Meandering to Recovery Post-Nargis Social Impacts Monitoring Ten Years After Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After World Bank Enlightened Myanmar Research Foundation Andaman Research & Advisory With financial support from the World Bank and the Global Facility for Disaster Reduction and Recovery © 2018 International Bank for Reconstruction and Development/The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclu- sions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemination of its knowledge, this work may be reproduced, in whole or in part, for noncommercial purposes as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Photos: © Enlightened Myanmar Research Foundation. Further permission required for reuse. Cover design: © Shepherd Incorporated. Further permission required for reuse. Contents Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii Abbreviations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi Chapter 1: Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Myanmar in 2017: The Context of the SIM 5 Research. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 Post-Nargis Recovery through Four Rounds of Research. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 Chapter 2: Livelihoods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Farming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6 Fishing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12 Labor. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15 Chapter 3: Credit, Debt, and Land . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Credit and Debt . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Land . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26 Chapter 4: Social Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Inter-Household Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 Relations within Village Groupings. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Relations across the Village. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31 Inter-Ethnic and Inter-Religious Social Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33 Factors Driving Changes to Village Social Relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Introduction of Social Protection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 Chapter 5: Leadership and Institutions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Leadership in the Delta. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37 Formal Leaders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Informal Leaders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 iii iv Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Chapter 6: Infrastructure and Housing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Village Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 Inter-Village Infrastructure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Village Housing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Chapter 7: Recovery and Resilience. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Recovery. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Resilience. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Preparedness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Chapter 8: Implications for Future Aid . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 For Future Post-Disaster Aid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 For Future Development Assistance to the Delta. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62 Chapter 9: Post-Disaster Social Impacts in the Literature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65 Chapter 10: Final Reflections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 Annex 1: Social Impacts Monitoring Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75 Annex 2: Location of the Sample Villages . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79 List of Tables Table 1: Livelihood Calendar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 Table 2: Farming Conditions by Township, 2013 and 2017. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Table 3: Farming Conditions by Degree of Cyclone Affectedness, 2013 and 2017. . . . . . . . . . . . . . . . . . . . . . . . . . 8 Table 4: Average Cost of Production by Acre . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 Table 5: Monsoon Paddy Yield Since before Nargis by Degree of Affectedness. . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Table 6: Monsoon Paddy Yield Variability by Degree of Affectedness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Table 7: Average Reported Price of Monsoon Paddy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Table 8: Variance in Harvest Prices over Time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 Table 9: Yields of Hilsa. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Table 10: Prices for Hilsa . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Table 11: Fishing Conditions by Degree of Affectedness over the Last 5 Years. . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 Table 12: Changes in Wages 2007–2017. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16 Table 13: Outmigration by Degree of Affectedness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Table 14: Indebtedness by Degree of Affectedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21 Table 15: Sources of Credit by Livelihood Group . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Table 16: Indicative Debt Levels by Degree of Affectedness across Various Livelihood Groups. . . . . . . . . . . . . . 25 Table 17: Frequency of Borrowing against Land by Degree of Affectedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 Table 18: Village-Level Social Relations between 2013 and 2017, by Degree of Affectedness. . . . . . . . . . . . . . . 29 Table 19: Village-Level Changes in Social Relations since 2013, by Degree of Affectedness. . . . . . . . . . . . . . . . 30 Table 20: Perceptions of Local Formal Leaders by Degree of Affectedness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Table 21: Most Important Local Leader by Degree of Affectedness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Contents v Table 22: Leader Age and Education, 2013 and 2017, by Degree of Affectedness. . . . . . . . . . . . . . . . . . . . . . . . . 40 Table 23: New Leaders Following Local Elections. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Table 24: Quality of Key Infrastructure, 2013 and 2017 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Table 25: Good Quality of Key Infrastructure, by Degree of Affectedness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 Table 26: Number of Villages with Good Quality of Key Infrastructure, by Connectivity. . . . . . . . . . . . . . . . . . . 45 Table 27: Connectivity by Degree of Affectedness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 Table 28: Villages ‘Left Behind’ by Infrastructure Development. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 Table 29: Infrastructure Funding Source by Sector . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47 Table 30: Number of Infrastructure Projects in Villages by Funding Source and Degree of Affectedness . . . . . 48 Table 31: Recovery of Villages by Degree of Affectedness, 2017. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 Table 32: Importance of Various Factors for Household Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Table 33: Importance of Various Factors for Community Recovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54 Table 34: Resilience of Villages by Degree of Affectedness, 2017 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55 Table 35: Capacity of Households to Recover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 Table 36: Capacity of Communities to Recover by Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56 List of Figures Figure 1: Economic Standing of Villages, 2013 and 2017 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5 Figure 2: Average Credit Sources per Village by Degree of Affectedness. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22 Figure 3: Average Debt Levels over Time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 Figure 4: Highest Debt Year for Farmers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 Figure 5: Proportion of Funders to Inter-Village Infrastructure Improvement. . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 Figure 6: Poverty Rates in Myanmar by Region/State. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 Figure 7: Welfare Index in Ayeyarwady and Yangon Regions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 List of Case Studies Case Study 1: A Farmer Invests in Embankments to Save His Livelihood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11 Case Study 2: New Nets Lead to Improved Earnings for a Fisher. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13 Case Study 3: Entrepreneurship Improves Livelihood Outcomes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 Case Study 4: Sisters Work Hard Together . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18 Case Study 5: Family Connections Help a Household Improve Their Lives. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30 Case Study 6: A Female Villager Becomes an Evergreen Chairwoman . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32 Case Study 7: Instances of Religious Tension. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 Case Study 8: A Non-Tract Village Ignored by Their VTA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Case Study 9: A Village Tract Elects a Female VTA for the First Time. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Case Study 10: Improved Transportation Infrastructure Results in Improved Livelihoods . . . . . . . . . . . . . . . . . 47 Case Study 11: Home Improvements Thanks to Migration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 Acknowledgments This report was prepared by Markus Kostner (World Bank), comments, Reena Badiani-Magnusson for her contribution Min Min Han (Enlightened Myanmar Research Founda- on poverty in the Ayeyarwady Delta, and Andrea Wood- tion), and Samuel Pursch (Andaman Research & Advisory). house, Patricia Fernandes, Patrick Barron, and Rina Meutia, Min Min Han led the field research and data analysis. The for their unwavering support throughout the SIM series. research team comprised Su Su Hlaing, Phyo Thitsar Kyaw, Yi Monn Soe, and Thaw Zin (research supervisors) and War Without the Global Facility for Disaster Reduction and War Lwin, Myo Thu Htun, Zay Yar Phyo, Soe Thiha, Saw Aung Recovery, post-Nargis social impacts monitoring would not Thein Myint, Noe Noe Nwe Lin, Khine Zin Yu Aung, Maung have been possible; it financed field research of four out of Phyoe, Thaw Zay Thwin, Shwe Yamin Oo, K Zin Thaw, Naw five rounds, including this round 5. Its critical contribution Htee Eh Hsoe, May Myat Noe Aye, and Aye Thandar Kyaw to the understanding of the social effects of a major disaster (associate researchers). Myat Thet Thitsar (national proj- over time is gratefully acknowledged. ect lead) administered the study with the support of Thi Thi Win, Lai Yin Myint, Kyaw Soe Win, Ye Win Aung, Zin The authors would also like to express their deep gratitude Mar Lwin, and Ye Naing Htwe from Enlightened Myanmar to the many villagers who generously devoted their time and Research Foundation. shared their journey out of the aftermath of Nargis with the research team. It is our hope that aid providers in future The authors would like to thank peer reviewers Marga- post-disaster situations will remember their experiences ret Arnold and Zuzana Stanton-Geddes for their insightful and heed the lessons they have helped us learn. vii Abbreviations DRR Disaster Risk Reduction GAD General Administration Department K Myanmar Kyat LUC Land Use Certificate MADB Myanmar Agricultural Development Bank NGO Nongovernmental organization SIM Social Impacts Monitoring VTA Village Tract Administrator ix Summary On May 2, 2008, Cyclone Nargis struck Myanmar’s Ayeyar- 1. Socioeconomic conditions: This focus area exam- wady Delta and swept across the region toward Yangon. By ines the conditions of livelihoods and the local socio- the time the storm had passed, it had killed over 140,000 economy in the context of Nargis’ destruction and the people, tearing apart families, destroying homes, and shat- evolving context of the rural economy across Myanmar tering livelihoods. In the months and years following Nargis, over the past five years. It looks at the three main live- communities, supported by the national and international lihood groups (farmers, fishers, and landless laborers) aid community, worked to rebuild their lives and repair and at key issues such as debt, land, and housing and the devastation that the cyclone had caused. Homes were local infrastructure. rebuilt, paddy field walls repaired, and new fishing boats purchased. 2. Social relations and leadership: This area assesses how communities have dealt with both the long-term However, even as the process of recovery inched forward, social upheaval caused by Nargis and the more recent villagers have had to contend with new and diverse shocks (but no less dramatic) changes that have accompanied and changes that have both enabled and slowed their efforts Myanmar’s political and economic transition. It reviews to rebuild. Among others, climate change has led to unpre- local relations within and between different social and dictable weather, hampering livelihoods, while the migra- identity groups and examines village leadership and tion boom to Yangon and elsewhere has provided economic institutions. opportunity even as it has altered the local social fabric. 3. Recovery and resilience: New to this round of SIM, the These more recent issues have had a complex inter-­ final analytical focus area identifies what recovery and relationship with changes wrought by Nargis. As time has resilience mean for households and communities in the passed, they have become the primary concern of most Ayeyarwady Delta, what factors are most important in villages studied by the social impacts monitoring (SIM) the recovery process and in building resilience, and to research. But the long-term effects of Nargis remain vis- what extent villagers have had and have the capacity to ible, combining with newer issues to create new challenges, develop both. exacerbate old problems, and, in some cases, even hasten the recovery process. SIM 5 placed particular emphasis on understanding change over time, both since 2013 (when the SIM 4 research was By focusing on a panel of 40 Nargis-affected villages across conducted) and prior to Cyclone Nargis. As much as pos- time, five rounds of SIM have been able to track how village sible, SIM 5 draws causal links between exogenous events life has changed both post-Nargis and, in more recent years, (such as cyclones, other natural disasters, political change, as villagers faced both new challenges and continued recov- and national economic development) and household and ery from Nargis. This fifth round of SIM (SIM 5) provides community actions. a snapshot of village economic and social life in 2017 and analyzes change over more than nine years since Nargis. It assesses three main areas: xi xii Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Socioeconomic conditions found it necessary to invest in new boats and nets, as well as nets with a tighter weave, which present sustainability In 2017, most villages’ economic situations improved com- concerns. Given these different experiences, overall income pared to four years ago, largely irrespective of how affected from fishing across the SIM panel was flat between 2013 they were by Nargis. Livelihoods overall remain mixed, but and 2017. the lower wages and stagnant local economy found in SIM 4 have been replaced by improved employment opportunities, Landless laborers saw the greatest better market access, and the ability of most households to improvement in livelihoods from 2013 as access a new income stream if their primary livelihood fal- wage increases, new credit access, migration, ters. Major challenges to economic development remain, and small business opportunities all boosted though they are most closely linked to systemic issues of incomes climate, lack of infrastructure, and access to improved farm- Myanmar’s national migration boom greatly affected the ing methods rather than long-term effects of Nargis or an Delta, with many landless laborers moving to Yangon or inability to find a job. other parts of the country in search of improved economic opportunities. This contrasts to previous rounds of SIM Farming livelihoods were mixed, with higher where migration was primarily a coping strategy. Many of costs of production offsetting price increases the migrants are women, moving to Yangon to work in the and yields increasingly varied and dependent rapidly expanding garment sector. The outmigration from on climate patterns SIM villages was also a significant contributor to higher The cost of inputs for farmers has risen compared to 2013, wages, as farmers had to pay laborers more to ensure their driving the cost of production for small farmers’ monsoon crops were harvested. Within the village, especially those paddy to almost double compared to pre-Nargis levels. villages with good access to larger towns, landless house- Summer paddy (increasingly planted by all farmers who holds frequently reported using their expanded access to can afford it thanks to higher and somewhat more reliable affordable credit (via both government programs and micro- yields) saw cost of production per acre increase by over finance lenders) to invest in new income-generating activi- 30 percent. The increases have been driven by farmers’ use ties, including small grocery shops, motorcycles for taxi of better quality and more expensive seed and fertilizer, as services, and even mobile phone shops. well as much higher labor costs. Expanded credit access has been accompanied Prices for paddy harvest have increased as well, with SIM by higher debt; some villagers are caught in a panel farmers reporting an average price (per 100 baskets debt trap of monsoon paddy) of K6.3 lakh,1 up from K3.7 lakh in 2013. Debt levels have expanded significantly since 2013 among It is yield, however, that remains the most determinative all livelihood groups, and debt repayment continues to be and variable indicator of the state of farming. Depending on a struggle within highly and moderately affected villages. the weather, experience of flooding and pests, or quality of Large and medium farmers have seen the largest increases the land, yields per acre for villages lightly affected by Nar- in debt, a reflection of how expensive it is to finance a crop gis have varied from 23 baskets per acre in the worst year of and the uncertainty of the harvest. Increased debt levels the last five, to 55 baskets per acre in best year. also reflect the significant increases in credit available within the village. In previous SIM reports, credit access was Returns from fishing livelihoods are similarly limited, particularly for landless households seeking afford- mixed based on geography, but low fish stocks able (i.e., non-private moneylender) loans. By 2017, there continue to threaten the sector were several new low-interest government credit programs In villages near the open ocean where catch levels have accessible by most villagers, and microfinance providers stayed relatively constant, increased prices and improved proliferated across the SIM panel. access to markets have benefited most fishers. However, fishers in villages along the Delta’s rivers have experienced Land consolidation has slowed, and for most significant catch declines—a change attributed by fishers to farmers, selling land is a last resort rather than unpredictable weather, increased river pollution, and the an economic opportunity presence of large fishing boats from the towns on their tra- SIM 4 reported that economic strains were driving more ditional fishing grounds. land transactions and more land inequality; in 2017, these processes slowed. Even farmers who are deeply in debt seek The costs associated with fishing have also expanded, as to retain their land as long as possible, knowing one or two license fees in some areas have gone up and fishers have successful harvests would allow them to escape the debt trap. When land rights are purchased, the purchaser tends to be larger farmers within the village or from a neighboring 1  One lakh equals 100,000 Myanmar kyats (K). Summary xiii village, though researchers found a notable increase in land- and the introduction of local, albeit indirect, democracy, all less households (generally young couples or households affected social relations within SIM villages. Overall, social with remittance income) seeking to invest in land for the relations, defined as the level of unity within the community, first time. Land prices vary significantly by village but also ability to work together, and trust in community leadership, by where the land is located. Less flood-prone land situated have coalesced in the middle. Villages, especially those near an access route could fetch five times the price of low highly affected by Nargis, that in 2013 had poor social rela- quality land. tions have seen some improvements, while others assessed to have good relations in 2013 (especially those moder- Transportation infrastructure and connectivity ately affected) have experienced a decline. This reflects the have improved since 2013, but many villages uncertainty that exists within many villages as their country continue to lack basic social infrastructure experiences massive change. Road infrastructure has improved across the Ayeyarwady Delta, but as many SIM villages are accessible only by boat, Village activity groups have evolved under the these improvements have had a limited effect. Of the 40 vil- transition toward democracy, though some lages in the SIM panel, just 15 have road access. All but have struggled to maintain membership and three villages have their own school buildings (at least at relevance the primary level), but many are in poor condition as vil- The self-reliance of the Delta villages and focus on self- lages do not receive funds for maintenance. Health access help as a necessary form of community activity has been varies and ten communities have neither a village-level well covered throughout the SIM series, and those instincts rural health center nor easy access to one nearby. For all remained strong in 2017. Under new legislation, villages are SIM villages, the most dramatic improvement in connectiv- not required to have cookie cutter standing committees, ity has been the introduction of mobile telephone services. and organic groups have taken their place. Free funeral ser- In 2013, mobile phones were virtually unheard of across vice organizations and other local charities have become Myanmar, but in 2017 all villages had access to 3G services particularly common, while independently created microfi- and mobiles were ubiquitous across the panel. nance groups have recently emerged in three villages. These village-directed groups sit alongside committees linked to Housing quality has improved significantly donor development programs which are now ubiquitous across the SIM panel, though a few notable across the Delta, and self-directed economic groups, such exceptions exist as the agricultural labor groups, are now present in over a A large majority of research respondents reported that their quarter of villages. houses are now of better quality than before Nargis, and that in recent years they have worked to improve them. Those Other groups have declined, however—for example, youth who reported the most significant improvements tended groups which have suffered from a lack of membership as to have either remittance income from a migrated house- young people migrate out in search of economic opportu- hold member, or multiple local income sources. In a small nity. For those that remain, there has been an increase in the number of communities highly affected by Nargis, villagers number of women included in their leadership. Also experi- reported that they did not see the point of rebuilding, as encing a decline were disaster risk reduction committees, a future Nargis-like storm would destroy them all. In these now present in just four villages across SIM, reflecting both villages, those households that could afford it invested in the lack of young people who had traditionally been active homes or property outside the village. Desire for better liv- in these committees and the fact that most villages have ing conditions, realization of the importance of a sturdy not experienced a major storm in recent years and see little home during a storm, and social pressure of comparing urgency for disaster risk management activities. homes with their neighbors were the main reported drivers of housing improvements. In some villages, there was an Relations between religious and ethnic groups increasing number of households taking on debt to improve are varied but overall have declined, due to or rebuild their home. both post-Nargis tensions and an increasing distrust between groups in recent years SIM includes several mixed-ethnicity (Karen and Bamar Social relations and leadership Buddhist) and mixed religious (Muslim or Christian and Buddhist) villages. In both cases, despite all respondents By 2017, the negative social effects of Nargis and the aid reporting no major challenges to local social harmony, effort that followed were largely dispersed, replaced in most there were more frequent reports of tension between the villagers’ minds by the forces affecting social change across groups. In Buddhist-Muslim villages, the communities do Myanmar. Migration, national political change, a fraying of still attend each other’s major religious festivals, but they the tight-knit social bonds that tied village groups together, live in de facto segregation, do not hire laborers from the xiv Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After other religious group, and rarely cooperate on village devel- While villages have recovered, few are resilient opment activities. In a small number of villages, Muslims in the face of another disaster: just six of the reported they had been prevented from building a mosque SIM panel villages are defined as ‘resilient’ and or practicing their religion freely and openly. Karen-Bamar eleven more as ‘somewhat resilient’ villages experienced similar issues, with the Karen seen by SIM sought to develop a working framework for resilience the Bamar as unified and self-supporting while the Bamar in the Ayeyarwady Delta and then to assess what factors (often a minority in these villages) suffered. This links back increased, or compromised, resilience. Based on existing lit- to the post-Nargis period when the broader Christian com- erature and the experience of previous rounds of SIM, resil- munity in Myanmar provided aid only to the villages’ Chris- ience in the Delta was defined as the ability to both reduce tian households. the impact of a disaster and recover more rapidly when a disaster strikes. Following local elections in late 2015, village tract leaders have consolidated power at the The most important factors that affect a household’s ability expense of village-level headmen, though to recover faster in the immediate aftermath of a disaster religious leaders and elders remain influential were identified as the survival of assets, local job oppor- Since 2012 legislation that reshaped local governance took tunities, remittance income, ability to manage household effect, administrative influence has become increasingly resources, aid, and the health and number of working fam- concentrated at the village tract level. The 2015 local elec- ily members. At the community level, diversified economic tions further consolidated this influence, and village tract activities within the village, credit access, leadership and administrators (VTAs) have an increasingly strong grip. social networks, good infrastructure, and external assis- Nevertheless, informal networks are still important, espe- tance were the factors most important for recovery. cially for those with the need to maintain a relationship with the township level, and have been enabled by the spread The ability of households and communities to of mobile technology. Villagers’ reported preferences for improve resilience exists across sectors, though village leader characteristics also evolved, with a stated is often informal or even a by-product of desire that leaders are educated, good managers, and speak actions done for other reasons English—though this was not borne out by local elections ­ Developing resilience consists of two broad baskets of which saw fewer high school graduates elected than in the activities: those that reduce a household’s or community’s first local election in 2013/14. Finally, there has clearly been exposure to the threat of damage and those that improve an expansion in the opportunities for women within local protection of people and assets. These activities can either governance. Several villages had multiple female 10 house- involve long-term planning that, over time, improves resil- hold heads (who are the electors of VTA), and one tract had ience or shorter preparation-type activities when a disaster chosen a woman to be the VTA. is already in the offing. Further, within the context of Ayeyar- wady Delta villages, all these activities happen at both the household and community levels. Recovery and resilience To reduce exposure to natural disasters, income diversifica- Across SIM’s 40 study villages, the research tion and, for those who were financially able to, purchasing team evaluated that 25, or almost two-thirds, land or a second home in town were the major steps taken by have now recovered from the effects of Nargis households in the Delta. Adding a small business or having a Unsurprisingly, a higher proportion of moderately (72 migrant family member working in Yangon lessened the risk percent) and lightly (80 percent) affected villages have that any one shock could wipe out their assets and income. recovered compared to 30 percent of villages that were Efforts at protection for farmers and fishers are limited, highly affected. The assessment of whether a village had given the relative immovability of their livelihood assets. recovered was based on the extent to which old liveli- Nonetheless, storing of paddy, seeds, cash, gold, and other hoods were now delivering sustainable positive returns, valuable assets and documents in protected places (bury- the effect of new economic opportunities (both in the vil- ing, storing in their home’s attic, or keeping them in town lage and beyond) on the household, the current state of were common methods) are perceived as important steps in social relations, perceptions of local leaders, and the state disaster protection. At the community level, and especially of village infrastructure. for villages heavily affected by Nargis or those with active leadership, evacuation plans, community food storage, and the planting of trees as windbreaks were all steps taken to reduce and mitigate disaster effects. Summary xv Long-term implications for aid both formal and informal. The gendered impacts of natural disasters must also not be overlooked. On the whole, the The experience of villages across the SIM panel over more post-disaster needs of women often do not receive the atten- than nine years since Cyclone Nargis demonstrates the tion they deserve. At the same time, post-disaster aid deliv- importance of post-disaster aid and some of the challenges ery mechanisms can lead to an increasing role of women such assistance has in helping households and communi- in local decision making. Consequently, aid that is not well ties rebuild. First, aid is not apolitical and can have lasting planned or executed is likely to create, or exacerbate, social effects, positive or negative, on local social relations. The tensions that can persist over a long time. importance of appropriate targeting of aid is clear from the experience of SIM communities, and the issues caused by A particular challenge is that disasters lead to greater debt, incorrectly targeted assistance have been a theme across and not only of the poor. As a matter of fact, while aid is SIM rounds. At the same time, the insistence of longer term generally focused on the poor, other livelihood groups (espe- aid projects to have women contribute to leadership struc- cially larger farmers) also merit recovery support as they are tures and decision making has been an important enabler of an important employer of local labor. Over time, disaster- new roles for women across various aspects of village life. related debt can lead to a more inequitable distribution of Given these longer term effects of aid, it is clear that, as land. Conversely, migration plays a critical role for house- soon as is practicable, aid delivery must take a long view in holds as it helps diversify income and makes households its delivery. Lifesaving aid is clearly the primary concern for less vulnerable to shocks. Finally, unless local disaster risk humanitarian actors, but as soon as possible, consideration management activities are linked with government struc- for how the livelihood and social contexts have changed tures and policies, they will not lead to resilience as com- (both directly from the disaster and due to broader shifts) munity memory of the event often fades over time. should play into aid delivery lest assistance be rendered irrelevant or misallocated.  Final reflections For future aid to the Ayeyarwady Delta, the importance of climate change as a determinative factor of future activities What would village life in the Delta be like today had Nargis is clear. Aid partners, government, and communities must not happened? It is not possible to offer an accurate answer work together to increase disaster preparedness, improve to this question for a variety of reasons. This does not mean, the resilience of local livelihoods, and empower community however, that long-term trends and patterns of Nargis’ direct development that equips villages to withstand future cli- and indirect effects could not be discerned. For this it is mate shocks.  important to understand the level of development of the Ayeyarwady Delta compared to the rest of the country. Post-disaster social impacts Despite being the rice bowl of Myanmar, the Ayeyarwady region has the third highest poverty rate in Myanmar; nearly in the literature one in two individuals in the region could be considered As a longitudinal study over nine years in a panel of disaster- poor. This development deficit is not a result of Nargis; affected villages, the Post-Nargis Social Impacts Monitoring however, available data indicate high rates of poverty in this is an important contribution to the field of post-disaster area (relative to other regions and states) years before the impacts analysis. Every disaster is different, affects indi- cyclone. The cyclone-affected townships in the Ayeyarwady viduals and communities in different ways, and is followed region fare even worse than the region’s average. All of them by a different recovery response. Not surprisingly, there- are in the lowest category with the least favorable wealth fore, many of the SIM findings are reflected in the literature, position. In other words, these townships are amongst the while others offer new insights. The most salient findings most disadvantaged in a region that is one of the most dis- include the following. advantaged in the country. Recovery is a continuous process, yet the direction of recov- A comparison of highly and lightly affected villages helps ery cannot be predetermined with any certainty. Disasters to pinpoint some noteworthy patterns. The differences in also deepen poverty and can lead to lasting structural debt, leadership, and connectivity contribute to a marked changes in the local economy. Any aid effort, therefore, difference regarding recovery, with lightly affected vil- ought to anticipate the need to adapt continuously. Two lages much more recovered than highly affected ones. Of critical ingredients to recovery and resilience that require these factors, debt is most closely related to Nargis, as the the attention of aid providers as well are social capital both cyclone broke the debt-harvest-repay cycle that had been within and across communities and with the local adminis- sustaining the main livelihood, farming, in the Delta. Other tration, and the quality and commitment of local leaders, factors that have shaped recovery would be at play regard- less but were accelerated by Nargis, such as mechanization, xvi Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After environmental degradation, and weakening social relations. the country’s economic and political transformations are At the same time, post-disaster aid helped not just to build offering. In other words, though unquantifiable, Nargis left back, but to build back better, and thereby contribute to a clear and lasting mark on the economic prospects of the some reduction in the area’s development deficit. villages it destroyed. Sadly, this is a common outcome of any major natural disaster. And yet, the SIM series has shed It can be surmised that the sample villages would still be poor unprecedented light on a recovery process from devastation today without Cyclone Nargis, but that they would be less in that has had as many twists and turns, and as many paths, debt and have more assets at their disposal, which would as the Ayeyarwady Delta itself. enable them to take fuller advantage of the opportunities 1Introduction Myanmar in 2017: The Context local socio-economy, social relations, and village institu- tions were evolving—both as a result of Nargis and due to of the SIM 5 Research the effects of aid provision. SIM research was undertaken in 40 villages spread across the Delta to provide in-depth The immediate effects of natural disasters are clear. Lives information on how village life was changing after the storm. are lost or torn apart, homes and village infrastructure are devastated, and both individual and community assets are Over the following two years, three rounds of SIM were com- damaged, in some cases beyond repair. In the twenty-first pleted under the auspices of the Tripartite Core Group. The century, these effects are seen all over the world, seared into first round, completed in November 2008, had assessed the consciousness of humanity. Response and assistance the immediate effects of Cyclone Nargis and the initial generally follow rapidly, with the country’s own government, response. SIM 2 was undertaken one year after Nargis from international governments, and a vast range of nongovern- May to June 2009, while SIM 3 was completed by May 2010, mental and intergovernmental organizations working to shortly before the Tripartite Core Group came to an end in bring aid to affected communities. July 2010. But disasters also have much longer term impacts, and these Recognizing that the longer term effects of major natural effects on communities persist long after images of suffering disasters are frequently understudied and little understood, have disappeared from the world’s media. A major disaster the Global Facility for Disaster Reduction and Recovery con- can alter the socio-economic trajectory of a household, a tinued the SIM series. SIM 4 was undertaken in April and community, and a region for at least several years, some- May 2013, in the same 40 villages studied in the earlier times several decades, and sometimes, forever. rounds. It began the process of understanding the continu- ing effects of Nargis on those who had survived the storm. When Cyclone Nargis hit the Ayeyarwady Delta on May 2, How had affected villages recovered economically? Has vil- 2008, it brought massive devastation. An estimated lage life retuned to what it had been before Nargis? What 140,000 people died in the storm, while damages and losses other factors shaped a village’s path to recovery? from the destruction are estimated at around US$4 billion. In the weeks following the storm international actors began With research in July and August 2017, SIM 5 returned to a wide-ranging humanitarian aid effort. To coordinate that these questions, looking back over the nine years follow- response, the Government of the Union of Myanmar, the ing Nargis’ devastation to understand how villages have Association of Southeast Asian Nations, and the United changed—and evaluating to what extent that change is Nations formed the Tripartite Core Group. directly due to Nargis, indirectly related to Nargis, or reflects other, broader, transformations in Myanmar. As in previous As part of its comprehensive monitoring system and to rounds, SIM 5 used a qualitative, village-level methodology ensure it responded to the changing needs of Nargis- of key interviews, focus group discussions, and participant