agricultural wastes and other by-products have not been widely developed. 3.2. Social dynamics The MKD is a populous region with a diverse population. In 2018, the population in the MKD was 17.6 million, distributed among the Middle subregion with 9 million (51 percent), the Upper subregion with 5 million (28 percent), and the Coastal subregion with 3.6 million (21 percent). The Upper subregion is home to the most diversified ethnic population, including Chinese, Khmer, and Cham, although the minority population is small, accounting for only 4 percent of the subregion’s community. Most minority groups live in the LXQ (particularly in An Giang Province). The Middle subregion has the largest ethnic minority population with about 800,000 ethnic people, mainly Khmer concentrated in Soc Trang and Tra Vinh Provinces. The Coastal subregion is home to 400,000 ethnic people, mostly Khmer. The Upper subregion is a hub of various religions, including Buddhism, Hoa Hao, Cao Dai, Chrism, Catholicism, and Islam. Owing to the earlier achievements in the agricultural sector, the MKD has experienced significant progress in poverty reduction and living standard improvement (Figure II-5). Based on Vietnam’s poverty standards,23 the incidence of poverty in the MKD fell from 36.9 percent in 1998 to 5.2 percent in 2016. The latter is below the national average (5.8 percent) and ranks third among all regions after the Southeast Region and the Red River Delta Region. The poverty rate in the Coastal subregion (5.9 percent) is higher than that in the Upper subregion (5.6 percent) and the Middle subregion (5.6 percent). The poor are mostly concentrated in the deep flooding areas of the northern part of the Upper subregion (near the Cambodian border), the western part of the Hau River of the Middle subregion, and the Eastern Coast of the Coastal subregion. Income inequality in the MKD is relatively high with the Gini coefficients of 39.9, although this is lower than the national average (42.4). Most people own durable goods (motorbike, television, telephone, and refrigerator) and access critical infrastructure (roads, electricity grid, post office, markets, schools, and clinics). 21 For example, the Cuu Long Delta Rice Research Institute, Southern Fruit Research Institute, Southern Irrigation Planning Institute, Southern Institute of Water Resource Science, Institute of Aquatic Economics, Institute of Aquaculture, Can Tho University (CTU), and An Giang University. 22 The area of rice land handled by machines has averaged 96 percent, with a seeding tool for 21 percent. Pesticide spraying machines reached 72 percent (nationwide 45 percent), harvesting machines over 90 percent, drying machines 46 percent, and milling machines 100 percent. 23 Poverty rate is calculated based on households’ monthly income per head, using the national poverty line for 2011-16. Chapter II - The Mekong Delta and its subregions 21 FIGURE II-5. Economic situations of farming households in the three MKD subregions Saving of households (VND million/household/year), 2010-2016 Poverty (%), 2016 40.3 39.9 39.2 37.7 37.2 36.7 36.1 35.9 35.4 34.7 34.8 33.1 33.3 32.6 32.6 32.4 32.0 30.4 29.4 25.0 2010 2012 2014 2016 2010 2012 2014 2016 2010 2012 2014 2016 2010 2012 2014 2016 2010 2012 2014 2016 Whole Whole Upper Middle Coastal country region Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ calculation from VHLSS and GSO (2018). The main social issues in the MKD are now its low-skilled and aging workforce. The total workforce in the MKD is 10.6 million, of which the Middle subregion has 6.6 million (62.1 percent), and the Upper subregion and the Coastal subregion each have 2 million (19 percent). About 68 percent of the workforce only had an elementary education or lower in 2016 (Figure II-6). Workers with college degrees in the MKD account for only 5 percent (country average of 6.32 percent). In 2017, the percentage of workers who received professional training was only 12.1 percent (country average of 21.4 percent).24 Among the three subregions, the workforce in the Middle subregion has a slightly higher educational attainment than the other areas. Still, it has recorded the slightest skills improvement (the percentage of workers without a college degree has been declining very slowly, a 2.1 percentage point decrease over six years from 2010 to 2016). The Upper subregion has the most significant improvement in higher education attainment in the MKD (the percentage of workers with a college degree and above increased by 2.84 percentage points over six years from 2010 to 2016). FIGURE II-6. Educational levels of the labor force in the three MKD subregions, 2016 100% 5.78 5.3 6.05 5.33 9.68 8.19 11.63 10.72 12.98 80% 20.11 20.02 21.52 19.78 22.49 College and above 60% 31.91 High school 34.54 33.92 39.15 40% 33.25 Primary 23.7 20% Elementary 26.54 30.28 25.23 27.32 14.6 0% No degree Whole country Whole region Upper Central Coastal Source: Authors’ calculation from VHLSS. 24 Authors’ calculation from VHLSS data. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 22 Alternative Scenarios and Policy Implications The MKD has experienced substantial out-migration (Figure II-7). Among the three subregions, the Upper subregion and the Coastal subregion have a high out-migration rate (Kien Giang and An Giang 9.0 percent, Ca Mau 8.0 percent, Bac Lieu 6.5 percent, and Soc Trang 5.5 percent).25 Most young people leave their farmland and seek other employment opportunities in big cities. However, most migrants can take only informal and seasonal jobs because of their lack of skills. There is a high possibility that, in the future, without further skills development, these people may have to return home unemployed in their old age. People who remain in the MKD are mainly middle-aged and more senior. The average age of agricultural workers in the MKD increased from 36 to 42 years during 2004- 16. The aging workforce was often accompanied by limited capacity in acquiring new techniques, thus keeping agricultural labor quality low. At some peak times, such as during land preparation and harvesting, it is harder to hire workers, pushing up labor costs and inducing greater mechanization. Consequently, this situation leads to households shifting from rice to other income sources, such as alternative crops and fishery. At the same time, the low income from rice production incentivizes the youth to leave agriculture. FIGURE II-7. Migration flows from the MKD Migration from the three MKD subregions Direction of out-migration flows (% of the population) from the MKD 12.0 Out-migration 10.0 8.0 26.3% 73.7% 6.0 Within the Mekong Delta Out of the Mekong Delta 4.0 70.1% 3.6% 2.0 5.1% 4.9% 3.9% Southeast Other regions 0.0 Tien Giang Kien Giang Can Tho 2005 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 50.9% 15.1% 4.1% Others 12.5% Ho Chi Minh Binh Duong Others Upper Middle Coastal Source: Authors’ calculation from statistical yearbooks (left), and Tran (2019) (right). Land consolidation remains a challenge. Given the high risk of losing jobs, migrants and their families still prefer to keep land as a kind of insurance or safety net. While land leasing has become more common, this usually occurs on an informal and short-term basis. This situation deters land renters from making significant investments in the land. Consequently, land consolidation accumulation is slow, and land productivity remains capped in many areas. 3.3. Biophysical dynamics An extensive water regulation system has been built in the delta to maximize rice production (Figure II-8). For flood control, there is a total of 13,000 km of dikes, including high dikes, to keep floodwater from entering rice fields in the flooding season. There are also flood bypass canals to drain floodwater to the West Sea to decrease flooding and August dikes or low dikes to prevent early- 25 Authors’ calculation from statistical yearbooks. Chapter II - The Mekong Delta and its subregions 23 season floodwater from entering fields to allow more time to complete the harvest of the second rice crop. In the Upper subregion, the high-dike system has been largely completed in the LXQ and partly in the Plain of Reeds (POR). In the Middle subregion and Coastal subregion, massive water regulation systems, sea dikes, and coastal sluice gates have been built to prevent saline intrusion and to bring freshwater to coastal areas for rice cultivation. FIGURE II-8. Water regulation engineering systems in the three MKD subregions Dike and sluice gate system Irrigation canal length Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ illustration from Southern Institute of Water Resources Research (2013) (left), Authors’ calculation from AgroCensus (2016) (right). The above flood and salinity control strategy has resulted in a mix of gains and pains: the expansion in output and trade as discussed above versus environmental and ecological costs, as discussed below. These pains may continue to amplify the challenges for the MKD in the future. Reduction in flood retention capacity in the Upper subregion. The rapid development of the high-dike system that enables rice triple cropping in the Upper subregion has taken up flood retention space. This has caused increased inundation in downstream provinces in the flood season and increased saline intrusion in the coastal provinces in the dry season (Dung et al. 2018). According to ICEM (2010), in the dike development decade (2000-10), a 42 percent reduction in retention capacity occurred in the flooded area within LXQ. Soil fertility depletion. Heightened dikes enable planting a third crop of rice inside diked areas during the flood season, but cut off the land from floodwater that brings sediments and associated nutrients. According to Chapman et al. (2017), farmers cultivating two crops in low-dike areas Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 24 Alternative Scenarios and Policy Implications received an average sediment deposit of about 2.5 cm, improving the average annual yield by 0.3 ton. However, the areas with fewer sediments deposited in the high-dike areas captured only about 0.5 cm of sediment deposit due to the high dike preventing floodwater from entering. This situation is more severe in the Upper subregion because of the large area devoted to triple rice. Stagnation of rivers and canals and water pollution. Many small rivers and canals have become disconnected from the main rivers and have lost tidal influence from the sea and flood pulse from the Mekong River. The loss of this self-purification capacity, while continuously receiving heavy loads of agricultural chemicals from intensive agriculture, makes the water heavily polluted and unsuitable for human consumption, and even for swimming or bathing (Chau et al. 2015; Sebesvari and Renaud 2017). It is increasingly urgent to study this problem further and address drinking water-related health risks. Groundwater depletion. Groundwater is used extensively in the Coastal subregion at the rate of 229,000 m3 per day to meet 87 percent of the demand for domestic purposes and 226,000 m3 per day to meet 16 percent of industrial production needs. Moreover, farmers in this subregion extract about 262,000 m3 per day to use in aquaculture production (input water for shrimp ponds) (Vietnam South Water Resources Federation 2013). The main reason for the heavy reliance on groundwater in this zone is the lack of freshwater in the dry season. In addition, the freshwater behind the controlled-salinity sluice gates is not suitable for human consumption as pollution is highly concentrated in stagnant rivers and canals during the dry season when the sluice gates are closed. Large-scale exploitation of groundwater has led to groundwater depletion in all subregions. This situation has contributed to the land subsidence of the whole MKD, and it also causes sea water with high saline concentrations to flow into freshwater aquifers. Loss of connectivity between the rivers and coastal water. Another environmental problem is the disconnect between the inland river systems behind the coastal dikes and the seawater from the sea in the dry season when all sluice gates are closed to keep freshwater for rice cultivation. Its multiple impacts include: (a) the loss of the brackish-water environment as the sluice gates, by separating the freshwater and saline water, eliminate the interface between these two types of environments; (b) the loss of migration routes for coastal fisheries to find foraging and spawning habitats in the brackish water and vice versa for freshwater species spending part of their life cycle in the brackish- water environment; (c) accumulation of pollution on the inland side of the rivers as the flows become stagnant, resulting in a loss of inland fishery resources; and (d) the pollution of coastal water with the periodic release of accumulated pollution behind the sluice gates to the sea. Soil erosion. The river and coastal banks experience severe soil erosion (Anthony et al. 2015). The leading causes include: (a) a significant decrease in suspended sediments from the Mekong River that may be linked to upstream dam retention of its sediment, (b) large-scale commercial sand mining along rivers in the MKD, and (c) mangrove forest degradation. MARD (2019) reported 136 erosion hot spots in the Upper subregion and 513 hot spots in the Coastal subregion. Forest loss. According to Thu and Populus (2007), over 36 years (1965-2001), the total coverage of mangrove forests in the MKD decreased by 50 percent. The forest area lost in the Coastal subregion Chapter II - The Mekong Delta and its subregions 25 was mainly due to coastal erosion and deforestation for aquaculture, fuelwood, and timber. However, Truong and Do (2018) argued that the allocation of mangrove forests in the MKD for protection, management, and logging, which allowed households to convert 40 percent of the allotted forests into other uses (mainly shrimp farming), is the leading cause of the mangrove forest destruction. The current mangrove forest cover in the MKD is 4.9 percent. The reduction in the mangrove forest area is another reason for reducing flood retention and flood regulation capacity of the delta and increasing soil erosion in all subregions of the MKD. Loss of wild capture fishery resources. The high-dike system has resulted in the loss of floodplain habitats in the flood season, thereby reducing fishery resources in the delta, especially in the Upper subregion26 (Baran, Guerin, and Nasielski 2015; Hashimoto 2001). As many canals and rivers have become slow-flowing or stagnant and some are entirely covered with water hyacinth, they are unsuitable for fish to live and spawn. Rapid spread of water hyacinth. No statistics exist on water hyacinth growth in the MKD. The scientific community has not paid due attention to the effect of water hyacinth proliferation in the delta. However, it has been reported by the media as a tremendous environmental problem in the delta’s waterways.27 Water hyacinth is causing a wide range of undocumented environmental problems in the delta, including: (a) loss of fishery resources; (b) loss of navigation ways; (c) impacts on the prices of agricultural products in areas covered with water hyacinth, as it clogs the waterways and boats cannot reach to those areas; and (d) changing of the aquatic environment as sunlight cannot pass into the water column beneath the thick blanket of water hyacinth. Many provinces have spent large amounts of money to remove water hyacinth with limited success.28 3.4. Institutions 3.4.1. Delta management Vietnam has no legal framework for an MKD regional authority. The government is piloting a regional link scheme. The provinces in the Upper subregion and the Middle subregion are currently pioneering regional links with two proposals to the government for LXQ (An Giang, Can Tho, Hau Giang, and Kien Giang) and POR (Dong Thap, Long An, and Tien Giang). In practice, some types of inter-provincial links have taken place but on information sharing only. Only Kien Giang and An Giang Provinces have official protocols for cooperation in water regulation. The lack of regional links is a significant constraint for the effective use and management of water resources and engineering construction operations. 26 The annual catch in An Giang declined from 95,000 tons in 2001 to less than 20,000 tons in 2017 (RIA2 2018). 27 http://moitruong24h.vn/tham-hoa-luc-binh-o-dbscl.html. 28 Tien Giang spent VND 8 billion in 2018 and Long An VND 3 billion in 2015 on water hyacinth weeding. https:// vnexpress.net/thoi-su/tien-giang-chi-hon-8-ty-dong-diet-luc-binh-bang-tay-3780695.html. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 26 Alternative Scenarios and Policy Implications 3.4.2. Regional governance The MKD has good local governance, as reflected by its Provincial Competitiveness Index (PCI) ranking (Figure II-9). The Upper subregion has the strongest governance system compared to the other subregions in the MKD. Dong Thap and Long An ranked second and third in the national PCI 2018 ratings, respectively, showing the consistent and considerable efforts of the local authorities to improve the business enabling environment.29 An Giang Province is in the mid-high group, ranking 6 out of 13 provinces of the MKD. In the Middle subregion, governance quality is better in the eastern part (Ben Tre, Vinh Long, and Tien Giang Provinces) than in the western part (Can Tho, Hau Giang, and Tra Vinh Provinces). In 2018, Ben Tre and Vinh Long Provinces were in the top ten nationally. Can Tho and Kien Giang are in the mid-high group, and Hau Giang and Tra Vinh are in the mid-low group.30 The Coastal subregion has the weakest governance. For PCI 2018 rankings, Bac Lieu was at the bottom of the mid-high group, and Soc Trang and Ca Mau Provinces were both in the mid-low group. Ca Mau ranks last among the 13 MKD provinces.31 FIGURE II-9. Provincial Competitiveness Index (PCI) of 13 provinces in the MKD, 2010-18 80 2010 2014 2018 60 40 20 0 An Dong Long Ben Tien Vinh Can Hau Soc Tra Bac Ca Kien Giang Thap An Tre Giang Long Tho Giang Trang Vinh Lieu Mau Giang Source: VCCI (2019). 3.4.3. Community value Community organizations and nongovernmental organizations (NGOs) play essential roles in communicating policies and facilitating farmer access to services. Farmers are typically members of one or several community organizations. In the agricultural sector, these organizations help disseminate policies; facilitate farmer training, extension, and access to microfinance; and support natural disaster rehabilitation. About 40 international NGOs are operating in the MKD, and all provinces have active international NGOs that support fund-raising and capacity building 29 Dong Thap Province has remained in the top five for the past 11 years. Long An has made significant improvements in creating a fairer business environment with less policy bias toward state-owned, foreign, or connected firms and through facilitation of land access and tenure for the private sector. 30 Ben Tre was rated positively on efforts to reform administrative procedures and for creating a conducive legal environment for business contraction and expansion. Vinh Long was the highest rated province in the country for decreases in informal charges. Tra Vinh received the nation’s lowest score for labor policy and Kien Giang the lowest score for business entry costs. 31 The coastal provinces have very low ratings in relation to informal charges, proactive leadership, and labor policy and law. Chapter II - The Mekong Delta and its subregions 27 for Vietnamese NGOs. Apart from those, Dong Thap Province in the Upper subregion initiated the establishment of ‘society’s premises’ (Hoi Quan) to promote the community spirit for rural people. During 2017–2019, Dong Thap Province established 63 society’s premises and attracted more than 3,000 members. The Society’s premises have become an effective forum for exchanging information and dialogue between provincial leaders and the business community. Nevertheless, there is a lack of effective consultation between local authorities and communities in the government’s decision-making process. Although communities have settled and exploited natural resources for generations, they have not been assigned the primary role for managing the resources.32 According to Hoang et al. (2009), in the MKD, formal complaints are increasing in diverse areas, and they are getting more complicated and complex. Community actions tend to be ad hoc, localized, and reactive, often based on immediate and visible economic and health threats rather than systematic preventive approaches to avoid long-term environmental damage (Boris 2011). 3.4.4. Producer organizations and value chain links Producer organizations, comprising collective groups and cooperatives, enable the collaboration of small farmers to achieve economies of scale. The MKD ranks first nationwide in the number of collective groups and the number of new-style cooperatives.33 The Middle subregion has the largest number of cooperatives and collective groups, concentrated mainly in the western part of the subregion (in Can Tho and Hau Giang Provinces). Next is the Upper subregion, concentrated more in Dong Thap and An Giang Provinces. In the Coastal subregion, the number of collective groups is twice those of the Upper subregion, but the number of cooperatives is the lowest in the delta, and most of them are shrimp cooperatives. However, most cooperatives mainly provide production support (67 percent of the cooperatives offer services in production, 0.13 percent in machinery, 3 percent in supplying inputs, and 11 percent in trade).34 The private sector plays a vital role in the MKD’s economy (Figure II-10). The total social investment capital35 of the MKD is about VND 180 trillion, of which 80 percent comes from private investment, including from enterprises and households. In 2017, the MKD had 3,124 enterprises 32 For example, in the case of mangrove management, two-fifths of the provinces assigned all their mangrove land to the Mangrove Management Board (Ben Tre and Kien Giang) and one province assigned most of its mangrove land to the Mangrove Management Board (73 percent of mangrove lands in Ca Mau). Coastal tidal flats are also mainly managed by the government; in some cases, the land is handed over to enterprises/individuals without the consent of the community. 33 The cooperatives are established and operated following Cooperative Law 2012. 34 Calculation from Enterprise Survey, 2018. 35 Social investment includes public investment, private investment, and foreign investment. It is the entire amount of spending of the whole society to increase or maintain capacity and resources for production including investment to generate fixed assets and increase current assets; spending on purchasing rare and precious assets; reserve gold in the form of goods; and storage of commodities in the resident and other investments to improve people’s knowledge, enhance social welfare, improve the ecological environment, and so on. It does not include investments which transfer the right of use or ownership among individuals, households, enterprises, or organizations, without increasing fixed assets and current assets in the locality. Available at https://www.gso. gov.vn/en/metadata/2019/03/investment-and-construction-2/. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 28 Alternative Scenarios and Policy Implications operating in agricultural production, trade, processing, and services.36 This number accounted for 25 percent of Vietnam’s total agricultural enterprises. The MKD ranks after the Southeast Region only in terms of agricultural enterprise numbers. However, the growth rate of enterprises in the MKD is lower than in other subregions. The Middle subregion has the largest number of agricultural enterprises (over 70 percent of the delta’s total in 2016). Enterprises in the Middle subregion have a larger scale in labor and capital and are more profitable than those in the other two subregions. The Coastal subregion has the lowest number of enterprises compared to the other two subregions. The household economy still dominates across the three subregions, making up 45.4 percent of the regional total social investment capital. FIGURE II-10. Contribution of the private sector in the MKD’s economy Social investment in three MKD subregions, Number of agricultural, forestry, and fishery 2018 (VND, billions) enterprises in the MKD, 2010–2016 Capital of households State investment State enterprises Cooperatives Capital of enterprises Private enterprises Foreign enterprises 46,624.0 34,719.5 14,440.2 18,435.2 16,544.3 2010 2011 2012 2013 2014 2015 2016 2010 2011 2012 2013 2014 2015 2016 2010 2011 2012 2013 2014 2015 2016 27,472.7 8,366.2 10,826.8 7,165.3 Upper Middle Coastal Upper Middle Coastal sub-region Source: Authors’ calculation from statistical yearbooks (left), Authors’ calculation from enterprise surveys (right). Agricultural value chains in the MKD are more developed for rice and fishery products than for fruits. In the Upper subregion, An Giang Province is the pioneer in promoting the large-scale rice field model37 and had the largest area under this model in 2018 at 56,000 ha. About 80 percent of the Pangasius fish area is now under the farmer-enterprise link scheme, involving coordinated input supply, harvest scheduling, and product purchases. The Coastal subregion has more links in the shrimp sector. Some big companies supply shrimp seed, feed, and technical assistance (TA) and then collect shrimp produce from the farmers. Most of them apply advanced technology in intensive shrimp farming, such as Minh Phu Seafood Corp. and Viet Uc Seafood Corp. Currently, some companies are cooperating with farmer groups to produce shrimp following the Aquaculture Stewardship Council (ASC) standards or organic standards in rice-shrimp or mangrove-shrimp. Fruit production in the Middle subregion is still fragmented, although more companies are starting to invest 36 This number includes enterprises in agricultural production, aquacultural production, and agriculture-related service and processing, of which agricultural production enterprises comprise crop production enterprises and livestock production enterprises. Related services refer to seedlings, breeds, animal feeds, and so on. 37 Large-scale rice field model is a type of production organization, in which enterprises or cooperatives establish a cooperative relationship with farmers to apply uniform production procedures through providing members with production inputs (including material and technical support) and buying outputs from producers or facilitating joint marketing of farm produce. Chapter II - The Mekong Delta and its subregions 29 in fruit processing. In general, the agricultural value chains in the MKD are still weak and vulnerable to market volatility. The trust between farmer organizations and agribusinesses still needs to be built and strengthened by a trustworthy conflict-solving mechanism. 4. Key external driving force - opportunities and challenges for future development 4.1. Commodities market As Vietnam is opening to the world and becoming a major agricultural exporting country, market demand is the driving force for the agricultural sector. This is particularly true for the MKD, the most commercialized agricultural production region, accounting for most of the national agricultural export value. Market opportunities exist for all three critical commodities of the region. 4.1.1. Rice It is projected that although the demand for rice as a staple will increase globally,38 the national demand for rice would be lower in 2030 than in 2017 as the decline in per capita consumption is expected to exceed the population growth rate. Calculations using VHLSS showed that per capita rice consumption in the MKD decreased from 10.9 kg/person/month in 2008 to 8.8 kg/person/month in 2016 and will continue to decline in the coming years as income increases. According to IPSARD’s estimates, rice consumption per capita by 2030 will diminish to 7.8 kg/person/ month. As Vietnam’s population is forecasted to increase to about 104 million people by 2030 (an increase of 9.3 million people compared to 2018), the amount of rice needed for domestic demand will be about 9.7 million tons. The World Bank (2016a) estimated that, in Vietnam, by 2030, rice consumption would decrease by 10 percent, and the rice surplus would be 6.9 million tons in general, with 5.8–6.4 million tons in the MKD in particular. In the domestic market, the demand for high-quality rice would increase while the demand for low-quality rice decreases. Also, the demand for rice for processed food in the domestic market would increase significantly, especially for nutritious food such as rice milk and rice cake. In the international market, the world’s rice trade is expected to increase as the world population grows, rising from the current 7.5 billion to about 10 billion by 2050. The total demand for rice will increase to 520 million tons by 2030 and up to 556 million tons by 2050 (Alexandratos and Bruinsma 2012). The supply may decline because of the impact of climate change and diseases. In addition, the high-end market for rice is still open for Vietnam, such as the European Union (EU) and Japan, especially when the Comprehensive and Progressive Agreement for Trans- Pacific Partnership (CPTPP) and the EU-Vietnam Free Trade Agreement (EVFTA) come into effect. The COVID-19 pandemic has pushed rice demand upward in the short term as countries built up rice reserves due to concerns about food security. 38 Total global rice trade is up 1.5 percent in 2016-24 (OECD and FAO 2015). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 30 Alternative Scenarios and Policy Implications Some importing countries, such as China, Indonesia, and the Philippines, are promoting increased rice production to achieve greater food security. In addition, there is a trend toward higher-quality standards and food safety requirements in many importing countries, including in traditional markets such as China and African countries. Regarding global rice supply, traditional rice exporters continue to sustain their production, and some new producers have emerged, including Cambodia and Myanmar. The international rice market is becoming more competitive, putting pressure on prices. According to the World Bank, the export price of Vietnamese rice is projected to decline by 10 percent from US$423 per ton in 2014 to US$380 per ton in 2025. 4.1.2. Vegetables and fruits Vegetables. The target market for vegetables from the MKD is the regional market and neighboring cities, particularly HCMC. Based on VHLSS data, it is projected that, by 2030, national vegetable consumption will increase to 7.5 million tons/year (VND 3.98 billion/year), of which the MKD will consume 1.3 million tons/year and HCMC 665,000 tons/year. Consumers will demand more safe and higher-quality products. Regarding the world market, the demand for vegetables of all kinds will continue to increase, while the current export of vegetables from the MKD remains minimal. This is understandable as the present number of processing enterprises and their scale of operations in the MKD are small. Besides, vegetable production currently remains fragmented, and the cost is high because of the many intermediaries involved in the value chain. Fruits. The supply of fruits at present is not meeting the increasing demand domestically and abroad. According to the World Bank (2016a), fruit demand in Vietnam is projected to increase from 5 million tons in 2009 to 7 million tons in 2030. Globally, the fruit market has a high potential and is forecast to continue to grow. The value of exports in the world fruit industry reached US$100 billion in 2016 and is expected to reach US$200 billion by 2030. Vietnam is an emerging exporter, with the total export value of fruits and vegetables increasing by more than 7.5 times from less than US$0.5 billion in 2010 to more than US$3.2 billion in 2020. The capacity to produce various fruit types and crops is an advantage for Vietnam. Fruits in the MKD are now mainly consumed domestically in fresh form. The export share is insignificant compared to the harvested output, not yet exploiting the full potential of destination markets. Vietnam’s participation in the CPTPP and EVFTA offers excellent market potential for processed fruits, given the significant and tariff reduction commitment by member countries compared with fresh and raw products. The market potential for the fruit and vegetable sub-sector is enormous and increasing in both domestic and international markets. The annual export turnover of fruits and vegetables is US$3.7 billion, but this accounts for only about 10 percent of the total value of domestic vegetable and fruit production. The primary destination export market is China (accounting for more than 70 percent of the total export turnover). Vegetables and fruits of the MKD have access to high-value markets, but exports remain limited. The challenges that the MKD’s vegetables and fruits have been facing include: (a) food safety problems because of high pesticide residues, (b) strict regulations on post-harvest handling and control with a lack of support services and high costs (high-quality certification services and irradiation, hot steam, and heat treatment services), (c) weak capacity to preserve fresh fruits and vegetables after harvest and export, (d) limited traceability due to weak links between stakeholders Chapter II - The Mekong Delta and its subregions 31 and complicated processes, and (e) weak brand building and protection. In addition, logistics services are not effective and are costly. Vietnamese logistics enterprises are small, fragmented, and not well connected, and the efficiency index is low. According to the World Bank, Vietnam is one of the countries having high logistics costs, higher than in many Asian countries (Thailand, Singapore, and Malaysia), and much higher than in the EU and the US. In agriculture, logistics costs account for 29.5 percent of the vegetable price and 30 percent of the export rice price. 4.1.3. Aquatic products Pangasius fish. The demand for Pangasius fish is projected to continue growing. The EU and North American markets are forecasted to remain firm, and Vietnam has not yet fully exploited the potential of these markets. The U.S. Department of Agriculture (USDA) has officially recognized the inspection and control system for Vietnamese catfish and catfish products to be equivalent to the US, allowing Pangasius products to enter these markets more easily. Pangasius products are also common with low-income groups in Asia, South America, and Africa. The EVFTA and CPTPP will provide potential opportunities for Vietnamese Pangasius exports to markets such as the EU, Mexico, Japan, and Chile. However, many countries worldwide are now beginning to switch to Pangasius farming, such as India, Bangladesh, Thailand, Indonesia, and China, competing with Vietnamese Pangasius products. Like other agricultural products, Pangasius products continue to carry warnings about food safety when exported to high-value markets such as the US, EU, and Eurasian Economic Union, and the risks of return prevail. Shrimp. It is projected that the world shrimp supply will not meet the increasing demand for shrimp. OECD and FAO (2015) indicated that, if not subjected to significant market fluctuations, pollution of the environment, disease, and especially the impact of climate change, by 2020, the world supply would be 3.8 million tons, and demand would be 4.56 million tons, resulting in a deficit of 0.76 million tons. By 2030, the figures are 4.22 million tons, 5.25 million tons, and 1.03 million tons, respectively. Because of climate change, the shortage could be more prominent. Like the Pangasius sub-sector, Vietnam’s shrimp sector will have opportunities to expand its markets when the EVFTA and CPTPP take effect, especially for processed shrimp products. However, the shrimp industry also faces trade barriers and protection measures in both traditional and new markets. Notably, unexpected market upheavals can occur, especially market price volatility. Vietnamese agro-products have to meet more strict regulations on quality and food safety and technical barriers, even in traditional ‘easy’ markets. Integration also means opening up and accepting more competition in local markets. In short, there is significant market potential. But to meet this potential, the MKD needs to adapt quickly to meet market demand in both quantity and quality. 