Adaptive Social Protection

INTERNATIONAL DE VELOPMENT IN FOCUS

Adaptive Social Protection Building Resilience to Shocks Thomas Bowen, Carlo del Ninno, Colin Andrews, Sarah Coll-Black, Ugo Gentilini, Kelly Johnson, Yasuhiro Kawasoe, Adea Kryeziu, Barry Maher, and Asha Williams

I N T E R N AT I O N A L D E V E L O P M E N T I N F O C U S

Adaptive Social Protection Building Resilience to Shocks

THOMAS BOWEN, CARLO DEL NINNO, COLIN ANDREWS, SARAH COLL-BL ACK, UGO GENTILINI, KELLY JOHNSON, YASUHIRO KAWASOE, ADEA KRYEZIU, BARRY MAHER, ANDÂ ASHA WILLIAMS

© 2020 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433 Telephone: 202-473-1000; Internet: www.worldbank.org Some rights reserved 1 2 3 4 23 22 21 20 Books in this series are published to communicate the results of Bank research, analysis, and operational experience with the least possible delay. The extent of language editing varies from book to book. This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved.

Rights and Permissions

This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http:// creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions: Attribution—Please cite the work as follows: Bowen, Thomas, Carlo del Ninno, Colin Andrews, Sarah Coll-Black, Ugo Gentilini, Kelly Johnson, Yasuhiro Kawasoe, Adea Kryeziu, Barry Maher, and Asha Williams. 2020. Adaptive Social Protection: Building Resilience to Shocks. International Development in Focus. Washington, DC: World Bank. doi:10.1596/978-1-4648-1575-1. License: Creative Commons Attribution CC BY 3.0 IGO Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation. Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank. Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-partyowned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images. All queries on rights and licenses should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org. ISBN: 978-1-4648-1575-1 DOI: 10.1596/978-1-4648-1575-1 Cover photo: © Steve McCurry / Magnum Photos. Used with permission of Steve McCurry / Magnum Photos. Further permission required for reuse. Cover design: Debra Naylor / Naylor Design Inc.

Contents

Foreword  vii Acknowledgments  ix About the Authors    xi Abbreviations  xv

Overview: A Framework for Adaptive Social Protection   1 Introduction  1 Resilience to shocks: The capacity to prepare, cope, and adapt   3 Poverty and vulnerability: Constraints to the capacity to prepare, cope, and adapt   4 Adaptive social protection: Building resilience by supporting the capacity to prepare, cope, and adapt   6 Notes  27 References  27

CHAPTER 1 Programs: Design Considerations for Building Resilience   33 Overview  33 Focusing on the role of safety nets in building household resilience  34 Understanding the challenge of building household resilience: No single program can “do it all”   34 Appraising the core design features of existing safety net programs for resilience building   36 Design features to support the capacity to prepare, cope, and adapt   40 Notes  53 References  53

CHAPTER 2 Data and Information: Understanding Risk and Household Vulnerability  59 Overview  59 Investing in an improved understanding of risk and household vulnerability  60 Social registries at the heart of the ASP information agenda   61 Data quality considerations for ASP   65 Filling the gaps: Linking ASP to early warning information and post-shock needs assessments    69 Considerations for institutionalizing information for ASP   73 Note  76 References  76  iii

iv | Adaptive Social Protection

CHAPTER 3 Finance: Applying a Disaster Risk Financing Approach   79 Overview  79 Disaster risk financing: From reactive to proactive resource mobilization  80 Why is a disaster risk financing approach important for ASP?   81 A closer examination of the disaster risk financing approach for ASP  82 Note  93 References  93

CHAPTER 4 Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages   95 Overview  95 Multisectoral coordination for ASP   96 Creating strategic partnerships across the government-humanitarian divide   103 Notes  110 References  110

APPENDIX A Resilience as It Relates to Adaptive Social Protection   113 APPENDIX B Considerations for Shock Response along the Social Protection Delivery Chain   127 Boxes 1.1 1.2 1.3 1.4 1.5 2.1 2.2 2.3 2.4 2.5 2.6 2.7 2.8 3.1 3.2 4.1 4.2 4.3 4.4 A.1 A.2 A.3

The United States: Role of the social protection system in insuring against disasters  35 Niger: Geographic targeting based on spatial vulnerability to climate shocks  38 Ethiopia, Mexico, and Pakistan: Contingency planning   47 Sahel: Productive inclusion to support the adaptive capacity of the extreme poor  51 India: Rural employment guarantee scheme creates assets to manage climate risk   52 Analyzing socioeconomic resilience and the potential impact of ASP   61 Lesotho, Mozambique, Pakistan, and the Philippines: Social registry coverage and utility for shock response   64 Pakistan: National Database and Registration Authority   67 Big data and technology leapfrogging: Power and limitations   68 Dominican Republic: Leveraging the social registry for early warning information  70 Uganda: Establishing satellite-based triggers for drought response   71 Chile: A postdisaster data collection tool to assess shock-impacted households  72 Jamaica: Household Disaster Impact and Needs Assessment   73 Mexico: The Natural Disasters Fund   88 Contingent credit instrument: Catastrophe Deferred Drawdown Option   89 Pacific Island countries: Disaster risk management institutions can help to drive the ASP agenda    97 The Philippines: Social protection embedded in disaster risk management frameworks  98 Ethiopia: Framework of a national, shock-responsive safety net   101 Cash transfers: A vehicle for ASP development across the government–humanitarian divide  107 Unbreakable: The poor suffer disproportionately from natural hazards   114 Resilience: A brief sample of prominent definitions in the literature and among development and humanitarian institutions   117 Social risk management: Vulnerability and resilience to poverty   118

Contents | v

A.4 A.5

Welfare mean versus welfare variance: Conceptualizing resilience to poverty  119 Household resilience “capacities” in the literature   120

Figures O.1

Natural disasters: Increasingly frequent and devastating impacts, 1967–2017  2 O.2 Africa: Chronic and transient poverty   5 O.3 Social safety nets: Global coverage compared to World Risk Index ranking  8 O.4 Framework for adaptive social protection: Four building blocks   9 O.5 Africa: Safety net beneficiaries tend to use the transfers to save   12 O.6 Social protection programs: Vertical and horizontal expansion   14 O.7 Social protection delivery chain   16 O.8 Lesotho, Mozambique, Pakistan, and the Philippines: Social registry coverage and utility for shock response   18 O.9 Kenya: Modeling the cost of responding to drought   22 O.10 Risk layering: Financial instruments, by frequency and severity of a shock   22 O.11 ASP delivery approaches: A mix across national and humanitarian actors  26 1.1 Africa: Safety net beneficiaries tend to use the transfers to save   42 1.2 Ethiopia: PSNP beneficiary and nonbeneficiary recovery trajectories   43 1.3 The Philippines: Multiple programs for differing household needs across differing post-shock time periods   46 1.4 Social protection delivery chain   48 1.5 Adaptive capacity: Welfare is less sensitive to shocks for beneficiary households receiving productive grants   49 1.6 Productive inclusion model: Graduation into sustainable livelihoods   50 2.1 Social registry as a gateway for multiple programs   62 2.2 Social registries: Global coverage and registration processes   63 B2.2.1 Coverage of selected social protection databases in four countries    64 B2.5.1 Hurricane simulation using IVACC to identify households that may be most affected   70 ­3.1 Humanitarian response: Trend in requirements and funding gaps, 2007–17  80 ­3.2 Kenya: Modeling the cost of responding to drought   85 ­3.3 Disaster risk financing: Visualizing risk layering   87 ­3.4 Ethiopia: PSNP integrates ex post humanitarian assistance within a risk financing strategy  90 ­4.1 Japan: History of postdisaster DRM reforms and ASP legislation   99 ­4.2 United Nations cluster system: Overview   105 ­4.3 Framework: Mix of delivery approaches across government and humanitarian actors  108 A.1 Africa: Chronic and transient poverty   115 BA.4.1 Vulnerability to poverty characterized by the mean and variance of welfare   119 B.1 Social protection delivery chain   128 B.2 Social registry intake and registration   129

Maps O.1 A.1

Dominican Republic: The Vulnerability to Climate Hazards Index   20 The Philippines: Vulnerability to poverty because of typhoons   115

Tables O.1 O.2 ­O.3

The social protection system: Objectives and types of social protection and labor programs  6 Adaptive social protection: Supporting the capacity to prepare, cope, and adapt    7 Summary of the key priorities and investments, by building block   10

vi | Adaptive Social Protection

O.4

Risk layering: Advantages and disadvantages of individual financial instruments  23 O.5 International humanitarian system: Features and implications for adaptive social protection  26 ­1.1 Adaptive social protection: Supporting the capacity to prepare, cope, and adapt  41 2.1 Beneficiary and social registries: Terminology   62 2.2 Implications of the different dimensions of data quality for adaptive social protection  66 2.3 Early warning systems: Triggers for rapid action   71 2.4 The maturity of ASP information systems and the roles of different actors in strengthening them  74 ­3.1 Uganda: A framework to scale up the NUSAF cash-for-work program   84 ­3.2 Kenya: A framework to scale up the Hunger Safety Net Program   85 ­3.3 Disaster risk financing: A comparison of instruments   87 ­3.4 Ethiopia, Kenya, Mexico, the Philippines, and Uganda: Financial layering for adaptive social protection programs   91 ­4.1 International humanitarian system: Features and implications for adaptive social protection  103 ­4.2 Kenya and Ethiopia: Features that influence the use of government systems for shock response versus humanitarian actors or NGOs   109 BA.3.1 Poor and nonpoor households grouped as vulnerable or resistant to poverty  118

Foreword

At the time of finalizing this publication on Adaptive Social Protection (ASP), the world entered the midst of the COVID-19 pandemic, which has left no country unaffected by its sweeping impacts. Although the long-term trajectory of these widespread health, economic, and social impacts is uncertain, its immediate consequences have already resulted in significant losses in terms of lives and livelihoods. A period of prolonged, often extreme, hardship is being endured by many who are undergoing social distancing and experiencing reduced income and diminished consumption. This is especially true for the poorest among us, with the lowest capacity to cope. As the crisis has taken hold, policy makers have been reminded of the value of having strong social protection systems in place that are capable of reaching affected households with immediate assistance. Toward the end of April 2020, as many as 133 countries had planned, introduced, or adjusted social protection programs in response to COVID-19. At the same time, the crisis is shining a light on both the enabling and constraining factors that affect governments’ ability to leverage social protection systems to address large, covariate shocks of this sort. At the World Bank Group, we consider ASP to be a dedicated area of focus within the wider field of social protection, examining and identifying the ways in which social protection systems can be prepared and enhanced ahead of large covariate shocks like COVID-19 to build the resilience of poor and vulnerable households—before, during, and after such shocks occur. The report begins by highlighting how, when designed appropriately, social protection programs that are delivered to the poorest and most vulnerable households can have a transformative impact on their resilience to these kinds of shocks. Through the provision of transfers and services to the poorest and most vulnerable households, adaptive social protection directly supports their capacity to prepare for, cope with, and adapt to the shocks they face. Over the long term, by supporting these three capacities, ASP can provide a pathway to a more resilient state for households that may otherwise lack the resources to move out of chronically vulnerable situations. Further, the organizing framework for ASP that is articulated in this report provides insights into the ways in which social protection systems can be made more capable of building household resilience. Through its four building blocks—programs, information, finance and institutional arrangements, and  vii

viii | Adaptive Social Protection

partnerships—the framework highlights both the e­ lements of existing social protection systems that are the cornerstones for building household resilience to shocks, as well as the additional priorities and core investments that will be instrumental in enhancing these outcomes and making the social protection system more prepared in advance of the next crisis. By way of some key examples, the report highlights the need to modify traditional targeting methods to factor in household vulnerability to shocks; integrate and layer programming among poor and vulnerable households in “hot-spot” areas of recurrent shocks; invest in delivery systems and contingency planning to enable the increased responsiveness of programs after a shock hits; expand coverage of social registries, with a focus on the inclusion of high-risk households; preposition risk financing to ensure funding is readily available to fund response programs in a timely manner; invest in fostering collaboration and coordination with nontraditional but essential partners across government—including those involved disaster risk management and climate change ­adaptation—as well as nongovernment, humanitarian actors. These are only a few of the priorities within the four building blocks that are outlined in the report. As the COVID-19 pandemic eventually begins to recede, other shocks and crises will remain on the horizon, many of which will become increasingly severe under the influence of climate change. The framework in this report can provide directions along the path toward the development of ASP in advance of those shocks materializing in the future. Indeed, the World Bank Group is increasingly working with governments to develop ASP in some of the poorest and highest-risk countries around the world. The report provides the basis for a structured approach to implementing these engagements, each of which will, in turn, continue to inform our collective learning on this evolving and important agenda. Michal Rutkowski Global Director Social Protection and Jobs Global Practice The World Bank Group

Acknowledgments

This report is the product of a sustained conversation among colleagues within the World Bank Group’s Social Protection and Jobs Global Practice over the past several years. In particular, many of the concepts, as well as the framework outlined here, were developed for the South-South Learning Forum 2018, “Building Resilience through Adaptive Social Protection.” The work conducted as part of the Sahel Adaptive Social Protection Program has also been key in driving many of the ideas outlined in this report. As such, the report has benefited enormously from the myriad views, insights, and experiences related to the evolving concept of adaptive social protection that have been shared by these many direct or indirect contributors, not all of whom can be captured in this acknowledgments section. The report was prepared by a team that was led by Thomas Bowen (social protection specialist) and included coauthors Carlo del Ninno (lead economist), Colin Andrews (program manager), Sarah Coll-Black (senior economist), Ugo Gentilini (senior economist), Kelly Johnson (senior social protection specialist), Yasuhiro Kawasoe ( junior professional officer), Adea Kryeziu (social protection specialist), Barry Maher (senior financial sector specialist), and Asha Williams (social protection specialist). In addition to the research completed by the authors, in several places this report draws heavily on unpublished background papers prepared by Oxford Policy Management (OPM). The OPM background paper team was led by Valentina Barca and included coauthors Sarah Bailey, Rodolfo Beazley, Andrew Kardan, Gabrielle Smith, and Ana Solórzano. Catherine Fitzgibbon, an independent consultant, also contributed to the background paper related to finance. All background paper references are cited throughout the text and are noted in the chapter reference lists. Overall guidance and quality control for this report were provided by Michal Rutkowski (global director), Margaret Grosh (senior adviser), Anush Bezhanyan (practice manager), and Jehan Arulpragasam (practice manager) of the Social Protection and Jobs Global Practice. Comments received during the peer review process from John Blomquist (lead economist), Yashodhan Ghorpade (economist), Aylin Isik-Dikmelik (senior economist), and Laura Rawlings (lead ­economist) helped to sharpen and enrich this report throughout. Cathy Ansell (financial sector specialist), Evie Calcutt (financial sector specialist),  ix

x | Adaptive Social Protection

Kenichi Chavez (senior social protection specialist), Aline Coudouel (lead ­economist), Jesse Doyle (young professional), Matthew Hobson (senior social protection s­ pecialist), Phillippe Leite (senior social protection economist), Olivier Mahul (practice manager), and Ruslan Yemtsov (lead economist) were also key ­contributors to the development of many of the concepts outlined in this report. The team would also like to thank members of the Sahel Adaptive Social Protection Program’s Partnership Council, who provided important insights and guidance related to the adaptive social protection building blocks and the ­concept of resilience that proved instrumental in helping to develop this report, namely: Laura Garn, Heidi Gilert, and Harriet McDonald (Department for International Development—DFID); Daniel Longhurst, Ralf Radermacher, and Anne-Sophie Vollmecke (Deutsche Gesellschaft für Internationale Zusammenarbeit—GIZ); and Thibault Van Langenhove (Agence française de développement—AFD). The messages contained within this report do not necessarily reflect the opinions of these Partnership Council members or the positions of their respective organizations. Lastly, the authors would like to sincerely thank Darcy Gallucio for her thorough, patient, and precise editorial assistance; Nita Congress and Andres de la Roche, who provided graphic design support to finalize the publication; as well as Janice Tuten who led the finalization of the publication as part of the World Bank’s International Development in Focus series. This report was commissioned and financed by the World Bank’s Global Facility for Disaster Reduction and Recovery (GFDRR).

About the Authors

Colin Andrews is a program manager in the Social Protection and Jobs Global Practice at the World Bank. He has more than 15 years of experience working in social protection across Africa and South Asia and at the global policy level. He leads the Partnership for Economic Inclusion, a multipartner initiative to support the scale-up of national economic inclusion programs, linking several ­sectors. Colin has managed several lending operations on safety nets and service delivery in the Africa region. He has published widely on safety net impacts, crisis response, and financing. Before joining the World Bank, he worked at the Food and Agriculture Organization of the United Nations, at the European Commission, and within international nongovernmental organizations. He holds an MA in economics from Trinity College, Dublin. Thomas Bowen is a social protection specialist in the Social Protection and Jobs Global Practice. He has worked extensively on issues related to adaptive social protection (ASP), focusing on the design and implementation of safety net programs as a means for building household resilience to disasters and climate change. Thomas has supported the development of ASP programs in the East Asia and Pacific Region, including in the Philippines, Vietnam, and the Pacific Island Countries. More recently, he has been supporting ASP engagements in West Africa, including in The Gambia, Sierra Leone, and the six Sahelian countries within the Sahel Adaptive Social Protection Program. He holds an MA in economics and international relations from the School of Advanced International Studies at The Johns Hopkins University in Washington, DC. Sarah Coll-Black is a senior economist in the Social Protection and Jobs Global Practice at the World Bank, with a focus on the social assistance and jobs in Eastern Europe and Central Asia. Previously, she worked extensively on social protection in East and West Africa, managing the design and delivery of national safety net programs, including links to disaster risk management and humanitarian response, and youth employment and productive inclusion programs. Before joining the World Bank, she worked in the Philippines with the World Health Organization Regional Office for the Western Pacific, the Canadian International

xi

xii | Adaptive Social Protection

Development Agency, and nongovernmental organizations. She holds an MPhil from the Institute of Development Studies at the University of Sussex and a BS in economics from the University of King’s College. Carlo del Ninno is a lead economist in the Social Protection and Jobs Global Practice for the Middle East and North Africa Region. Over the past 18 years, he has worked on analytical and operational issues on safety net programs covering several countries in South Asia and Sub-Saharan Africa. In the Africa Region, he worked on safety net policies and programs and managed the Sahel Adaptive Social Protection Program. Before joining the World Bank, he worked on food security policy for the International Food Policy Research Institute in Bangladesh, and on poverty analysis in several countries for the Policy Research Division of the World Bank and for Cornell University. He has published on safety nets, food policy, and food security and holds a PhD in agriculture and applied economics from the University of Minnesota. Ugo Gentilini is a global lead for social safety nets in the Social Protection and Jobs Global Practice at the World Bank. With more than 20 years of experience, he has worked across the analytics and practice of social assistance as it relates to disaster risk management, labor markets, urbanization, subsidy reforms, ­displacement, humanitarian assistance, and food security in high-, middle-, and low-income countries. His latest book, Exploring Universal Basic Income: A Guide to Navigating Concepts, Evidence, and Practices (World Bank, 2020) provides a comprehensive analysis of universal basic income. Ugo produces a newsletter that reaches more than 10,000 practitioners weekly (ugogentilini.net). He holds a PhD in economics. Kelly Johnson is a senior social protection specialist in the Office of the Managing Director at the World Bank. Previously, she worked in the Social Protection and Jobs Global Practice in the Africa Region on Eswatini, Ethiopia, and Lesotho and in the South Asia Region on Afghanistan, Pakistan, and Sri Lanka. She joined the World Bank in 2010 in the Fragile and Conflict-Affected Countries Group. She specializes in emergency response, cash transfers, safety nets, and institutional reform. Kelly holds an MSc from the London School of Economics and a BA with distinction from Queen’s University in Canada. Yasuhiro Kawasoe is a junior professional officer in the Social Protection and Jobs Global Practice at the World Bank. His interests encompass the analytics and practice of social protection and disaster risk management, with a focus on the role of cash transfer programs in building household resilience to disasters. Yasuhiro holds an MA in architecture from the Graduate School of Creative Science and Engineering at Waseda University in Japan. Adea Kryeziu is a social protection specialist in the Social Protection and Jobs Global Practice at the World Bank. She has published extensively on social protection matters, including on linkages with climate change, disaster risk management, and energy subsidy reform. Her analytical background has been complemented by operational experience in East Asia, the Middle East and North Africa, and West Africa. She serves as co-chair and task team leader of the Intra-Agency Social Protection Assessments Partnership, playing a key role in managing relationships with donors and partner organizations. Before joining

About the Authors  | xiii

the World Bank, Adea served as a sustainable development fellow at the Aspen Institute and a researcher in the Ministry of Economic Development in her native Kosovo. Adea holds an MA in international economics and energy, resources, and environmental policy from The Johns Hopkins University and a BA in economics from John Cabot University in Rome. Barry Maher works in the Finance, Competitiveness, and Innovation Global Practice at the World Bank, based in Pretoria, South Africa. He leads the financial resilience policy dialogue in several African countries and coordinates the work on crisis and disaster risk financing in the Africa Region. Barry is a qualified actuary with experience in the non-life insurance sector, the carbon and renewable energy markets, financial inclusion, social protection, and disaster risk financing. He has worked for a Lloyds of London reinsurance syndicate, as a chief actuary in an insurance agency, and with the United Nations to help drive financial inclusion in the Pacific. At the World Bank, Barry has led the work on developing public-private partnerships to support agriculture insurance in Bangladesh, India, Kenya, Rwanda, South Africa, and Uganda; he spearheaded the work on developing financing strategies to support shock-­responsive safety nets. Barry has also been working closely on the joint workplan between the African Risk Capacity and the World Bank. He holds a BA with distinction in actuarial and financial studies from the University College of Dublin and a Masters in statistics with distinction from the University of Oxford. Asha Williams is a social protection specialist in the Social Protection and Jobs Global Practice at the World Bank. Her analytic and operational work on social protection spans more than a decade and covers a range of topics, including adaptive social protection, social protection systems, social safety nets, labor market programs, youth, and people with disabilities. Before joining the World Bank, she worked at the Organization of American States, where she managed a multicountry skills training program for at-risk youth; in her home country of Trinidad and Tobago, she was a researcher with the Ministry of Social Development. Asha has authored several journal articles and reports. She is a Fulbright scholar and holds an MA in international development and graduate certificate in Latin American studies from the University of Pittsburgh, and a BS in government from University of the West Indies.