affected communities, the Tripartite Core Group developed observation, revisiting the same 40-village panel across the Post-Nargis Social Impacts Monitoring (SIM) project. eight townships (Annexes 1 and 2). SIM aimed to understand the changing needs and priorities of the affected communities over time, as well as how the 1 2 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Alongside SIM’s core research questions, taking stock of impacts of the disaster. Undertaken within two months of changes to a village’s socio-economy, social relations, and the cyclone, the analysis was necessarily preliminary and development situation, SIM 5 added a new component in tentative. It developed a set of hypotheses on forms of an effort to reflect on how Nargis, and the ensuing deliv- social impact that might play out in the post-Nargis period. ery of aid, altered the trajectory of communities and their SIM tracked these issues over time. Field visits during the recovery process. It has also sought to understand what the assessment observed a high level of unity and social cohe- concept of resilience means in the context of Nargis-affected sion among survivors, who had no doubt been brought villages. It examined how villages have adapted to changes together by their common efforts to survive and rebuild. in the environment in the face of climate change, how they prepare for natural disasters, and what factors are most The analysis also identified the risks of redistributing land important for enabling faster recovery when disaster strikes. away from small-scale farmers to those with larger holdings and of loan-based responses that would require already Since 2010, Myanmar has undergone major political, social, indebted villagers to take on additional debt. Furthermore, and economic changes. The dissolution of the State Peace the analysis underlined the importance of working with and Development Council in March 2011, began a multiyear informal local institutions and local leaders. Based on this process of change that included national elections in 2015, analysis, the joint assessment report included in its guiding several local and by-elections, as well as significant eco- principles the need to involve communities at all stages of nomic reforms that have opened up the country to foreign the relief process, including decision making and feedback business and investment. Together, these moves signaled on the quality of the relief and recovery effort; maximizing that the country was steadily moving toward democracy the use of local initiative, resources, and capacities; and and a market-oriented economy, signals reinforced in 2015 building the capacity of local communities to participate in by the historic landslide election of Aung San Suu Kyi’s the relief and recovery effort. National League for Democracy. All along, the economy has been growing rapidly. Dashed hopes of a speedy recovery: Some of these broader issues directly affect SIM villages while others contribute only indirectly to the context SIM 1 within which villagers live and work. SIM 5 does not seek SIM 1, conducted six months after the cyclone, noted that to explicate these broader changes nor does it offer policy relief and recovery assistance had reached even the most prescriptions. Instead, by returning to 40 villages across remote villages, though levels of aid varied between and the Ayeyarwady Delta and giving villagers the opportunity within villages. Aid appeared to be helping villages recover— to reflect on their social and economic experiences since there was a link between the amount of aid received and the Nargis, SIM aims to improve global understanding of the speed of recovery. long-term effects of natural disasters and the role of disas- ter relief. However, the level of cyclone impact was a larger determi- nant of recovery. Even where larger amounts of aid were being delivered, many moderately or highly affected villages were not recovering quickly. This suggested that the scale of Post-Nargis Recovery through aid and/or its effectiveness was still insufficient. Although Four Rounds of Research relief assistance had reached all villages, much more assis- tance was needed for communities to recover, particularly Villages’ paths toward recovery have evolved over time, in the form of cash grants. Moreover, villagers thus far had differing not only by degree of affectedness, but also by had few opportunities to influence the aid effort, which had location and by type of livelihood. This section summarizes led to the provision of some kinds of aid not adapted to local salient findings from the first four rounds of SIM to provide a needs. longitudinal snapshot of how the village context has evolved across the panel over more than nine years. Nargis had caused farming productivity and crop yields to decrease significantly. Reductions in harvests were driving increased debt burdens for farmers, many of whom could Setting the stage: The Post-Nargis not repay old loans and also had to borrow for consump- tion. Fishing was severely affected too, as fishers who had Joint Assessment lost their fishing gear were struggling to recover. This lack of The Post-Nargis Joint Assessment was the first post-disaster fishing equipment affected those at every point in the fish- assessment that included a specific analysis of the social ing value chain. Introduction 3 Moreover, Cyclone Nargis drastically reduced the opportuni- Casual laborers also continued to struggle. The inability of ties for paid work for laborers, who faced reduced demand big farmers to cultivate their land as intensely as before fur- for their labor from the bigger farmers and fishers who ther reduced job opportunities for casual labor. Moreover, would normally employ them. Laborers received food aid some farmers and fishers themselves became casual labor- but less livelihood support compared to other groups. As ers, so competition increased for the jobs that did exist. a result, many were facing immense difficulties in getting Debt levels were still rising, and interest rates remained by. The most recurrent theme was an increasing debt bur- high. Farmers began to lose land to moneylenders as they den. In every village studied, villagers were worried about failed to meet their debt payments. being able to meet their loan repayments and having enough money for consumption in the following year. Yet, social capital was still strong overall. Young people con- tinued to play an active role in relief and recovery activities, Cyclone Nargis had a major impact on socioeconomic life such as repairing and renovating schools and pathways. in Delta villages. However, despite its immense shock, Formal leaders played a strong role in aid-related affairs, social relations and local capacity remained strong. Villag- though the roles of village elders in aid affairs had decreased ers worked together to overcome immediate challenges, somewhat since SIM 1. Similarly, the role of religious leaders which in turn strengthened social relations. Communities had evolved, with only Christian and Muslim religious lead- remained resilient and functioning. Nonetheless, SIM 1 con- ers engaged in day-to-day recovery activities. There was an cluded that if people’s livelihoods and village economies did increasing number of instances where aid-related tensions not begin to recover soon, there would likely be profound weakened social relations. longer term impacts, such as migration out of Delta villages and a tearing of the social fabric. If people could not break out of the Nargis debt trap, there was also a risk of longer Deceptive indications of a recovery: term redistribution of assets from many to a few. SIM 3 By 2010, the steady decline in aid since the cyclone was Still a mixed picture: SIM 2 now accompanied by a significant change in aid provider priorities. The decline of credit provision and fishing inputs Cyclone survivors continued to prioritize livelihood assis- was particularly noteworthy. Support for small and medium tance. However, a year after the cyclone, the needs of farmers remained widespread, but there were concerns that affected communities had evolved, with villagers also priori- levels of aid were insufficient to help restore livelihoods. tizing health, education, and small-scale community infra- Overall, SIM 3 noted a significant disjuncture between aid structure. This reflected a shift from emergency to longer provision and the needs of different groups in the commu- term recovery priorities. As in SIM 1, aid in the form of cash nity. Aid providers continued to be the main decision mak- or affordable credit appeared to be more effective—and was ers, determining both the type of project and the process preferred by villagers—than in-kind assistance. of implementation. Accountability relationships between aid providers and the villagers remained limited, with little However, aid levels dropped sharply and were too low to evidence of transparency measures working effectively at enable cyclone survivors to adequately recover their live- the community level. lihoods. Furthermore, aid providers continued to make aid-related decisions themselves, and aid distribution was There were signs of a recovery in the farming sector. Farming formalized through official leaders and village emergency yields had rebounded since the disaster by over 15 percent committees that were established for the purpose of dis- on average and by many more farmers growing both mon- tributing aid. Villagers often lacked clear information about soon and summery paddy. Simultaneously, however, inten- aid, which led to cases of (perceptions of) misuse and local sive investments in the fishing sector during the previous conflict over delivery. year led to overfishing and a drastic reduction in output and employment. As a result, many fishers lost their traditional There was little overall progress toward livelihoods recov- livelihoods and became casual laborers at a time when farm- ery. Socioeconomic conditions remained challenging and ers were reducing their demand for labor even further. In were exacerbated by the wider economic environment. almost half of SIM villages, laborers became significantly Farmers were unable to afford adequate inputs and culti- worse off than other villagers. vated their land less intensively. There appeared to be some progress with restarting fishing, though as most aid pro- Debt continued to undermine the prospects for recovery. vided was small-scale fishing gear, this enabled fishers to Across all occupational groups, the average maximum debt earn a subsistence income, rather than reviving the fishing across villages two years after Nargis was higher than before value chain. 4 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After the cyclone. Farmers went to great lengths to remain current affected villages in the sample, only three moderately on at least their interest payments and were selling assets affected ones did not have repayment problems. Indebted- other than land in order to do so. The story of debt in the ness was directly related to the degree of cyclone affected- Delta two years after Nargis was thus one of continued asset ness. Coping strategies had not changed over the previous depletion but not yet of widespread default. three years but were being used more intensively. Migration continued to increase as a coping mechanism and as a way Due to a strong social fabric, Ayeyarwady Delta communities to escape job scarcity. Other forms of coping had potentially stayed resilient despite the challenges. In most aspects of negative social and health consequences. village life, there was little change and social relations were good. There were indications that women were becoming In about three-quarters of villages, social relations were more empowered through their increased involvement in considered good or fair, including in most highly affected aid-related committees. At the same time, the role of youth villages. Where social relations were good, the community in the aid effort decreased. Some signs of tensions between was organized and villagers undertook collective activities, villagers and their leaders were reported, most of which mostly socioreligious tasks and regular community works. related to aid provision. Where social relations were poor, villagers were divided into different groups, and communal activities occurred only within these groups. In most cases, poor social rela- Livelihoods fall backwards: SIM 4 tions were a remnant of the aid effort. Five years after Nargis, Delta villages still found them- At the same time, however, villagers had less time and fewer selves in a dire economic situation. This situation could not resources for social interaction. They also felt that relations be attributed to the cyclone alone; multiple other factors were worse than before Nargis in about half of the villages. were now influencing villages’ paths to recovery, with more The passing away of religious leaders during Nargis contrib- recent climatic shocks playing key roles. Highly affected vil- uted to strained social relations five years on, especially in lages were more exposed to these events. The combination villages with both Muslims and Buddhists. of Nargis and these more recent events had affected liveli- hoods in some villages to such a degree that they appeared The role of women and youth continued to evolve. With aid to have lost their ability to self recover. In 2013, most of the having largely ceased, momentum toward women’s empow- 40 sample villages were struggling and had yet to recover erment had evolved along new tracks. In some cases, women socioeconomically. Only six lightly and moderately affected were making inroads in broader village affairs. In two vil- sample villages were considered to be in good standing, lages, women were elected as ten household leaders in while 16 were in poor standing. 2013, the first time this had happened. At the same time, SIM 4 saw a resurgence of youth engagement. There were Villagers in over one-third of the villages were caught in more youth-led activities in many villages, potentially a a debt trap, unable to repay even the debts that they had reflection of the recent elections that had brought younger owed from before Nargis. Of the 30 highly and moderately leaders to power. 2 Livelihoods Previous rounds of SIM research highlighted the variabil- was ‘good’ or ‘fair’ increased from 23 in SIM 4, to 38 in SIM 5 ity of villages’ economic situations in the years following (Figure 1). While farmers in particular face high income vari- Nargis, with communities experiencing some improvements ation due to weather shocks, all livelihood groups reported one round and then falling back the next. For example, farm- positive changes compared to 2013.3 ing returns captured by SIM 3 suggested the sector was on a positive trajectory, but by SIM 4, farmers were struggling This improvement in economic standing for the SIM panel again. SIM 4 also found that an increased labor supply, the reflects a similar slow, but positive, evolution of socioeco- result of fishers and farmers leaving their livelihoods, had nomic conditions across Myanmar. In the Delta, Cyclone led to lower wages and limited employment opportunities. Nargis is no longer the defining livelihood issue for most communities, as more time- and location-specific issues In contrast, SIM 5 research found that the economic situa- have taken a central role. Other changes, including the tion improved for most villages, across levels of affected- evolution of market access, price changes, and especially ness. The number of villages where the economic situation2 Figure 1: Economic Standing of Villages, 2013 and 2017 14 12 10 8 6 4 2 0 SIM 4 SIM 5 SIM 4 SIM 5 SIM 4 SIM 5 Good standing Fair standing Poor standing Highly affected Moderately affected Lightly affected and easy access to microfinance to invest in income-­  generating activities; “poor,” few local job opportunities exist, villages with 2  Economic standing is defined as: ”good,” a notable increase in many households without multiple working family members, and remittance levels and local job opportunities, i.e., most households limited access to credit meaning high daily interest rates. have multiple working family members and few households in the 3  Throughout SIM 5, time comparisons were made either to the village have high levels of debt; “fair,” most households have at period before Nargis or to the period following the summer paddy least one regular income-generation activity (i.e., a daily income), harvest of 2013. SIM 4 research took place following the Thingyan i.e., there is a medium level of migration compared to other villages, celebrations of April 2013. 5 6 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Myanmar’s internal migration boom, which has in turn An overview of farming in the Delta affected labor availability, reflect broader shifts affecting communities across the country. The primary crop across the Delta region is paddy rice, grown (depending on the location and on the farmer) across one or Economic activity for many of Myanmar’s rural communi- two harvest cycles each year. The monsoon paddy season is ties varies significantly depending on the time of year, and planted in June with the onset of the monsoon rains and har- where in the country the community is located. Weather that vested in October. Less reliant on additional inputs than the varies dramatically from season to season, fish spawning second summer season, the monsoon harvest also has higher patterns, and government regulations on fisheries all help prices and is the traditional primary crop for most farm- shape the Ayeyarwady Delta’s annual livelihood calendar. ers. The second ‘summer’ harvest is planted in November Table 1 provides an example of the types of economic activi- and then harvested between late January and early March. ties undertaken by different households over the course of Requiring much greater investment (particularly in fertilizer a year. This is an indicative rather than definitive map, but and labor costs), summer paddy makes up for this with the helps illustrate that while households may continue to have high yields that offset both the expense of the second crop a primary livelihood, in reality most seek diverse sources of and the generally lower prices available in January. Among income across a twelve-month period. The calendar is pre- farmers interviewed, they reported that yields in the most sented from May–April, in keeping with both the Myanmar recent season for monsoon paddy were approximately 45 traditional calendar (where New Year’s day is in mid-April) baskets per acre. For summer paddy, however, yields were and the planting and harvest cycles described by farmers. double that, with large farmers averaging over 90 baskets per acre.5 Within the 50 percent of SIM villages where sum- mer paddy is grown, the farmers who can afford to finance the more expensive summer crop do so, though often only Farming on their highest quality (and least flood-prone) land. SIM 4 research found that farmers faced poor conditions4 While farming activities are often perceived as a primarily in almost half (17) of the panel’s villages, primarily in com- male economic livelihood, within Delta households, women munities either heavily or moderately affected by Nargis make vital contributions to the success or failure of a sea- (Table 2). This reflected how villages that had been badly son’s harvest. While the division of labor varies, men are damaged by Nargis had been unable to recover to pre-Nargis generally responsible for the physical process of land prepa- levels. As a result, many farmers faced a debt trap. There ration and planting, while women often organize the vari- was also a significant difference between the worst-hit ous inputs required, including hiring labor and/or machines; townships along the coast—Labutta, Bogale, Pyapon, Kung- purchasing petrol, seeds, or fertilizer; managing cattle or yangon, and Dedaye—where farming conditions were mostly oxen; and in the case of informal borrowing, arranging loans. poor, and communities in townships were further inland. Formal loans from Myanmar Agricultural Development Bank SIM 5 research found a somewhat more positive, if highly (MADB) are in the name of the registered land user, generally variable, situation facing farmers where improving prices the male head of household. For women-headed households and the possibility of good yields are regularly undermined with land, the process is a little different, but women play a by weather shocks and poor quality harvests. more active role in supervising laborers in the fields as they must manage the process of land preparation directly. Table 3 captures a trend toward the middle as there was a clear reduction of villages with both good and poor con- ditions. In 2017, three-quarters of villages experienced fair farming conditions. Half of the highly affected villages as Production costs have increased well as one-quarter of moderately affected ones were able The costs associated with farming have consistently gone to improve their condition from poor to fair. up, thus harvest price increases have done little to improve the condition of most farmers (Table 4). Many farmers, especially those with more than a few acres 4  Farming conditions are defined as: “good” = access to improved of land, depend on daily wage agricultural labor during peak seeds, ability to grow both summer and monsoon paddy, alterna- times of planting and harvesting. A smaller number of laborers tive summer crops, and a high proportion of paddy land without are often hired seasonally to assist with land preparation flood risk either due to location or embankments; “fair” = good access to water for irrigation, access to improved seed, ability to invest in farming inputs, and a medium proportion of paddy land that was not flooded; “poor” = experiencing too much rainfall For the SIM study, large farmers are defined as those with ten 5  resulting in lots of fallow land due to flooding, lack of any embank- acres of land or more, medium farmers as those with between five ments, and high incidence of snail and other pests damaging crops. and ten acres, and small farmers as those with less than five acres. Table 1: Livelihood Calendar Activity/ Livelihood May June July August September October November December January February March April Group Fa Fis La Fa Fis La Fa Fis La Fa Fis La Fa Fis La Fa Fis La Fa Fis La Fa Fis La Fa Fis La Fa Fis La Fa Fis La Fa Fis La Farming Subsistence fishing Livelihood fishing Open sea fishing Agricultural labor Other daily jobs Seasonal migration Unemployed Note: “Fa” means farmers, “Fis” means fishers, and “La” means laborers. 7 8 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Table 2: Farming Conditions by Township, 2013 and 2017 Farming Condition (2013) Farming Condition (2017) Difference Township Good Fair Poor Good Fair Poor Good Fair Poor Bogale  5 4 1 6 2 1 1 –2 Dedaye  3 4 5 2 0 2 –1 Kungyangon  1 2 2 1 0 1 –1 Kyaiklat 3 1 2 2 –1 2 –1 Labutta  2 3 5 0 4 –3 Mawlamyinegyun 4 2 2 –4 2 2 Ngapudaw 2  1 1 4 –2 2 –1 Pyapon  2 2 4 0 2 –3 Total 9 14 17 3 30 7 –6 16 –10 Table 3: Farming Conditions by Degree of Cyclone Affectedness, 2013 and 2017 Farming Condition (2013) Farming Condition (2017) Difference Affectedness Good Fair Poor Good Fair Poor Good Fair Poor Highly affected 0  4  8 1  9 2  1  5 –6 Moderately affected 4  6  8 1 13 4 –3  7 –4 Lightly affected 5  4  1 1  8 1 –4  4 0 Total 9 14 17 3 30 7 –6 16 –10 Table 4: Average Cost of Production by Acre (in lakhs6) Monsoon Paddy Summer Paddy Farmer Size Before Nargis 2013 2017 Before Nargis 2013 2017 Large 0.8 1.1 1.6 1.2 1.6 2.3 Medium 0.7 1 1.4 1.2 1.4 2.2 Small 0.7 1.2 1.3 1.2 1.6 2.1 6 or crop care while the paddy is growing. Since SIM 4 One result of this search for alternatives has been the expan- and across all the villages in the panel, labor costs have sion of mechanization. The number of machines used in agri- increased significantly due to the outward migration boom culture, ranging from small, two-wheeled tractors to larger in recent years as well as the expansion of nonfarm jobs that harvesting and threshing machines has expanded signifi- have reduced the available supply of labor. cantly. The uptake in use has been driven by both farmers’ desire to reduce labor costs and the increasing availabil- The higher costs of hiring labor have had particular negative ity of financing arrangements, particularly hire–purchase effects on medium farmers, who are less able to fall back on schemes supported by government and/or donor programs. family labor (the coping strategy of choice for smaller farm- The purchase of agricultural machinery, especially larger ers) and do not have the financial options of larger farm- machines, also has a multiplier effect, as such equipment is ers who can more easily access capital, choose to rent out often rented out to other villagers, reducing labor costs for farmland, or even sell land to raise cash. The challenges of other farmers while helping the owners recoup their invest- paying for labor are such that farmers of all sizes have begun ment more rapidly. to seek alternatives that reduce their reliance on workers, even if they require larger investments. Moreover, the cost of production has been driven up by expanding fertilizer use. Farmers across the panel reported using more and more fertilizer to ensure that yields 6  One lakh equals 100,000 Myanmar kyats (K). Livelihoods 9 remained high, while some also have begun to use chemical of year, and location. The most common weather shock was weed killers and pesticides as an alternative to hiring labor- flooding, which villagers, especially those living along the ers to weed the paddy fields. Such new techniques, quite banks of larger rivers, reported was an almost annual chal- apart from any effect on the harvest, have negative environ- lenge. One village flooded every month when the tides were mental impacts, affecting water quality in local streams and high. Half of the SIM panel reported 2016 monsoon flooding rivers as well as fish stocks. Fishers across the SIM panel had negatively affected their harvest’s yield and quality. In often attributed (anecdotally) declines in catches to water eleven villages, some farmers told researchers that they had pollution from agriculture. As well as these enabling inputs, chosen to let parts or all of their land lie fallow during the seed costs have gone up as farmers are more likely to pur- 2017 monsoon (either from the beginning of the season or chase higher yielding seeds rather than rely on seeds from choosing not to replant after an early shock), as they did not their own harvests. want to risk taking out additional loans that they were not sure they would ever be able to pay back. Yields have been variable Other weather shocks included unexpectedly timed rain (i.e., before or after monsoon) which damaged crops, hail, Cost of production increases have not been met by consis- and, perhaps surprisingly, drought. The lack of rain in the tent improvements in yields. Instead, farmers have faced Delta outside of monsoon means that many communities extremely variable harvests due to exogenous factors. face water shortages in April and May, challenges that Table 5 shows how yields for monsoon paddy have evolved have worsened in recent years due to shorter (albeit more in the years following Nargis. SIM 5 data indicate that yields intense) monsoon periods. Pest shocks, associated with have generally recovered to pre-Nargis levels. However, this damaging flooding, have worsened outcomes further. does not capture the variability of farmers’ experiences. Table 5: Monsoon Paddy Yield Since before Nargis Prices by Degree of Affectedness (baskets per acre) Prices for paddy have increased since SIM 4, with farmers 2007/ reporting that monsoon paddy in 2016/2017 had an average Affectedness Pre-Nargis 2008 2013 2017 price of 6.3 lakhs per 100 baskets (Table 7).7 This compares Highly affected 47 22 32 46 to 4.3 lakhs per 100 baskets found in SIM 4 for higher grade Bay Gyar paddy and just 3.1 lakhs per 100 baskets for lower Moderately affected 44 31 33 42 grade Achon paddy. Part of the change is attributable to Lightly affected 45 37 42 47 the phasing out of lower quality paddy and farmers’ focus on better seeds. However, farmers also reported that bet- Farmers reported that for every good harvest they experi- ter access to markets and lower transportation costs were enced, they faced one or two where they had little or no positively affecting the price they received. return due to problems with weather or other shocks. As shown in Table 6, the difference between a good harvest’s yield and a poor one is significant, with yields collapsing Table 7: Average Reported Price of Monsoon Paddy almost 50 percent. Some farmers reported that in particu- (lakhs per 100 baskets) larly bad years, they do not even bother to harvest their crop as the cost of labor outweighs any potential return. Year 2007 2008 2013 2017 Price 4.05 3.15 3.70 6.34 Table 6: Monsoon Paddy Yield Variability by Degree of Affectedness (baskets per acre) Worst Yield in the Best Yield in the This positive picture for prices is, however, tempered by Affectedness Last Five Years Last Five Years two issues. First, there is significant variation from harvest Highly affected 29 47 to harvest. The variation in yield is shown in Table 6, and Table  8 shows the best and worst prices large and small Moderately affected 23 46 farmers have received for both monsoon and summer har- Lightly affected 25 55 vests since Nargis, as well as the spread between them. The Variable weather and its effect on both yields and har- Pre-2017 numbers are an average of prices for the two most com- 7  vest quality have the power to wipe out all positive gains. mon types of paddy, Bay Gyar and Achon. 2017 prices are an aver- Weather shocks were common across the SIM panel, with age of prices for all paddy types. This change reflects more seed varying levels of affectedness depending on the year, time diversity. 10 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Table 8: Variance in Harvest Prices over Time found that the quality of soil was an important determinant (lakhs per 100 baskets) of what a farmer could expect his land to produce. In SIM 5, in large part due to more frequent flooding across the Large Best Price Worst Price SIM panel, soil quality’s importance has declined and the Farmers since Nargis since Nargis Spread location of a farmer’s land (i.e., whether it is at a higher ele- Monsoon 7.3 4.9 2.3 vation away from major rivers) is now the most important Summer 4.6 3.3 1.4 determinant of yield. Land at higher elevations produced more reliable yields, even if the soil was not necessarily as Small fertile as some low-lying plots. As discussed in the following Farmers chapter, this change is reflected in land value, with plots Monsoon 6.9 5.1 1.8 deemed to be in good locations having significantly higher Summer 4.6 3.3 1.3 sale prices. Training data indicate that while in some years prices are high, there Trainings were identified as important to helping farmers is little consistency and farmers face frequent swings in the counter a lack of information available on improved farming amount they can expect to earn from their harvests. methods. While Myanmar does have an Agricultural Exten- sion Department, it is chronically understaffed and under- The second issue is harvest quality, which is linked to both funded. No village reported receiving assistance from the yield and price, but not fully dependent on either. For exam- department unless it requested advice on a specific issue, ple, some farmers reported that in some years their paddy such as one village which asked for help dealing with a snail harvest had average yields, but the quality was poor, gener- infestation (an issue the department was unable to resolve). ally meaning high moisture rates, discoloration of the rice, and higher risk of mold, resulting in low prices and a result- Instead, extension services and agricultural trainings fall to ing loss overall for their investment from that season’s crop. three groups. First, there are the many non-governmental organizations (NGOs) active in the Delta, many of whom have focused on improving the quality of the seeds used by Timing farmers.8 Such efforts have led to yield improvements, in some villages increasing the number of harvested baskets Most farmers continue to be unable to store paddy them- per acre from 35 to 55. Second, there are the companies selves and must sell their harvest almost immediately. This selling inputs to farmers. Many of these, especially fertil- means they are subject to the market, as described by the izer salespeople, also provide information on best prac- broker, at a time when paddy is most available and the price tices, though there are clearly conflict of interest issues at is lowest. A frequent complaint is that if they were able to play. Finally, there are a range of social enterprises seek- store the paddy themselves, farmers could sell later in the ing to commoditize agricultural extension via soil testing year when prices have improved from their post-harvest low kits, demonstration plots, and other extension products. point. Farmers reported that two or three months after the While farmers reported mixed results from all three types of peak harvest season, the price for paddy, especially the high- actors, improved seed was recognized as an important first est quality (i.e., export) varieties, could be 15–50 percent step, as was improved knowledge on how much fertilizer to higher, depending on the year, than what they are normally apply and when. able to sell their crop for. So, for paddy sold at 6 lakh per 100 baskets, the price 2–3 months later could be anywhere The availability of trainings is independent of the degree of from 7 to 9 lakh per 100 baskets. For lesser quality paddy Nargis affectedness. Of the twelve villages where extension varieties the price difference is less pronounced as brokers services and agricultural training were reported, four were are less interested in storing these types. highly affected, five were moderately affected and three were lightly affected. Factors contributing to farming outcomes Location Beyond challenges posed by weather shocks, variable yields, and fluctuating prices, the outcomes of farming are influ- 8  The multi-donor Livelihoods and Food Security Trust Fund has a enced by several other factors. Perhaps the most important large seed program working with multiple NGOs across the Ayeyar- of these is land location. Previous rounds of SIM research wady region. Livelihoods 11 Adaptions to evolving farming context Embankments In response to flooding and its associated effects on farming Changing farming methods livelihoods, farmers across SIM villages perceive a need for Across SIM villages, farmers are taking steps (such as leav- embankments that can withstand flooding and protect har- ing flood-prone land fallow, or attempting to build embank- vests. Until recent years (and as shown in earlier rounds of ments) in an effort to respond to the changing context for SIM), villages prioritized transportation infrastructure such their livelihoods, namely the progressively worsening cli- as roads and improved jetties that enabled improved market mate issues (as reflected by increased flooding and more access. pest problems). In SIM 5, however, when asked what infrastructure their vil- The other main livelihood adaptions have been attempts lage still needed, embankments were clearly a higher prior- to reduce reliance on hired labor such as altered farming ity than in the past. Across the panel, six villages have built practices (e.g., broadcasting seed rather than transplanting their own embankments since Nargis (at least two villages seedlings and increasing fertilizer use), seeking alterna- in the last two years), financed and constructed by (groups tive sources of labor (such as relying solely on family mem- of) farmers and serving to protect the group’s paddy fields bers), or investing in equipment or inputs that require fewer (Case Study 1). Another six villages have older embankments workers. built by the government before Nargis that protect both the village itself and the surrounding paddy fields. Of these vil- “The soil has become sour and we have to use more fertilizer; lages with embankments, six reported that they were in poor if we do not use it, the plants will not grow well.” condition and needed repair. Another five villages without embankments said they viewed them as a high priority. The —Farmer effect of embankments is clear: of the twelve villages that Case Study 1: A Farmer Invests in Embankments to Save His Livelihood Small farmer Ko Aung Tun owns two acres of paddy land and rents land from other farmers as well. He lives with his parents, both 65 years old, as his relatives have started their own families. Before Nargis, his parents bought five acres of farmland together with their relative, and the relative took three acres and his parents took two acres. In those years, his job was looking after the family’s flock of ducks while his parents farmed. In 2008, right before Nargis, he took over farming responsibilities. Nargis destroyed the paddy, leaving the family with K300,000 in debt for agricultural inputs. But in the months following Nargis, the family did not have problems with their basic needs thanks to external assistance. In 2009, the family started farming again but his parents’ health had begun to worsen and they could no longer work in the fields. Alongside his farming, he also worked as a casual laborer when the planting season ended, and as a carpen- ter. Once he even went to Rakhine state and earned K6,000 per day for two months as a carpenter, returning only when the planting season arrived. As his land is in an area that floods regularly, he can plant only in the summer season. So, during the monsoon season, he works as a permanent farming laborer and sometimes gets payment for the season from the farmers up front, though the total is never more than K300,000. By 2013, there were more credit sources available and Ko Aung Tun took on more debt. While he was able to repay loans taken from two suppliers, his third loan caused him problems. He often had to take a bridge loan from a private money lender with 5 percent interest per month in order to make the required payments every two weeks. In 2015, his parents’ health was getting worse still and they had to see the doctor in Pyapon. He took on even more debt that, even though he is working more and more, he cannot repay. Now, he has a debt of K 1.2 million to repay. However, Ko Aung Tun is hopeful that things will improve because of a new embankment around his paddy field. Last year, having witnessed a large farmer build protective embankments to prevent monsoon flooding, he built his own embankments around his fields. At the time of the SIM research the seedlings were in good condition. He was hoping that he would be able to repay some of his debts. The farmer said that two good harvests in a row would allow him to become free of debt. 12 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After currently have embankments, all but one reported that farm- friends. These can be further divided based on location. ing conditions in the village were good or fair, and all but Some, living in villages that dot the coastline, fish primar- three were also considered recovered. ily in near-offshore waters of the Bay of Bengal. Their catch tends to be larger fish and, as shown later in this chapter, Diversification and migration these fishers experience greater returns from fishing. The A wide variation in yields from harvest to harvest has also second group are fishers from villages that lie along the led to some changes in farmers’ behavior, with farmers major rivers forming the main waterways of the Delta. These beginning to diversify their income sources. In previous rivers can reach a mile wide and have traditionally provided rounds of SIM farming, households were much less likely to a sustainable livelihood for the fishers living along their diversify than fishers or casual laborers. Farmers’ reticence banks, though since Nargis this livelihood is on the decline. to diversify reflected both a comparatively reduced need to Both of these groups use a range of net sizes, generally work- migrate as a coping strategy (given land’s value as an asset) ing from small, uncovered motorboats with ‘fishtail’ propel- and the farming-specific need to remain close to their land lers. The proportion of this sort of fishers in any given village during the growing season. For the first time, SIM 5 found varies depending on the village’s location, but they are by a notable expansion in the number of farmers diversifying far the most common type of fishers across the SIM panel. their income sources, especially in the ‘off season’. For farm- ers who do not plant summer paddy, this begins in Novem- Finally, there are the ‘employed’ fishers, generally land- ber and ends in April; for those planting summer paddy, the less laborers who have left their village and work in town low season is from February to April. For most of these farm- centers for larger scale fishing boats. These boats fish fur- ers, diversifying either meant doing local casual labor jobs, ther offshore and are generally away at sea for a few weeks or seasonal migration to the township center or to Yangon. at a time. While not a traditional village-level livelihood, improved access and social networks in some villages have “This year, I diversified the business and I am both raising meant that subsistence or small-scale fishers have increas- ducks and planting betel vines. Prices of everything we buy ingly turned to employment on fishing boats to either sup- are increasing and it is not good to depend on farming only.” plement or replace their other livelihood. —Farmer In fishing households, the act of setting and retrieving the nets is normally done by the male head of household and/or the eldest son. If additional labor is required, households try to hire labor, but as catch sizes have dwindled it has become Fishing more common for women to go on the boats, especially when retrieving the nets. Once the catch is landed, women Fishing in the Delta sort the fish into categories and determine which will be sold, which will be eaten by the household, and which will Three types of fishing take place in the Ayeyarwady Delta. be processed (either for consumption, sale as fish paste, or First, there is subsistence fishing conducted on a very small used as bait). Women also participate in repairing nets. The scale, generally by landless, asset-poor households in the physical requirements of fishing from a boat mean there are small streams that crisscross the region. These fishers also few women-headed households who fish in this way, but sometimes fish in paddy fields when they flood during the some do engage in subsistence fishing using small boats and depths of the monsoon. They fish using lines or small nets nets that do not require great strength, expensive licenses, from man-powered canoes. Typically, their catch is used for or the need to venture far from the village. their own consumption (either fresh or dried for future use) though they might sell some to the local fish broker/collector What fishers catch if they have excess. In previous rounds of SIM, subsistence The Ayeyarwady River, its Delta, and the offshore waters fishers made up a small but measurable proportion of most villages’ population. By 2017, due to the increased avail- hold a wide variety of fish and seafood species. For the pur- ability of employment and expanded migration opportuni- poses of this report, ‘fishing’ as a livelihood includes the catching of prawns, shrimp, crab, and langouste as fish. The ties, there are now very few who fish at subsistence levels most common fish species is the hilsa, but fishers catch, eat, for their primary livelihood, though it is still common as a and sell a wide variety of fish both as fresh and for drying or secondary activity, especially during periods of little agri- production into fish paste (ngapi, a staple of the Myanmar cultural activity. diet). Second, there are the small-scale fishers who either work independently or in small groups of family members or Livelihoods 13 Returns from fishing For various fish types, these rules can range from a simple K5,000 licenses for individual fishers to K2.5 million con- The costs associated with fishing have cessions to large businesses which subcontract smaller fish- increased ing enterprises. The licensing structure often changes on a Since Nargis, the costs of being a fisherman have steadily yearly basis, making long-term planning for village fishers a increased. In the years following the cyclone, there was the challenge. Further, it is becoming costlier to obtain licenses immediate need to reinvest to replace lost boats, engines, and there is little benefit to doing so—when the fishers face and nets. In some villages, every fisher lost all their live- a problem, such as when large fishing boats tear up their lihood equipment and was forced to start from scratch. nets, the fisheries department offers no support. Some received assistance from external sources, Myanmar or international, while others sought other ways to restart The income from fishing has remained flat their livelihood. Several fishers interviewed spent the years Fishers broadly reported that increased prices and dimin- after Nargis working in Yangon or elsewhere in order to save ished yield offset one another, and the net effect is that enough to buy a new boat and equipment. returns of fishing have been essentially flat in recent years. As time has passed and in response to reduced catches, Stagnant yields . . . the fishers continually invested in new equipment so as to sustain their livelihood (Case Study 2). Improved boats and In most villages with a substantial number of fishers, low engines allow fishers to go further afield, reaching fishing and unchanging catches were consistently reported no mat- plots where catch levels are highest. Fishers are also using ter what fish or net was being used. The few exceptions to nets with a tighter weave, allowing them to catch smaller this were in villages that lie directly on the coast and have fish, though this practice comes with significant sustainabil- access to deeper water and open ocean. For villages along ity concerns. the Delta’s major rivers, reports of declining catches were universal (Table 9). Prawn catches were reported to have In recent years, changes to license fees have accounted for recovered back to 2010 levels, though catches for most more and more of the ongoing costs for fishers. Licensing types of fish remain a fraction of what they were before rules vary significantly, not just depending on the type of Nargis. fish being caught or the types of nets used, but from town- ship to township as well. Township-level fisheries depart- Fishers attributed a range of explanations for the reduced ments appear to develop their own licensing procedures. catch levels. The two most commonly cited were increased Case Study 2: New Nets Lead to Improved Earnings for a Fisher The primary livelihood in a village in Bogale township is fishing; there are only a few farming households. During Nargis, all the houses in the village were destroyed and it took a year for the village to be reconstituted. One fisher, U Zaw Win, has four young children. He is the only family member working, though his wife helps him collect the fish after the catch. Nargis destroyed his nets and boat. In the years following the storm he worked as a farm laborer. In 2009, he borrowed K80,000 to buy a net for hilsa fishing. He also borrowed a boat from a fisherman, but this meant that for every four times other fishers went out, he could only go out once. As a result, he could not earn a living as a fisher. So, at the end of the hilsa fishing season U Zaw Win went to Yangon to work as a casual laborer. In 2013, an NGO pro- vided eleven boats to village fishers who did not have one, including U Tin Khine. He also bought himself a new net. From 2013 to 2015, the fish catch was good because of the new net. By 2016, the fish catch had declined, a problem he attributed to weather variation, and the fisher decided to get a new “Kyar” net. He took a loan of K500,000 to pay for the license fees and the new net from the village fish collector, and as part of his repayment sells his fish only to that collector. Although he is aware of the difference in price between the village and township fish collectors, his engine boat is not sturdy enough to go to the township. It has been five months since U Zaw Win started Kyar net fishing and his family is one of the better-off households in the village. He said “though the debt from purchasing a license is always between K400,000 and 500,000 we can pay it back fully every year.” 14 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Table 9: Yields of Hilsa (viss per year) Table 11: Fishing Conditions by Degree of Affectedness over the Last 5 Years Before Nargis 2008 2009 2010 2011 2013 2017 Affectedness Good Fair Poor Total Yield 100 100 100 75 48 40 45 Highly affected 1 6 1 8 (of 12) Moderately affected 3 4 1 8 (of 18) Lightly affected 1 0 1 2 (of 10) “We cannot now say how many viss we catch; instead we can Total 5 10 3 18 (of 40) say how many individual fish we catch.” —Fisherman from village on the banks of a river that has experienced a significant decline in catch Contributing factors Fishing’s profitability is very much dependent on vil- lage location. Fishing villages along or near the coastline reported better catch levels than those set on the banks of competition from fishers from other places, including larger rivers forming the Ayeyarwady Delta. Further, transporta- fishing businesses from the township, and a perceived tion improvements for these coastal villages, including sev- decline in water quality. The latter was blamed on both eral newly accessible via road, has meant improved market changes to the rivers’ flow which had seen the water become access and accompanying higher prices. more brackish and also on fertilizer-polluted run-off from farmers’ fields. Other common issues include lots of garbage Fishing and the degree of affectedness are linked in that being found in nets (fishers in several villages reported they a larger proportion of highly affected villages have fishing sometimes catch more plastic waste than fish) and that sectors. Of the 18 SIM villages where the team was able to tides, and the resulting catch, have become less predictable evaluate the condition of fishing,9 eight were highly affected in recent years. (of twelve highly affected villages overall) while just two were lightly affected (of ten). However, as Table 11 shows, . . . are offset by price increases beyond this linkage there does not appear to be a correla- tion between the degree of cyclone affectedness and the Nonetheless, declining catches have been offset for many condition of the fishing sector over the past five years. fishers by significant increases in the market price for fish, especially prawn. This contrasts to SIM 4 which found a similar narrative on catch decline, but no counterbalanc- ing price increases. In 2017, the price has been buoyed Adaptions to the evolving by a variety of factors (Table 10). First, improved market fishing context access has reduced the number of middlemen for fishers in many villages. In one village, the lobster fishers now “It takes time to return to fishing. We had to try hard to go bypass township-level markets entirely and sell directly to back because of the frightening experience in Nargis” a broker in Yangon for several times the price available to them locally. Second, fish scarcity has driven prices up. This —Fisher from Zin Ywe Gyi scarcity reflects both the reduced catch and a reduction in the number of fishers. Finally, for some fishers, increased The number of active fishers in SIM villages has gone down. investment in their fishing equipment allows them to catch Exact figures are not reported, but fishers, village leaders, different types of fish for which there continues to be a and the research team itself all reported that there has been strong market. At the same time, many villages reported that a notable decline in the number of households that have market prices for seafood catches with low barriers to entry, fishing as the primary livelihood. especially crab, had increased as well. 9  Fishing conditions are defined as: “good,” prices have increased and remained strong, villagers have been able to invest in motor- Table 10: Prices for Hilsa (kyats per viss) boats as well as improved nets, often with a tighter weave; mar- ket access has improved and fishing license costs have declined; Before “fair,” prices are strong, license costs have declined, and fishers Nargis 2008 2009 2010 2011 2013 2017 have access to improved nets; “poor,” catch has diminished signifi- cantly and the number of fishing households has declined, almost Price 16,700 17,200 20,250 24,650 25,900 30,400 31,400 disappearing as a livelihood in some villages. Livelihoods 15 The expansion of larger scale fishing enterprises based gen- for agricultural jobs, new job opportunities, and expanded erally at the township level has contributed to this decline. migration. Fishers complained that large fishing boats now fish in shore and along the river rather than staying offshore. For While a variety of new job opportunities for laborers existed those fishers who remain in the village, diversification is across SIM villages, new livelihoods were especially appar- very common, with many choosing casual labor locally given ent in communities with improved connectivity to township improved wages. Some fishers choose to migrate, either sea- centers. The implications for laborers is that there are both sonally during the low fishing period or, for families with expanded opportunities to work for themselves rather than multiple working adults, one person may migrate perma- for a business and that they have improved access to better nently to Yangon or further afield. jobs in town rather than being restricted to opportunities within their village. A lack of support Interviewed fishers also noted that they have received little assistance in recent years, highlighting the inequity Agricultural labor between livelihood assistance programs for farmers and the paucity of fishing-focused activities by both donors and the Peak versus non-peak activities government. This was linked to a growing anger over license Agricultural labor in the Delta region can be divided between fees, with fishers feeling that despite increasing costs, they peak and non-peak employment. During nonpeak times received no active support from the fisheries department (most of the year) most small and medium farmers require and when wronged, they had no recourse. A common com- only a small handful of laborers—and some small farmers plaint was that when, as often happens, offshore fishing use only family labor in these periods. The predominant boats run over their nets, damaging or destroying them, nonpeak activities tend to be preparation before planting there is no one for the small fishers to turn to for justice. (including rebuilding paddy walls and irrigation channels) and the weeding and crop care that takes place before the harvest. For these activities, larger farmers often hire sea- Problematic fishing methods sonal or permanent laborers who are paid either monthly or for the whole season. Certain fishing methods in the Delta pose broader chal- lenges for both the environment and future catches. The use Since before Nargis, there were two peak seasons for agri- of poison, explosives, and car batteries in fishing (the for- cultural labor: the monsoon planting and monsoon har- mer kills fish via the concussive force while the latter kills vesting season. The expansion of summer paddy crops to via voltage) are indiscriminate and can damage river beds more and more farmers has brought a third season: summer and other animal life. These methods are illegal, but remain paddy harvest.10 The first monsoon peak season occurs in common practice, and have been applied more frequently May and June and involves the planting and transplantation since Nargis. of paddy seedlings. In recent years due to flooding or other shocks, a farmer may need to replant at least once, result- ing in increased job opportunities for laborers who do not migrate. Labor The monsoon harvest is in September and is the period of SIM 4 found casual laborers facing a dire situation. While highest labor demand as there is often just a short window in the immediate aftermath of Nargis, food aid had meant for farmers to harvest their crops. The second harvest period that laborers faced reduced food insecurity, in the medium- lasts from late January till March. term agriculture’s stagnation and lack of secondary liveli- hood opportunities were placing an increasing strain on Wages laborer households. Particularly in moderately and highly Wages across the SIM panel have increased over the pre- affected villages, the socioeconomic well-being of laborers vious four years, and in some villages peak season daily was closely tied to that of farmers and their village’s overall wages have more than doubled their 2013 levels. Table 12 economic condition. highlights how average labor wages across villages during peak seasons have changed for both men and women since By SIM 5, the situation evolved and of the three main liveli- before Nargis. These changes have been driven by a range of hood groups, laborer livelihoods are in many ways the most fully recovered since Nargis. Their improved position is a reflection of three major interlinked trends that exist across 10  Farmers use ‘broadcasting’ methods to plant summer paddy the SIM panel and across much of Myanmar: higher wages crops, so little additional labor is required. 16 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Table 12: Changes in Wages 2007–2017 Male Labor Wage (kyats) Female Labor Wage (kyats) 2007 2013 2016–2017 2007 2013 2016–2017 Daily Monthly Daily Monthly Daily Monthly Daily Daily Daily 1,947 58,500 2,693 76,429 4,988 87,015 1,383 1,832 4,138 factors that can be broadly categorized as labor action and enforceability facing these arrangements has contributed to farming context. the attractiveness of labor groups, where the leader’s repu- tation ensures attendance. Labor groups and advance wages There are two labor actions that have affected wages and Other factors affecting agricultural the relationship between farmers and laborers. The first is labor demand the expansion of labor groups. Prior to Nargis labor groups Demand for agricultural workers has evolved in response were active in three villages; now there are labor groups in to a number of changes in farming practices—which in turn eleven villages, including two groups formed in recent years have been caused or exacerbated by changes in labor sup- that focus on nonagricultural labor. These groups, made of ply. For example, as labor has become more expensive, farm- up of at least 20 but often as many as 50 laborers, work as ers have turned to mechanization, broadcasting instead of a team and are led by a foreman, usually a relatively well-off seedling transplanting, and increasing the use of chemical and skilled laborer with a strong network beyond the village. additives to reduce their labor costs. Some farmers have Two groups are led by women and all include both men and even left some of their land fallow because it was too expen- women as workers. There is gender pay equality within the sive to hire labor. groups. At the same time, other farming changes have resulted in According to both farmers and laborers, labor groups repre- increased demand, such as the introduction of a second sent an improvement over individual hiring. Farmers benefit summer paddy crop, or improved the position of workers, from risk reduction, namely the assurance of a large number such as improved connectivity allowing workers to access of workers in a timely fashion to plant or harvest with the other employment opportunities. Overall, demand for labor reduced burden of finding workers and ensuring they show has outstripped supply leaving laborers in a stronger posi- up. For laborers in the group, they get more and more certain tion and farmers scrambling for new ways to reduce labor job opportunities, regular wages that are higher than indi- costs. vidual wages, the freedom not to worry about finding the next job, and prearranged transportation to the work loca- tion even when it is in another village or even the next town- Changes to local livelihoods ship. For example, if wages for the individual were K4,000 per day, workers in a group can earn between K5,000 and In previous rounds of SIM, nonfarm enterprises that existed K7,000 for the same activities, thanks to the increased pro- in villages tended to offer just basic retail goods. Villages ductivity associated with group labor. Also, with the fore- had small grocery shops, fertilizer shops, or other business man in charge of finding employment for the group across catering to either basic needs or servicing the three primary village tracts and townships, the season for high agricultural livelihood groups. By SIM 5, location was a key indicator of labor demand expands from two months to four months. the extent of new opportunities for laborers. The second labor action has been the expansion of advance Various nonfarm enterprises have declined, particularly wage payment by farmers to secure the services of labor- those that required substantial labor. Charcoal making, salt ers during peak times. Well before the planting or harvest- production, fish paste production, and nipa palm thatch ing seasons, farmers provide advance amounts in exchange making were all fixtures of the pre- and post-Nargis village for a promise that the laborer will come and work on their economy through SIM 4. In 2017, however, small enterprise fields at a designated time. For the farmer, this is meant to owners and laborers reported that these businesses have ensure that labor will be available, while for the laborer it declined or disappeared entirely as new opportunities had provides cash up front at what otherwise is a low season emerged for laborers. for agricultural jobs. However, in practice farmers face dif- ficulties enforcing the informal contract with laborers not At the same time, there has been a noticeable increase in the showing up or demanding additional wages. The lack of number of nonfarm small enterprises active in SIM villages Livelihoods 17 Case Study 3: Entrepreneurship Improves Livelihood Outcomes A village in Bogale township, where most households are farmers, was moderately affected by Nargis. In the years fol- lowing the cyclone, one household stopped farming paddy entirely, turning to new livelihood activities. Prior to Nargis, U Mg Mg was a well-off farmer: he owned over 20 acres of land and two boats while also working as a rice broker. He had a small grocery shop that he had run for 13 years. In 2015, after several years of bad harvests due to bad weather and expensive labor, he decided to sell his land to another farmer in the village. He told him to pay for it when he had enough money. Meanwhile he continued to work as a rice trader. He then also sold both his boats: after Nargis more farmers began to own boats and sell directly to Bogale town, reducing the income he could make by rice trading. Prior to Nargis, their grocery shop was the only one in the village and benefited from being near the village’s high school. Though the shop was destroyed during Nargis, the family did not lose the gold they had saved and was able to restart their business by pawning gold. In 2009, he decided to recover these expenses by starting a betel business. U Mg Mg planted 2,600 betel trees on the plot of land where they live. From these trees, he now gets K15,000 in daily income. He had decided on betel because his village was close to town, and there was easy access to the market there. From 2013 the household also had a business charging batteries with a solar panel. “For five years, I have charged bat- teries for people in the village. When solar started to be used in 2012 I bought a solar panel (300 W) in town just for home use. But I realized it could be a business when my neighbors came over and asked me to charge their batteries. Then, I invested 12 lakhs and bought the biggest solar panel I could find. To start with I made more than K10,000 each day. After two years, other people began to copy me but I was not worried about the competition because I had already got back my investment. Now, I get about K3,000 a day, which is fine for extra income for use around the home. I am always thinking about what kind of business can be done in the village and always have new ideas before the others.” At the time of SIM 5 research the household’s main businesses are the grocery shop and betel plantation. The grocery shop’s sales have gone up thanks to its proximity to both the high school and Bogale market—villagers from other vil- lages often come to his shop. U Mg Mg is able to save K10,000 per day from these two livelihoods. He remembers that after Nargis he had to restart his businesses by pawning gold; he is saving again (in gold) in order to improve his ability to recover if faced with another disaster. With the savings, he was able to rebuild his home in 2014 so it has wooden walls, a brick floor, and a tin roof; he now intends to build a new house in brick. Commenting on the changes to his life after changing from a large farmer to a small businessman, he said “I do not want to hire labor for my business. Instead, I will save money myself. So, I work only on my small businesses, not a big farm.” (Case Study 3), reflecting the increased level of connected- in mechanization), pharmacies in villages closer to towns, ness villages have with their township centers and beyond. local tailors and embroidery shops, tuition services, and the New or improved roads and jetties have allowed more rapid first ‘beer stations.’ Other new nonfarm opportunities men- travel to the markets at the township level, while the now tioned in at least a small number of villages included money ubiquitous presence of mobile 3G allows communities to transfer shops, beauty salons, housing material shops and, connect, economically and socially, to employers, buyers, in villages closer to township centers, restaurants. friends, and relations well beyond the village. The new opportunities are attractive because they offer By SIM 5, well-connected villages experienced an expansion higher wages and/or allow the laborer to be self-employed of service-based businesses. They included motorbike taxi rather than dependent on a local employer. There is little and goods-carrying services that are now ubiquitous in road- difference between highly, moderately, and lightly affected connected villages while some villages even have mobile villages when it comes to new employment opportunities. phone stores. Other enterprises seen more often include Even more remote villages have improved employment grocery shops of various sizes, repair shops for motor- opportunities though they are more limited. bikes and agricultural machinery (reflecting the increase 18 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Women and income diversification Table 13: Outmigration by Degree of Affectedness Women, within both male- and female-headed households, have made particularly important contributions to a house- High Medium Low Affectedness Levels Levels Levels Total hold’s ability to diversify incomes via nonagricultural jobs (Case Study 4). Alongside their contributions to a house- Highly affected 4 5 3 12 hold’s main income, be it fishing or farming, women are Moderately affected 7 7 4 18 also often responsible for the day-to-day management of a Lightly affected 5 3 2 10 secondary income source—whether a small grocery shop, a snack shop selling mohinga or, in larger villages, a more fully Total 16 15 9 40 stocked grocery store. Other activities identified by the SIM team as contributing to income diversification and mainly organized by women included livestock raising, clothes member who had migrated (18 percent domestically, 2 per- making/tailoring, supplementary agricultural labor, and cent internationally).12 working as local fish collectors/brokers. Of SIM’s 40 villages, 16 were assessed as having high levels of migration (at least 25 percent of households having at least one migrant member) with an additional 15 with mod- Migration erate levels of migration (at least 10 percent of households Migration across Myanmar has expanded rapidly in recent with a migrant) (Table 13). Put another way, almost 80 per- years, but the growth has been especially pronounced in the cent of villages have experienced substantial outmigration. Delta. The region’s close proximity to Yangon and the rap- idly growing construction and garment sectors there have The effects of this phenomenon have been clear across this proven massive draws for villagers across the SIM panel. chapter (and are seen in later chapters examining changes to A recent study found that workers from the Ayeyarwady social relations, leadership, and resilience). Farmers strug- region account for 29 percent of garment sector workers gle to hire labor that has left for better opportunities in the in Yangon,11 while a World Bank survey in 2015 found that towns and Yangon and even migrate seasonally themselves. 20 percent of Delta households had at least one household Fishers discontented with reduced catches and uncertain Case Study 4: Sisters Work Hard Together Daw Mya Aye and her sister live in a village in Pyapon township and own a tailoring business, mostly sewing bags. They moved to Yangon in 2007 and started to work in a weaving factory where they were paid just K30,000 for the first month. In the second month, because they were more experienced, they were paid K50,000, but decided to move to a new factory where they were paid K100,000. While the pay in the new factory was better, by 2008 they decided to start their own business: they felt that they would never be able to be independent if they continued working in the factory. Therefore, they resigned and opened their own tailoring shop, accepting orders through their personal network from the bag shops in downtown Yangon’s Thein Gyi clothing market. In 2011, they took what they had learnt in Yangon and opened a business in their village. Since their return, they have continued to make bags, selling not only to their Yangon contacts but also to shops in Pyapon and Myin Ka Kone towns. The shops send them a sample bag and then they make copies of it. They currently have three clients in Yangon, two in Myin Ka Kone, and one in Pyapon with orders for approximately 500 bags per week. They make K14,000 of profit for every 100 bags they produce. Daw Mya Aye and her sister would like to hire more local women to work for them and expand the business, but they have been unable to do so because most of the young women in their village have migrated to Yangon. As a result, they have continued to rely on the labor their family can provide, and are unable to expand their business. Enlightened Myanmar Research Foundation and Andaman 11  Research & Advisory. 2017. A Baseline Survey of Yangon’s Garment World Bank and Livelihoods and Food Security Trust Fund. 2016. 12  Sector Workforce. A Country on the Move. Livelihoods 19 returns migrate seasonally, or in some villages have taken “Before Nargis, only men were responsible for their house- their whole household and left the village behind. But it is hold. Now it is the women who migrate to Yangon and work at for laborers that the migration boom has been most trans- the factories to look after their families.” formational, providing important new opportunities for income diversification and advancement, particularly for —Landless laborer households with more than one adult able to work. Migrating women are a key source of remittances back to Unlike in other parts of Myanmar where migrants tend to the villages—over 60 percent of garment workers send back skew male, the number of migrants out of SIM villages money, often every month—and this income has allowed are slightly more female. While women do still face social many families to improve their quality of life, finance edu- norms against migration, such as the expectation that they cation, build new homes, and invest in new or additional look after children, parents, or other family members, a livelihoods.13 Indeed, as Chapter 7 shows, remittances are a higher proportion of Delta women have overcome these and central contributor to a household’s resilience. For women migrated. Most go to Yangon where jobs in the economic who return to the village, they often bring back both hard capital’s burgeoning industrial base are easy to come by skills and a new outlook on life enabling them to identify and offer permanent positions—compared to the construc- needs in the village and set up businesses to address those tion sector, a major employer of male migrants—that offers needs. only day-based positions. The garment sector, as in other Southeast Asian countries, is a particular draw for migrating 13  Enlightened Myanmar Research Foundation and Andaman young women. Research & Advisory. 2017. Op. cit. 3 Credit, Debt, and Land In 2013, SIM 4 research found that the debt burden in most provided by the Myanmar Agricultural Development Bank villages had increased. Of the panel’s 30 highly and mod- (accessible only to farmers) and various private sources, erately affected villages, just three moderately affected vil- there is now a wide variety of credit offerings ranging from lages did not have repayment problems. Small farmers were government programs, microfinance enterprises, NGO found to be the most affected by increased indebtedness, schemes, and private lenders. There is little difference in followed by laborers who frequently had to borrow for con- credit access between highly and moderately affected vil- sumption and, therefore, were greatly affected by increases lages, with lightly affected villages having on average about in prices for food and other necessities. one additional credit source (Figure 2). The expansion of credit sources is explored in more detail below. Five years later, debt levels have continued to grow, and indebtedness continues to be a challenge identified by households across the Delta region (Table 14).14 However, fewer villages, especially those moderately affected, are in Credit and Debt debt traps and, among lightly and moderately affected vil- lages, there is a clear increase in the number of households Credit and debt cycles have been a constant in the Ayeyar- who are able to regularly settle their outstanding loans. 14 wady River Delta since well before Cyclone Nargis. The main borrowers have traditionally been farmers, especially small The increased debt levels instead reflect increased credit and medium farmers who require loans in order to finance availability across the panel. While in SIM 3, credit was seeds, labor, and other inputs for their paddy rice crop. Until Table 14: Indebtedness by Degree of Affectedness Villages Able to Settle Their Villages Unable to Repay Loans Regularly Loans Regularly Villages in a Debt Trap Affectedness SIM 4 SIM 5 SIM 4 SIM 5 SIM 4 SIM 5 Highly affected 0  4  6  4  6 4 Moderately affected 3  5  7  9  8 4 Lightly affected 5  8  5  2  0 0 Total 8 17 18 15 14 8 14  “Settle their loans regularly” means borrowers able to repay interest, but not the principal; “in a debt trap” means unable to both 14  interest and principal; “unable to repay” means able to repay maintain interest payments. 21 22 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Figure 2: Average Credit Sources per Village by Degree of Affectedness Lightly affected Moderately affected Highly affected 0 1 2 3 4 5 6 7 recent years, these farmers borrowed from one government Credit sources in 2017 source, the Myanmar Agricultural Development Bank, and a range of local private money lenders (including village or As in SIM 4, Delta villages continue to rely on a variety of townships traders, fertilizer shops, and larger farmers). credit sources, but their choice has increased significantly (Table 15). New programs from government, such as the For other livelihood groups, while credit was both needed Evergreen Village Development Program and the Coopera- and desired, it was much less available prior to Nargis and tive Department’s loan scheme, now offer more affordable in the years that followed. Landless casual laborers relied on borrowing opportunities to nonlandowners than was avail- credit in the leaner months, borrowing for consumption and able previously. Microfinance activities, by both NGOs and accruing debt they hoped to be able to pay off through agri- private sector entities, have also increased. cultural wages during seasons of high demand for labor. As there were no government sources of credit for the landless, Myanmar Agricultural Development Bank they would rely on larger farmers or professional money and other government credit expand lenders, who charged very high interest rates (as much as As in SIM 4, MADB remains the most common and the most 50 percent per month). important credit source for farmers across the panel vil- lages. In 2013, it provided farmers with K50,000 of cheap The SIM 4 research reflected these findings, as well as the credit per acre of registered paddy land (up to a maximum high demand for additional credit sources by all livelihood of 10 acres). As this was not enough to pay for all inputs, groups. MADB’s expansion and the expanding presence of the government has worked to increase MADB loan size over microfinance since 2013 received a positive response from time. For example, the loan size for the 2017 harvest was villagers, though fishers reported that they continued to K150,000 per acre for both the monsoon and summer crops. have less access to finance than other groups. Since then, MADB continues to practice a group-lending approach where there have been dramatic changes to credit availability in loans are distributed to groups of farmers (it used to be ten, the Delta. but can now be as few as three) rather than to individuals. However, farmers regularly complain that disbursement comes late. Table 15: Sources of Credit by Livelihood Group Small Medium Large Small Large Casual Farmers Farmers Farmers Fishers Fishers Laborers MADB 39 40 40  0  0  0 Cooperative 31 30 28  9  8 16 Evergreen 11 11 11  7  6 11 PACT microfinance 24 23 21 14 13 27 NGO microfinance 20 18 18  9  9 18 Private moneylender 31 32 28 23 22 35 Other private source 29 30 28  9  8 22 Village fund  6  6  6  3  2  8 Note: Number of villages where livelihood groups have borrowed from a given source. Credit, Debt, and Land 23 The importance of MADB has led to a variety of coping strat- microfinance that have limited borrowers’ ability to improve egies when harvests fail and farmers want to ensure they their livelihoods and living standards. remain eligible for future loans. In some cases, village tract administrators have set up schemes that provide farmers Many interviewees cited the inflexibility of microfinance with a way to remain current, but at a cost: the Village Tract credit as a major issue; villagers frequently criticized the Administrator (VTA) (or a wealthy businessman using the requirement that borrowers make repayments every two VTA as a middleman) provides a lump sum to the farmers weeks. For most villagers, income streams are not steady to repay MADB; shortly thereafter, farmers re-borrow from over time, making such regular repayment schedules impos- MADB and pay back the VTA, plus a 5–6 percent service fee. sible and leading many to borrow from private sources at This keeps the farmers current with MADB for another six much higher interest rates in order to make these interim months. payments. Instead of providing affordable finance to help people escape the debt trap, these loans are worsening the In other villages, farmers have realized that MADB’s enforce- debit situation of the Delta’s most vulnerable households ment is piecemeal or completely absent and simply failed and have led some individuals and households to flee their to repay their loans. Farmers who had taken this approach village in order to escape debt. These debts are having often had one member of their group who was unable to pay. further effects on the social relations of the communities, Recognizing that any consequence for nonpayment would examined in more detail in the following chapter. affect the whole group, as a group they simply skipped repayment. At the time of research, they had faced no sanc- Microfinance was also criticized by villages for not offering tion from MADB. In three villages, a different manipulation enough credit to substantially alter a household’s economic of the MADB system took place. With the help of their village trajectory. Now that food insecurity is much less common tract administrator, farmers created fake Land Use Certifi- and even the poorest laborers rarely need to borrow for cates to ensure they could access loans. consumption, the focus for most borrowers is on improving their livelihoods. However, while several hundred thousand The Cooperative Department’s loan scheme offers varying kyat, the most offered by many of the microfinance services, amounts to all livelihood groups, and has become more can finance extra fertilizer for a farmer or a new motorcycle popular among villagers, though some still object to its for a taxi service, loans of this size do not allow for liveli- forced savings mechanisms. The Evergreen Village Develop- hood transformation. For example, they would not be nearly ment Program, which is financed by a loan from the Chinese enough to allow a landless laborer to invest in land and government, started in 2015 and was already active across become a farmer nor enough for a farmer to invest in new the country in about 4,500 villages by 2016; it has a pres- agricultural machinery. In some ways, much of the Delta’s ence in 14 SIM villages (which is proportionally high for a population has moved beyond the micro-finance programs 40-village sample). The Evergreen program provides a cash designed for their needs five years ago and are increasingly grant of K30 million to the village level. These grants are in need of credit opportunities that allow for longer term, then disbursed as individual or group loans by a locally cre- larger scale investments in economic activities. ated committee, according to terms and interest rates set by that same committee with guidance from the Department Private moneylenders of Rural Development. The grant is treated as a revolving Private sources of credit remain common throughout SIM fund, with interest collected intended to be used for village villages where they range from semiformal lenders to more development activities. The low interest rates and locally informal arrangements, such as large farmers who lend to determined payment terms have made the project popular their smaller farmer neighbors or local input suppliers pro- among villagers in the SIM panel, with many interviewees viding seeds or fertilizer on loan. While almost all villages commenting on how easy Evergreen loans are to repay. have some sort of private lending source, evidence from interviews suggested that the increasing number of credit Microfinance sources (both government and donor-driven) has reduced PACT Global Microfinance is the most common microfi- reliance on private sources. SIM 4 identified the early stages nance program. It has had a presence in the Delta since of this trend, noting the increasing use of MADB, and with the well before Nargis, but has continually worked to expand expansion of low(er) interest credit sources it has continued. its programming and is now active in 32 of the panel vil- lages. Other microfinance programs, including those funded However, it should be noted that relative to farmers and now by the United Nations Development Programme and the to casual laborers as well, fishers continue to have much multi-donor Livelihoods and Food Security Trust Fund less access to non-private sources of credit. also have a presence across several townships. Their pres- ence, combined with the Evergreen program, has greatly Variance in terms and conditions improved access to credit for landless households, though This expansion of credit access since earlier rounds of SIM, interviewees cited a number of challenges accompanying thanks to a mix of national and international NGOs, and 24 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After government programs, has offered villagers the opportu- programs, such as bookkeeping training or learning to set nity to begin comparing different credit sources and make up small tailoring shops with support from a village revolv- decisions on which source best suits their needs. Villagers ing fund. In sum, while there are few formal barriers for in SIM 5 were more aware of credit terms, of interest rates women to access credit (and in many households women and models for calculating them, and of social security fund are responsible for day-to-day household finances), social incentives for various credit sources. norms regarding activities suitable for women and house- hold responsibilities do limit how women borrow. Gender and credit access Levels of debt With the exception of MADB, where loans are disbursed only to the person registered on the Land Use Certificate (LUC),15 Previous SIM research found many households carried sig- women do not face any regulatory barriers to credit access nificant levels of interest-bearing debt from a variety of compared to men. All individuals can access Cooperative government-subsidized and private sources. Over the past Loans and Evergreen program finance (reflective of the five years, debt levels have continued to rise, though these rules set up by the local committee) while PACT and other increases have been unevenly distributed among livelihood microfinance and revolving fund programs actively enable groups. women to borrow independently of their husbands by set- ting up women’s groups for this purpose. 