4.2. Science and technology The world is at an early stage of Industry 4.0 (including digital technology, artificial intelligence, big data, Internet of Things, social, mobile, analytics, and cloud services [SMAC]); such development offers opportunities to address the technology constraints faced by the MKD. In addition to new varieties, environment-friendly inputs, processing technology, Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 32 Alternative Scenarios and Policy Implications and advanced technology can support flood monitoring and forecasting in the Upper subregion and improve the understanding of flood-based agriculture such as floating garden initiatives. For the Middle subregion, there are initiatives on mechanization and integrated farming in the fruit areas. For the Coastal subregion, there are initiatives on integrated mangrove-shrimp, closed circulation shrimp-farming systems, micro water treatment systems for small households, salinity or polluted- water treatment systems for drinking water, salinity monitoring and forecast, sea-based aquaculture, and so on. On the other hand, the negative impacts of new technologies on the workforce in the MKD come from the effects on agricultural jobs and industrial and service jobs that most people rely on for nonfarm income sources. According to the World Economic Forum (2018), Vietnam is among those that are not currently ready for Industry 4.0. It ranks 90th in technology and innovation and 70th in human capital among 100 countries. As predicted by Oxford Economics (2018), in Vietnam, in 2028, about 3.4 million jobs in agriculture would be displaced, accounting for 17.1 percent of the agricultural workforce. In other words, more than 90 percent of the redundant jobs of the country will occur in the agricultural sector alone. The ILO (2018) forecasted that 85 percent of the employment in agriculture and 86 percent in Vietnam’s textile and footwear industries would be at risk due to automation. The majority of the affected workforce would be young people, mainly unskilled laborers and skilled agricultural workers, including crop and livestock farmers, machine operators, and trade workers (ILO 2018; Oxford Economics 2018). Even in the agricultural sector, the skill requirements for labor would also be different, so new types of training would be required (that is, information technology [IT] skills, management skills, and so on). By gender, the ILO (2018) found that female workers would be at a higher risk of being replaced by automation than males. In Vietnam, 79.7 percent of female workers would be affected compared to 61.7 percent of male laborers. With a low-skilled workforce and high out-migration rate, it is expected that the MKD will be at increased risk of having insufficient jobs for household members who switch to local nonagricultural employment or migrate to cities for formal employment. Only a few livelihood models with high labor productivity will be attractive to labor in the region, such as catfish and shrimp production (World Bank 2016a). 4.3. Natural hazards and climate change Vietnam faces many natural hazards, including floods, typhoons, saline intrusion, and coastal erosion, and the most severely affected regions are the Red River and Mekong deltas. Agricultural crops, especially those in coastal areas, are significantly exposed to flooding, and rice is the most threatened crop. Studies showed that a combined riverine and coastal flood with a 25-year return period would inundate 17 percent of all rice paddies in coastal provinces, and it would increase dramatically to 34 percent for a 1-in-50-year flood, and 63 percent for a 1-in-100-year flood (Rentschler et al. 2020). Storm surges associated with typhoons and rising sea levels cause more severe coastal flooding and saline intrusions which are destructive to agricultural crops. In the MKD, excessive groundwater extraction and droughts caused by erratic precipitation patterns in upstream areas have led to the depletion of freshwater aquifers and high saline concentrations to flow into freshwater aquifers. The heightened salinity levels are detrimental to water intensive rice production, irrigation more generally, and high-quality water supply to households and firms. Chapter II - The Mekong Delta and its subregions 33 In addition to exposure to natural stresses resulting from rare and extreme events, the country also experiences frequent El Niño–Southern Oscillation (ENSO), which occurs roughly every four years. The most recent El Niño event lasted for 18 months from November 2014 to May 2016. This El Niño caused high temperatures and reduced precipitation, leading to severe droughts and salinity intrusion, especially in the MKD. The peak of the drought, from February to May 2016, had large human impacts. An estimated 2 million people in the country had limited access to water for consumption and domestic use, and millions of people were food insecure. The poor and women were the most vulnerable groups. The total crop production area affected by drought was 820,000 ha, which comprised 355,000 ha of rice and 465,000 ha of other crops. The total land left to fallow was 127,000 ha. In total, this El Niño caused about US$674 million in economic losses in Vietnam. It is estimated that drought and saltwater intrusion during the 2014–2016 event caused an estimated US$3.6 billion in damages to agriculture, fisheries, and aquaculture alone (Sutton et al. 2019). Labor movement mitigates or exacerbates ENSO-related GDP losses. When agricultural production faces a climate shock such as El Niño, the economy adapts by reallocating labor, which, unlike land, is mobile between farm and nonfarm sectors, and it triggers more seasonal out-migration to cities. Climate change and upstream development also seriously challenge the sustainable development of the MKD. Vietnam is among the most vulnerable countries to climate change. Among the 84 coastal developing countries heavily affected by sea level rise, Vietnam ranks first in terms of consequence to population and GDP performance and second in terms of influence on land area and agricultural production (World Bank 2016b) and is one of the 30 ‘extreme risk countries’ in the world according to the 2014 Climate Change Vulnerability Index (CCVI) of Maplecroft (Nguyen et al. 2017). According to the average climate change scenario Representative Concentration Pathway (RCP) 4.5, in 2050, sea level rise compared with the average of the 1985-2006 period is projected to increase by 22 cm on the east coast and by 23 cm on the west coast. During 2016-35, the average temperature could increase by 0.4-1.2oC, rainfall could increase by 5.2-12.3 percent, rainfall amounts in the rainy season could increase more, and those in the dry season could decrease more (MONRE 2016). In addition, the Mekong flows are influenced by a complex combination of drivers from the upstream Mekong. In addition to the current large reservoirs on the Lancang in the Upper Mekong Basin and on tributaries in the Lower Mekong Basin, 11 runoff-river dams are proposed and under construction in the Mekong Basin. The operation of the dams decreases streamflows in the wet season and increases streamflows in the dry season (Räsänen et al. 2017). Operators generally let water pass through within a day in ordinary years and significantly delay flows in dry years from 1.5 to 18 days each (ICEM 2010). Climate change and upstream developments could amplify the current environmental concerns and pose more challenges than opportunities for the MKD in the coming years. Rice yields are expected to be reduced by 4.3 percent over 2016–2045 period, below the level it would be in the absence of climate change. Sea level rise and salinity intrusion are expected to reshape the geography of rice production; however, aquaculture has a more promising outlook, with adapted species and innovative management systems (Nguyen et al. 2017). Impact on land subsidence. The MKD is sinking at an alarming rate. A study by Utrecht University reported that in the 25 years from 1991 to 2016, the average rate of groundwater extraction-induced subsidence of the delta was 1.0–4.7 cm/year and is accelerating (Minderhoud et al. 2017). During 1991- 2016, the delta subsided an accumulated depth of 18 cm and up to 53 cm at specific locations. Using Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 34 Alternative Scenarios and Policy Implications the same study model, the current rate of subsidence estimated is 1.1 cm/year, with the maximum speed at 2.5 cm/year in some areas, exceeding the rate of sea level rise. Another study (Erban, Gorelick, and Zebk 2014) projects that, if groundwater pumping continues at the present rate, a cumulative 0.88 m (0.35–1.40 m) of land subsidence can be expected by 2050. The Coastal subregion faces the most severe problems, while the Upper subregion is the least affected by land subsidence. Impact on salinity intrusion. Saline intrusion in the MKD is conventionally seasonal, but in recent years, it is on an increasing trend in the dry season. In unusual dry years, about 35 percent of the delta’s area could be affected with salinity higher than 4 g/L (MARD 2019). The Coastal subregion is the most affected: salinity can intrude in 60-90 percent of the area and last for 40-100 days, with 30-35 g/L. The artificially ‘freshened’ areas in the Middle subregion are also vulnerable to salinity intrusion in the dry season. The severity of salinity intrusion in the delta is predictable as the dry season’s salinity intrusion peaks from March to early April and is closely linked with the peak of the Mekong flood season in the previous year. In recent years, climate change in the Mekong Basin upstream of the delta has been exacerbated. In severe years, saline intrusion causes damage to crops in the areas that have not been intruded by salinity before, causing a shortage of drinking water and increased extraction of groundwater and an increased concentration of salinity in aquaculture ponds that result in additional costs to farmers. The IPSARD (2016) research found that the MKD farmers’ adaptation measures were not sufficiently effective in coping with severe droughts and salinity intrusions occurring in 2016. Empirical evidence shows that saline concentrations of over 4 milligrams per liter make monocultural rice production unviable. In 2016, 22 percent of rice paddies in the MKD were exposed to salinity levels above this threshold, equivalent to 12 percent of national rice production and 8 percent of national agricultural GDP, directly affecting the livelihoods of millions of rice farmers. Impact on flood regime. Under the combined effects of climate change, upstream hydropower reservoirs, and high-dike systems, water flows are anticipated to be more extreme, either with worse floods in La Niña years or worse droughts in El Niño years (Hoang et al. 2019). High floods of 4 m and above might occur more frequently in the Upper subregion, rendering the third rice crop in the flood season and the associated high dikes highly vulnerable to damage. Nguyen et al. (2015) showed that sediment dynamics would be seriously altered because of the operation of upstream hydropower plants. Impact on coastal erosion. The general trend in the near future is fast-paced erosion and more erosion than deposition. In the coming years, coastal erosion will increase. According to ICEM (2010), all six coastal provinces will experience coastal erosion at an annual rate of 34-44 m, twice that of the past 40 years. Impact on water supply and drainage. Given the current flood and salinity control infrastructure and intensive farming practices in the MKD, some regions (especially Can Tho, Hau Giang, and the northern area of Quan Lo-Phung Hiep) are experiencing local flooding due to the inability to drain water when high tide and flash floods occur together. Some areas of the Ca Mau Peninsula, which are located far away from large tributaries, continue to be at risk of severe freshwater shortage in the dry season. The lack of waste treatment in aquaculture and the intensive paddy farming system result in pollution of surface water resources (including rainwater) and lead to increased groundwater exploitation and escalating subsidence. Chapter II - The Mekong Delta and its subregions 35 CHAPTER III Assessment of main and potential livelihood models of the three Mekong Delta subregions Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 36 Alternative Scenarios and Policy Implications In this chapter, an MCA was employed to assess the characteristics and suitability of a variety of farming systems as alternatives to rice monocropping. The research team conducted assessments for existing farming models in the Upper subregion, Middle subregion, and Coastal subregion of the MKD based on the four main criteria: natural condition suitability and climate resilience, environmental suitability, economic and financial, and social and gender aspects. The findings highlight the most promising production systems for the MKD farmers and policy makers to consider when moving forward with the agricultural transformation program for the MKD in the coming decades. 1. Upper subregion 1.1. General situation The Upper subregion is the most productive rice production center of the MKD. Over the past two decades, this subregion has witnessed a rapid transformation from a 1-crop rice system to 2 and 3 high-yield rice crop systems. One key factor driving this change was the government’s food security policy, which promoted high-dike construction to protect this subregion from floods from July to November each year. The total rice area in this subregion increased by 8.5 percent from 2006 to 2014 and decreased slightly from 2015 to 2017, reaching 734,000 ha (producing 9.6 million tons of rice) in 2017. In 2017, 2-crop and 3-crop rice systems were still dominant, accounting for 60 percent and 36 percent of the total rice land in the Upper subregion, respectively. Recently, because of the low profitability of rice production, farmers in some districts of Dong Thap Province shifted to two high-quality rice crops or one high-quality rice crop combined with flood-based lotus or rice-fish intercropping. Nevertheless, these models have not yet been adopted widely because the markets for these new products have not been developed. Fruit production has expanded in the past five years. Despite its present small area of 48,500 ha (15 percent of the total area), fruit production in the subregion had the highest growth rate, about 7 percent per year during 2012–2017. Horticulture developed rapidly in places where farmers converted the low-value 3-rice crop systems to fruit tree gardens such as in Cho Moi (3,500 ha), An Phu (432 ha), Tri Ton (352 ha), Tan Chau (351 ha), and Chau Phu (302 ha) Districts of An Giang Province. Some farmers started establishing specialized gardens for citrus, mango, and dragon fruit, although mixed types of fruit tree gardens are still prevalent. Vegetable production has also emerged recently in some districts of An Giang Province. Freshwater fisheries are highly developed in the Upper subregion, combining aquaculture and capture fisheries. Pangasius is the main culture species in An Giang and Dong Thap Provinces, accounting for 80 percent of the production of the delta. Dong Thap is the leading province in Pangasius production in the MKD, with an area of 2,500 ha and an annual output of 466,000 tons. Initially, Pangasius cage culture was developed in the mainstreams of the Hau and Tien rivers, and then it was shifted to pond culture. Flood-based aquaculture during the flood season, including giant freshwater prawns, snakehead fish, and others, has been developed for the past 10 years in the areas outside high dikes. However, developing and securing markets for these new products is still a major challenge, especially when large-scale transformation occurs. Flood-based fishing, a traditional Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 37 livelihood strategy, has diminished markedly due to depleted fishery stock and less predictable floodwater of the Mekong River caused by upstream hydropower plants. The animal husbandry sub-sector is underdeveloped in this subregion. The number of buffalos decreased by 29 percent from 2010 to 2018. And the pig herd decreased from 545,000 to less than 300,000 during 2005-15. Poultry (ducks and chickens) raising is concentrated in Phu Tan and Cho Moi Districts in An Giang Province and Thanh Hoa District in Long An Province, with a total of 500 million heads. Free-range duck raising in rice fields, which used to be widely practiced in the Upper subregion, has declined and there has been a shift to caged duck-raising systems. 1.2. Key agricultural livelihood models In this study, local livelihood models were reviewed to identify the most common ones (Figure III-1). The criteria used included the production area, production value, and the quantity of participating households. The AgroCensus statistical calculations show that 3-crop rice and 2-crop rice systems occupied the most agricultural land in the subregion (690,000 ha or 79 percent) and contributed 54 percent of total production value. Pangasius fish production covered only about 3,000 ha but contributed almost 10 percent of the total regional production value (Table III-1). Vegetable farms and fruit orchards accounted for about 6 percent of the subregion’s agricultural land use and had a share of 13 percent of the agricultural production value. The vast majority of farmers engage in rice production, with more than 310,000 households working in 3-crop and 2-crop rice models. In terms FIGURE III-1. Map of key livelihood models in the Upper subregion Key livelihood models in Mekong Delta upper sub-region Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ illustration from Dang et al. (2019) and survey data. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 38 Alternative Scenarios and Policy Implications of participation, fruit and vegetable production is next in prominence with some 55,000 households. Only about 500 households are engaged in Pangasius production due to the high investment and production costs (and risks) involved. TABLE III-1. Coverage of the most common livelihood models in the Upper subregion % of % of Number of Production % of % of agricultural agricultural house- value agricultural agricultural Livelihood Area land in land in MD holds (billion production production models (000 ha) Upper (000) VND) value of value of MD Region Upper Region 3-crop rice 348.8 40 11 174.4 42,489 33 6 2-crop rice 342.0 39 10 143.2 25,781 21 4 Rice- 8.6 1 0 3.6 2,713 2 0 vegetable Fruit 41.3 5 1 48.7 12,267 10 2 plantation Specialized 26.1 3 1 20.9 10,656 9 2 vegetable Pangasius 2.0 0 0 0.5 13,200 11 2 fish Source: Authors’ calculation from AgroCensus 2016 and Dang et al. (2019). 1.3. Assessment of key and potential livelihood models 1.3.1. Natural condition suitability and climate resilience The index of future national condition suitability is used to assess the natural condition suitability of livelihood models. It was calculated based on the following two indicators: (a) the share of the most suitable and suitable land of a livelihood model in the subregion’s total agricultural land and (b) the percentage change in the most suitable and suitable land of the model during 2019- 30. The natural condition suitability shows that the flood-based (that is, rice-lotus and duck-raising) livelihood models are the most suitable to the water and soil conditions of the Upper subregion in 2030. Two-crop rice, rice-aquaculture, rice-fishing, Pangasius fish, fruits, and specialized vegetable models are ranked as moderately suitable. Three-crop rice and rice-vegetables models are the least suitable to the natural conditions in 2030 (Table III-2). Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 39 TABLE III-2. Ranking of models: suitability to future conditions, Upper subregion Area of most Proporption Change of Index of Index of Index of suitable and of most suit- most suitable column A column B suitability Livelihood suitable land able and suit- and suitable with future models in 2030 able land in land 2019- natural (thousand ha) 2030 (%) (A) 2030 (%) (B) condition Rice-lotus 591 56.9 62.8 3.0 5.0 4.0 Duck 1,000 96.2 0.0 5.0 3.0 4.0 2-crop rice 666 64.1 −17.8 4.0 1.0 2.5 Rice- 350 33.7 0.0 2.0 3.0 2.5 aquaculture Rice-fish 350 33.7 0.0 2.0 3.0 2.5 capture Pangasius fish 350 33.7 0.0 2.0 3.0 2.5 Fruit tree 443 42.6 −3.1 3.0 2.0 2.5 Specialized 440 42.3 −4.3 3.0 2.0 2.5 vegetable Rice- 461 44.4 −24.4 3.0 1.0 2.0 vegetable 3-crop rice 457 44.0 −25.0 3.0 1.0 2.0 Source: Authors’ calculation from Dang et al. (2019) and survey data. Note: The index of column A is computed as 80-100%: 5; 60-79%: 4; 40-59%: 3; 20-39%: 2; and 0-19%: 1. The index of column B is computed as 20-100%: 5; 1-20%: 4; 0%: 3; (−1)–(−20%): 2; (−20)–(−100%): 1. The index of future natural condition suitability is the average of the indexes of columns A and B. Concerning climate change resilience, assessments of suitable livelihood models in the Upper subregion are presented in Table III-3. The rice-lotus, rice-aquaculture, rice-fishing, and duck- raising livelihood models appear as the most resilient to climate change, while fruit trees, specialized vegetables, 3-crop rice, and Pangasius fish exhibit the lowest climate change resilience. These ranking results are based on the comparison of their absorptive ability,39 flexibility,40 and transformative ability.41 Lessons learned from the projects funded by the International Union for Conservation of Nature (IUCN) in 2018 showed that rice-lotus models in Dong Thap and An Giang Provinces had proved their resilience to flooding. Rice-giant freshwater shrimp and rice-fish systems can develop well in deep-flood conditions. Moreover, these models help improve soil fertility in the rice fields by allowing floodwater and sediments to come into rice fields. They also help increase the water retention capacity in the Upper subregion and reduce floods and saline intrusion for downstream provinces. The rice, vegetable, and fruit tree production models are less resilient to climate change. An IPSARD (2016) study shows that rice farmers working in conventional farming systems can cope with small extreme events but cannot overcome large-scale disasters for several reasons. First, farmers lack climate-tolerant rice varieties to deal with unexpected prolonged severe floods and droughts. 39 The ability of an agricultural livelihood model to absorb negative impacts of climate change. 40 The ability to change to other agricultural practices in a livelihood model. 41 The ability to transform to other agricultural livelihood models. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 40 Alternative Scenarios and Policy Implications Second, they cannot change their production calendar given their tight production schedule. Moreover, a transition from current key livelihood models to a flood-based one would require a structural modification in the interregional dike system (especially closed dikes). Pangasius fish and fruit production livelihood models are more resilient than rice because they are located in areas with adequate infrastructure to cope with disasters and other shocks. However, because of the high initial investments in fishponds or irrigated gardens, farmers operating these models often want to stick to them rather than shift to other livelihoods. TABLE III-3. Ranking of models for climate change resilience, Upper subregion Livelihood Absorptive Flexibility Transformative Climate change models ability ability resilience index Rice-lotus 5.0 4.0 5.0 4.7 Rice-aquaculture 5.0 5.0 3.0 4.3 Rice-fish capture 5.0 4.0 3.0 4.0 Duck 5.0 1.0 5.0 3.7 2-crop rice 2.0 3.0 5.0 3.3 Rice-vegetable 3.0 3.0 3.0 3.0 Pangasius fish 4.0 1.0 1.0 2.0 3-crop rice 2.0 1.0 3.0 2.0 Specialized vegetable 2.0 2.0 2.0 2.0 Fruit tree 3.0 1.0 1.0 1.7 Source: Authors’ evaluation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Looking ahead to 2030, considering both natural condition suitability and climate change impact, the livelihood model suitability map is expected to change markedly. The area suitable for 3-crop rice and rice-vegetable farming will decrease by some 24 percent (160,000 ha). Rice production in the western coastal areas of Kien Giang Province (that is, Hon Dat, Tan Hiep, and Rach Gia) and the deep-flood zone of the POR in Long An Province (that is, Vinh Hung, Tan Hung, Moc Hoa, and Thu Thua) will no longer be viable under a 3-crop rice system. The highly suitable and suitable area for the 2-crop rice model will also decline by 18 percent (140,000 ha). Other models would be only slightly affected by climate change. 1.3.2. Economic and financial aspects Three indicators (total investment, operational cost, and profitability) are used to compare the expected economic performance of the livelihood models. Households’ investment requirement varies significantly across livelihood models. The surveyed data show that 3-crop rice, 2-crop rice, and rice-vegetable systems need small investments from farmers. Farmers have to invest in intra-field irrigation with an investment amount of VND 3 million per ha and VND 5 million per ha for the 2-crop rice model and 3-crop rice model, respectively. In contrast, in the specialized vegetable, fruit, and Pangasius fish models, farmers need to make significant investments in in-field infrastructure (Figure III-2). For the fruit tree model, farmers need to invest in pumps, protective nets, pipes, and Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 41 field bed raising (and the latter may entail purchasing more land to achieve sufficient scale). The total investment for the fruit tree model is estimated at VND 50 million per ha. The model for growing specialized vegetables requires daily watering while avoiding waterlogging. Therefore, it requires more laborers. Some households have invested VND 150 million per ha to install an automatic irrigation system to reduce labor costs. The investment requirement for the Pangasius fish model is high, especially for pond design and preparation, water treatment, purchase of a feed-making machine, and so on. The estimated investment cost for the Pangasius fish model is expected to be VND 500 million per ha. FIGURE III-2. Investment in the infrastructure of the key models, Upper subregion Pangasius 500 Fruit tree 50 Vegetable 75 Rice-Vegetable 5 2-crop rice 3 3-crop rice 8 Unit: Million VND/ha Source: Authors’ calculation from survey data. There is a broad range of operational costs associated with the different models (Figure III- 3). The Pangasius production model has the highest production cost of about VND 6.4 billion/ha/year (about 80 percent of feed and veterinary drugs can be bought on credit). Therefore, only households with access to large working capital can adopt this model. The vegetable model also involves a considerable production cost of about VND 350 million/ha/year. The models of rice-vegetables and fruit trees have production costs ranging from VND 80 million/ha/year to 115 million/ha/year. In these three models, farmers have to pay more for purchased inputs such as chemical pesticides, fertilizers, and hired laborers (particularly at harvesting time). Rice farmers have lower operating costs compared to the other models. In the 2-crop rice and 3-crop rice models, the farmers often buy fertilizers and FIGURE III-3. Total production cost per hectare of key livelihood models, Upper subregion 0 1000 2000 3000 4000 5000 6000 7000 Pangasius Fruit tree Vegetable Rice-Vegetable Buying on credit 2-crop rice Immediate payment 3-crop rice Unit: VND, millions per ha per year Source: Authors’ calculation from survey data. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 42 Alternative Scenarios and Policy Implications pesticides on credit, and they do not need a lot of hired labor. Therefore, the operational costs for these two models are low at only VND 40-55 million/ha/year (50 percent of which can be paid with credit). High production cost is one of the main barriers for low-income households to shift from the rice production model to other high-value models. Generally speaking, rice production has a low but less variable profit than other farming systems (Figure III-4). The annual financial returns of the 3-crop rice and 2-crop rice models are small, at only VND 44–51 million per ha. The annual profit from the Pangasius farming model is the highest: VND 525 million per ha. The rice-vegetable model, specialized vegetable model, and fruit tree model also result in a significantly higher profit than rice farming (especially the 3-crop rice model). However, in the long run (10 years), the 3-crop rice and 2-crop rice are expected to provide a more stable annual income than other livelihood models. The standard deviations for the financial profitability of those models show that high volatility exists, particularly for the high-value models such as Pangasius farming or specialized vegetable growing. FIGURE III-4. Average annual profit of key models, Upper subregion, 2009-19 800 VND, millions per ha 3-crop rice 700 2-crop rice 600 Rice-vegetable 525 500 Vegetable 400 Pangasius 311 300 Duck (1000 heads) 191 Rice + Lotus 200 106 Fruit tree 77 77 55 100 51 44 Rice + Capture shery 0 Source: Authors’ calculation from survey data The surveyed data in this research show that, from 2009 to 2019, households that adopted the Pangasius fish farming model, specialized vegetable model, fruit tree model, or duck- raising model made significant profits (Table III-4). However, they also witnessed losses at some points. In the meantime, the specialized rice-farming models or rice rotation models offer low but stable profits over the past 10 years. Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 43 TABLE III-4. Ranking of livelihood models for economic performance, Upper subregion Net profit Economic index Livelihood models Investment cost Operating cost (weight=2) (weight=2) Rice-lotus 5.0 5.0 3.0 4.0 Rice-aquaculture 3.0 5.0 4.0 4.0 Fruit tree 3.0 3.0 5.0 4.0 Rice-fish capture 5.0 5.0 2.0 3.5 Rice-vegetable 4.0 3.0 3.0 3.3 Pangasius fish 1.0 1.0 5.0 3.0 Specialized 2.0 2.0 4.0 3.0 vegetable Duck 3.0 2.0 3.0 2.8 2-crop rice 4.0 4.0 1.0 2.5 3-crop rice 4.0 4.0 1.0 2.5 Source: Authors’ calculation from survey data. Note: For cost, 1 = very high; 2 = high; 3 = medium; 4 = low; 5 = very low. For profit and economic index, 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Economic index (column 5, with weight = 2) = average of investment cost (column 2), operating cost (column 3), and net profit (column 4), with weight = 2. The main reason for the variability in profitability among the livelihood models comes from their difference in risk profiles. Converting from a lower-value to a higher-value livelihood model also implies higher risks for farmers. Table III-5 shows that farmers adopting rice models are less exposed to market and other risks than high-value commodity models. Compared with those of other commodities, rice value chains are well established with a large number of rice collectors and traders who can go directly to rice fields to collect wet paddy. Moreover, because the international and domestic markets for rice have been well defined for the past 20 years, they help keep the rice price more stable than that for other products. In contrast, some potential livelihood models such as rice-aquaculture (with giant freshwater prawns as the main product) or rice-lotus (with lotus flowers and lotus seeds as the main products) face considerable challenges in finding a stable market for their products. Lack of market access and unstable prices are also critical problems of other commodities such as fruits, vegetables, ducks, and Pangasius fish. Farmers cannot supply directly international and big city markets such as HCMC because they do not possess their own pre/processing and storage facilities. The ranking shows that the Pangasius fish model faces the most prominent risks compared with other products. The international market for Pangasius collapsed several times during 2010-19, arising from changing trade policies in importing countries and the poor production coordination among Vietnamese Pangasius producers. 42 Sudden price changes, products stuck at border, growing competition, contract breach, and so on. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 44 Alternative Scenarios and Policy Implications TABLE III-5. Ranking of livelihood models for risk-coping ability, Upper subregion Number of years in 2010-2019 (10 Risk coping ability years) that a farming household enjoys Model No serious No serious No serious Market Disease Disaster Index of market disease disaster access risk coping shock (weight =2) ability 2-crop rice 7.5 7.0 7.5 4.0 4.0 4.0 4.0 3-crop rice 7.5 7.0 7.5 4.0 4.0 4.0 4.0 Rice-aquaculture 6.7 8.4 9.0 3.0 5.0 5.0 4.0 Rice-lotus 5.7 9.0 9.0 2.0 5.0 5.0 3.5 Duck 6.5 6.5 7.0 3.0 3.0 4.0 3.3 Rice-capture fish 5.0 8.0 5.0 2.0 5.0 2.0 2.8 Rice-vegetable 5.8 6.5 7.5 2.0 3.0 4.0 2.8 Fruit tree 5.5 6.5 7.5 2.0 3.0 4.0 2.8 Specialized 5.5 6.0 7.5 2.0 3.0 4.0 2.8 vegetable Pangasius fish 0.0 6.0 8.0 1.0 3.0 5.0 2.5 Source: Authors’ calculation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Risk-coping ability index (column 5) = average of market access with weight = 2 (column 5), disease (column 6), and disaster (column 7). 1.3.3. Employment generation Farming systems contribute to job creation in rural areas at different levels. The Pangasius farming model requires the most labor for production: 543 person-days/ha/year. The specialized vegetable model and the fruit tree model also create considerable employment: 300 person-days/ ha/year. In contrast, most rice-farming models require less labor, only 20 person-days/ ha/crop, because rice cultivation in MKD has become highly mechanized. This partly explains the high rate of underemployment in the delta and the high proportion of young people from rice-farming households migrating to cities or finding work in the industrial parks. In areas of vegetable farming, labor shortages are a recurring problem. The ranking shows that Pangasius and specialized vegetable production had the highest level of social impact as they created the highest employment demand and also contributed significantly to the income of households that employed these systems (Table III-6). In contrast, rice-fish capture, duck-raising, and 2-crop rice systems had the lowest positive social impact as they often do not create much employment. Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 45 TABLE III-6. Ranking of livelihood models for employment impact, Upper subregion Employment Contribution to total Models Social index creation income Pangasius fish 5.0 5.0 5.0 Specialized vegetable 5.0 5.0 5.0 Rice-lotus 4.0 4.0 4.0 Fruit tree 3.0 5.0 4.0 Rice-vegetable 4.0 4.0 4.0 Rice-aquaculture 4.0 3.0 3.5 3-crop rice 3.0 3.0 3.0 Duck 2.0 3.0 2.5 Rice-fish capture 3.0 2.0 2.5 2-crop rice 1.0 1.0 1.0 Source: Authors’ calculation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Social index (column 4) = average of employment creation (column 2) and contribution to total income (column 3). 