Abbreviations

ASP BRACED

adaptive social protection Building Resilience and Adaptation to Climate Extremes and Disasters DRM disaster risk management HSNP Hunger Safety Net Program IVACC Vulnerability to Climate Hazards Index (Índice de Vulnerabilidad ante Choques Climáticos) NGO nongovernmental organization PSNP Productive Safety Net Programme SASPP Sahel Adaptive Social Protection Program SP social protection UNICEF United Nations Children’s Fund All dollar amounts are US dollars unless otherwise indicated.

Overview

A FRAMEWORK FOR ADAPTIVE SOCIAL PROTECTION

INTRODUCTION Today’s global landscape is fraught with interconnected and often devastating covariate shocks such as natural disasters, economic crises, pandemics, conflicts, and forced displacement.1 In the last 50 years, natural disasters have followed an increasing trend in terms of occurrence and human devastation (figure O.1).2 Climate change is expected to exacerbate these trends and, without climate-­informed development, to push an additional 100 million people into extreme poverty by 2030 (Hallegatte et ­al. ­2016). Forced displacement also has hit record highs in recent years, with an estimated 20 persons fleeing their homes every 60 seconds and more than 64 million people being displaced worldwide in 2016 (UNHCR 2 ­ 016). In addition, the COVID-19 pandemic is providing a vivid reminder of the devastating potential impact of pandemics on the lives and livelihoods of those who are directly and indirectly a­ ffected. Adaptive social protection (ASP) is a response to widespread demand for the use of social protection as a tool to build the resilience of poor and vulnerable households to these kinds of covariate ­shocks. ASP is outlined in this report as a specific focus area within the wider field of social protection that is dedicated to identifying the ways in which social protection can be leveraged and enhanced to build household resilience to these kinds of ­shocks. In doing so, this report draws inspiration and insight from the concept of ASP promulgated by researchers at the Institute of Development Studies (for example, Arnall et ­al. 2010; Davies et ­al. ­2009, 2012). These authors first highlighted the value of integrating the often disconnected social protection, ­disaster risk management (DRM), and climate change adaptation sectors for a mutually reinforcing approach to reduce household vulnerability and build household ­resilience. In equal measure, the report draws on the proliferating literature on and operational experiences related to shock-responsive social protection, especially the Oxford Policy Management shock-responsive social protection series, ­2015–18. Finally, the report draws on and adapts the Building Resilience and Adaptation to Climate Extremes and Disasters (BRACED) framework developed by Bahadur et ­al. (2015), as the primary basis for its definition of household ­resilience.  1

2 | Adaptive Social Protection

FIGURE O.1

700

600

600

500

500

400

400

300

300

200

200

100

100

0

Number of disasters

700

0

19 6 19 7 69 19 7 19 1 7 19 3 7 19 5 77 19 7 19 9 8 19 1 8 19 3 85 19 8 19 7 8 19 9 9 19 1 9 19 3 9 19 5 9 19 7 9 20 9 0 20 1 03 20 0 20 5 0 20 7 09 20 1 20 1 1 20 3 15 20 17

Millions of people affected

Natural disasters: Increasingly frequent and devastating impacts, 1967–2017

Number of disasters Number of disasters–trend

Number of people affected Number of people affected–trend

Source: EM-DAT: The Emergency Events Database, Université catholique de Louvain (UCL)—CRED, w ­ ww.emdat​ .­be, accessed May ­2019.

In response to this growing demand for ASP, this report outlines and elaborates on a concise framework to help inform its design and i­mplementation. To do so the report first outlines a working definition of ASP that is anchored to a definition of household r­ esilience. Building from these foundational definitions, the main contribution of the report is an organizing framework for ASP that is composed of four building blocks—programs, data and information systems, finance, and institutional arrangements and ­partnerships. In developing this framework, the report highlights the specific priorities and core investments aligned to each building block that support the design and implementation of ­ASP. In this way, the report identifies several priorities and investments that are above and beyond those that are business as usual for regular social protection, generated by the unique demands of building household resilience to covariate ­shocks. This report focuses primarily on elaborating this framework in relation to natural disasters and climate ­change. Each type of covariate shock transmits its impacts to households in a different way: primarily, if not exclusively, through the labor market for economic shocks, through food insecurity for drought, and through asset loss for destructive shocks such as e­ arthquakes. This implies different policy and programmatic prescriptions to mitigate the impacts, including, for example, the timing of an intervention and the most appropriate type of ­assistance. Natural disasters lay at the intersection of those covariate shocks where more is known on the role of social protection—such as economic and financial crises—and those where lessons are only beginning to emerge—­ including pandemics such as COVID-19 and forced d ­ isplacement. The building blocks and the priorities that are outlined in this report offer a foundation for a structured approach to advance ASP globally, across each type of s­ hock.

Overview: A Framework for Adaptive Social Protection | 3

RESILIENCE TO SHOCKS: THE CAPACITY TO PREPARE, COPE, AND ADAPT To understand how ASP can build household resilience to shocks, it is important to first define ­resilience. The concept of resilience has enjoyed widespread adoption across (as well as outside of ) international development organizations and sectors, from finance to health to infrastructure, to name but a f­ ew. The concept has gained traction, significance, and influence in part because it highlights a positive capacity for a unit of analysis to manage adversity (that is, a system, a society, a community, a household, or a person; for detailed synopses, see Béné et ­al. 2012; de Weijer 2013; and appendix ­A). Definitions for resilience abound and any given two definitions are rarely the same across or even within ­institutions. For conceptual clarity, in this report resilience is defined as: The ability for a household to prepare for, cope with, and adapt to shocks in a manner that protects their well-being: ensuring that they do not fall into poverty or become trapped in poverty as a result of the ­impacts.

A household’s resilience to a shock can be thought of as the product of its capacity to prepare for, cope with, and adapt to i­ t. Drawing inspiration from the BRACED framework (Bahadur et ­al. 2015), conceptually, a more resilient household will possess three interlinked capacities that help to minimize and resist a shock’s negative i­mpacts. The higher the household’s capacity to prepare, cope, and adapt, the lesser the implied impact of the shock on well-being and the increased likelihood that the household will “bounce back faster” (Schipper and Langston 2015), recovering to pre-shock levels of w ­ ell-being. By extension, vulnerability and resilience can be simplistically seen as “two sides of the same coin” (Jorgensen and Siegel 2019), where a household is vulnerable to a shock because of a limited capacity to prepare, cope, and adapt, translating into an inability to minimize and resist the negative impacts, bouncing back slowly, if at ­all. For greater precision, taking each interlinked capacity in turn, a more ­resilient household can do the ­following. • Prepare for a shock: mitigating the impacts, informing and enabling ­coping and ­adaptation.3 First, the capacity to prepare is, to a large extent, determined by a household’s access to information on the risks it faces, enabling a better understanding of the factors that drive its own exposure and vulnerability to those risks (Bahadur et ­al. ­2015). Adequate information on risk is essential for informing the actions needed to minimize exposure and vulnerability, including through preparing to cope with the immediate impact of a shock, as well as strategies for long-term ­adaptation. At the same time, a more resilient household tends to have access to savings in the form of cash and assets to create a buffer that it can draw upon after a ­shock. Similarly, a more resilient household is typically more prepared as a result of having access to a range of private (insurance) and public (social protection) instruments to draw upon when savings are depleted and/or a shock is especially ­severe. • Cope with a shock: minimizing the immediate impact of a shock on well-­ being in the short ­term.4 The capacity to cope with a shock is highly correlated to the capacity to ­prepare. A more resilient household possesses a higher capacity to cope with the impact because it can draw upon its savings

4 | Adaptive Social Protection

and leverage private (insurance) and public (social protection) resources as appropriate to smooth consumption and lost ­income. Together, these strategies and instruments help to resist the negative impact on their well-being and enable households to bounce back to their pre-shock state as quickly as ­possible. • Adapt to a shock: reducing exposure and vulnerability over the long term, enabling a movement toward a more resilient s­ tate. With sufficient adaptive capacity, a more resilient household can make investments that reduce both its exposure and vulnerability to shocks over the longer ­term. This includes diversifying or adjusting livelihood portfolios away from sources of income that are especially vulnerable to the impacts of a shock; building a larger and more diversified asset base, including productive, financial, and human ­capital-related assets to enable these adjustments in livelihood portfolios; and/or leveraging such assets to relocate away from an area of spatially concentrated ­risk. Indeed, the ultimate expression of adaptive capacity may be the household’s ability to reduce its exposure to a shock altogether through relocation and planned migration when in situ adjustments to livelihood and assets portfolios fail and where remaining in place would lead to chronic vulnerability and even ­maladaptation.5

POVERTY AND VULNERABILITY: CONSTRAINTS TO THE CAPACITY TO PREPARE, COPE, AND ADAPT Shocks disproportionately impact poorer households, who tend to be particularly exposed to shocks and more vulnerable to their impacts (Hallegatte et ­al. ­2016). The generalized vulnerability of poorer households to shocks can be ascribed to a deficit in terms of the capacity to prepare, cope, and ­adapt. For example, it is widely documented that poorer households resort to ­“negative coping mechanisms” to smooth consumption, including by cutting ­consumption, selling productive assets, and removing children from school (Hill, Skoufias, and Maher 2 ­ 019). Poverty also can prevent the adoption of livelihood strategies and higher-risk investments in support of greater preparedness and longer-term adaptation, leading to a state of chronic vulnerability to shocks (Bahadur et ­al. ­2015). For many poorer households, the ability to bounce back to a pre-shock state of well-being is acutely limited, creating poverty traps and, at a societal level, undermining poverty reduction (UNISDR ­2015). Shocks routinely impoverish nonpoor households when their capacity to prepare, cope, and adapt is ­overwhelmed. The data and research are replete with examples of how local and national poverty rates increase substantially after severe and less severe shocks (see appendix ­A). Many households live close to the poverty line, meaning they are especially vulnerable to poverty as a consequence of even small variances in income and consumption (figure O.2). In this way, households that are vulnerable to poverty due to shocks often possess similar constraints as poor households to prepare, cope, and adapt to ­shocks. Particularly severe shocks—especially those that are rapid-onset, destructive shocks such as earthquakes and severe typhoons—can erase assets and livelihoods and impoverish even wealthier ­households. Further, within a

Overview: A Framework for Adaptive Social Protection | 5

FIGURE O.2

77.3 24.2 17.2 28.0 11.6 18.1 16.2 7.3 6.5

Congo, Dem. Rep. Madagascar Malawi Mozambique Rwanda Burkina Faso Zambia Togo Sierra Leone Average Uganda Senegal Ethiopia Tanzania Chad Ghana Eswatini Côte d´Ivoire Cameroon Nigeria Botswana Mauritania

39.0 15.1 99.4 53.5 14.0 27.4 1.3 22.7 23.3 182.2 2.3 4.1 0

20

40

60

80

Total population (millions)

Africa: Chronic and transient poverty

100

Percent Chronically poor

Downwardly mobile

Upwardly mobile

Never poor

Source: Dang and Dabalen 2017, as cited in Beegle, Coudouel, and Monsalve 2 ­ 018. Note: Poverty statistics are from the latest household survey year for each ­country. “Chronically poor” are households that were poor in both periods of the analysis; “downwardly mobile” are households that fell into poverty in the second period; “upwardly mobile” are those that were poor in the first period but not in the second; and “never poor” are households that were nonpoor in both ­periods.

household, women, children, the disabled, and the elderly are often found to be especially vulnerable to the impacts from shocks (see, for example, Holmes 2019; UNICEF ­2018). Adapting to shocks and “bouncing back better” after they hit is critical for poor and vulnerable ­households. Priority 4 of the Sendai Framework for Disaster Risk Reduction emphasizes that reconstruction after a disaster offers an opportunity to build more resilient societies (Hallegatte, Rentschler, and Walsh ­2018). The concept of “building back better,” aligned to Priority 4, highlights the necessity of not r­ e-creating the same vulnerabilities that exacerbated the impacts of the previous ­disaster. Applying the same principle in relation to household resilience, it is critical to ensure that poor and vulnerable households can “bounce back better” to a more resilient state of lower exposure and vulnerability (Frankenberger et ­al. 2012; Manyena et ­al. ­2011). Further, under the influence of climate change, and alongside societal adaptation initiatives, a household’s ability to adapt over the long term to increased uncertainty and worsening climatic conditions will become increasingly ­critical. The limited capacity of poorer households to adapt to climate change means they are likely to be among the hardest hit by the worsening impacts (Hallegatte et ­al. ­2016).

6 | Adaptive Social Protection

ADAPTIVE SOCIAL PROTECTION: BUILDING RESILIENCE BY SUPPORTING THE CAPACITY TO PREPARE, COPE, AND ADAPT ASP can help to build the resilience of poor and vulnerable households to shocks by directly investing in their capacity to prepare, cope, and a­ dapt. As such, the report defines ASP in the following way: Adaptive social protection helps to build the resilience of poor and vulnerable households by investing in their capacity to prepare for, cope with, and adapt to shocks: protecting their wellbeing and ensuring that they do not fall into poverty or become trapped in poverty as a result of the impacts.

This definition of ASP promotes government-led investment in the three resilience capacities of households who are particularly vulnerable to shocks along the pre- and post-shock continuum, through social protection programs (table O.1). Together, social safety nets, social insurance, and labor market programs constitute the social protection “system” along with the policies that guide them and the delivery systems that underpin them (ILO 2017; Robalino, Rawlings, and Walker 2012; World Bank ­2012). The pronounced ability of safety net programs in particular to build the resilience of poor and vulnerable households can be harnessed and enhanced in relation to covariate ­shocks. Unemployment insurance and social insurance programs are widely understood to be instruments that can help households to cope with the impacts of a shock, if they have access to these ­programs. That said, in many countries, the share of the formal labor force is limited, and access to unemployment insurance is highly constrained, especially among the poorest ­households. Safety nets, on the other hand, r­ outinely reach among the

TABLE O.1  The social protection system: Objectives and types of social protection and labor programs SOCIAL PROTECTION AND LABOR PROGRAM

OBJECTIVES

TYPES OF PROGRAMS

Social safety nets/ social assistance

Reduce poverty and inequality

• • • • • • •

Ensure adequate standards in the face of shocks and life changes

• Contributory old-age, survivor, and disability pensions • Sick leave • Maternity/paternity benefits • Health insurance coverage • Other types of insurance

Improve chances of employment and earnings; smooth income support during unemployment

• Active labor market programs: training, employment intermediation services, and wage subsidies • Passive labor market programs: unemployment insurance and early retirement incentives

Noncontributory

Social insurance Contributory

Labor market programs Contributory and noncontributory Source: World Bank 2 ­ 018.

Unconditional cash transfers Conditional cash transfers Social pensions Food and in-kind support School feeding programs Public works projects Fee waivers and targeted subsidies • Other interventions

Overview: A Framework for Adaptive Social Protection | 7

poorest households with various forms of noncontributory assistance—most typically unconditional or conditional cash transfers, in-kind support such as food and nonfood items, and public works programs. For these reasons, while there is ample room to explore the role of all types of social protection programs in building resilience, the focus of this report is squarely on safety net ­programs. The impact of assistance delivered to a poor or vulnerable household through a safety net can be transformative across its resilience c­ apacities. A cash transfer, for example, provides a supplementary source of income that can enable the beneficiary household to undertake preparedness measures (such as saving) and to invest in higher-risk, higher-return livelihoods, supporting ­adaptation. If a shock hits, the beneficiary household is better able to smooth consumption and to avoid negative coping s­ trategies. Moreover, after a shock, if preparedness measures are overwhelmed (for example, depleted savings), the continued provision of transfers can directly support the beneficiary household’s capacity to c­ ope. More specifically, table O.2 summarizes the impact of safety net programs across the three ­resilience capacities—prepare, cope, and adapt—highlighting the ways ASP can build the resilience of poor and vulnerable ­households. TABLE O.2  Adaptive

social protection: Supporting the capacity to prepare, cope, and adapt

PREPAREDNESS

COPING

A more resilient household

• More savings (cash, assets) to draw upon if a shock occurs • Access to public (social protection) and private (insurance) instruments if needed after a shock • Access to information on their own exposure and vulnerability to shocks (including early warning information) to inform action

• Activates coping mecha- • Capable of making long-term nisms: acting on investments to reduce exposure and information (including vulnerability over time early warning informa- • Adjustment of asset and livelihood tion), leverages savings, portfolios away from sources of risk assets, public and private and vulnerability instruments to smooth • Planned movement and migration consumption and to away from areas of spatially supplement lost income ­concentrated, chronic risk

Poor and vulnerable households

• Limited savings and assets to draw on if a shock occurs • Limited or no access to public (social protection) and private (insurance) instruments if needed should a shock occur • Limited access to information on their exposure and vulnerability (including early warning information) to inform action

• Fewer resources with which to make • In the absence of adequate savings and long-term investments in adaptation access to social through adjustments in livelihood protection and/or and asset portfolios that can lead to private insurance, • Maladaptation and chronic resort to negative vulnerability coping strategies—­ • Forced displacement and cutting consumption, unplanned migration removing children from school, distress sale of ­assets, among others

Role of safety • Increased access to safety nets among • Support to post-shock net programs in the poor and vulnerable, especially coping through supporting those identified as at-risk from shocks continued delivery preparedness, • Transfers to at-risk households before during and after a coping, and shocks occur to support savings and shock to existing adaptation asset accumulation beneficiaries among the poor • Safety nets leveraged to transmit • Shock-responsive and vulnerable information on exposure and vulneraprograms capable of households bility, enabling the increased anticipaadjusting benefit tion of shocks, and informing actions in package and temposupport of preparedness, coping, and rarily increasing the adaptation number of beneficiaries as needed based on post-shock needs Source: World Bank.

ADAPTATION

• Support to long-term adjustment of asset and livelihood portfolios, including through cash, cash plus, and productive inclusion ­interventions • Community asset-building projects through public works programs that address key drivers of ­community-level vulnerability • Support to human capital ­accumulation for intergenerational adaptation through increased opportunity

8 | Adaptive Social Protection

An overriding priority for ASP is the continued extension of access to safety net programs, especially for the households that are identified as being most vulnerable to ­shocks. Recently, safety net coverage has increased dramatically, globally (see, for example, Beegle, Coudouel, and Monsalve 2018; World Bank 2018), providing a strong platform for their use in building resilience to covariate ­shocks. However, while the rise of safety nets has been impressive and is at the heart of several increasingly ambitious social protection–related agendas (including ASP), the undercoverage of and limited access to safety net programs, particularly among the poorest households, remain widespread (ILO 2017; World Bank ­2018). Low social protection coverage of those most vulnerable to covariate shocks inevitably limits the role of social protection in building resilience (Bastagli and Holmes 2 ­ 014). Indeed, many countries at high risk of natural disasters have especially low coverage, as highlighted in figure O.3. In that sense, the development of ASP is consistent with and integral to the advancement of the universal social protection agenda: access to social protection for all in need, when they need it, including in relation to ­shocks.6 The remainder of this report highlights how specific priorities and core investments can enhance the ability of safety net programs to build household resilience to covariate ­shocks. In order to highlight these priorities and core investments, the report outlines a framework that delineates four key building blocks for the development of ASP: (1) programs, (2) data and information ­systems, (3) finance, and (4) institutional arrangements and partnerships ­(figure O.4). This report is structured around these building blocks, with each chapter dedicated to expanding on the key priorities and core investments aligned to e­ ach. In that way, chapter 1, “Programs: Design Considerations for Building Resilience,” focuses on some of the design features that can enhance the ability of safety net programs to build resilience by supporting preparedness, coping, and ­adaptation. Chapter 2, “Data and Information: Understanding FIGURE O.3

ASPIRE: Social safety net coverage, latest year (all programs)

Social safety nets: Global coverage compared to World Risk Index ranking 100 90 80 70 60 50 40 30 20 10

0

20

40

60

80

100

120

140

160

World risk index, 2016 (169 = Most at risk; 1 = Least at risk) Africa

East Asia and the Pacific

Middle East and North Africa

Europe and Central Asia South Asia

Latin America and the Caribbean Expon. (ALL)

Sources: Atlas of Social Protection (ASPIRE), h ­ ttp://datatopics.worldbank.org/aspire/; UNU-EHS ­2016.

Overview: A Framework for Adaptive Social Protection | 9

FIGURE O.4

Framework for adaptive social protection: Four building blocks

Data and information

Finance

Programs

Institutional arrangements and partnerships

Source: World Bank.

Risk and Household Vulnerability,” identifies some of the data and information requirements that underpin the design and implementation of these ­programs. Chapter 3, “Finance: Applying a Disaster Risk Financing Approach,” then focuses on outlining the role of risk financing in enabling timely response to shocks with ­A SP. Lastly, chapter 4, “Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages,” unbundles some of the multisectoral institutional arrangements and partnerships that are critical for ASP both across government line ministries as well as with nongovernment ­partners. These key priorities and core investments are summarized in table O.3 and the remainder of this overview s­ ection.