15 Farmers Large and medium farmers have experienced the largest Nonetheless, social norms and pressures do factor into debt increases, a reflection of the challenges facing them women’s ability to access credit. In some cases, loans are from weather shocks, variable harvests, and their reliance disbursed by household, which may limit a woman’s access on increasingly expensive labor. As Figure 3 shows, debt lev- relative to her husband. In other cases, training programs els among these groups have gone up particularly fast. Small that accompany loans, such as for agricultural or livestock farmers have been the least affected by debt increases of rearing techniques, do not fit well with women’s other any livelihood group, showing only a marginal increase over responsibilities within the household. Women are instead the previous four years. Their ability to avoid debt increases more likely to engage with other types of credit adjacent is due to both their limited need for finance (restricted to Figure 3: Average Debt Levels over Time (in lakh) 50.0 40.0 30.0 20.0 10.0 0.0 Before One year Two years Five years Nine years nargis after after after after Large farmers Medium farmers Small farmers Medium/large fishermen Small/subsistence fishermen Casual laborers 15  The registered individual receiving the loan is generally the head daughters as necessary, to ensure they receive as large an MADB of the household. However, as MADB only allows loans for ten loan as possible. In these cases, women do technically receive acres of land per person, larger farmers often register their land MADB loans, but usage of that loan would be determined by the 15  in the names of other family members, including their wife and household head. Credit, Debt, and Land 25 inputs) and their inability to take on much debt: once a small input that is then used for several years. For example, after farmer’s debt grows too high, he would likely need to sell Nargis when many boats were destroyed, the biggest chal- (part of) the land. lenge for fishers was being able to afford to replace their boats, engines (for those who had them), and nets. Today, Asked when over the past five years they had the highest the fishing context requires a boat with an engine—a large and lowest levels of debt, in over two-thirds of SIM villages, expense that most village-level credit sources (including farmers said they had the most debt in 2017; over 80 per- microfinance providers) are ill-equipped to address. As a cent reported lowest debt levels in 2012 or 2013. This sug- result, debt among fishing households reflects large debts gests that debt levels have gotten progressively worse, with taken on for livelihood investment, often from township- a growing number of farmers, especially medium farmers level businesses or fish brokers. who need to borrow but have the least ability to service debt, unable to escape from a debt trap (Figure 4). Casual laborers Casual laborers’ debt levels are also up substantially, though this likely reflects the greater access to credit available to Figure 4: Highest Debt Year for Farmers them in 2017 as much as needing to borrow for consump- tion. Since 2013 the number of village-level credit programs % of farmers citing accessible to landless households has grown significantly 2012 or 2013 with both government (Cooperative Department, Evergreen lowest debt Loan program) and microfinance providers now offering % of farmers loans. citing 2017 as highest debt These credit programs have increasingly focused on pro- viding landless households with alternative or diversified 60% 65% 70% 75% 80% 85% income sources in an effort to reduce reliance on agricul- tural labor. From group loan programs supporting animal husbandry to individual loans that have enabled households Among farmers, there are many who have struggled to to start family businesses, debt levels of landless house- remain eligible for government loans while others have con- holds have risen alongside local economic opportunities. sciously defaulted. Despite the high levels of debt facing Delta households, however, many continue to borrow, with farmers in particular saying they have little choice. Without borrowing each planting season, many farmers have no abil- Debt and the degree of cyclone ity to buy agricultural inputs. For certain livelihood groups, Nargis’ effects continue to reflect increased debt burdens. This is particularly clear for Fishers large farmers: large farmers from highly affected villages Fishers reported a similar increase in debt, though starting tended to have significantly more debt than similarly sized from a lower level. For small or subsistence fishers, average farmers in lightly affected villagers. The same correlation debt levels in 2017 were three times those of 2013. While existed, but is much weaker, for small and medium farmers there are fewer large fishers making it difficult to identify as well as small fishers. However, there is little correlation the drivers of their increased debt load, for small fishers the between debt load and affectedness for casual laborers, expanded costs of equipment and licenses were cited. Fur- likely reflecting both the improved livelihood situation ther, fishers tend to have the most limited access to credit. for laborers discussed in the previous chapter and the relatively lower debt held overall by non-fishing, landless Credit needs for fishing livelihoods differ structurally from households. Table 16 highlights this distribution of (self- farming; instead of annual inputs, fishing requires a capital reported) debt. Table 16: Indicative Debt Levels by Degree of Affectedness across Various Livelihood Groups (in lakh) Large Medium Small Small/Subsistence Affectedness Farmers Farmers Farmers Fishers Laborers Highly affected 5.28 2.93 1.46 0.74 0.35 Moderately affected 4.72 2.66 1.02 0.66 0.46 Lightly affected 3.20 2.72 1.34 0.55 0.34 26 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After How households deal with debt traps highlighted in SIM 4, were found to have declined in both number and severity.16 Households stuck in debt traps generally first respond by the sale of assets—in the case of farmers this means pawning land or selling it and in the case of fishers, selling their boat. Land Sales Many farmers, rather than sell their land outright, attempt to pawn it so as not to lose the rights to it completely and even Sellers have one final attempt at a successful harvest. However, in SIM 5 research found that with the exception of a handful of cases where they are unable to redeem it, this results in fur- villages, there were just one or two land sales per year. Of ther debt and an inability to benefit from its sale. those who were selling their land, most tended to be small or medium farmers whose debt burden had reached a point Once a household’s major assets have been sold, migration where they could no longer avoid land sales. In some vil- of at least one working-age adult is an extremely common lages, larger farmers were also active land sellers, generally response. Either in the town or, more frequently in recent driven by two factors. First, given the challenges of finding years, Yangon, it is possible for most able-bodied adults to labor, increasing input costs, and highly variable returns, find work that allows them to live and send small amounts the advantages of having large amounts of land (especially of money back to their family. While these remittances can if that land is far from the village or in a bad location) are provide enough for survival, the experiences of many debt- less attractive. Second, MADB limits credit disbursement ridden villagers suggest they still often struggle to pay the to 10 acres per farmer, and while in the years immediately interest on their loans, let alone the principal. following registration large farmers could get around this by registering land under multiple family members’ names, In the most extreme case, villagers leave their homes and MADB has begun cracking down on this practice and there- run away to escape debts. Mostly the head of household fore making it much more expensive to borrow for input on (who holds the debts) will leave, but the SIM 5 research anything over 10 acres. As a result, larger farmers either found numerous examples where whole families would run sought to rent out or frequently sell land they viewed as to other parts of the Delta or to Yangon in an effort to flee extra to their needs, especially given the more diverse from their creditors. There were 15 villages where this was investment opportunities that now exist in many SIM vil- reported as having happened at least once. lages or the opportunity to buy land or houses in town. Buyers Land Land buyers tended to come from within the village itself, and many were moneylenders, highlighting the challenge The SIM 4 research found that economic strains were lead- of debt facing small and medium farmers. The land owned ing to more land transactions, resulting in increasing land by moneylenders is then frequently rented out to landless inequality where significant amounts of farmland were held households (sometimes even the previous owners) for them by a small number of individuals. A quarter of SIM villages to farm. Other types of local buyers included those who had had experienced a high number of land transactions between received remittances and sought to invest in land (including 2010 and 2013, including half of the highly affected villages, fishers working to diversify) and new couples who wanted with further consolidation on the horizon: many farmers told a housing plot of their own. Several instances of buy- researchers they planned to sell their land once it had been ers from outside the community were reported, including registered under the 2012 Farmland law. However, SIM 5 research indicates that the consolidation 16  Discussions with the research team and analysis of the data sug- of land has slowed and that for small and medium farmers gest that while the registration process may have caused some ten- sions, for villages where all farmers have Land Use Certificates and land sales continues to be a last resort. Even when facing the issue of who owns what land has been largely settled, LUCs significant debt burdens, these farmers continue to seek serve to reduce uncertainty over ownership and have reduced innovative ways to retain their land and plant it, knowing intra-village disputes over land. Nonetheless, it is worth noting that one or two good harvests can allow them to escape the that LUCs have not solved long-standing disputes over land, par- debt trap, while the market value of the land will do little ticularly where land has been seized by parties outside the village, but furnish consumption costs for a year or more. Further, especially government or government-linked companies. Such land disputes, which were a major theme and potential risk instances, however, are uncommon across the SIM panel. Credit, Debt, and Land 27 businessmen from the township who bought up land along than the 7–10 percent charged for collateral-less loans. This the side of the main road leading through the village. Such borrowing model is the newest of the three as it is reliant outside activity in village land transactions was reported in on farmers having an LUC, something which was only intro- four villages. duced following the 2012 Farmland Law and only widely held in SIM villages in the past few years. Prices Land prices vary significantly from village to village, though Second, there is the more traditional land pawning, where in general there has been an increase in land prices over instead of the LUC, the land itself is the collateral. For both time. The lowest reported price was K300,000 per acre for this and the LUC-based loans, the farmer is still able to work paddy land located outside a protective embankment. Most his farmland, while also able to access finance, which is of the interview respondents who had sold land reported used to pay for inputs, other farming costs and, sometimes, sales prices of between K600,000 and K900,000 per acre, their own consumption in advance of the harvest. with much of the variance depending on the location (land near roadways and access routes was notably more valu- Finally, there is the ‘money return–land return’ system, able) and when the land was sold. The highest reported whereby land used as collateral is used by the lender. Under land prices were between K1,500,000 and K1,800,000 per this system, a farmer gives over usage rights to a plot of land acre, though in one part of Dedaye township prices were for a fixed term, generally for three years, for a fixed sum, reported at K6,000,000. The more highly valued land was often approximately 50 percent of the land’s value, with no either along the main roads purchased by businessmen from interest accompanying the agreement. For the three-year outside the village, or was high quality paddy land (i.e., at period, the land is ‘owned’ by the lender who can farm it a higher elevation) bought by larger farmers from indebted himself or rent it to a tenant farmer. After three years, the smaller farmers. farmer either repays the debt and recovers his land or devel- ops a new agreement, either extending the ‘money-return, land-return’ period, or by agreeing to sell the land com- pletely, in which case the farmer receives an additional sum. “Money Return–Land Return” and the importance of land pawning For both land pawning and ‘money return–land return’ sys- tems, the amount of cash the farmer can expect to receive is Borrowing against the value of land has a long history in significantly less than the market value—generally between the Delta. Table 17 indicates that this happens commonly or one-half and one-third of the mutually accepted market occasionally in just over half of the sample villages. Previ- value of the land. However, the flexible nature of the loan ously, these transactions were undertaken in a very infor- and informal nature of the process make it appealing. Repay- mal manner. Since land registration occurred, however, they ment periods for these borrowing models are generally up have gained an extra level of formality. There are three ways to three years but can be as much as five, while the informal in which land is used as collateral for loans. process, generally between neighbors with local witnesses and social pressure as enforcement, limits the need to deal First, there is the use of the Land Use Certificate as collat- extensively with beyond-village authorities. The lenders for eral for loans from private sources. Interest rates for such all three types of land loans tend to be larger farmers within loans are generally 5–6 percent per month, somewhat less the village, relatives, or farmers from neighboring villages. Table 17: Frequency of Borrowing against Land by Degree of Affectedness Affectedness Common Occasional Never Total Share of Never Highly affected 2  7  3 12 25% Moderately affected 0  8 10 18 56% Lightly affected 0  4  6 10 60% Total 2 19 19 40 48% 4 Social Relations Early rounds of SIM examined the social strength of villages some areas, challenging intercommunal relations (Table 19). in the Delta following Nargis and, in later rounds, the evolv- Despite these issues, some of which are found in almost all ing effects of assistance and aid delivery on social bonds. villages, social cohesion remains strong. Tensions over distribution persisted into SIM 4, albeit at lower levels as by 2013, social challenges were more directly linked to economic ones. Nonetheless, SIM 4 found that overall, social relations remained strong. By SIM 5 the direct Inter-Household Relations effects of Nargis-related issues, including post-disaster aid disbursement, have largely dispersed (Table 18).17 The importance of extended family Instead, with a small number of exceptions, the forces affect- The presence of family nearby has proven to be an increas- ing social dynamics in SIM villages are reflective of broader ingly important indicator of a household’s ability to restart trends found across Myanmar that are, in some cases, exac- a business or generate new sources of income following erbated by Nargis’ legacy. These forces include migration, a shock. The extent to which the extended family helps economic change, an increasingly connected society and, in one another has not changed and remains an important Table 18: Village-Level Social Relations between 2013 and 2017, by Degree of Affectedness 2013 2017 Affectedness Good Fair Poor Good Fair Poor Highly affected  3  5 4  2  9 1 Moderately affected  8  7 3  2 11 5 Lightly affected  5  3 2  6  2 2 Total 16 15 9 10 22 8 Villages with “good” social relations tended to have high lev- 17  development activities, and clear divides between supporters els of unity/homogeneity and a high level of collective action and detractors of the current leadership; “poor” social relations and trust in village leadership; those with “fair” social relations occurred in villages with multiple religious groups, where villagers tended to have less unity (e.g., with some groups not participating felt discriminated against by village or township authorities, and in village affairs), some disagreement within the village on village where there was little unity or collective action. 29 30 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After contributor to recovery. This assistance often takes the form However, in SIM 5 some interviewees reported that the will- of renting farmland, lending farming tools, giving loans, and ingness of friends and neighbors to respond to requests has offering space to live or raise livestock (most commonly begun to change. They generally attributed it to communi- ducks) (Case Study 5). ties becoming more ‘business-minded’ and individualistic while at the same time improved job availability and better wages meant people have less free time to help one another. Table 19: Village-Level Changes in Social Relations since 2013, by Degree of Affectedness Stayed the Relations within Village Affectedness Improved Worsened Same Groupings Highly affected  3  4  5 Village life across the SIM panel is increasingly ordered Moderately affected  2  9  7 along activity and identity grounds that reflect the actual Lightly affected  5  3  2 make-up of the village, marking a step away from the Total 10 16 14 required, but often formulaic, ordered institutions under the previous military government that required every village to have cookiecutter organizations such as the fire brigade or maternal and child health group. Reliance on neighbors starting to shift Households across the Delta rely on each other for assis- Activity-based groups tance on specific needs; households mostly continue to help each other where they can. For example, since well before As the required groups disappeared, there has been an Nargis, farming households would ask for labor sharing accompanying increase in the number of local charity-type help from other farmers when ploughing or planting their organizations formed at the village level. These take vari- fields. This has continued into SIM 5. For social requests, ous shapes, but particularly common are free funeral service the types of needs have changed over time. In 2017, some of organizations, likely a reflection of the popularity and sup- the most frequent requests were to borrow a motorcycle as port for the national free funeral service charity that has transport to visit a health clinic or borrow an identity card become well known across Myanmar. Twelve SIM villages (as national identity cards are far from ubiquitous in Nargis- reported having this kind of local group active in their com- affected villages) which is needed when trying to register munity. Three villages also reported the emergence of inde- with employers in Yangon or elsewhere. Requests for vol- pendently created microfinance groups, modeled on a local unteer assistance when constructing and repairing homes microfinance program but initiated and run completely by were also common. members of a given village. Case Study 5: Family Connections Help a Household Improve Their Lives A 34-year old woman owns a grocery shop and her husband is a village administrator. Right after getting married, the couple moved to another village where her husband lived. They farmed as tenant farmers on four acres rented from her mother-in-law, and in 2006 also started raising 200 ducks. Nargis destroyed their farming equipment, but they were able to borrow a plough from her brother-in-law and started tenanting again. In 2009, her elder brother asked her and her husband to tenant his 8-acre farm, so she and her husband moved to yet another village where the farm was. They continued to raise ducks; 300 this time. By 2010, her parents-in-law had asked them to return to their original village, so they sold the ducks and returned to tenant their farm. She and her husband wanted to open a grocery shop, as his brother ran one and it was a good business. They heard that there was no grocery shop in a village located in the same tract and moved there in 2013. Several months later they borrowed K700,000 at 5 percent interest per month from the brother-in-law from whom they had gotten the idea and set up their shop. They did well and were able to repay the debt after just one year. Social Relations 31 The increase in government or government-linked/donor- women opportunities to create some independence within funded programs has also driven an expansion of local the household. institutions. Both the Evergreen program and the World Bank-funded National Community Driven Development The expansion of mobile connectivity and 3G services that Program (active in seven villages across Kungyangon and now exist in all SIM villages have also begun to gradually Ngapudaw townships) mandate village-level coordination shift gender norms. All villagers have increased access to groups. These new institutions were also credited by villag- media, including social media, that shows women in differ- ers as helping give rise to the potential for increased female ent contexts, such as social and economic leaders. Aware- leadership locally as they require the inclusion of women in ness of gender issues has also increased, especially among their committee structures. SIM’s female respondents who reported they had more information on issues such as violence against women18 and As highlighted in the discussion on livelihoods, agricultural concerns over human trafficking. They also reported that the labor groups are more common than ever before and in two increased exposure through media of women taking on dif- villages nonagriculture labor groups have also started up. ferent roles in life had made it harder for traditional local The laborers reported they like to work with the group as male leaders to dictate how women should and should not there are many advantages, including increased productiv- act regarding village social affairs. ity, expanded number of working days, and more job oppor- tunities. In addition, working groups provide social benefits The broader political transition has made a significant con- from working with the same people every day, and friend- tribution toward women leadership as well. Aung San Suu ships and cooperation is built through that collaboration. Kyi is seen as an inspiration and as making it acceptable for women to be leaders in the eyes of both men and women (Case Study 6). This is discussed in detail in the following Identity groups chapter on village leadership. Youth groups’ declining role in village affairs Beyond political engagement, migration is having a broader Youth participation in all forms of village activities is down, impact on female economic empowerment. There is more primarily due to migration. Just five villages now have active female migration than male migration thanks to the develop- youth groups, which has had a particular effect on village ment of the garment industry in Yangon. Women who have socioreligious events. Prior to the migration boom, such migrated are more able to identify economic opportunities events were often organized by youth groups; now, only a in the village when they return, and have sought to take handful of youth are available to participate. advantage of village needs. The villagers reported women have more ability to save money as they do not (have the Villagers across the SIM panel report that there is only strong chance to) participate in social, entertainment, gambling, or youth participation for ceremonies at Thingyan (New Year) drinking. In addition, returned female migrants have larger in April and Thadingyut (end of Buddhist Lent) in October, roles in social affairs of the village. the two holidays when migrant workers often return to their home villages. As a result, some villages have reduced the number of other socioreligious events that they hold in an effort to maintain participation levels among those villagers Relations across the Village who remain. Again, communities reported that enthusiasm Villages continue to cooperate and work together at similar for such events remains, but the ability for a village to hold levels and for similar goals, both before Nargis and in previ- them and for villagers to participate is more limited. Youth ous rounds of SIM research. However, social interaction and groups, where they exist, have become somewhat more village cooperation have changed, reflecting the broader accepting of women in leadership positions. trends identified above. Changing norms for female involvement Contributions to village activities ranging from socioeco- social relations nomic development works to socioreligious activities have Women are more involved in local organizational activities evolved away from labor contributions by individual house- thanks to NGOs and development activities that have man- holds. Prior to 2013, villagers would regularly be asked dated a gender balance in committees and other aid-linked to contribute labor alongside a small amount of money development projects. Women are encouraged to participate for village activities—particularly those linked to village and have an opportunity to receive training, including skills like accounting as well as vocational training. The credit pro- grams by international donors also have contributed to cre- 18  As ever, instances of domestic abuse and sexual violence are ating additional livelihood/income sources, and have given almost certainly underreported in this study. 32 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Case Study 6: A Female Villager Becomes an Evergreen Chairwoman A village in Kyaiklat township is led by a 100 household head. One of the villagers, a wealthy lady and a strong supporter of the National League for Democracy (NLD) has taken an active role in improving the welfare of the village. She also has connections with the local NLD party leadership—she was named village tract NLD chair after the general election—and contested the 2015 VTA election as a 10 household head, though she was defeated. In 2016, the village was approved to become part of the Evergreen program. The new VTA wanted loans from the scheme to be accessible to all villages in the tract. However, the influential lady said that her village had been accepted into the program thanks to her proposal at an NLD party meeting and that it was only for one village according to the program’s policy. The VTA responded that the approval had been granted due to his administration’s work and his vil- lage should also benefit. This difference of opinion led to tension between the two. A meeting was held at the woman’s house to set up the Evergreen program’s local management committee. The NLD Member of Parliament attended, as did district- and township-level officers from the Department for Rural Development (DRD). All of the villagers attended and DRD staff explained that the committee would be responsible for the proper management of the loans. That same woman was selected chair of the Evergreen committee and ensured that all the other members were people who had a good relationship with her. As the loans began to be disbursed, her village’s leader and some other villagers became dissatisfied about how the money was being distributed. The chairwoman allowed households with which she had good relations to borrow between five and fifteen lakhs; others could only borrow two to five lakhs. The village leader worked with the VTA to submit a complaint to the DRD township office about the chairwoman. At the monthly NLD party meeting the Member of Parliament advised her directly to explain the case to the more senior district-level DRD office. In 2016, DRD eventually ruled that the committee would need to be reconstituted based on an open village election. The Evergreen chairwoman was interviewed about the program as part of the SIM 5 research. She explained that loan amounts were determined based on the livelihood of the borrower, the level of return that could be expected from their proposed investment, and their track record as a hard worker. She said she had continued to help improve the village after the DRD decision by visiting Naypyitaw with the support of the Member of Parliament and requested support to renovate the village road. Her request was approved and the new road was built. The election for the Evergreen commit- tee had not yet occurred by the time of this research. development. The cash contributions continue largely as reported feeling isolated and excluded from village affairs normal, but labor contributions are now much less common. and religious ceremonies. This is because many households either cannot contribute (because those who might volunteer labor have migrated) or Village self-help activities have also changed, due in part they are busy with their own livelihoods. Instead, they pay to the declining role of the 100 household head, discussed village laborers to take their place, generally for a wage of in the following chapter on leadership. The effect of this between K3,000 and K5,000 per day. decline in village social life stems from leaders’ reduced ability to organize the community for collective action. This While the tendency to utilize more cash payments versus has been the case especially in villages where local leaders, in-kind labor for village contributions was seen across the either 100 household head or VTAs, have found it difficult SIM panel, the burden of village cash contributions remains to adjust to a new political context where they cannot order a heavy one for the poorest households. Village leaders reg- villagers to take part. ularly collect sums of money from villagers for monastery improvements, socioreligious events, or local infrastruc- Several villages where there is no VTA (and thus when the ture maintenance, and while the amounts are generally organization of collective action for village development is based on household incomes, poor households struggle to dependent on the 100 household head) reported a declining give what is expected. In some cases, especially vulnerable frequency in activity. This happened despite the need for households, such as women-headed or child-headed house- such efforts remaining undiminished. holds, were released from the requirement to pay, but then Social Relations 33 Community leaders organize far fewer village meetings than economic relations between the two religious groups with before 2013, a change attributed to a lack of villager interest. both sides saying they buy from and sell to members of the Village leaders complained it is now very difficult to orga- other religion and do not feel cheated. At the same time, nize such a meeting if there was no incentive for villagers to there were no labor sharing and few areas of joint decision attend—for instance, the promise of outside assistance. As making as the two groups largely live separately. a result, since the most recent election in 2015, mass meet- ings are normally only called to discuss government service Mixed villages reported that both Muslims and Buddhists delivery or other outside support, and most meetings report- would attend each other’s major religious festivals and cel- edly focus on education access, the arrival of MADB loans, ebrate jointly, though Muslims would not join in celebratory or National Community Driven Development and Evergreen meals during Buddhist festivals. In small day-to-day needs, Village Development program activities. such as funeral contributions or supervising children walk- ing to school, both communities helped each other. But as they lived in different parts of the village, sharing of farming Intra-village conflict equipment or the hiring of laborers from the other religion was uncommon. Further, as Bamar Buddhists could more and conflict resolution easily move to Yangon, they had higher migration rates, Small-scale conflicts, generally between households, are while Muslim farmers took care to provide job security in relatively frequently of minor issues across many SIM vil- the village for their co-religionists. lages. Such disagreements often revolve around issues of public nuisance (generally related to alcohol abuse), small- There are other social differences, such as the fact that while scale debt repayment issues, disagreements over advance both groups in some villages could cooperate on commu- payments to agricultural laborers, and conflict between the nity infrastructure, this was not always the case. And Bamar livelihood groups. The latter includes, for example, fishers’ Buddhists said that in one village the infrastructure in the anger over perceived pollution of streams by farmers’ fertil- Muslim portion of the village was much better, a difference izer or if a landless household’s duck flock eats a farmer’s perceived as being rooted in stronger religious unity. Most young paddy plants. village leaders are Bamar, though at least one 10 household head is Muslim. Resolution of such disputes generally occurs at the village level. If the disputants cannot find a solution, the first step However, as shown in Case Study 7, religious minorities have is to involve the village leader or village elder and, in most faced difficulties practicing their religion freely or having cases, the issue ends there. If either side feels they were their own place of worship. While tensions between groups unfairly treated, they are able to appeal to the village tract were reported in previous rounds of SIM, in previous years administrator but, given the frequently large distances they were generally due to aid distribution. This finding of between villages, this is only done in rare cases or if the a more general tension between religious groups reflects a issue is particularly important, such as dealing with a large concerning trend of growing localized intolerance toward debt dispute. An exception to this model is in the case of religious minorities across Myanmar (Case Study 7). villages where the VTA lives or villages that are close to a VTA’s home. In these instances, the parties to the dispute may involve the VTA immediately. Bamar-Karen relations Alongside the three Buddhist-Muslim villages, the SIM panel includes several villages with mixed Christian and Buddhist Inter-Ethnic and Inter-Religious populations. The Karen demographics in SIM villages consist of one village which is entirely Karen Christian, three which Social Relations are mixed Karen Christian and Bamar Buddhist, one which is Karen Buddhist and Bamar Buddhist, and one where the Buddhist-Muslim relations majority Karen population has both Christian and Buddhist households. Bamar Buddhist and Burmese Muslim minorities coexist in three villages across the SIM panel. Both communities In one of these villages, there are poor relations and a lack of were asked about their relationships with the other religious trust between the Karen and the Bamar, stemming from the group and, at least in conversation with outsiders, no major aftermath of Nargis. The village was badly affected by the challenges to social harmony were reported. For example, in cyclone, and in the following years many Bamar migrants one village the minority Muslim population expressed their moved to the village because of the large amounts of outside willingness to contribute funds for improvements to the assistance the village received. Karen people have perceived local monastery as it is used as the local shelter. There are these migrants as taking what is not theirs and associate 34 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Case Study 7: Instances of Religious Tension In a village in Bogale township, there were 15 Muslim households before Nargis. After Nargis, the village faced many difficulties. When external assistance came to the village, it was shared across the whole village. Later, when Muslim households received additional aid from Muslim communities in other parts of the country, they did not share it with other villagers. Moreover, a few years later as violence against Muslims erupted in Rakhine state and elsewhere, there was some Bamar-Muslim violence in the township as well. Consequently, the villagers became cautious in their interac- tion with households of a different religion. Moreover, Muslims were not allowed to celebrate their Eid al-Adha. The relationship between the two religious groups has thus been damaged. Since then, some of the Muslim households have moved to Bogale town, and there are only seven households left in the village. In another village, there are several Muslim households. Their primary livelihood is farming and their economic situa- tion is fair. They hire labor only from their Muslim community, which has caused some resentment among Bamar labor- ers who feel they have fewer job opportunities. The Muslim community is not allowed to celebrate their Eid feast and sometimes when they travel, they were asked what religion they had, and other invasive questions. These issues have created divisions between the two groups and both sides feel insecurity and distrust toward the other. them with local misbehavior. There is little social interaction has expanded. At the same time, their ability and willing- between the two groups on any issues and they do not invite ness to respond to requests for help has declined. When the other group to religious or other events. interviewed, villagers attributed the change to communi- ties becoming more ‘business-minded’ and ‘individualistic’ More broadly, even in the other mixed villages, there is no while also recognizing that improved job availability and joint decision making. The Karen are always the major- better wages meant people have less free time to help one ity group; as such, they make most of the decisions and another. the Bamar villagers are unable to be an influence in the ­decision-making process. There is also a gap in terms of rich and poor: the Bamar households tend to be poorer, working Impact of migration as fishers, and living further from the center of the village. Even without the ethno-religious differences, their village Migration is a contributing factor to the decline in socio- position would reduce their ability to be involved in com- religious occasions celebrated in many villages. Youth par- munity decision making. ticipation in village affairs has decreased significantly, with the exception of a small number of occasions when migrant The groups tend to celebrate social occasions on their own, workers return to their homes. Villagers interviewed said and though they might invite people from the other religion, “we only realized how many people were leaving when they it is not very common for people to attend. Inter-household came back for a day of religious significance.” The younger assistance is generally asked from and provided to only generation no longer wants to work in the fields as youths people from the same religious group, though this has been prefer working conditions in town, but still enjoy returning the case historically. There is no sharing of livelihood equip- to the village when they can. ment, mainly because the groups tend to have different pri- mary livelihoods. Location Peri-urban villages have become ever more urbanized in Factors Driving Changes their social practices, such as in organizing weddings, novi- to Village Social Relations tiation ceremonies, and other religious affairs. These vil- lages have a variety of activities not seen in more remote villages, such as December sport competitions. However, Economic standing peri-urban villages are also taking on less positive charac- Economic standing for many villagers has improved over teristics of urban spaces, showing less hospitality to out- the last five years, but in doing so it has led to households siders and being less welcoming to guests from beyond becoming more individualistic. As their economic standing the village. For example, during the SIM field work, three improves, villagers’ ability to contribute more cash for vil- out of four research teams were told they could not stay in lage improvements, religious affairs, and social activities peri-urban villages overnight. These same villages generally Social Relations 35 expressed a reluctance to participate in the research project unexpected events. Villagers came together to raise funds, compared to remote villages, a sentiment that did not come especially in the case of funerals or hospitalizations. While up in previous rounds of SIM. government support to the neediest has begun to increase, communities continue to serve as the backbone of social protection. There are new examples across the panel vil- Affectedness lages of associations that are developing social security funds. Funeral funds to lighten the financial burden on The humanitarian aid following Nargis resulted in intra- a household after a death are perhaps the most common. village social tensions in highly affected villages receiving The member households contribute a fixed amount ranging more assistance that took a long time to disperse. At the from K500 to K1,000 monthly depending on their economic same time, highly affected villages now report very strong standing. Such funds can be found in 13 SIM villages. social relations and cohesion, based at least in part on their close brush with death. As a result, highly affected villages The PACT Global Microfinance Fund, active across 32 of the today seem to work together for all kinds of development SIM villages, has introduced social protection services since activities more effectively than less affected villages. 