1.3.4. Environmental aspects Three groups of indicators, natural resource use (NRU), on-site environmental impacts (EIs), and interregional impacts (IRIs), are used to assess the impacts of different livelihood models on the environment (Table III-7). Concerning the NRU, household surveys and local expert consultations found that the 3-crop rice had the highest land-use intensity. On average, the 3-crop rice model occupied about 11 months per year, and the total time between crops was just about one month. The Pangasius production model consumed the highest electricity for water pumping, paddle wheels for pond aeration, and processing. The 3-crop rice, vegetables, and Pangasius models were the biggest consumers in terms of water use. Notably, some vegetable and Pangasius farms started pumping groundwater for their production in the dry season, when river water was insufficient. The ranking shows that the duck-raising, rice-fish capture, rice-lotus, and rice-aquaculture models consumed the least natural resources because they are nature-based systems. In contrast, the vegetable, Pangasius, and 3-crop rice models were the most significant users of natural resources. The specialized vegetable model uses the most agrochemicals in production. This model uses about 1.5 tons of fertilizers and 73 L of pesticides per ha per year. The fruit model uses the most pesticides at 155 L/ha/year and about 200 kg of chemical fertilizers/ha/year. The three-crop rice model uses 1.3 tons of chemical fertilizers/ha/year and about 45 L of pesticides/ha/year. The 2-crop and upland rice models use fewer fertilizers and pesticides, about 65 percent of those in the 3-crop rice model. The flood-based models such as rice-lotus and rice-aquatic products use the least fertilizers and pesticides, only 30 percent of the quantities used in the specialized 3-crop rice models. The Pangasius model contributes to environmental pollution by discharging wastewater and creating pond muds. The ranking shows that the 3-crop rice model ranks at the top as having the worst environmental impact, followed by the Pangasius production and specialized vegetable models. The free-range duck-raising, rice-lotus, rice-fish capture, and rice-aquaculture models, in general, are the most environmentally friendly. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 46 Alternative Scenarios and Policy Implications Livelihood models have different interregional environmental impacts. Some livelihood models, such as rice-lotus and rice-aquaculture, do not cause a significant impact on the development of other sectors or other regions. In contrast, the 3-crop rice and fruit tree production models have interregional spillover impacts. The high-dike system protecting these livelihood models reduced the water retention capacity of the Upper subregion, thus exacerbating the flooding and salinity intrusion problems in downstream provinces. The dominance of 3-crop rice with an alarming rate of chemical use degraded surface water quality and forced farmers to switch to groundwater, especially for aquaculture, which is the critical factor causing land subsidence of the delta. To sum up, the ranking of the environmental index shows that the flood-based models (rice-lotus, duck raising, rice-fish capture, and rice-aquaculture) have a low negative impact on the environment and natural resources. On the other end, the 3-crop rice model has severe negative impacts on the environment and the delta’s sustainable development. Other models, such as specialized vegetables, fruit trees, and Pangasius fish, fall in between the flood-based models and the 3-crop rice model. TABLE III-7. Ranking of livelihood model suitability based on environmental impact, Upper subregion Rice- Rice- Special- Rice Pan- Rice- fish aqua- 2-crop Fruit ized 3-crop Model Duck vege- gasi- lotus cap- cul- rice tree vegeta- rice table us ture ture ble Surface water 4.0 5.0 5.0 3.0 3.0 3.0 2.0 4.0 2.0 2.0 Groundwater 5.0 5.0 5.0 5.0 4.0 5.0 3.0 4.0 3.0 5.0 NRU Soil use intensity 5.0 5.0 5.0 5.0 4.0 3.0 5.0 3.0 3.0 1.0 Energy (electricity) 5.0 5.0 5.0 4.0 4.0 5.0 1.0 4.0 3.0 4.0 NRU index 4.8 5.0 5.0 4.3 3.8 4.0 2.8 3.8 2.8 3.0 Water pollution 5.0 5.0 5.0 5.0 3.0 2.0 1.0 3.0 3.0 1.0 On-site environmental (weight = 2) Soil pollution 5.0 5.0 5.0 5.0 3.0 3.0 2.0 3.0 1.0 1.0 impact Air pollution 5.0 5.0 5.0 5.0 3.0 3.0 3.0 1.0 1.0 2.0 Greenhouse gas 4.0 5.0 4.0 4.0 5.0 2.0 4.0 5.0 4.0 1.0 (GHG) emissions Landscape change 5.0 5.0 5.0 4.0 5.0 5.0 1.0 3.0 4.0 3.0 EI Index 4.8 5.0 4.8 4.7 3.7 2.8 2.0 3.0 2.7 1.5 Foundation 5.0 5.0 5.0 5.0 5.0 4.0 5.0 3.0 4.0 1.0 subsidence Salinity intrusion 5.0 5.0 5.0 5.0 4.0 4.0 5.0 2.0 4.0 2.0 Biodiversity loss 5.0 3.0 3.0 5.0 4.0 3.0 5.0 4.0 3.0 2.0 IRI Flooding in urban 5.0 5.0 5.0 5.0 5.0 5.0 5.0 2.0 4.0 2.0 area IRI index 5.0 4.5 4.5 5.0 4.5 4.0 5.0 2.8 3.8 1.8 Environmental index 4.9 4.8 4.8 4.6 4.0 3.6 3.3 3.2 3.1 2.1 Source: Authors’ calculation from survey data. Note: 1 = very seriously impacted; 2 = seriously impacted; 3 = impacted; 4 = lightly impacted; 5 = negligible impact. Environmental index (last row) = average of NRU index, EI index, and IRI index. Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 47 1.3.5. Gender In general, the workload of women is similar among different livelihood models. Only the rice-vegetable and specialized vegetable livelihood models required more work from female workers. The household survey found that rice production was the most common livelihood model in this subregion. This model required the lowest person-days (82.9 person-days for 3-crop rice and 53.6 person-days for 2-crop rice). Most of the time, women did not participate in any stage of the primary production of rice. Only in households with especially small farms and employing family labor are women actively participating in rice production. Another case is when they worked as hired labor in other people’s fields. Even in this case, women were generally in charge of weeding and transplanting and so on. Men mostly performed the activities that required technical and mechanical skills (such as driving tractors and combine harvesters) or were hazardous (that is, spraying pesticides and herbicides). Overall, female person-days accounted for about 25 percent of the total person-days in the 3-crop rice and 34 percent in the 2-crop rice models. In the 3-crop rice system, as more pesticide application was required, the use of male labor was greater. Unlike in the rice models, women play a crucial role in the vegetable production model (Table III-8). Women participated in planting, weeding, watering, applying fertilizers, and harvesting. The survey showed that, on average, women contributed 62.5 percent of the 346.2 required person- days in the vegetable production model and 46.7 percent of the 188.1 person-days in the rice- vegetable model. Apart from harvesting, some women were also responsible for selling their produce in the market. Women were least involved in the Pangasius production model for some reasons. First, Pangasius production requires higher production skills such as controlling and monitoring water pumping systems. Second, they may not have the strength to perform the heavy work (for example, carrying the Pangasius feed bags). Third, because of the high level of risk in Pangasius production and the high value of Pangasius, a superstition exists in Vietnam that women can bring bad luck to the production. Therefore, women tend to avoid working on Pangasius farms. TABLE III-8. Ranking of livelihood models for gender aspects, Upper subregion Negative impacts on Livelihood models Power of women Gender index women (workload) Rice-lotus 4.0 5.0 4.5 Rice-aquaculture 4.0 5.0 4.5 Duck 4.0 5.0 4.5 Fruit tree 4.0 5.0 4.5 Rice-fish capture 4.0 5.0 4.5 Pangasius fish 4.0 5.0 4.5 2-crop rice 3.0 5.0 4.0 Rice-vegetable 5.0 3.0 4.0 3-crop rice 4.0 4.0 4.0 Specialized vegetable 5.0 2.0 3.5 Source: Authors’ calculation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Gender index (column 4) = power of women43 (column 2) and negative impacts on women (column 3). 43 Power of women means the voice of women and how they influence the production decisions at the household level. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 48 Alternative Scenarios and Policy Implications 1.3.6. Conclusion The most suitable and sustainable livelihood models are identified based on the suitability and sustainability index, which is computed from all seven types of indexes: future natural condition suitability, climate change resilience, risk-coping ability, economic, social, environmental, and gender aspects. The ranking shows that the rice-lotus, rice-aquaculture, and duck-raising models are the most suitable and sustainable systems for the Upper subregion. The fruit trees, rice-fish capture, rice-vegetables, Pangasius, and specialized vegetable models are ranked as suitable systems. The 3-crop rice and 2-crop rice systems are classified as the least fit and sustainable agricultural livelihood models in the new context. TABLE III-9. MCA ranking of livelihood models for suitability and sustainability, Upper subregion Suitabil- Climate Economic Risk Social Environ- Gender Final ity with change index coping index ment index suitability Livelihood natural resil- (weight=2) ability index and sustain- models condition ience index ability index (weight index =2) Rice-lotus 4.0 4.7 4.0 3.5 4.0 4.9 4.5 4.2 Rice-aquaculture 2.5m 4.3 4.0 4.0 3.5 4.6 4.5 3.8 Duck 4.0 3.7 2.8 3.3 2.5 4.8 4.5 3.6 Rice-fish capture 2.5 4.0 3.5 2.8 2.5 4.8 4.5 3.4 Fruit tree 2.5 1.7 4.0 2.8 4.0 3.2 4.5 3.2 Rice-vegetable 2.0 3.0 3.3 2.8 4.0 4.0 4.0 3.1 Pangasius fish 2.5 2.0 3.0 2.5 5.0 3.3 4.0 3.1 Specialized 2.5 2.0 3.0 2.8 5.0 3.1 3.5 3.0 vegetable 2-crop rice 2.5 3.3 2.5 4.0 1.0 3.6 4.5 2.9 3-crop rice 2.0 2.0 2.5 4.0 3.0 2.1 4.0 2.7 Source: Authors’ calculation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Final suitability and sustainability index (column 9) = average of all columns from 2 to 8 (considering the weights of 2 of columns 2 and 4). This ranking indicates some main directions for transforming key and potential livelihood models in the Upper subregion (Table III-10). Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 49 TABLE III-10. Potential development directions of livelihood models, Upper subregion Livelihood models Orientation Target models Phasing out and transforming to High-quality rice-vegetables 3-crop rice other models Specialized vegetables Improve, scale up. Hygienic, biologically safe Fruit Maintain High-tech, sustainable Specialized vegetable Adopting food safety and scaling up Hygienic and biologically safe Pangasius fish Maintaining High-tech and sustainable Adopting food safety and scaling up Hygienic and biologically safe Rice-vegetable with consideration Rice-fish capture Maintaining Sustainable and high-quality rice 2-crop rice Transforming to other models 2-crop high-quality rice Improve, scale up Large-scale production zones, Rice-aquaculture complete value chain establishment Promote, scale up Large-scale production zones, Rice-flood-based crops complete value chain establishment Adopting food safety and scaling up Hygienic and biologically safe Fruit trees with consideration Duck raising Adopting biosecurity and scaling up Hygienic and biologically safe Developing value chains and scaling Large-scale production zones and Rice-aquaculture up complete value chain establishment Promoting and scaling up Large-scale production zones and Rice-flood-based crops complete value chain establishment Source: Authors’ evaluation from survey data. 2. Middle subregion 2.1. General situation Rice is an important crop in the Middle subregion. Following the government’s project diverting freshwater to the Nam Mang Thit area since the early 1990s, a shift occurred from the 1-crop rice system to 2-crop and 3-crop systems. Subsequently, during 2005–2017, a significant part of the paddy area was converted to aquaculture, horticulture, or high-quality rice varieties. Generally speaking, the rice area in this subregion has remained relatively stable over time, although some changes occurred between the 3-crop rice and 2-crop rice systems near the Hau River of Can Tho and Hau Giang Provinces. The 1-crop rice model still exists but occupies only a small proportion of the area. The Middle subregion is Vietnam’s leading area for fruit production. The fruit tree production area has been stable at around 250,000 ha over time. The main area is located between the Tien River and Hau River, such as Chau Thanh and Lai Vung Districts of Dong Thap Province, Cho Lach and Chau Thanh Districts of Ben Tre Province, Tam Binh and Long Ho Districts of Vinh Long Province, and some other river districts of Hau Giang Province. Cho Lach District of Ben Tre Province is the leading producer of seedlings for the entire MKD. This subregion also has the largest total area of Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 50 Alternative Scenarios and Policy Implications coconut trees, amounting to 119,000 ha in 2017. This represents 80 percent of the area under coconut nationally. This subregion has more specialized horticultural zones than the other subregions, but mixed fruit orchards still prevail. Some fruit tree orchards have also diversified, adding ornamental plants under the canopy of fruit trees for earning additional income. In the Middle subregion, the Pangasius farming area in Can Tho and Hau Giang Provinces decreased during 2013–2016, due to market instability and concerns about environmental impacts, but has recovered in recent years. Leading districts for Pangasius farming in this subregion are Chau Thanh District of Dong Thap Province, Tam Binh and Long Ho Districts of Vinh Long Province, and Phu Tan District of An Giang Province. Total production of these districts reached 203,600 tons, equivalent to 16 percent of the MKD’s total production in 2017. Besides Pangasius, many freshwater fish species are also farmed, such as bronze featherback, blue biota, marble goby, giant river prawn, tilapia, several species of eel, and others. To improve production efficiency and accommodate better flood adaptation, in some areas outside of flood-control dikes, farmers have combined their rice systems with giant freshwater prawn or fish farming or switched to giant freshwater prawn farming. In Kien Giang Province, farmers practice marine fish cage culture (that is, groupers, cobia, and recently Trachinotus blochii). The Middle subregion is the leading area in the MKD for raising animals (buffalo: 12,510 heads, poultry: 56.03 million heads, cattle: 723,710 heads, and pigs: 2.6 million heads). Between 2010 and 2018, particularly strong growth was experienced in raising beef cattle and producing poultry. For beef cattle raising, Ba Tri District (Ben Tre Province), Duc Hoa District (Long An Province), and Cau Ngang District (Tra Vinh Province) have been most prominent. Pig production has been most significant in Cai Be and Cho Gao Districts (Tien Giang Province), followed by Mo Cay Bac and Mo Cay Nam Districts (Ben Tre Province). 2.2. Key agricultural livelihood models As with the Upper subregion, common livelihood models were identified for the Middle subregion using the same criteria (Figure III-5 and Table III-11). The 3-crop rice, specialized fruit production, coconut production, and 2-crop rice models are the most common, accounting for 75 percent of the subregion’s total agricultural area and 51 percent of the subregion’s agricultural production value. Pangasius fish and vegetables occupy only 5 percent of the subregion’s agricultural land yet together account for 28 percent of the total production value. Rice, coconut, and fruit production are the main livelihoods of more than 1.7 million households. Vegetables, chicken, rice- vegetable rotation, beef cattle, and industrial crops are the livelihood models of another 470,000 households. Similar to what occurs in the Upper subregion, Pangasius is cultured by enterprises and by only 900 households with high financial capacity. Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 51 FIGURE III-5. Map of key livelihood models in the Middle subregion Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ illustration from Dang et al. (2019). TABLE III-11. Coverage of the most common agricultural livelihood models, Middle subregion Area % of % of ag- Number of Produc- % of ag- % of ag- (ha, agricul- ricultural house- tion value ricultural ricultural thou- tural land land in holds (VND, produc- produc- Livelihood models sands) in Middle MKD (thou- billions) tion value tion value subregion sands) of Middle of MKD subregion 3-crop rice 587.2 42 18 469.0 62,188 25.8 8.9 2-crop rice 229.4 16 7 274.0 17,189 7.1 2.5 Rice-vegetable 12.6 1 0 46.0 3,826 1.6 0.5 Specialized vegetable 71.2 5 2 105.0 32,819 13.6 4.7 Fruit 305.5 22 9 Mango 41.8 3 1 100.6 12,540 5.2 1.8 Durian 7.0 1 0 8.8 2,730 1.1 0.4 Dragon fruit 12.1 1 0 42.5 12,100 5.0 1.7 Citrus 73.0 5 2 168.2 7,300 3.0 1.0 Coconut 130.0 9 4 459.1 1,300 0.5 0.2 Other fruits 41.6 3 1 217.7 17,966 7.5 2.6 Industrial crops 27.5 2 1 36.0 5,124 1.2 0.4 Pangasius fish 3.4 0 0 0.5 22,440 9.3 3.2 Rice-shrimp 35.2 3 1 8.0 4,612 2.7 0.9 Beef cattle farming (*) n.a. n.a. n.a. 57.6 7,397 3.0 1.0 Pig farming (*) n.a. n.a. n.a. 41.2 23,900 9.0 3.2 Chicken farming (*) n.a. n.a. n.a. 183.1 7,265 3.0 1.0 Source: Authors’ calculation from AgroCensus 2016 and Dang et al. (2019). Note: (*) Because livestock farming by farmers in the MKD is mainly at a backyard scale and its land use is attached to cultivation or aquaculture, it is impossible to estimate the land area of this model. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 52 Alternative Scenarios and Policy Implications 2.3. Assessment of key and potential livelihood models 2.3.1. Natural condition suitability and climate change resilience The assessment shows that rice-giant freshwater shrimp, 2-crop rice, and 3-crop rice systems are the most suitable livelihood models under current natural conditions in the Middle subregion. Some 94 percent of the agricultural land in this subregion is either highly suitable or suitable for the rice-giant freshwater shrimp model. About 80 percent of the subregion’s agricultural land is either highly suitable or suitable for the 3-crop rice and 2-crop rice systems. Only about 13 percent of the current rice area is not biophysically suitable. It is concentrated in locations where the dike systems do not ensure effective flood prevention for the summer–autumn rice crop and autumn–winter rice crop. The area suitable for specialized vegetable production is just below 65 percent of the subregion’s agricultural land. About 60 percent of the regional agricultural land is considered highly suitable or suitable for fruit orchards (that is, durian, mango, dragon fruit, citrus, coconut, or other fruit). However, the natural conditions will quickly change soon (and certainly by 2030), resulting in less favorable conditions for most crops in the Middle subregion (Table III-12). The area deemed to be unsuitable for rice-giant freshwater shrimp, 2-crop rice, and 3-crop rice systems will increase significantly, with the 3-crop rice model becoming especially vulnerable to adverse natural conditions. Because of the dual impacts of sea level rise and changes in upstream river flows, some 366,000 ha of land now suitable for the 3-crop model will no longer be so. Livestock production, rice- aquaculture, and industrial crops are considered likely to be the most suitable models based upon the projections for the natural conditions in 2030, although the fruit tree, rice-duck, 2-crop rice, rice- vegetables, Pangasius, and vegetable models will also still have a potential for future development. TABLE III-12. Ranking of livelihood models for suitability to future conditions, Middle subregion Area of most Proportion of Change of Index of Index of Index of suit- suitable and most suitable most suitable column column ability with Livelihood models suitable land and suitable and suitable A B future natural in 2030 (ha, land in 2030 land 2019– condition thousands) (%) 2030 (%) (B) Beef cattle farming 1,775.0 100.0 0.0 5.0 3.0 4.0 Chicken farming 1,775.0 100.0 0.0 5.0 3.0 4.0 Pig farming 1,775.0 100.0 0.0 5.0 3.0 4.0 Rice-aquaculture 895.5 50.5 1.0 3.0 4.0 3.5 Industrial crops 1,094.0 61.6 −6.8 4.0 2.0 3.0 Coconut-aquaculture 928.0 52.3 −7.5 3.0 2.0 2.5 Fruit tree 969.0 54.6 −9.7 3.0 2.0 2.5 Rice-duck 1,151.0 64.8 −22.7 4.0 1.0 2.5 2-crop rice 1,151.0 64.8 −22.7 4.0 1.0 2.5 Rice-vegetable 1,083.0 61.0 −23.4 4.0 1.0 2.5 Pangasius fish 750.0 42.3 −4.6 3.0 2.0 2.5 Vegetable 1,055.0 59.4 −7.3 3.0 2.0 2.5 3-crop rice 1,062.0 59.8 −24.9 3.0 1.0 2.0 Source: Authors’ calculation from survey data and Dang et al. (2019). Note: The index of column A is computed as 80-100%: 5; 60-79%: 4; 40-59%: 3; 20-39%: 2; and 0-19%: 1. The index of column B is computed as 20-100%: 5; 1-20%: 4; 0%: 3; (−1)–(−20%): 2; (−20)– −100%): 1. The index of future natural condition suitability is the average of the indexes of columns A and B. Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 53 Concerning climate change resilience, the small-scale cattle-raising model (backyard) is the most resilient to climate change, followed by the rice-aquaculture, coconut-aquaculture, and 2-crop rice models (Table III-13). The second group, which may be resilient to climate change, includes the fruit tree, rice-duck, rice-vegetable, and chicken/pig production model. Farmers in the Middle subregion have practiced these models for a long time (over 50 years). However, the 3-crop rice, rice-vegetable, specialized vegetable, and industrial crop models are less resilient because it is difficult to change their production calendars. The transition from these models to flood-based livelihood systems is also not easy because of the existing interregional dike system (especially the enclosed dikes). High-investment models such as Pangasius fish farming can remain owing to the in- built infrastructure, which can help protect them from flooding and salinity intrusion. However, the problem with these models is that it would be difficult for farmers to convert these systems to other models because their infrastructure investments have already been made. That is a sunk cost which cannot be readily reversed. TABLE III-13. Ranking of livelihood models for climate change resilience, Middle subregion Livelihood Absorptive Transformative Climate change Flexibility models ability ability resilience index Beef cattle farming 5.0 5.0 5.0 5.0 Rice-aquaculture 5.0 5.0 3.0 4.3 Coconut-aquaculture 5.0 5.0 3.0 4.3 2-crop rice 2.0 3.0 5.0 3.3 Fruit tree 3.0 3.0 3.0 3.0 Rice-duck 4.0 2.0 3.0 3.0 Rice-vegetable 3.0 3.0 3.0 3.0 Chicken farming 3.0 3.0 2.0 2.7 Pig farming 3.0 3.0 2.0 2.7 Industrial crops 2.0 3.0 2.0 2.3 3-crop rice 2.0 1.0 4.0 2.3 Pangasius fish 4.0 1.0 1.0 2.0 Vegetable 2.0 2.0 2.0 2.0 Source: Authors’ evaluation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Climate change resilience index (column 5) = average of absorptive ability (column 2), flexibility (column 3), and transformative ability (column 4). 2.3.2. Economic and financial aspects All crop models require public investment in dikes, canals, and pump stations, whereas livestock and aquaculture models require household-level investment in infrastructure. Among the crop models, rice households only need to invest VND 3-5 million per ha in their in-field infrastructure (Figure III-6). For vegetables, coconut, and fruit trees, households have to invest in in- field and garden/orchard infrastructure. Livestock models also require significant investments from farmers for animal cage construction. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 54 Alternative Scenarios and Policy Implications FIGURE III-6. Investment in the infrastructure of key models, Middle subregion44 Chicken (1000 head) 15 Pig (100 head) 15 Beef cow (1 head) 10 Pangasius 500 Industrial crops 20 Fruit tree 500 Coconut 50 Vegetable 75 Rice-Vegetable 5 2-crop rice 3 3-crop rice 5 Unit: VND, millions per ha Source: Authors’ calculation from survey data. The variable production cost of different livelihood models also differs significantly (Figure III-7). As in the Upper subregion, Pangasius production in the Middle subregion has the highest operating cost, at about VND 6.4 billion/ha/year (with 80 percent of feed and veterinary drugs sometimes purchased on credit). So, only households with very strong financial capacity can afford this model. Vegetable farming also requires significant operating costs of about VND 330 million/ha/ year; rice-vegetable and fruit tree systems have operating costs ranging from VND 80 million/ha/year to VND 115 million/ha/year. These three models also use more agrochemicals, fertilizer, and hired laborers (particularly at harvesting time). In the rice models, the operating costs tend to be much lower. Farmers tend to buy fertilizer and agrochemicals on credit and generally use household labor. Therefore, the operational cost for rice cultivation is generally VND 40-55 million/ha/year (50 percent of which can be paid using credit). High production operating cost is one of the main barriers for low-income households to replace their rice production with other higher-value production models. Among livestock models, beef cattle raising has a lower production cost than the chicken and pig production models. FIGURE III-7. Total production cost of key livelihood models, Middle subregion 0 50 100 150 200 250 300 350 400 3-crop rice 2-crop rice Rice-Vegetable Vegetable Coconut Fruit tree Industrial crops Beef cow (1 head) Buying on credit Pig (100 head) Immediate payment Chicken (1000 head) Unit: VND, millions per ha per year Source: Authors’ calculation from survey data (2019). 44 Investment in infrastructure for Pangasius fish (that is, pond construction, water supply and treatment systems, and so on) and fruit trees (that is, land preparation, water supply, and drainage systems) will support operations over many years. Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 55 All of the livelihood models other than rice mono-cropping generate higher profits, yet greater year-to-year volatility in returns. The profits of the 2-crop rice and 3-crop rice models are very low, approximately only VND 44–51 million/ha/year. Among the other models, the Pangasius production model has the highest profit of about VND 515 million/year, 10 times higher than rice (Figure III-8). The vegetable and fruit models are profitable too, although the rice models’ profit has been more stable than the vegetable and Pangasius models in the last decade. FIGURE III-8. Average profit of key models, Middle subregion 800.0 Unit: Million VND/ha 2-season rice 3-season rice 700.0 Rice-vegetable 600.0 Vegetable 514.2 Coconut 500.0 Fruit tree 400.0 350.3 Industrial crops Pangasius 300.0 Beef cow (1 head) 200.0 Chicken production (1000 head) 142.6 106.9 108.1 Rice-duck/aquaculture 86.6 72.9 100.0 64.1 38.8 43.6 25.8 Coconut + Shrimp 3.2 0.6 Pig production (1 head) 0.0 Source: Authors’ calculation from survey data. The economic performance ranking indicates that the rice-aquaculture, beef cattle raising, and fruit production models appear to have the best profiles in terms of economic outcomes (Table III-14). In contrast, the chicken production, pig production, 3-crop rice, and 2-crop rice models provide farmers with the lowest positive economic impacts. All other remaining models have a relatively good economic performance. TABLE III-14. Ranking of livelihood models for economic performance, Middle subregion Livelihood Investment Operating Net profit Economic index models cost cost (weight=2) (weight=2) Rice-aquaculture 3.0 5.0 4.0 4.0 Beef cattle farming 4.0 4.0 4.0 4.0 Fruit tree 3.0 3.0 5.0 4.0 Coconut-aquaculture 3.0 4.0 3.0 3.3 Industrial crops 5.0 4.0 2.0 3.3 Rice-vegetable 4.0 3.0 3.0 3.3 Rice-duck 3.0 4.0 3.0 3.3 Pangasius fish 1.0 1.0 5.0 3.0 Vegetable 2.0 2.0 4.0 3.0 Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 56 Alternative Scenarios and Policy Implications TABLE III-14. (Cont.) Livelihood Investment Operating Net profit Economic index models cost cost (weight=2) (weight=2) Chicken farming 3.0 2.0 3.0 2.8 Pig farming 3.0 2.0 3.0 2.8 3-crop rice 5.0 4.0 1.0 2.8 2-crop rice 4.0 4.0 1.0 2.5 Source: Authors’ evaluation from survey data. Note: For cost, 1 = very high; 2 = high; 3 = medium; 4 = low; 5 = very low. For profit and economic index, 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Economic index (column 5, with weight =2) = average of investment cost (column 2), operating cost (column 3), and net profit (column 4), with weight = 2. Risks of markets, diseases, and disasters are crucial factors influencing the profit stability of all livelihood models. Generally speaking, high profitability is associated with higher risks. Rice farmers face few market risks, but the rice price remains low. Meanwhile, farmers in the Pangasius, fruit tree, and vegetable models face higher market risks. The 3-crop rice model has an increased risk of pests and diseases because rice is cultivated year-round. The ranking shows that beef cattle raising, 2-crop rice, 3-crop rice, and coconut-aquaculture stand as the safest agricultural livelihood models in the subregion. In contrast, rice-vegetables, specialized vegetables, Pangasius, and pig raising appear the riskiest livelihood models. The main risks come from market instability and disease outbreaks. TABLE III-15. Ranking of livelihood models for risk-coping ability, Middle subregion Number of years in 2010–2019 (10 Risk coping ability years) that a farming household enjoys Livelihood No serious No serious No serious Market Disease Disaster Risk coping models market disease disaster access ability shock (weight =2) index Beef cattle farming 8.0 9.0 10.0 5.0 5.0 5.0 5.0 Coconut- 8.0 7.5 8.0 5.0 4.0 5.0 4.8 aquaculture 3-crop rice 7.5 7.0 8.0 4.0 4.0 5.0 4.3 2-crop rice 7.5 7.0 7.5 4.0 4.0 4.0 4.0 Rice-aquaculture 6.7 8.4 9.0 3.0 5.0 5.0 4.0 Chicken farming 7.5 6.5 9.0 4.0 3.0 5.0 4.0 Fruit tree 7.0 6.5 7.5 4.0 3.0 4.0 3.8 Rice-duck 6.5 6.5 7.0 3.0 3.0 4.0 3.3 Rice-vegetable 5.8 6.5 7.5 2.0 3.0 4.0 2.8 Specialized 5.5 6.0 7.5 2.0 3.0 4.0 2.8 vegetable Pangasius fish 0.0 6.0 8.0 1.0 3.0 5.0 2.5 Industrial crops 4.5 7.5 6.5 1.0 4.0 3.0 2.3 Pig farming 4.5 4.5 10.0 1.0 1.0 5.0 2.0 Source: Authors’ evaluation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Risk-coping ability index (column 5) = average of market access with weight = 2 (column 5), disease (column 6), and disaster (column 7). Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 57 2.3.3. Employment generation The level of work effort required varies across production systems. The Pangasius fish model uses the most person-days, up to 598 person-days/ha/year (Table III-16). Next are the vegetable, fruit, and coconut-aquaculture models with 380, 334, and 326 person-days, respectively. By contrast, for rice-based systems, work is seasonal, especially for the 2-crop rice, rice-ducks, and rice-aquaculture. This seasonality of labor demand and low wages are the main reasons why many young people from rice households migrate to the cities seeking additional and higher incomes. It is difficult for rice households to shift to other livelihoods because high investments for infrastructure and operations are needed. In addition, the vegetable, fruit tree, and aquaculture models are also increasingly mechanized to reduce the high dependence on labor. TABLE III-16. Ranking of livelihood models for social impact, Middle subregion Livelihood models Employment creation Contribution to total income Social index Pangasius fish 5.0 5.0 5.0 Vegetable 5.0 5.0 5.0 Fruit tree 4.0 5.0 4.5 Chicken farming 4.0 4.0 4.0 Pig farming 4.0 4.0 4.0 Rice-vegetable 4.0 4.0 4.0 Rice-aquaculture 4.0 3.0 3.5 Beef cattle farming 3.0 3.0 3.0 Coconut-aquaculture 3.0 3.0 3.0 3-crop rice 3.0 3.0 3.0 Rice-duck 2.0 3.0 2.5 Industrial crops 1.0 2.0 1.5 2-crop rice 1.0 1.0 1.0 Source: Authors’ evaluation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Social index (column 4) = average of employment creation (column 2) and contribution to total income (column 3). 2.3.4. Environmental impacts This ranking shows that 3-crop rice, vegetables, Pangasius, and fruit trees have the highest impacts on the environment (Table III-17). The specialized vegetable model uses the most chemicals in production with a total fertilizer use of about 4 tons/ha/year and 73 L of agrochemicals, followed by the fruit model, with about 200 kg of fertilizer/ha/year and 155 L/ha/year of chemicals. Rice- based systems use fewer chemicals, and among them, the 2-crop rice uses the least agrochemicals. Livelihood models in the Middle subregion also have an interregional spillover environmental impact. Several households that produce industrial crops and trees began using groundwater in the dry season, contributing to the delta’s land subsidence. Large-scale livestock development (with poor manure management) and chemical overuse in the 3-crop rice, fruit tree, and vegetable systems (causing water pollution) have also damaged the ecosystem. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 58 Alternative Scenarios and Policy Implications TABLE III-17. Ranking of livelihood model suitability based on environmental impact, Middle subregion Rice-vegetable Pangasius fish aquaculture aquaculture Pig farming 2-crop rice 3-crop rice Beef cattle Vegetable Rice-duck Industrial Coconut- Fruit tree Chicken farming crops Rice- Models Surface water 5.0 5.0 5.0 3.0 5.0 3.0 3.0 5.0 3.0 4.0 2.0 2.0 2.0 Groundwater 5.0 5.0 5.0 5.0 5.0 4.0 3.0 5.0 5.0 4.0 3.0 3.0 5.0 Soil use 5.0 5.0 5.0 5.0 5.0 4.0 4.0 5.0 3.0 3.0 5.0 3.0 1.0 NRU intensity Energy (elec- 5.0 5.0 5.0 4.0 3.0 4.0 4.0 3.0 5.0 4.0 1.0 3.0 4.0 tricity) NRU index 5.0 5.0 5.0 4.3 4.5 3.8 3.5 4.5 4.0 3.8 2.8 2.8 3.0 Water pollu- On-site environmental impact tion (weight 5.0 5.0 5.0 5.0 5.0 4.0 4.0 3.0 2.0 3.0 1.0 3.0 1.0 = 2) Soil pollution 5.0 5.0 5.0 5.0 5.0 4.0 4.0 2.0 3.0 3.0 2.0 1.0 2.0 Air pollution 5.0 4.0 5.0 5.0 4.0 4.0 4.0 2.0 3.0 1.0 3.0 1.0 2.0 GHG emis- 5.0 4.0 4.0 4.0 4.0 5.0 5.0 3.0 2.0 5.0 4.0 4.0 1.0 sions Landscape 5.0 5.0 5.0 4.0 4.0 5.0 5.0 4.0 5.0 3.0 1.0 4.0 3.0 change EI Index 5.0 4.7 4.8 4.7 4.5 4.3 4.3 2.8 2.8 3.0 2.0 2.7 1.7 Foundation 5.0 5.0 5.0 5.0 5.0 5.0 3.0 5.0 4.0 5.0 5.0 4.0 3.0 subsidence Salinity intru- 5.0 5.0 5.0 5.0 5.0 4.0 4.0 5.0 4.0 2.0 5.0 4.0 3.0 sion Biodiversity IRI 5.0 5.0 3.0 5.0 4.0 3.0 3.0 3.0 3.0 3.0 5.0 3.0 2.0 loss Flooding in 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 5.0 2.0 5.0 4.0 3.0 urban area IRI 5.0 5.0 4.5 5.0 4.8 4.3 3.8 4.5 4.0 3.0 5.0 3.8 2.8 Environmental index 5.0 4.9 4.8 4.6 4.6 4.1 3.9 3.9 3.6 3.3 3.3 3.1 2.5 Source: Authors’ evaluation from survey data and literature reviews. Note: 1 = very seriously impacted; 2 = seriously impacted; 3 = impacted; 4 = lightly impacted; 5 = negligible. Environmental index (last row) = average of NRU index, EI index, and IRI index. 2.3.5. Gender impacts There is no significant difference in the participation of women in rice production and vegetable production between the Middle subregion and the Upper subregion. The role of women in fruit tree production and industrial crop production in the Middle subregion is also small. They contribute only about 25 percent of the total 334 person-days required in the fruit tree production model. Women mainly take part in the least heavy and toxic work (that is, work like watering and applying pesticides is primarily done by men). Hired female labor is generally for planting, taking Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 59 care of the fruit (that is, putting on fruit covers for mango and guava, shaping the dragon fruit, and picking unwanted grapefruits/oranges), and harvesting. In coconut production, women participate more actively (that is, weeding, cutting dry coconut leaves, and so on) as gardens usually are close to their houses. Both husband and wife are involved in coconut harvesting (man climbs and cuts the coconuts, and the woman picks up and gathers them). In pineapple production, women do most of the harvesting. Livestock production activities attract more female work effort than crop production (Table III-18). As most livestock are raised in their yards, women can do both animal raising and housework. So, labor division tends to be such that the husband takes care of the fruit orchard, and the wife takes care of livestock production and housework. Since small-scale beef cattle raising is a good livelihood model adapting to climate change in the Middle subregion, especially in areas affected by salinity intrusion, women can participate very well in transitioning from rice to livestock farming. Although women do not directly participate in rice and fruit tree production activities, their role in the household’s decision-making is highly visible. They often decide where to borrow money, choose buyers, sell products, or look for new livelihoods. Women are also familiar and skillful in working with traders to negotiate prices. Therefore, they play a vital role in choosing buyers and handling trade transactions. TABLE III-18. Ranking of livelihood models for gender aspects, Middle subregion Negative impacts on Livelihood models Power of women Gender index women (workload) Pangasius fish 5.0 5.0 5.0 Rice-aquaculture 4.0 5.0 4.5 Coconut-aquaculture 4.0 5.0 4.5 Fruit tree 4.0 5.0 4.5 Chicken farming 5.0 4.0 4.5 Pig farming 5.0 4.0 4.5 2-crop rice 4.0 5.0 4.5 Beef cattle farming 3.0 5.0 4.0 Rice-duck 3.0 5.0 4.0 Industrial crops 3.0 5.0 4.0 Rice-vegetable 5.0 3.0 4.0 Specialized vegetable 5.0 2.0 3.5 3-crop rice 3.0 4.0 3.5 Source: Calculated using MCA by the research team based on household surveys (2019). Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Gender index (column 4) = power of women (column 2) and negative impacts on women (column 3). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 60 Alternative Scenarios and Policy Implications 2.3.6. Conclusion The 3-crop rice is the least suitable and sustainable model in the Middle subregion (Table III-19). The 3-crop rice model generates very low profit but results in many negative impacts on the environment because of the high use of agrochemicals. This model also does not create many jobs, pushing rice farmers to migrate out of the subregion. Sustaining a large area of triple rice in the Middle subregion continues to come at high environmental costs due to the high-level use of inputs. In addition, an extensive high-dike system to protect year-round rice production from flooding and saline intrusion will continue blocking the flow of rivers and canals and cause more severe water pollution. Therefore, the 3-crop rice model in the Middle subregion needs to be transformed to be more sustainable. In contrast, the rice-aquaculture and small-scale beef cattle-raising models are more suitable and sustainable for the future in the Middle subregion (Table III-19). They are both resilient to climate change and provide good economic returns with a low environmental footprint. Since disease outbreaks are still a significant challenge in the livestock sector, cattle production should be maintained at a small-scale, low-stock density. Also, there should be greater focus on improving biosecurity and reducing GHG emissions as well as proper waste management and recycling. TABLE III-19. MCA ranking of models by suitability and sustainability index, Middle subregion Suitabil- Climate Economic Risk Social Environ- Gender Final ity with change index coping index ment index suitabil- Livelihood natural resil- (weight=2) ability index ity and models condition ience index sustain- (weight index ability =2) index Beef cattle farming 4.0 5.0 4.0 4.5 3.0 4.9 4.0 4.2 Rice-aquaculture 3.5 4.3 4.0 4.5 3.5 4.6 4.5 4.1 Fruit tree 2.5 3.0 4.0 5.0 4.5 3.3 4.5 3.7 Coconut- 2.5 4.3 3.3 3.8 3.0 5.0 4.5 3.6 aquaculture Chicken farming 4.0 2.7 2.8 3.3 4.0 4.6 4.5 3.6 Pig farming 4.0 2.7 2.8 2.8 4.0 3.9 4.5 3.5 Rice-vegetable 2.5 3.0 3.3 3.5 4.0 4.1 4.0 3.3 Pangasius fish 2.5 2.0 3.0 2.5 5.0 3.3 5.0 3.2 Rice-duck 2.5 3.0 3.0 2.3 2.5 4.8 4.0 3.1 Industrial crops 3.0 2.3 3.3 3.0 1.5 3.9 4.0 3.0 Vegetable 2.5 2.0 3.0 2.3 5.0 3.0 3.5 3.0 2-crop rice 2.5 3.3 2.5 4.0 1.0 3.7 4.0 2.9 3-crop rice 2.0 2.3 2.8 4.0 3.0 2.5 3.5 2.8 Source: Authors’ estimation. Note: 1= very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Final suitability and sustainability index (column 9) = average of all columns from 2 to 8 (considering the weights of 2 of columns 2 and 4). Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 61 This ranking indicates some main directions for transforming key and potential livelihood models in the Middle subregion (Table III-20). TABLE III-20. Tentative development directions of livelihood models, Middle subregion Livelihood models Development directions Target models Reducing and gradually shifting to other models High-quality rice-vegetables, (about 165,000 ha near the Cai Lon–Cai Be Rivers 3-crop rice specialized vegetables, fruits, and and northwest of Tien Giang Province and in beef cattle districts near the Eastern Coastal Zone) Fruits, high-quality 2-crop rice, 2-crop rice Transforming to other models rice-fishery, and rice-flood-based crops (Dong Thap) Cleanliness, safety, biosecurity, Rice-vegetable Upgrading and expanding processing, and storage Cleanliness, safety, biosecurity, Fruit tree Upgrade, expand, and integrate with livestock processing, and storage Fruit tree (low resilience Reducing and only developing in suitable areas, capacity to salinity and Other fruits especially between the two rivers flood such as durian) Cleanliness, safety, and Specialized vegetable Upgrading and expanding biosecurity Standard procedures, production Rice-aquaculture Upgrading and expanding zoning, value chain development, and large market channel Integrated coconut- Commercial production area and Upgrading and expanding aquaculture value chain development Rice-duck Upgrading and expanding Value chain development In the current context of complicated diseases, cattle production should maintain its small production scale and focus on biosecurity, Cleanliness, safety, biosecurity, Beef cattle farming reducing emissions, and a circular economy. value chain development, and In the medium term and long term, cattle cooperative development production could be considered to expand for households with a strict biosecurity protocol. Cleanliness, safety, biosecurity, Chicken farming Developing and expanding value chain development, and cooperative development Cleanliness, safety, biosecurity, Pig farming Maintaining and expanding with consideration value chain development, and cooperative development Cleanliness, safety, biosecurity, Industrial crops Maintaining and processing Source: Prepared by authors (2019). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 62 Alternative Scenarios and Policy Implications 3. Coastal subregion 3.1. General situation Aquaculture is the main industry in the Coastal subregion. Shrimp production in the MKD was 789,200 tons in 2018, representing nearly a 70 percent increase over the output in 2012. The Coastal subregion accounted for 90 percent of this total. Shrimp production is concentrated mainly in Ca Mau, Bac Lieu, Kien Giang, Soc Trang, Ben Tre, and Tra Vinh Provinces, with a total area of 650,820 ha. The current distribution of farming areas for white-leg shrimp and tiger shrimp is 13 percent and 87 percent, respectively, while their share of the volume is 57 percent and 43 percent, respectively. A trend of expanding shrimp-farming areas in this subregion occurred in recent years resulting from increased saline intrusion, which forced farmers to change from freshwater rice cultivation to brackish-water shrimp farming. The shrimp-farming area increased by more than 30,000 ha during 2012–2016. The level of shrimp-farming intensification varies among provinces; Bac Lieu Province is the most intensive, followed by Soc Trang and Ca Mau Provinces. However, the total area of intensive and highly intensive farming accounts for less than 10 percent, while the extensive and improved extensive farming models account for around 90 percent. Integrated shrimp- farming models were also promoted, such as mangrove-shrimp farming in Ca Mau and Bac Lieu Provinces. Other coastal aquaculture models are also promoted, including blood cockle farming in Kien Giang Province and clam farming in Ben Tre, Tien Giang, and Tra Vinh Provinces. The Coastal subregion does not have favorable conditions for rice production compared to the other subregions. Starting in the 1990s, the government invested in mega-irrigation projects to divert freshwater from the Hau River to the Coastal subregion to support rice production. A shift from the 1-crop rice to 2-crop and even 3-crop rice systems occurred after that. However, in the past decade, when aquaculture brought higher profits and rice farming was not efficient due to increased saline intrusion, the government allowed converting some mono-rice areas to shrimp-rice farming or shrimp farming. This mainly happened in the coastal districts of Ca Mau, Bac Lieu, and Soc Trang Provinces. In some locations, farmers even converted their rice fields to shrimp farming spontaneously. The area of the 3-crop rice now is marginal. The 2-crop rice model in 2018 was around 100,000 ha. This model continues to be replaced by the 1-crop rice rotated/intercropped with shrimp or converted to intensive shrimp farming (that is, in An Minh and An Bien Districts of Kien Giang Province; Vinh Thuan, Thoi Binh, and Cai Nuoc Districts of Ca Mau Province; and Cau Ngang and Duyen Hai Districts of Tra Vinh Province), at an annual rate of 1.6 percent. The Coastal subregion has featured relatively little development of fruit production. Coconut production has experienced some growth, with its area coverage now standing at just under 26,000 ha, about a 20 percent increase from a decade ago. Elsewhere in this subregion, most fruit production is on small and mostly mixed orchards, scattered in Tran Van Thoi District (Ca Mau Province); U Minh District (Kien Giang Province); and Vinh Thuan, Cu Lao Dung, and Tan Phu Dong Districts (Soc Trang Province). Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 63 Livestock production is also limited in the Coastal subregion. Although the number of beef cattle raised doubled from 2005 to 2018, the pig and buffalo population declined. The most significant decrease came from the buffalo herd (by half) because farmers do not use buffalo as draft animals anymore. The poultry population in the subregion remained stable from 2010 to 2018 at around 7.5 million. In addition, this area has some new lines of animal product production as with goats and bees in Ca Mau Province and the neighboring areas. 3.2. Key agricultural livelihood models The five most common livelihood models in the Coastal subregion include brackish-water aquaculture (mainly shrimp), aquaculture-mangrove, rice-aquaculture, 2-crop rice, and industrial crops (Figure III-9 and Table III-21). In 2018, the extensive shrimp-farming model covered the largest area (38 percent) of the Coastal subregion, yet it contributed only 6 percent of the total value of agricultural production. Semi-intensive shrimp and intensive shrimp models covered about 9 percent of the farming land but contributed 54 percent of the subregion’s total agricultural production value. These two models attracted the participation of 97,200 and 59,000 households, respectively. The 2-crop rice and shrimp-rice models are the main choices of more than 100,000 families in the Coastal subregion. Only a few enterprises and some wealthy households are engaged in the super-intensive shrimp model, making up only 0.1 percent of the area and contributing 1 percent of the subregion’s agricultural production value. The 2-crop rice and industrial crop models (that is, sugarcane, pineapple, and jute) account for very small areas (around 100,000 ha and 3,000 ha, respectively). FIGURE III-9. Map of key livelihood models in the Coastal subregion Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ illustration from Dang et al. (2019) and survey data. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 64 Alternative Scenarios and Policy Implications TABLE III-21. Coverage of key livelihood models in the Coastal subregion Area (ha, % of ag- % of ag- Number of Produc- % of ag- % of ag- thou- ricultural ricultural house- tion value ricultural ricultural Livelihood mod- sands) land in land in holds (VND, produc- produc- els Coastal MKD (thou- billions) tion value tion value subregion sands) of Coastal of MKD subregion 2-crop rice 137.7 13.1 4.10 61.8 7,889 2.4 1.1 Rice-shrimp 154.7 14.7 4.70 45.5 20,800 6.3 3.0 Mangrove-shrimp 121.2 11.5 3.60 30.0 6,058 1.8 0.9 Shrimp 424.3 40.3 12.80 aquaculture • Extensive aquaculture/ improved 393.0 31.7 10.10 97.2 16,713 5.0 2.4 extensive shrimp • Intensive/semi- 89.0 8.4 2.70 59.0 178,000 53.5 25.6 intensive shrimp • Super-intensive 1.0 0.1 0.03 4,500 1.4 0.6 shrimp Source: Calculated by authors based on AgroCensus 2016, Dang et al. (2019), and digital mapping data (2019). 3.3. Assessment of the most common livelihood models 3.3.1. Natural condition suitability and climate change resilience Under current natural conditions, the specialized shrimp production models (extensive/ improved extensive, intensive/semi-intensive, and super-intensive farming scheme) are the most suitable in the Coastal subregion (Table III-22). More than 53 percent of the subregion’s agricultural area is suitable or highly suitable for these models. Only 14 percent of the area is suitable for the 2-crop rice model, located in areas with the ability to store rainwater (Tran Van Thoi District in Ca Mau Province) or having infrastructure systems to prevent salinity (in Kien Giang Province). In the future, with increased saline intrusion, land subsidence, and sea level rise, the freshwater areas will be reduced, especially in the dry season, so rice production will no longer be viable. Instead, it will be more suitable for the brackish-water economy, including shrimp production. Therefore, the specialized shrimp-farming model may go up significantly by 110,000 ha, and the rice-shrimp area may increase by about 10,000 ha. The mangrove-shrimp coverage will likely remain unchanged (assuming that the mangrove area remains unchanged). Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 65 TABLE III-22. Ranking of livelihood models for suitability to future natural conditions, Coastal subregion Area of most Proportion of Change of Index of Index of Index of suit- suitable and most suitable most suitable column A column B ability with Livelihood suitable land and suitable and suitable future natural models in 2030 (ha, land in 2030 land 2019– condition thousands) (%) (A) 2030 (%) (B) Super-intensive 772 66.9 17.0 4.0 4.0 4.0 aquaculture Extensive 772 66.9 17.0 4.0 4.0 4.0 aquaculture/ improved extensive aquaculture Semi-intensive/ 772 66.9 17.0 4.0 4.0 4.0 intensive aquaculture Organic 121 10.5 0.0 1.0 4.0 2.5 mangrove- shrimp Rice-shrimp 170 14.7 6.3 1.0 4.0 2.5 2-crop rice 73 6.3 -53.5 1.0 1.0 1.0 Source: Authors’ calculation. Note: Index of column A is computed as 80-100%: 5; 60-79%: 4; 40-59%: 3; 20-39%: 2; and 0-19%: 1. The index of column B is computed as 20-100%: 5; 1-20%: 4; 0%: 3; (−1)–(−20%): 2; (−20)–(−100%): 1. The index of future natural condition suitability is the average of the indexes of columns A and B. Regarding climate change resilience capacity, the 2-crop rice model is the least resilient because it is difficult to change the crop schedule (Table III-23). While it is possible to change the input mix (including using salinity-tolerant varieties), the beneficial effect is not high. In addition, this model is also relatively easy to switch to other more effective models with modest investment costs. The shrimp models have a high capacity to adapt to climate change because of their high salinity tolerance. The extensive, improved extensive, semi-intensive, and intensive shrimp models are easy to switch back and forth by adjusting stocking densities. However, the super-intensive shrimp model is difficult to change because the initial investment is very large. The organic mangrove-shrimp model has an excellent ability to absorb and tolerate climate change. Still, it is impossible to switch to other models because farmers must maintain mangroves as coastal protection forests. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 66 Alternative Scenarios and Policy Implications TABLE III-23. Ranking of livelihood models for climate change resilience, Coastal subregion Absorptive Transformative Climate change Livelihood models Flexibility ability ability resilience index Rice-shrimp 5.0 5.0 3.0 4.3 Organic mangrove-shrimp 5.0 3.0 4.0 4.0 Super-intensive 5.0 2.0 1.0 2.7 aquaculture Extensive/improved 3.0 3.0 1.0 2.3 extensive aquaculture Semi-intensive/intensive 3.0 3.0 1.0 2.3 aquaculture 2-crop rice 2.0 2.0 5.0 3.0 Source: Authors’ evaluation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Climate change resilience index (column 5) = average of absorptive ability (column 2), flexibility (column 3), and trans- formative ability (column 4). 3.3.2. Economic and financial performance The livelihood models in the Coastal subregion can be classified into two categories: (a) the rice-linked models that are associated with salinity control infrastructure and (b) the brackish-water livelihoods that require energy-related infrastructure (Figure III-10). The average public investment for the 2-crop rice model and rice-shrimp model is about VND 32 million per ha, and the main structures required include dikes, sluice gates, pump stations, and canals. Public investment for the extensive, improved extensive, and semi-intensive/intensive aquaculture models ranges from VND 4.5 million per ha to VND 50 million per ha, and the main public investment required relates to electricity supply. For the super-intensive shrimp model, the public investment requirement is much larger at VND 38 million per ha because this model requires a lot of energy to operate. In the rice-shrimp, extensive, and improved extensive shrimp models, farmers need to spend VND 28 million per ha for digging ditches or ponds next to the water pumping system. The 2-crop rice model does not require a large investment at the household level, requiring only a water pump system and an investment of VND 3 million per ha. Significant differences exist in variable production costs among various farming systems (Figure III-11). The extensive and improved shrimp models require operating costs of VND 24-25 million per ha per year, mainly for fingerlings. These models require no other substantial operational expenditure. Meanwhile, the operating costs for the rice-shrimp, semi-intensive, and intensive shrimp models are higher at VND 110 million per ha and VND 330 million per ha per year, respectively, due to the cost of feed and veterinary, which farmers can buy only partly on credit. The super-intensive shrimp model requires very high operating costs (VND 3.5 billion per ha per year) together with a large initial investment. The operating costs are mainly for fingerlings, feed, veterinary drugs, and electricity. Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 67 FIGURE III-10. Investment in infrastructure of key models, Coastal subregion Super-intensive aquaculture 2500 Intensive/Semi-intensive 200 Aquaculture Extensive/Improved 28 Extensive Aquaculture Rice-shrimp 28 2-crop rice 3 Unit: VND, millions per ha Source: Authors’ calculation from household surveys in this research (2019). FIGURE III-11. Total production cost of key livelihood models, Coastal subregion Super-intensive aquaculture Intensive/Semi-intensive aquaculture Extensive/Improved Extensive aquaculture Buying on credit Rice-shrimp Immediate payment 2-crop rice Unit: VND, millions per ha per year 0 500 1000 1500 2000 2500 3000 3500 4000 Source: Authors’ calculation from household surveys in this research (2019). The profits generated by the different livelihood models vary considerably (Figure III-12). The super-intensive shrimp model can generate a profit of VND 4.3 billion/ha/year. The semi-intensive and intensive shrimp models can generate profits of VND 500 million/ha/year. The 2-crop rice model offers the lowest profit: VND 34 million/ha/year. Other models can generate profits of VND 70-80 million ha/year. The standard deviation for the profitability of those models shows that high volatility exists for high-profit models. In the super-intensive model, the difference can be as high as VND 1 billion/ha/year. In some periods during 2009–2019, high-profit models (semi-intensive, intensive, and super-intensive shrimp farming) suffered big losses, from VND 600 million/ha/year to VND 1,500 million/ha/year. The specialized rice-farming models or rice rotation models consistently offered low but stable profits over the past 10 years (Table III-24). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 68 Alternative Scenarios and Policy Implications FIGURE III-12. Average profit of key models, Coastal subregion, 2009-19 6,000.00 5,000.00 Unit: VND, millions per ha per year 2-season rice 4,000.00 Rice-shrimp Extensive/Improved 3,000.00 Extensive Aquaculture Semi-intensive/Intensive 2,000.00 aquaculture Super-intensive 1,000.00 aquaculture Organic forest-shrimp 0.00 Source: Authors’ calculation from household surveys in this research (2019). TABLE III-24. Ranking of livelihood models for economic performance, Coastal subregion Investment Net profit Economic index Livelihood models Operating cost cost (weight=2) (weight=2) Rice-shrimp 3.0 5.0 4.0 4.0 Organic mangrove-shrimp 3.0 4.0 4.0 3.8 Semi-intensive/intensive aquaculture 3.0 2.0 4.0 3.3 Super-intensive aquaculture 1.0 1.0 5.0 3.0 Extensive/improved extensive 4.0 4.0 2.0 3.0 aquaculture 2-crop rice 5.0 5.0 1.0 3.0 Source: Authors’ evaluation from survey data. Note: For cost, 1 = very high; 2 = high; 3 = medium; 4 = low; 5 = very low. For profit and economic index, 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Economic index (column 5, with weight = 2) = average of investment cost (column 2), operating cost (column 3), and net profit (column 4), with weight = 2. The profitability of various livelihood models varies from year to year and within a year between different areas. The main reason is risk profile. Table III-25 shows that the 2-crop rice model is less risky in terms of both the consumption market and crop disease, but it is more vulnerable to climate change impacts, such as drought and salinity intrusion. Meanwhile, the shrimp-farming models can be exposed significantly to the consumption market and disease risks, but they are less affected by increased salinity caused by climate change. The low-value models (2-crop rice, rice- shrimp, and extensive and improved extensive shrimp farming) face fewer risks than the high-value models (semi-intensive, intensive, and super-intensive shrimp farming). Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 69 TABLE III-25. Ranking of livelihood models for risk-coping ability, Coastal subregion Number of years in 2010–2019 (10 Risk coping ability years) that a farming household enjoys Livelihood models No serious No serious No serious Market Disease Disaster Risk coping market disease disaster access ability shock (weight =2) index Organic mangrove- 7.0 9.0 8.0 4.0 5.0 5.0 4.5 shrimp Super-intensive 7.0 9.0 8.0 4.0 5.0 5.0 4.5 aquaculture 2-crop rice 8.0 7.5 7.5 5.0 4.0 4.0 4.5 Rice-shrimp 7.0 9.0 7.0 4.0 5.0 4.0 4.3 Extensive/improved 7.0 3.0 6.0 4.0 1.0 3.0 3.0 extensive aquaculture Semi-intensive/ 7 3 6 4.0 1.0 3.0 3.0 intensive aquaculture 7.0 3.0 6.0 4.0 1.0 3.0 3.0 Source: Authors’ evaluation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Risk-coping ability index (column 5) = average of market access with weight = 2 (column 2), disease (column 3), and disaster (column 4). 3.3.3. Employment generation Different livelihood models have different labor requirements. The semi-intensive, intensive, and super-intensive shrimp-farming models require the largest number of workers during the production stage (543 working days in a year), but they mainly recruit young and skilled male workers to do heavy work (that is, feeding) and protect the farms from thieves (Table III-26). These models often face a labor shortage of young and skilled workers. In response to this, mechanization and automation systems are increasingly used in these models to address the labor shortage and, at the same time, improve production efficiency. In contrast, the 2-crop rice, rice-shrimp, and extensive and improved extensive shrimp models require fewer workers, as they require only 30-75 person-days per year. So, young people from these farming households prefer to migrate to big cities to find work and other higher-income opportunities than work in rural farms. This explains why this subregion has the highest proportion of migrants in the MKD. TABLE III-26. Ranking of livelihood models for social impact, Coastal subregion Employment Contribution to Livelihood models Social index creation total income Super-intensive aquaculture 5.0 5.0 5.0 Organic mangrove-shrimp 4.0 4.0 4.0 Semi-intensive/intensive aquaculture 4.0 4.0 4.0 Rice-shrimp 4.0 3.0 3.5 Extensive/improved extensive aquaculture 3.0 4.0 3.5 2-crop rice 2.0 2.0 2.0 Source: Authors’ evaluation from survey data. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Social index (column 4) = average of employment creation (column 2) and contribution to total income (column 3). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 70 Alternative Scenarios and Policy Implications 3.3.4. Environmental aspects Different livelihood models have different environmental footprints (Table III-27). The extensive and improved extensive shrimp models have the least negative impacts on the environment because of the restricted use of chemicals and water exchange during the whole farming cycle. The 2-crop rice model has the most negative impacts on the environment because of the overuse of chemicals. In addition, the 2-crop rice model relies on rain irrigation and needs to store water from six months to a year. The fertilizers and pesticides accumulated from the 2-crop rice model also cause pollution in the production areas. Meanwhile, the semi-intensive shrimp model and more advanced models show little impact on land resources. Still, they have a substantial effect on water resources, especially the semi-intensive and intensive models without adequate water treatment systems. In addition, sludge from shrimp-farming activities is a concern because it affects other neighboring production systems. TABLE III-27. Ranking of livelihood model suitability based on environmental impact, Coastal subregion Organic Super- Semi-inten- Extensive man- Rice- intensive sive/Inten- Livelihood models 2-crop rice aquacul- grove- shrimp aquacul- sive aquacul- ture/ shrimp ture ture Surface water 4.0 5.0 5.0 3.0 3.0 3.0 Groundwater 5.0 5.0 5.0 5.0 4.0 5.0 NRU Soil use intensity 5.0 5.0 5.0 5.0 4.0 3.0 Energy (electricity) 5.0 5.0 5.0 4.0 4.0 5.0 NRU index 4.8 5.0 5.0 4.3 3.8 4.0 Water pollution On-site environmental 5.0 5.0 5.0 5.0 3.0 2.0 (weight = 2) Soil pollution 5.0 5.0 5.0 5.0 3.0 3.0 impact Air pollution 5.0 5.0 5.0 5.0 3.0 3.0 GHG emissions 4.0 5.0 4.0 4.0 5.0 2.0 Landscape change 5.0 5.0 5.0 4.0 5.0 5.0 EI Index 4.8 5.0 4.8 4.7 3.7 2.8 Foundation 5.0 5.0 5.0 5.0 5.0 4.0 subsidence Salinity intrusion 5.0 5.0 5.0 5.0 4.0 4.0 Biodiversity loss 5.0 3.0 3.0 5.0 4.0 3.0 IRI Flooding in urban 5.0 5.0 5.0 5.0 5.0 5.0 area IRI 5.0 4.5 4.5 5.0 4.5 4.0 Environmental index 4.9 4.8 4.8 4.6 4.0 3.6 Source: Authors’ evaluation from survey data and expert consultations. Note: 1 = very seriously impacted; 2 = seriously impacted; 3 = impacted; 4 = lightly impacted; 5 = negligible. Environmental index (last row) = average of NRU index, EI index, and IRI index. Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 71 3.3.5. Gender aspects The 2-crop rice production in the Coastal subregion mostly depends on rainwater, and more labor is required for irrigation and water management activities, which are performed mainly by men (Table III-28). This is why the 2-crop rice model in this subregion has slightly less involvement of female labor than the other two subregions. Generally speaking, the participation of women in rice production is limited. In the rice-shrimp model, women are even less involved. The most labor-intensive stage of the rice-shrimp model is harvesting, mainly carried out by male labor (for both rice and shrimp). Therefore, the contribution of women in this model is even lower than in the 2-crop rice model. The situation in extensive and improved extensive shrimp is much different. In this livelihood model, more female labor is required to prepare shrimp feed, feed, run the water aeration system, and observe the health of shrimp. Men are mainly in charge of choosing shrimp seeds, feeding, water quality management, and sleeping near their ponds to take care of the shrimp. Like in the Pangasius production model, there is a superstition that women can bring bad luck to shrimp ponds. Although this superstition is fading away, owners of shrimp ponds always try to avoid strangers coming to their pond to reduce disease risks. Since women often go to the market and they can carry potentially harmful pathogens with them, they are less welcome to shrimp ponds. TABLE III-28. Ranking of livelihood models for gender aspects, Coastal subregion Power of Negative impacts on Gender Livelihood models women women (workload) index Rice-shrimp 4.0 5.0 4.5 2-crop rice 3.0 5.0 4.0 Extensive/improved extensive aquaculture 2.0 5.0 3.5 Organic mangrove-shrimp 2.0 4.0 3.0 Super-intensive aquaculture 2.0 4.0 3.0 Semi-intensive/intensive aquaculture 2.0 4.0 3.0 Source: Authors’ evaluation from survey data and expert consultation. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Gender index (column 4) = average of power of women (column 2) and negative impacts on women (column 3). As its production value is very high for the intensive shrimp model, a small risk could push a household into an economic crisis. As a mitigation measure, most of these households hire one or two permanent workers, mainly men, to take care of their ponds, who rarely leave the ponds to minimize disease risks. The husband visits the shrimp ponds regularly, while the wife does not come to the ponds until harvest. This is why female labor accounts for only 7 percent of the total of 356 working days for 1 ha of intensive shrimp production. The situation is similar for the super-intensive shrimp model but with more labor (2-4 male laborers per farm). 3.3.6. Conclusion In the Coastal subregion, the 2-crop rice model shows relatively poor adaptation to future natural conditions of sea level rise, erratic salinity intrusion on a larger scale, and greater Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 72 Alternative Scenarios and Policy Implications intensity. Although this model can take advantage of some salinity-tolerant varieties, the economic efficiency is low and creates only limited jobs. Therefore, the 2-crop rice model will only be suitable in restricted areas, while production in other areas will need to shift to the rice-shrimp model. The extensive and improved extensive shrimp models are the most sustainable because they are suitable for brackish-water ecosystems. The investment costs and profits of these models are average. However, the biggest drawback of this model is that it does not create many job opportunities. In the future, these models should be upgraded to enhance biosecurity and achieve ecological certificates to increase product value, thereby increasing income. The semi-intensive and intensive shrimp models can adapt well to future climate change conditions and generate good income and employment. However, these models, investments in water treatment and filtration systems, and farm biosecurity are critical to managing disease risks. Otherwise, farmers should reduce stocking density toward extensive ecological farming. Those households with the financial capacity can switch to the super-intensive farming model in which all farming parameters are well monitored and strictly controlled. The rice-shrimp and super-intensive shrimp models are ranked as the top livelihood systems for the Coastal subregion as they are suitable for future natural conditions, bring good profits with less disease risk, and create many jobs (Table III-29). Mangrove-shrimp integrated farming is another good model suitable for natural conditions and generates a good income. It is essential for this model to standardize the techniques according to the ecological standards and build eco-product consumption value chains with a higher value than conventional products. TABLE III-29. MCA ranking of livelihood models for suitability and sustainability, Coastal subregion Suitabil- Climate Eco- Risk Social Environ- Gender Final ity with change nomic coping index ment index suitabil- natural resil- index ability index ity and Livelihood models condition ience (weight index sustain- (weight index =2) ability =2) index Rice-shrimp 2.5 4.3 4.0 4.3 3.5 4.9 4.5 3.8 Organic mangrove- 4.0 3.0 3.0 4.5 5.0 4.4 3.0 3.8 shrimp Super-intensive 2.5 4.0 3.8 4.5 4.0 5.0 3.0 3.7 aquaculture Extensive/improved 4.0 2.7 3.0 3.0 3.5 3.6 3.5 3.4 extensive aquaculture Semi-intensive/ 4.0 2.3 3.3 3.0 4.0 2.9 3.0 3.3 intensive aquaculture 2-crop rice 0.8 3.0 3.0 4.5 2.0 4.3 4.0 2.8 Source: Authors’ evaluation from survey data and expert consultation. Note: 1 = very low; 2 = low; 3 = medium; 4 = high; 5 = very high. Final suitability and sustainability index (column 9) = average of all columns from 2 to 8 (considering the weights of 2 of columns 2 and 4). Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 73 This ranking indicates some main directions for transforming key and potential livelihood models in the Coastal subregion (Table III-30). TABLE III-30. Tentative development directions of livelihood models, Coastal subregion Livelihood models Development directions Target models Key models 2-crop rice Decrease, improve Shrimp-rice, 2-crop high-quality rice Rice-giant head shrimp Decrease Rice–tiger shrimp Production process standardization, Rice-tiger shrimp Improve, expand production zone planning, value chain development, market channels Extensive aquaculture/improved extensive Hygienic, biologically safe, organic, Improve aquaculture certificates Disease-free shrimp using physical and biological water treatment Semi-intensive/intensive aquaculture Improve technology to rebuild natural conditions at the household level Super-intensive aquaculture Improve, expand High-tech models, sensor, indoor Shrimp-forest combining with eco-tourism Improve, expand Standard models with certificates Source: Prepared by authors (2019). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 74 Alternative Scenarios and Policy Implications CHAPTER IV Sustainable agricultural livelihood transformation Chapter III - Assessment of main and potential livelihood models of the three Mekong Delta subregions 75 In this chapter, two vastly different scenarios for agricultural development in the MKD are considered to illustrate the benefits of agricultural transformation for the MKD. To this end, key policy options are recommended. The time horizon covers the period to 2030. Contrasts are provided between a BAU scenario and the PAT scenario. Major differences between the two scenarios pertain to whether or not a significant process of land-use change occurs and is facilitated by the government and the degree to which the government is effective in enabling and leveraging private investment in agricultural innovation, services, and value chains in the MKD. Either the agro- economy in the MKD will adapt to climate change and adjust effectively to changing demographic and socioeconomic conditions—and prosper in doing so—or it will struggle to adjust, with negative consequences for its people and the nation. 