ASP building block 1: Programs As noted, investing in a stronger, more comprehensive social protection system composed of multiple programs with high coverage provides the foundation for building ­household resilience. Moreover, beyond the traditional social protection system itself, ASP highlights the need for strong coordination with the disparate programs working on building the resilience of households to s­ hocks from other sectors. Take for example the many agriculture, human development (health and education), and disaster risk reduction programs that explicitly or

10 | Adaptive Social Protection

TABLE ­O.3

Summary of the key priorities and investments, by building block

BUILDING BLOCK

Programs

Data and information

Finance

PRIORITY/INVESTMENT

DESCRIPTION

Strengthen the overall social protection system and expand coverage

A stronger social protection system with higher coverage across several programs provides more avenues for reaching poor and vulnerable households with assistance before and after shocks

Appraise and adjust the design parameters of existing programs within the system

Adjusting targeting approaches to integrate risk and household vulnerability into eligibility criteria and beneficiary selection, as well as fine-tuning benefit parameters to enhance resilience-building outcomes among those households

Design features to support preparedness

Promote increased savings and financial inclusion; disseminate risk information within at-risk communities to inform strategies and actions for household preparedness, coping, and adaptation

Design features to support coping

Invest in preparing shock-responsive, flexible programs that are backed by adequate preparedness measures and contingency plans

Design features to support adaptation

Promote more productive and resilient livelihoods including through asset and livelihood diversification; support to human capital accumulation; building resilient community assets that address sources of vulnerability in the community

Household risk and vulnerability assessments

Integrating poverty and vulnerability data with disaster risk assessments for a spatial understanding of household vulnerability to shocks

Social registries

Expanding social registry coverage within high-risk areas, enabling more frequent updating and ensuring the data contained in registries are useful in the assessment of household vulnerability to shocks

Early warning systems

Linking to early warning systems as a basis for predicting needs and promoting timely action based on predefined triggers and thresholds for action

Post-shock needs assessment

Investing in the capacity to conduct post-shock assessments, or linking to assessment from other sectors, to ensure an up-to-date understanding of household needs—especially after less predictable, destructive shocks

Data sharing platforms and protocols

Facilitating exchange of data between social protection and relevant line ministries, including DRM, as well as nongovernment partners

Cost estimation of shock response Use historical shock data to analyze the predicted cost of future responses with social protection Preplanned risk financing and risk layering for shock response

Preposition financial instruments to cover those costs, layering different instruments for different risks and ensuring timelier responses

Linking to disbursement mechanisms

Ensure that programs and their payment platforms are prepared to efficiently disburse available funds to beneficiaries once released

Secure long-term financing in support of resilience building

Financing for the expansion of long-term programs, supporting household resilience, including through preparedness and adaptation

Institutional Government leadership arrangements and partnerships Policy coherence and cross-sector collaboration

Internalizing responsibility to build the resilience of poor and vulnerable households to shocks, owning the ASP agenda and setting government objectives and strategy accordingly Especially among the core sectors of social protection, DRM, and those involved in climate change adaptation

Institutional capacity

Beyond policy coherence and coordination mechanisms: investing in the additional human, financial, and physical capacity required for ASP delivery

Strategic partnerships with nongovernment actors

Pursuing a collaborative, coordinated approach with nongovernment partners engaged in building resilience

National and nongovernment actor specificity in roles and responsibilities

Beyond simple dichotomies, identifying specific comparative advantages in design and delivery of ASP programs across humanitarian/government divide

Source: World Bank. Note: ASP = adaptive social protection; DRM = disaster risk management.

Overview: A Framework for Adaptive Social Protection | 11

implicitly seek to build household resilience to covariate s­ hocks. Similarly, after a shock hits, many emergency response and recovery programs deliver from a multitude of ministries, departments, and agencies as well as from nongovernmental and humanitarian organizations to help people cope with the i­mpacts. Where such coordination, coherence, and integration of programming is achieved in practice, household gains in resilience building could be more ­significant and sustainable; see, for example, the integrated and layered programmatic approaches to building resilience undertaken by the World Food Programme (WFP 2015, ­2018). More specifically, traditional approaches to safety net beneficiary selection need to be re-evaluated to ensure coverage of the households that are most ­vulnerable to s­ hocks. Geographic targeting that is based on a spatial understanding of risks and that prioritizes extending and/or deepening coverage within high-risk areas will enhance the ability of safety net programs to support resilience ­building. Within program eligibility criteria, measures of ­vulnerability to covariate shocks can further enhance the ability to identify and reach households most vulnerable to shocks in support of preparedness, coping, and adaptation (del Ninno and Mills ­2015). For example, climate-smart targeting incorporates area and household data to help identify the households v­ ulnerable to natural hazards and climate-change risks (ADB 2018; Bastagli and Holmes 2014; World Bank ­2013). Appraising and adjusting existing program benefit package parameters can enhance their resilience-building i­ mpact. Building on risk and vulnerability– informed beneficiary selection, an assessment of existing safety net program benefit packages can inform specific adjustments to maximize their impact on resilience ­building. For example, it is worthwhile considering how the benefit package design parameters—type (cash, vouchers, food), timing, frequency, duration, and amount—mediate their impact on preparedness, coping, and ­adaptation. Concretely, smaller, more frequent cash transfers in support of consumption smoothing are associated with support to coping, especially when they are timed with predictable shocks such as lean ­seasons. Larger, lump-sum, infrequent transfers are more likely to spur investments in support of adaptation and p ­ reparedness. More generally, where transfers are not predictable and reliable, they will undermine resilience-building impacts, with beneficiaries more likely to continue to resort to negative coping strategies and not factor the transfers into longer-term investment decisions, hampering preparedness and ­adaptation. Safety net support to the capacity to prepare for shocks

Savings and financial inclusion can directly increase the preparedness of poor and vulnerable households, enhancing their ability to cope with and adapt to a ­shock. Financial inclusion can be explicitly supported where safety net beneficiaries are given access to a store-of-value transaction account (increasingly common practice for cash transfers) and encouraged to save and/or are given access to savings groups in their ­community. Even where social protection does not explicitly support financial inclusion and saving, beneficiaries often use the transfer for this purpose, especially in contexts of recurrent c­ rises. Recent evaluations of safety net programs indicate significant impacts on increased savings, improved creditworthiness, and reduced debt (Andrews, Hsiao, and Ralston 2018; Bastagli et ­al. 2016; Hidrobo et ­al. 2018; Ulrichs and Slater 2 ­ 016). In Mexico, beneficiaries of the former national

12 | Adaptive Social Protection

conditional cash transfer program (Prospera) who lived in communities highly exposed to droughts and hurricanes largely used the transfer “to save for the bad times” (Solórzano 2 ­ 016). In Africa, safety net beneficiary households are 4–20 ­percentage points more likely to save relative to comparable nonbeneficiary households (figure O.5); given the initial low savings rate among such households, this implies an expansion by a factor of almost two in the incidence of savings (Beegle, Coudouel, and Monsalve ­2018). Financial inclusion provides the additional benefit of making beneficiaries more easily reachable with swift electronic cash transfer assistance after a shock. Safety net programs can also provide channels for communicating early warning information, disaster preparedness training, and guidance on adaptation to recipient h ­ ouseholds. Access to early warning systems is low and biased against poor households in developing countries (Hallegatte et ­al. ­2017). FIGURE O.5

Africa: Safety net beneficiaries tend to use the transfers to save a. Percentage change in savings rate from baseline 300 % change from baseline

250 200 150 100 50 0 –50 –100 Tanzania TASAF

Kenya HSNP

Zambia ZCGP

Program impact Mean impact

Kenya CTOVC

Ghana LEAP

Sierra Leone CFW

Lesotho LCGP

95% confidence interval of program impact 95% confidence interval of mean impact

b. Estimated change in savings rate 100 Savings rate (%)

80 60 40 20 0 –20 Tanzania TASAF

Kenya HSNP

Zambia ZCGP

Kenya CTOVC

Impact estimate

Ghana LEAP

Sierra Leone CFW

Lesotho LCGP

Baseline mean

Source: Beegle, Coudouel, and Monsalve 2 ­ 018. Note: The mean value of the household transfer (in 2011 US$ purchasing power parity) is Tanzania Social Action Fund $48; Kenya’s Hunger Safety Net Program $47; Zambia’s Child Grant Program $27; Kenya Cash Transfer for Orphan and Vulnerable Children $71; Ghana’s Livelihood Empowerment against Poverty $24; Sierra Leone Cash for Work $83; and Lesotho Child Grants Program $34.

Overview: A Framework for Adaptive Social Protection | 13

This deficit is notable insofar as households can only be so prepared without access to early warning ­information to anticipate the coming shocks. Safety net programs rely on a network of implementers who often reach into the poorest communities, including social workers and v­ illage/community l­eaders. Leveraging these networks and the behavioral change sessions that increasingly accompany program delivery within communities can provide the means and venues for communicating this ­information. These venues can also be utilized to disseminate information on household and community disaster risk, risk reduction, and adaptation measures to beneficiary households and the wider community that are otherwise hard to reach (ADB ­2018). For example, in the Philippines, Family Development Sessions, an integral component of the national conditional cash transfer program (Pantawid Pamilya Pilipino Program), are used as a vehicle and venue for delivering disaster preparedness information to all beneficiaries (Bowen ­2015). Safety net support to the capacity to cope with shocks

Safety net programs have well-documented, positive impacts on a poor household’s capacity to cope with shocks, supporting food security and lessening the need to resort to negative coping alternatives (Ulrichs and Slater ­2016). Of the resilience capacities, safety nets tend to demonstrate the strongest impact on supporting a household’s capacity to ­cope. Evaluations of safety net programs across six African countries describe “unambiguous” increases in the food security of beneficiary households (Asfaw and Davis 2 ­ 018).7 The receipt of transfers through Ethiopia’s Productive Safety Net Programme reduced the initial impact of a drought on beneficiaries by 57 ­percent, eliminating the adverse impact on food security within 2 years (Hidrobo et ­al. ­2018). In the context of covariate shocks, safety net programs can provide extraordinary support to help households cope with often devastating i­mpacts. In their shock-responsive social protection framework, O’Brien et ­al. (2018) outline five potential ways that social protection programs can be leveraged to respond to large-scale shocks: • Design tweaks are small adjustments to a routine social protection ­program. They can introduce flexibility to maintain the regular service for existing beneficiaries in a shock (for example, by waiving c­ onditionalities). Alternatively, they can address vulnerabilities that are likely to increase in a crisis, through adjustments to program coverage, timeliness or predictability (for example, by altering the payment schedule), without requiring a flex at the moment of the ­shock. • Vertical expansion is the temporary increase of the value or duration of a social protection intervention to meet the additional needs of existing ­beneficiaries. For such vertical expansions to be relevant, the program or programs must have good coverage of the disaster-affected area and also of the neediest ­households. • Horizontal expansion is the temporary inclusion of new beneficiaries from disaster-affected communities into a social protection program, by extending geographic coverage, enrolling more eligible households in existing areas, or altering the enrollment ­criteria. • Piggybacking occurs when an emergency response uses part of an established system or program while delivering something ­new. Exactly which and how many elements of the system or program are borrowed will vary; it could

14 | Adaptive Social Protection

be, for example, a specific program’s beneficiary list, its staff, a national database, or a particular payment ­mechanism. • Alignment describes designing an intervention with elements resembling others that already exist or are planned, but without integrating the ­two. For example, this could be an alignment of objectives, targeting method, transfer value, or delivery ­mechanism. Governments may align their systems with those of humanitarian agencies or vice versa, either because an existing intervention is not operational as needed in a crisis or because it may not yet exist (O’Brien et ­al. ­2018). Where a safety net exists and has a good degree of coverage among affected households, vertical expansion offers a relatively simple method of providing more assistance to existing beneficiaries that have been affected by a shock (­figure O.6). Recent examples include the vertical expansion of the social protection system in Fiji following Tropical Cyclone Winston in 2016 and of the national conditional cash transfer program in the Philippines, through additional grants from humanitarian actors (the World Food Programme and UNICEF) following Typhoon Haiyan (Yolanda) and Typhoon Ruby. In the case of Fiji, an impact assessment, conducted 3 months after the disaster, found that households that received the vertical expansion were more likely to report having recovered from the shock more quickly; for instance, they were 8–10 ­percent more likely to have recovered from housing damage than nonbeneficiaries (Mansur et ­al. ­2017). However, vertical expansions generally do not reach shock-affected, nonbeneficiary households that may be in equal or greater need of assistance (Barca and O’Brien 2 ­ 017). As such, the ability to at least temporarily reach additional households that may be equally or more in need of support to their coping capacity but that may not be regular beneficiaries of social protection programs is critical for shock-responsive social FIGURE O.6

Social protection programs: Vertical and horizontal expansion Benefit amount Vertical expansion

Temporarily increased benefit amount

Regular benefit(s) amount

Regular social protection system parameters

Horizontal expansion

Population Core beneficiaries of social protection system Source: World Bank.

Those not in receipt of regular benefits but affected by a shock

Overview: A Framework for Adaptive Social Protection | 15

­ rotection. This can be achieved through horizontal expansion, or dedicated p emergency p ­ rograms that may piggyback on social protection delivery systems. Horizontal expansion enables a safety net program to temporarily expand its caseload after a shock to include new households based on eligibility from a shock’s impacts (figure O.6). Introducing the ability to horizontally expand in this manner is far more complex than undertaking vertical expansion to existing ­beneficiaries. Horizontal expansion benefits from significant ex ante investment in the processes and procedures for delivering the program, often in challenging ­postdisaster ­settings. Several countries have invested in the capacity to horizontally expand a safety net program, including most prominently Ethiopia with the Productive Safety Net Programme and Kenya with the Hunger Safety Net Program; each is prepared to undertake horizontal expansions based on ­household needs generated by drought and related food ­insecurity in drought-prone parts of the country. Some countries use a dedicated emergency program with characteristics similar to a safety net (cash, in-kind, and public works), which may piggyback on core safety net delivery systems and c­ apacity. Emergency programs have dedicated response objectives and operate alongside an existing safety net ­program. Such programs can be located within or outside of the social protection ministries, departments, and agencies and can leverage underlying safety net delivery systems such as social registries, payment systems, and front-line social protection s­ taff. In Pakistan, one such emergency program, the Citizen’s Damage Compensation Program, responded to widespread flooding in 2010 (World Bank ­2013). The Citizen’s Damage Compensation Program model has since been adopted as a permanent approach to reaching those affected by ­shocks in Pakistan. In the Sahel, Mauritania has developed a dedicated response program (“Elmaouna”) that piggybacks on existing social protection social registries and payment platforms for its ­delivery. Whichever approach is taken, the timeliness of shock-responsive social protection is critical for the protection of household well-being and is a function of adequate preparedness ­measures. Specifically, contingency planning is a critical preparedness measure that enhances the timeliness of r­ esponse. Indeed, operational processes for shock response need to be clearly defined in advance—who does what, when—in relevant operational manuals, standard operating procedures, and the wider government shock response ­plans. Ultimately, such planning can better ensure faster, more effective, and more coordinated ­implementation. To a large extent, social protection programs across countries rely on common phases of delivery to ensure that programs provide the right amount/composition of benefits and services, to the right persons at the right ­time. This “delivery chain” is centered on four implementation phases: assess, enroll, provide, and manage (figure O.7). The delivery chain provides a useful schematic for considering the preparedness measures and contingency plans that are required at each phase of delivery to enable the operationalization of shock responsive social protection. These considerations for shock response ­. along the social protection delivery chain are explored further in appendix B Safety net support to the capacity to adapt to shocks

Alongside supporting short-term coping after a shock, governments can use safety nets to invest in the capacity for poor and vulnerable households to adapt to shocks over the long ­term. There has been an increasing and

16 | Adaptive Social Protection

FIGURE O.7

Social protection delivery chain Assess

Enroll

Outreach

Intake and registration

Assess needs and conditions

1

2

3

Eligibility Determine and benefits enrollment and service decisions package

4

5

Provide

Notification and onboarding

Benefits and/or services

6

7

Manage

Beneficiaries, Exit decisions, notifications, grievances, compliance case outcomes

Recurring cycle

8

9

Periodic reassessment

Source: Lindert et ­al., ­forthcoming.

justifiable focus on the role of shock-responsive social protection in supporting post-shock c­ oping. That said, ASP and the wider definition of resilience used here highlight the central importance of supporting a vulnerable household’s ­longer-term adaptation in order to reduce its vulnerability to a shock over ­time. By broadening the focus in this way, safety net programs can provide pathways toward a more resilient state for poor and vulnerable households (see also Tenzing ­2 019). By extension, where successful, these investments may serve to reduce future post-shock needs over t­ ime. Concretely, safety nets can support adaptive capacity when designed to help the poor and vulnerable households accumulate and diversify assets and livelihoods (Bahadur et ­al. 2015; FSIN 2015; Jorgensen and Siegel 2 ­ 019). The promotion of more productive and resilient livelihoods among poor and vulnerable households is one of the primary ways in which safety net ­programs can ­support adaptive ­capacity. Interventions that promote more productive and resilient livelihoods have the potential for empowering beneficiaries to diversify their asset and livelihood portfolios and to reduce their exposure and ­vulnerability to s­ hocks. For example, a study by Macours, Premand, and Vakis (2012) found that the provision of vocational training or a productive investment grant in addition to a cash transfer to beneficiaries vulnerable to drought in Nicaragua provided full protection against drought shocks 2 years after the end of the intervention (relative to a control group that only received a cash ­transfer). Similarly, safety nets can contribute to livelihood promotion through specific programs that link cash transfer recipients to complementary interventions in other sectors (for example, agricultural inputs, training, and microfinance), leading to positive—yet varied—impacts on production and diversification into on-farm and off-farm opportunities (FAO 2016; Mariotti, Ulrichs, and Harman ­2016). As such, productive inclusion programs are emerging as powerful instruments for supporting the adaptive capacity of the poorest by supporting ­transitions into more productive a­ nd resilient livelihoods. Productive ­inclusion complements and links the provision of routine transfers with other ­interventions. These other interventions include skills and micro-­ entrepreneurship training tailored to livelihood opportunities; promotion of

Overview: A Framework for Adaptive Social Protection | 17

and support for saving groups; provision of seed capital and productive grants; linkage to existing value chains and markets; and mentoring, behavior, and life skills to build confidence and reinforce existing skillsets, among others (Bossuroy and Premand 2016; PEI 2016; Roelen et ­al. ­2017). Additionally, climate-sensitive public works programs enable beneficiaries to build assets that address structural vulnerabilities within their c­ ommunity. When designed to do so, public works programs can engage communities in climate-smart agriculture and integrated natural resource management, including a focus on waste management, reforestation, rainwater harvesting, soil/water conservation, and drought-resistant horticulture, among o ­ thers.8 A series of case studies of India’s Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) found that it can help to build resilience to various climate ­shocks. The MGNREGS was found to do so by providing integrated natural resource management and soil conservation infrastructure, agriculture-based investments, and other local infrastructures (Esteves et ­al. 2013; Kaur et ­al. ­2017). Lastly, safety net programs that contribute to building human capital can equip future generations with the tools to adapt to s­ hocks. Promoting the accumulation of human capital among poorer households is critical in terms of connecting those households with the skills to adapt over the long t­ erm. Indeed, human capital can empower the next generation with the means to move out of at-risk areas toward employment opportunities in lower-risk livelihoods or ­lower-risk ­areas. To encourage the accumulation of human capital among beneficiaries, safety net benefits often come with conditions such as, most prominently, those aligned to conditional cash transfer (CCT) ­programs. CCT programs typically provide cash transfers to households when a household meets conditions related to investing in the education and health of its children. In cases where the capacity to monitor compliance with “hard” conditions in CCT programs may be lower, “soft” conditions are increasingly being used. For example, behavioral change sessions are increasingly accompanying cash transfer programs in Africa, delivered in the community to transmit information on health, nutrition, and education to ­beneficiaries.

ASP building block 2: Data and information Information on household vulnerability to shocks and their relative ­capacity to cope and recover is crucial for the design and implementation of ASP ­programs. Critical questions for ASP include: What kinds of hazards does the country face? How frequently? Where? Which assets and population groups are exposed, and among them, which are the most vulnerable? The analysis of disaster risk is a core pillar of work conducted by the DRM ­sector. As highlighted above, ASP will need to draw on these analyses and assessments of disaster risk, integrating them with assessments of household poverty and vulnerability to poverty to provide an informed, needs-based foundation for policy dialogue and program ­design. Beyond foundational analyses of risk and vulnerability, the global expansion of social registries is framing much of the current discussion around the ASP information agenda (Barca 2017; Bastagli 2014; Bastagli et ­al. 2016; IEG 2011; Kuriakose et ­al. ­2012). Social registries are information systems that

18 | Adaptive Social Protection

support outreach, intake, registration, and determination of potential eligibility for inclusion in one or more social ­programs. While many technical considerations are involved in designing and implementing social registries, their role in social policy is simple: provide a “gateway” for potential inclusion of intended populations into social programs (Leite et ­al. ­2017). Social registries have been noted as especially useful tools for estimating the effects of a disaster on a household and for providing information on social protection beneficiaries and nonbeneficiaries that can enable shock-responsive social ­protection. However, the business-as-usual expansion of social registries alone may not meet the information requirements for A ­ SP. Many countries operate registries with “fixed lists” of registrants and program beneficiaries, and they generally update the lists every 4–5 ­years. Thus, social registries often comprise dated information and partial population coverage. For example, in Ecuador, only 15 ­percent of households in the database of affected households collected after the 2016 earthquake, Registro de Damnificados, were linked to the country’s flagship social assistance program, Bono de Desarrollo Humano (Beazley ­2017). Figure O.8 compares the coverage of social registries in four countries to demonstrate the varying population shares that are more or less easily identified and reached with post-shock ­assistance. Even with a complete social registry, existing information may not be fully up to date or

FIGURE O.8

Lesotho, Mozambique, Pakistan, and the Philippines: Social registry coverage and utility for shock response a. Lesotho

22%

b. Mozambique

CGP beneficiaries

NISSA social registry

5%

100%

b.

Beneficiaries across all program

Lesotho population

a. c.

NSER social registry

d. The Philippines

100%

20%

Pakistan population

BISP beneficiaries a. b.

Social registry data

a.

Mozambique population

c.

c. Pakistan

85%

100%

8%

75%

Listahanan social registry

c.

Beneficiary data for noncontributory social protection program(s)

100%

21%

Pantawid beneficiaries

a. b.