2013, a first for many Delta communities. PACT provides various forms of social assistance based on an assessment of village needs. The support includes maternal and child health care, funeral expenses, loan write-offs, and financ- Introduction of Social Protection ing its members’ recovery and reconstruction from shocks and disasters. Other NGOs are providing similar assistance, Government social security funds have been introduced in including Marlin, Save the Children, and the International some villages, though these programs only support the most Organization for Migration, but none are as prevalent or well vulnerable. Traditionally, Delta communities have been on thought of as PACT. their own when there was need following accidents or other 5 Leadership and Institutions Over the last five years, local leadership structures in vil- linked to it and receives a stipend from it. Above the VTA lages across Myanmar have changed. In the years imme- lies the township administration, primarily a GAD organiza- diately after Nargis, contact with state bodies and other tion but with assigned officers from other line ministries, outside organizations did expand notably compared to the the number and assignments of whom vary by location. pre-Nargis context. But it was after the 2010 election and The VTA is administratively responsible for a group of three the installation of the reform-minded government under to ten villages and, from the SIM perspective, the most U Thein Sein in 2011 that changes to leadership within and senior leader included in the research. Below the VTA are beyond the village really began to evolve as new government the 100 household heads who, while no longer recognized services were delivered and democratic reforms introduced. by government, provide important and frequently formally appointed assistance to the VTA, and 10 household heads, These changes were not limited to the Ayeyarwady Delta; who continue to have an official role as the electors of the a recent report found government assistance projects more VTA but few other administrative duties. In larger villages than tripled between 2013 and 2015.19 Alongside service with more than 100 households, one of the 100 household expansion came new legislation, massive improvements in heads is sometimes informally designated as the ‘village connectivity (particularly technology but also for transpor- administrator.’ tation infrastructure), and both a level of increased asser- tiveness on the part of villagers and a reduced ability of Legislation in 2012 paved the way for a form local leaders to engage in local authoritarianism. All these of local democracy forces have shaped how leaders are selected, how they are Indirect elections via the 10 household heads for the posi- evaluated in the eyes of the communities, and what their tion of VTA were introduced in 2012. Up to early 2011, VTAs local role has become.20 were appointed by the township administration department with or without consultation with the community. The very first VTA election was then held in 2012 according to the newly passed Village Tract and Ward Administration Law. Leadership in the Delta Every ten households selected a representative who became the 10 household head. These people, mostly men, gathered The most grassroots level of official leadership recognized to elect the village tract administrator. At the same time, a by the Myanmar government is the Village Tract Administra- committee of community elders from across the village tract tor who, while not a formal employee of the central bureau- was created. This body, literally known as the Village Elderly cracy, the General Administration Department (GAD), is and Respected Persons committee, was tasked with setting up, supervising, and monitoring the VTA election process. 19  Enlightened Myanmar Research and World Bank. 2016. Liveli- hoods and Social Change in Rural Myanmar. In most village tracts, once the VTA was elected, they would 20  For a detailed study on leadership, institutions, and change then select candidates to be the 100 household heads in in Myanmar villages, please see World Bank. 2017. Document- the villages around the tract. While local-level politicking ing Myanmar’s Social Transformation Insights from Six Rounds of was relatively nascent in 2012, VTAs generally selected Research on Livelihoods and Social Change in Rural Communities. people who supported them and who they felt had strong Working Paper 8055. 37 38 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After motivations to work for village development. Support of the development reflects broader changes in local leadership in ruling political party (the Union Solidarity and Development rural Myanmar. Party at the time) was on occasion also a factor. In some cases where the first selection was objected to by the villag- Local influence of the VTA has expanded ers, the VTA asked the households to select their preferred since 2013 . . . leader. Since the first election of VTAs, their local influence and administrative power has expanded. Their formal duties Mobile technology’s role in changing relations have an emphasis on local administration, and they are also between leaders responsible for engagement with township-level authorities Myanmar’s mobile telecommunication boom is perhaps and development of the village tract. As there is no longer one of the country’s most significant changes since SIM 4. any officially designated role for the 100 household heads, The results of newfound access to digital connectivity has the VTA role has gradually taken over many of the duties changed social and economic interactions, and has had a they previously had, such as issuing letters or bearing wit- major impact on how local leaders interact with each other. ness to agreements. VTAs are also responsible for dispute In SIM 4 and all prior SIM reports, the ability of village tract resolution, leading village activities focused on both social leadership to engage with villages or represent them to the and development efforts, conveying news to and from the township, was dependent on ease of access to the township township, and other activities deemed by the VTA to be in center and the dedication of the local leader. the village tract’s interest. Moreover, they are key actors in engagement with nongovernment institutions from outside For most local leaders, attending the monthly township the village, such as international development actors. meeting was the total of engagement with the township administration—and leaders from more remote communities Since 2013 and the accompanying expansion of govern- struggled even to attend these. Now, all 17 VTAs who were ment services, VTAs also liaise more with other government interviewed reported that they were in a Viber, a mobile departments both in the village tract itself and in visits to messaging app, social chat group with all other VTAs in the the township. These departments frequently include MADB, township. This group allows the township administrator to the education department, the settlement and land record rapidly disseminate information across the township, allows department, and the health department. VTAs to provide updates to the township administrator, and allows VTAs to communicate with each other, sharing rel- The concentration of administrative responsibility has had evant news. Several village leaders also reported that they mixed effects on SIM villages depending on their relation- used Viber to provide updates from their village to their VTA, ship with their VTA. Six of the 23 SIM villages that do not reflecting an informal but digital communications structure have a VTA living in their village reported their VTA had little that now connects individual villages with the township. interest in working to improve their villages. Instead, their VTA was focused on ensuring that opportunities from the township went to his own home village, rarely even visit- ing other tract villages (Case Study 8). One ethnic minority Formal Leaders village reported being discriminated against by their VTA. Nine years after Nargis, 80 percent of villages had positive . . . as other formal leadership positions have or neutral perceptions of their leaders in 2017 (Table 20). taken a backseat However, there was a strong shift from positive to neutral. Non-VTA leaders have seen their local roles diminish. The By SIM 4, about half the villages had positive perceptions experience of most 100 household heads is the clearest of their formal leaders. By contrast, four years later only example of this decline. 100 household heads interviewed about one-quarter of villages had positive perceptions. This in the SIM research reported having much less autonomy to sign official letters, issue judgments over village disputes, or approve land transactions. Their main responsibilities Table 20: Perceptions of Local Formal Leaders are now to organize activities as directed by the VTA, such by Degree of Affectedness as spreading information about MADB loans, collaborating with the local midwife on a vaccination program for small Affectedness Positive Neutral Negative n.a Total children, or very localized dispute resolution such as marital Highly affected  3  6 3 0 12 disputes or fights between drunks. Moderately  6  8 4 0 18 affected Both villagers and 100 household heads also reported Lightly affected  2  5 1 2 10 reduced ability to organize the villages for local collective works or joint village activities. This change was attributed Total 11 19 8 2 40 Leadership and Institutions 39 Case Study 8: A Non-Tract Village Ignored by Their VTA A village lies in Mawlamyinegyun township, 30 minutes from the township center but one hour by boat (the only pos- sible transport) from the main village tract village. As a result, the VTA, who has held that title since the first VTA elec- tions in 2013, rarely visits the village and only when he cannot avoid it. In 2016, the nurse assigned to the village submitted paperwork to the township administrative office and township health department requesting a new and improved building for the health center. The process took time, and midway through the process, the nurse was reassigned to a new location. Despite the application for a new center still being processed, the VTA made no effort to get approval from the township administration. Villagers felt that the VTA had been uninterested in the health center (and broader village development) from the start. As a result, the existing building is not safe and the roof is damaged. Moreover, the existing health center serves all villages of the tract. However, only the village administrator of the village where the health center is located is asked to pay for the delivery of medicines and other supplies. The VTA does not collect funds from the other village administrators, which villagers consider unfair. In addition, the VTA asked villagers to pay for his transportation should he need to visit the village, for instance regarding MADB loans. The villagers com- plained that the VTA’s neglect delays the development of the village. to both their reduced status vis-à-vis the VTA and broader 10 household heads work for the 100 household heads to political changes that have lessened local leaders’ ability to maintain village arrangements, though in reality they were instruct villagers to do things. In cases where 100 household more likely colleagues. Now, however, their sole responsi- heads have retained local influence, their role has either bilities are organizing villagers for mass meetings, and every shifted toward a focus on socioreligious issues or they are in few years voting in the elections for the VTA. particularly remote villages infrequently visited by the VTA. Village meetings in 2017 only occur occasionally and VTA There are some indications that in the villages most affected elections are even more infrequent. As a result, some villag- by Nargis, the reduction of the 100 household head’s influ- ers found it difficult to remember who was their 10 house- ence has been slower than in less affected villages. The field hold head as the role had so little weight. Whereas in teams identified the most influential leader within each vil- previous years the position was seen as a key associate of lage (across sectors and limited to within-village influence); the 100 household head, helping to maintain village order for nine of the 12 highly affected villages the most important and ensuring security, today their role is greatly reduced. leader was the 100 household heads. This compares to the Without the formal assignment of elector, which is written moderately affected villages where of the 18 communities into the 2012 Ward and Village Tract Administration law, this categorized as such, just five had the 100 household heads position could well disappear entirely. as most important and ten identified the VTA (Table 21). “The role of a 10 household head is just to vote for the village Ten household heads also reported having both a diminished tract administrator. Now no one really knows anymore who role and diminished authority in the village. Traditionally, the 10 household head in the village is.” —Fishers Table 21: Most Important Local Leader by Degree of Affectedness Leadership characteristics: perceived 100 and actual Household Informal Affectedness VTA Head Leader Total Village respondents frequently attributed recovery from Highly affected  3  9 0 12 disasters and improved development standing to the quali- fications and abilities of their village leader. In many cases, Moderately affected 10  5 3 18 the ability of the village to gain improved infrastructure, be Lightly affected  3  3 4 10 included in government programs (the Evergreen program Total 16 17 7 40 is a notable example) and other benefits that come from 40 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After beyond the village, is based on the leader’s ability to net- village leaders in SIM 5 saw a similar proportion of young work, and willingness to volunteer their time and incur the leaders (between the ages of 18 and 45) but a significant financial costs of engaging at the township. reduction in the proportion of well-educated leaders (either a high school graduate or having attended university). Economic standing plays a key role in whether a leader can be effective, largely because wealth is frequently a positive Leadership changes following influence when dealing with township authorities, peer col- the 2015 elections leagues, and villagers. Having a higher education is another The most recent VTA election was held in late 2015/early criterion for strong leadership, yet it falls behind wealth and 2016 following Myanmar’s national election and led to nota- having a strong network. ble local leadership changes (Table 23). Particularly note- worthy was the high turnover in highly affected villages as a The qualities of a good leader as perceived by result of both elections. Of the 40 villages in the SIM panel, communities are changing, but are not backed 18 are so-called ‘tract villages,’ where the VTA tends to live. up by election results Of these, 10, or 59 percent, had changed VTA by March 2016. Interviewees across villages reported that their views on Of the remaining 22 villages, 13 (59 percent) changed 100 what was required from a good leader had evolved in recent household heads shortly after the election. Though there is years and since the election; however, this is not reflected no formal process for 100 household head selection, as they in the identity of their leaders following the 2015 local are perceived as assistants to the VTA when a new VTA is elections. elected, new 100 household heads often follow. Given their position as the most visible local leaders, such turnover has In early SIM rounds, the traditional characteristics of ‘Benev- implications for local leadership dynamics—especially as olence, Interest, Sacrifice’ (Say Ta Nar, War Ta Nar, Anit Nar) the appointment to 100 household head is often based on was what communities identified as important for village personal relationships with the VTA rather than on ability or leaders. In SIM 5, interviewed villagers identified other administrative skill. attributes they felt were necessary for a good leader. These included improved education, a strong network of contacts Factors driving leadership changes both in the village and beyond, ‘intelligence’ (defined as the The common reasons behind the shifts in leadership can be ability to think critically about an issue), and the ability to divided into two categories. First, there are the cases, five speak some English—which is particularly important when during the most recent local elections, where the previous dealing with NGOs. leader voluntarily vacated the position. The reasons for this tended to be old age, a reluctance to continue shouldering “Today, a leader should be educated and able to speak Eng- the burden of local leadership, or the ability to maintain lish. When we visit the town, we can find that signboards are family control by (informally) passing the position on to a written in English. It is not good if we cannot read those. We relative. The other group of reasons, accounting for 18 of do not even know where to go. So, as a leader, he should be the local leader changes across the SIM panel, centered on able to speak and understand English.” the dissatisfaction of the community and involved cases —Small farmer where the village leader lost out to a competitor. Examples included perceptions of localized corruption and a view that However, the data gathered on current village-level leaders the old leader was not educated enough or unable to create does not reflect this change (Table 22). Compared to SIM 4, the networks that would advance the interests of the village. Localized corruption was defined as the leader asking for money when called upon to resolve local disputes, distrib- Table 22: Leader Age and Education, 2013 and 2017, uting aid to the village unfairly, and, most importantly, the by Degree of Affectedness High School Table 23: New Leaders Following Local Elections Leader Graduate Age 18–45 or University 2012/13 Local 2015/16 Local Affectedness 2013 2017 2013 2017 Affectedness Elections Elections Highly affected  5  6  6  4 Highly affected  8  8 Moderately  9  7 14  7 Moderately 12 10 affected affected Lightly affected  4  2  6  4 Lightly affected  6  5 Total 18 15 26 15 Total 26 23 Leadership and Institutions 41 view that the VTA only delivered opportunities from the replaced, and did not have the beyond-village network that township to their own home village. In these cases, villag- remains a key determinant of government support a village ers reported that the VTA would rarely visit other tract vil- will receive. lages and, in response, the other villages in the tract would try hard to remove that VTA and install one from their own The new recommended criteria on age and education set by village. the government for the VTA under the 2012 law also con- tributed to leadership changes in 2015, namely that the VTA The leaders who resigned tended to be those originally must be at least 25 years of age, have ‘sufficient resources appointed by the GAD prior to the reform period and who for his livelihood’ and a ‘fair education.’ While not compul- had won their first election in 2012. These leaders had sory, in many cases the GAD provided an interpretation that proven their willingness to contribute time, money, and new VTAs should be well educated, be at least a high school effort working for the development of the village. But after graduate and preferably a university graduate, and be under many years of work and the increasing burden on the VTA the age of 30. Party affiliation was a further driver of leader- without any salary,21 many were getting tired of the position ship turnover: most of the new VTAs were strong supporters and sought to resign. The increasing willingness of villagers of the National League for Democracy and those VTAs who to be vocal in their opposition to certain VTA decisions was were members of Union Solidarity and Development Party also a reason cited by some former VTAs as a contributing were less likely to win. factor to their retirement. New roles for women Retirements create leadership gaps Village leadership opportunities, previously almost exclu- The departure of some village leaders created a gap where sively restricted to men, are expanding, albeit slowly, to few qualified candidates for the position existed and the women. Six villages across the SIM panel now have a female abilities of local leaders declined. Villagers who were well 10 household head, and despite the decline in regular influ- qualified to lead the community had little interest, as they ence of this position, the increase means that women now thought it more important to focus on their own livelihoods have a more direct voice in local elections to select VTAs. and businesses, while those who were willing to run for One result of this change was the election of the first female the position often lacked management and communication VTA in the SIM panel (Case Study 9). While data following skills, had less influence in the community than those they the latest VTA elections is not available, prior to the 2015 Case Study 9: A Village Tract Elects a Female VTA for the First Time A 48-year-old woman was elected as the village tract administer in 2015. She is a university graduate whose livelihood is running a grocery shop, a small bamboo plantation, and trading in paddy. She and her household are among the wealthiest in the village. Her extended family has a history of village leadership (her father was the village leader), and they all are active in their work for the development of the community. For example, her younger brother founded a free school in the village and later became a Yangon Regional Member of Parliament. When the 2015 election arrived, she did not plan to participate because she was occupied with her business. However, everyone in the village wanted her to become the 10 household head and vote in the election. She agreed and then the ya mi ya pha (village elder) and the local monk lobbied her to run for VTA. She agreed to that and ended up winning the election unopposed. As a VTA, she is also responsible for the challenges facing other villagers. For example, she is currently negotiating with farmers whose land lies along the route where they plan to build an intra-village road. Her ability to manage both the social and financial aspects of the VTA role has won her the respect of most villagers, though she is disliked by some casual laborers because she has imposed fines for certain activities (including unlicensed fishing and cutting bamboo) and providing only those people she thinks are financially deserving with the recommendation required to take a loan. VTAs are supposed to receive a stipend from GAD plus allowances for the cost of their duties. However, the disbursement of this money 21  appears extremely varied across townships. 42 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After election there were just 42 female VTAs across the entire Wealthy businessmen in the village also often have an infor- country, or 0.25 percent. mal leadership role. Their activities are varied, with different individuals engaging in many different roles with the excep- The broader political transition has made a significant con- tion of administrative duties that are the sole preserve of tribution toward women leadership. Daw Suu is seen as an the VTA. Mostly, however, informal leaders engage in event inspiration and is making it acceptable for women to be organization for religious occasions, organizing collective leaders in the eyes of both men and women. While this pro- action for village infrastructure development and social wel- cess is understood to have begun in 2012 with her election fare, dealing with nongovernment outside actors for local to a parliamentary position, it has expanded since 2016 and needs, and a very informal role in local dispute resolution her ascension to the role of state counsellor. (providing advice where to go and whom to approach). The expanded participation in leadership roles for women Even in villages where they have little formal roles, religious was attributed in large part to the Evergreen program and leaders (generally monks, though pastors in Christian vil- other aid projects that required women’s involvement in lages) have had persistently high levels of influence in vil- local management committees. Women also played an lages since well before Nargis. expanded leadership role in some cases by default: as liveli- hood opportunities improved, men were frequently too busy Wealthy individuals in the village have also maintained to engage in village affairs and so women replaced their hus- their influence. In other cases, newer informal leaders bands’ positions, including in some cases as 10 household have appeared, such as political party representatives who heads. appeared in many villages prior to the 2015 election and worked aggressively. Since the election, such activity is Women leaders were perceived by communities as being reduced, but in general local activism among informal lead- more supportive and open and were popular among ers has increased. In some cases, such leaders have sought research respondents. Where they existed, younger leaders to bypass the VTA or other local administrative positions also received good reviews, as they were more willing to and bring issues to the attention of the township or even travel to the township on behalf of the village than older local Members of Parliament directly. leaders and are seen as more active on behalf of the village, especially when it comes to developing a strong network of In general, informal leaders maintain good relationships contacts at the township level. with formal leaders and hold different spheres of direct influence. Nonetheless, there are occasional cases of power competition between the two. While power competition does arise, for the most part the village leadership is able to Informal Leaders maintain a balance between the administrative roles played by the 100 household head and VTA, and the more socio- Informal leaders in Ayeyarwady Delta villages include the religious side of the village led by informal leaders. village elders, village monks or pastors, political party orga- nizers, and the heads of village development committees. 6 Infrastructure and Housing The condition of village infrastructure and housing reflects improvements have been to local roads, to monasteries, how the condition of villages has evolved since Nargis and bridges, schools, and water ponds—all were found in at least what the continuing effects of the disaster are. While there half of the SIM panel villages. has been significant investment in infrastructure from both government and donors since Nargis, SIM villages con- While village context is the most important factor in deter- tinue to face major constraints. Roads have been improved, mining the quality of local infrastructure, Nargis’ shadow but the lack of quality or inability to use them during the remains. Villages highly affected by the cyclone continue to monsoon means villagers are unable to rely on them. This show less likelihood of having improved across the board analysis first assesses the state of village-level infrastruc- and infrastructure is no different (Table 25). For infrastruc- ture before turning to the question of housing and whether ture, the challenges of restarting livelihood activities due communities’ houses have returned to (or improved beyond) to asset and manpower losses combined with high levels pre-Nargis levels. of debt and a lack of confidence undercut the ability of highly affected villages to improve infrastructure on their own. Only two of the twelve highly affected villages demon- strated improved infrastructure and this was thanks mainly Village Infrastructure to strong local leadership. Overall, since SIM 4, most villages have experienced Local leaders in these communities included wealthier indi- some improvement in local infrastructure despite the sig- viduals who could make large cash contributions to their nificant gaps that remain (Table 24).22 The most common villages’ development, could pay for labor costs where oth- ers could not, and had both a strong interest in developing 22  Infrastructure quality is defined as follows: housing (“good”: the village and good networks beyond the community upon most houses are of a high quality, made of concrete, some with which they could call. Villages where infrastructure devel- two stories, all are in good condition; “fair”: most houses are opment or renovation was slow, often attributed it to a lack made of wood with zinc/tin roofing, most are in good condition; of manpower, poor connectivity to a township, and few job “poor”: broken down, homes in need of repair, many are made opportunities compared to other villages. of bamboo walls and nipa palm roofs); schools (“good”: made of concrete, well constructed to withstand storms, large enough for the number of students to grow; “fair”: older building in need of updating but fit to purpose and just large enough for current needs, individual households have to store own water); health students; “poor”: run-down building in need of repair, not large (“good”: has an own functioning health center or is close to town enough for current student body; “missing”: no school building in with access to a district hospital; “fair”: has health care staff in the village); water/­sanitation (“good”: easy access to ground water village but not well equipped beyond basic services for children and/or storage tanks and/or well-maintained water ponds; “fair”: and pregnant women; “poor”: no permanent presence, but regular enough water ponds, no concerns over water shortages, access to visits by health care staff for natal care/child vaccinations; “miss- ground water via pump; “poor”: water ponds are in bad condition, ing”: no staff in village and rarely visited because of remoteness vulnerable to salt water intrusion and villagers have to buy water and poor transport options, villagers have to travel to other village during dry season; “missing”: not enough water ponds for village for basic care, rely on traditional medicine). 43 44 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Table 24: Quality of Key Infrastructure, 2013 and 2017 2013 2017 Good Fair Poor None Good Fair Poor None Housing 13 19 8  0 13 18 9  0 Schools 23 10 2  5 19 11 7  3 Water 11 19 8  2 14 17 8  1 Health 10 3 0 27 16  7 7 10 Table 25: Good Quality of Key Infrastructure, On education, there has been notable progress since Nargis by Degree of Affectedness on expanding the number of school buildings across SIM vil- lages. The number of villages without any school buildings has Water/ declined from seven villages in SIM 4 to just three in SIM 5, Affectedness Housing Schools Sanitation Health most of which are very small villages (no more than 44 house- Highly affected  3  5  2  1 holds). The one exception is near a very large village that can Moderately  6 11  7 12 provide education up to the high school level. Comparing affected23 education access to affectedness, there is little correlation; instead, school buildings tend to reflect the size of a village. Lightly affected  4  2  5  1 Total 13 18 14 14 Since 2008, SIM has counted 50 school improvement proj- ects across 36 SIM villages, a mix of repair projects, and construction of new schools. However, the SIM research team found that subcontractors responsible for construc- Infrastructure status across SIM villages tion frequently failed to build suitable buildings. Further, The status of infrastructure differs from village to village, apart from providing the money, the government failed to but there are clear patterns that emerge across the SIM inspect either the construction process or monitor schools panel, with certain sectors of infrastructure having been the after completion. As a result, seven villages reported that focus of recovery efforts while others remain in poor quality. school buildings were in poor condition despite having been built recently. Social service infrastructure 23 Transportation and connectivity Even though five health centers were built over the past four years, for a total of 17 in the panel, more than half of SIM The quality of transportation infrastructure varies widely villages still do not have one. Further, having a village health across the SIM panel and is closely correlated with the center is not determinative of ‘good’ access to health ser- degree of affectedness in Nargis. Most villages with “poor” vices, as many of the centers that do exist, as well as many connectivity24 in 2017 were highly affected by Nargis, schools, are often of poor quality. During the research, ten accounting for eight of the twelve poorly connected commu- villages highlighted the lack of a health center and said that nities. Furthermore, villages in the highly affected category they wanted the government to build one, but that no assis- are almost exclusively only reachable by boat—an inherently tance had not been forthcoming. In the remaining villages, more vulnerable and uncertain mode of transport. when interviewees were asked about health care, they cited that they were either very small villages with relatively easy While transportation infrastructure improvements were a access to a larger village with a clinic, or were near towns regularly cited need in many communities and there is some that offered improved health care services. evidence of a new focus by both government and self-help, community driven projects, the villages most vulnerable to Assessed by affectedness, of the twelve highly affected vil- Nargis-like storms remain both poorly connected to urban lages, only one had good access to health care compared to centers and fully dependent on water transport. Table 26 12 of 18 moderately affected villages. Lightly affected vil- illustrates the linkage between infrastructure quality and lages also lagged behind, likely a reflection of aid disparities between lightly affected villages and others in the SIM panel. 