1. Different development scenarios 1.1. Introducing two scenarios Scenario 1: BAU. In this scenario, legacy land-use patterns are more or less maintained, and only some changes are made at the margins. Many farmers prove reluctant to shift their land use to potentially more lucrative endeavors, perhaps because of labor and/or financial constraints or concerns about more significant production and market risks. And under this scenario, many local governments fail to promote more diversified land-use and farming systems. Why might this occur? One factor is systemic—budgetary allocations from the central government have often been based on the needed infrastructure to protect rice production. If that approach does not change, the motivation to promote change may be weakened. Additional factors might come into play. It is not inconceivable that there will be a year or season in which weather-related shock results in a decline in rice output and an increase in consumer prices. This may induce some local leaders to ‘put the breaks’ on land-use change and revert to strategies of protecting rice lands, even at very high financial and opportunity costs. Other factors could include the lack of detailed guidance from the central government for local governments or detailed studies to inform the transformation or necessary infrastructure to support the new systems to make them viable and sustainable. As a result, in this scenario, most farmers would continue their current farming models until they find better opportunities, decide to retire, or experience a major shock (for example, production, health, or financial) that may force them to lease or transfer their land-use rights to others. Also, the government would continue investing in more large-scale infrastructure to protect current farming systems (especially rice production) with limited attention to longer-term environmental sustainability. Scenario 2: PAT. In this scenario, the government would plan for and accelerate agricultural transformation toward climate-resilient and green production and do so by facilitating land-use changes, shifting the nature and focus of public investments, and creating a much more favorable environment for private investment and service delivery. In this scenario, agriculture becomes more market driven with the government continuing to provide a strategic vision and selected core public goods, yet with the bulk of service delivery shifted over to cooperatives, the private sector, and one or another type of public-private partnership (PPP) initiative. This is akin to the longer-term vision laid out in the ARP. In this scenario, the government would adopt (and locally apply) a more flexible land- Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 76 Alternative Scenarios and Policy Implications use policy, adopt policies in support of digital applications to agriculture, and implement policies and programs to promote private investment and incentivize improvements in farming practices. To calculate and project the outcomes of this scenario, the natural condition maps under the climate change scenario to 2030 prepared by CTU were used. Based on these maps, the adaptive areas for the specific types of livelihood model were calculated for subregions, provinces, and districts. Based on those, the cost and benefit are calculated. Data of the livelihood models were collected from the household surveys and expert consultations conducted by the research team in 2018. The data of the costs and benefits, labor use (especially high-tech and high-skilled labors), and shifting areas of the livelihood models were used for calculating the gross agricultural production value, agricultural added value, and employment generation for this scenario. 1.2. Climate change and land-use changes 1.2.1. Climate change The climate change predictions would be similar for the two scenarios. Upper subregion. In the next 10 years, one of the fundamental natural condition changes in the upstream area will be flooding. Due to its favorable drainage conditions to the West Sea, the level of floodwaters in Dong Thap Muoi is currently higher than in LXQ. However, because of sea level rise, the ability to drain water will be much more difficult, so the floodwater levels in LXQ will be as high as in Dong Thap Muoi, back to the time of floating-rice cultivation. The flood level in the Ha Tien area of Kien Giang Province will be from below 0.3 m now to 0.6-0.9 m, and flooding time will last longer, from under 3 months now to 3-6 months. The flood level will also rise from the current 0.3-0.6 m to 0.6-0.9 m in Hon Dat District in the future, and flooding time will also increase from less than 3 months to 3-6 months. The coastal areas of the Upper subregion will suffer the longest flooding season. In Dong Thap Muoi, the flood will increase from 0.6-0.9 m at present to more than 0.9 m, with longer flooding time, in areas that are far from the river in Thap Muoi District of Dong Thap Province and areas of Vinh Hung and Kien Tuong Districts of Long An Province. Middle subregion. In the context of climate change, the Middle subregion will find it more difficult to discharge water to the West Sea, and a large area south of the Hau River will become flooded. In areas of Giong Rieng, Go Quao, Chau Thanh, and south of Tan Hiep District of Kien Giang Province and Long My, Vi Thanh, and Co Do Districts of Hau Giang Province, the water level will increase to 0.6-0.9 m where the previous flood level was 0.3-0.6 m. Flooding time will increase from less than 3 months to nearly 6 months, thereby changing the hydrological regime on a large scale. In addition, for the northern region of Cai Be and Cai Lay Districts of Tien Giang Province and the south of Thap Muoi District of Dong Thap Province, the inundation level will be higher during the flooding season. The current water level (0.6-0.9 m) will increase to more than 0.9 m. Flooding time is forecasted to increase from 3-6 months to 6 months in the future. For the area bordering the East Sea, salinity intrusion will intensify on a small scale. In Go Quao District of Kien Giang Province, the northern area of Cu Lao Dung of Soc Trang Province and Can Duoc, Can Giuoc, and Ben Luc Districts of Long An Province, salinity intrusion will increase from less Chapter IV - Sustainable agricultural livelihood transformation 77 than 3 months to 3-6 months. Also, more new areas will be affected by saline intrusion of less than 3 months. Nearly a quarter of the Middle subregion will have to convert to other livelihoods because of salinity intrusion and inundation that will affect the current agricultural production structure. Sea level rise makes it challenging to allow drainage to the sea. The impact of increasing the level of flooding and the time of inundation is projected to be more serious than the impact of salinity intrusion. Coastal subregion. For the next 10 years, the Coastal subregion of the MKD will experience more severe salinity intrusion. A salinity level of higher than 4 g/L and intrusive duration of more than 6 months will dominate most of the eastern Coastal subregion (at present, nearly half of this area is currently exposed for 3-6 months). The coastal area of Ha Tien District (with current salinity of 4 g/L to less than 8 g/L) and the southern area of Cai Lon and Cai Be Rivers (with present salinity of 4 g/L to less than 8 g/L) will experience salinity of more than 12 g/L to less than 20 g/L. In the entire east of Ca Mau Cape, salinity will double from the current rate of more than 20 g/L. For the east coast, salinity will increase from less than 12 g/L to more than 20 g/L. In addition, the west coastal area of Hon Dat District will be inundated to more than 0.9 m, and the inundation duration will last for up to 6 months. The southern part of Cai Lon River in An Bien District, Kien Giang Province, will be flooded from 0.6 m to 0.9 m, and the flooding time will last for up to 6 months. 1.2.2. Land-use changes (a) BAU scenario The land-use map for the MKD in 2017 shows the dominance of rice cultivation, either in 2-crop or 3-crop systems. The 3-crop rice areas comprise most of the land in the Upper subregion and Middle subregion and cover the stretch of land in between the Tien and Hau Rivers. The area affected by salinity during the dry season in this region is currently allocated to rice-shrimp farming on a large scale. These two production systems generate very large volumes of output, yet also consume a lot of natural resources (land and water) and have had a very heavy environmental footprint in terms of water pollution, biodiversity loss, and GHG emissions. In recent years, there has been some reduction in rice cultivation: some marginal 3-crop rice land areas have gradually shifted to 2-crop rice, rice-vegetable, or rice-aquaculture systems. The rice areas in the rice-shrimp model have also been decreasing, switching to specialized shrimp farming or other forms. Meanwhile, the area under fruit trees has expanded, replacing some of the areas under the 3-crop rice model. In the BAU scenario, the government and farmers would invest more in infrastructure to protect rice production to maintain the present rice production areas until it becomes much clearer that this is unsustainable due to the risks and costs involved. In the Upper subregion, flood prevention works would be further developed. In the Middle subregion, protection dikes would be consolidated, and pumps would be used against local floods. In the Coastal subregion, additional large-scale salinity prevention structures would be built. Overall, the investments for maintaining and improving irrigation structures would be significant. Production costs and risks would further increase while efficiency and farm profits would, almost certainly, decrease. Despite significant investments in extensive protective infrastructure, residential areas and non-adaptive production systems would remain vulnerable to climate change and Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 78 Alternative Scenarios and Policy Implications extreme weather events. They would depend on the government’s maintenance and operation of preventive structures. If the state budgets for infrastructure development are insufficient and regional cooperation in managing these structures is weak, then the productive capacity of the MKD would be challenged and, probably, weakened. In such circumstances, farmer incomes would be at risk, and the government’s goal of maintaining rice production would not be achieved. Under this scenario, the complex impacts of decreased growth and lower economic efficiency would generate further social pressures and lead to accelerated out-migration. The environmental pressure would also increase, and the consequences, such as ground subsidence (due to groundwater exploitation), pollution (due to chemical use), and landslides (due to covering dikes) would manifest more frequently. Many smallholder farmers would be forced to exit from agriculture—in numbers far in excess than what would otherwise be expected from the normal process of economic structural change. An aging cadre of farmers (averaging 50 years old now) would not be substantially replaced by a younger, better-educated generation as the latter would be deterred from remaining/entering into agriculture if restricted to non-viable options. The sector’s human capital quotient would thus decline. Perhaps, there would be more corporate farming, although not many firms would be interested in investing in rice production in vulnerable areas. There would be a slight reduction in marginal rice land area but without a corresponding increase in other things. This scenario is not sustainable from an economic, social, or environmental perspective and would involve public investments which, in retrospective, could involve much ‘regret’. (b) PAT scenario In the next 10 years, with the increased frequency and intensity of climate change impacts (that is, floods and droughts, sea level rise, and saline intrusion), most farmers in marginal and vulnerable areas would change to more adaptive livelihood models if they remain in their same locations. Inaction will not be a viable choice. As land-use restrictions are removed and as supportive infrastructure (and services) is put in place, the shifts will begin to accelerate, both in pace and scale. The order of magnitude of expected shifts is highlighted for each of the three subregions, in turn. In the Upper subregion, the 3-crop rice areas would be reduced in the next 10 years, and in some places, the third rice crop would be replaced by aquaculture or flood-based crops. This rollback to traditional livelihood models would help improve soil fertility and farmers’ incomes, and it also would help reduce the frequency and intensity of floods in downstream areas. That being said, it would be expected that nearly two-thirds of the current 3-crop rice area, especially that in locations where high dikes are in place, would more or less maintain current production systems. With new or repurposed infrastructure and with a relaxation of land-use restrictions, about one-third of these areas would shift, with the most common shift being to rice-flood-based crop models. A small portion may shift to other rotation models or entirely different patterns of specialization. For the current 2-crop rice areas, about one-half will be maintained, about one-third would shift to freshwater aquaculture, and much of the balance would become rice-flood-based crop models. In the Upper subregion, three-fourths of the current areas under fruit trees would be retained, with the remaining shifting to rice-freshwater aquaculture or other production systems. In the aggregate, such changes Chapter IV - Sustainable agricultural livelihood transformation 79 would lead to a shift away from 3-crop and 2-crop mono-crop systems on a scale of approximately 150,000 ha each for: (a) rice-freshwater aquaculture systems and (b) rice-flood-based crop (that is, rice- lotus and rice-floating vegetables) systems and a modest decline in the area under fruit trees. In the Middle subregion, the current areas without regular floods would face more frequent floods in the next 10 years, and some of the current 3-crop rice, fruit trees, and rice crop areas would be replaced by flood-based livelihood models. It is envisaged that about 60 percent of the 3-crop rice areas would be retained, 30 percent would switch to rice-freshwater aquaculture, and the balance would shift to other more flood-tolerant models. A much more significant change would occur in the current 2-crop rice production areas. Only 40 percent of this area would be retained, with another 40 percent converted to rice-freshwater aquaculture and the remaining 20% involving rice intercropped with floodplain crops or animal production (that is, ducks). In the Middle subregion, about 90 percent of the current fruit tree areas could be retained, with the remaining 10 percent which is more vulnerable to flooding being switched to rice-freshwater aquaculture. For salinity-affected areas, salinity-resilient livelihood models would be adopted (that is, coconut trees and brackish-water aquaculture). In the remaining areas of the Middle subregion, where the current livelihood models (that is, 3-crop rice, fruit trees, specialized crops, and rice-other crops) are still efficient, they would move toward greater degrees of specialization. In the Coastal subregion, salinity intrusion would be more severe in the coming 10 years with deeper and more prolonged inundation in some areas and freshwater crops being replaced by brackish-water aquaculture or salinity-resilient crops. Based on the land suitability maps, it is envisaged that more than half of the (more limited) 2-crop rice area would be converted to salinity-resilient crops, about one-fifth to brackish-water shrimp farming, and the remaining areas to the rice-shrimp rotation model. For fruit tree areas, only about 10 percent would be retained, and the remaining areas would be switched to shrimp farming or other types of livelihoods. It is also anticipated that about one-fifth of the current rice-shrimp area (which currently accounts for a large area in the subregion) would move towards more intensive shrimp farming, provided that biosecurity measures can effectively control against disease outbreaks. The current livestock sector (that is, poultry, pigs, and cattle) in the Upper subregion and Middle subregion is mainly of small scale (backyard) and can develop alongside crop cultivation or aquaculture as part of the families’ integrated farming systems. Therefore, growing small-scale family livestock does not exclude land use for other livelihood models. Land-use changes in the two different development scenarios are illustrated in Figure IV-1, Figure IV-2, and Figure IV-3. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 80 Alternative Scenarios and Policy Implications FIGURE IV-1. Land-use changes in BAU and PAT, Upper subregion, 2030 BAU PAT Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ illustration. FIGURE IV-2. Land-use changes in BAU and PAT, Middle subregion, 2030 BAU PAT Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ illustration. Chapter IV - Sustainable agricultural livelihood transformation 81 FIGURE IV-3: Comparison of BAU and LUR scenarios, Coastal Region, 2030 BAU PAT Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ illustration. TABLE IV-1. Area changes of key farming models in BAU and PAT scenarios Subre- Difference Difference Difference Livelihood models BAU PAT gions BAU-PAT BAU - LUR BAU - SCAC 3-rice 358,570 224,641 -133,928 -133,928 -133,928 2-rice 357,017 195,792 -161,224 -161,224 -161,224 Fruit tree 39,665 30,777 -8,888 -8,888 -8,888 Upper Rice-vegetable 8,964 6,101 -2,863 -2,863 -2,863 Rice-aquaculture 10,000 161,660 151,660 151,660 151,660 Rice-food-based crops 1,000 145,467 144,467 144,467 144,467 3-rice 589,679 334,979 -254,700 -254,700 -245,700 2-rice 253,949 106,814 -147,135 -147,135 -147,135 Middle Fruit tree 306,673 306,673 0 0 0 Rice-vegetable 12,645 108,663 96,018 96,018 96,018 Vegetable 10,970 60,970 50,000 50,000 50,000 Rice-aquaculture 1,000 226,344 225,344 225,344 225,344 2-rice 98,998 34,136 -64,862 -64,862 -64,862 Rice-shrimp 161,073 174,000 12,927 12,927 12,927 Extensive/improved extensive shrimp 357,566 146,149 (211,417) 22,434 (211,417) Semi-intensive/Intensive aquaculture 71,513 - (71,513) 7,873 (71,513) Coastal Super-intensive aquaculture 715 7,826 7,111 3,978 7,111 Household-level circular shrimp system - 121,682 121,682 - 121,682 Low density large-scale circular - 200,000 200,000 - 200,000 shrimp system Organic mangrove-shrimp 121,165 121,165 0 0 0 Source: Authors’ calculation. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 82 Alternative Scenarios and Policy Implications 2. Outcomes of agricultural transformation scenarios 2.1. Financial and economic outcomes The BAU and PAT scenarios would result in very different outcomes from a sustainable development perspective. The contrasted economic, social, and environmental outcomes are briefly summarized as follows: 2.1.1. Gross agricultural production value In terms of the production value of the main crops and aquaculture commodities, it is estimated that the BAU would have a lower production value than even the current level, meaning that BAU would result in a contraction of the MKD’s output (Figure IV-4). This is because the marginal rice land areas will become entirely unviable. The loss of output there would not be compensated for by a corresponding increase elsewhere. With an increased impact of climate change (that is, more prolonged floods and droughts, saline intrusion, and increased sea level rise) and more serious impacts from environmental damage in rice mono-cropped systems, a large number of small farmers will fail and considerable land could become idle (or spontaneously converted to nonagricultural uses) if the government does not have a comprehensive plan or enough resources to support the transformation on the ground (that is, maps and plans promoting land-use changes at local levels; supporting public infrastructure; and financing credits, value chains, and market). FIGURE IV-4: Comparison of gross agricultural outputs between different scenarios PAT 1,038 BAU 374 Current 402 Unit: VND, trillions in 2019 price Source: Authors’ calculation. In contrast, the production value gains in the PAT are estimated to be significantly higher than those in the current situation (an increase of 158 percent). The improvements are mainly attributed to the policy changes shifting the nature and focus of public investments and creating a much more favorable environment for private investment in more productive systems, advanced technologies, and service delivery to improve farm productivity substantially. In this scenario, it is assumed that the government would proactively restructure the farming systems in the MKD to be more adaptive and resilient to climate change and more competitive. More specifically, the government would implement a policy reform to allow more flexible land use to accelerate diversification away from rice for rice-growing areas which are no longer efficient and increasingly vulnerable. This is critically important to legalize and orient the process and mobilize farmers and other actors in the value chains. Public investments would shift from large-scale infrastructure fighting against climate change (that is, Chapter IV - Sustainable agricultural livelihood transformation 83 high dikes to prevent floods and saline intrusion) to missing infrastructure supporting the conversion to new production systems and leveraging private investments in biological and digital technologies in support for productive, sustainable, and resilient farming systems. And to enable all this to be commercially viable, continued improvements would be made in the business enabling environment to attract increased private investment in agro-logistics and processing/distribution facilities. As defined, the PAT is projected to be a highly favorable transformation scenario, although it is one which requires a significant shift in the role of the government to a mode in which it is ‘leading less but facilitating more’ within the agro-food system (World Bank 2016a). In this scenario, the government’s public investment focus would shift from primarily being infrastructure- centric to being more holistic and seeking to facilitate and better leverage private sector investments in innovation, human capital, service delivery, logistics, and value-adding facilities. Agriculture development in this scenario would also be market driven, with the government continuing to provide a strategic vision and selected core public goods and the bulk of service delivery would be shifted over to cooperatives, the private sector, and one or another type of PPP initiative. The government’s role, in this case, is a facilitator to enable private investment, incentivize improvements in farming practices, and facilitate the further development of modern agricultural value chains. It should be noted that the MKD’s agricultural transformation would be technically feasible at the farm and subregional levels; however, its success and efficiency would depend on market development, risk management, and the level of advanced technologies to be adopted. In driving value chain development, although the private sector will have a primary role, the role of the government is still essential in creating a conducive PPP platform. This is not difficult, but it requires a strong commitment and attention to support private sector development. Besides, along with the necessary policy reforms and investments, improvement in the institutional arrangements would be essential. It is necessary to strengthen the MKD center in Can Tho to coordinate the regional development projects, infrastructure works, and services. At the subregional level, the provinces would need to coordinate to share information and co-manage natural resources. At the community level, there would be a need for mechanisms to mobilize stakeholders in natural resource management, in upgrading basic infrastructure, and, where feasible, in collaborative modes of production. At the farm level, the government would need to ensure farmers have better access to land and capital to become professional producers. There would be a need to have a breakthrough in promoting the cooperative economy to connect companies and farms. Last but not least is a need for an effective monitoring system to gauge progress and incorporate lessons learned from field implementation to continually improve the quality of support and facilitation programs and to make policy adjustments, where warranted. A suggested transformation monitoring and assessment framework is included in Appendix 5. 2.1.2. Agricultural added value In terms of agricultural value added, it is estimated that the results for the PAT scenario would be nearly 2.5 times higher by 2030 than under the BAU scenario (Figure IV-5). This difference primarily comes from the changing composition of the region’s output with there being some reduction in rice output (a commodity with relatively low value addition) and considerable Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 84 Alternative Scenarios and Policy Implications increases in the output of higher-value aquatic products and of other crops. The perishable nature of many of these products makes it necessary that additional investments are made in logistics, especially cold chain logistics, as well as processing and distribution facilities. Much of these investments is expected to be undertaken by the private sector, although the upgrading of public market infrastructure in cities will enable the marketing of safe, high-nutrient foods. FIGURE IV-5. Comparison of agricultural added value between different scenarios PAU 595 BAU 224 Current 239 Unit: VND, trillions in 2019 price Source: Authors’ calculation. 2.2. Social outcomes 2.2.1. Employment The PAT scenario is projected to create much greater opportunities for remunerative employment than is the case currently or under the BAU scenario. Under all scenarios, direct employment in primary production would decline along with mechanization and the application of other labor-saving technologies. Still, under the PAT, the more diverse pattern of agricultural production would give rise to greater employment in the downstream functions of agricultural logistics and processing and food services. The total working days in the PAT scenario is estimated to increase by about 300 million, equivalent to about 1.2 million laborers (an increase of 190 percent compared to the BAU case) (Figure IV-6). This is mainly attributed to the reduction in the 3-crop rice and 2-crop rice areas and the increase of aquaculture (including rice-aquaculture) and higher-value crops. Also, as it assumes the adoption of advanced technologies, the PAT scenario would require more high-skilled laborers (Figure IV-7). As the MKD often has limited nonfarm job opportunities, the PAT scenario assumes improvements in human resources and would have a positive impact on the whole MKD region. Therefore, the transformation would generate more better-quality jobs and enhance labor skill formation for the region. This, in turn, would help trigger the reform and improvement of education and training systems in the MKD. Chapter IV - Sustainable agricultural livelihood transformation 85 FIGURE IV-6. Comparison of total required working days/year between different scenarios 700 629 Current PAT 600 506 500 Million days 400 331 300 264 200 100 67 56 45 22 - Total Production Services Processing Source: Authors’ calculation. FIGURE IV-7. Comparison of required employment between different scenarios High-tech agro-engineers High-skilled laborers PAT 13.9 PAT 833 BAU 7.0 BAU 333 Current 7.0 Current 333 Unit: Thousand labor Source: Authors’ calculation. 2.2.2. Food security In the PAT scenario, the rice cultivation area is projected to be reduced by one-third compared to the BAU scenario (that is, 3.1 million ha/year compared to 4.4 million ha/year), but still meet the national food security target. In the PAT scenario, since only the marginal rice crop/land is converted to other more productive systems and advanced technology will be adopted to increase productivity in the remaining, the paddy production volume is estimated to be more than 18 million tons (or 11 million tons of milled rice), which is enough to satisfy domestic consumption in the MKD, Central Highlands, and Southeast Region (about 3.2 million tons) and export targets (4.5 million tons). As such, the agriculture transformation not only will enhance the sustainability of the production systems but also does not compromise the government’s food security target. 2.2.3. Poverty reduction Poverty reduction is an important social goal that has to be achieved in any selected future scenario. Among all MKD subregions, the Coastal subregion is the poorest (multidimensional Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 86 Alternative Scenarios and Policy Implications poverty rate of 5.88 percent), has the highest inequality index, and has the highest migrating rate (net rate of 5.31 percent). The present study shows that the Coastal subregion has many livelihood models that can generate excellent economic efficiency. Therefore, if there are good policies, a conducive investment environment, and good markets, the Coastal subregion is likely to be able to accelerate its poverty reduction. This subregion can become the center for skilled laborers and high-income people. If proper advanced technologies and improved farm management are adopted, intensive shrimp farming can be a breakthrough for agricultural development in the Coastal subregion, particularly for enterprises and farmers with financial and technical capabilities. 2.3. Environment In the PAT scenario, the selected livelihood models are the ones that have a good environmental and natural index (see Chapter III). They are environment friendly, ecologically balanced, climate adaptive, and resilient. Available data show that intensive rice farming in the MKD generates about 0.5 ton of GHG emissions for each ton of paddy rice produced. If 500,000 ha of rice land will be diversified to other environmentally friendly crops, it will help reduce about 1.7 million tons of CO2 equivalent per year. Moreover, if GAP (that is, alternate wet and dry irrigation) is introduced, it will help further reduce half of the GHG emissions in the remaining rice areas (about another 12 million tons of CO2 equivalent each year). These will reposition Vietnam’s agriculture to be a cleaner and greener sector. 2.4. Agricultural transformation in relation to GHG emission reduction GHG emissions from Vietnam’s agriculture sector has increased in the past decades, especially during 1990–2010. Agricultural expansion in this period caused large-scale deforestation and land-use changes (mainly for tree crop development in the Central Highlands and shrimp farming in the MKD). The past decade witnessed accelerated agricultural intensification with the overuse of agrochemicals. In the MKD, extensive large dike systems were constructed to allow the cultivation of a third rice crop. According to the Vietnam Third Biennial Updated Report to the United Nations Framework Convention on Climate Change in 2020, total gross and net (after subtracting carbon sequestration/absorption) GHG emissions of the agriculture, forestry, and other land use (AFOLU) in 2016 were around 81.5 million tons of CO2e and 44.1 million tons of CO2e, respectively (Table IV-2). These data show that while forest land contributed the highest absorption (around −54.6 million tons of CO2e), rice farming was the largest emitter (around 49.7 million tons of CO2e), followed by livestock (around 18.5 million CO2e). As Vietnam ratified the Paris Agreement on Climate Change in 2016, Vietnam’s Nationally Determined Contributions (NDCs) included a commitment to reducing GHG emissions by 8 percent between 2021 and 2030 compared to BAU levels using domestic resources, and up to 25 percent conditional on receiving international support. The Action Plan to Implement the Paris Agreement on Climate Change is outlined in Decision 2053/QD-TTg dated October 28, 2016, which includes activities for adaptation and mitigation in the agricultural sector. To achieve the government’s NDC target, it is crucial to prioritize and address the causes of GHG emissions in the key economic sectors to bring them back to the sustainable and green growth path. Resolution 120 /NQ-CP on sustainable and climate-resilient development for the MKD in 2017 and Decision 34/NQ-CP on national food security to 2030 in March 2021 provide a Chapter IV - Sustainable agricultural livelihood transformation 87 breakthrough opportunity for the region and for many farmers as they call for the reduction of the protected rice lands by some 500,000 ha. If effectively implemented, an assisted process of land-use diversification would result in higher incomes for farmers, much reduced levels of water pollution, and a reduction in the region’s GHG emissions from agriculture, contributing to achieving the NDC target. TABLE IV-2. GHG emissions/removals of the AFOLU sector in 2016 (thousands of tons of CO2e) IPCC Total net GHG source/sink categories CO2 CH4 N2O code emissions 3 AFOLU -37,489.34 66,544.64 15,014.44 44,069.75 3A Livestock 15,553.10 2,960.27 18,513.37 3A1 Enteric fermentation 12,421.74 12,421.74 3A2 Manure management 3,131.36 2,960.27 6,091.63 3B Land -39,491.24 -39,491.24 3B1 Forest land -54,657.78 -54,657.78 3B2 Cropland 3,637.60 3,637.60 3B3 Grassland 1,383.64 1,383.64 3B4 Wetlands 1,046.90 1,046.90 3B5 Settlements 1,919.14 1,919.14 3B6 Other land 7,179.27 7,179.27 Aggregate sources and non-CO2 3C 2,001.90 50,991.54 12,054.16 65,047.60 emission sources on land 3C1 Emissions from biomass burning 1,298.52 325.61 1,624.13 3C2 Liming 565.79 565.79 3C3 Urea application 1,436.11 1,436.11 Direct N2O emissions from managed 3C4 7,754.11 7,754.11 soils Indirect N2O emissions from managed 3C5 3,752.55 3,752.55 soils Indirect N2O emissions from manure 3C6 221.90 221.90 management 3C7 Rice cultivations 49,693.02 49,693.02 Source: Vietnam Third Biennial Updated Report 2020. Note: Negative value (−) shows the GHG absorption of sinks. When considering climate-smart cropping systems for the MKD, to be sustainable, it is important to consider all three objectives: making technologies available to produce food sustainably and bring higher incomes for producers, adapting to changing and fluctuating climate and environment, and reducing GHG emissions. When selecting farming systems and technologies, it is essential to consider all options to improve the net GHG emissions balance via three levers: less N2O, CH4, and CO2 emissions; more carbon storage; and green energy production (Debaeke, Pellerin, and Scopel 2017). Since N2O emissions are linked to the application of inorganic fertilizers, to reduce them, applying GAPs and shifting from inorganic to organic fertilizers together with proper soil management (that is, increasing the proportion of legumes in the rotation) will be an option. The most promising options for reducing CH4 emissions in paddy fields include intermittent irrigation or Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 88 Alternative Scenarios and Policy Implications alternate wetting and drying (AWD) irrigation techniques. Options to mitigate CO2 include storing more carbon in soil and biomass through promoting no tillage and eliminating the burning of rice straw. International and local experiences in applying these technologies are available and ready for scaling up on large scales if backed by appropriate policies. 2.4.1. Promoting replacement of 3-rice and 2-rice with climate-smart cropping systems Findings from Chapter III show that about 760,000 ha of 3-rice and 2-rice cultivation areas in the MKD are currently no longer suitable for rice farming due to the impact of climate change. They include 295,000 ha of 3-rice and 2-rice areas in the Upper subregion, 400,000 ha of 3-rice and 2-rice areas in the Middle subregion, and 65,000 ha of the 2-rice crop in the Coastal subregion. These marginal rice lands should be replaced with other climate-smart cropping systems without adversely affecting the country’s exceptionally strong food security status. Experiences from an ongoing World Bank-financed project45 showed that farmers in Dong Thap Province (the Upper subregion) could earn higher alternative incomes by replacing the third rice crop with a fish farming crop or trapping/ harvesting wild fish entering the rice field during the flooding season (that is, income from fish farming and wild fish harvesting is 15–35 percent higher than that of rice). Regarding GHG emissions, according to GHG measurement data from the Vietnam Sustainable Agriculture Transformation Project (VnSAT) financed by the World Bank, each hectare of rice crop in the MKD generates 3–5 tons of CO2e per crop (or 4 tons of CO2e per ha per 90-day crop, on average). According to Mathewa et al. (2017), grass and cereals exhibit the highest potential for carbon sequestration because they have high biomass production and carbon accumulation. Other studies show that fruit trees also have a high potential for carbon sequestration (Janiola and Marin 2016), and among economic trees, nut trees sequester more carbon than fruit trees (Hou et al. 2019). Applying crop rotation, intercropping, and proper soil management (that is, replacing chemical fertilizers with organic fertilizers) will also help increase soil microbial biomass and enhance carbon sequestration (Wang, Li, and Alva 2010). Assuming 50 percent of these marginal rice areas of 2-rice and 3-rice would shift to other lower-carbon farming systems (that is, with 50 percent or 70 percent less GHG emissions), it would help cut down 1.5–4.2 million tons of CO2e each year (Table IV-3). It is suggested that further research be conducted to recommend suitable climate-smart cropping systems for each MKD subregion. 