Philippines population

c.

Households potentially affected by shock

Source: Barca and O’Brien 2 ­ 017. Note: BISP = Benazir Income Support Program; CGP = Child Grants Programme; NISSA = National Information System for Social Assistance; NSER = National Socioeconomic Registry. Figures do not represent the totality of social protection databases in each c­ ountry. The original source material also referred to Listahanan’s coverage in the Philippines as 60 ­percent, which has since expanded to the 75 ­percent pictured ­here. a = households that can be reached through vertical expansion or piggybacking (on the beneficiary databases); b = households that can be easily reached through horizontal expansion or piggybacking (on the social registry); c = households less easily reached through horizontal expansion or piggybacking (not covered by existing social protection databases).

Overview: A Framework for Adaptive Social Protection | 19

accessible or may not be fully complete to reflect the multidimensional data requirements to inform a response after a shock (Barca and O’Brien 2 ­ 017). With that said, social registries can enhance their relevance for ASP by expanding into and within high-risk areas, updating information in those areas more frequently, and including variables related to household ­vulnerability. The prioritization of identified hotspot areas, in tandem with the expansion of programming to those areas, can help to increase the relevance of the social registry for ­ASP. In Mauritania, the government developed a methodology to determine the ideal number of households in the social registry in each commune to ensure it was more capable of informing response to ­drought. The analysis recommended including an additional 50,000 households that were expected to be food insecure. Additionally, the social registry data collected on households in those high-risk areas can be adapted to include key indicators related to their livelihoods and vulnerability to the ­hazards they ­face. In the Dominican Republic, the Vulnerability to Climate Hazards Index (Índice de Vulnerabilidad ante Choques Climáticos; IVACC) quantifies the likelihood of a household being vulnerable to hurricanes, storms, and ­floods. The index uses data from the country’s national social registry, Sistema Unico de Beneficiarios, which covered approximately ­85.5 ­percent of the population in 2015 (UNDP-UNEP PEI 2018) (map O.1). ASP also highlights a significant need to link to information systems that are typically disconnected from the social protection ­sector. Early warning systems continue to play a critical role in providing and monitoring information for response and in triggering early action, especially in a context of growing climate-related ­risks.9 Drought–food security hybrid systems typically use a range of information on food production, access, and livelihood outcomes from national agencies and international assessments (such as the Famine Early Warning Systems Network and the Integrated Food Security Phase Classification) and merge the information into an assessment of the food-­security status and likely risk (Wilkinson et ­al. ­2018). More recently, forecasts have started using a growing range of climate i­ nformation. Systems using probabilistic forecast information typically draw on products from international, regional, and national forecasting ­centers. Products from international and regional forecasting centers are most common, as these are freely available and considered r­ eliable. Where appropriate, these are complemented with products from national hydrological and meteorological ­services. Indeed, countries are already linking social protection responses to early warning information and developing index-based triggers for response, particularly for slow-onset s­ hocks. In Uganda, satellite data and the Normalized Difference Vegetation Anomaly Index provide the basis for triggering earlier response to d ­ rought through the Northern Uganda Social Action Fund’s (NUSAF) cash-for-work program. In addition to early warning systems, post-shock data collection can play a key role in reflecting socioeconomic conditions and household needs, especially after fast-onset, less predictable, and destructive d ­ isasters. A postdisaster household assessment helps to gather real-time information and data for better understanding a disaster’s impact on household well-being and livelihoods, thereby informing the choice of response programs and the appropriate benefit ­package. Social registry information on households can help inform ­ rocess. In and can be informed by the postdisaster household assessment p Chile, for example, the electronic Basic Emergency Sheet (Ficha Básica de

20 | Adaptive Social Protection

MAP O.1

Dominican Republic: The Vulnerability to Climate Hazards Index (IVACC) a. Zones, by level of vulnerability

b. María Trinidad Sánchez province

Vulnerability level 0.35–0.46

0.56–0.59

Vulnerability level

0.47–0.55

0.60–0.65

0.60–0.65

c. Municipalities of María Trinidad Sánchez

Vulnerability level

d. Neighborhoods of María Trinidad Sánchez

Vulnerability level

0.54 - Municipality of Cabrera

0.30–0.54

0.59 - Municipality of Río San Juan

0.55–0.69

0.62 - Municipality of Nagua

0.70–0.87

0.67 - Municipality of El Factor

Source: UNDP-UNEP 2 ­ 018. Note: The Vulnerability to Climate Hazards Index has a scale of 0–1, where provinces, municipalities, neighborhoods, and households with values close to 0 are the least vulnerable, and those with values close to 1 are the most vulnerable.

Emergencia—FIBE) collects and links postdisaster household assessment data to the social registry, providing a model for merging existing social registry data with up-to-date, post-shock needs assessments and for facilitating virtual, two-way data ­flows. That said, when balancing the trade-off between a timely versus an accurate shock response, speed is more ­important. The literature on the topic is unequivocal: overall timeliness is more important than full targeting accuracy, especially in the first phase of assistance (Beazley, Solórzano, and Sossouvi 2016;

Overview: A Framework for Adaptive Social Protection | 21

O’Brien et ­al. 2018; Pelham, Clay, and Braunholz ­2011). Specifically, inclusion errors can and should be tolerated in the short term, especially as they can contribute to controlling tensions within recipient ­communities. As shock responses evolve and refocus on longer-term recovery, more precise targeting of losses and needs will become increasingly important to identify the households most in need of longer-term ­support.

ASP building block 3: Finance Disaster risk financing is part of a global shift in thinking from seeing disasters as unpredictable humanitarian crises to predictable events that can be planned for and managed to minimize their ­impact. This involves moving from a reactive approach that addresses the impact of shocks once they happen to a more proactive approach, putting in place the required systems and financing to respond to shocks before they take p ­ lace. This approach highlights the need for governments to develop risk financing strategies that enable funding to flow in the event of a shock and thus enable a faster response to ­disasters. The application of these principles and related risk financing instruments to ASP can transform the ability to mobilize a faster response through a social protection system. Indeed, this highlights a strong synergy: when disaster risk financing strategies are established, a shock-responsive safety net program represents a preprepared mechanism through which financial instruments can disburse directly to affected h ­ ouseholds. Conversely, the availability of the kinds of risk financing instruments outlined in this building block and the extent of their linkage to safety net programs will to a large extent determine the speed of the response to affected households. As a first step, financial modeling can better forecast the costs of responding to shocks through safety net ­programs. Leveraging a long time series of historic shock data, models can assess the retrospective incidence and scale of shocks to extrapolate future cost ­scenarios. In Kenya, the Hunger Safety Net Program is capable of horizontally expanding to drought; using a 15-year time series, financial models were able to avail policy makers of the program’s cost implications to be planned for accordingly (figure O.9). Based on such analyses, a disaster risk financing strategy can be developed for shock-responsive social ­protection. No single financial instrument can or should cover all risk financing r­ equirements. Risk-layering considers how to meet the financial cost of response using a menu of financial instruments ­(figure O.10). Each instrument has its own terms and conditions and, therefore, advantages and disadvantages (table O.4). When assessing how to finance contingent liabilities from adaptive social protection, assessing which instruments are the most appropriate, adequate, and cost-effective is ­critical. In most cases, multiple financial instruments will be required to meet the financial cost of the anticipated response(s). Establishing effective disbursement mechanisms (that is, how funding reaches beneficiaries) and linking disaster risk financing instruments to them is as important as securing funds in the first ­place. Having funds available in-country is of limited benefit if they cannot be transferred in a timely manner to the relevant institutions and, in turn, to the shock-affected households. A key factor affecting the disbursement of funds to affected households is the

22 | Adaptive Social Protection

FIGURE O.9

Kenya: Modeling the cost of responding to drought 45 40 35

Million $

30 25 20 15 10 5

Extreme payout

Severe payout

14

12

13

20

20

11 20

20

10 20

09 20

08 20

07 20

06

05

20

04

20

03

20

02

20

20

20

01

0

Average annual payout

Source: Maher, Fitzgibbon, and Solórzano 2 ­ 018. Note: Annual scalability costs are totaled across the four Hunger Safety Net Program ­counties.

FIGURE O.10

Risk layering: Financial instruments, by frequency and severity of a shock

Low frequency/ high severity

Hazard type

Financing instrument Market-based instruments Risk transfer for assets such as property insurance or agricultural insurance and risk transfer for budget management like parametric insurance, catastrophe bonds/swaps Contingent financing

High frequency/ low severity

Financial instruments that provide liquidity immediately after a shock Budgetary instruments Reserve funds specifically designated for financing disaster-related expenditures, general contingency budgets, or diverted spending from other programs

Source: Financial Protection Forum 2 ­ 018.

existence of effective safety net payment ­systems. Countries use different approaches for the deliv­ery of cash-based emergency responses including manual systems (“over-the-counter”), electronic transfers to bank accounts, or via mobile phone payments. Indeed, e-payment systems are emerging as a ­preferred option to deliver postdisaster assistance as they have the advantages of speed and flexibility, even in challenging environments (Maher, Fitzgibbon, and Solórzano ­2018).

Overview: A Framework for Adaptive Social Protection | 23

TABLE O.4  Risk TYPE

layering: Advantages and disadvantages of individual financial instruments ADVANTAGES

DISADVANTAGES

BEST SUITED

Contingency/ reserve funds

• Can be cheap, particularly for frequent shocks • Fast • Allows implementers to plan • Approach has been used in many contexts; thus, experience is available for countries to build upon

• Requires fiscal discipline • High opportunity cost of funds, given high rates of return on other government investments • Can be hard to defend given the opportunity cost

Low risk layer such as frequent low-level events (annual flooding, localized drought, conflict)

Contingent credit

• Can be cheap, particularly for midfrequency shocks • Fast, when conditions for disbursement are met • Allows implementers to plan • Can incentivize proactive actions to reduce risk (for example, policy actions in disaster risk reduction and DRM)

• Has conditionality • Opportunity cost of loan • Adds to country’s debt burden, must be repaid • Current low (but growing) uptake of Cat DDOs as some countries prefer investment projects guaranteed resources over contingent instruments

Mid-risk layer such as higher-magnitude, less frequent events whose damages exhaust the resources of national contingencies (widespread flooding, hurricanes)

Market-based risk transfer instruments

• Can be cheap, particularly for extreme shocks • Can be fast • Allows implementers to plan • Supports fiscal discipline • Risk diversification

• Can be expensive for frequent shocks • Can be vulnerable to criticism and “regret” • Can miss need • Need a level playing field to negotiate • Trade-off between the cost of premiums and the frequency or scale of the pay-out

High-risk layer such as extreme, less frequent events, less than every 5–10 years (severe droughts, hurricanes, earthquakes)

Humanitarian assistance

• Flexible—can respond to need • Doesn’t have to be repaid

• Can be slow to be mobilized • Can be unreliable • Undermines preplanning

Only as a last resort

Other ex post instruments

• Approach has been used in many contexts; thus, experience is available for countries to build upon

• Can be slow • Can have negative impact on longterm development/investment programs • Can be expensive

Only as a last resort

Ex ante

Ex post

Source: Maher, Fitzgibbon, and Solórzano 2 ­ 018. Note: Cat DDO = Catastrophe Deferred Drawdown Option.

Lastly, where finance for ASP is concerned, interventions that support people’s ­longer-term adaptation could reduce future household needs, and by extension, response costs—but more evidence is ­needed. Programs that support household adaptation over the longer term can be costly at scale (for example, in contexts of widespread chronic and severe poverty in highrisk ­areas). Yet, initial evidence indicates that more expensive investment scenarios are broadly offset by the avoided cost of humanitarian response (Cabot Venton 2018; Wilkinson et ­a l. ­2 018). In Bangladesh, the Chars Livelihoods Program focused on building an annual contingency budget into its project design for disaster response, but the need for this contingency fund decreased over time because of the program’s specific focus on reducing vulnerabilities and supporting the adaptation of poor households living in the chars to regular flooding (ADB 2 ­ 018). Where resources are limited, more evidence is needed on the cost-effectiveness of and trade-offs between ex ante r­ esilience-building interventions in support of adaptation and risk reduction at the household level over the long term versus the cost of ex post shock response to support ­short-term ­coping.

24 | Adaptive Social Protection

ASP building block 4: Institutional arrangements and partnerships A defining feature of ASP is the many actors within government that may be involved in its ­implementation. The inherent multidisciplinary and interagency nature of resilience building across the three capacities of preparedness, coping, and adaptation requires diversified expertise and coordination among ­actors. Indeed, the number of potential actors and complementary programs aligned to ASP objectives calls for institutional arrangements that anchor the planning, management, and delivery of this ­assistance. In practice, the development of ASP in many countries has shifted attention from a singular focus on national social protection systems, the policies that guide them, and the organizations that deliver social protection programs to a wider focus inclusive of the policies, organizations, and programs involved in DRM and climate change ­adaptation. Strong government leadership is necessary to ensure coordination of the often disconnected actors, based on a clear articulation of respective roles and ­responsibilities. Concretely, governments are required to lead the ASP agenda by setting resilience-related objectives in policies and strategies, including social protection, DRM, and climate change ­adaptation. Policy commitments can instill the necessary budgetary allocations for national ministerial structures to translate objectives into outcomes among poor and vulnerable h ­ ouseholds. In practice, government leadership also includes establishing the standards and procedures to guide the integration of nongovernmental organizations and humanitarian actors into ASP ­implementation. National social protection policies and strategies should provide the foundation for ­ASP. Most countries have social protection policies and strategies that set out the government’s ­vision for the sector. The extent to which these policies and strategies are rooted in legislation varies (see, for example, Beegle, Coudouel, and Monsalve ­2018). The functions of social protection often are equity, which provides protection against deprivation; resilience, which is insurance against shocks; and, opportunity, which seeks to promote human capital and access to income earning opportunities (World Bank ­2012).10 Articulated in this manner, these policies and strategies provide a foundation for the aims of further elaboration of and commitment to ASP objectives. At the same time, the strategies of other sectors such as—prominently— DRM also can support the advancement of A ­ SP. This is particularly true given the ongoing shift from disaster response to disaster preparedness within the DRM community, as encapsulated in the Sendai Framework for Disaster Risk ­Reduction.11 In Kenya, ASP emanated out of the government’s resolve to address poverty and vulnerability as a cause of and outcome from drought emergencies; a f­ ramework developed by the government (Ending Drought Emergencies) laid the policy foundation for the Hunger Safety Net Program, which expands vertically and horizontally for drought ­emergencies. Where the political appetite for social protection is low, national DRM policies, and the DRM sector more broadly, can provide additional impetus for introducing A ­ SP as part of a comprehensive disaster risk management strategy. This suggests a government’s commitment to ASP can come from sectors other than social protection ­itself. Policy commitment for ASP, to be credible, needs to be backed with appropriate implementation capacity, financing, and ­accountability. To be effective, these

Overview: A Framework for Adaptive Social Protection | 25

policies require enough capacity within the parts of government charged with delivery as well as clear roles and responsibilities, such as those set out in contingency plans and the decision-making in response to early warning d ­ ata ­(chapters 1 and 2). In addition, they also need to be backed by the levels of financing required to achieve the stated ­objectives (chapter 3). The source of this financing may be from national governments or development partners, depending on the prevailing context in the c­ ountry. Accountability mechanisms and feedback loops are central to help ensure that citizens are aware of available programs, inform governments when services are failing, and ultimately, hold governments accountable to their commitments and ­objectives. Additionally, in many contexts of limited national government capacity and/or especially severe shocks, the development and coordination of ASP with humanitarian actors is ­essential. However, humanitarian assistance tends to be provided in parallel to national ­structures. Only 1­ .0–2.5 ­percent of global humanitarian flows channel through host governments (Gentilini, Laughton, and O’Brien ­2018). Factors for this often include a risky operating environment; the need for timely assistance in life-threatening situations; possible lack of government sovereignty over a territory in full or in part; legislation preventing domestic assistance to particular groups; concerns about the impartiality of governments, especially in relation to conflicts; low government capacity; and ensuring transparency and accountability of r­ esources. The humanitarian Grand Bargain and the increasing shift to cash-based assistance are strengthening linkages with national social protection systems and providing impetus for closer i­ ntegration. The 2016 World Humanitarian Summit, and the resulting Grand Bargain, created high-level policy support to strengthen humanitarian linkages with social ­p rotection. The Grand Bargain is a series of 10 commitments to improve assistance to crisis-affected ­populations and included a commitment to “increase social protection programmes and strengthen national and local systems and coping mechanisms in order to build resilience in fragile contexts” (Grand Bargain 2016, ­14). Cash transfers are simultaneously on the rise in national social protection systems and humanitarian programming. Cash now claims about 10 ­percent of global humanitarian assistance, highlighting a strong synergy with national cash ­ 018). Table O.5 summatransfer programs (CaLP 2018; World Bank 2016, 2 rizes these and other features of the humanitarian system, along with their implication for ­ASP. Identifying the precise roles and responsibilities of government and humanitarian actors can help establish actionable, operational partnerships for the delivery of ­ASP. Conceptually, government and humanitarian actors are often viewed simplistically in “either-or” ­terms. A framework laid out by Seyfert et ­al. (2019) attempts to facilitate the identification of workable pathways for progress among national and humanitarian actors (figure O.11). Instead of falling back on the “either–or” choice, the framework lays out four strategic options (parallel systems, alignment, piggybacking, and national-led s­ ystems). It also discusses how collaborations may emerge around select programmatic “functions” and the “degrees” of possible connection between national and humanitarian actors within a given ­function. While a work-in-progress, such a granular analytical approach holds the potential to move beyond strategic dialogue and strategies in support of coordination; that is, coordination toward an operationally relevant delineation of roles and responsibilities based on relative comparative advantages in differing country ­contexts.

26 | Adaptive Social Protection

TABLE O.5  International

humanitarian system: Features and implications for adaptive social protection

FEATURE

CHARACTERISTICS

IMPLICATIONS

Policy commitments and the Grand Bargain

High-level policy support for building resilience and increasing the role of social protection

Opportunities to advance aims of ASP but need to be translated into more concrete and strategic actions

Bifurcation of humanitarian/ development and rise of resilience building

Humanitarian and development assistance often underpinned by different financing channels, coordination structures, mandates, and principles

Divide between humanitarian and development systems may remain an obstacle; need for specificity on “resilience building”

Humanitarian financing

Very little direct funding goes to national governments; significant flows to fragile and conflict settings, and year-on-year to the same places

Limited potential for humanitarian financing to be channeled to governments for national safety nets; scope to fund nongovernmental organizations operating within national frameworks for ASP in some countries

Shares of humanitarian financing go to some areas and populations supported by national safety nets Humanitarian principles

Humanitarian assistance is guided mainly by the four principles of humanity, impartiality, independence, and neutrality Differing views on flexibility of principles exist, but they are not incompatible with working with governments

Humanitarian principles should inform the response function of ASP to shocks; can be referenced to advocate for a principled engagement around ASP with governments by humanitarian agencies

Coordination

Established mechanisms for coordination (see the cluster system, chapter 4) but varying coordination approaches because of differing levels of national involvement in those mechanisms

Need for engagement of ASP at various levels of humanitarian operational and strategic coordination and for bilateral engagement with major donors and aid agencies

Increasing shift to cash transfers

Cash transfers increasingly accepted as mainstream tool of humanitarian response, but programs often fragmented and still represent only a small share of total assistance

Offers an entry point for engagement of national ASP programs with humanitarian system

Source: Bailey 2 ­ 018.

FIGURE O.11

ASP delivery approaches: A mix across national and humanitarian actors Parallel systems

Alignment

Financing Legal and policy framework Setting eligibility criteria and qualifying conditions Setting transfer type, level, frequency, duration Governance and coordination Outreach Registration Enrollment Payment Case management Complaints and appeals Protection VAM/M&E Information management Source: Seyfert et ­al. ­2019. Note: VAM/M&E = Vulnerability analysis and mapping/monitoring and e ­ valuation.

Piggybacking

National-led systems

Overview: A Framework for Adaptive Social Protection | 27

NOTES  1. Shocks may be either idiosyncratic or covariate in n ­ ature. An idiosyncratic shock is felt by an individual or household, with the negative impacts to their well-being typically not shared more widely by others outside of the immediate h ­ ousehold. Idiosyncratic shocks include things such as ill health, injury, disease, disability, a death in the family, and job ­loss. Covariate shocks are larger in scale, affecting multiple individuals and households at once, with the negative impacts to well-being spread across a (typically large) number of ­persons.  2. An undetermined share of this increase is undoubtedly due to better recording of events and their impact during that time ­period.  3. This capacity also is referred to as “anticipatory” capacity in the BRACED 3As framework (Bahadur et ­al. ­2015). The term “preparedness” is used here to more explicitly reflect the meaning of the capacity as used in this ­report. The capacity to anticipate a shock based on appropriate information is recognized as a critical component of the capacity to prepare, informing appropriate ­action.  4. This capacity also is referred to as “absorptive” capacity in the BRACED 3As framework (Bahadur et ­al. 2015; similar in Béné et ­al. ­2012). The term “coping” is chosen here because of its widespread use in the social protection community and its interchangeability with the term ­“absorptive.”  5. Defined as a failure to adjust adequately or appropriately to a s­ hock.  6. Led by the “International Labour Organization (ILO) and the World Bank Group (WBG), in partnership with the African Union, the Food and Agriculture Organization (FAO), the European Commission, the Inter-American Development Bank (IDB), Organization of Economic Cooperation and Development (OECD), the United Nations Development Programme (UNDP) and its International Poverty Centre for Inclusive Growth (IPC-IG), the United Nations Children’s Fund (UNICEF), and others, along with Belgian, Finnish, French, and German development cooperation, and international civil society organizations such as HelpAge, the International Council of Social Welfare (ICSW), Save the Children, among others” (World Bank and ILO ­2018, 1). ­ ambia.  7. The six African countries are Ethiopia, Ghana, Kenya, Lesotho, Malawi, and Z  8. These are sometimes classified as “soft resilience measures” typically low cost and adaptable to deliver benefits in changing conditions (Cabot Venton et ­al. ­2012).  9. The United Nations International Strategy for Disaster Reduction defines early warning systems as “an integrated system of hazard monitoring, forecasting and prediction, disaster risk assessment, communication and preparedness activities systems and processes that enables individuals, communities, governments, businesses and others to take timely action to reduce disaster risks in advance of hazardous events” ­(www.unisdr.org/we​ /­inform/terminology). 10. See also Devereux and Sabates-Wheeler (2004), who set a similar framework of protection, prevention, promotion, and ­transformation. 11. Sendai Framework, ­https://www.unisdr.org/we/coordinate/sendai-framework.