24  Villages with a travel time to the nearest town of 30 minutes The relatively better quality of infrastructure in moderately 23  or less are defined as having “good” connectivity; villages with a affected villages may be attributed to better connectivity and travel time of 30 minutes to 1 hour have “fair” connectivity; vil- higher aid levels. lages with a travel time of over 1 hour have “poor” connectivity. Infrastructure and Housing 45 Table 26: Number of Villages with Good Quality unsafe motorbike driving on the roads had led to more acci- of Key Infrastructure, by Connectivity dents. In villages close to town, local shopkeepers reported they were struggling as villagers could now access township Houses Schools Water Health stores that had cheaper prices and more variety in their Good (16 villages)  7  8  6  8 products. Fair (12 villages)  7  6  9  4 Connectivity is not only about the ability to move people and Poor (12 villages)  2  0  1  2 goods, but also information. The rise of mobile telecommu- nications around Myanmar has been well covered by jour- Total 14 14 16 14 nalists and researchers alike, and its impact is seen across the SIM panel. In 2013 during the SIM 4 research period, not one of the SIM villages had mobile Internet. In 2017, every connectivity to township centers. Villages that have good or village had a 3G network, and interviews with key leaders fair connectivity are more likely to have good access to vari- and focus groups were often arranged via mobile phone or ous forms of infrastructure. This reflects an improved abil- even via Viber. Though almost one-third of SIM villages con- ity to access materials, expertise, support, and engagement tinue to struggle with poor physical connectivity, they are from authorities compared to the most remote communities. now connected to the outside world via the Internet. Since SIM 4, there have been some improvements in com- Good access to 3G Internet is beginning to affect the local munities’ transportation infrastructure (Table 27). Seven vil- economy, as well as allowing villagers to easily communi- lages reported either that a new road to the township center cate with friends and family beyond the village. While not had been constructed, or that an existing road had been yet widespread, SIM research found examples of villag- improved. Seventeen villages have had improvement work ers using 3G to check commodity prices, find out about done on inter-village roads (many of which are just wide job opportunities in town or in Yangon, and use GPS and enough for a motorcycle) while 13 villages reported that weather forecasts in their fishing activities. there were new public transport options, generally motor- cycle taxis or local bus/truck services, available to them. Water, sanitation, and hygiene infrastructure At the village level, as water, sanitation, and hygiene infra- A small number of villages also reported that boat access structure primarily reflects drinking water infrastructure, had improved their connectivity, with one village located community toilets (especially on school compounds) are near a town becoming a junction for trading boats following also important assets. Since Nargis, the provision of a con- improvements to their jetty (rebuilt in steel instead of wood). tinuous supply of drinking water to communities has been a focus of government and international donors alike, gener- Improved connectivity has enabled labor groups to orga- ally taking the form of drinking water pond improvements. nize more easily and access employment opportunities in Yet, nine years after Nargis just three villages reported hav- other villages, eased resupply for shopkeepers, and given ing improved water sources—defined as bore wells or other local brokers (especially fish brokers) access to new markets direct access to ground water. including Yangon. Some farmers reported that they now sell their paddy harvest directly to brokers in town rather than Water storage has received increased investment, with many waiting for them to come to the farm gate. villages since SIM 4 having a 5,000-gallon water storage facility installed in their villages, which is generally located However, improved access in and out of villages does come in the school compound so as to ensure water supply for stu- with a social cost. For example, villagers reported an increase dents. However, villages often still face challenges accessing in the number of youth who drank alcohol excessively, while fresh water during the dry season, when salt water intrusion Table 27: Connectivity by Degree of Affectedness Connectivity to Nearest Town Mode of Transport Affectedness Total Good Fair Poor Boat Road Highly affected 12  1  3  8 11  1 Moderately affected 18 10  7  1  8 10 Lightly affected 10  5  2  3  6  4 Total 40 16 12 12 25 15 46 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After into the water ponds is common. During these periods, vil- Table 28: Villages ‘Left Behind’ by Infrastructure lagers have to either travel to access water ponds in other Development villages or purchase drinking water. Villages Left Behind Other public services   (11) One of the most important issues for communities in the Highly affected (12 villages) 4/33% years following Nargis was ensuring there was a village Moderately affected (18 villages) 3/17% building able to withstand bad storms in the future. Such Poor connectivity (12 villages) 6/50% shelters would provide space for both storing supplies that they needed to survive a storm, and also a place where those Fair connectivity (12 villages) 3/25% with less resistant homes could shelter. While in a few cases formal cyclone shelters were constructed, for the most part improvements to existing structures were made. The most common, well-built structure in most villages is the monas- A new focus on embankments: tery (or in Christian communities, the church). Improving changing infrastructure priorities these religious structures (strengthened monasteries were observed in eight villages, churches in two villages) served During SIM 4 research and in earlier rounds, villagers pri- a dual purpose therefore, both as a religious building and oritized transportation, education, and health care infra- as a way to reduce the impact of disasters, and has been structure as their highest priority needs. In 2017, however, a key focus of community self-help activities since Nargis. villagers reported that they increasingly valued livelihood- In some cases, both government and NGOs have supported related infrastructure and electricity as the most important such efforts, but outside funding to improve religious build- needs for their communities. ings is uncommon. The clearest example of this shift toward prioritization of Other public services that have been funded and imple- different forms of infrastructure is embankments, newly mented by either the community itself or outside actors emphasized for their contributions to farming livelihoods include embankments, electricity provision (two villages, and village safety. Six villages had received embankment the first indication of the national electrification project), decades before Nargis as part of a World Bank project, and mangrove plantations (three villages), a village administra- more and more villages are either asking the government to tion office (two villages), and a community hall (two vil- support embankments or trying to build them themselves. lages). Notably, all communities have access to electricity through self-financed home solar panels, though the pro- In one example, farmers from 17 different villages in one portion of households who have been able to afford them part of Bogale township joined together in 2017 and sub- varies from village to village. Since SIM 4, there has been mitted a proposal to the government. Their plan was to occasional donor support for communities to purchase create one large embankment protecting 20,000 acres of solar panels but no systematic efforts to develop localized farmland from all the villages. In other cases, farmers have electrification. worked to build small embankments, either by hand or by renting machines, around their own paddy land in an effort to reduce both flooding and the risk of pests. However, such Villages left behind investments are expensive and out of reach for many small farmers. Villages where infrastructure remains of a poor qual- ity across sectors tend to suffer from at least one signifi- cant problem that has prevented them from realizing any Financial support for infrastructure improvements. The research team found that the chal- lenges that could retard infrastructure development to this comes from different sources—and level generally revolved around affectedness and location funders have varying priorities (Table 28). Overall, about one-quarter of villages are con- sidered left behind.25 Infrastructure improvements have been driven by a number of factors, including early on NGO intervention, increasing government funding over time, and collective action at the Villages were defined as “left behind” if at least two types of 25  village level. The political transactions and a shift toward infrastructure (housing, schools, water/sanitation, health) were investment in infrastructure have also been major contribut- identified as either of poor quality or nonexistent. ing factors for improvements. Infrastructure and Housing 47 Government, (national and international) NGOs, and village combined. At the same time, NGOs have been much more self-help groups are the most common sources of funding focused on improving water and sanitation (29 projects) for local infrastructure projects. As shown in Table 29, there while communities have targeted improvements of religious have been a higher number of transportation infrastructure buildings as their key focus (22 projects). projects (116) than any other sector, with a roughly even split between government-funded, NGO-funded, and village- Reviewing infrastructure projects by affectedness, there funded activities. This distribution highlights the impor- appears to be some correlation between affectedness and tance placed on transportation and improving the quality of the number of projects but this depends on the funding intra-village tract roads and motorcycle tracks as part of a source (Table 30). Of the two largest funders, government broader effort to improve the local economy and livelihood and NGOs, the government has financed fewer projects, on opportunities (Case Study 10). average, in highly affected communities. By contrast, highly affected communities have received as much support from This unity over prioritization ends at transportation, how- NGOs as moderately affected ones. The limited sample ever. Health and education infrastructure is dominated size notwithstanding, the evidence suggests that the post- by government funding, with a total of 36 projects across Nargis NGO efforts to reach the most affected communities the panel compared to 21 for NGO and villages sources have been very important for recovery. At the same time, Table 29: Infrastructure Funding Source by Sector Religious Water and Roads Schools Clinics Buildings Sanitation Other Government  29 25 11  0  8 12 NGOs  25 12  0  3 29 16 Village initiative  39  6  3 22  2  7 Private sector  10  5  1  7  6  2 Joint  13  2  0  3  1  5 Total 116 50 15 35 46 42 Note: Number of projects across the 40 panel villages since Nargis. Case Study 10: Improved Transportation Infrastructure Results in Improved Livelihoods One village in Pyaypon township is near the open sea, with an economy that is largely dependent on fishing. It is five miles from the main village in the tract and is accessible by car, either along the road or via the beach. Most villagers work as fishers and a smaller number as fish collectors who then sell the catch to larger markets. Some households also have secondary livelihoods running grocery shops of different sizes or workshops for motorcycles. Before Nargis the road that connected the village to the tract flooded regularly and villagers could only go by foot, with water often up to their knees. The road condition in the years after Nargis was little better. In 2016, however, the com- munity came together to construct a better road. Now that they have this new road, both cars and motorcycles can easily access the village and a few grocery stores even sell petrol and diesel to outsiders passing through. It has become much easier for shopkeepers to keep goods in stock. Before, they had to ask boatmen to go to town and buy goods, which could often take two days. With the road, they can access the township market within half a day. The new road has benefited more than shopkeepers, however. There are new job opportunities to work as motorbike taxi drivers and cycle carriers, which have allowed fishers to supplement their incomes. Some villagers have opened up workshops to repair both bicycles and motorbikes. There is now a travel service from the village tract via ferry bus, and two local shops act as points of sale for tickets. Perhaps most importantly however, the village’s lobster catch can be transported directly to Yangon, where it receives as much as five times the price of the market they had previously sold to. 48 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Table 30: Number of Infrastructure Projects in Villages by Funding Source and Degree of Affectedness Affectedness Government NGO Village Initiative Private Sector Joint Contribution Highly affected 17 37 21  8  1 Moderately affected 45 39 32 19 15 Lightly affected 23 9 26  4  8 Total 85 85 79 31 24 government initiatives to expand service delivery have been Those households reporting improvements tend to have at hampered by the difficulties of reaching the Delta’s most least one of the following characteristics: migrant income, remote and vulnerable communities. multiple local income sources, many working family mem- bers, or external assistance. Migration was particularly criti- cal: 70 percent of households with migrant family members reported having built entirely new houses since Nargis. Inter-Village Infrastructure For the 20 percent who have been unable to improve the Investment in inter-village infrastructure is much more reli- quality of their housing, it was generally attributed to at ant on government than village infrastructure and as such as least one of the following challenges: reliance on only one lagged behind. SIM 5 found eight inter-village road projects income source, which meant a high ratio of dependent family in six villages and township road projects in five villages members, health shocks, a lack of education to get a better (Figure 5). The majority are solely funded by government, job, or social problems like alcohol or gambling addictions. though inter-village road projects are also supported by vil- lages and private sector contributions. Township roads are In a small number of communities particularly badly affected fully reliant on the government, with the exception of one by Nargis, villagers reported that they were reluctant to village where the community has collected money to help invest in improved housing, as a future storm would simply advance the project. destroy the houses. In these villages, those who could afford it invested in alternative housing in the township rather than in stronger village homes. Beyond these villages, there was little difference in housing improvements by degree of Village Housing affectedness, as housing was more a reflection of economic ability to invest than the longer term effects of Nargis (Case Housing quality across SIM villages Study 11). has improved Approximately 80 percent of households reported the qual- ity of their house had improved compared to before Nargis. Figure 5: Proportion of Funders to Inter-Village Infrastructure Improvement 5 4 3 2 1 0 Government Village initiative Joint NGO Village to town road Inter-village road Infrastructure and Housing 49 Case Study 11: Home Improvements Thanks to Migration U Hla Mg is 57 years old and the head of a group of carpenters in a village in Dedaye township. He has three sons and two daughters and participates in many village development activities. Before Nargis, his family used to be among the poorest in the village, living in a home with a nipa palm roof and bamboo walls. He worked as a small fisherman, carried water to government staff houses, and during the summer season worked as a permanent farming laborer. Later he did construction work as day laborer. His wife also did various jobs, including picking vegetables for the farmers that she would sell in town, opening a carom-board and betel shop, and selling boiled corn and watermelon slices in the village. Although they worked hard, their financial condition was not getting better, and their eldest son had to quit school because they could not afford the expenses. Their second son made grass mats and sold them at wholesale prices to pay for his school expenses; he passed the matriculation exam. After matriculating he went to Yangon and made a plan with some friends to move abroad. His family pawned the plots of land where their home was along with that of two other family members. They raised K1.2 million with 10 percent per month interest and borrowed a further K600,000 at 20 percent interest from a private money lender. With this money, the second son was able to move to Malaysia and found a good factory job with a friendly employer. On holidays, he cleaned the factory compound for a side income and also worked as a security guard at night. At first, he made K200,000 per month, but after one year had doubled it to K400,000. He eventually became a super- visor after five years. Within two years, his remittances allowed his family to repay their debts and reclaim their house plots. In 2010, his elder brother followed him to Malaysia but worked in a different factory and had a bad experience, returning to his home village after just one year. He tried again, however, and this time got a job in the same factory as his brother. Together they could send K1 million a month to their parents. In 2011, their parents invested the remit- tances by buying house plots. They bought five plots in the name of their daughters and sons. Following improvements, the house plots they bought for K5 million are now worth K30 million. In 2012, they started to build a new house with remittance income. They took their time completing it, but in 2016 their house was finally finished. It is a very modern house, with amenities like a TV, electricity transformer, refrigerator, and other furniture. It cost K25 million. In 2013, the youngest son followed his brothers after failing the matriculation exam. From then on they sent back remit- tances every two or three months, each time sending around K6 million. In 2014, the eldest son returned and married a women from their village. His parents gave him the house plot they had bought in his name. That year they also donated K3 million to the monastery. In 2016, the youngest son also returned to the village and married a Shan woman. Then they both migrated to Thailand for work. The middle son had come back for holidays in 2012 before inter-religious violence erupted. His parents and other family members asked him not to go back to Malaysia. He thus stayed in the village and started working as a money lender, providing loans at low interest rates to villagers. He also ran a grocery shop. He made friends with Pact Myanmar committee members and put his money in the committee to lend. His youngest sister is now attending high school. The youngest son helps manage his parents’ financial situation and has saved K10 million for each of them in a bank account. Although their financial condition is much improved, U Hla Mg still works as the head of a group of carpenters in the village and always participates in the village development activities. He said: “I have never thought that I will have this kind of life, not even dreamt of this life.” 50 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Factors driving new home designs a result, they would need to take a small loan to put the final touches on their home. While this appeared to be a common and building decisions and largely anodyne practice, there were cases where the Households are altering the kinds of homes they build based expense of a new home had driven households deeply into on their experience with Nargis and other weather shocks, debt. as well as expectations of what the effects of climate change may bring. Many households, especially in moderately and lightly affected villages, sought to invest in homes that Social factors behind new houses would be better able to withstand future cyclones or other extreme weather events. Social pressure now also influences villagers’ decisions to improve their homes. SIM 5 research found that among Home improvements primarily meant improved materials, households where at least one person had migrated, informal though the research team also observed new designs in competition existed to have good quality housing. While the anticipation of flooding or high winds. In the past, poorer first objective for families with remittance income is a bet- homes were made of bamboo with nipa palm thatch roofs, ter ability to cope with daily basic needs, when remittances while better off homes had wood frames and tin roofs. SIM reach a certain level, the family members who remain in the 5 found that newly improved homes tend to be constructed village turn to opportunities, investing in income generating of concrete bricks and cement with wooden upper frames activities, buying land, and making housing improvement. and tin roofs. Prior to Nargis, such homes were very rare in the Delta and especially uncommon in villages. Now, there One calculation that households make is that they want to are several such homes in most villages. Better post-footings be able to host guests of the migrated family member, in foundations are also increasingly common as they better some cases the migrant worker’s senior colleague or older withstand termites—a constant threat after flooding. While friend, in a good quality house. Further, housing improve- improved protection is an important driver, the change also ment is a change in the village that is easy to visualize and reflects the desire to have homes that are similar to ones to show off to both other households with migrant work- found in towns—and the increasing cost of wood as a con- ers—and to the household member themselves when they struction material to the point where it is almost as expen- return as evidence of their hard work on behalf of the family. sive as concrete and cement. Villagers reported that they were even considering building their bathroom and toilet In one mixed village where both Buddhist and Muslim vil- within their home, an amenity that has almost never been lagers live, the Buddhists sought to improve their housing found in rural areas. conditions. In this village, while in a minority, the Muslim community tends to own more land than the Buddhists and Going into debt to finance home construction is increasingly are wealthier. Buddhists in the village reported that they common, though households mostly reported that such debt wanted to ensure that when guests came to the village, they was manageable. Such households generally have saved would not be embarrassed by the low quality of their accom- almost enough to build a new home but reported that the modation compared to the Muslims. construction costs almost always exceeded expectations. As 7 Recovery and Resilience The SIM 4 report examined five years of post-Nargis recov- But at the same time most villages remain highly vulner- ery efforts and found that while some villages or groups able to weather and climate shock, are experiencing social within villages were on track to recover, or had recovered, change stemming from the Delta’s migration boom, and con- there was more variation than pattern. Villages and groups tinue to rely on primary livelihoods that are as uncertain as (socioeconomic, livelihood, and other) within those villages ever. While the situation facing SIM villages has no doubt recovered differently and at different paces, with few lin- improved over the past four years, their position remains ear trajectories for recovery visible in the experiences of tenuous. the SIM panel. And most importantly, it was clear that there were many factors determining the pace of a village’s recov- As SIM 5 seeks to understand not just how communities ery and that the interplay between those factors resulted in have recovered from Nargis but how they will fare in the many possible outcomes. In sum, the situation in 2013 was future and in future disasters, this chapter also presents an complex, with just a handful of communities that could be assessment of resilience in the Ayeyarwady Delta. Through said to have recovered. both a review of existing literature and fieldwork experi- ence, the SIM research defines resilience as a household’s Four years later, the situation is quite different, though the or community’s ability to recover faster from a disaster and complexity of what drives recovery remains. Evaluated for reduce the impact of a disaster.26 economic recovery and the standing of social relations, almost two-thirds of villages, especially those only lightly or moderately affected, can be described as recovered to pre-Nargis levels (Table 31). Recovery from a disaster for Recovery either a household or a community is dependent on a wide range of factors that stretch from the degree of affectedness Using SIM’s longitudinal data and its field experience, the to geographic location to assets that survived the event, as research team identified a variety of factors that contrib- discussed in this chapter. uted to the economic and social recovery of households and communities over time. Researchers assessed the impor- tance of these factors within 1–2 years after Nargis (SIM 2 and 3) and 9 years after Nargis (SIM 5). Table 31: Recovery of Villages by Degree of Affectedness, 2017 Affectedness Recovered Not Recovered Total 26  The analysis in this chapter is based on the assessment of recov- Highly affected  4  8 12 ery and resilience by the SIM 5 research team. Researchers were Moderately affected 13  5 18 able to compare villages across the panel whereas villagers tended Lightly affected  8  2 10 to compare the situation of their village only with that of other villages in the tract. Consequently, data may on occasion diverge Total 25 15 40 somewhat from those provided by villagers. 53 54 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Household recovery Factors of a more social nature generally scored as less important, perhaps because relations were strong and could The SIM team identified 18 factors that helped households be relied on. Good relations with family, the importance of recover from Nargis (Table 32). There is no difference in the having a male household head, and the ability to be entre- importance of these factors by degree of affectedness. preneurial were all seen as moderately important in the early stages of disaster recovery. In the immediate aftermath of Nargis there are wide range of important factors that determined a household’s ability to Over time, however, the importance of various factors has recover and the bulk of those that scored as very important changed. In 2017, the importance of migration and remit- (i.e., ‘1’) were livelihood-centric. The survival of economic tances has increased as has the ability to be entrepreneur- assets, availability of jobs, management of resources, avail- ial. At the same time, the importance of credit access has ability of affordable credit, and number of family members diminished, likely a reflection of the comparative ease of fit and able to work were all identified as very important to borrowing in 2017 compared to 2008, as has the impor- household recovery in the years immediately following Nar- tance of external aid. This latter finding indicates that in gis. A clear exception to this was ownership of land, which relative terms, the impact of post-disaster assistance dimin- fared poorly, likely due to the fact that in the aftermath ished over time as livelihoods recovered. of the cyclone, land itself could bring little benefit until a return to farming was possible. Community recovery Table 32: Importance of Various Factors A similar exercise was undertaken at the village level, this for Household Recovery time examining the importance of 16 specific factors to com- munity recovery after the cyclone (Table 33). While there is 1–2 Years 9 Years some overlap between household and community factors, after after Household Factors Nargis Nargis Having economic assets survive Nargis 1 2 Table 33: Importance of Various Factors for Community Recovery Owning a large amount of land 2+ 2– Owning a small amount of land 3 3 1–2 Years 9 Years Availability of local job opportunities 1 1 after after Community Factors Nargis Nargis Ability of at least a family member 2+ 1– to migrate Level of development before Nargis 2 2 Remittances from migrated family 1– 1 Diversified village economic activities 1 1 member(s) Availability of land 2 2– Improved access to markets 2– 2– Availability of labor 2 2 Availability of affordable credit 1 2 Improved access to markets 2 2+ Entrepreneurial spirit and initiative 2+ 1– Higher migration levels 2 2+ Good management of household 1 1 Availability of affordable credit 1 2 resources Good leadership 1 1 Having a male household head 3+ 3 Social networks 1 1– Having a good number of working 1 1 High levels of membership in social 1– 2+ family members or livelihood groups Family members having good health 1 1 Mutual help freely given (household 2+ 2 Good relations within the household 2 2 to household) Good relations with extended family 2 2 Level of participation in cultural events 2 2 Having good social networks beyond 2+ 2+ Joint village infrastructure activities 1– 1– the family Good quality village infrastructure 1 1 Having good quality housing 2– 2– Livelihood aid 1 1– Availability of externally provided aid 1 2 Infrastructure aid 1 1 Note: The codes are: “1” very important, “2” of moderate importance, and Note: The codes are: “1” very important, “2” of moderate importance, and “3” less important for recovery. “3” less important for recovery. Recovery and Resilience 55 there are also key differences, particularly surrounding social relations and leadership. Again, no difference was Resilience identified with regard to the degree of affectedness. Analysis of the SIM 5 research and evaluations by the research team found that of the SIM villages, just six vil- The findings at the community level are more balanced lages were considered resilient with another eleven some- between economic and social factors than at the household what resilient. Table 34 shows that approximately one-half level, especially in the shorter term. Factors such as leader- of highly and moderately villages were resilient or some- ship and social networks are identified as extremely impor- what resilient but less than one-third of the lightly affected tant within two years after Nargis, though several social ones. The low proportion of lightly affected villages defined factors have declined somewhat in the longer term. This as resilient reflects their lower vulnerability to shocks like reflects the importance of a village’s ability to work together Cyclone Nargis and, therefore, their lack of preparedness or in the immediate aftermath of the disaster, whereas in the perceived need to prepare for disasters. longer term, as things were returning to a semblance of normal, more individualistic behaviors emerged. Nonethe- This sets the scene for an in-depth analysis of what factors less, the importance of leadership remains, a reminder of are important to helping both households and communities the essential role that local leaders continue to play not in build resilience in the expectation of a future disaster. The keeping order, but in getting villagers to work together for remainder of this chapter examines these two determinants mutual benefit. of resilience in depth, first interrogating what factors enable households and communities to recover faster before exam- Diversification of village livelihood activities is clearly iden- ining the ways that they seek to reduce the impact of future tified as a key indicator of a community’s recovery abilities disasters. in both the short and long term. This clearly fits with the findings in Chapter 2, which indicated that new livelihoods Research indicates that vulnerability of SIM households and in the villages were closely correlated with an improved eco- villages has less to do with government services than it does nomic context. High levels of migration and improved access with their own characteristics. In turn, this indicates that to markets were found to be of only moderate importance efforts to enable improved resilience must be focused as in the early post-disaster period when aid was more signifi- much on providing disadvantaged groups with a safety net cant, but increasingly important as time went on. For both, as on enabling the village or region as a whole to develop. this likely also reflects the limited connectivity and oppor- tunities (for either economic migration or improved prices) in pre-reform Myanmar and the rapid change on both fronts that has occurred over the past five years. Household resilience Table 35 shows the capacity for recovery of the households Aid is ranked not only as highly important in the shorter in the 40 SIM villages by the factors ranked as important term, but also remains a key factor in the longer term. The (i.e., ‘1’) in Table 32 above within the first two years and nine slight decline in livelihood centric aid reflects a logical years after Nargis (as well as social networks). By 2009– decline in the impact of livelihood assistance, as people’s 2010, the ability of households, regardless of affectedness to occupations were returning to an appearance of normal. recover via migration and credit (even if expensive), is clear. At the same time, the infrastructure deficit that predates At the same time, the challenges of being entrepreneurial Nargis has meant greater infrastructure needs of villages in the longer run. As with households, credit access begins as extremely important but moderates, reflecting the expan- Table 34: Resilience of Villages by Degree sion of affordable credit covered in Chapter 3. of Affectedness, 2017 In sum, we can see that in the early recovery period follow- Somewhat Not ing a disaster, social ties bringing the village together to Affectedness Resilient Resilient Resilient Total act and the ability to generate cash (whether through aid, Highly affected 0 5 7 12 diverse livelihood activities, or loans) were the key factors Moderately 4 5 9 18 for community recovery. As time went on, other market fac- affected tors began to play ever larger roles, while social ties receded Lightly affected 2 1 7 10 somewhat. Total 6 11 23 40 56 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Table 35: Capacity of Households to Recover Table 36: Capacity of Communities to Recover by Factor 1–2 Years 9 Years after after 1–2 Years 9 Years Household Factors Nargis Nargis after after Economic assets 1/ 1–2–3 2 Community Factors Nargis Nargis Local jobs opportunities 2 1 Diversified economic activities 2 1 Migration 1 1 Credit 1 1 Remittances 2 1 Leadership 2 1 Credit 1 1 Group membership 2 3 Entrepreneurial spirit 3 2 Social networks 2 2 Management of household resources 2 1 Joint infrastructure activities 2 3 Number of working family members 3 1 Good quality infrastructure 2 2 Good health 2 1 Note: The codes are: “1” enough capacity for recovery; “2“ some capacity for recovery; “3” little capacity for recovery. Social networks 2 2 Note: The codes are: “1” enough capacity for recovery; “2“ some capacity for recovery; “3” little capacity for recovery. Community resilience 1/ Indicates differences by degree of affectedness 1–2 years after Nargis Within the first two years of Nargis and across degrees of (lightly, moderately, highly, respectively). affectedness, there were few differences in communities’ capacity for recovery across the important factors identified in the aftermath of a disaster and the lack of enough work- in Table 33. As Table 36 indicates, villages tended to have ing family members to assist in rebuilding livelihoods were some capacity for recovery with all factors, and credit was evident challenges. In addition, there is a range of factors identified as the one where enough was accessible to enable where households faced significant constraints in their abil- recovery even if often at very high interest rates. ity to recover. Nine years after Nargis, communities are more diversified Unsurprisingly, the main difference across levels of affected- economically and have stronger leadership. Availability of ness is the level to which households could rely on existing more (affordable) credit also remains an important asset. At economic assets. In lightly affected villages, households had the same time, despite numerous village projects, an infra- a higher level of capacity when it came to using their existing structure deficit still exists across the panel villages. Fur- economic assets for recovery, while the villages more badly thermore, communities demonstrate lower social resilience affected by Nargis had seen most of their assets destroyed as both their capacity to undertake infrastructure activities and could not count on them to aid in recovery. Furthermore, jointly and their group membership have weakened over revisiting the data on aid at the time of SIM 327 indicates that time, an unintended outcome of increased outmigration. the aid provided tended not to be the one that, with hind- As the SIM series indicates with regard to aid, emphasis sight, helped most households in their recovery. on the “what”—diverse economic activities, good quality infrastructure—tended to overlook important “soft” factors By 2017, households continue to draw on migration (espe- of recovery, the “how”—leadership, social network support, cially to Yangon), remittances, and access to (more afford- and joint activities. The trend in Table 36 confirms, however, able) credit. In addition, local job opportunities also the importance for development assistance that aims to pro- strengthen their resilience capacity, as do better health of vide sustainable benefits to place particular emphasis on family members, a larger number of working family mem- strengthening social relations. bers, and a greater ability to manage household resources. At the same time, households still do not command a strong entrepreneurial spirit, and their access to economic assets remains more limited, which weakens their capacity to Preparedness recover. Moreover, they face an important deficit in social The team also investigated the extent to which a SIM house- networks—all important entry points for future aid. hold or community can act to reduce the effects of a disaster. The ability to engage in such harm reduction is a reflection of two broad sets of actions. First, there are the actions taken 27  Tripartite Core Group. 2010. Post-Nargis Social Impacts Moni- by households and communities to recognize the potential toring, April 2010; page 6, Figure 1. ways to reduce the impact of a potential disaster. Second, Recovery and Resilience 57 there are the specific steps that can be taken to reduce the Similar examples appear across thematic areas. Income effects of a disaster when it does occur through planning diversification via new job opportunities available in the and preparation. Furthermore, resilience can emerge as the village reduces the disaster exposure of livelihoods for both by-product of other socioeconomic decisions. slow and rapid onset challenges. Migration is an even more effective tool to reduce livelihood exposure as it largely Throughout this analysis, care must be taken to recognize removes a working adult’s income from the ebbs and flows the distinctions between slow and rapid onset disasters. of the rural economy. Identifying where a household or com- Over the previous nine years, SIM has focused on how com- munity is exposed to disaster and taking action to limit it are munities have recovered in the aftermath of a rapid onset the first components for reducing disaster impact. disaster such as Nargis. In 2017, however, while SIM villages are at a high risk of another rapid onset disaster, they also Exposure reduction can also occur at the community level face an almost certain slow onset disaster in the form of when steps are taken to reduce a village’s risk of being climate change (see box below). The early chapters of this affected by a natural disaster, such as removing trees that report have shown the effects climate change have had on might fall, improving roadways and paths so they do not Delta communities to date. This chapter includes an analysis wash away, or constructing shelters and windbreaks. of how SIM communities are working to reduce their vulner- ability to this uncertain future. Protecting from disaster effects However, it is not possible to reduce exposure for all aspects Reducing exposure to disaster of a households’ existence. Paddy fields cannot be moved One clear way to lessen the impact of a disaster is to reduce and villages are also (mostly) immobile. Physical assets how exposed one is to it and its long-term effects. For exam- such as cash, gold, harvested paddy, and food stocks are ple, the few Delta households that are able to buy and build inherently at risk because of their location, and it is difficult a second home in the township are inherently more resilient for many households to move these goods or eliminate their to the effects of a rapid onset disaster because they face exposure to a potential disaster. reduced exposure to being without shelter, thereby also reducing the risk that they might lose their lives in another Instead of exposure reduction, villagers have taken steps cyclone. to protect such assets from the effects of a disaster. For  Climate Change—Observed Effects Myanmar is one of the world’s most vulnerable countries to climate change, with a coastline stretching more than 6,000 kilometers and a majority of its population living in areas deemed ‘climate change exposed.’ The Delta is one of Myanmar’s most vulnerable regions, given its proximity to the ocean and cyclones, its low elevation, the reliance of its population on agricultural production, and its limited infrastructure. Beyond the destruction caused by Nargis, the Ayeyarwady Delta has experienced regular flooding, occasional droughts, pest infestations, and a wide variety of other weather and climate-linked challenges. While the SIM report does not seek to offer a specific climate change diagnosis for the Delta, there are a number of observed changes that are linked to climate change. Though villagers themselves do not tend to have a strong under- standing of what climate change is more broadly, they have a clear picture of how changes in weather patterns each year have become more noticeable, almost all having a detrimental effect on community livelihoods. For farmers, the monsoon rains and their schedule, intensity, and duration are perhaps the most important indicators of livelihood outcomes. Across the SIM panel, farmers reported that the monsoon in recent years has become unpredict- able in its timing, shorter in its duration, and more intense in its storms. Fighting their farmland’s flooding and drought now occurs within months of each other. These changes have had negative effects on harvests and driven farmers to leave some land fallow, change crops, and seek alternative or supplementary income sources. Fishers see different indicators but the result is the same: less predictable patterns of weather that have changed significantly in recent years. Winds from the south that accompanied the hilsa fishing season have been replaced by northerly breezes, and they have been accompanied by a decline in the size of fish that can be caught. Higher and more violent tides are common, with the fishing villages that dot the Myanmar coast and lie alongside the Delta’s major rivers reporting regular flooding—something that prior to Nargis was an annual or even more infrequent occurrence. 58 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After physical assets, this has generally meant some sort of physi- In some cases, fishers reported that they would scuttle cal barrier. For instance, building an embankment or a sluice their boats to protect them from a storm surge, refloating gate is a protection activity for both homes and paddy fields them after the storm had passed. During cyclone season, in expectation of a slow onset disaster, while burying food, fishers watch the tides carefully and are much less likely water, and important items for post-disaster consumption to fish offshore. is now a common protection step for a rapid onset disaster. Beyond livelihood-specific activities, households take a range of longer term steps to mitigate the effects of a bad Planning and preparation storm, saving in gold or replanting trees in an effort to create a windbreak. Many households consider potential evacua- Identifying where exposure can be reduced or protection is tion uses when contemplating whether to buy a car, motor- needed is just the first step toward reducing the impact of a cycle, or power boat, while others who can afford it, buy disaster. Action is then taken: both longer term planning for property in the nearest town. households and communities in the event of a disaster and immediate preparation steps when it becomes apparent that In some villages, households have used the experience of a rapid onset disaster is imminent. Nargis and other post-Nargis storms to inform their house building decisions. In communities that experienced high This analysis of planning and preparation is differentiated levels of flooding, households build new homes further off between the household and the village level. While there is the ground than previously, in some cases elevating their some overlap, households clearly operate as a more consoli- homes on stilts several meters above the ground. Other com- dated and tighter unit with more of a focus on the individual munities which experienced high winds but are situated socio-economy compared to a village’s focus on local infra- away from waterways on higher elevations, build homes structure and social ties. with a larger footprint that are closer to the ground. Across the sample villages, villagers have taken to strengthening At the household level their roof and fastening it better to the frame. Some of the steps taken by households to reduce the impact of a disaster vary by livelihood group, others vary by eco- If a rapid onset disaster is identified as imminent, house- nomic wealth, while others are consistent across villages. holds take a range of preparatory steps. Most common and Many of the activities undertaken reflect an instinctual reported in almost all moderately and heavily affected vil- assessment on the part of villagers about their own ability lages is the packing and burying of food and water (and, to effect change and take steps to mitigate the effect of a potentially, valuables and important documents) in water- disaster on their livelihood. proof plastic bags or containers. In some villages where new, higher homes have been built specifically to withstand Concerned about slow onset climate change, for example, flooding, paddy and other important goods are stored in farmers advocate and work to improve or build embank- the eaves of the roof where flooding is unlikely to reach. In ments and sluice gates and higher paddy walls to better con- particularly vulnerable villages, households might visit rela- trol water flow and mitigate flooding. Many farmers decided tives in town until the storm has passed. on using a harvester to harvest the crop faster so that it would not be standing for longer than necessary. Those who At the village level are able to do so have often taken to storing their harvested Villages take a range of steps to plan and prepare for disas- paddy in town where there is reduced flood risk. Farmers ters that include both risk reduction and prevention steps. are also now more likely to plant more paddy seedlings than While there are few formal plans and most are unwritten, previously necessary in anticipation of the need to replant. most villages and village leaders, primarily in highly and But there are few steps farmers have been able to take that moderately affected villages, reported both taking steps to would effectively reduce exposure to pests or protect their reduce risk and planning actively for the potential of a new crops from another Nargis. disaster. At the same time, fishers, while unable to effectively One of the clearest risk reduction steps in SIM villages has address slower onset problems like water quality or been the planting and/or removal of specific groups of trees increased pollution, have been able to take steps to reduce in the years following Nargis. Almost half of all villages have their own risk during rapid onset disasters. Fishers across taken steps in recent years to reduce the threat of damage the SIM panel reported being much more cautious about caused by trees during storms. Fourteen villages reported when they took their boats out, more aware of the weather that they had worked together as a village to plant more forecasts and storm potential, and more likely to take steps trees in an effort to block high winds that had damaged or to prevent their equipment from being damaged, like prop- destroyed homes in the community. Villagers also reported erly fastening nets to the riverbed with long bamboo sticks. that larger trees are some of the only things that were not Recovery and Resilience 59 washed away in the storm. A further four villages had taken “I have added U Tun Lwin on my Facebook page and always steps to cut down trees around the village, as homes had follow him.” been destroyed in previous storms when the trees were blown down. —Fisher As with the changes to house construction, this reflects communities taking the experiences of previous storms and doing their best to respond to them. However, it should Village leadership be noted that these measures have been taken within vil- Village leaders play an important role in the planning and lages; there were no indications that any communities had preparation for potential disasters. In some villages with an replanted mangroves or even viewed mangroves as useful active leader involved in disaster preparedness, s/he keeps a for flood or storm surge reduction. While in a small number list of vulnerable people in the village, including the elderly, of SIM villages, NGOs have tried to emphasize mangrove children, and disabled. These same leaders often determine restoration via cash-for-work or nursery garden programs, food contributed by the village that is kept at a central com- communities have not shown an interest in pursuing these munity building (often the monastery), make arrangements independent of externally driven activities. for the wealthier villagers to contribute to local prepared- ness activities, and make advance plans for villagers who Most villages now have designated shelter areas, reflecting need shelter in town during a storm. The VTA is also respon- a focus on community infrastructure improvement despite sible for informing the township administration when a the continued absence of formal cyclone shelters. Nine flood or other disaster occurs in the tract and for passing years after Nargis, very few SIM villages have purpose-built information, including on the weather, from the township to cyclone shelters, and they are not high on villagers’ priority other villagers, generally by use of a loudspeaker. list for self-help given the wide variety of other needs and high costs. However, villages do tend to have a designated Local collective action or activities by NGOs to encourage structure(s) where the community plans to take shelter dur- disaster risk reduction (DRR) has been relatively stagnant ing bad storms. The most common shelters were monaster- over the past four years, with DRR committees now either ies but other structures, including community centers and absent or inactive in most villages. There are DRR commit- schools, are also used for this purpose. These designated tees in only four villages, with several other villagers report- areas have been the focus of ongoing collaborate village ing their DRR committees had gone defunct in recent years. efforts to maintain and improve them, to ensure they are By contrast, 17 DRR committees were reported in SIM 2. (In secure and effective shelters when needed. In some of the SIM 1, no villages had DRR committees but 27 had formed SIM panel’s most vulnerable villages, a collective evacuation emergency committees to manage post-Nargis relief activi- plan to go to town was created. Villagers often also came ties.) The most common reason for such committees to together to build stronger jetties that could withstand storm cease activities was that it has been over nine years since surges. Nargis and no bad storm had come so there was considered to be less need. Migration has also played a role: DRR com- Communities are more focused on awareness as a key link mittees were often run by the younger generation, many of between planning for possible disasters and preparing for whom have since left the village. oncoming storms or other shocks. Radio, and more recently satellite television, are now used to get up-to-date weather “There have been no big events, shocks or disasters after forecasts. For fishers, especially, these bulletins determine Nargis. DRR committees have gradually disappeared.” their livelihood activities, as they will no longer take their boats out if a storm is expected. —Member of a DRR committee The ubiquity in 2017 of mobile devices and 3G has meant For those DRR committees that remain, their normal activi- that information delivered via social media has rapidly ties are to provide demonstration exercises every six months become one of the most important tools for village pre- of what to do in the event of a disaster, spread information paredness. Nationally known meteorologist U Tun Lwin is about expected storms and cyclones to everyone in the vil- a favorite and his Facebook page is closely followed across lage, and ensure all households (especially the vulnerable Myanmar. His name was mentioned several times by respon- ones) know what to do in the event of a storm—what to dents in different parts of the Delta. store, and where to go. They also play a coordinating role between the local fire brigade, youth, and Red Cross com- “We look at the Facebook page of U Tun Lwin if we want to mittees (where these exist) so that everyone knows their know news about the weather.” responsibility in a storm. In one village, a particularly active DRR committee has a system of flags based on different —Member of a disaster risk reduction committee storm severities that are raised so that all in the village can see the alert level. 60 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Resilience as a by-product for example, a farmer bought a house in town to provide his children a place to live while at school, but when the village Not all steps taken by villagers and communities that reduce flooded, the house offered a place for the entire family to exposure or improve protection are taken with the specific live while repairs were being made. goal of resilience in mind. In fact, often it is some other goal that motivates people, and any improved ability to reduce While these are large-scale investments, there are other the impact of a disaster is secondary, a by-product of this examples available to more households where the by- goal. Of course, this is not a binary decision; activities lie product of a decision or economic activity is improved along a spectrum with actions undertaken by households household resilience. Examples from the SIM research purely focused on resilience (building a new home higher off include the motorcycle taxi used for evacuation, building a the ground), some fully focused on socioeconomic improve- new home out of concrete because of the increase in wood ments (buying a motorcycle to start a new taxi business but prices resulting in improved home sturdiness, and the pur- also allowing the household to rapidly evacuate if neces- chase of a satellite dish for entertainment that is now used to sary), while others emerge from mixed motivations (improv- view weather forecasts and improve information awareness. ing paddy field drainage). The by-product of resilience also exists at the community Migration to Yangon and the purchase of land and houses by level. For example, efforts to develop roads and bridges that migrants are likely the most important examples of socio- improve a village’s connectivity to town by motorcycle or economic decisions that have massive positive effects on car allow for much faster evacuation if villagers are forced households’ ability to reduce disaster impacts on the Delta. to flee a storm. Buying a home in town has a similar impact: in one case, 8 Implications for Future Aid For Future Post-Disaster Aid the delivery of aid. But aid is not unproblematic, particu- larly when it comes to distribution and what groups receive In the aftermath of Nargis, aid28 delivery saved lives and then what aid. SIM 4 highlighted the need for proper targeting sought to rebuild lives. It first provided affected communi- and the need to design aid responses that meet the needs of ties with immediate necessities of food, water, and shelter the affected groups. Four years on, the same lessons can be and then, as the region began to rebuild, with the tools to drawn, with the additional point that outside assistance can rebuild homes, communities, and livelihoods. Early aid pro- have both longer term positive and negative social effects. viders were almost entirely local, though the response grew and eventually helped lay the groundwork for international Throughout this report, it is clear that women’s role in donors and aid agencies to assist Myanmar more broadly. village social affairs has expanded. The drivers behind this change are varied and their interplay complex, but a The experience of villages across the SIM panel over the recurring theme heard by researchers from respondents nine years since Cyclone Nargis demonstrates both the was that aid programs that required women be involved in strengths of post-disaster aid and some of the challenges community-level, decision-making bodies were an impor- such assistance has in helping households and communi- tant component. Whether quotas within larger committees, ties rebuild their lives and livelihoods. On the basis of five women-only bodies, or some other model, involving women rounds of social impacts monitoring, the following consid- in decision making both gave them the tools and confidence erations have become clear. to put themselves forward and helped socialize men to the idea of formal female leadership and women’s voices being heard in social affairs, previously a rare occurrence in Myan- mar village life. Aid is not apolitical and can have lasting effects on the social relations However, there have been negative social effects as well. of a community Aid delivery worsened relations between religious groups and led to splits within Christian and Buddhist, and Mus- The experience of aid across the SIM panel was generally lim and Buddhist villages that persist to this day. Trust in a positive one, and no one village looked back in anger at the other religious group was undercut due to aid and, with that mistrust strengthened by broader forces around 28  For the purposes of SIM, ‘aid’ refers to all types of aid received, religious nationalist and ethnic identity across Myanmar, both cash and in-kind and across multiple sectors. All types of aid these divides have remained. Taking such local dynamics providers from outside villages, such as government, local and into account will be of paramount importance for any future international NGOs, United Nations agencies, private sector aid post-disaster aid effort. providers, and religious groups, are referred to in the SIM series as ‘aid providers’, as villagers themselves did not tend to distinguish among different kinds of sources of assistance. 61 62 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After The long-term trajectory of aid For Future Development and its role in livelihood development Assistance to the Delta should be considered as soon as The Ayeyarwady Delta has in many ways recovered from the practically possible most obvious effects of Nargis. Houses and much village The provision of immediate, life-sustaining aid must clearly infrastructure have been rebuilt or improved, most paddy be the primary concern of any aid response to a disaster. fields are planted each year (sometimes twice), and mar- At the same, transparency, accountability, and participa- kets and new roads bustle with increased traffic. But the tion ought to be central features of aid even at this stage. region has changed irrevocably as a result of both Nargis Then, as soon as is practical, aid providers should work and Myanmar’s broader political and economic transition. to develop a longer term strategy for rebuilding the liveli- The Delta’s economy is already beginning to move away hoods of affected populations and enabling new economic from labor intensive farming and fishing methods even as opportunities for those whose livelihoods are irrevocably village social structures work to adapt to a new reality where altered. Without a strategy, aid risks becoming irrelevant or village populations are smaller with far fewer young people misapplied as the context within which it is being offered to maintain traditions. is changing. As development actors look to develop longer term sustain- Over the lifetime of the SIM series, the core Delta liveli- able programming, the Delta’s current context is instructive hoods of farming, fishing, and agricultural labor have ebbed and provides lessons for the national government, donors, and flowed. New equipment has allowed fishing boats to go and implementation agencies alike. further afield, and better quality seed and market access improved farmers’ returns per acre. But these livelihoods face systemic threats of climate change and labor short- Climate change adaptation ages that will limit their role as both pillars of income and and mitigation employment for the Delta population. Post-disaster aid that seeks to rebuild a shattered socio-economy cannot afford to As discussed above, climate change must be recognized by ignore the trajectories of these livelihoods and must antici- all actors working in the Ayeyarwady Delta as perhaps the pate and adapt to the changing shape of Myanmar’s rural, most powerful force affecting livelihoods and social change. rice producing economies. Programs that fail to take into account the roles of unpre- dictable weather conditions, varied monsoon periods, dry season droughts, increased storm frequency, and higher sea Disasters alter the social fabric levels will fail to deliver positive and sustainable results to the communities they work with. There are a number of key of communities and may lead lessons to be learnt from the experience of SIM villages in to a permanent reordering of recent years, particularly around disaster preparedness, community life embankments, and the role of locally driven development. Aid must ensure it grapples with these broader changes Disaster preparedness rather than seeking to rebuild a society that is no longer In the years immediately following Nargis, disaster pre- there. The clearest disruption to the social fabric has been paredness was seen as an important household and com- the massive migration outflow from the Delta to Yangon and munity activity across many SIM villages, not just those that beyond that has left many villages with few young people to had been highly affected. In the tenth year after the disaster, take on either their parents’ livelihoods or carry on village the memories of Nargis have faded somewhat, and despite social life. regular flooding and the experience of smaller cyclones in the past several years, the urgency to spend significant While this is not unique to the Delta (and is not uncommon time and money preparing for disasters is gone from all but in a rapidly urbanizing world), it is a continued challenge for the villages worst affected by Nargis. Initiatives such as SIM communities, many of which already lost one genera- DRR committees have disappeared as donors have moved tion of village elders when Nargis struck. As migration con- on, and most communities would rather invest in religious tinues and village populations continue to age and shrink, buildings or paddy storage facilities than a purpose-built aid providers in the Delta will start to face slowly disappear- cyclone shelter. ing villages, an issue that will require concerted and focused attention. Implications for Future Aid 63 Preparing for a future disaster can seem to lack clear ben- The momentum that these programs have created can and efits, and programs on how to survive or withstand a storm should be used to make community-based disaster risk can seem immaterial to a farmer who is struggling to deal management (CBDRM) a standard feature of any local level with drought-induced pests. But the experiences of house- project. Disaster impacts can erode or reverse the hard-won holds across the SIM panel show that those who were pre- achievements of participatory development in improving pared; who knew to store rice, water, oil, and documents; the situation of poor communities, including the diver- who knew to evacuate from their vulnerable village; and sion of scarce developmental resources to support disaster who had access to radio warnings fared better than those responses. A CBDRM perspective can be included in opera- who did not. Continuing to ensure Delta communities are as tions in multiple ways, for instance: building awareness and prepared as possible for a future disaster must be a focus of support for disaster risk management and climate change government and donors alike by continuing to invest in the adaptation during the social preparatory stage and other DRR mechanisms that were first set up in the wake of Nargis. key junctures of the project cycle; pursuing a broader policy dialogue regarding ways to strengthen the linkages between Embankments village development planning and township level planning; In the coming years, it is a question of when rather than if and creating project implementation partnerships with the Delta will experience major flooding, and planning by NGOs/ and civil society organizations engaged in commu- development actors should account for this uncertain future. nity-based disaster risk management.29 SIM 5 research made clear that one of the most important indicators of a village’s agricultural outcomes was the pres- ence of embankments protecting at least paddy fields (if not Livelihood development the entire village) from the damaging effects of flooding and salt water intrusion. Trainings for farmers Agricultural extension services are an important tool to Well-built embankments are time- and labor-consuming to improving agricultural returns. While better access to build and beyond the financial capacities of most local self- improved inputs is an important step that many farmers in help groups, though some have tried. The need for embank- the Delta now have, their ability to take full advantage of the ments in the Delta is not a new finding—some of the existing potential increases in return is limited by a lack of access to ones were constructed decades ago by the government knowledge about alterative farming methods. with World Bank funding. A program to provide embank- ments across the Delta would serve multiple purposes, from To date, extensions services, especially trainings, across improving agricultural returns, to providing a new livelihood the SIM panel have been both infrequent and the exclusive for landless laborers, to enhancing DRR. domain of NGO providers. Expanding farmer access to infor- mation on best practices with regard to new fertilizers, soil Community driven development/ quality tests, and other extensions services is an important Evergreen-type interventions step toward sustainable agriculture development, particu- Beyond a few large-scale challenges, such as embankments, larly in a context where labor is less available and farmers the development needs of communities across the Delta are beginning to turn to pesticides and fertilizers as cost- tend to be local, and communities are best placed to identify saving alternatives. Without training, use of these chemical and design programs to address them. In recent years, the inputs can have negative effects on agricultural outcomes, expansion of the government’s Evergreen program, the Live- on the local environment, and the quality of the crop itself. lihoods and Food Security Trust Fund, and the World Bank- financed National Community Driven Development Program Justice for fishers have sought to enable local solutions to local challenges of Fishers across the SIM panel expressed frustration at what credit access and community infrastructure. they see as an unequal playing field. After the first year after Nargis, their livelihood has received less support compared While neither program is a panacea, there remains signifi- to farming,30 even as the regulatory and competitive land- cant scope for financing community-level efforts to identify scape of fishing has become harder in recent years. The and address local challenges. Many of the persistent chal- lenges identified by communities regarding local infra- structure, access to finance, and livelihood diversification 29  World Bank. 2018. Integrating Community-Based Disaster Risk could be at least partially addressed by such programs. In so Management into Myanmar’s National Community-Driven Develop- doing, they would reduce many of the vulnerabilities Delta ment Program: Guidance Note. Draft. villages continue to suffer from. 30  Tripartite Core Group. 2010. Op. cit. 64 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After regulatory context for village-level fishers is complex and including poison, explosives, and electric shock both harm appears to vary from township to township and year on year. the environment and hamper those fishers who attempt to Small fishers often suffer the negative externalities (loss of fish legally. Greater efforts by both government and devel- nets, loss of fishing territory) of larger scale fishing opera- opment actors to create a fair playing field for village-level tions. Further, illegal but commonplace fishing techniques fishers are required. 9 Post-Disaster Social Impacts in the Literature While considerable research has been undertaken on com- until the next event. The longer recovery takes, the stron- munity resilience to natural disasters more broadly, the ger the impacts of indirect effects of the original event and/ literature on the long-term social impacts of a natural disas- or new hazards and developments. The literature also dis- ter is limited.31 Various aspects have been researched, for tinguishes between recovery of individuals and recovery instance, in Peru after the 1970 Ancash earthquake, in Sri of communities and notes that vulnerability to disaster Lanka after the 2004 Indian Ocean tsunami, and in New impacts is dynamic and shaped by interconnected shocks Orleans after Hurricane Katrina in 2005. Few longitudinal and stresses. studies have been undertaken, and those followed commu- nities and households for relatively short periods of a few SIM clearly demonstrates that recovery is a dynamic process years only. As a longitudinal study over nine years in a panel that evolves over time in response to a myriad of external of disaster-affected villages the Post-Nargis Social Impacts influences subsequent to the initial disaster. Every new Monitoring is an important contribution to the field. round of SIM identified a new trajectory due to changes in the environment that could not have been foreseen at the Disagreement remains about the long-term economic and time. For instance, two years after Nargis, laborers found social consequences of natural disasters. This is not surpris- themselves in the direst predicament, yet seven years later ing since every disaster is different, affects individuals and they were overall the most recovered livelihood group. communities in different ways, and is followed by a differ- ent recovery response. Not surprisingly, therefore, many of Though difficult to distinguish in detail, it is also evident the SIM findings are reflected in the literature, while others that higher order losses, those caused by impacts that were offer new insights. This section summarizes the most salient themselves caused by the direct effects of Nargis, have cast findings of SIM in relation to key aspects of post-disaster a long shadow on the recovery of villages in the Delta. Had impact and recovery identified in the literature. Nargis not happened, villagers would have been better able to cope with subsequent natural events. The SIM findings over time confirm the original hypothesis, namely that con- Recovery as a continuous process tinued monitoring of the socioeconomic activities and social relations is critical to devising recovery (and later develop- The literature points out consistently that recovery is an ment) assistance that meets the prevailing yet changing ongoing process that unfolds over time and can continue needs of the disaster-affected population. 31  See, for instance, Martin-Breen, Patrick, and J. Marty Anderies. Deepening poverty and structural 2011. Resilience: A Literature Review. The Rockefeller Foundation; changes in the local economy Burton, Cynthia. 2014. Social Risk in Post-Disaster Contexts: Guid- ance Note. The World Bank; Noy, Ilan, and William duPont IV. 2016. The literature provides multiple examples of countries The Long-Term Consequences of Natural Disasters–A Summary of the where poor individuals, communities, and regions have been Literature. SEF Working paper 02/2016. Victoria University of Wel- confronted with a permanent decline in incomes as a result lington; and World Bank. 2018. Community Resilience and Social of exposure to a natural disaster. Population movements, Impacts of Natural Disasters: An Annotated Select Bibliography. 65 66 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After declines in income and asset prices, and apparent perma- The long-term trend that SIM found, however, indicates that nent shifts in sectors of economic activity are amongst the while social relations within and between villages are still most prevalent manifestations. good a decade after Nargis, they are less strong then they used to be. A more individualistic way has emerged which SIM findings confirm these observations on two levels. First, became quite pronounced five years after the cyclone, when even in villages that are considered recovered, there are many families could no longer afford to attend community households that had been poor or near-poor prior to Nargis social events. The increase in outmigration that followed the and have found themselves in destitution ever since. Others start of the economic transition in Myanmar has amplified became poor as a result of the cyclone. In terms of post- this trend. disaster assistance, this puts a premium on targeting, and again underlines the usefulness of SIM to track household status over time in a participatory manner. The central role of local leadership Second, there have been strong and continuous shifts in eco- Numerous studies attest to the important role that lead- nomic activity ever since Nargis across all livelihood groups, ership plays for recovery. Local leaders, whether formal and in multiple directions. Farmers would become fishers, or informal, have been shown to be more effective when fishers laborers, and laborers farmers. Nargis has also had they are able to harness a community’s social capital in the an important impact on population movements. Outmigra- recovery process, and facilitate collaborative decision mak- tion has been a critical coping mechanism throughout the ing. Leadership is easier when social capital is strong, and recovery period, later significantly amplified by Yangon’s good leaders make social capital stronger. The challenge on strong economic growth that started several years after the local leaders is two-fold. First, they need to be accountable disaster. to and earn the trust of community members to mobilize them for collective action. Second, they need to develop a strong relationship with outside actors—higher levels of gov- The importance of social capital ernment and aid providers—to gain resources for recovery. The literature assigns social capital a critical role in the SIM confirms the central place of local leadership in the recovery process, with some arguing that it serves as the core recovery process. Over time, good leadership combined with engine of recovery. Communities with strong social capital increasing outmigration have been the central determinants are seen as recovering more quickly than those with weaker of whether or not a village recovered. At the same time, SIM social ties. Social cohesion can increase after a disaster, at offers a unique window into how the political transition least temporarily, as a community works together to face process that started in 2011 has played out at the village the aftermath. level. Immediately following Nargis, the ability of formal, government-appointed leaders and informal leaders (espe- Important attributes of communities with strong social cially religious leaders and respected individuals) to work capital include high levels of trust (in community members together helped during the critical relief and early recovery as well as government officials), participation in social phase. SIM has also demonstrated that the roles of leaders, events, and the ability to mobilize collectively. Households and the types of leaders, change over time. with strong ties to family and friends also recover faster as they may receive more support in the form of, for example, Elections of local leaders, as of 2013, disempowered village shelter, transport, or tools. Moreover, the literature notes leaders in favor of village tract leaders. However, good rela- that social capital can be strengthened or weakened after tions between them facilitated recovery. Interestingly, vil- a disaster depending on how well aid interventions are lagers considered English language skills important for local designed and/or implemented. leaders, as this would enable them to engage more effec- tively with external aid providers. SIM also confirmed that SIM confirms the importance of social capital for the recov- good relations of local leaders with township leaders helped ery of panel villages, though perhaps not to the extent of mobilize additional resources for their communities. being the “core engine of recovery.” Significant coopera- tion helped those villages with strong social capital get by in the first years after Nargis. Throughout the SIM period, A gendered perspective moreover, community members have cooperated economi- cally both within and across villages. And, those villages that The literature abounds with examples of post-disaster aid suffered from poorly targeted aid, or aid targeted to a spe- not being sufficiently gender sensitive. Due to socioeco- cific social group, had lower social capital, which impacted nomic conditions, cultural beliefs, and traditional practices, recovery negatively. disaster impacts are often not distributed uniformly within Post-Disaster Social Impacts in the Literature 67 a population. Gender differences are frequently most telling systems has been identified as a key element in reducing in the aftermath of a disaster, and as in most cases, the mor- social tensions. By way of example, if the process of ben- tality rates for women are higher than those of men. More- eficiary identification is seen to be equitable and impartial, over, post-disaster relief and recovery tend to be dominated communities develop a sense of ownership over the aid by men, which time and again leads to a strong male bias received and become more engaged in the recovery and to the detriment of women’s needs and livelihood concerns. reconstruction process. In terms of the “what,” for exam- In addition, women face entrenched societal barriers to ple, an oversupply of fishing boats in Aceh after the 2004 participation: they are often poorly represented in decision tsunami put added pressure on fish stocks that had been making and in project planning, implementation, and moni- declining prior to the disaster and increased preexisting toring. As a result, women’s voices and concerns tend to be vulnerabilities. less heard, and their different and specific needs remain repeatedly unaddressed. The post-Nargis response reflects many of these findings. Aid was important for the recovery process, and on the In broad terms, SIM finds that the post-Nargis response is no whole, recovery has been stronger in villages that received exception to this general male bias. Women suffered more more aid. However, the level of aid did not correlate well heavily from Nargis, accounting for 61 percent of the people with the degree of affectedness, and many highly affected who died that night (with the number much higher in some villages received less aid than some of the less affected villages).32 Those who survived had to carry the additional ones. SIM pointed out that external aid providers often burden of lost male family members which, given rural chose to bypass formal leaders due to their association with Myanmar’s traditional distribution of intra-household roles, the State and Peace Development Council, an approach that arguably put a greater strain on them as compared to sur- contributed to a long-lasting deterioration of community viving widowers (who could also remarry more easily). The social relations in several villages. post-disaster relief and recovery aid had only limited spe- cific support for women, though clearly a range of activities Aid after Nargis was more often than not provided in a top- benefited them as much as or more than men, for instance down manner where aid providers identified the priorities in water and sanitation. The most important specifically tar- and beneficiaries. On the one hand, this led to an imbalance geted activity SIM researchers found was the establishment in the recovery process. An overemphasis on the provision of microcredit funds for women-only groups. By contrast, of fishing boats to the detriment of farming inputs in the especially external aid providers placed important empha- first year after the disaster contributed directly to the sharp sis on including women in the decision-making process. This decline of the fishing sector over time. Moreover, in the rush approach eventually contributed to a broader recognition of to provide support immediately after the cyclone, providers the role of women in leadership positions. distributed aid that was ill suited to the Ayeyarwady Delta’s climate and ecology, including with regard to shelter and farming inputs. Aid and its perils On the other hand, villagers or their leaders were frequently The literature indicates that aid reduces the impact of a not included in the identification of beneficiaries and the natural disaster. A higher amount of aid does not necessar- distribution of aid. In the best of circumstances, villagers ily reflect greater affectedness, however. Evidence from Sri mobilized to redirect the assistance according to their own Lanka, for instance, suggests that the districts that benefited identified needs—with or without the agreement of the aid the most from aid were those that were perceived as tsu- provider. In the worst of cases, poorly targeted aid facilitated nami affected (and therefore received assistance) but were corruption and weakened the community’s social capital. less heavily damaged. Further to this, disaster relief and These findings stress the importance of institutional learn- recovery efforts may sometimes generate new risks instead ing on the part of aid providers in post-disaster situations. of facilitating the conditions required to build longer term social resilience—by setting up parallel delivery mecha- Furthermore, previous SIM reports pointed to a lack of nisms in the absence of state measures or bypassing local ‘systems’ or sector-level approaches to livelihood recov- communities and community groups in decision making. ery. Instead, the focus was largely on short-term needs. In the case of fishing, for instance, SIM found that support Moreover, the “how” and “what” of aid are often seen as to small-scale fishers who had lost their boats overlooked equally, if not more important than the “how much.” Com- the vital role that middlemen, who had been directly and munity involvement in beneficiary selection and verification indirectly impacted by Nargis, were playing for the proper functioning of the local fishing sector. This also meant that recovery aid did not aim to address structural constraints and policies governing development in the Delta. For formal 32  Tripartite Core Group. 