45 Mekong Delta Integrated Climate Resilience and Sustainable Livelihoods Project (MKDICRSL) financed by World Bank. Project Progress Reports. Chapter IV - Sustainable agricultural livelihood transformation 89 TABLE IV-3. GHG emissions in different diversification scenarios Diversifying to less GHG emission cropping Upper Middle Coastal Total systems Region Region Region 30% 50% 70% 100% 3-rice area (ha) -133,900 -254,700 0 -388,600 (116,580) (194,300) (272,020) (388,600) 2-rice area (ha) -161,200 -147,100 -64,800 -373,100 (111,930) (186,550) (261,170) (373,100) Total rice crop can be reduced -295,100 -401,800 -64,800 -761,700 (ha) Est. GHG reduction in 3 regions by (914,040) (1,523,400) (2,132,760) (3,046,800) reducing 1 rice crop (tons CO2e) Est. GHG reduction in 3 regions by (1,828,080) (3,046,800) (4,265,520) (6,093,600) reducing 2 rice crop (tons CO2e) Source: Authors’ calculation. 2.4.2. Promoting CSA and GAPs Climate-smart agriculture (CSA) technologies and GAP present opportunities to address climate change challenges, improve farm income, and reduce environmental pollution, thereby enhancing agriculture competitiveness. There are so many technologies and approaches around the world on CSA and GAP. Some of the most frequently suggested practices are those related to irrigation management in almost all crop systems (that is, rice, coffee, tea, citrus, cashew, maize, rice, pepper, and so on). In rice farming, the most applicable practices include applying water-saving technologies and AWD techniques. Rice lands in the MKD have become increasingly degraded. leading farmers to overuse fertilizer to maintain yield levels. Various studies showed that most rice farmers in the MKD applied up to 20– 30 percent more fertilizer than recommended. Overuse of fertilizer has been costly to the farmers and the environment. It is estimated that every year, around 140,000 tons of N, 82,000 tons of P, and 66,000 tons of K are wasted due to overfertilization in rice farming in the MKD. This has a financial cost of US$150 million per year (Nguyen 2017). Similar problems are found in the use of other inputs such as pesticides and water for irrigation. Most farmers used pesticides and water for irrigation about 20–30 percent higher than the recommended levels. Overuse of fertilization, pesticide, and water have resulted in high environmental costs, including water and soil pollution. To address these and other unsustainable practices, efforts have been made by the government in the past 10 years to promote CSA and GAP through the ‘3 Reductions-3 Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 90 Alternative Scenarios and Policy Implications Gains’ program (3R3G),46 followed by the ‘1 Must do-5 Reductions’ program (1M5R).47 While the 3R3G has been initiated by MARD in the early 2000s based on the integrated pest management program (IPM), the 1M5R was developed in collaboration with IRRI and first piloted in seven MKD provinces (Dong Thap, An Giang, Kien Giang, Long An, Tien Giang, Soc Trang, and Can Tho) in the last two years (2012–2013) of the World Bank-financed Agriculture Competitiveness Project (ACP). The ‘1M5R’ pilot model helped reduce seeding rate by 29–50 percent compared to the control group, inorganic fertilizer by 22–50 percent, water use by 30–50 percent (by applying AWD techniques), pesticide applications by 20–33 percent, and production costs by 22 percent (VND 4 million per ha) while increasing yield by 5.2–7.9 percent and profits by 29–67 percent (Table IV-4). These encouraging results were later scaled up under the World Bank-financed follow-on VnSAT project in the core rice zones of eight MKD provinces (Long An, Tien Giang, Can Tho, Kien Giang, An Giang, Dong Thap, Soc Trang, and Hau Giang) from 2015 to 2022.48 In the VnSAT, similar positive results have been reported (Table IV-5). After five years of implementation from 2015 to 2021, the VnSAT project has supported some 200,000 farmers adopt 3R3G and 1M5R over 175,000 ha. Under both the ACP and VnSAT, it was found that project beneficiaries achieved significant GHG emissions reductions from adopting the improved farming models than those in the control group. Results from the ACP conducted in Tien Giang and Can Tho Provinces in the 2013 summer–autumn crop found that GHG emissions from the 1M5R model were reduced by 5.4–6 tons of CO2e per ha per crop compared to the traditional practices (equivalent to 60–70 percent of the total). The 2013 autumn–winter crop (the second crop) showed a greater reduction (reducing 15–19 tons of CO2e per ha per crop, equivalent to 85–95 percent of the total). Under the VnSAT project, GHG emissions from 1M5R project beneficiary farms in 2020 were found to have been reduced by 20 percent compared to non-project farms.49 Using the Food and Agriculture Organization (FAO) of the United Nations Ex-Ante Carbon-balance Tool (EX-ACT) for calculations, it is estimated that the project helped reduce about 1.5 million tons of CO2e (about 8.5 tons of CO2e/ha/year on average). Based on conservative assumptions that each rice crop applying the 1M5R would reduce 8 tons of CO2e/year (two crops) and the 1M5R model would be applied to 500,000 or 1,000,0000 ha of the remaining core rice zones in the MKD, the improved farming practices would mitigate GHG emissions about 4–8 million tons of CO2e each year. 46 3R3G: Reducing seeding rate, fertilizers, and pesticides; Gaining higher yield, higher quality, and higher income. 47 1M5R: Must use certified seed; Reducing seeding rate, fertilizers, pesticides, water use (through applying AWD techniques), and post-harvest losses. 48 Both ACP and VnSAT are financed by the World Bank. Both projects received TA from IRRI. The ACP received CAD 3.3 million grant funds for the TA. 49 There are small differences in GHG emission reduction reported under the ACP and VnSAT, which could be due to the spillover effects of the ACP. In fact, farmers in the VnSAT areas inherited the results from the ACP, and most of them have partially adopted some of the 1M5R technique. Chapter IV - Sustainable agricultural livelihood transformation 91 TABLE IV-4. Summer–autumn rice crop in Tien Giang in 2013 Items Control 1m 5R+AWD Reduction/Increase Seed (kg/ha) 170 120 -29% Fertilizer (kg/ha) • Urea 220 170 -23% • DAP 100 100 0% • KCI 100 100 0% Pesticide spraying (times) 6 4 -33% Water pumping (m3/ha) 3,250 2,250 -31% Water pumping (times) 8 5 -37% Total costs (VND mil./ha) 19,755 16,033 -19% Break-even 3,314 2,557 -23% Revenue (VND mil./ha) 32,787 34,503 +5% Gross margin (VND mil./ha) 13,032 18,467 +42% Returns to capital (%) 66 115 +74% Source: World Bank (2014) Note: DAP = Di-ammonium phosphate. TABLE IV-5. Benefits of 1M5R adoption under the VnSAT Net benefits (%) of the VNSAT-1M5R treatment compared with farmer practices DSR -1M5R DrumS -1M5R MecT -1M5R Winter-Spring Nitrogen-use efficiency -24.5 (0.23) -25.1 (0.20) 3.5 (0.23) Phosphorus-use efficiency 15.2 (0.69) 15.8 (0.65) 26.0 (0.69) Potassium-use efficiency -8.5 (0.38) -8.1 (0.32) 21.4 (0.38) Reduced pesticide application 5.8 (0.02) 5.8 (0.02) 37.5 (0.02) Labor productivity 9.4 (0.39) 11.8 (0.34) 6.7 (0.39) Grain yield 10.1 (0.01) 10.1 (0.01) 11.3 (0.01) Energy efficiency 7.8 (0.22) 6.8 (0.16) 10.6 (0.22) Reduced GHGE 26.6 (8.28) 26.7 (8.20) 32.0 (8.28) Net income 7.0 (1.61) 8.7 (1.16) 13.2 (1.61) Summer-Autumn Nitrogen-use efficiency 18.8 (0.10) 12.9 (0.11) 18.8 (0.12) Phosphorus-use efficiency 24.6 (0.53) 18.9 (0.51) 26.3 (0.66) Potassium-use efficiency -40.5 (0.28) -50.7 (0.30) -37.8 (0.35) Reduced pesticide application 46.2 (0.01) 38.5 (0.01) 46.2 (0.01) Labor productivity 26.5 (0.03) 22.1 (0.31) -29.9 (0.21) Grain yield 2.3 (0.00) -4.0 (0.00) 5.1 (0.00) Energy efficiency 1.3 (0.11) -7.9 (0.16) 4.3 (0.14) Reduced GHGE 29.9 (7.09) 29.8 (7.06) 30.2 (7.03) Net income 17.6 (0.97) 6.9 (7.06) 9.0 (1.35) Source: IRRI, 2019. Note: DSR = Direct seeded rice; DrumS = Drum seeders; GHGE = GHG emission; and MecT = Mechanical transplanters. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 92 Alternative Scenarios and Policy Implications 2.4.3. Reducing GHG emissions from other sources in the rice value chain In the rice value chain, three other major sources generate high GHG emissions, including by-products management (i.e., rice straw), water and energy use, and rice processing (i.e., milling efficiency and post-harvest losses). It is estimated that about 53 million tons of rice straw are produced each year in Vietnam, of which around 29 million tons come from the MKD (Tran et al. 2021), and up to 80–90 percent of farmers in the MKD burn rice straw after paddy harvest (Nguyen 2017). Open burning of crop residues has been common because it is a cheap and quick way to eliminate waste and prepare land for the next crop (short turnaround time between rice crops in 2–3 cropping systems). However, this practice directly contributes to air pollution and human health problems as it emits gases such as sulfur dioxide (SO2), oxides of nitrogen (NOx), carbon dioxide (CO2), carbon monoxide (CO), black carbon (BC), organic carbon (OC), methane (CH4), volatile organic compounds (VOCs), non-methane hydrocarbons (NMHCs), ozone (O3), aerosols, etc. (Nguyen 2017). In addition, burning crop residues causes increased soil temperature, decreased microorganisms, and reduced organic matter in the soil. On average, open burning of 1 kg of dry straw would emit 1.4–2 kg CO2e (Arai et al. 2015). Using conservative assumptions for a quick calculation that open burning of rice straw in the MKD is around 80 percent, the moisture content of rice straw in the rice field is about 30 percent, and burning 1 kg of dry straw would generate about 1.5 kg of CO2e, the open burning of 29 million tons of rice straw in the MKD would generate some 24 million tons of CO2e each year. If half of this rice straw amount could be reduced by alternative uses or recycled for production purposes (with less GHG emissions, that is, reducing GHG emissions by 30 percent or 50 percent compared to open burning), the improved practices could help the MKD reduce GHG emissions up to 6–10 million tons of CO2e each year. There are alternatives for environmentally and economically better management of rice straw than open burning (Nguyen et al. 2020). Rice straw collection is only economically viable when the rice straw collection is done by machine (that is, balers). In the VnSAT project, farmer cooperatives were provided matching grants to procure balers to collect rice straw for resale. In those areas, the open burning of rice straw by targeted farmers almost eliminated. Another alternative to rice straw burning is soil incorporation using fungal inocula (Trichoderma) to speed up the decomposition rate of rice straw right in the rice fields and to incorporate it back into the soil. The Mekong Delta Integrated Climate Resilience and Sustainable Livelihoods Project (MKDICRSL) project provided training to farmers in An Giang Province. Local farmers accepted this technique well as it helped reduce applications of inorganic fertilizers by about 15–20 percent. However, this soil incorporation technique requires sufficient time between crops for decomposition and careful attention to avoid anaerobic conditions after soil incorporation as it will increase CH4 emission. Another option is composting rice straw with animal manure and enzymes to produce organic fertilizers or soil conditioners or using rice straw for mushroom cultivation. All these techniques are locally available for scale-up, but further in- depth research is needed to maximize their benefits and minimize impacts on GHG emissions. In agricultural value chains, water and energy efficiency are critical factors affecting production costs. In rice farming in the MKD, irrigation is mainly done by pumping. Field results from the ACP and VnSAT projects showed that rice farmers could save up to 20–30 percent of their Chapter IV - Sustainable agricultural livelihood transformation 93 pumping cost by adopting the AWD practices without affecting rice yield and quality. In aquaculture, the results from the World Bank-financed Coastal Resources for Sustainable Development Project (CRSD)50 showed that shrimp farmers could save up to 90 percent of water use by applying closed water recirculation systems. These improved farming practices allow better disease risk management and environmental pollution, reducing energy costs, thereby reducing GHG emissions. Although energy consumption by the agriculture sector currently is only about 1.2 percent of the total energy consumption,51 promoting energy savings and renewable energy (that is, solar energy) is essential to all agricultural value chains to reduce production costs and GHG emissions further. A survey conducted in 2018 estimated physical product losses of 32 percent for fruits and vegetables, 14 percent for meat, and 12 percent for fish in major supply chains servicing HCMC.52 If such rates of product loss were representative of the national picture, then total national losses of such high-value perishable foods would amount to US$3.9 billion per year, equivalent to 12 percent of Vietnam’s agricultural GDP. In rice value chains, post-harvest losses are estimated at around 13 percent, with the highest losses occurring in harvesting, drying, and storage processes (Table IV-6). The main causes include early or delayed harvesting due to the unavailability of combine harvesters, the lack of drying and storage facilities at the community level, and the double milling practices in which rice is de-husked by small mills and then the brown rice is sold to larger millers and provincial food companies (that is, state-owned food companies) for final milling. Due to the lack of drying facilities at the community level, when rice is de-husked, it still has a high moisture content (that is, > 20 percent), resulting in large physical or quality losses. After a few months of storage, milled rice often turns yellow. When it comes to the final milling and polishing, another TABLE IV-6. Comparing post-harvest loss round of processing losses occurs. This multiple in rice between Vietnam and Thailand milling practice substantially affects the quality Vietnam Stage Thailand and standardization of rice, which affects value (MRD) and the opportunities to export high-quality rice. Harvesting 2-3% 2% It is estimated that if post-harvest losses in rice Distribution 0.9% 0.4% in the MKD were reduced from 13 percent to 8 Drying 4.2% 1.7% percent, the MKD could save about 1.2 million Storage 2.6% 1.2% tons of milled rice (or 1.8 million tons of paddy) Milling 3% 2.3% and reduce GHG emissions approximately by 1 Total 13.7% 7.6% million tons of CO2e each year.53 Source: Khoi (2017). 50 CRSD ICR, 2019. 51 Currently, the energy sector is the largest GHG emitter in Vietnam, around 205 million tons of CO2e in 2016 (Vietnam Third Biennial Updated Report, 2020). 52 https://www.cel-consulting.com/single-post/2018/08/10/Food-Losses-in-Vietnam-the-shocking-reality. 53 The estimate is based on data from VnSAT that in the MKD, on average, producing 1 ton of paddy will generate about 0.5 ton of CO2e. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 94 Alternative Scenarios and Policy Implications In summary, to fully realize the benefits of the agricultural transformation, all possible options should be considered to also minimize GHG emissions and make the sector greener. Comprehensive and integrated intervention programs should be launched and implemented, including: (a) replacing unsuitable and unsustainable 3-rice and 2-rice with low-carbon climate-smart cropping systems adapted to changing and fluctuating climate and environment; (b) promoting CSA and GAP in the remaining core rice areas to mitigate GHG emission; and (c) optimizing the use of agricultural by-products, improving water and energy efficiency, and reducing post-harvest losses to reduce GHG emissions. Such programs would require a strong role of the public sector in setting standards and improving policy and regulations and facilitating greater private sector investment and provision of technical and other services. It will also require the mobilization of international TA and financial resources. Based on the above discussion, if all these measures are included, it is quite possible to reduce GHG emissions from the MKD from 14–25 million tons of CO2e each year, contributing to meeting the NDC target. 3. Policy options and imperatives The recommendations in this section are mainly based on this present context and a vision until 2030. Looking beyond 2030 will need a comprehensive assessment of the growth trajectory the Vietnam would take, and what kind of land use plans this would entail. The government’s policies play a crucial role in creating incentives for efficient transformation in the MKD. In this process, the government’s role should be facilitation rather than implementation, investing more selectively and focusing on core public goods and services and the enabling environment to leverage investments by farmers and the private sector. Following these principles, apart from general national policies for the agricultural sector, it is recommended that the government focus on the following policies for the MKD. 3.1. Land policy To achieve a scale of land-use change closer to the optimum for effective agricultural transformation in the MKD, the government should adopt a policy that promotes more flexible and diverse land uses and presumes that adjustments will be not only permitted but facilitated. This is because the agricultural transformation in the MKD is tied critically to matters of land-use change in the face of changes in natural conditions (including climate change), demographics, diets, and socioeconomic conditions. Restrictions will be applied only in special circumstances which will be clearly defined by the government through its land-use planning instrument. Yet, neither the national policy target of 3.5 million ha of rice land nor the scenarios provided in this report provide a blueprint for guiding a progressive and sustained process of land-use change from the current to one which is more optimal from the perspectives of households, communities, the agricultural sector, and the regional and national economies. Both despite and because of concerns about climate change, it is clear that the optimal shift in land use is not just a modest one but a substantial one. The economy of the region can be greatly improved if the total planted area devoted to rice is reduced by about 25–40 percent. This scale of adjustment, even if it were to occur over the span of a decade, cannot take place Chapter IV - Sustainable agricultural livelihood transformation 95 if local governments do not have decision-making authority over which ‘rice lands’ can and cannot be converted to alternative agricultural uses. Therefore, the central government should develop clear guidelines on facilitating more flexible and diverse land uses to allow local governments to make decisions on land-use changes based on comprehensive studies and analysis of market demands and comparative advantages, land suitability, climate change, and socioeconomic factors. Along with the flexible land-use policy, it is crucial to support the agricultural transformation by reorganizing agricultural production systems through land consolidation and cooperative farming models. Over the past decades, the government has supported cooperative or collaborative farming via the ‘small farm large field model’. This has typically involved 30–100 farmers joining their fields and doing either synchronized or centrally managed land preparation, planting, water management, harvesting, and other functions. This cooperative farming is an important precondition to enable farmers to adopt GAP and AWD. Upward of 10 percent of the rice- growing areas in the MKD already follow this model. Support for the scaling-up of the cooperative and other clustering approaches would be essential in the future, especially as many older farmers look to reduce their involvement in agriculture. Also, the government should facilitate the development of a more active agricultural land lease market, which will enable individual entrepreneurial farmers, cooperatives, and agribusiness companies to expand their farm size. Various approaches for developing land lease markets, involving local government agencies, brokerage services, and financial institutions, have been adopted successfully in other countries, such as in China, Japan, and Republic of Korea, with complementary efforts also made to facilitate the entry of a new generation of younger farmers into farming. It is expected that the upcoming decree on land consolidation (under preparation) could provide the needed legal framework for these initiatives. 3.2. Fostering private investment and private provision of agricultural services For effective agricultural transformation, the government should aim to facilitate and better leverage private sector investments in MKD agriculture and agricultural value chains and the private delivery of agricultural technical, management, and other services. And over time, especially as networks of agricultural cooperatives and associations of agro-related industries become stronger, the government should explore opportunities for regulatory co-management with such entities, especially in areas related to food safety, biosecurity, and environmental protection. The government should provide incentives to the private sector in priority areas that will facilitate effective agricultural transformation through financing missing public goods and infrastructure. This should be considered on a case-by-case basis as there is no universal formula for all cases. The government may allow large corporations to invest in common shared infrastructure and then attract other enterprises to invest in their own infrastructure and services. Local small and medium enterprises (SMEs) are encouraged to invest in production support centers for key specialized production areas. At the same time, large companies can be encouraged to invest in regional rice clusters. The government should gradually withdraw its direct investment and market interventions in production and supply chains and promote the socialization of research and advisory services. Local government agencies can also act to promote more active and competitive markets for the provision Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 96 Alternative Scenarios and Policy Implications of mechanization and other agricultural services by accrediting and incorporating private providers into local programs, encouraging links between them and farmer organizations, and ensuring that such service providers are able to participate in existing small business promotion schemes. Agencies can create an information platform to enable farmers to rate and comment on the quality of services provided to them. Then, the PPP investment scheme should continue for production support centers, processing zones, logistics centers, and so on. In this period, the government should completely phase out its direct production activities and agro-trade control (by completing the privatization of state- owned enterprises [SOEs]) and open the research and innovation systems to the private sector. The government will act more effectively as a facilitator by providing preferential policies and enabling the business environment at the highest level. Greater efforts to further support contract farming or similar links between farmer groups and agribusiness companies are needed. Useful functions which the Ministry of Agriculture and Rural Development (MARD) or local government agencies can play might include: (a) the development of ‘codes of practice’ for contract farming relationships to help guide company and farmer organization behavior; (b) the dissemination of analytical tools for assessing the feasibility of such arrangements and for monitoring/evaluating their performance and guidelines for problem/conflict resolution in such relationships; (c) the development of lines of credit with a focus on supporting companies and farmers involved in such partnerships; and, possibly, (d) a challenge fund, utilizing matching grants, to incentivize especially innovative schemes, perhaps involving new products or in especially challenging locations. 3.3. Stimulating innovation The transformation of MKD agriculture is going to require a much more robust system for agricultural innovation as well as the strengthening of the knowledge base of farmers and cooperative managers. In the coming years, the MKD Center, which is being established by the Ministry of Natural and Environment (MONRE) and located in Can Tho, is expected to serve as an information hub and integrated data system for the MKD. Unfortunately, it could not be passed onto a research consortium as per its original plan. However, establishing an innovation hub for the MKD region is essential and needs initial public support. The MKD innovation hub can be established as part of a university or a capable private sector company operating in the region. It will support research and innovation programs for the MKD and promote research, training, and the application of emerging technologies to adapt to climate change and reduce agriculture’s environmental footprint (including GHG emissions) at the farm, community, landscape, and value chain levels. Both public and private funds can be mobilized. The program could focus first on seeds, breeds, and technologies to improve input use efficiency (that is, especially for water and energy); reduce post-harvest losses; and optimize the use of agricultural by-products (that is, rice straw). It is recommended that the government foster reform in research management mechanisms by simplifying procedures and granting incentive policies for the private sector to participate in research and advisory and other technical services and technology transfer. Links among research institutes, universities, extension organizations, businesses, cooperatives, and local authorities should also be promoted. In pursuing these efforts, lessons can be learned from an array of agricultural incubation Chapter IV - Sustainable agricultural livelihood transformation 97 centers and models which have been applied elsewhere. Agribusiness incubation has emerged over the past 15–20 years to stimulate commercial agriculture and transform comparative advantages in commodity markets into competitive advantages in differentiated product markets. A defining characteristic of agribusiness incubators is that they directly engage with startups to help them grow, usually offering them a range of advisory and business development services geared to improving firms’ competitiveness and access to markets. Government involvement in agribusiness incubation varies significantly. Incubators enjoy different degrees of financial and political autonomy. Many are non-profits and start out with the public sector and other external sources of funding from which they wean themselves to varying degrees. Incubators are generally able to cover some, if not all, of their operating costs by charging firms for access to their services and facilities (by charging consulting, business development, marketing, franchising, rental, and other fees). Over time, certain incubators invest in the firms they incubate as well as in their intellectual property, allowing them to share in their profits and royalties. 3.4. Strengthening human resources The current human resources capacity in the MKD remains weak compared to those in other regions of Vietnam. This problem will become more serious in the context of MKD’s agricultural transformation in which science and technology are expected to be key driving forces. The government should develop a comprehensive capacity-building policy to strengthen human resources for the MKD. The policy should aim at developing and implementing professional training programs for farmers and managers of their organizations and other key actors in agricultural value chains, focusing on sustainable farming practices, business management skills, food safety standards, and application of digital technologies to improve production efficiency and reduce environmental footprint (including GHG emissions). The training programs should be closely linked with the transformation pathway in localities. There should also be a policy to attract young experts and intellectuals to work in research and training centers and public services in the agricultural sector in the MKD. Mainstreamed in such efforts, there should be programs to promote gender equality in skills development and opportunities in remunerative farming and employment (in agricultural value chains and nonfarm sectors). 3.5. Setting targets and promoting low-carbon agricultural value chains To achieve the NDC target, GHG emission reduction targets should be set and implemented for all key agricultural value chains, and it should first start with the rice value chain in the next 10 years as it has the greatest potential in cutting the GHG emission. Comprehensive and integrated intervention programs should be launched and implemented, including (a) replacing unsuitable and unsustainable 3-rice and 2-rice with low-carbon climate-smart cropping systems adapting to changing and fluctuating climate and environment, (b) promoting CSA and GAP in the remaining core rice areas to mitigate GHG emission, (c) and optimizing the use of agricultural by- products, improving water and energy use efficiency, and reducing post-harvest losses to reduce GHG emissions. Such programs would require a decisive role of the public sector in setting standards and improving policy, legislation, and the enabling environment and greater private sector involvement and investment in innovation, agricultural services, and marketing strategy. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 98 Alternative Scenarios and Policy Implications 3.6. Market vision and strategy Market development and trade promotion for agribusinesses are an essential part of the MKD’s agricultural transformation. There should be a comprehensive set of policies targeting domestic and export markets that aim for diversification and high-value-added products and low- carbon branding. In addition to traditional export markets, rice export should gradually shift to higher- end markets. The sector should also pay closer attention to the domestic market to supply high-quality and specialty rice for medium- and high-income consumers, especially in Hanoi and Ho Chi Minh City and medium-quality rice for industrial kitchens. It is also suggested that rice companies in the MKD be linked to traders in big cities to promote MKD low-carbon rice. It is also essential to develop a system to monitor the rice supply and demand. For fruit commodities, the aim should be retaining the domestic market by quality and safe food through supermarkets and convenience stores. At the same time, thorough market research aiming at potential countries under the Comprehensive and Progressive Agreement for Trans-Pacific Partnership (CPTPP) and EU-Vietnam Free Trade Agreement (EVFTA) should be conducted. For aquaculture commodities, frequent monitoring of the export market should be undertaken to respond to changing policies. For livestock products, the sector should aim to satisfy the demand of the regional market and partly the Ho Chi Minh City market by stable quantity, safe food, and traceability products. The rice sub-sector should develop a strategy to penetrate and compete in high-end markets under Vietnam’s low-carbon brand names. This will provide signals for the change to high-quality rice varieties in the field and processing technology application. For fruits, producers are encouraged to increase the proportion of processed products and strictly control food safety to take advantage of tariff reductions in potential markets. For fishery products, the sub-sector should develop geographical indications and trademarks, especially for eco-farming products, to raise the added value. Livestock producers are encouraged to explore niche export markets for specialty and high-value products such as bird nests, honey, duck eggs, and so on. Geographical indications and trademarks should be developed to expand to the high-end markets to improve value addition. 3.7. Food safety and biosecurity Food safety and biosecurity should be given priority as this will help protect human and animal health, facilitate market access, raise value addition, and protect the environment. The government should renovate and improve the quarantine system to ensure biosafety and hygiene for imported and exported goods. It should also develop fully enforced anti-counterfeiting regulations to effectively control the trade of counterfeit production inputs (fertilizers and pesticides). In addition, it is recommended to build an independent management system or reorganize the existing system, thus ensuring accountability in quality management and food safety management among sectors and levels and strengthening coordination among ministries and sectors at all levels. This requires shifting from the current consignment-based and product-based inspection approaches to value chain- based traceability for quality, food safety, and hygiene management. This also requires establishing a systematic database system and strengthening inter-sector and inter-locality coordination in the management and traceability of agricultural, forestry, and fishery products. It is essential to promote the participation of various actors and society in quality certification and recognition. Chapter IV - Sustainable agricultural livelihood transformation 99 The government should move toward the judicious sharing of responsibilities—among government levels and private sector entities—and proactive, risk-based approaches to protecting consumer safety. The government should promote a preventive food safety management approach by better managing the whole value chains by food business operators and consumers rather than the reactive approach based on the inspection of final products by the government’s entities. This model is considered as co-regulation, which relies on greater private sector involvement to manage the risks. Technical and other forms of support from the public sector will be needed to enable private food operators (especially SME food processors and informal-sector food distributors) to improve their safe food management practices. Adopting a risk-based approach will help provide a set of clear foci for public interventions. 3.8. Institutional strengthening Policy innovations should go with improved institutional arrangements at all levels. It is critical for the MKD to reach an economic scale for commercial production and efficiently coordinate resource allocation and use, especially for investment capital, land, and water resources. There should be breakthrough solutions to promote a cooperative economy to improve value chain efficiencies, especially where market intermediation is highly fragmented. It is suggested that the central and local governments facilitate the setup of professional groups to support the cooperative establishment and its access to policies and value chain link. At the regional level, the government should set up a joint regional database for decision-making at the MKD Center in line with the Prime Minister’s Decision 1619/QD-TTg dated October 20, 2020, on developing a multi-sectoral, integrated database system for the MKD and Decision 825/QD-TTg dated June 12, 2020, on establishing a multi-stakeholder Regional Coordination Council (RCC) to support and promote sustainable livelihoods and economic transformation of the region and enhance the efficiency of interprovincial investments, plans, and policies. The government should also strengthen the cooperation and coordination among existing institutions, including central and local governments and private sector representatives, to coordinate delta-level investment items, water regulation, production planning for key commodities, and research and innovations. At the regional level, the provinces should promote inter-provincial links through a coordinating board to share information and co-manage natural resources (that is, freshwater use, flood control, salinity prevention, natural disaster prevention, value chain management, and so on). 