REFERENCES ADB (Asian Development ­Bank). ­2018. “Strengthening Resilience through Social Protection ­Programs.” Guidance ­Note. ADB, ­Manila. Andrews, C ­ ., A ­ . Hsiao, and ­L. R ­ alston. 2 ­ 018. “Social Safety Nets Promote Poverty Reduction, Increase Resilience, and Expand O ­ pportunities. In Realizing the Full Potential of Social Safety Nets in Africa, edited by K ­ . Beegle, ­A. Coudouel, and E ­ . Monsalve, 87–137. Washington, DC: World ­Bank. Arnall, ­A., ­K. Oswald, ­M. Davies, ­T. Mitchell, and ­C. ­Coirolo. ­2010. “Adaptive Social Protection: Mapping the Evidence and Policy Context in the Agriculture Sector in South ­Asia.” IDS Working Paper, Institute of Development Studies, Brighton, U ­ K. Asfaw, ­S., and ­B. ­Davis. ­2018. “Can Cash Transfer Programmes Promote Household Resilience? Evidence from Sub-Saharan ­Africa.” In Climate Smart Agriculture, edited by ­L. Lipper, ­N. McCarthy, D ­ . Zilberman, ­S. Asfaw, and ­G. Braca, ­227–50. Rome: Food and Agriculture Organization of the United ­Nations. ­http://www.fao.org/3/a-i7931e.pdf.

28 | Adaptive Social Protection

Bahadur, A ­ ., ­K. Peters, ­E. Wilkinson, ­F. Pichon, ­K. Gray, and ­T. ­Tanner. ­2015. “The 3As: Tracking Resilience across ­BRACED.” Working Paper, Building Resilience and Adaptation to Climate Extremes and Disasters (BRACED), ­London. Bailey, S ­. 2 ­ 018. “Institutions for Adaptive Social Protection: External Linkages and the Humanitarian ­Sector.” Background ­Paper. Oxford Policy Management, Oxford, ­UK. Barca, ­V. ­2017. Integrating Data and Information Management for Social Protection: Social Registries and Integrated Beneficiary ­Registries. Canberra: Commonwealth of Australia, Department of Foreign Affairs and ­Trade. ­https://dfat.gov.au/about-us/publications​ /­Documents/integrating-data-information-management-social-protection-full.pdf. Barca, ­V., and ­C. ­O’Brien. ­2017. “Factors Affecting the Usefulness of Existing Social Protection Databases in Disaster Preparedness and ­Response.” Policy ­Brief. Oxford Policy Management, Oxford, ­UK. ­https://assets.publishing.service.gov.uk/media/5a942c50ed915d57d4d0ef98​ /­Policy-Brief-Factors-affecting-usefulness-existing-social-protection-databases.pdf. Bastagli, F ­.2 ­ 014. “Responding to a Crisis: The Design and Delivery of Social P ­ rotection.” ODI Working Paper, Overseas Development Institute, ­London. Bastagli, F ­ ., J ­ . Hagen-Zanker, ­L. Harman, V ­ . Barca, G ­ . Sturge, and T ­.S ­ chmidt. 2 ­ 016. “Cash Transfers: What Does the Evidence Say? A Rigorous Review of Programme Impact and of the Role of Design and Implementation F ­ eatures.” Overseas Development Institute, ­London. Bastagli, ­F., and ­R. ­Holmes. ­2014. “Delivering Social Protection in the Aftermath of a Shock: Lessons from Bangladesh, Kenya, Pakistan, and Viet ­Nam.” Overseas Development Institute, ­London. Beazley, ­R. ­2017. “Study on Shock-Responsive Social Protection in Latin America and the Caribbean: Ecuador Case ­Study.” Oxford Policy Management, Oxford, ­UK. Beazley, ­R., ­A. Solórzano, and K ­ . ­Sossouvi. ­2016. “Study on Shock-Responsive Social Protection in Latin America and the Caribbean: Theoretical Framework and Literature R ­ eview.” Oxford Policy Management, Oxford, ­UK. Beegle, ­K., ­A. Coudouel, and E ­ . ­Monsalve. ­2018. Realizing the Full Potential of Social Safety Nets in ­Africa. Washington, DC: World ­Bank. ­http://documents.worldbank.org/curated​ /­en/657581531930611436/pdf/128594-PUB-PUBLIC.pdf. Béné, ­C., ­S. Devereux, and ­R. ­Sabates-Wheeler. ­2012a. “Shocks and Social Protection in the Horn ­ thiopia.” IDS Working of Africa: Analysis from the Productive Safety Net Programme in E Paper 395, Institute of Development Studies, Brighton, ­UK. Béné, C ­ ., R ­ . Wood, A ­ . Newsham, and M ­ . ­Davies. 2 ­ 012b. “Resilience: New Utopia or New Tyranny? Reflection about the Potentials and Limits of the Concept of Resilience in Relation to Vulnerability Reduction ­Programmes.” IDS Working Paper 405, Institute of Development Studies, Brighton, ­UK. Bossuroy, ­T., and P ­ . ­Premand. ­2016. “Boosting Productive Inclusion and Resilience of the Poor: Perspectives from the Sahel Adaptive Social Protection ­Program.” ­Presentation, World Bank, Washington DC. Bowen, ­T. ­2015. “Social Protection and Disaster Risk Management in the Philippines: The Case of Typhoon Yolanda ­(Haiyan).” Policy Research Working Paper 7482, World Bank, Washington, ­DC. ­http://documents.worldbank.org/curated/en/681881468181128752/pdf​ /­WPS7482.pdf. Cabot Venton, ­C. ­2018. “Economics of Resilience to Drought in Ethiopia, Kenya and Somalia: Executive S ­ ummary.” Center for Resilience, U ­ .S. Agency for International Development, Washington, ­DC. Cabot Venton, ­C., ­C. Fitzgibbon, ­T. Shitarek, ­L. Coulter, and O ­ . ­Dooley. ­2012. “The Economics of Early Response and Disaster Resilience: Lessons from Kenya and Ethiopia: Economics of Resilience Final ­Report.” Department for International Development, ­London. CaLP ­2018. The State of the World’s Cash Report: Cash Transfer Programming in Humanitarian ­Aid. Oxford, ­UK. ­http://www.cashlearning.org/downloads/calp-sowc-report-web.pdf. Dang, H ­ .-A., and A ­ .D ­ abalen. 2 ­ 017. “Is Poverty in Africa Mostly Chronic or Transient? Evidence from Synthetic Panel ­Data.” Policy Research Working Paper 8033, World Bank,

Overview: A Framework for Adaptive Social Protection | 29

Washington, ­DC. ­http://documents.worldbank.org/curated/en/172891492703250779​ /­pdf/WPS8033​.pdf. Davies, M., C. Béné, A. Arnall, T. Tanner, A. Newsham, and C. Coirolo. 2012. “Promoting Resilient Livelihoods through Adaptive Social Protection: Lessons from 124 Programmes in South Asia.” Development Policy Review 31 (1): 27–58. Davies, ­M., ­B. Guenther, ­J. Leavy, ­T. Mitchell, and ­T. ­Tanner. ­2009. “Adaptive Social Protection: Synergies for Poverty ­Reduction.” IDS Bulletin 39 (4): ­105–12. de Weijer, ­F. ­2013. “Resilience: A Trojan Horse for a New Way of Thinking?” ECDPM Discussion Paper 139, European Centre for Development Policy Management, ­Maastricht. del Ninno, C ­ ., and B ­ . Mills, e­ ds. 2 ­ 015. Safety Nets in Africa: Effective Mechanisms to Reach the Poor and Most V ­ ulnerable. Washington, DC: World B ­ ank. ­http://documents.worldbank.org​ /­curated/en/869311468009642720/pdf/Safety-nets-in-Africa-effective-mechanisms-to​ -reach-the-poor-and-most-vulnerable.pdf. Devereux, ­S., and ­R. ­Sabates-Wheeler. ­2004. “Transformative Social P ­ rotection.” IDS Working Paper 232, Institute of Development Studies, Brighton, ­UK. Esteves T ­ ., K ­ . ­V. Rao, B ­ . Sinha, ­S. ­S. Roy, ­B. ­B. Rai, I­ . ­B. Rao, N ­ . Sharma, S ­ . Rao, V ­ . Patil, ­I. ­K. Murthy, ­J. Srinivasan, R ­ . ­K. Chaturvedi, J ­ . Sharma, ­S. ­K. Jha, ­S. Mishra, ­A. ­B. Singh, ­H. ­S. Rakhroy, ­S. Rai, ­R. Sharma, ­S. Schwan, ­K. Basu, ­N. Guerten, ­I. Porsché, ­N. Ranjan, ­K. ­K. Tripathy, and ­N. ­H. ­Ravindranath. ­2013. “Environmental Benefits and Vulnerability Reduction through Mahatma Gandhi Rural Employment Guarantee Scheme (NREGS): Synthesis R ­ eport.” Ministry of Rural Development, Government of India and Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), New ­Delhi. FAO (Food and Agriculture Organization of the United N ­ ations). ­2016. Adapting Agriculture to Climate C ­ hange. FAO’s Work on Climate Change A ­ daptation. Rome: F ­ AO. ­http://www.fao​ .­org/3/a-i6273e.pdf. Financial Protection Forum. 2018. “Disaster Risk Finance: A Primer, Core Principles, and Operational Framework.” World Bank, Washington, DC. https://financialprotectionforum​ .org/publication/disaster-risk-finance-a-primercore-principles-and-operational​ -framework. Frankenberger, ­T., ­T. Spangler, ­S. Nelson, ­M. ­Langworthy. ­2012. “Enhancing Resilience to Food Insecurity amid Protracted C ­ risis.” United Nations High-Level Expert Forum, R ­ ome. ­http:// www.fao.org/fileadmin/templates/cfs_high_level_forum/documents/Enhancing​ _­Resilience_FoodInsecurity-TANGO.pdf. FSIN (Food Security Information Network). ­2015. “Measuring Shocks and Stressors as Part of Resilience ­Measurement.” Technical Series ­5. Resilience Measurement Technical Working Group, FSIN Secretariat, World Food ­Programme. ­http://www.fsincop.net/fileadmin/user​ _upload/fsin/docs/resources/1_FSIN_­TechnicalSeries_5.pdf. Gentilini, ­U., ­S. Laughton, and ­C. ­O’Brien. ­2018. “Lessons on Better Connecting Humanitarian ­ rotection.” Social Protection and Labor Discussion Paper 1­ 802. Assistance and Social P World Bank, Washington, ­DC. ­http://documents.worldbank.org/curated/en/94640154​ 2689917993/pdf/Human-itarian-Capital-Lessons-on-Better-Connecting-Humanitarian​ -Assistance-and-Social-Protection.pdf. Grand ­Bargain. ­2016. “The Grand Bargain: A Shared Commitment to Better Serve People in ­Need.” World Humanitarian Summit, I­ stanbul. h ­ ttps://reliefweb.int/sites/reliefweb.int​ /­files/resources/Grand_Bargain_final_22_May_FINAL-2.pdf. ­ . Fay, ­T. Kane, U ­ . Narloch, J ­ . Rozenberg, D ­ . Treguer, Hallegatte, ­S., ­M. Bangalore, ­L. Bonzanigo, M ­ overty. and ­A. ­Vogt-Schlib. ­2016. Shock Waves: Managing the Impacts of Climate Change on P Washington, DC: World ­Bank. Hallegatte, ­S., ­J. Rentschler, and ­B. ­Walsh. ­2018. “Building Back Better: Achieving Resilience ­ econstruction.” World Bank, through Stronger, Faster, and More Inclusive Post-Disaster R Washington, ­DC. ­http://documents.worldbank.org/curated/en/420321528985115831​ /­pdf/127215-REVISED-BuildingBackBetter-Web-July18Update.pdf. Hallegatte, ­S., ­A. Vogt-Schilb, ­M. Bangalore, and ­J. ­Rozenberg. ­2017. Unbreakable: Building the Resilience of the Poor in the Face of Natural ­Disasters. Washington, DC: World ­Bank. ­https:// www.gfdrr.org/sites/default/files/publication/Unbreakable_FullBook_Web-3.pdf.

30 | Adaptive Social Protection

Hidrobo, M ­ ., ­J. Hoddinott, ­J. Kumar, and M ­ . ­Oliver. ­2018. “Social Protection, Food Security, and Asset ­Formation.” World Development 101: ­88–103. Hill, ­R., ­E. Skoufias, and ­B. ­P. ­Maher. ­2019. The Chronology of Disaster: A Review and Assessment of the Value of Acting Early on Household ­Welfare. Washington, DC: World ­Bank. ­http:// documents.worldbank.org/curated/en/796341557483493173/pdf/The-Chronology-of-a​ -Disaster-A-Review-and-Assessment-of-the-Value-of-Acting-Early-on-Household​ -Welfare.pdf. Holmes, ­R. ­2019. “Promoting Gender Equality and Women’s Empowerment in Shock Sensitive Social ­Protection.” ODI Working Paper 549, Overseas Development Institute, ­London. ­ roup). 2 ­ 011. Social Safety Nets: An Evaluation of World Bank IEG (Independent Evaluation G Support, ­2000–2010. Washington, DC: World ­Bank. ILO (International Labour ­Organization). ­2017. World Social Protection Report 2017–19: Universal Social Protection to Achieve the Sustainable Development ­Goals. Geneva: ­ILO. ­https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/---publ/documents​ /­publication/wcms_604882.pdf. Jorgensen, S ­ ., and P ­.S ­ iegel. 2 ­ 019. “Social Protection in an Era of Increasing Uncertainty and Disruption: Social Risk Management 2 ­ .0.” Social Protection and Jobs Discussion Paper 1930, World Bank, Washington, ­D C. ­h ttp://documents.worldbank.org/curated​/­e n/2637​ 61559643240069/pdf/Social-Protection-in-an-Era-of-Increasing-Uncertainty​ - and​ -Disruption-Social-Risk-Management-2-0.pdf. Kaur, ­N., ­D. Steinbach, ­A. Agrawal, ­C. Manuel, ­S. Saigal, ­A. Panjiyar, ­C. Shakya, and ­A. ­Norton. ­2017. “Building Resilience to Climate Change: MGNREGS and Climate-Induced Droughts in ­Sikkim.” IIED Issue ­Paper. International Institute for Environment and Development, ­London. Kuriakose, ­A., ­R. Heltberg, ­W. Wiseman, ­C. Costella, ­R. Cipryk, and ­S. ­Cornelius. ­2012. “ClimateResponsive Social ­Protection.” Social Protection and Labor Discussion Paper 1210, World Bank, Washington, ­DC. ­https://siteresources.worldbank.org/SOCIALPROTECTION​ /­Resources/SP-Discussion-papers/430578-1331508552354/1210.pdf. Leite, ­P., ­T. George, ­C. Sun, ­T. Jones, and ­K. ­Lindert. ­2017. “Social Registries for Social Assistance and Beyond: A Guidance Note and Assessment T ­ ool.” Social Protection and Labor Working Paper 1704, World Bank, Washington, ­DC. ­http://documents.worldbank​ .­o rg/curated​ /­e n/698441502095248081/pdf/117971-REVISED-PUBLIC-Discussion​ -paper-1704.pdf. Lindert, ­K., ­T. George, ­I. Rodriguez-Caillava, and Kenichi N ­ ishikawa. ­Forthcoming. A Sourcebook on the Foundations of Social Protection Delivery ­Systems. Washington, DC: World ­Bank. Macours, K ­ ., ­P. Premand, and R ­ . ­Vakis. 2 ­ 012. “Transfers, Diversification and Household Risk Strategies: Experimental Evidence with Lessons for Climate Change A ­ daptation.” Policy Research Working Paper 6053, World Bank, Washington, D ­ C. ­http://documents.worldbank​ .org/curated/en/275241468340175496/pdf/WPS6053.pdf. Maher B ­ ., C ­ . Fitzgibbon, and A ­ . ­Solórzano. 2 ­ 018. “Emerging Lessons in Financing Adaptive Social ­Protection.” Background Paper for the World Bank, Oxford Policy Management, ­London. Mansur, ­A., ­J. Doyle, ­J. Gerome, and ­O. ­Ivaschenko. ­2017. “Social Protection and Humanitarian Assistance Nexus for Disaster Response: Lessons Learnt from Fiji’s Tropical Cyclone ­Winston.” Social Protection and Labor Discussion Paper ­1701. World Bank, Washington, ­DC. ­http://documents.worldbank.org/curated/en/143591490296944528/pdf/113710-NWP​ -PUBLIC-P159592-1701.pdf. Manyena, ­S., ­G. O’Brien, ­P. O’Keefe, and ­J. ­Rose. ­2011. “Disaster Resilience: A Bounce back or Bounce forward Ability?” Local Environment 16 (5): ­417–24. Mariotti, C ­ ., M ­ . Ulrichs, and ­L. H ­ arman. 2 ­ 016. “Sustainable Escapes from Poverty through Productive Inclusion: A Policy Guide on the Role of Social P ­ rotection.” CPAN Policy Guide ­7. Chronic Poverty Advisory Network, ­London. O’Brien, ­C., ­Z. Scott, ­G. Smith, ­V. Barca, ­A . Kardan, ­R. Holmes, ­C. Watson, and ­J. ­Congrave. ­2 018. “Shock-Responsive Social Protection Systems Research: Synthesis R ­ eport.”

Overview: A Framework for Adaptive Social Protection | 31

Oxford Policy Management, Oxford, ­UK. ­https://www.opml.co.uk/files/Publications​ /­a 0408-shock-responsive-social-protection-systems/srsp-synthesis-report.pdf ?​ noredirect=1. PEI (Partnership for Economic I­ nclusion). ­2016. “Increasing the Income Earning Opportunities of Poor and Vulnerable P ­ eople.” World Bank, Washington, D ­ C. h ­ ttps://www​ .­microfinancegateway.org/sites/default/files/announcement/pei-brochure.pdf. Pelham, ­L., ­E. Clay, and ­T. ­Braunholz. ­2011. “Natural Disasters: What Is the Role for Social Safety Nets?” SP Discussion Paper 1102, World Bank, Washington, ­DC. ­https://www.gfdrr.org​ /­sites/default/files/documents/Social%20Safety%20Nets.pdf. Robalino, ­D., ­L. Rawlings, and ­I. ­Walker. ­2012. “Building Social Protection and Labor Systems: Concepts and Operational I­ mplications.” Social Protection and Labor Working Paper 1202, World Bank, Washington, ­DC. ­http://siteresources.worldbank.org/SOCIALPROTECTION​ /­Resources/SP-Discussion-papers/430578-1331508552354/1202.pdf. Roelen, ­K., ­S. Devereux, ­A. ­G. Abdulai, ­B. Martorano, ­T. Palermo, and ­L. ­P. ­Ragno. ­2017. “How to Make ‘Cash Plus’ Work: Linking Cash Transfers to Services and S ­ ectors.” Innoncenti Working Paper ­2017-10. United Nations Children’s Fund Office of Research, F ­ lorence. Schipper, E ­ ., and ­L. ­Langston. 2 ­ 015. “A Comparative Overview of Resilience Measurement ­Frameworks.” ODI Working Paper 422, Overseas Development Institute, ­London. Seyfert, ­K., ­V. Barca, ­U. Gentilini, ­M. Luthria, and ­S. ­Abbady. ­2019. “Unbundled: A Framework for Connecting Safety Nets and Humanitarian Assistance in Refugee S ­ ettings.” Social Protection and Labor Discussion Paper 1935, World Bank, Washington, D ­ C. ­https:// openknowledge.worldbank.org/ bitstream/ handle/10986/32467/Unbundled-A​ -Framework​-for-Connecting-Safety-Nets-and-Humanitarian-Assistance-in-Refugee​ -Settings​.pdf?sequence=1&isAllowed=y. Solórzano, ­A. ­2016. “Can Social Protection Increase Resilience to Climate Change? A Case Study of Oportunidades in Rural Y ­ ucatan, Mexico.” IDS Working Paper 465, Centre for Social Protection and Institute of Development ­Studies, Brighton, UK. ­ eview.” Tenzing, ­J. ­D. ­2019. “Integrating Social Protection and Climate Change Adaptation: A R WIREs Climate Change 11 (2): ­e626. Ulrichs, ­M., and ­R. ­Slater. ­2016. “How Can Social Protection Build Resilience? Insights from Ethiopia, Kenya and ­Uganda.” Working Paper, Building Resilience and Adaptation to Climate Extremes and Disasters (BRACED), ­London. UNDP-UNEP (United Nations Development Programme–United Nations Environment ­Programme). 2018. Vulnerability to Climate Hazards Index: Lessons Learned and Systematization of the Design Process and Application of the IVACC Index—Dominican Republic. Poverty-Environment Initiative. Panama City: Panama. UNHCR (United Nations High Commissioner for ­Refugees). ­2016. “Global Trends in Forced ­Displacement.” UNHCR, ­Geneva. UNICEF. 2 ­ 018. “Resilience, Humanitarian Assistance and Social Protection for Children in ­ . ­https://www.unicef.org​ Europe and Central ­Asia.” Social Protection Regional Issue Brief: 2 /­eca/media/2671/file/Social_Protection2.pdf. UNISDR (United Nations International Strategy for Disaster R ­ eduction). 2 ­ 015. Making Development Sustainable: The Future of Disaster Risk Management: Global Assessment Report on Disaster Risk ­Reduction. Geneva: ­UNISDR. UNU-EHS (United Nations University–Institute for Environment and Human ­Security). ­2016. World Risk Report ­2016. Berlin: Bündnis Entwicklung ­Hilft. WFP (World Food ­Programme). ­2015. “Policy on Building Resilience for Food Security and ­Nutrition.” ­WFP/EB.A/2015/5-C. WFP, ­Rome. ­https://documents.wfp.org/stellent/groups​ /­p ublic/documents/eb/wfpdoc063833.pdf?_ga=2.20959473.817428444.1582152603​ -752767465.1554223343. WFP (World Food Programme). ­2018. “Scaling Up for Resilient Individuals, Communities and Systems in the ­Sahel.” Fact ­Sheet. WFP, ­Rome. ­https://docs.wfp.org/api/documents/WFP​ -0000110238/download​/?_ga=2.26639604.817428444.1582152603-752767465​.15542​23343.