2008. Post-Nargis Joint Assessment, page 26. credit, the policy environment has become more enabling 68 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After over time, as SIM 5 attests. Conversely, for fishing licenses, Migration the situation nine years after Nargis is arguably worse than it had been before the cyclone. This confirms the importance Diversification by the poor of their income through migra- of policy-level engagement already in the immediate post- tion is a common phenomenon in post-disaster contexts. disaster period, which in turn needs to be grounded in a Research amply shows that remittances after a disaster play deep understanding of context. an important part in a household’s ability to recover. No clear pattern has been observed as to how many households migrate, whether en masse or only select household mem- Disasters, debt, and land bers. Post-disaster migration can alter gender roles in the affected communities as well, with women more involved An established finding in the literature is that demand in paid work outside the home and community than before by households for loans increases after a disaster. Poorer the event. households are under greater pressure to borrow for daily purchases, housing construction, and to procure other The importance of migration to the Delta economy after goods. As natural disasters can significantly impact house- Nargis cannot be overemphasized. Despite restrictions on hold income, a household’s ability to repay loans is dimin- migration during the time of the military regime, household ished, and household debt increases. This is compounded by members migrated to Yangon after the cyclone in search of the fact that poorer households often lack access to formal work. This migration, and the remittances it brought, helped credit and have to take out loans from informal lenders at those who stayed behind to recover. The massive expansion much higher interest rates. At the same time, households of migration over the past few years, however, has led to a that are credit constrained for want of collateral find them- quantum change that has had a fundamental impact on the selves in a direr predicament still. Oftentimes, those with local economy: higher disposable income for migrant house- land are forced to sell it, and at a lower price if the earning holds and less exposure to income variation due to other potential from land is affected by the disaster. More unequal external events. land distribution may ensue. SIM findings are consistent with all these experiences. The Disasters and vulnerability debt burden of all livelihoods groups increased significantly as a result of Nargis and a constrained aid effort. As Nargis The literature reveals that those who are already poor and broke the debt-harvest-repay cycle, the majority of house- socially vulnerable, or who have not yet recovered from a holds were unable to repay pre-Nargis debts, with many previous disaster, are at a higher risk, as they are more likely ending up in a vicious debt cycle. Until the opening of the to lack adequate resources to buffer them against hazard credit market in the Delta several years after the disaster, impacts. Furthermore, poor people are often exposed to fre- households had to pay usurious interest rates of up to quent, low-intensity events, such as the recurrent floods. In 50  percent per month. In particular, until out-migration this sense, disasters not only increase the short-term eco- opened a much-needed escape route from poverty for land- nomic and social vulnerability of the poor, but also erode less laborers, their ability to take out loans was constrained their ability to cope with future shocks. by their lack of creditworthiness. Vulnerability is seen as the ability of an exposed population In the long run, small farmers, the mainstay of the Delta to withstand a hazard and reduce its social, economic, and economy, were impacted by Nargis and subsequent events personal impacts. Location is a critical determinant of vul- the most, a fact reflected in the number of villages experi- nerability, as is the quality of the built environment, includ- encing formal and informal land transactions. As expected, ing infrastructure and housing. Communities reliant on a the price of land has depended largely on location; land that single economic sector for their livelihood are more vulner- lies outside an embankment and is frequently flooded hails able than those with more diversified economies. Addition- a lower price than land located near a highway connecting ally, inequality as well as social marginalization increase towns. Though difficult to quantify, not least since the land household and community vulnerability, including through tenure system changed significantly with the introduction limited access to information. of the Farmland Law enacted in 2012, land ownership has become more unequal in several villages by SIM 4, with a SIM suggests that Delta villages remain vulnerable to future smaller proportion of households owning their own farms. potential disasters, though to a somewhat lesser degree However, larger landowners have also lost land over the past than before Nargis. The Delta is highly exposed to the risk of decade, which has in turn reduced the disparity in farm sizes cyclones. At the same time, the infrastructure has improved within a given village. markedly over the years, especially with regard to housing, Post-Disaster Social Impacts in the Literature 69 transport, and telecommunications. The Delta economy preparatory activities, such as improving housing quality, relies more than in the past on farming, because the fish- storing documents safely, or replanting trees. Social links to ing sector suffered from long-term decline. And yet, out- towns and Yangon also provide a safety net. But these are migration has offered important opportunities for employ- not sufficient for area-wide resilience. More disconcertingly ment that lessen the exposure in terms of income. Another still, the actual and potential impacts of climate change are critical feature reducing vulnerability today is easy access poorly understood, and not yet acted upon in a comprehen- to information, in particular with more trusted and more fre- sive manner. quently accessed weather information. Community-based disaster Resilience risk management In hazards research, resilience incorporates the capacity The literature indicates that the less prepared people and to reduce or avoid losses, contains the effects of disasters, institutions are for natural disasters, the more devastating and recovers with minimal disruptions. Communities and their impact will be. As a result, in many post-disaster situ- people who are more resilient are better able to cope with ations, aid providers aim to strengthen community capacity and recover from the consequences of disasters that can- to plan and prepare for a possible future disaster. Often- not be avoided. Factors that are generally agreed to enable times, this is achieved through the establishment of disaster resilience include asset ownership, quality infrastructure, risk reduction committees. However, the international expe- community evacuation plans, local knowledge, improve- rience also suggests that the memories of many disaster- ments in communications, social cohesion and community affected communities fade with time. Consequently, unless leadership, and connectedness to social networks outside preparedness systems are linked to government, especially of the community. local government, with regular resourcing and periodic sim- ulations led or coordinated by relevant responsible agen- The literature also underlines that the resilience of a commu- cies, they often disappear. nity is not static, as communities are complex and dynamic social structures. Awareness of a community’s history and Following Nargis, NGOs undertook a significant number of the ability to learn from successes and failures of the past community-based disaster risk management activities cen- are other important ingredients of resilience. And, the larger tered around the establishment of DRR committees. Exten- the number of resilient households within the community is, sive training was offered to its members, and where feasible, the more resilient is the community. accompanying infrastructure built (especially storm shel- ters). Over time, the experience of the SIM villages mirrored Overall, SIM 5 paints a somber picture of the Delta’s disaster those of similar efforts in other countries. Since no major resilience. Only a minority of villages may be able to with- storm hit the area since Nargis, the recognition of this con- stand another major cyclone. Even lightly affected villages stant threat has waned. More social learning has taken place are in their majority not adequately prepared. Location is in highly and moderately affected villages and yet, in only the primary factor, but the missing translation of awareness about 10 percent of these villages have DRR committees sur- into (collective) action is an important other one. Social vived for a decade (and without external funding for many learning from Nargis has not taken place to the extent years). At the same time, aid providers have not linked their needed. Disaster committees and evacuation plans remain post-Nargis DRR efforts to local and national government in only a few villages, and in lightly affected villages, many structures. As a result, most of the investments in prepared- villagers do not even see the need to invest in resilience. ness have been lost. Individual households and communities do undertake some 10 Final Reflections What would village life in the Delta be like today had Nargis climate change and the dual transformation in the country not happened? It is not possible to offer an accurate answer have already had a noticeable impact. This does not mean, to this question for a variety of reasons. however, that long-term trends and patterns of Nargis’ direct and indirect effects could not be discerned. For this it • First, up until recently, data (and data quality) in Myan- is important to understand the level of development of the mar, whether quantitative or qualitative, were very Ayeyarwady Delta compared to the rest of the country. limited—and even more so from the township level ­ down. The eight townships in the Ayeyarwady Delta most affected by Nargis were no exception. Consequently, no Structural poverty reliable baseline data existed that could be used to com- pare the before and after situations. Despite being the rice bowl of Myanmar, the Ayeyarwady Delta has been one of the poorest areas of the country. • Second, while the initial sampling aimed to include four Ayeyarwady region is amongst the most populous regions/ unaffected control villages, it became apparent during states, with an estimated 6.2 million residents.34 It has the SIM 1 research that these had been struck by Nargis third highest poverty rate in Myanmar, tied with Shan state; as well.33 SIM data thus do not allow for a comparison the latest state/region level estimates from 2009/10 place between villages affected and those not affected by the poverty in the region at 45 percent, implying that nearly one cyclone. in two individuals in the region could be considered poor (Figure 6).35 Given the substantial population density of • Third, the political and economic transformations that the area, Ayeyarwady region is home to nearly one in five started in 2012 affected the entire country. As SIM 5 of Myanmar’s poor—18 percent of the poor were estimated shows, these transformations have had wide-ranging to live in this region in 2009/10. This development deficit effects on villages in the Delta—and in the rest of the is not a result of Nargis, however; available 2004/05 data country. for Ayeyarwady region indicate high rates of poverty in this area (relative to other regions and states) years before the • Fourth, climate change effects have become prevalent cyclone. and visible over time. On the one hand, these have started to overshadow the direct effects of Nargis. On the Multiple human development indicators—and in particular other hand, these effects are also not limited to villages those related to health—place Ayeyarwady region among the affected by the cyclone. bottom performers. The region is home to the second largest Consequently, even with a baseline and control villages, a counterfactual scenario could not have been established as 34  Ministry of Immigration and Population. 2015. 2014 Myanmar of SIM 4, which was undertaken in 2013 and by which time Housing and Population Census: The Union Report. Census Report Volume 2. Department of Population, The Republic of the Union of Myanmar: Nay Pyi Taw. Access restrictions at the time of SIM 1 preparation did not allow 33  35  World Bank. 2014. Myanmar: Ending Poverty and Boosting Shared preliminary field visits before identifying the sample villages. Prosperity in a Time of Transition: Systematic Country Diagnostic. 71 72 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Figure 6: Poverty Rates in Myanmar by Region/State 90 Number of poor, by state or region (millions) 80 Rakhine 2.3 70 Chin 60 0.2 Shan (S) Poverty rate (%) 0.7 50 Shan (E) Taninthayi Ayeyarwady 0.2 0.4 2.8 40 0.4 Shan (N) Bago (W) Magway Yangon 0.6 0.5 1.2 1.8 30 Mandalay Kayah Sagaing 2 20 0 Kayin 1 Bago (E) 0.3 Mon 0.6 10 0.4 0 0 2 4 6 8 10 12 14 16 18 Share (%) of national population Source: World Bank. 2014. Myanmar: Ending Poverty and Boosting Shared Prosperity in a Time of Transition: Systematic Country Diagnostic. Note: The data in this figure are based on the population weights from 2009/10. Census data of 2014 show Ayeyarwady to be second to Yangon. population of stunted children under the age of 5 and has Figure 7: Welfare Index in Ayeyarwady and Yangon the second highest infant mortality rate in the country, with Regions 87 children dying for every 1,000 live births. The poor socio- economic outcomes are partly a reflection of the economic infrastructure and geography of the region. Only 12 percent of households reported access to electricity for lighting, the second lowest rate of access in Myanmar.36 Moreover, data limitations notwithstanding, it is estimated that nearly one in four of those who work in agriculture are landless despite the region’s focus being on agriculture. The seven cyclone-affected townships in Ayeyarwady region37 fare even worse than the region’s average (Fig- ure 7).38 All of them are in the lowest category with the least >1 favorable wealth position. In other words, these townships >0 to 1 are amongst the most disadvantaged in a region that is one >–1 to 0 of the most disadvantaged in the country. Consequently, –2 to –1 <–2 36  Ministry of Immigration and Population. 2015. Op. cit. Source: World Bank. 2014. Myanmar: Ending Poverty and Boosting Shared 37  Ayeyarwady region covers seven of the eight most affected Prosperity in a Time of Transition: Systematic Country Diagnostic. townships by Nargis, the eight being in Yangon region. 38  An index of wealth was constructed using asset and demo- Note: The townships most affected by Nargis are marked in red. A higher graphic indicators from the 2014 Population and Housing Census. welfare index (darker color) signals that a township has a more favorable The index is constructed at the township level, using principle wealth position. component analysis to determine the relative weights of the indi- cators. The choice of indicators was guided by secondary analysis of the literature on correlates of wellbeing in Myanmar. Final Reflections 73 villages hit by Nargis had to contend not just with the dev- livelihood, farming, in the Delta. The inability of many villag- astation wrought upon by the disaster but also historically ers to escape a vicious debt cycle is a direct result of Nargis, high poverty and underdevelopment compared to the rest and one that has had substantial long-term consequences of the region and country. This was a veritable challenge.39 for both household and community recovery. A within-sample comparison What if? While a counterfactual scenario cannot be established, a The wealth of SIM data over five rounds and nine years comparison of highly and lightly affected villages helps to allows some observations of what the situation in Delta vil- pinpoint some noteworthy patterns. SIM 5 analysis shows lages today would be in the absence of the cyclone. First, that there is little difference in the overall economic stand- there are those factors that would be at play regardless, but ing of highly and lightly affected villages, with regard to both the impact of which was exacerbated by Nargis: farming and fishing. The most striking economic difference is in villagers’ indebtedness, where 80 percent of highly • Nargis led to labor shortages that in turn led farmers to affected villages face severe debt problems vis-à-vis 20 per- use more machinery and fertilizer. Out-migration after cent of the lightly affected ones. Similarly, debt levels are the economic transformation began would have cre- higher for all livelihood groups in highly affected villages, ated labor shortages at any rate, but due to Nargis these and there is more borrowing against land. appeared earlier. No major differences can be found for social relations whilst • The destruction of spawning grounds and the pressure highly affected villages had a noticeably higher turnover of on livelihoods, linked to a post-disaster aid emphasis on leaders in the past two elections. In terms of housing, qual- providing fishing equipment within the first year after ity is clearly higher in lightly affected villages. At the same Nargis, led to overfishing in many rivers and streams. time, there is little difference overall for social infrastruc- Climate change would have eventually led to similar ture. A critical difference relates to connectivity, which is results, but the cyclone accelerated this trend. distinctively lower for highly affected villages, reflecting their remoteness. • The cyclone damaged embankments, mangrove forests, and other vegetation, which led to more frequent flood- The differences in debt, leadership, and connectivity con- ing and saltwater intrusion. The increasingly unpredict- tribute to a marked difference regarding recovery, with able weather pattern due to climate change would have lightly affected villages much more recovered than highly had similar effects on flooding, though these would have affected ones. Connectivity (the absence of year-round via- been lessened had the embankments and vegetation ble transport infrastructure in particular) is clearly an aspect remained in place. of the area’s development deficit. It also complicated the aid effort and slowed recovery. • With regard to social relations, the post-Nargis aid effort, an indirect effect of the cyclone, worsened inter- Leadership changes in isolation are not related to Nargis religious and inter-ethnic relations. And yet, the inter- but are the result of the political transition that led to two ethnic/­religious violence that has been occurring across rounds of local-level elections. Then again, the frequency different parts of the country since 2012 would have had of changes in highly affected villages is a reflection of the an inevitable impact on relations in the Delta as well. difficulty of leadership in times of a massive shock to vil- lage social and economic life and the need to catch up on • Post-disaster aid had an important effect on altering the development more broadly. gender balance in local-level decision making—another indirect effect of Nargis. While later development aid Debt in itself is also not related to Nargis, as it has been applied a similar approach to enhance women’s roles, the lubricant of the local economy for generations. However, these efforts were arguably facilitated by the earlier the cyclone had an immediate impact in that it broke the experience with aid providers. debt-harvest-repay cycle that had been sustaining the main • Furthermore, many formal (and informal) leaders died during Nargis, requiring leadership changes in numerous This dilemma is also revealed in the fact that 25 villages are con- 39  sidered recovered, but only eight are in good economic standing, villages. However, new leaders would have emerged in implying that the pre-Nargis standing to which they recovered had any event due to the political transition and accompany- not been good. ing local elections. 74 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Without Nargis . . . engagement with the international community in various • households would have had more able-bodied family ways.40 members to take greater advantage of evolving economic opportunities without having suffered a dramatic loss of • Nargis helped increase the appreciation by villagers in life; high-risk areas of the importance to prepare for and pro- tect themselves from natural disasters. • environmental protection from mangroves and other vegetation against climate change impacts would be It can be surmised that the sample villages would still be stronger; poor today without Cyclone Nargis, but that they would be less in debt and have more assets at their disposal, which • fewer land transactions under debt distress would have would enable them to take fuller advantage of the opportu- taken place; nities the country’s economic and political transformations are offering. In other words, though unquantifiable, Nargis • villagers would command more assets to adapt to climate left a clear and lasting mark on the economic prospects of change and be better able to invest in order to develop the villages it destroyed. Sadly, this is a common outcome of and protect their livelihoods; and any major natural disaster. And yet, the SIM series has shed unprecedented light on a recovery process from devastation • social relations would have become more individualistic that has had as many twists and turns, and as many paths, regardless as a result of increasing out-migration. as the Ayeyarwady Delta itself. At the same time . . . • The post-disaster aid helped not just to build back, but to By way of example, the World Bank’s re-engagement in 2012 40  through the first approved project in 25 years (the National Com- build back better, and thereby contribute to some reduc- munity Driven Development Project) built on the findings of three tion in the area’s development deficit. rounds of SIM and accompanying relations with civil society and development partners. SIM also paved the way for Qualitative • At a time when Myanmar was still under international Social and Economic Monitoring which, in six rounds from 2012 sanctions, post-cyclone aid (coordinated through the to 2017, expanded understanding of village social and economic Tripartite Core Group) paved the way for eventual life across Myanmar significantly (see World Bank. 2017, op. cit.). Annex 1 Social Impacts Monitoring Methodology The SIM series used time-tested social research methods. a unique window into the daily lives of villagers in an area Based on the early SIM experience, the methodology and severely devastated by a major natural disaster well beyond tools were adapted to post-disaster contexts to capture the timeframe of normal recovery assessments. In order to post-disaster social impacts across thematic areas (relief, ensure the ability to make longitudinal comparisons, the recovery, and resilience; socioeconomic impacts; social methodology implemented in SIM 5 copied that utilized in relations and cohesion; and local leadership and institu- previous rounds of SIM research. tions) and identify and monitor the effectiveness of recov- ery and aid efforts in the medium to longer term.41 SIM 4 affirmed the relevance post-disaster social analysis but also Focus areas pointed out some challenges to doing good social analysis in post-disaster settings, notably time (in the immediate post- SIM 1–4 provided assessments of how 40 villages affected disaster period) and local research capacity (particularly in by the cyclone were recovering six months, and one, three, countries with weak capacities). SIM 5 confirmed the valid- and five years after the storm, and what changes had ity and relevance of the approach, methodology, and tools occurred in the social and socioeconomic structures of the for post-disaster qualitative research. villages. SIM was the first time that the social impacts of a natural disaster had been assessed periodically as a core The Enlightened Myanmar Research Foundation conducted part of a post-disaster needs assessment and formal moni- the field research and analysis for SIM 5, with technical toring system. By SIM 4, the focus of the research had come support from the World Bank and Andaman Research & to include the socioeconomic impacts of the disaster and aid Advisory. The field researchers were all from Myanmar civil response, local social relations and institutions, and the sta- society with extensive experience working in remote vil- tus of village infrastructure, much of which required repair lages and conducting social assessments. The social impacts and reconstruction after Nargis. monitoring team represented a mix of Myanmar’s ethnic and religious groups. Women formed over half the team and SIM 5 included the focus areas of the previous rounds of three of the four team leaders. Research was conducted in research while adding new focus areas seeking to capture the local languages. long-term or permanent effects from Nargis and subsequent events and how villages have responded to the changed While findings from SIM are representative only of the vil- context. These new focus areas are recovery and resilience. lages where fieldwork was conducted, triangulation with Recovery examines the extent of villages’ cross-sector recov- other data sources suggests that many findings may apply ery from Nargis’ effects, assessing the importance of certain more broadly across Nargis-affected areas. The particular factors in both the near term and longer term recovery. It value added by SIM 5, however, is the longitudinal compari- also looked at how the differing vulnerabilities across the son over nine years across a panel of 40 villages, offering panel (as indicated by subsequent disasters) has affected households’ and communities’ ability to recover. World Bank and Global Facility for Disaster Reduction and 41  The resilience focus explored how households and commu- Recovery. 2015. Analyzing the Social Impacts of Disasters; Vol- nities adapt to the dual contexts of a post-natural disaster ume I: Methodology, Volume II: Tools. 75 76 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After reality and the effects of climate change that have increased assets they had lost to the disaster than the number of lives the frequency of weather shocks and other associated chal- lost in the community. Consequently, SIM 4 used the fol- lenges. It examined both issues of preparedness, i.e., the lowing definitions, as assessed by villagers and researchers: extent to which communities in the Delta take explicit actions in expectation of another cyclone or a drought, and • Ten lightly/not affected villages: the level of damage to how the socio-economy of these communities has changed housing and infrastructure was very low, or there was no in efforts to adapt to a post-Nargis context. damage Together these two analytical focus areas build a picture • Eighteen moderately affected villages: almost every of how communities act in an inherently vulnerable region house was slightly damaged but only a few were washed in the years after a massive disaster, while also facing an away; schools, roads, and bridges were damaged but to increased frequency of climate-related shocks. This picture a milder degree in turn seeks to help national and international policy mak- ers and aid practitioners identify what is most important to • Twelve highly affected villages: almost every house and help communities prepare for shocks, recover from them, the assets they contained were washed away, and impor- and build the capacity to lessen the impact of such shocks tant community infrastructure were destroyed. in the future. SIM 5 has used the same definition for affectedness devel- oped for the SIM 4 research. The village panel The 40 villages studied in SIM 5 were the same as in Overview of research topics SIM 1–4 in order to allow changing conditions to be tracked over time. In the months after Nargis, panel villages located and questions in the eight most disaster-affected townships were selected By and large, SIM 5 focused on the same topic areas as to ensure: (a) wide geographic coverage to take account of SIM  1–4: socioeconomic situation, social relations, lead- the level of remoteness; (b) variation in predominant village ership, and infrastructure and housing. Villagers, however, livelihood types (in rough proportion to the importance of were also asked to reflect on the nine years since Nargis those livelihoods in the Delta economy); and (c) variation in and on how their household and their village had recovered, the degree of initial affectedness by Nargis.42 as well as what steps they had taken (again, at both house- hold and community levels) to prepare themselves for future These criteria were hypothesized to affect recovery. Using disasters. The introduction of this new focus area, encapsu- qualitative panel data over a nine-year period from this lated within the ‘Recovery and Resilience’ analysis chapter, purposefully selected sample of villages provided unprec- sought to examine the longer term effects of Nargis on vil- edented insights into how villages recover from a major lages’ trajectories, identify where and how resilience has (or natural disaster and their trajectories once post-disaster has not) been improved, and understand lessons that could aid flows diminish. be applied in future post-disaster contexts. Initially, the degree of affectedness by Nargis was measured The key research questions for SIM 5 were: by the number of people who died during the cyclone.43 Assessing village life five years after Nargis for SIM 4, Socioeconomic conditions researchers found that the level of damage explained a village’s socioeconomic condition better than the number • Have there been any changes in the livelihood activities of deaths. In other words, the capacity of the survivors to in the village over the past four years? recover from a major disaster depended more on how many • What are the main factors that affect, implicitly or explic- itly, the current livelihood situation of villagers? 42  Full sampling procedures are outlined in the SIM 1 report: Tri- partite Core Group. 2009. Post-Nargis Social Impacts Monitoring, • What is the debt situation in the village? November 2008; page 43 ff. Four villages thought to be less- affected, control villages, were also included in the sample. These villages turned out to have been affected by Nargis, although to a • What are people’s livelihood coping strategies today? lesser extent than most other villages. One of the farming villages (not a control village) turned out not to have been directly affected Social relations by the cyclone. • How are social relations and social interaction within the 43  See Tripartite Core Group. 2009. Op. cit., page 7, footnote 7. village today, and have they changed over time? Social Impacts Monitoring Methodology 77 • How have interactions between different groups of vil- other occupations; (potentially) vulnerable groups, includ- lagers changed over the last four years? Since Nargis? ing female- headed households, the handicapped or injured, (includes male-female, rich-poor, inter-ethnic/religions and the elderly; and young men and women. To the extent (where appropriate), and young-old interactions.) possible, the researchers tried to get the perspectives on the same topics from each group in order to triangulate the • Who in the village are the most marginalized, and how information received. do they cope? • What about social relations and interaction with neigh- Fieldwork boring villages? Following an intensive week of training in early July, 2017, Leadership the research questions were tested and validated in four of • Who are the formal and informal leaders in the village the sample villages. This was followed by a debriefing and and what is their age, education, and wealth status? revision workshop to clarify any questions from the team and make any necessary adjustments to the field guide. The • Do leaders work together or do they compete with each entire team of 16 researchers was present for the training, other? the field test, and the debriefing session. • Are the leaders effective, and do they support the village? Two rounds of fieldwork were then conducted from July to September, 2017. Four teams of three researchers and • Has there been any change in leadership with the 2015 one team leader each covered the sample villages in two elections, and are new leaders more or less effective? phases. The one-week period between the phases was used Why/why not? for the team to regroup and complete documentation before embarking on the second phase. Each research team spent Village infrastructure and housing approximately three days in each village, with additional time allocated for travel. • What infrastructure exists in the village, and in which condition is it? • Was it destroyed by Nargis and rebuilt subsequently? Research instruments What is still missing? Three research instruments were used. • What is the quality of housing in the village? First, in-depth interviews were conducted with a wide Recovery and resilience range of elite and non-elite villagers. Interviews were semi-­ structured or unstructured, meaning that the researchers • Have households and the villages overall returned to the had the flexibility to focus on particular issues that the infor- pre-disaster situation in terms of social and economic mant had information on and to follow interesting lines of standing? Are they better off today than before Nargis? inquiry. Guiding questions were provided to the researchers to help focus interviews, but the researchers were free to • How have other disasters and external events since 2008 amend and adapt these as necessary, in particular, to make shaped the trajectory of the village’s recovery? sure that they fitted with the local context. A total of 467 key informant interviews were conducted during SIM 5. • How do households and communities prepare them- selves for potential future disasters? Has this changed Second, focus group discussions were held with different since Nargis? Was this change due to Nargis directly? To livelihood and identity groups within the panel villages. other more recent disasters? Depending on the village context, the following groups were invited to participate in focus group discussions: farmers, fishers, laborers, small business owners, religious or ethnic Research respondents minorities, women (particularly female head of households), members of local groups (such as women’s lending groups or Within villages, sampling protocols were the same as for DRR committees), youth, and other vulnerable groups iden- SIM 1–4. The research team sought to interview a wide tified by the research team. In each, between four and eight cross-section of the community. This included: the village informants were interviewed together. Inviting participant head and other official village leaders; village elders and ‘peers’ helped ensure the openness of the discussions. Over- religious leaders; others who were involved in aid deci- all, 132 focus group discussions were conducted. sions in the village; farmers, fishers, laborers, and those in 78 Meandering to Recovery: Post-Nargis Social Impacts Monitoring Ten Years After Across 40 villages, the SIM research engaged 1,243 partici- Second, qualitative village summary sheets were written for pants, including 454 women and 789 men. each village, and qualitative data were entered into spread- sheets to enable rapid analysis of trends. Third, the researchers also conducted informal inter- views and participant observation. This included many Finally, case studies on the experiences of particular fami- late-night discussions with those with whom they stayed, lies and groups in Nargis-affected villages were created. and discussions with individuals and groups over meals. These allowed for in-depth investigations of how Nargis Directly observing dynamics and impacts also provided and subsequent events have affected the lives of different much information on the way villages were functioning, people. Case studies focused first on people and families the interplay between the long-term effects of Nargis and whose experiences were representative of (large subsec- other disasters or exogenous effects on the village, and tions of) the general population. This allowed for an extrap- how broader changes around Myanmar were felt at the vil- olation from the experiences of particular individuals or lage level. households. Case studies also outlined the experiences of ‘exceptional’ Field data cases, those who have had particularly positive or negative experiences. In these, researchers made efforts to deter- Three types of data were utilized in the SIM 5 report. mine why the individual or household had done worse or better than others. Again, this helped determine sources First, the research team collected standardized data on vari- of variation in outcomes and experiences, between and ous dimensions of village life. These data were used to gen- within villages. A selection of the case studies is presented erate tables that allow for a mapping of broad patterns and throughout the report.44 consideration of sources of variation. Many of the indicators collected were based on those used in previous rounds of 44  Names mentioned in the case studies are not the real names of SIM. the respondents. Annex 2 Location of the Sample Villages 79 The Global Facility for Disaster Reduction and Recovery (GFDRR) is a global partnership that helps developing countries better understand and reduce their vulnerabilities to natural hazards and adapt to climate change. Working with over 400 local, national, regional, and international partners, GFDRR provides grant financing, technical assistance, training, and knowledge sharing activities to mainstream disaster and climate risk management in policies and strategies. Managed by the World Bank, GFDRR is supported by 33 countries and 11 international organizations.