3.9. International cooperation There are relevant global experiences, lessons learned, and operational models that Vietnam can learn from to enhance the MKD’s transformation effectiveness. International cooperation will remain important to support the MKD’s agricultural transformation in the coming decades. In addition, the government should develop incentive policies to encourage farmers and producers to seek international TA and global funds to support the adoption of low-carbon and green development and CSA. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 100 Alternative Scenarios and Policy Implications Bibliography Alexandratos, N., and J. Bruinsma. 2012. “World Agriculture towards 2030/2050: The 2012 Revision.” ESA Working Paper 12-03, FAO, Rome. Anthony, E., G. Brunier, M. Besset, M. Goichot, V. Nguyen, and P. Dussouillez. 2015. “Linking Rapid Erosion of the Mekong River Delta to Human Activities.” Scientific Reports 5:14745. doi:10.1038/ srep14745. Arai, H., Y. Hosen, N. P. H. Van, T. T. Nga, N. H. Chiem, and K. Inubushi. 2015. “Greenhouse Gas Emissions from Rice Straw Burning and Straw-Mushroom Cultivation in a Triple Rice Cropping System in the Mekong Delta.” Soil Science and Plant Nutrition 61 (4): 719–735. doi:10.1080/00380768.20 15.1041862. Baran, E., E. Guerin, and J. Nasielski. 2015. Fish, Sediment and Dams in the Mekong: How Hydropower Development Affects Water Productivity and Food Supply. Penang, Malaysia: WorldFish, and CGIAR Research Program on Water, Land and Ecosystems (WLE). http://hdl.handle. net/10568/71117. Baran, E., T. Jantunen, and C. Chong. 2007. Values of Inland Fisheries in the Mekong River Basin. Phnom Penh, Cambodia: World Fish Center. Betcherman, G., H. Iftekharul, and M. Melissa. 2019. “Exploring Livelihood Transitions in the Mekong Delta.” Singapore Journal of Tropical Geography 42 (2): 222–240. Blood, R., and D. M. Wolfe. 1960. Husbands and Wives: The Dynamics of Married Living. Glencoe, IL: Free Press. Bolak, H. C. 1997. “When Wives Are Major Providers: Culture, Gender and Family Work.” Gender and Society 11:409-433. Boris, F. 2011. “Think Global, Act Global in the Mekong Delta? Environmental Change, Civil Society, and NGOs.” In Advances in Global Change Research 45, edited by M. A. Stewart and P. A. Coclanis, 7-34. Hanoi. doi:10.1007/978-94-007-0934-8_2. Chapman, A., S. Darby, E. Tompkins, C. Hackney, J. Leyland, P. D. Van, and A. Nicholas. 2017. “Sustainable Rice Cultivation in the Deep Flooded Zones of the Vietnamese Mekong Delta.” Vietnam Journal of Science, Technology and Engineering 59 (2): 34–38. Chau, N. D., Z. Sebesvari, A. Amelung, and F. Renaud. 2015. “Pesticide Pollution of Multiple Drinking Water Sources in the Mekong Delta, Vietnam: Evidence from Two Provinces.” Environmental Science and Pollution Research 22 (12): 9042–9058. https://doi.org/10.1007/s11356-014- 4034-x. Chapter IV - Sustainable agricultural livelihood transformation 101 Conklin, G. H. 1988. “The Influence of Economic Development on Patterns of Conjugal Power and Extended Family Residence in India.” Journal of Comparative Family Studies 19:187–205. Dang, K. S, K. K. Dang, T. H. L. Le, and T. D. Le. 2019. “Scientific Foundation for Agricultural and Aquaculture Transformation in Mekong River Delta.” Vietnam’s Government Research Project 2017–2019, Ha Noi. Debaeke, P., S. Pellerin, and E. Scopel. 2017. “Climate-Smart Cropping Systems for Temperate and Tropical Agriculture: Mitigation, Adaptation and Trade-offs.” Cah. Agric. 26 (3): 34002. doi:10.1051/cagri/2017028. Dung, T. D., G. van Halsema, P. J. Hellegers, L. P. Hoang, Q. T. Tho, M. Kummu, and F. Ludwig. 2018. “Assessing Impacts of Dike Construction on the Flood Dynamics of the Mekong Delta.” Hydrology and Earth System Sciences 22 (3): 1875–1896. Duong, P. 2013. “The Role of Urban Areas in Economic Development in the Mekong Delta.” Ho Chi Minh City. Dutch Government, and Vietnam Government. 2013. Mekong Delta Plan. Vietnam Government. Erban, L. E., S. M. Gorelick, and H. A. Zebk. 2014. “Groundwater Extraction, Land Subsidence, and Sea-Level Rise in the Mekong Delta, Vietnam.” Environmental Research Letters 9 (8): 1–6. doi:10.1088/1748-9326/9/8/084010. FAO (Food and Agriculture Organization of the United Nations). 2010. Climate‐Smart Agriculture: Policies, Practices and Financing for Food Security, Adaptation and Mitigation. Rome: FAO. Giesecke, J., N. Tran, E. Corong, and S. Jaffee. 2013. “Rice Land Designation Policy in Vietnam and the Implications of Policy Reform for Food Security and Economic Welfare.” Journal of Development Studies 49 (9): 1202–1218. Giesecke, J., N. Tran, S. Jaffee, and D. Nguyen 2012. Policy Reforms Associated with Vietnam’s Rice Production and Trade - Simulating the Macroeconomic, Sectoral and Distributional Impacts. Washington, DC: World Bank. Goletti, F., R. Kopicki, E. R. Hansen, and J. Thaller. 2011. “Growing Food, Products, and Businesses: Applying Business Incubation to Agribusiness SMEs.” InfoDev, World Bank, Washington DC. Ha, T. T., and S. R. Bush. 2010. “Transformations of Vietnamese Shrimp Aquaculture Policy: Empirical Evidence from the Mekong Delta.” Environment and Planning C: Government and Policy 28 (6): 1101-1119. Hashimoto, T. 2001. Environmental Issues and Recent Infrastructure Development in the Mekong Delta: Review, Analysis and Recommendations with Particular Reference to Large-Scale Water Control Projects and the Development of Coastal Areas. Sydney: Australian Mekong Resource Centre. Hoang, N. G., V. S. Tran, T. T. V. Nguyen, V. M. Dinh, V. K. Nguyen, and T. H. Nguyen. 2009. The Mechanism to Settle Administrative Complaints in Vietnam: Challenges and Solutions. Summary Report. Hanoi: The Asia Foundation and the Research Institute for Policy, Law and Development. Hoang, L. P., M. T. H. van Viet, M. Kummu, H. Lauri, J. Koponen, I. Supit, and F. Ludwig. 2019. “The Mekong’s Future Flows under Multiple Drivers: How Climate Change, Hydropower Developments and Irrigation Expansions Drive Hydrological Changes.” Science of the Total Environment 649:601- 609. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 102 Alternative Scenarios and Policy Implications Hou, G., C. O. Delang, X. Lu, and R. Olschewski. 2019. “Valuing Carbon Sequestration to Finance Afforestation Projects in China.” Forests 10 (9): 754. doi:10.3390/f10090754. ICEM (International Center for Environmental Management). 2010. MRC Strategic Environmental Assessment (SEA) of Hydropower on the Mekong. Hanoi: Mekong River Commission Secretariat. ILO (International Labour Organization). 2018. ASEAN in Transformation: The Future of Jobs at Risk of Automation. ILO. IPSARD (Institute of Policy and Strategy for Agricultural and Rural Development). 2016. Assessing Farmer Responses to Climate Change—Adjustment Policy Options. Australian Centre for International Agricultural Research (ACIAR) technical report. https://www.aciar.gov.au/sites/ default/files/project-page-docs/final_report_adp-2011-039.pdf. Jaffee, S., Nguyen Do Anh Tuan, Nguyen Ngoc Que, Dao The Anh, Nguyen The Dung, Nguyen Ngoc Mai, Vu Nguyen, and Nguyen Anh Phong. 2012. Moving the Goal Posts. Vietnam’s Evolving Rice Balance and Other Food Security Considerations. Washington, DC: World Bank. Janiola, M. D. C., and R. A. Marin. 2016. “Carbon Sequestration Potential of Fruit Tree Plantations in Southern Philippines.” Journal of Biodiversity and Environmental Sciences 8 (5): 164–174. Khoi, D. K. 2017. Situation, Policy and Innovations on Reducing Post-Harvest Loss in Vietnam. Presentation at the 2017 APEC Expert Consultation on Food Losses and Waste Reduction, Taipei City, June 12–13, 2017. MARD (Ministry of Agricultural and Rural Development). 2019. Master Program for Sustainable Agriculture of the Mekong Delta in the Context of Climate Change (Draft 6). Mathewa, I., H. Shimelisa, M. Mutemaa, and V. Chaplot. 2017. “What Crop Type for Atmospheric Carbon Sequestration: Results from a Global Data Analysis.” Agriculture, Ecosystems and Environment 243:34–46. Mendoza, G. A., and H. Martins. 2006. “Multi-Criteria Decision Analysis in Natural Resource Management: A Critical Review of Methods and New Modelling Paradigms.” Forest Ecology and Management 230 (1): 1-22. Minderhoud, P., G. Erkens, V. Pham, V. Bui, L. Erban, H. Kooi, and E. Stouthamer. 2017. “Impacts of 25 Years of Groundwater Extraction on Subsidence in the Mekong Delta, Vietnam.” Environmental Research Letters 12 (6): 1–13. IOP Publishing Ltd. DOI:10.1088/1748-9326/aa7146. MONRE (Ministry of Natural Resources and Environment). 2016. Climate Change and Sea Level Rise Scenarios of Vietnam up to 2100. Ha Noi: MONRE. Nguyen, T. H. 2017. “An Overview of Agricultural Pollution in Vietnam: The Crops Sector.” Prepared for the World Bank, Washington, DC. Nguyen, D. C., L. T. Duong, V. S. Nguyen, and F. Miller. 2007. “Livelihoods and Resource Use Strategies of Farmers in the Mekong Delta.” In Challenges to Sustainable Development in the Mekong Delta, edited by T. B. Tran, T. S. Bach, and F. Miller, 66–98. Bangkok: Sustainable Mekong Research Network. Nguyen, H. Q., D. Korbee, H. L. Ho, J. Weger, P. T. T. Hoa, N. T. T. Duyen, P. D. M. H. Luan, T. T. Luu, D. H. P. Thao, N. T. T. Trang, L. Hermans, J. Evers, A. Wyatt, X. Q. C. Nguyen, and H. L. Phi. 2019. “Farmer Adoptability for Livelihood Transformations in the Mekong Delta: A Case in Ben Tre Province.” Chapter IV - Sustainable agricultural livelihood transformation 103 Journal of Environmental Planning and Management 62 (9): 1603-1618, doi:10.1080/0964056 8.2019.1568768. Nguyen, V. H., M. C. Maguyon-Detras, M. V. Migo, R. Quilloy, C. Balingbing, P. Chivenge, and M. Gummert. 2020. “Rice Straw Overview: Availability, Properties, and Management Practices.” In Sustainable Rice Straw Management, edited by M. Gummert, V. H. Nguyen, P. Chivenge, and B. Douthwaite, 1–13. Springer Open. https://doi.org/10.1007/978-3-030-32373-8. Nguyen, V., V. Nguyen, N. Nguyen, M. Kummu, B. Merz, and H. Apel. 2015. “Future Sediment Dynamics in the Mekong Delta Floodplains: Impacts of Hydropower Development, Climate Change and Sea Level Rise.” Global and Planetary Change 127: 22-33. Nguyen, T. T. N., F. Roehrig, G. Grosjean, D. N. Tran, and T. M. Vu. 2017. “Climate Smart Agriculture in Vietnam.” CSA Country Profiles for Asia Series. International Center for Tropical Agriculture (CIAT), The Food and Agriculture Organization, Hanoi, Vietnam. OECD (Organisation for Economic Co-operation and Development), and FAO. 2015. “Agricultural Outlook 2015–2024.” OECD and FAO. Oxford Economics. 2018. “Technology and the Future of ASEAN Jobs.” Oxford. Phi, H. L., L. M. Hermans, W. J. A. M. Douven, G. E. van Halsema, and M. F. Khan. 2015. “A Framework to Assess Plan Implementation Maturity with an Application to Flood Management in Vietnam.” Water International 40:984–1003. doi:10.1080/02508060.2015.1101528. Räsänen, T. A., P. Someth, H. Lauri, J. Koponen, J. Sarkkula, and M. Kummu. 2017. “Observed River Discharge Changes due to Hydropower Operations in the Upper Mekong Basin.” Journal of Hydrology 545:28-41. https://doi.org/10.1016/j.jhydrol.2016.12.023. Rentschler, J., S. de Vries Robbé, J. Braese, D. Huy Nguyen, M. van Ledden, and B. Pozueta Mayo. 2020. Resilient Shores: Vietnam’s Coastal Development between Opportunity and Disaster Risk. Washington, DC: World Bank. RIA2 (Research Institute of Aquaculture 2). 2018. Report: Monitoring Changes in Inland Fisheries Resource in Mekong River Delta. Sebesvari, Z., and F. Renaud. 2017. “What Mekong Delta Communities Can Teach Us about Access to Safe Water.” Our World. https://ourworld.unu.edu/en/what-mekong-delta-communities-can- teach-us-about-access-to-safe-water. Shakhovskoy, M., C. Colina, and M. C. Hook 2019. Pathways to Prosperity: Rural and Agricultural Finance - State of the Sector Report. Funded by the Feed the Future Program of the U.S. government. Accessed December 12, 2019. https://pathways.raflearning.org/. South Water Resources Federation. 2013. Report on Groundwater Use and Demand in Mekong Delta. Thanh, H., and C. Tacoli. 2009. “Transforming Livelihoods and Settlements: Fruit Production and Urbanisation in Vietnam’s Mekong Delta.” In Rural-Urban Dynamics: Livelihoods, Mobility and Markets in African and Asian Frontiers, edited by Jytte Agergaard, Niels Fold, and Katherine Gough. Routledge Taylor & Francis Group. Thu, P., and J. Populus. 2007. “Status and Changes of Mangrove Forest in Mekong Delta: Case Study in Tra Vinh, Vietnam.” Estuarine Coastal and Shelf Science 71 (1–2): 98-109. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 104 Alternative Scenarios and Policy Implications Tong, Y. 2015. A Cost and Benefit of Dyke Heightening in the Mekong Delta. World Fish - Economy and Environmental Program for Southeast Asia (EEPSEA). Tran, D., G. van Halsema, P. Hellegers, F. Ludwig, and A. Wyatt. 2018. “Questioning Triple Rice Intensification on the Vietnamese Mekong Delta Floodplains: An Environmental and Economic Analysis of Current Land-Use Trends and Alternatives.” Journal of Environmental Management 217:429-441. Tran, T. 2019. “Land Use Change Driven Out-Migration: Evidence from Three Flood-Prone Communities in the Vietnamese Mekong Delta.” Land Use Policy 88:104157. Tran, T. C, A. Le. Hoang, M. K. Nguyen, A. H. Pham, T. L. Le, V. T. Nguyen, D. T. Ngo, and X. H. Nguyen. 2021. “Renewable Energy from Biomass Surplus Resource: Potential of Power Generation from Rice Straw in Vietnam.” Scientific Reports 11:792. https://doi.org/10.1038/s41598-020-80678-3. Tran, T., and H. James. 2017. “Transformation of Household Livelihoods in Adapting to the Impacts of Flood Control Schemes in the Vietnamese Mekong Delta.” Water Resources and Rural Development 9:67-80. Truong, T. D., and H. L. Do. 2018. “Mangrove Forests and Aquaculture in the Mekong River Delta.” Land Use Policy 73:20-28. VCCI (Vietnam Chamber of Commerce and Industry). 2019. The Provincial Competitiveness Index (PCI). http://eng.pcivietnam.org/. Vietnam South Water Resources Federation. 2013. Report on Groundwater Use and Demand in Mekong River Delta. Sutton, W. R., J. P. Srivastava, M. Rosegrant, J. Thurlow, and L. Sebastian. 2019. Striking a Balance: Managing El Niño and La Niña in Vietnam’s Agriculture. Rep. No 132068. Washington, DC: World Bank. Wang, Q., Y. Li, and A. Alva. 2010. “Cropping Systems to Improve Carbon Sequestration for Mitigation of Climate Change.” Journal of Environmental Protection 1 (3): 207–215. doi:10.4236/ jep.2010.13025. World Bank. 2012. Vietnam Rice, Farmers and Rural Development: From Successful Growth to Sustainable Prosperity. Washington, DC: World Bank. World Bank. 2014. Agriculture Competitiveness Project. Implementation Completion and Results Report. Washington, DC: World Bank. World Bank. 2016a. Transforming Vietnamese Agriculture: Gain More from Less. Ha Noi: World Bank. World Bank. 2016b. Viet Nam 2035: Toward Prosperity, Creativity, Equity and Democracy. Washington, DC: World Bank. World Bank. 2019. Coastal Resources for Sustainable Development Project. Implementation Completion and Results Report. Washington, DC: World Bank. World Bank. 2021. Vietnam Sustainable Agriculture Transformation Project. Progress Reports. Washington, DC: World Bank. World Economic Forum. 2018. Report: Readiness for the Future of Production. Zorya, S., S. Jaffee, T. Nguyen, Q. Nguyen, H. Nguyen, and T. Truong. 2018. “Policy Note: Vietnam’s Rice Outlook 2030 and Implications for the Designated Rice Area.” World Bank, Hanoi. Chapter IV - Sustainable agricultural livelihood transformation 105 APPENDICES Appendix 1. Study Scope and Methods Annex 1.1. Zoning the Mekong River Delta Depending on their research and policy purposes, agencies have different ways to divide the MKD. In the agricultural sector so far, there are four main ways to zone the delta (Figure A1-1). FIGURE A1-1. Four main ways of zoning the MKD MDP of Dutch experts Rural agricultural planning in the MKD Zoning criteria: (Decision No. 639/QDBNN-KH)54 hydrological regime from west to east Zoning criteria: agricultural ecosystem Irrigation planning of the MKD (Decision No. 59356) (Decision No. 1397/QD-TTg)55 Zoning criteria: river system from north to south Zoning criteria: expected integration based on reality between provinces according to administrative boundaries Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Authors’ illustration. 54 Decision No. 639/QD-BNN-KH dated 2 April 2014 of MARD on “Planning of agriculture and rural areas in the Mekong Delta up to 2020, with a vision to 2030 under climate change conditions.” 55 Decision No. 1397/QD-TTg of the Prime Minister on ‘Approving the planning on irrigation in the Mekong River Delta for the period of 2012–2020 and orientations to 2050 in the context of climate change and sea-level rise’. 56 Decision No. 593/QD-TTg dated April 6, 2016, of the Prime Minister on the ‘Pilot Regulation on Socio-Economic Development Linkages in the Mekong Delta’. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 106 Alternative Scenarios and Policy Implications The present study used the MKD zoning according to the three subregions in the MDP by the Dutch experts to be consistent with the draft Mekong Delta Agricultural Transformation Program in the framework of Resolution 120/2018/NQ-CP prepared by the MARD. This way of zoning is based on the hydrological regime, especially flood and drought. Accordingly, the MKD is divided horizontally into the Upper subregion (deep flood, freshwater), the Middle subregion (shallow, brackish water), and the Coastal subregion (salt water) (Figure A1-2). However, this new way of delta zoning has at least two weaknesses. First, no clear borders exist among these three subregions (as this division is based on water movement caused by climate change impact, which is relatively unstable). Second, although this zoning method can reflect the climate change impact on farmers’ livelihoods, it may not capture this precisely enough and may not separate some areas with significantly different farming systems and livelihood systems. For agriculture, in addition to the hydrological regime, ecological zoning must consider other factors such as geomorphology, soil, and cultivation history. In addition, through the development process, several irrigation infrastructure systems have been built together with a strongly shifting cultivation system, which is difficult to reverse, so the regional development orientation needs to be calculated according to these factors. Therefore, instead of dividing it into three large areas as in the MDP, a further division is needed to have detailed orientation and be close to the reality of each region. FIGURE A1-2. Three subregions in the MKD MAP OF MEKONG RIVER DELTA ZONING LEGEND Regional boundaries Provincial boundaries District boundaries Upper Sub-region Middle Sub-region Disclaimer: Coastal Sub-region The boundaries, colors, denominations and other information shown on any Project: Identifying sustainable map in this work do not imply any judgement on the part of cliamate-resilient agriculture, The World Bank concerning aquaculture, and horticulture the legal status of any livelihood models and transition territory or the endorsement strategies in the Mekong Delta to or acceptance of such adapt with climate change. boundaries. Source: Prepared by authors (2019). Appendices 107 The six-subregion division of Decision 639 can be engaged in harmony with the three-region zoning of MDP: (a) most of the LXQ and Dong Thap Muoi POR areas correspond to the Upper subregion of the MDP; (b) a part of Ca Mau Peninsula and the eastern coastal estuaries corresponds to the estuarine and coastal zones the MDP; and (c) much of the western part of the Hau River, the area between the Tien and Hau Rivers, and the interior of Ca Mau Peninsula and the eastern coastal estuaries correspond to the Middle subregion of the MDP. Thus, the present study briefly divides the three large regions in the MDP into six smaller subregions (Figure A1-3): Eastern Upper, Western Upper, Northern Middle, Southern Middle, Eastern Coastal, and Western Coastal. During this work, the authors will use both ways of delta zoning, three-region zoning and six subregion division, depending on the purpose of the analyses (Table A1-1). FIGURE A1-3. Six subregions in the MKD MAP OF MEKONG RIVER DELTA ZONING LEGEND Regional boundaries Provincial boundaries District boundaries Eastern upper Sub-region Western upper Sub-region Southern middle Sub-region Northern middle Sub-region Disclaimer: The boundaries, colors, Eastern coastal Sub-region denominations and other Western coastal Sub-region information shown on any map in this work do not imply any judgement on the part of Project: Identifying sustainable The World Bank concerning cliamate-resilient agriculture, the legal status of any aquaculture, and horticulture territory or the endorsement livelihood models and transition or acceptance of such strategies in the Mekong Delta to boundaries. adapt with climate change. Source: Prepared by authors (2019). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 108 Alternative Scenarios and Policy Implications TABLE A1-1. The area of three subregions and six subregions of the MKD (ha, thousands) Upper Middle Coastal Eastern Province Major part of Major part of North of Hau South of Hau Ca Mau coastal POR LXQ River River Peninsula estuaries Long An 257.4 180.1 12.5 Tien Giang 33.6 192.7 13.6 Ben Tre 171.0 60.3 Dong Thap 202.3 0.2 136.9 Vinh Long 153.1 Tra Vinh 177.3 47.4 Can Tho 11.8 0.4 132.6 Hau Giang 0.1 162.9 An Giang 0.9 355.0 0.1 Kien Giang 178.0 148.4 241.6 Soc Trang 0.3 243.2 23.5 59.7 Bac Lieu 71.6 176.7 Ca Mau 0.0 0.0 520.0 Source: Calculated by authors (2019). Annex 1.2. Desk-review methodology To conduct a summary on key features of the three MKD subregions, the research team consolidated national statistics for 13 MKD provinces from three main databases, including annual provincial statistical yearbook 2018, AgroCensus57 2016, and Vietnam Household Living Standard Survey58 (VHLSS) 2000-16. International and domestic research and studies related to agriculture, farmer livelihood, rural areas, upstream development, internal socioeconomic development, and climate change in the MKD will also be collected. For the information from provincial sources, the research team synthesized and summarized the policies, strategies, development plans, land-use maps, and administrative reports that CAP/ IPSARD had already collected from DARD, DONRE, DoC, DoIT, and DOLISA of 13 provinces in the MKD for the work on preparing the Agricultural Transformation Program in the framework of Resolution 120. 57 This survey has been implemented countrywide for the whole farmer population by the GSO every five years from 2001 to the present. 58 This survey has been implemented countrywide (with the sample of about 45,000 households) by the GSO every two years from 2002 to the present. Appendices 109 Annex 1.3. Stock-taking method of livelihood models in the three MKD subregions The research team consolidated materials from previous studies related to farmer livelihood in the MKD, which were conducted by CAP/IPSARD, Center of Water Management and Climate Change (WACC) of Vietnam National University, CTU, An Giang University, IUCN, and other organizations (Table A1-2). The materials were used to prepare the following: • A stock-taking of common conventional livelihood models in the three subregions covering descriptions of farming systems, geographic location, production size/scale, number of participating households, and strengths and weaknesses • A stock-taking of potential climate-resilient livelihood models in the three subregions covering descriptions of farming systems, geographic location, production size/scale, number of participating households, and strengths and weaknesses. TABLE A1-2. Some available key databases Institution Description Period A database from a survey of 300 farming households in Ben Tre, Tra Vinh, and An Giang Provinces on farmer measures for climate change in the Australian Centre for CAP/IPSARD 2013-2016 International Agricultural Research (ACIAR) project ‘Assessing farmer responses to climate change: adjustment policy options’ A database from a survey of 600 farming households on their farming systems in six provinces: An Giang and Dong Thap (POR and LXQ), Hau Giang (West Hau River), CAP/IPSARD and Ben Tre (eastern coastal estuary), and Soc Trang and Ca Mau (Ca Mau Peninsula) 2017 CTU in the MKD in the National Research Program ‘Studying the scientific basis for sustainable development of agriculture and aquaculture in the Mekong Delta of Vietnam’ WACC/Vietnam A database from a survey of 600 farmer households on assessing the resilience of National their livelihood to climate change using the MOTA method in Ben Tre Province in 2018 University the National Research Program A database related to some surveys about pilot livelihood models in the preparation IUCN 2016 of the MKDICRSL project funded by the World Bank Source: Prepared by authors (2019). Based on data collected in previous studies related to CSA, climate-resilient agriculture, sustainable livelihoods, and resilient livelihoods, the research team prepared a list of 59 livelihood models (Table 1-3). For the most common model group, three main indicators (area of production, number of households participating in the model, and production value of the model) were used. After that, the research team conducted one expert consultation meeting in Can Tho City59 and consulted with many other experts to select potentially sustainable models based on a set of indicators (economic, environmental, natural condition suitability, social, gender, and climate change resilience). 59 Experts include local experts coming from DARD of 10 MKD provinces and commodity experts (institutes, universities, and enterprises) who have good knowledge on rice, fruit, aquaculture, and livestock commodities. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 110 Alternative Scenarios and Policy Implications TABLE A1-3. List of key livelihood models selected for the research Key livelihood Potentially sustain- Improvement model able model 3-crop rice 1M5R, 3R3G, contract farming, and cooperatives High-quality variety, organic rice, contract farming, and 2-crop rice cooperatives Upper Rice-vegetables Contract farming and cooperatives Rice-lotus, rice- subregion Automatic irrigation, contract farming, cooperatives, and Vegetables shrimp, rice-fish, and plastic/net house rice-ducks Automatic irrigation, contract farming, improved variety, Fruit trees and cooperatives Pangasius Ducks Large commercial farm with certification 3-crop rice 1M5R, 3R3G, contract farming, and cooperatives High-quality variety, organic rice, contract farming, and 2-crop rice cooperatives Rice-vegetables Contract farming and cooperatives Automatic irrigation, contract farming, cooperatives, and Vegetables plastic/net house Combined coconut- Automatic irrigation, contract farming, improved variety, Middle Fruit trees shrimp, rice-ducks; and cooperatives subregion combined fruit trees- Automatic irrigation, contract farming, improved variety, Industrial crops livestock and cooperatives Pigs Large commercial farm with certification Large commercial farm, combined with vegetable Beef cattle production to use by-products Chicken Pangasius fish High-quality variety, organic rice, contract farming, and 2-season rice cooperatives Rice-shrimp Extensive aquaculture/ improved High-tech physical Coastal extensive and biological subregion aquaculture shrimp model Semi-intensive/ intensive aquaculture Super-intensive aquaculture Source: Prepared by authors (2019). Appendices 111 Annex 1.4. Soil and water suitability assessment of key livelihood models in the three MKD subregions at present (2017) and in the future (2030) This methodology includes the following steps: Step 1. Spatial database development The research team adopted the map prepared by CTU in the governmental project ‘Establishing the scientific foundation for sustainable agricultural and fisheries development in the Mekong Delta’ co-implemented by IPSARD and CTU (2019), authored by Dang et al. (2019). In that research, the maps were edited using Qgis’s GIS toolkit, and the Vietnam national coordinate reference system (VN2000). Base maps include administrative boundaries (provincial and district levels), transportation (national roads, provincial roads, and district roads), and hydrology (main rivers and canals). In addition, they used the current agricultural land-use map (prepared in 2010 and updated to 2015) of the Southern Institute of Agriculture Planning and the land review map of the MKD provinces in 2014. The crop structure was identified using the Moderate Resolution Imaging Spectroradiometer (MODIS) satellite pictures. During field trips, the research team also worked with technical staff in DARDs of 10 provinces in the MKD to update agricultural land use in 2019. Step 1.1. Preparation of specialized maps The research team adopted the soil map prepared by CTU (Dang et al. 2019), which updated the soil map 2014, and the MKD land map based on the following data sources: (a) land map 2009 of the Ministry of Natural Resources and Environment (MONRE) and CTU; (b) the national- level research project ‘Research on the proper use of acid sulfate soil in the Mekong Delta in the context of climate change’, Code: BDKH-57, of Prof. Ngo Ngoc Hung. With data from the land map 2014, the research team developed some thematic maps. Soil texture map: Soil texture is classified into four types (sandy, loam, clay, and clay loam) (Figure A1-4). FIGURE A1-4. Map of soil texture of the three subregions in the MKD MAP OF MEKONG RIVER DELTA ZONING LEGEND Regional boundaries Sand Loam clay Loam Clay loam Project: Identifying sustainable cliamate-resilient agricul- ture, aquaculture, and horticulture livelihood models and transition strategies in the Mekong Delta to adapt with climate change. Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Prepared by authors (2019) based on GIS data in Dang et al. (2019). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 112 Alternative Scenarios and Policy Implications Acid sulfate soil occurrence maps (Figure 1.5) The depth of acid sulfate soil and potential acid sulfate soil occurrence is classified into four types: non-occurred, depth less than 50 cm, depth 50-100 cm, and depth more than 100 cm. FIGURE A1-5. Active acid sulfate soil map and potential acid sulfate soil map Active acid sulfate Potential acid sulfate Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Prepared by authors (2019) based on GIS data in Dang et al. (2019). Salinity maps Based on the available data on water conditions (salinity and flood) under climate change scenarios until 2030 in the CLUES research project of Dragon Institute - CTU (2015), in Dang et al. (2019), CTU conducted surveys to update salinity and flood conditions in 2017. The information on salt water is shown in the salinity duration map and salinity degree map (Figure A1-6). Salinity duration can be classified into four groups: non-salinity, salinity under 3 months, salinity from 3 to 6 months, and salinity surpassing 6 months. Salinity degree can be classified into five groups: non-salinity, salinity of 4-8 g/L, salinity of 8-12 g/L, salinity of 12-20 g/L, and salinity of more than 20 g/L. Appendices 113 FIGURE A1-6. Map of salinity duration and salinity degree at present (2017) Salinity degree Salinity duration Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Prepared by authors (2019) based on GIS data in Dang et al. (2019). Flood maps In Dang et al. (2019), flooded areas come from the CLUES research project, and satellite image interpretation results come from a research project by Dr. Vo Quoc Tuan and updated with 2017 information. The locations of flooded areas are shown in a flood duration map and flood depth map (Figure A1-7). Flood duration can be classified into three groups: non-flood, flood under 3 months, and flood from 3 to 6 months. Flood depth can be classified into five groups: non-flood, flood under 30 cm, flood from 30 to 60 cm, flood from 60 to 90 cm, and flood more than 90 cm. FIGURE A1-7. Flood duration map and flood depth map, 2017 Flood depth Flood duration Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Prepared by authors (2019) based on GIS data in Dang et al. (2019). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 114 Alternative Scenarios and Policy Implications Precipitation map Precipitation allocation comes from the data set of the Mekong Delta Master Plan to 2020 of the Institute of Agriculture and Rural Development. Precipitation amount is classified into four groups: under 1,200 mm, from 1,200 to 1,600 mm, from 1,600 to 2,000 mm, and from 2,000 to 2,400 mm. Freshwater capacity map The capacity to supply freshwater for agricultural production activities is identified using the current status of the paddy cultivation areas, regions with Mekong River water supply, and the data on rainfall distribution during the year collected by the MKD meteorological stations. Freshwater capacity is classified into three groups: freshwater available year-round, water available for 6 months, and water available for 6 to 9 months. Step 1.2. Land unit map From Dang et al. (2019), the research team used land characteristics collected from thematic maps, soil maps (texture, depth of acid sulfate soil, and potential acid sulfate soil), and water maps (salinity degree, salinity duration, flood depth, flood duration, freshwater availability during the year, and so on) to prepare a land unit map (Table A1-4). These data in 2019 were updated during the field trips in 10 provinces to match with the current status of the MKD. The land unit map includes 260 units along with their specific descriptions of soil texture, depth of acid sulfate soil and potential acid sulfate soil occurrences, flood depth, flood duration, salinity degree, salinity duration, and freshwater availability. TABLE A1-4. Characteristics for land unit classification Characteristics Group Texture ­ • Sandy, loam, clay, clay loam ­• Non-occurrence Acid sulfate • ­ Occurrence under 50 cm • ­ Occurrence of 50-100 cm ­• Non-occurrence • ­ Occurrence under 50 cm Potential acid sulfate • ­ Occurrence of 50-100 cm ­ • Occurrence surpassing 100 cm • Salinity of <40/00 ­ • Salinity of 480/00 Salinity degree ­ • Salinity of 8120/00 ­ • Salinity of 12200/00 • ­ Salinity of >200/00 • Non-salinity • Salinity under 3 months ­ Salinity duration • Salinity of 3-6 months ­ ­ • Salinity surpassing 6 months Appendices 115 TABLE A1-4. Characteristics for land unit classification (Cont.) Characteristics Group • Non-flood • Flood under 30 cm ­ Flood depth • Flood of 30-60 cm • Flood of 60-90 cm ­ • Flood of more than 90 cm ­ • Non-flood Flood duration • Flood under 3 months ­ • Flood of 3-6 months ­ • Available year-round ­ • Available 9 months Freshwater availability • ­ Available 6-9 months ­ • Available 6 months ­ • Available less than 3 months Source: Adapted from Dang et al. (2019). Step 2. Natural condition suitability assessment of key livelihood models in the three MKD subregions for the present (2017) Each key livelihood model is analyzed based on a typology using the FAO land assessment framework of 1976. The four ranges of suitability are S1 (highly suitable), S2 (moderately suitable), S3 (poorly suitable), and N (unsuitable). The assessment results were verified with technical experts and local governments and by comparing the suitability ranges with the actual productivity. The results are illustrated by eight thematic maps to evaluate the suitability of key livelihood models in the MKD. Step 3. Natural condition suitability assessment of key livelihood models in the three MKD subregions for the future (2030) When analyzing the climate change scenario until 2030, to focus on the main impact factors, the calculation is based on the assumption that the soil conditions (texture, depth of acid sulfate soil, and depth of potential acid sulfate soil) and characteristics of freshwater availability in each year in the MKD have not changed. The only two main changes are the conditions of salinity intrusion (salinity degree and salinity duration) and flooding (flood depth and flood duration). The data on salinity intrusion and flood by 2030 used for calculation were adopted from the CLUES project of the Institute for Climate Change Research (CTU), conducted in 2015 and updated in 2017. Change in salinity According to MONRE scenario 2012,60 based on sea level rise and flow changes in the Mekong River downstream, in the next 50 years, about 47 percent of the MKD will be affected by 4 g/L salinity, and up to 64 percent of the area will be affected by 1 g/L salinity. Based on the updated CLUES 60 A2: High Greenhouse Emission Scenario; B2: Medium Greenhouse Emission Scenario. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 116 Alternative Scenarios and Policy Implications research project, the research group classified the MKD by salinity degree: the area with salinity lower than 4 g/L zone will decrease, accounting for 71.52 percent of the total suitability zones. In addition, the salinity degree of the smallest size is the salinity 4 g/L–8 g/L zone, accounting for 1.08 percent of the total. The area with salinity 8 g/L-12 g/L zone accounted for 2.15 percent of the total, concentrated in coastal areas. The salinity area of the 12 g/L-20 g/L zone (with a total size of 283,272 ha) accounted for 7.06 percent of the total. Finally, 18.20 percent is the percentage for the area of salinity higher than 20 g/L zone. Regarding salinity duration, the MKD will be classified into four groups (Figure A1-8). The area of the non-salinity zone will be assessed as 2,869,703 ha by 2030, accounting for 71.52 percent. The area of the salinity under 3 months zone will be calculated at 34,516 ha, accounting for 0.86 percent. The area of salinity for the 3-6 months zone will be calculated at 293,260 ha. The area of salinity for the more than 6 months zone will be 819,714 ha, accounting for about 20.31 percent. FIGURE A1-8. Map of salinity duration and salinity degree in the future (2030) Salinity degree Salinity duration Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Prepared by authors (2019) based on GIS data in Dang et al. (2019). Changes in flood The results of flood forecast indicate that non-flooded area still accounts for 72.58 percent of the regional area (2,912,194 ha), distributed in almost all provinces. The zone with 0-30-cm flood depth will be 411,689 ha. The 30-60-cm flood depth zone will decrease to 107,591 ha. The 60-90-cm flood depth zone is estimated to increase to 259,355 ha. The more than 90 cm flood depth zone, currently accounting for 1.89 percent, will rise sharply to 18.28 percent and reach 733,498 ha (Figure A1-9). Appendices 117 FIGURE A1-9. Flood duration and flood depth in the MKD by 2030 Flood depth Flood duration Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Prepared by authors (2019) based on GIS data in Dang et al. (2019). The calculation results show that flood duration in the MKD can be classified into three zones. The area of the non-flooded zone will be almost unchanged, accounting for 72.58 percent of the natural area, distributed in nearly all provinces. The area of flood under 3 months by 2030 is assessed at 42,353 ha, accounting for only 1.06 percent of the region’s natural area. The area of flood from 3 months to 6 months is expected to increase to 387,952 ha, accounting for 9.67 percent. The area for flood surpassing 6 months by 2030 is calculated at 670,140 ha, accounting for 16.70 percent of the natural land area. From eight thematic maps that focus on changes in flooding and salinity, the research group developed 225 units along with specific descriptions of soil, water, and climate conditions in the MKD for the scenario to 2030. The maps in Figure A1-10 illustrate the locations of the classification of the characteristics for the land unit. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 118 Alternative Scenarios and Policy Implications FIGURE A1-10. Suitability assessment of key livelihood models in the MKD in 2017 and 2030 Disclaimer: The boundaries, colors, denominations and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Source: Prepared by authors (2019). Appendices 119 Annex 1.5. Sampling and survey Local expert consultations The research team conducted field trips to 10 provinces (except Can Tho, Tra Vinh, and Bac Lieu) in three subregions to (a) collect background information on localities, (b) consult with local experts related to selecting potentially sustainable livelihood models, (c) interview local experts on incentive and capacity of local agencies for the MOTA framework, and (d) update the land-use map. Expert consultations were conducted in the form of face-to-face discussions, group discussions, and closed questionnaires. The research team also interviewed a diversified group, including the Department of Planning and Investment, Provincial Women’s Union, District Division of Agricultural and Rural Development, leaders of communes, leaders of commune women’s unions, and leaders of cooperatives. Household surveys To obtain households’ detailed data for CBA, in-depth MCAs, feasibility studies, and surveys in 10 provinces in the three MKD subregions were implemented. Samples were selected to represent typical households of each livelihood model. The number of samples for each livelihood model depends on the number of participating households, likewise for production practices and recommendations from the local government. The total number of interviewed households was 600, for which 450 were interviewed using a closed questionnaire and 150 were selected to join focus group discussions. The household interviews and focus group discussions were designed to provide information for all three research methods: CBA, MOTA, and gender analysis. The allocation of samples is presented in Table A1-5. TABLE A1-5. Household survey sample Number of HHs Province Livelihood model District Total HHs for survey HHs for FDGs Vegetables Chau Thanh 20 15 5 Long An Intensive fruit (dragon fruit) Chau Thanh 20 15 5 Livestock (pigs) Chau Thanh 20 15 5 2-crop rice Upper Region Tan Phuoc 50 40 10 Tien Giang Intercropping fruits Tan Phuoc 30 20 10 Mango Cai Be 20 15 5 Coconut Ba Tri 25 20 5 Ben Tre Poultry (chickens) Ba Tri 20 15 5 Beef cattle Ba Tri 20 15 5 Rice-vegetables Long Ho 25 20 5 Vinh Long Vegetables Long Ho 30 25 5 Industrial crops (pineapple) Vi Thanh 25 20 5 Hau Giang Industrial crops (sugar cane) Phung Hiep 25 20 5 Rice-shrimp An Bien 15 10 5 Kien Giang Intensive shrimp farming An Bien 15 10 5 Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 120 Alternative Scenarios and Policy Implications TABLE 1.5: Household survey sample (cont.) Number of HHs Province Livelihood model District Total HHs for survey HHs for FDGs Rice-shrimp Dam Doi 15 10 5 Ca Mau Intensive shrimp farming Dam Doi 15 10 5 2-crop rice Lower Region Tran Van Thoi 50 40 10 3-crop rice Thoai Son 40 30 10 An Giang Rice-vegetables Cho Moi 25 20 5 Catfish farming Thanh Binh 15 10 5 Dong Thap Poultry (ducks) Thanh Binh 20 15 5 Citrus Lai Vung 30 20 10 Soc Trang Semi-intensive shrimp farming Tran De 30 20 10 Source: Prepared by authors (2019). The six interviewers were experienced researchers from Vietnam National University - HCMC and the Vietnam National University of Agriculture. Data were collected using paper versions and were stored and analyzed using the Statistical Package for the Social Sciences (SPSS version 20) and Stata software. Annex 1.6. Cost and benefit analysis To assess the profitability of key livelihood models for farmers in each region of the MRD, the research team used the data on economic performance of such models from the household surveys and the existing data from the literature and conducted cost-benefit analyses. Survey questionnaires with detailed questions related to costs and revenue of households participating in key livelihood models were used. For each livelihood model, the total costs incurred when using that intervention and benefits were identified and the net benefit computed for that particular practice in monetary terms. NB = ∑TB – ∑TC where NB represents the net benefits, TB represents the total benefits, and TC represents the total costs. Expected values of net profit cash flows in a period of 10 years (2009-2019) were calculated. Then, net present values were measured because they discount the future benefits to present values: NPV = ∑ (Bt - Ct) / (1 + r)t where Bt = total benefits in the year t, Ct = total costs in the year t; and r = discount rate.61 61 The interest rate used in this research was the lending rate of the Agricultural Bank (Agribank). Appendices 121 Besides net profit, public investment in surrounding or supportive infrastructure, private investment in in-field infrastructure, and working capital for daily livelihood model operation were calculated to evaluate the cost of establishing and upgrading livelihood models or transforming to other livelihood models of farmers. Two methods were used to quantify risks affecting the cost, revenue, and profit of key livelihood models. Probability distribution. The degree of uncertainty can be defined and measured in terms of the analyst’s probability distribution: probability estimates associated with each possible outcome. In its simplest form, a probability distribution could consist of an optimistic, pessimistic, and most likely estimate or, alternatively, high, low, and “best guess” estimates. In this research, three factors are examined: state of market, state of disease, and state of disaster. The optimistic or high estimate is realized if the marketing prospects of the products are booming, with no disease and no disaster. Pessimistic or low estimates prevail under conditions of market recession, experiencing diseases and disasters. Most likely or best guess estimates prevail during normal conditions. All information was collected from farmer group discussions and local authority interviews. Sensitivity analysis. The purpose of this analysis is to identify the variables to which NPV is most sensitive, and it is normally used to measure the extent to which these variables can change before the investment results in a negative NPV. Sensitivity analysis is useful for determining consequences of specified changes in variables in the future such as product price, sales volume, and input costs. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 122 Alternative Scenarios and Policy Implications Annex 1.7. Multi-criteria analysis MCA is a method of engaging stakeholders in evaluating alternatives based on set criteria and sub-criteria (Mendoza and Martins 2006). This study implemented an analytic hierarchy process (AHP) MCA, the most widely used method to assess a set of different aspects of livelihood models in the three MKD subregions (economic, socio-institutional, and environmental sustainability in addition to resilience assessment) (Figure A1-11). It aims to evaluate agricultural alternatives in farming systems, including cultivation, poultry and livestock production, and aquaculture, under a perspective of exploring sustainable climate-resilient agriculture, aquaculture, and horticulture livelihood models. FIGURE A1-11. MCA-AHP framework applied for the project Source: Prepared by authors (2019). Appendices 123 With this aim, the MCA involved a complex evaluation of three main alternative livelihood models (in the Upper, Middle, and Coastal subregions) and conducted assessments based on four criteria and 18 sub-criteria. The four criteria include resilience (five sub-criteria), economy (five sub-criteria), social institutions (four sub-criteria), and environment (four sub-criteria) (Table A1-6). This study conducted MCA-AHP using a two-phase approach to evaluate alternatives in an agriculture-based livelihood model and organized three MCA workshops, one with experts and two with farmers, to assess the farming systems based on the criteria defined. To perform the rankings and assess the consistency and coherence of participants’ scores and responses, we used the AHP. In the first phase, a questionnaire was formulated with five sections to send to 10 Vietnamese experts for the evaluation of the criteria and sub-criteria. All the experts were required to have experience of more than 10 years working in agriculture, horticulture, and aquaculture. Especially, they should have been working with farmers, and each should be knowledgeable about farming systems and farmers’ livelihoods in the MKD. The questionnaire began with an overview of the study’s objective, methodology, the farming systems included, and a ‘how-to’ scoring guide for the MCA. In the second phase, 10 focus group sessions were conducted with 63 officials in 10 provinces of the MKD. Each focus group session involved five to nine invited officials working in different departments of agriculture, aquaculture, planting, animal health, and women’s union. For each focus group discussion, the officials first discussed in the whole group to specify alternatives in livelihood models, which were then evaluated individually using the questionnaires. In addition, 17 officials of DARDs of 13 districts of the 10 provinces were appointed to evaluate the MCA questionnaires, providing different perspectives compared with those from the provincial level. As mentioned earlier, the criteria and sub-criteria were considered based on two aspects of resilience and sustainability. Resilience criteria consist of risk-coping ability, adaptability, recovery, transformability, and vulnerability, whereas sustainability is assessed following three main pillars: economy, social institutions, and environment. To determine such criteria/sub-criteria and indicators, two steps are followed: Step 1. Criteria, sub-criteria, and indicators were identified from the peer-reviewed literature and gray literature published by major international and national agencies. Criteria and sub-criteria were preliminarily listed with definitions. Step 2. Possible criteria/sub-criteria and indicators were selected under the evaluation of experts and scientists. These criteria/indicators were then included in the questionnaires for scoring. Questions were formulated and pretested carefully for possible answers. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 124 Alternative Scenarios and Policy Implications TABLE A1-6. Results of MCA in ranking current livelihood models in the Upper, Middle, and Coastal subregions of the Vietnamese MKD Source: Prepared by authors (2019). Appendices 125 Annex 1.8. Gender analysis Gender analysis consisted of three main aspects: (a) the participation of women in agricultural activities, (b) the role of women in making production decisions, and (c) the ability of women to access common and household resources. The participation of women in agricultural activities would provide information on the current situation regarding the contribution of women in the local workforce in general and in specific livelihood models. The data were analyzed to provide information on the changes in female labor supply and demand with each transformation scenario for each subregion. This would be an important aspect of poverty and social impact analysis (PSIA) to minimize the negative impacts of the transformation and involve more women in agricultural activities. The role of women in making decisions has been an important topic in social studies. The framework on women’s voice was adopted from Blood and Wolfe (1960), in which the voice in decision-making originated from the contribution of resources, especially income, educational level, and employment status. The research team also supplemented factors from Conklin (1988), in which women who live with the husband’s family have a lower voice and women have a stronger voice in a core-family than in an expanded family. Bolak (1997) noted that women’s participation in housework would also be recognized even when it does not create more income. In addition, local custom plays an important role in the allocation of decision-making power (Blood and Wolfe 1960). Apart from the decision-making power, the ability to access common and family resources is also an important factor in identifying the gender equality of a region. This study considered three main public resources: water resources, transportation resources, and health care resources. At the family level, the research team looked at the role of women in managing the household’s financial resources (who keeps money and makes purchase decisions). In addition, the research team used the PSIA method to assess the social impacts of transformation, particularly gender equality. Accordingly, the research team provided evidence of the impact of the transition on women, especially female workers, and made appropriate policy recommendations to minimize the adverse effects. Annex 1.9. MOTA analysis To assess the feasibility of upscaling potential sustainable models and their transformative pathways, this study employed the MOTA (Motivation and Ability) framework (Figure A1-12). The MOTA framework was developed to analyze the feasibility of implementing a plan or policy through the lens of stakeholder capacity and willingness to adopt a proposed plan (Nguyen et al. 2019; Phi et al. 2015). The MOTA framework takes a multi-stakeholder and multi-level approach to assess and compare projects and plans, centering on the integrated relationship between ‘trigger- motivation-ability’. The outcomes are then conveyed through a combination of motivation and Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 126 Alternative Scenarios and Policy Implications the ability of multiple stakeholders at different co-existing levels in the implementation process (Nguyen et al. 2019). FIGURE 1.12: MOTA framework: from trigger This framework recognizes that the influence to action of a trigger on motivation is shaped by one’s perception of threats and opportunities or subjective assessment of that trigger, which may be either positive or negative and range in intensity from weak to strong. In other words, a specific trigger may be perceived as a threat, as neutral, or as an opportunity. And this is where stakeholders become (or do not become) involved in the execution of a project. One of the relevant aspects of the MOTA tool is that, through surveys, it can offer quantitative indicators for the motivations and Source: Adapted by authors from Phi et al. (2015). abilities of different actors. This quantification is useful for visualizing whether a stakeholder lags in motivations and/or abilities toward a specific plan (Figure A1-13). FIGURE A1-13. MOTA score mapping Source: Prepared by authors (2019). Appendices 127 Identifying the degree to which specific motivations and abilities associate the stakeholders as, for example, leaders or opposers is useful for then identifying the strategies to follow to make a behavior change happen. A noticeable result of such use of the MOTA framework is that an option with a low benefit-cost ratio but with a high implementation probability is not necessarily worse than an option with the highest benefit-cost ratio but a low probability of implementation. More specifically, capacity- and consent-building strategies were considered in developing pathways in livelihood transformation. The following are the steps for implementing the MOTA framework: 1. Problem definition of when to use MOTA. Why use MOTA? In which situations? Adequate preliminary research should be done here to identify the key elements for which MOTA can contribute. 2. Specifying the relevant MOTA elements. Here, we will translate the challenges found in the previous step into specific relations with motivations, abilities, and triggers. 3. Survey preparation. How will the MOTA components be measured? By surveys (namely questionnaires and interviews). Here, we will see how to translate MOTA attributes into questions. 4. Survey implementation. This step presents the fundamental advice for an adequate survey implementation. 5. Data processing and analysis. Once data are gathered, what do you do with them? How do you process them? This step presents the main tools used: MOTA scoring and (multi-variable) statistical analyses. 6. Synthesis and recommendations. How do you translate the findings into useful information for planners and decision makers? Here are some tips on how to translate this information. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 128 Alternative Scenarios and Policy Implications Appendix 2. Mekong Delta and its subregions FIGURE A2-1. Elevation map of the three subregions in the MKD Source: Prepared by authors based on the International Center for Environmental Management (ICEM) map (2019). Appendices 129 FIGURE A2-2. Land subsidence in the three subregions in the MKD Source: Adapted by the authors based on Minderhoud et al. (2017). FIGURE A2-3. Map of land erosion points in the three subregions in the MKD, 2018 Source: Prepared by authors based on the Directorate of Natural Disaster Control and Prevention (2019). http://satlodbscl. phongchongthientai.vn/#9/9.7862/105.7050/c0c1. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 130 Alternative Scenarios and Policy Implications FIGURE A2-4. Inland fishery resources in the Upper subregion of the MKD 120 Production (thousand tons) An Giang Vinh Long 100 Dong Thap Can Tho 80 Hau Giang 60 40 20 0 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 19 19 19 19 19 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 20 Source: RIA2 2018. FIGURE A2-5. Average after-tax profit of agriculture, forestry, and fishery enterprises in the MKD, 2018 Unit: Million VND in real price Upper Middle Coastal 35000.0 29,247 30000.0 25000.0 20,552 20000.0 15000.0 12,093 10000.0 3,728 4,014 139 2,153 5000.0 3,204 194 1,235 641 0.0 State enterprises Cooperatives Private Enterprises Foreign Enterprises -5000.0 -378 Source: Authors’ calculations from Enterprise Surveys. Appendices 131 TABLE A2-1. Comparison of current groundwater mining with water demand in the MKD, 2010 Drinking Industry Agriculture Province Volume Proportion of Volume Proportion of Volume Proportion of (m3/day) total need (%) (m3/day) total need (%) (m3/day) total need (%) 1 An Giang 37,482 22.6 22,864 20.3 34,191 0.4 2 Bac Lieu 63,072 95.7 18,393 27.6 167,263 2.7 3 Ben Tre 11,872 14.3 4,210 5.6 1,905 0.1 4 Ca Mau 87,493 99.2 31,480 16.9 40,145 0.4 5 Can Tho 101,398 85.1 44,916 12.3 42,530 1.0 6 Dong Thap 37,155 31.5 69,873 39.0 9,141 0.1 7 Hau Giang 31,548 57.8 5,600 8.8 25,395 0.5 8 Kien Giang 122,650 94.6 800 0.7 73,991 0.6 9 Long An 96,663 94.7 67,091 19.8 31,800 0.4 10 Soc Trang 73,190 77.2 31,180 20.6 140,480 1.6 11 Tien Giang 59,596 52.0 41,349 29.2 39,719 0.8 12 Tra Vinh 65,330 93.8 9,150 11.0 150,293 2.7 13 Vinh Long 14,281 20.1 5,426 5.4 12,766 0.3 MRD 801,730 62.8 352,332 16.9 769,619 0.8 Source: South Water Resources Federation (2013). TABLE A2-2. Average number of laborers in agriculture, forestry, and fishery enterprises State Private Foreign Cooperatives enterprises enterprises enterprises 2010 401.8 87.5 18.4 511.6 2011 229.4 411.5 17.8 512.2 2012 303.6 17.0 20.4 409.2 Upper 2013 297.6 17.0 18.9 455.0 subregion 2014 276.8 18.1 18.3 434.3 2015 256.0 19.2 17.6 413.5 2016 252.2 18.0 16.9 439.1 2010 240.6 176.5 22.5 427.3 2011 248.7 139.5 22.1 415.4 2012 198.8 15.5 24.0 485.0 Middle 2013 214.7 15.1 22.8 541.4 subregion 2014 196.7 15.2 22.6 548.5 2015 178.8 15.2 22.4 555.6 2016 177.1 13.5 21.9 570.1 2010 221.2 380.1 16.7 283.1 2011 181.6 195.5 17.6 208.2 2012 179.7 25.8 17.8 73.4 Coastal 2013 189.9 23.0 17.0 167.5 subregion 2014 197.9 24.4 17.6 171.2 2015 205.8 25.9 18.2 174.9 2016 215.5 25.4 17.6 195.8 Source: Authors’ calculations from Enterprise Surveys. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 132 Alternative Scenarios and Policy Implications Appendix 3. Assessment of Main and Potential Livelihood Models of Three Mekong Delta Subregions FIGURE A3-1. Average, minimum, and maximum profit of key livelihood models in the Upper subregion in 2009-19 (VND, millions/ha/year) 300 3,000 250 Average Average 2,500 200 2,000 150 1,500 100 1,000 50 500 0 0 -50 -500 3-crop rice 2-crop rice Rice-vegetable Duck (1000 Rice + Lotus Rice + Capture -1,000 heads) Pangasius Fruit tree Vegetable shery Source: Calculated by authors from the household surveys in this research (2019). FIGURE A3-2. Average, minimum, and maximum profit of key livelihood models in the Middle subregion in 2009-19 (VND, millions/ha/year) 120 3,000 Average Average 100 2,500 80 2,000 60 1,500 40 20 1,000 0 500 -20 0 -40 -500 3-seasons rice 2-season rice Coconut Industrial crops Beef cow (1 head) Pig production (1 Chicken production Rice-duck/aquaculture -1,000 (1000 head) Rice-vegetable Vegetable Fruit tree Pangasius Coconut + Shrimp head) Source: Calculated by authors from the household surveys in this research (2019). Appendices 133 FIGURE A3-3. Average, minimum, and maximum profit of key livelihood models in the Coastal subregion in 2009-19 (VND, millions/ha/year) 300.00 Average 8,000 Average 250.00 7,000 6,000 200.00 5,000 150.00 4,000 3,000 100.00 2,000 50.00 1,000 0 0.00 -1,000 -50.00 -2,000 2-season rice Rice-shrimp Sparsed aquaculture/ intensive/Intensive Super-intensive Improved sparsed aquaculture aquaculture aquaculture Semi- Source: Calculated by authors from the household surveys in this research (2019). FIGURE A3-4. Role of women in decision-making in agricultural production in the Upper subregion (%) 0.00 10.00 20.00 30.00 40.00 50.00 60.00 Getting loans 46.74 Inputs/output trading 20.00 Farming technique 27.34 Hiring labor 14.17 Decide buyers 50.00 Selling main products 54.00 Switch agricultural activity 38.75 Looking for new non-farm income 49.47 Source: Authors’ calculations from survey data in this research. Note: This is the answer to the question “On a scale of 0 to 100, what is the weight of women’s voices in the making of these following decisions?” Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 134 Alternative Scenarios and Policy Implications FIGURE A3-5. Role of women in decision-making in agricultural production in the Middle subregion 0.00 10.00 20.00 30.00 40.00 50.00 60.00 Getting loans 44.29 Inputs/output trading 39.58 Farming technique 30.11 Hiring labor 30.44 Decide buyers 51.15 Selling main products 52.86 Switch agricultural activity 36.04 Looking for new non-farm income 44.71 Source: Authors’ calculations from survey data (2019). Note: This is the answer to the question “On a scale of 0 to 100, what is the weight of women’s voices in the making of these following decisions?” FIGURE A3-6. Role of women in decision-making in agricultural production in the Coastal subregion 0.00 10.00 20.00 30.00 40.00 50.00 60.00 Getting loans 42.50 Inputs/output trading 32.69 Farming technique 30.80 Hiring labor 29.20 Decide buyers 51.07 Selling main products 53.33 Switch agricultural activity 36.54 Looking for new non-farm income 38.80 Source: Authors’ calculations from survey data (2019). Note: This is the answer to the question “On a scale of 0 to 100, what is the weight of women’s voices in the making of these following decisions?” Appendices 135 TABLE A3-1. Social aspects of key livelihood models in the Upper subregion (1 ha/year) Maximum Production Number of Contribution of Total no. number of stages permanent the livelihood Livelihood models of work- laborers requiring high- laborers model in total ing days required at skilled labor required income (%) peak time Key livelihood models 3-crop rice 102.9 7.0 0 0.3 71.0 2-crop rice 53.6 7.0 0 0.2 55.4 Rice-vegetables 188.1 12.3 0 1.0 87.0 Specialized vegetables 346.2 11.2 0 1.7 89.0 Fruit 303.8 13.5 0 1.9 88.7 Pangasius 543.8 22.0 1 2.3 75.0 Potentially sustainable models Ducks (000 heads) 15.0 3.0 0 1.0 58.8 Rice + lotus 92.2 12.3 0 1.0 80.0 Rice + capture fish in 81.3 12.3 0 2.0 70.0 flood season Source: Prepared by authors based on household group discussions in this research (2019). TABLE A3-2. Social aspects of key livelihood models in the Middle subregion (1 ha/year) Maximum Production Number of Contribution of Total no. number of stages permanent the livelihood Livelihood models of work- laborers requiring high- laborers model in total ing days required at skilled labor required income (%) peak time Key livelihood models 3-crop rice 113.2 7.35 0 0.3 60.0 2-crop rice 59.0 6.72 0 0.2 38.8 Rice-vegetables 96.9 11.81 0 1.1 68.7 Vegetables 380.8 10.75 0 1.6 62.3 Fruit trees 334.2 14.31 0 1.9 70.1 Industrial crops 89.1 7.68 0 0.7 58.2 Pangasius 598.2 22.57 1 2.2 91.0 Beef cattle (1 head) 60.0 1.00 None 0.25 34.4 Pig production (1 head) 5.0 1.00 None 0.04 62.3 Chicken production 80.0 4.00 Vaccination 1.0 68.8 (000 head) Potentially sustainable models Rice-ducks 65.3 6.72 1.2 43.7 Rice-aquaculture 82.4 6.72 0 0 55.4 Coconut + aquaculture 326.7 6.30 0 0 66.2 Source: Calculated by authors from the household surveys in this research (2019). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 136 Alternative Scenarios and Policy Implications TABLE A3-3. Social aspects of key livelihood models in the Coastal subregion (1 ha/year) Maximum Production Number of Contribution of Total no. number of stages permanent the livelihood Livelihood models of work- laborers requiring laborers model in total ing days required at high-skilled required income (%) peak time labor 2-crop rice 61.4 6.1 0 0.3 68.7 Rice-shrimp 75.6 5.7 0 0.2 73.7 Extensive aquaculture/ improved 33.6 3.0 0 0.7 69.6 extensive aquaculture Semi-intensive/intensive aquacul- 356.2 4.9 1 1.0 91.5 ture Super-intensive aquaculture 898.2 7.6 1 2.3 95.4 Source: Calculated by authors from the household surveys in this research (2019). TABLE A3-4. Number of working days and percentage implemented by women in the Upper subregion Livelihood models Total working days (person-days/ha) % working days by women Key livelihood models 3-crop rice 82.9 24.5 2-crop rice 53.6 34.1 Rice-vegetables 188.1 46.7 Vegetables 346.2 62.5 Pangasius 543.8 8.1 Ducks (000 heads) 15.0 14.6 Potentially sustainable models Rice + lotus 102.2 15.4 Fruit trees 303.8 40.0 Rice + capture fishery in flood season 81.3 20.0 Source: Authors’ calculations from survey data in this research. Appendices 137 TABLE A3-5. Number of working days and percentage implemented by women in the Middle subregion Livelihood models Total working days % working days by women Key livelihood models 3-crop rice 113.2 29.5 2-crop rice 59.0 38.1 Rice-vegetables 142.3 46.7 Vegetables 380.8 62.5 Coconut 42.3 38.3 Fruit trees 334.2 25.4 Industrial crops 89.1 29.8 Pangasius 598.2 8.1 Beef cattle (1 head) 60.0 60-70 Pig production (1 head) 5.0 50.0 Chicken production (000 head) 80.0 70.0 Potentially sustainable models Rice-ducks 65.3 31.1 Rice-aquaculture 82.4 38.2 Coconut + aquaculture 326.7 18.4 Source: Authors’ calculations from survey data in this research (2019). TABLE A3-6. Number of working days and percentage implemented by women in the Coastal subregion Livelihood models Total working days % working days by women Key livelihood models 2-crop rice 61.4 30.2 Rice-shrimp 75.6 24.6 Extensive aquaculture/improved 33.6 42.3 extensive aquaculture Semi-intensive/intensive aquaculture 356.2 6.8 Super-intensive aquaculture 898.2 3.2 Source: Authors’ calculations from survey data (2019). Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 138 Alternative Scenarios and Policy Implications Appendix 4. Suggested Transformation Monitoring and Assessment Framework Below is a recommended list of monitoring and assessment indicators to monitor and facilitate the MKD’s agricultural transformation. They should be integrated into the provincial 5-year socioeconomic development plans and provincial 5-year agricultural and rural development plans for two periods, 2021-25 and 2026-30. Structural land use Upper subregion. The proportions of lands for the 3-crop rice, 2-crop rice, and rice-flood-based crop models in total agricultural land. Middle subregion. The proportions of lands for the 3-crop rice, 2-crop rice, rice-aquaculture, and horticulture models in total agricultural land. Coastal subregion. The proportions of lands for the closed water circulation shrimp production, intensive shrimp production, and rice-shrimp production models in total agricultural land. Infrastructure, science and technology, and logistics Upper subregion. The proportion of irrigation system suitable to rice-flood-based crops, the ratio of high-value seed used for production (rice, Pangasius fish, and horticulture), the ratio of mechanization, the ratio of agricultural land applying 4.0 technology, and the ratio of value-added processed agricultural products in total agricultural production. Middle subregion. The proportion of irrigation system suitable to rice-flood-based crops, the proportion of irrigation system suitable to rice-aquaculture, the ratio of high-value seed used for production (rice, Pangasius fish, shrimp, horticulture, and livestock), the ratio of mechanization; the ratio of agricultural land applying 4.0 technology, and the ratio of value-added processed agricultural products in total agricultural production. Appendices 139 Coastal subregion. The proportion of irrigation system suitable to shrimp production system, the proportion of irrigation system suitable to rice-aquaculture, the ratio of high-value seed used for production (rice, Pangasius fish, shrimp, horticulture, and livestock), the ratio of mechanization, the ratio of agricultural land applying 4.0 technology, and the proportion of value-added processed agricultural products in total agricultural production. Economic aspects Value added in agricultural production, profit per ha of land (cultivation, forestry, and aquaculture) or profit per unit of production (livestock), labor productivity in the agricultural sector, income per household, the proportion of production via complete value chain/business-farmer link/ proportion of sale via contract farming, average land area per household, the proportion of production via collective groups and cooperatives, the number of production support centers and cluster-cum-logistics hubs built and operated, the proportion of production going through these production support centers and clusters, and the number of enterprises operating in agriculture and agriculture-supported industries. Social aspects The proportions of migrants out of and into the MKD, the number of new jobs and new high-skill jobs created in the agricultural sector and food sector in the MKD, the number of cooperatives and farmer groups, and the number of commodity boards and associations. Environmental aspects The proportion of agricultural production area applying environmentally friendly practices in total agricultural land (crops, aquaculture, and integrated systems), the proportion of production units applying environmentally friendly practices in total production units (livestock), the average amount of chemical pesticide and fertilizer per ha (crops, aquaculture, and integrated systems), the average amount of surface water used per ha (crops, aquaculture, and integrated systems), and the average amount of groundwater used per ha (crops, aquaculture, and integrated systems). Gender aspects The proportion of female farmers participating in training courses and vocational programs out of the total trained farmers, the proportion of female residents migrating out of and into the MKD, female farmers’ income versus male farmers’, and satisfaction rate of female. Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 140 Alternative Scenarios and Policy Implications HONG DUC PUBLISHING HOUSE Vietnam Lawyers Association Hong Duc Publishing House • Address: 65 Trang Thi Street, Hoan Kiem District, Hanoi • Email: nhaxuatbanhongduc@yahoo.com • Tel: (84 24) 3 9260024 – Fax: (84 24) 3 9260031 Publishing Manager: Bui Viet Bac, Director Editor: Nguyen Thi Phuong Mai Cover photo: Cao Thanh Long © The World Bank Print 100 copies in English, size 20.5cm x 28.5cm at Xuan An printing industrial development and investment company limited. Address: 199 Quan Tho 1, Cho Dua, Dong Da, Hanoi, Vietnam. Publishing permit: 528/QĐ-NXBHĐ. Publishing registration plan: 2940-2022/CXBIPH/02 - 87/HĐ. ISBN: 978-604-380-546-8. Completed and archived in 2022 Agricultural Land Use and Sustainable Livelihoods in the Mekong Delta: 142 Alternative Scenarios and Policy Implications With support from: 8 Dao Tan Street, Ba Dinh District, Hanoi, Vietnam Telephone: +84 24 37740100 Facsimile: +84 24 37740111 Website: www.dfat.gov.au 8th Floor, 63 Ly Thai To Street, Hoan Kiem District, Hanoi, Vietnam Telephone: +84 24 39346600 Facsimile: +84 24 39346597 Website: www.worldbank.org/en/country/vietnam