32 | Adaptive Social Protection

Wilkinson, ­E., ­L. Weingartner, ­R. Choularton, ­M. Bailey, ­M. Todd, ­D. Kniveton, and ­C. Cabot ­Venton. 2 ­ 018. “Forecasting Hazards, Averting Disasters: Implementing Forecast-Based Early Action at ­Scale.” Overseas Development Institute, ­London. World B ­ ank. 2 ­ 012. “Resilience, Equity, and Opportunity: The World Bank’s Social Protection and Labor Strategy 2 ­ 012–2022.” World Bank, Washington, D ­ C. ­http://documents.worldbank. org/curated/en/443791468157506768/pdf/732350BR0CODE200doc0version0​ REVISED.pdf. World Bank. ­2013. Building Resilience to Disaster and Climate Change through Social Protection: Synthesis ­Note. Washington, DC: World ­Bank. ­http://documents.worldbank.org/curated​ /­en/187211468349778714/pdf/796210WP0Build0Box0377381B00PUBLIC0.pdf. World Bank. ­2016. “Cash Transfers in Humanitarian Contexts: Strategic ­Note.” World Bank, Washington, ­DC. ­http://documents.worldbank.org/curated/en/697681467995447727​ /­pdf/106449-WP-IASC-Humanitarian-Cash-PUBLIC.pdf. World Bank. 2 ­ 018. The State of Social Safety Nets 2­ 018. Washington, DC: World B ­ ank. h ­ ttps:// openknowledge.worldbank.org/ bitstream/ handle/10986/29115/9781464812545​ .­pdf?sequence=5&isAllowed=y. World Bank and ILO (International Labour ­Organization). ­2018. Universal Social Protection: Country ­Cases. Global Partnership for Universal Social Protection U ­ SP2030. Washington, DC: World Bank and ­ILO. ­https://www.social-protection.org/gimi/gess/RessourcePDF​ .­action?id=55072.

1

Programs

DESIGN CONSIDERATIONS FOR BUILDING RESILIENCE

OVERVIEW To enhance their impact on resilience building, safety net programs need to be explicitly designed to support the capacity of poor and vulnerable households to prepare for, cope with, and adapt to the shocks they ­face. First and foremost, this requires asking fundamental questions of existing programs—are they reaching those that are most vulnerable to shocks, and does the assistance they provide maximize resilience-related outcomes among beneficiaries? Such appraisals can inform the adjustment of existing programs, or the introduction of new programs to enhance the impact of the social protection system on building resilience to s­ hocks. An emphasis on encouraging savings can help to enhance people’s preparedness by creating a cash or asset based “buffer” to be drawn upon after a shock has ­hit. Moreover, financial inclusion of beneficiaries can both support saving in this way while also making those beneficiaries more accessible with assistance after a shock has h ­ it. Preparedness can also be supported when beneficiaries are informed about the risks they face and the appropriate strategies and actions they can take, including through even very basic disaster preparedness training and information on the timing of lean or rainy ­seasons. Now widely utilized, communications sessions that are provided to deliver information on health and education to beneficiaries of safety net programs and their wider communities can provide under-exploited venues for the communication of this pertinent preparedness i­ nformation. The continued provision of safety net transfers to beneficiaries who have been affected by a shock can be especially effective in helping them to cope with the impact, smoothing consumption and reducing the need to use negative coping ­mechanisms. Critically, safety net programs will need to be explicitly prepared to respond after a shock, so that they are more capable of providing timely assistance to help people to cope—in most cases, reaching above and beyond those people typically reached through existing safety net programs delivering in times of acute need and environments of heightened operational complexity. Lastly, over the long term, cash transfers can be transformative in helping people make the investments needed to ­adapt. This impact may be enhanced when cash transfers are accompanied by additional measures that encourage income diversification and adoption of more productive livelihoods, reducing the reliance on livelihoods that are at high risk of s­ hocks. When cash transfers also are designed to stimulate human capital investments among poor and vulnerable households, they may better equip the next generation to access opportunity that enables moving away from areas and out of livelihoods of concentrated r­ isk. In complement, public works programs can be designed to enable beneficiaries to implement projects that address some of the sources of vulnerability at the community level, including water insecurity, soil erosion, and frequent f­ looding.

Programs

This chapter draws extensively from the background paper by Barca (­ 2018).  33

34 | Adaptive Social Protection

FOCUSING ON THE ROLE OF SAFETY NETS IN BUILDING HOUSEHOLD RESILIENCE A social protection system comprises multiple programs that can support and sustain the capacity of households to prepare for, cope with, and adapt to s­ hocks. Social protection programs translate government policies and objectives into outcomes among beneficiary h ­ ouseholds. To date, a significant body of knowledge has accrued on social protection program design and implementation, especially in terms of social safety net programs and their use in the pursuit of poverty reduction and human capital-related objectives, from foundational texts such as Grosh et ­al. (2008) to more recent treatments such as Subbarao et al. (2012) and Lindert et ­al. (forthcoming), to name only a select ­few. Where building resilience to covariate shocks is concerned, however, the knowledge base around program design and implementation is only beginning to emerge (see ADB 2018; McCord 2013; O’Brien et ­al. ­2018a). Within the social protection system, social safety net programs can be harnessed and enhanced to become increasingly capable of building the ­resilience of poor and vulnerable households to covariate s­ hocks. The impact of assistance provided through a safety net program can be transformative across a beneficiary’s resilience capacities: preparedness, coping, and ­adaptation. The delivery of a safety net transfer can provide a supplementary source of income which can empower beneficiaries to take critical preparedness measures (such as accumulating ­savings). In turn, this preparedness can help beneficiaries to cope once a shock hits by smoothing consumption and lessening the need to use negative coping ­strategies. The continued provision of safety net transfers following a shock can be instrumental in supporting coping capacity, especially when preparedness measures are overwhelmed by a severe shock. Shocks of this nature will invariably generate needs among households beyond core safety net beneficiaries that can also be addressed by temporarily expanding access to safety net programs during and after crises. Over the longer term, safety net transfers that are integrated with productive measures, livelihood diversification, and human capital initiatives also can contribute to the capacity of a vulnerable household to adapt, reducing its exposure and ­vulnerability. This chapter highlights an emerging body of knowledge on the role of safety net programs in building resilience to covariate s­ hocks. It begins by reviewing the ways that core design parameters for all safety nets determine their contribution to resilience ­building outcomes. This includes programmatic objectives and resultant decisions on who is reached and with what kinds of ­assistance. The chapter then addresses fundamental design considerations for supporting the three capacities—preparedness, coping, and adaptation—with safety net ­programs.

UNDERSTANDING THE CHALLENGE OF BUILDING HOUSEHOLD RESILIENCE: NO SINGLE PROGRAM CAN “DO IT ALL” A common thread throughout much of the literature on building household resilience to shocks is that one program cannot “do it all” (ADB 2018; Asfaw and Davis 2018; Bastagli and Holmes 2014; Fallavier 2014; OPM 2 ­ 017). The social protection system comprises social assistance, social insurance, and labor

Programs: Design Considerations for Building Resilience | 35

market programs—each reaching population subsets with unique forms of assistance and each with varying resilience-building ­utility. As such, a program-specific discussion can be misleading in lieu of a discussion of the wider social protection system, the position of any one social protection program within it, and its comparative advantage in building resilience across the three capacities among which subset of the ­population. Evidence suggests that investing in a stronger, more comprehensive social protection system composed of multiple programs provides the requisite foundation for building the resilience of poor and vulnerable h ­ ouseholds. It is important to take stock of the capacity of a country’s existing social protection system to build the resilience of the poor and vulnerable to covariate s­ hocks. As noted in the literature, stronger, more comprehensive systems with a mix of social insurance, social assistance, and labor market programs better position a country when a shock ­hits. This enables a country to draw from a larger toolbox and facilitates diverse points of access to support vulnerable households to cope with the impacts (Grosh et ­al. 2011; IEG 2011; Isik-Dikmelik 2012; Marzo and Mori 2012; O’Brien et ­al. ­2018a). In the United States, for example, the wider social protection system has been found to provide substantial assistance across its programs to households affected by hurricanes, implying people are more “insured” against the impacts from these shocks than previously thought ­(box 1.1). Beyond the traditional social protection system itself, adaptive social protection (ASP) highlights the need for strong coordination between social ­protection and the disparate actors and programs each working on building the resilience of vulnerable households to ­shocks. Take, for example, the many agriculture, human development (health and education), and disaster risk reduction programs that explicitly or implicitly seek to build household resilience to covariate ­shocks. For those other nonsocial protection sectors, coordination with social ­protection actors will often enable increased access among the poorest households, which may otherwise not be reached through their ­programming. Similarly, after a shock hits, many emergency response and recovery

BOX ­1.1

The United States: Role of the social protection system in insuring against disasters A study by Deryugina found that in the United States, government transfers through the social protection system to areas affected by hurricanes substantially increase in the decade after a ­hurricane, suggesting that US households are better insured against hurricanes than currently ­recognized. By extension, the fiscal costs of disasters in the United States have been ­underestimated because they have not accounted for the transfers through the social protection system in the Source: Deryugina ­2016.

decade after a hurricane, serving as a form of insurance to the affected h ­ ouseholds. By extension, the study suggests that victims in developed countries with comprehensive social protection systems are better insured against disasters than previously ­thought. This also suggests that “expanding social safety nets provides benefits not only to those affected by idiosyncratic shocks and general economic downturns, but also to victims of natural ­disasters” (Deryugina 2016, 197).

36 | Adaptive Social Protection

programs deliver from a multitude of ministries, departments, and agencies as well as from nongovernmental and humanitarian organizations to help people to cope with the i­ mpacts. Where such coordination, coherence, and integration of programming among vulnerable households is achieved in practice, household gains in resilience building could be more significant and sustainable; see, for example, the integrated and layered programmatic approaches to building resilience undertaken by the World Food Programme (2015, ­2019). Stress testing the social protection system as a whole and its ability to build resilience to the shocks faced by a country can provide the basis for modifying existing programs or introducing new p ­ rograms. Much of the literature related to ASP, especially shock-responsive social protection, highlights ex ante stress testing of the social protection system against potential shocks (see, for example, Atkinson 2009; Barca 2018; Kanbur 2009; McCord ­2013). Indeed, leveraging the analyses outlined in chapter 2, a risk-informed appraisal of the social protection system can highlight the coverage of social protection programs among the most vulnerable households and the collective impact and effectiveness of the programs on resilience building with regard to the shocks faced by the ­country. Such analyses could inform the programs that are available to households after a shock hits and elucidate the flexibility and capacity, or lack thereof, of those programs in responding to shocks of varying magnitudes and ­types. Such analyses also could provide the basis for the needs-based alterations of existing programs and/or for the introduction of new ­programs.

APPRAISING THE CORE DESIGN FEATURES OF EXISTING SAFETY NET PROGRAMS FOR RESILIENCE BUILDING Returning to the program level, there are three factors to consider in appraising the core design features of a safety net program and its contribution to building household resilience to covariate shocks: the objectives of the safety net, who it reaches with assistance, and the specific parameters of the benefits it p ­ rovides.

Programmatic objectives: The right tool for the job? The objectives of most safety net programs do not explicitly frame the p ­ rogram as an instrument for building resilience to covariate shocks—which can mean that design features are not directly aligned to such ­outcomes. Most often, safety net programs pursue objectives related to poverty reduction, human-capital accumulation, and the promotion of access to ­opportunity. These objectives do each themselves tacitly support resilience building, especially for idiosyncratic shocks, often among a subset of the population that is inherently vulnerable to covariate ­shocks. This has served to elevate the recognition of the potential role of safety nets as instruments to build resilience to covariate s­ hocks. However, without explicit objectives related to such outcomes, the potential for safety nets to build resilience may be ­constrained, remaining only tacit. For example, the provision of support to the capacity to cope with covariate shocks through the delivery of shock-­responsive social protection may contrast with the priorities that underpin social protection development in noncrisis times and have several implications for the design

Programs: Design Considerations for Building Resilience | 37

and delivery including the tensions and trade-offs with regular social protection planning and the design and implementation details that facilitate timely, adaptable and adequate social protection programs (Bastagli 2 ­ 014). The objectives of existing safety net programs can be reimagined and adjusted, or new programs with explicit objectives introduced, as the basis for building resilience to covariate ­shocks. Social protection programs that have been specifically designed to build resilience to climate-induced food insecurity, such as Kenya’s Hunger Safety Net Program (HSNP) and Ethiopia’s Productive Safety Net Programme (PSNP), illustrate that integrating resilience objectives will more explicitly frame the program and its consequent design features in these ­terms. For example, by design, the PSNP connects to high-level policies on social protection, disaster risk management, and climate change to promote “resilience to shocks and enhance livelihoods, improve food security and nutrition for rural households vulnerable to food insecurity” (Government of Ethiopia ­2014, 21). Nevertheless, beyond the importance of possessing explicit resilience-building objectives on paper, consequent design features and, ultimately, the effectiveness of implementation will determine the extent of the program’s impact on resilience building (Ulrichs and Slater ­2016).

Who is being reached? Are they the most vulnerable to shocks? Often, safety net programs only reach a subset of those most vulnerable to ­shocks. The subset of the population reached by different safety net programs is a function of the safety net’s objectives and resultant decisions on eligibility criteria and ­coverage. Low social protection coverage or coverage of those who are not the most vulnerable to covariate shocks inevitably limits the role of social protection in building resilience (Bastagli and Holmes ­2 014). Uniformly broadening coverage across and within programs widens the safety net, expanding the reach of the safety net ­system. However, universal access to social protection among the population is a long-term proposition for most countries (Gentilini ­2019). In the mean time, when looking to fill the holes in the safety net that are highlighted by covariate shocks, traditional approaches to beneficiary selection need to be reevaluated. Countries operating narrow poverty-­ targeted programs with very low coverage are likely to require strategies for expanding access to the poor and vulnerable that are at risk from shocks (Isik-Dikmelik 2012; O’Brien et ­al. ­2018a). While the poor are more vulnerable to shocks, the near-poor and nonpoor also are vulnerable to falling into poverty because of a shock (Grosh et al. 2011; Hallegatte et al. 2016). After a shock, the overlap between the poor, those vulnerable to poverty because of the shocks, and beneficiaries of poverty-targeted programs coexisting also depends on the type of ­s hock. For example, economic shocks and slow-onset (or seasonal) food-security crises tend to affect low-income households disproportionately, while rapid-onset disasters can impact the affected population transversally (O’Brien et ­al. ­2018a). Where there are categorically targeted programs (such as social pensions or child grants) that have broad coverage they also may support resilience building among beneficiaries that are acutely vulnerable to shocks (that is, women, children, the elderly, and the d ­ isabled). That said, any single categorical program also will be limited in its reach among the totality of poor and vulnerable ­households.

38 | Adaptive Social Protection

As a foundation, risk-informed geographic targeting can increase the reach of a safety net in areas of spatially concentrated risk and high household ­vulnerability to shocks. To date, geographic targeting for social protection is not typically assessed through a covariate-shock l­ ens. Most often, geographic targeting assesses the depth and breadth of poverty on a subnational ­basis. It is the first of often multiple targeting mechanisms to select where the program should go, broadly s­ peaking. Within those areas, households are then selected using additional targeting methods such as ­community-based targeting and the proxy means t­ est. In Bangladesh, social protection is prioritized where poverty and vulnerability to shocks are intertwined, including for the lean season in the Northwest (monga), monsoons in the Northeast (haors), and tropical storms in the Southwest (Coastal Belt) (Bastagli and Holmes 2 ­ 014). Similarly, in Niger, a composite index of geographic vulnerability to climate shocks was used as one of the criteria to determine where to target the safety net program (box ­1.2). In addition to geographic targeting, household targeting criteria that include measures of vulnerability to shocks can increase the impact of a safety net program on resilience ­building. Climate-smart targeting incorporates household-level data in addition to geographic assessments of spatial vulnerability to help identify the specific households that are more vulnerable to natural hazards and climate change risks and that may be, as a result, eligible for programs that can build resilience (ADB 2018; Bastagli and Holmes 2014; World Bank ­2 013). In the Dominican Republic, the Vulnerability to Climate Hazards Index (IVACC) calculates the probability of a given household being affected by climate shocks using three factors: housing characteristics, estimated income, and proximity to a hazardous ­ 017a). In Niger, natural element (such as a river, stream, or ravine) (Beazley 2

BOX ­1.2

Niger: Geographic targeting based on spatial vulnerability to climate shocks In Niger, vulnerability to climate shocks was taken into consideration for the geographical targeting of the safety net ­program. The selection of departments within each region is based on three geographical targeting criteria: poverty, food insecurity, and vulnerability to ­shocks. In each region, the departments whose poverty rate is above the regional median as well as departments in which the share of severe and moderate food insecurity exceeds 20 ­percent are ­selected. A composite index of geographic vulnerability to climate shocks is then ­applied. Specifically, the index integrates: indicators of rainfall and Normalized Difference Vegetation Index (NDVI) anomalies to capture the recurrence Source: World Bank ­2018b.

of drought events; an indicator of price shocks; the ratio of cereal use for consumption to domestic production at the department level; and the vulnerability score given to each department annually by the National Framework for the Prevention and Management of Food Crisis (Dispositif National de Prévention et de Gestion des Catastrophes et des Crises ­A limentaires). All the departments whose vulnerability index is found to be above the 80th ­p ercentile are included into the p ­ rogram. Having selected the geographic areas at the department level in this way, additional targeting criteria are applied to select the specific communes within them and then, finally, the eligible ­households.

Programs: Design Considerations for Building Resilience | 39

in addition to geographically targeting areas that are vulnerable to climate shocks, as outlined above, the proxy means test targeting formula used for selecting beneficiary households was also adjusted to take into account the different factors that contribute to poverty and vulnerability based on a household’s location in different ­agricultural-ecological zones (World Bank ­2018b). For example, in a zone where the main livelihood is in agriculture, ownership of land will result in a higher weight in the proxy means test than other v­ ariables. In a zone where livestock is the predominant livelihood, livestock will result in a higher weight in the proxy means test than other variables, including versus ­land. This is done to sensitize the proxy means test to the fact that livelihoods have different effects on a household’s level of poverty as well as their vulnerability to climatic shocks based on where they ­live.

What kind of assistance are beneficiaries receiving? Does it maximize resilience-related outcomes? When a safety net designed to build resilience to covariate shocks reaches the most vulnerable households, the design parameters of the benefits it provides will play a mediating role in enabling or restraining resilience-enhancing ­outcomes. Broadly speaking, the design of the benefit package of, say, a cash transfer program can significantly impact resilience building and a program’s ability to support a beneficiary household’s capacity to prepare for, cope with, and adapt to a covariate shock (Barca ­2018). A program’s transfer values can be adjusted and set in relation to ­resilience-​ building ­objectives. Due to the many programmatic objectives pursued by safety net programs, the amount transferred to beneficiaries can vary significantly across i­nterventions. Safety nets tend to cover a small share of a poor person’s income/consumption—13 ­percent on average in low-income countries (World Bank ­2018a). An increasing body of evidence from Sub-Saharan Africa indicates programs that include transfers amounting to over 20 ­percent of per capita income produce more significant results on resilience building (Beazley and Farhat 2016; Daidone et ­al. 2015; Davis ­2014). Conversely, lack of impacts across recent evaluations has often been attributed to modest transfer sizes and erosion of value over time due to inflation (see, for example, Arnold, Conway, and Greenslade 2011; Asfaw et ­al. 2014; McCord et ­al. ­2016). Beyond the general point on the need for adequate transfer levels, frequent transfers of lesser value tend to favor consumption-smoothing and spending on smaller assets that are beneficial in support of coping with shocks, but which may not stimulate investment in adaptation; less frequent, lump-sum payments tend to increase productive investment in support of adaptation and savings in support of preparedness (Beazley and Farhat 2016; Haushofer and Shapiro 2013) as well as longer-term recovery and the reaccumulation of assets following a large-scale ­disaster. The timing of the safety net transfers can enhance their impact on resilience ­building. Where the needs of poor and vulnerable households are predictable in relation to shocks (such as rainy or lean seasons), the provision of benefits can be timed accordingly to align with this ­seasonality. This way, beneficiaries receive assistance ahead of the need to resort to negative coping mechanisms materializing, enhancing preparedness and the ability to cope (Barca et ­al. 2015; Beazley, McCord, and Solórzano ­2016).

40 | Adaptive Social Protection

The duration of time that a household is a beneficiary of a program can impact the resilience-enhancing ­outcomes. Evaluations of cash transfer programs showcase a “number of improvements in outcomes arising from increased duration [of time spent in a program], including some improvements in health behaviors and child growth outcomes, higher expenditure and food expenditure, lower likelihood of early marriage, pregnancy and greater contraceptive use” (Bastagli et ­al. ­2016, 11). In Ethiopia, Kenya, and Uganda short-term assistance has been found to limit the program’s impacts on seasonal and chronic food insecurity (Ulrichs and Slater 2 ­ 016). Evidence also is increasing on the detrimental effects of households that stop receiving transfers, questioning the longer-term sustainability of impacts and shedding light on the need for responsible “exit” and/or “graduation” strategies to ensure that those who exit do not return to a position of vulnerability1 (Bastagli et ­al. ­2016). Predictable transfers are fundamental to building r­ esilience. No matter what the parameters of the transfers—amount, frequency, timing, duration—a regular and predictable benefit is critical to generating resilience-building ­outcomes. A predictable transfer can be relied upon by a beneficiary to smooth consumption and preempt negative coping; it provides certainty with which to better plan, thus prepare; and it supports risk-taking behavior that contributes to long-term ­adaptation. In the context of covariate shocks, where a transfer is not predictable and reliable, the resulting uncertainty undermines the ability for beneficiaries to not undertake negative coping ­strategies. The importance of predictability is extensively discussed across evaluations and reviews of existing social protection interventions (Andrews, Hsiao, and Ralston, 2018; Barca et ­al. 2015; Bastagli et ­al. 2016; Daidone et ­al. 2015; del Ninno, Pierre, and Coll-Black 2016; Fallavier 2014; Solórzano 2016; Ulrichs and Slater ­2016).

DESIGN FEATURES TO SUPPORT THE CAPACITY TO PREPARE, COPE, AND ADAPT In addition to these core design considerations across programs, the impacts of safety nets across the three resilience capacities can be examined more closely. Any one program can provide support across all three capacities, and all three capacities are i­nterlinked. However, the remainder of this chapter disentangles the three capacities and analyzes the design features and types of safety net programs that can serve to support each of the three capacities among identified, at-risk, poor, and vulnerable households (table ­1.1).

Support to the capacity to prepare for shocks Safety nets can support the capacity to prepare for shocks when they reach the most vulnerable households in advance—stimulating savings, financial inclusion, and the communication of information on how to better prepare, cope, and ­adapt. Safety net beneficiaries often use transfers to save, especially in contexts of recurrent s­ hocks. Evidence indicates that safety nets have significant impacts on increased savings, improved creditworthiness, and reduced debt, though the extent varies based on the context and design (Andrews, Hsiao, and Ralston, 2018; Bastagli et ­al. 2016; Hidrobo et ­al. 2018; Ulrichs and

Programs: Design Considerations for Building Resilience | 41

TABLE ­1.1

Adaptive social protection: Supporting the capacity to prepare, cope, and adapt PREPAREDNESS

COPING

A more resilient household

• More savings (cash, assets) to draw upon if a shock occurs • Access to public (social protection) and private (insurance) instruments if needed after a shock • Access to information on their own exposure and vulnerability to shocks (including early warning information) to inform action

• Activates coping • Capable of making long-term mechanisms: acting on investments to reduce exposure and information (including vulnerability over time early warning informa• Adjustment of asset and livelihood tion), leverages savings, portfolios away from sources of risk assets, public and private and vulnerability instruments to smooth • Planned movement and migration consumption and to away from areas of spatially supplement lost income concentrated, chronic risk

Poor and vulnerable households

• Limited savings and assets to draw on • In the absence of • Fewer resources with which to make if a shock occurs adequate savings and long-term investments in adaptation • Limited or no access to public (social access to social through adjustments in livelihood protection) and private (insurance) protection and/or private and asset portfolios that can lead to instruments if needed should a shock insurance, resort to • Maladaptation and chronic occur negative coping vulnerability • Limited access to information on their strategies—cutting • Forced displacement and exposure and vulnerability (including consumption, removing unplanned migration early warning information) to inform children from school, action distress sale of ­assets, among others

Role of safety • Increased access to safety nets among • Support to post-shock net programs in the poor and vulnerable, especially coping through supporting those identified as at-risk from shocks continued delivery preparedness, • Transfers to at-risk households before during and after a shock coping, and shocks occur to support savings and to existing beneficiaries adaptation asset accumulation • Shock-responsive among the poor • Safety nets leveraged to transmit programs capable of and vulnerable information on exposure and adjusting benefit households vulnerability, enabling the increased package and temporarily anticipation of shocks, and informing increasing the number actions in support of preparedness, of beneficiaries as coping, and adaptation needed based on post-shock needs

Slater ­2016). In Mexico, beneficiaries of the former national conditional cash transfer program (Prospera) who lived in communities highly exposed to droughts and hurricanes use the transfer “to save for the bad times” and their creditworthiness in local markets increases because the communities know the recipient households and the payment intervals (Solórzano 2 ­ 016). Similar insights were consistent across programs in Ethiopia, Kenya, and Uganda (Ulrichs and Slater ­2016). Indeed, in Africa, safety net beneficiary households are 4–20 ­percentage points more likely to save relative to comparable nonbeneficiary households (figure ­1.1). Given the initial low savings rate among poor households, this implies an expansion by a factor of almost two in the incidence of savings (Beegle, Coudouel, and Monsalve ­2018). Safety nets can encourage financial inclusion, which can be instrumental for enhancing the capacity to prepare, cope, and ­adapt. Financial inclusion seeks to extend financial services to all citizens by ensuring that services are accessible, affordable, and appropriate (ISPA ­2017). Access to financial services can help households reduce their vulnerability to shocks by facilitating saving in support of preparedness; enhancing opportunities for income generation, asset accumulation, and consequently adaptation (Cull, Ehrbeck, and Holle 2014); and even making beneficiaries reachable more easily and quickly with assistance in the form of electronic payments after a shock has h ­ it. In the context of social

ADAPTATION

• Support to long-term adjustment of asset and livelihood portfolios, including through cash, cash plus, and productive inclusion ­interventions • Community asset-building projects through public works programs that address key drivers of ­community-level vulnerability • Support to human capital accumulation for intergenerational adaptation through increased opportunity

42 | Adaptive Social Protection

FIGURE ­1.1

Africa: Safety net beneficiaries tend to use the transfers to save a. Percentage change in savings rate from baseline 300 % change from baseline

250 200 150 100 50 0 –50 –100 Tanzania TASAF

Kenya HSNP

Zambia ZCGP

Program impact Mean impact

Kenya CTOVC

Ghana LEAP

Sierra Leone CFW

Lesotho LCGP

95% confidence interval of program impact 95% confidence interval of mean impact

b. Estimated change in savings rate 100 Savings rate (%)

80 60 40 20 0 –20 Tanzania TASAF

Kenya HSNP

Zambia ZCGP

Kenya CTOVC

Impact estimate

Ghana LEAP

Sierra Leone CFW

Lesotho LCGP

Baseline mean

Source: Beegle, Coudouel, and Monsalve ­2018. Note: The mean value of the household transfer (in 2011 US$ purchasing power parity) is Tanzania Social Action Fund $48, Kenya’s Hunger Safety Net Program $47, Zambia’s Child Grant Program $27, Kenya Cash Transfer for Orphan and Vulnerable Children $71, Ghana’s Livelihood Empowerment against Poverty $24, Sierra Leone Cash for Work $83, and Lesotho Child Grants Program $34.

protection, financial inclusion can be explicitly supported where beneficiaries are given access to a “store-of-value” transaction account, encouraged to save, and/or given access to savings ­groups. In addition, safety net programs can provide channels for communicating information on exposure and vulnerability to poor and vulnerable households that are hard to r­ each. Social protection programs rely on a network of implementers who often reach into the community, including social workers and village/­community leaders, and may be involved in program i­mplementation. These unique channels into the community level often are used to transmit information to beneficiaries through community that may be mandatory (or not) for cash-transfer ­beneficiaries. Such ­“behavioral change communication” sessions can provide venues and conduits for the communication of early warning information to beneficiaries and the wider community that are otherwise

Programs: Design Considerations for Building Resilience | 43

hard to reach, as well as broader information on disaster risk, risk reduction, and adaptation measures (ADB ­2018). In the Philippines, Family Development Sessions, an integral component of the national conditional cash transfer program (Pantawid Pamilya Pilipino Program), are used as a vehicle and venue to deliver information and practices on disaster risk management to all beneficiaries—including what to pack and where to go (Bowen 2 ­ 015). In Mexico, the former conditional cash transfer program Prospera and civil protection together provided disaster preparedness training to social protection beneficiaries (Beazley, Solórzano, and Barca ­2019).

Support to the capacity to cope with shocks Safety net programs have well-documented, positive impacts on a household’s capacity to cope with shocks, through the delivery of support after a shock has ­hit. Of the three resilience capacities, safety nets tend to demonstrate the strongest impact on supporting coping capacity: supporting consumption, lessening food insecurity, and providing alternatives to negative coping (Ulrichs and Slater ­2016). Complementing a transfer’s impact before a shock and its potential to generate a buffer to be leveraged after a shock hits, the continued receipt of transfers during and after a shock supports post-shock ­coping. Impact evaluations of safety net programs in six African countries describe “unambiguous” increases in the food security of beneficiary households2 (Asfaw and Davis ­2018). In Ethiopia, the PSNP reduced the initial impact of a drought by 57 ­percent on beneficiaries, eliminating the adverse impact on food security within 2 years (Knippenberg and Hoddinott 2017) (figure ­1.2). Relatedly, safety nets can play a role in reducing the need to resort to negative coping strategies that trigger ­longer-term detrimental effects (Barca et ­al. 2015; Dammert et ­al. 2018; Hill, Skoufias, and Maher ­2019).

FIGURE ­1.2

Ethiopia: PSNP beneficiary and nonbeneficiary recovery trajectories 12 11

Months food secure

10

Drought

9 8 7 6 5 4 3 2

0

1

2

3

Years since drought No PSNP

PSNP

Total

Source: Knippenberg and Hoddinott ­2017. Note: Representative household at mean level of food ­security. Blue bands represent 95 ­percent confidence ­intervals. PSNP = Productive Safety Net ­Programme.

4

44 | Adaptive Social Protection

In the context of covariate shocks, safety net programs with resilience-­ building objectives need to be shock-responsive to provide extraordinary support to help households cope with often devastating i­ mpacts. In their framework for shock-responsive social protection, O’Brien et ­al. (2018b) outline five potential ways that social protection programs can be leveraged to respond to largescale shocks: • Design tweaks are small adjustments to a routine social protection ­program. They can introduce flexibility to maintain the regular service for existing beneficiaries in a shock (for example, by waiving c­ onditionalities). Alternatively, they can address vulnerabilities that are likely to increase in a crisis, through adjustments to program coverage, timeliness, or predictability (for example, by altering the payment schedule), without requiring a flex at the moment of the ­shock. • Vertical expansion is the temporary increase of the value or duration of a social protection intervention to meet the additional needs of existing ­beneficiaries. For such vertical expansions to be relevant, the program or programs must have good coverage of the disaster-affected area and also of the neediest ­households. • Horizontal expansion is the temporary inclusion of new beneficiaries from disaster-affected communities into a social protection program, by extending geographical coverage, enrolling more eligible households in existing areas, or altering the enrollment ­criteria. • Piggybacking occurs when an emergency response uses part of an established system or program while delivering something ­new. Exactly which and how many elements of the system or program are borrowed will vary; it could be, for example, a specific program’s beneficiary list, its staff, a national database, or a particular payment ­mechanism. • Alignment describes designing an intervention with elements resembling others that already exist or are planned, but without integrating the ­two. For example, this could be an alignment of objectives, targeting method, transfer value, or delivery ­mechanism. Governments may align their systems with those of humanitarian agencies or vice versa, either because an existing intervention is not operational as needed in a crisis or because it may not yet ­exist. Where a safety net exists and has a good degree of coverage among affected households, vertical expansion offers a relatively simple method of providing ­ eneficiaries. Recent examples include the more assistance to existing, affected b vertical expansion of the social protection system in Fiji following Tropical Cyclone Winston in 2016 and of the national conditional cash transfer program, through additional grants from humanitarian actors (the World Food Programme and UNICEF) following Typhoon Haiyan (Yolanda) and Typhoon Ruby in the ­Philippines. In the case of Fiji, an impact assessment found that households that received the vertical expansion were more likely to report having recovered from the shock more quickly; for instance, they were 8–10 ­percent more likely to have recovered from housing damage than nonbeneficiaries (Mansur et ­al. ­2017). However, vertical expansions generally do not reach nonbeneficiary, shock-­ affected households that may be in equal or greater need of a­ ssistance. As noted, programs with traditional safety net objectives related to poverty reduction, for

Programs: Design Considerations for Building Resilience | 45

example, reach subsets of a shock-affected p ­ opulation. Vertical expansions alone therefore risk missing shock-affected households (Barca and O’Brien 2 ­ 017). In Ecuador, only 15 ­percent of households within the Registro de Damnificados (the database of affected households established in the aftermath of the 2016 earthquake) received the country’s flagship social assistance program, Bono de Desarrollo Humano (Beazley 2 ­ 017b). In Mozambique, the government estimate on the median population affected by the 2016 droughts was 15 ­percent across the 71 affected d ­ istricts. The median coverage of the country’s largest social assistance program in these districts was only 9 ­percent, suggesting that “even if the recipients of the Basic Social Security Program were indeed the population most affected by the drought, there was still a large population in need of support” (Kardan et ­al. ­2017, 43). As such, the ability to temporarily reach additional households that may be equally or more in need of support to their coping capacity but may not be regular beneficiaries of social protection programs is critical for shock-responsive social ­protection. This can be achieved through horizontal expansion and dedicated emergency programs or through coordination and alignment with non–safety net and/or nongovernmental interventions, including humanitarian ­assistance. Horizontal expansion enables a safety net program to temporarily expand its caseload after a shock to include new households based on eligibility from a disaster’s ­impacts. Introducing the ability to scale out in this manner is far more complex than undertaking vertical expansion to existing ­beneficiaries. Scaling out requires significant investment in the processes and procedures for delivering the program, often in challenging settings. Several countries have invested in the capacity to horizontally expand a safety net program, including most prominently Ethiopia with the PSNP and Kenya with the HSNP; each is prepared to undertake horizontal expansions based on needs generated by drought and related food ­insecurity. In the United States, the Supplemental Nutrition Assistance Program (SNAP) routinely expands its caseload horizontally in response to economic shocks; Disaster-SNAP (D-SNAP) provides temporary assistance to nonregular participants who have suffered significant losses by relaxing program requirements to ease access and relieve administrative burdens on staff (FNS ­2014). Some countries use a dedicated emergency program with characteristics similar to a safety net (cash, in-kind, and public works), which may piggyback on core safety net delivery systems and ­capacity. Emergency programs have dedicated response objectives and exist outside of an existing safety net p ­ rogram. Relative to expanding safety net programs, the use of a dedicated emergency program with a singular objective to respond to a shock holds the advantage of ring fencing an existing safety net with its existing, non–­disaster response objectives and design f­ eatures. It enables the development of a dedicated instrument with appropriate design f­ eatures. It also can reduce confusion around entry and exit decisions for temporary beneficiaries, some of whom may be transitioned into longer-term safety net programs (depending on their longer-term needs following the ­disaster). Such programs can be located within or outside of the social protection ministries, departments, and agencies and can leverage underlying safety net delivery systems (such as social registries, payment systems, and front-line social protection s­ taff ). This process has been referred to as piggybacking (O’Brien et ­al. ­2018a).

46 | Adaptive Social Protection

In Pakistan, the Citizen’s Damage Compensation Program was initially implemented as an emergency flood response that utilized a stand-alone beneficiary registration and payment distribution system in partnership with commercial banks and linked to the national civil registry (World Bank 2 ­ 013). The cash-based disaster response in Pakistan has since evolved to encompass a combined ­approach. The Benazir Income Support Program (BISP) is used for vertical expansions to extreme poor ­households. Horizontal expansion to non-program beneficiaries is undertaken by the National Database and Registration Authority with the use of dedicated one-stop shops (Citizen Facilitation Centers) when support is required by segments of the population over and above the program’s poverty cut-off or where tailored responses are ­required. In the Sahel, with support from the Sahel ASP Program, Mauritania has developed a dedicated emergency response program (“Elmaouna”) that piggybacks on existing social protection social registries and payment ­platforms for its ­delivery (Mauritania 2019). Whichever approach to shock-responsive social protection is undertaken, the different phases of postdisaster support generally require multiple programs to reach different households with different kinds of a­ ssistance. Where multiple safety net programs exist and cover different subsets of the population, more than one program may need to be used. This includes the use of different programs across the post-shock phases, depending on the relative strengths and weaknesses of each program type, the precise needs generated by a shock, and how they change ­over time. For example, this may include transitioning from supporting life-saving basic needs in the short term to assistance that supports longer-term recovery of livelihoods and assets (figure ­1.3). Operational processes and contingency plans need to be clearly defined in advance—including who does what when—to better ensure faster ­implementation. Countries should outline clear processes for the delivery of

FIGURE ­1.3

The Philippines: Multiple programs for differing household needs across differing post-shock time periods Community driven

National Community-Driven Development

Core Shelter Assistance Program

Programs

Shelter

Emergency Shelter Assistance

Temporary shelter assistance

Cash for Building Livelihood Assets Cash transfer programs

Pantawid conditional cash transfer delivered unconditionally Cash for Work Nonfood items

Relief Food for Work Family food packs Relief

Early recovery Response

Recovery

Reconstruction Recovery and rehabilitation Time

Source: Bowen ­2015.

Programs: Design Considerations for Building Resilience | 47

assistance in post-shock environments in operations manuals or contingency plans (as in the case of Ethiopia, box ­1.3). These manuals serve as a key guide for implementing ­staff. After a destructive shock, the often difficult operating environment makes it hard to implement response actions effectively if they are defined in an ad hoc manner only after the fact, contributing to ­delays. Operational processes can be defined as a subsection of existing program manuals (as in the case of Mexico, box ­1.3) or as stand-alone d ­ ocuments. Ideally, they should be as comprehensive as possible, detailing how processes vary from established procedures, among others: time frames for waiving conditions; mechanisms for triggering payments in information systems; rules for taking on and exiting new beneficiaries; grievance redress procedures; and clear details on data management and information sharing ­protocols. The social protection delivery chain can help identify the critical implementation processes and contingencies required for shock-response programs, to be captured in contingency plans and protocols, ex a­ nte. The delivery chain,

BOX ­1.3

Ethiopia, Mexico, and Pakistan: Contingency planning In Ethiopia, the PSNP identified the district (woreda) for collecting household information and designed contingency plans to ensure that in the event of shocks, transitory and regular beneficiaries received support in the same ­manner. The five steps of the PSNP contingency planning include (1) context analysis, based on early warning information, historical data, and community needs; (2) scenario assessment, identifying hazards and their potential impacts on food security and estimating the number of potential additional beneficiaries; (3) response planning, including targeting of transitory beneficiaries, public works activities, and budgeting; (4) operational support planning, including setting up “shelf projects” for public works and identifying the necessary decision makers, resources, systems, and structures; and (5) revising contingency plans, based on new information, early warning systems, and annual ­updates. In Mexico, most cash transfer programs, including the former national conditional cash transfer program Prospera, have rules of operation for implementing ­interventions. These rules are updated annually and broadly define changes in operational processes in the event of a disaster, declared emergency, or ­epidemic. In the case of the former national conditional cash transfer program Prospera, these changes include, among others, (1) paying cash transfers without Sources: Coll-Black et ­al., forthcoming; Mexico ­2018.

verification of co-responsibilities for no more than four consecutive months (unless the National Coordinating Council authorizes an extension) in emergency situations where delivery of education and health services are impeded, (2) adjusting operational processes of the National Coordinating Council, (3) delaying recertification (with prior authorization from the National Coordinating Council) by 1 year for households in affected areas, and (4) deploying Prospera personnel to affected ­areas. In Pakistan, the government developed a national strategy for managing catastrophic events, the Federal Disaster Response Action Plan, which outlines contingency plans and the minimum resources and swiftest approval processes required to respond to s­ hocks. The plan clearly defines the cash response model for emergencies and the roles and responsibilities of the respective partner agencies critical for future r­ esponses. These include the national and provincial disaster management authorities, the Benazir Income Support Program, the National Database and Registration Authority, the Ministry of Finance, and commercial ­banks. The processes outlined in the Federal Disaster Response Action Plan have since been implemented during response to floods in Sindh province (2012–13), as well as to internally displaced persons of the conflict-affected Federally Administered Tribal Area regions ­(2015).

48 | Adaptive Social Protection

FIGURE ­1.4

Social protection delivery chain Assess

Enroll

Outreach

Intake and registration

Assess needs and conditions

1

2

3

Eligibility Determine and benefits enrollment and service decisions package

4

5

Provide

Notification and onboarding

Benefits and/or services

6

7

Manage

Beneficiaries, Exit decisions, notifications, grievances, compliance case outcomes

Recurring cycle

8

9

Periodic reassessment

Source: Lindert et ­al., ­forthcoming.

pictured in figure ­1.4, outlines core steps for the implementation of any safety net program to provide the right amount of benefits and services, to the right households, at the right t­ ime. When designing contingency plans for shock-response programs, the delivery chain forms a helpful basis for thinking through the key questions and considerations as well as the ex-ante investments in adapting the delivery chain that are required to undertake those processes and deliver the response program (for further detail, see appendix ­B).

Support to the capacity to adapt to shocks Alongside supporting short-term coping after a shock, governments can use safety nets to invest in the capacity for poor and vulnerable households to adapt to shocks over the long t­ erm. There has been an increasing and justifiable focus on the role of safety nets in supporting post-shock ­coping. That said, ASP and the wider definition of resilience used here highlight the central importance of supporting a household’s longer-term adaptation in order to reduce its vulnerability to a shock over ­time. By broadening the focus in this way, safety net programs can also be seen as promising tools for providing pathways toward a more ­resilient state for poor households that are vulnerable to shocks (see also Tenzing ­2019). By extension, where successful, these investments may serve to reduce future post-shock needs over ­time. As a foundation for supporting adaptive capacity, the provision of longterm, predictable, and adequate cash transfers can be instrumental in supporting income generation, livelihood diversification, asset accumulation, and human capital accumulation (Asfaw and Davis 2018; Barca 2018; FAO ­2019). In Africa, safety net programs enhance the ownership of productive assets and strengthen livelihoods; one of the most striking results is the significant rise in livestock ­ownership. Across seven programs, livestock ownership improved 34 ­percent relative to baseline levels (Beegle, Coudouel, and Monsalve ­2018). Expenditures on durables (tools and other equipment for farms and businesses) exhibited a smaller, but still significant, i­mprovement: a 10 ­percent increase relative to the ­baseline. Ethiopia and Malawi demonstrate improved fertilizer and seed use, which may indicate a shift to ­higher-risk, higher-return agricultural ­practices.

Programs: Design Considerations for Building Resilience | 49

Concretely, safety nets can support adaptive capacity when designed to help the poor accu­mulate and diversify assets and livelihoods (Bahadur et al. 2015; FSIN 2015; Jorgensen and Siegel 2019). The promotion of more productive and resilient livelihoods among poor and vulnerable households is one of the core elements of supporting adaptive ­capacity. Interventions that promote more productive and resilient livelihoods have the potential to empower beneficiaries to diversify their asset and resilient livelihood portfolios and to reduce their exposure and vulnerability to shocks. For example, a study by Macours, Premand, and Vakis (2012) found that the provision of vocational training or a productive investment grant in addition to a cash transfer to beneficiaries vulnerable to drought in Nicaragua provided full protection against drought shocks 2 years after the end of the intervention (relative to a control group that only received a cash transfer) (figure ­1.5). Similarly, safety nets can contribute to livelihood promotion through specific programs that link cash transfer recipients to complementary interventions in other sectors (for example, agricultural inputs, training, and microfinance), leading to p ­ ositive—yet varied—impacts on production and diversification into on-farm and off-farm opportunities (FAO 2016; Mariotti, Ulrichs, and Harman 2016 as cited in FAO ­2019). For this reason, productive inclusion programs are emerging as potentially powerful instruments for supporting the adaptive capacity of the poorest households by supporting transitions into more productive and resilient ­livelihoods. Productive inclusion provides an integrated package of assistance, in addition to cash transfers, that is designed to overcome barriers preventing households from moving into more productive and resilient livelihoods. Integrated packages of intervention often include cash transfers plus a mixture of the following: skills and micro-entrepreneurship training tailored to local livelihood opportunities; promotion of and support for saving groups;

FIGURE ­1.5

Log (total per capita consumption)

Adaptive capacity: Welfare is less sensitive to shocks for beneficiary households receiving productive grants 9.4 9.3 9.2 9.1 9.0 8.9

−1.0

−0.5

0 0.5 Shock intensity Control

1.0

1.5

Treatment

Source: Macours, Premand, and Vakis ­2012. Note: Households eligible for the productive investment package and control ­only. Shock intensity is standardized by subtracting the mean and dividing by the standard ­deviation. Estimates exclude shocks more than 1 standard deviation removed from ­others. Fan regressions with bandwidth of ­1.5. Graph trimmed at 5 ­percent highest and lowest values of shocks ­intensity. Estimates using fan regressions as in Fan and Gijbels ­(1996).

50 | Adaptive Social Protection

provision of seed capital and productive grants; linkage to existing value chains and markets; and mentoring, behavior, and life skills to build confidence and reinforce existing skillsets, among others (Bossuroy and Premand 2016; PEI 2016; Roelen et ­al. ­2017). Impact evaluations over the past 15 years have shown the resilience-enhancing value of such interventions, both in terms of breaking the cycle of chronic poverty and mitigating the risks of “backsliding” into poverty after a shock (Banerjee et ­al. 2015; PEI 2016; Samson 2015) (figure ­1.6 and box ­1.4). In complement to the support for more productive livelihoods, public works programs can be designed to support a community’s capacity to adapt to ­shocks. Through careful design and planning, assets created with public works programs can help build the resilience of communities (ADB 2018; Asfaw and Davis 2018; Steinbach et ­al. 2017a, 2017b, 2017c; Subbarao et ­al. ­2013). The following channels facilitate community-level adaptive capacity: • Engaging communities in climate-smart agriculture and integrated natural resource management, including a focus on waste management, reforestation, rainwater harvesting, soil/water conservation, and drought-resistant horticulture, among ­others.3 • Supporting wider disaster-reduction activities and climate-proofing of existing assets (such as building or maintaining local infrastructure to higher standards of ­resilience). This includes shelter belts and mangrove plantations, building cyclone shelters, and raising embankment h ­ eights. • Enhancing agriculture-based livelihoods by building, among others, irrigation channels, livestock shelters, and water and grain storage s­ tructures. In Ethiopia, the public works component of the flagship safety net invests in building the resilience of communities to climatic ­shocks. Public works focus on creating community assets to reverse the severe degradation of watersheds and to provide more reliable water supply under different climatic c­ onditions. Its community-based Participatory Watershed Management Planning Process

FIGURE ­1.6

Productive inclusion model: Graduation into sustainable livelihoods Extreme poverty

Sustainable livelihoods Mentoring Seed capital/ employment

Training Access to financial services Consumption assistance Market/value chain analysis Targeting Start Source: PEI ­2016.

Month 3

Month 6

Month 24

Month 36

Programs: Design Considerations for Building Resilience | 51

BOX ­1.4

Sahel: Productive inclusion to support the adaptive capacity of the extreme poor In the Sahel, traditional cash transfers are a core component of many social protection s­ ystems. However, policy makers in the region are interested in complementing these programs to improve the ability of recipient households to become more productive and by extension more resilient to climatic ­shocks. Burkina Faso, Chad, Mali, Mauritania, Niger, and Senegal are implementing productive inclusion packages, including the following interventions: •

•

•

Basic consumption ­support. Beneficiaries receive regular cash transfers of about $15 per ­month. Coaching and group ­formation. Beneficiaries form groups to meet with coaches on an ongoing ­basis. Coaches provide support and guidance to the groups, ensure that program components are being implemented effectively, and promote group access to local ­markets. They also provide individualized follow-up when participants need additional ­support. Savings ­groups. Beneficiaries form groups to pool their ­savings. The intervention trains and

•

•

•

•

supervises participants to establish and manage Village Savings and Loans Associations for a source of lending ­funds. Workshops. Beneficiaries and the wider ­community attend a screening of a video and engage in a facilitated ­after-discussion. The aim is to lift aspirations and to address community norms that prevent beneficiaries from making productive ­investments. Life-skills ­training. Beneficiaries are offered a week-long training program designed to promote socio-emotional skills such as self-­ esteem, effective decision-making, and conflict ­resolution. Micro-entrepreneurship ­training. Beneficiaries receive a week-long ­training program covering basic business skills for agricultural and nonagricultural ­activities. Large, lump-sum cash ­grant. Beneficiaries receive a one-time grant of $100–$200 meant to promote investment in productive ­activities.

Source: Karlan et ­al. ­2017.

targets areas and p ­ rojects. These projects have led to re-greening in previously arid areas and have contributed to the rehabilitation of entire community ­watersheds. These positive environmental impacts are enhancing agriculture productivity and livelihoods (World Bank ­2013b). Similarly, a series of case studies of India’s Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) found that it can help to build and strengthen resilience to various climate ­shocks. The MGNREGS was found to do so by providing (1) integrated natural resource management and soil conservation infrastructure, such as check dams, ponds and trenches, afforestation, and land development works; (2) agriculture-based investments, such as irrigation channels; and (3) other local i­ nfrastructure. The authors use household surveys to construct both a livelihood vulnerability index and an agriculture vulnerability index, finding that vulnerability falls across the board because of the MGNREGS assets and in particular the natural resource–based assets constructed (Esteves et ­al. 2013; Kaur et ­al. ­2017). Projects can similarly support wider disaster risk reduction activities and climate-proofing of existing ­a ssets. They can be designed to enhance ­agriculture-based livelihoods by producing irrigation channels, livestock

52 | Adaptive Social Protection

shelters, and water and grain storage structures, among others (ADB 2018; Asfaw and Davis 2018; Steinbach et ­al. 2017a, 2017b, 2017c; Subbarao et ­al. ­2013). Further results are synthesized in box ­1.5. Lastly, safety net programs that contribute to building human capital can help to equip future generations with the capacity to adapt to ­s hocks. Promoting the accumulation of human capital among poorer households can be critical in terms of connecting those households with the skills to adapt over the long t­ erm. Indeed, human capital can empower the next generation

BOX ­1.5

India: Rural employment guarantee scheme creates assets to manage climate risk Drawing on field research on India’s Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) and specific climate-related shocks in different states (winter drought in Sikkim, cyclones in Andhra Pradesh, drought in Jharkhand, and flooding and drought in Odisha), the following insights emerge on the assets/infrastructure constructed: •

•

In Sikkim, Dhara Vikas Program (local) and the MGNREGS Springshed Development Program (national) have recharged groundwater, increasing discharge rates from ­4.4 to ­14.4 liters per minute in ­2010/11. This gave households 10–15 ­percent more water for consumptive and productive use during the lean ­period. Other private assets created through MGNREGS include cardamom and broom plantations, livestock sheds, water storage tanks, and infrastructure to produce organic ­fertilizer. The resulting improvements to soil quality and irrigation have increased arable land and enabled the introduction of new crops, while the new structures have contributed to further improving agricultural ­production. This has led to an average 18 ­percent increase in crop ­yields. Public assets such as roads and rural marketing centers, on the other hand, have increased access to ­markets. In Andhra Pradesh, most projects focused on irrigation, land productivity improvements, horticulture and sericulture plantations, and developing fallow lands, such as into coconut

•

•

plantations for landless ­households. Increased access to water for irrigation and reduced flooding in adjacent farmland during and after cyclones has reduced the sensitivity of farmers’ livelihoods to climate ­impacts. For example, decreased waterlogging in the paddy fields and increased availability of water for irrigation have improved agricultural production from 15–17 to 25–30 bags of rice per ­acre. Moreover, the amount of land under cultivation for a second paddy crop in the rabi (winter) season doubled or tripled from 50 to 100–150 ­acres. In Jharkhand, the creation of dobas (farm ponds), dug wells, and irrigation pumps resulted in 33 ­percent of respondents reporting improvements in water availability and irrigation, 29 ­percent increased availability of agricultural land, 21 ­percent increased agricultural productivity, and 46 ­percent increased crop ­diversity. Mixed results across the region largely depended on program implementation, such as construction of dobas that did not adhere to ­specifications. In Odisha, a focus on horticultural services, such as the creation of guava, cashew, and mango plantations, and irrigation resulted in 90 ­percent reporting improvements in agricultural production, 85 ­percent increased crop diversity, and 40 ­percent increased irrigation and availability of agricultural ­land.

Sources: Barca 2018, drawing on Kaur et ­al. 2017 and Steinbach et ­al. 2017a, 2017b, and ­2017c.

Programs: Design Considerations for Building Resilience | 53

with the means to move out of at-risk areas and toward employment opportunities in lower-​risk livelihoods. To encourage the accumulation of human capital among beneficiaries, safety net benefits often come with conditions such as, most prominently, those aligned to conditional cash transfer programs encouraging investments in the health and education of a beneficiary households’ c­ hildren. In cases where the capacity to monitor compliance with “hard” conditions in conditional cash transfer programs may be lower, softer conditions are increasingly being i­mplemented. For example, behavioral change sessions, highlighted for their role in supporting beneficiary preparedness, are increasingly accompanying cash transfer programs, ­delivered in the community to transmit information on health, nutrition, and ­education to ­beneficiaries.

NOTES 1. Program exit refers to exclusion from the program of those who have passed away or no longer qualify (based on predetermined program c­ riteria). Importantly, program exit does not depend on a participant’s behavior or economic status, as is the case for graduation (Samson ­2015). 2. The six countries are Ethiopia, Ghana, Kenya, Lesotho, Malawi, and ­Zambia. 3. These are sometimes classified as “soft resilience measures” and typically are low-cost and adaptable to deliver benefits in changing conditions (Cabot Venton et ­al. ­2012).

REFERENCES ADB (Asian Development ­Bank). ­2018. “Strengthening Resilience through Social Protection ­Programs.” Guidance ­Note. ADB, ­Manila. Andrews, C ­ ., A ­ . Hsiao, and L ­ . ­Ralston. 2 ­ 018. “Social Safety Nets Promote Poverty Reduction, Increase Resilience, and Expand ­Opportunities. In Realizing the Full Potential of Social Safety Nets in Africa, edited by K ­ . Beegle, A ­ . Coudouel, and ­E. Monsalve, 87–137. Washington, DC: World ­Bank. Arnold, ­C., ­T. Conway, and ­M. ­Greenslade. ­2011. “Cash ­Transfers.” Evidence ­Paper. Department for International Development, L ­ ondon. h ­ ttps://webarchive.nationalarchives.gov.uk​ /+/http://www.dfid.gov.uk/Documents/publications1/cash-transfers-evidence-paper.pdf. Asfaw, ­S., and ­B. ­Davis. ­2018. “Can Cash Transfer Programmes Promote Household Resilience? Evidence from Sub-Saharan ­Africa.” In Climate Smart Agriculture, edited by ­L. Lipper, ­N. McCarthy, ­D. Zilberman, ­S. Asfaw, and ­G. Braca, ­227–50. Rome: Food and Agriculture Organization of the United ­Nations. ­http://www.fao.org/3/a-i7931e.pdf. Asfaw, S ­ ., B ­ . Davis, J ­ . Dewbre, S ­ . Handa, and ­P. W ­ inters. 2 ­ 014. “Cash Transfer Programme, Productive Activities and Labour Supply: Evidence from a Randomized Experiment in ­Kenya.” Journal of Development Studies 50 (8): ­1172–96. Atkinson, ­T. ­2009. “A Stress Test for the Welfare ­State.” In Aftershocks: Economic Crisis and Institutional Choice, edited by ­A . Hemerijck, ­B. Knapen, and ­E. Van Doorne, 2 ­ 07–11. Amsterdam: Amsterdam University ­Press. Bahadur, ­A., ­K. Peters, ­E. Wilkinson, ­F. Pichon, ­K. Gray, and ­T. ­Tanner. ­2015. “The 3As: Tracking Resilience across ­BRACED.” Working Paper, Building Resilience and Adaptation to Climate Extremes and Disasters (BRACED), ­London. Banerjee, ­A., ­E. Duflo, ­N. Goldberg, ­D. Karlan, ­R. Osei, ­W. Parienté, ­J. Shapiro, ­B. Thuysbaert, and ­C. ­Udry. ­2015. “A Multifaceted Program Causes Lasting Progress for the Very Poor: Evidence from Six ­Countries.” Science 348 (6236): ­1260799. Barca, ­V. ­2018. “Setting the Scene: Building on Existing Systems for Adaptive Social P ­ rotection.” Background Paper, Oxford Policy Management, Oxford, ­UK.

54 | Adaptive Social Protection

Barca, ­V., S ­ . Brook, J ­ . Holland, and M ­ . ­Otulana. 2 ­ 015. Qualitative Research and Analysis of the Economic Impacts of Cash Transfer Programmes in Sub-Saharan Africa, Synthesis ­Report. Rome: Food and Agriculture Organization of the United ­Nations. Barca, ­V., and ­C. ­O’Brien. ­2017. “Factors Affecting the Usefulness of Existing Social Protection Databases in Disaster Preparedness and ­Response.” Policy ­Brief. Oxford Policy Management, Oxford, ­UK. ­https://assets.publishing.service.gov.uk/media/5a942c50ed915d57d4d0ef98​ /­Policy-Brief-Factors-affecting-usefulness-existing-social-protection-databases.pdf. Bastagli, F ­.2 ­ 014. “Responding to a Crisis: The Design and Delivery of Social P ­ rotection.” ODI Working Paper, Overseas Development Institute, ­London. Bastagli, ­F., ­J. Hagen-Zanker, L ­ . Harman, ­V. Barca, ­G. Sturge, and ­T. S ­ chmidt. 2 ­ 016. “Cash Transfers: What Does the Evidence Say? A Rigorous Review of Programme Impact and of the Role of Design and Implementation F ­ eatures.” Overseas Development Institute, L ­ ondon. https://www.odi.org/sites/odi.org.uk/files/resource-documents/11316.pdf. Bastagli, ­F., and ­R. ­Holmes. ­2014. “Delivering Social Protection in the Aftermath of a Shock: Lessons from Bangladesh, Kenya, Pakistan, and Viet ­Nam.” Overseas Development Institute, ­London. Beazley, R. 2017a. “Study on Shock-Responsive Social Protection in Latin America and the Caribbean: Dominican Republic Case Study.” Oxford Policy Management, Oxford, UK. https://www.opml.co.uk/files/Publications/a1537-shock-responsive-social-protection​ -latin-america-caribbean/dominican-republic-case-study.pdf?noredirect=1. Beazley, ­R. ­2017b. “Study on Shock-Responsive Social Protection in Latin America and the Caribbean: Ecuador Case ­Study.” Oxford Policy Management, Oxford, ­UK. Beazley, ­R., and M ­ . ­Farhat. ­2016. “How Can Lump-Sum Cash Transfers Be Designed to Improve Their Productive ­Potential.” Working Paper, Oxford Policy Management, Oxford, ­UK. Beazley, ­R., ­A. McCord, and ­A. ­Solórzano. ­2016. “Public Works Programmes for Protection and Climate Resilience: Theory of Change and Evidence in Low-Income ­Countries.” One Pager ­335. International Policy Centre for Inclusive Growth, B ­ rasilia. Beazley, ­R., ­A. Solórzano, and ­V. ­Barca. ­2019. “Study on Shock-Responsive Social Protection in Latin America and the Caribbean: Summary of Key Findings and Policy ­Recommendations.” Oxford Policy Management, Oxford, ­UK. Beegle, ­K., ­A. Coudouel, and ­E. M ­ onsalve. ­2018. Realizing the Full Potential of Social Safety Nets in ­Africa. Washington, DC: World ­Bank. ­http://documents.worldbank.org/curated​ /­en/657581531930611436/pdf/128594-PUB-PUBLIC.pdf. Bossuroy, ­T., and P ­ . ­Premand. ­2016. “Boosting Productive Inclusion and Resilience of the Poor: Perspectives from the Sahel Adaptive Social Protection ­Program.” ­Presentation. World Bank, Washington, DC. Bowen, ­T. ­2015. “Social Protection and Disaster Risk Management in the Philippines: The Case of Typhoon Yolanda ­(Haiyan).” Policy Research Working Paper 7482, World Bank, Washington, ­DC. ­http://documents.worldbank.org/curated/en/681881468181128752/pdf​ /­WPS7482.pdf. Cabot Venton, ­C., ­C. Fitzgibbon, ­T. Shitarek, ­L. Coulter, and ­O. ­Dooley. ­2012. “The Economics of Early Response and Disaster Resilience: Lessons from Kenya and Ethiopia: Economics of Resilience Final ­Report.” Department for International Development, ­London. Coll-Black, ­S., ­C. Holmemo, ­J. Sandford, and ­W. ­Soer. ­Forthcoming. From Programs to Systems: A Decade of Lessons Learned from Designing and Implementing Ethiopia’s Productive Safety Net ­Program. Washington, DC: World ­Bank. Cull, ­R., ­T. Ehrbeck, and ­N. ­Holle. ­2014. “Financial Inclusion and ­Development. Recent Impact ­Evaluation.” Focus Note ­92. Consultative Group to Assist the Poor, Washington, ­DC. Daidone, ­S., ­L. Pellerano, ­S. Handa, and ­B. ­Davis. ­2015. “Is Graduation from Social Safety Nets Possible? Evidence from Sub-Saharan ­Africa.” IDS Bulletin 46 (2): ­93–102. Dammert, A ­ ., ­J. ­J. de Hoop, ­E. Myukiyehe, and ­F. ­C. ­Rosati. ­2018. “Effects of Public Policy on Child ­Labor. Current Knowledge, Gaps, and Implications for Program ­Design.” World Development 110: ­104–23.

Programs: Design Considerations for Building Resilience | 55

Davis, ­B. ­2014. “The Impact of Social Cash Transfers on Labour Market Outcomes: The Evidence from Sub-Saharan ­Africa.” International Seminar and Policy Forum, Instituto de Pesquisa Econômica Aplicada Headquarters, Brasilia, September ­2014. del Ninno, ­C., ­F. Pierre, and ­S. ­Coll-Black. ­2016. Social Protection Programs for Africa’s ­Drylands. Washington, DC: World ­Bank. ­http://documents.worldbank.org/curated/en​/ ­736221​ 471343475745/pdf/107854-PUB-PUBLIC-PUBDATE-8-9-16.pdf. Deryugina, ­T. ­2016. “The Fiscal Cost of Hurricanes: Disaster Aid versus Social I­ nsurance.” American Economic Journal: Economic Policy 9 (3): ­168–98. Esteves, T., K. V. Rao, B. Sinha, S. S. Roy, B. B. Rai, I. B. Rao, N. Sharma, S. Rao, V. Patil, I. K. Murthy, J. Srinivasan, R. K. Chaturvedi, J. Sharma, S. K. Jha, S. Mishra, A. B. Singh, H. S. Rakhroy, S. Rai, R. Sharma, S. Schwan, K. Basu, N. Guerten, I. Porsché, N. Ranjan, K. K. Tripathy, and N. H. Ravindranath. 2013. “Environmental Benefits and Vulnerability Reduction through Mahatma Gandhi Rural Employment Guarantee Scheme (NREGS): Synthesis Report.” Ministry of Rural Development, Government of India and Deutsche Gesellschaft für Internationale Zusammenarbeit (GIZ), New Delhi. Ethiopia, Ministry of Agriculture. ­2014. “Productive Safety Net Programme Phase IV, Programme Implementation ­Manual.” Ministry of Agriculture, Addis ­Ababa. ­https://www.usaid.gov​ /­sites/default/files/documents/1866/psnp_iv_programme_implementation_manual_14​ _­dec_14.pdf. Fallavier, P ­. 2 ­ 014. “Social Protection in the Sahel’s Drylands: A Review of the Literature on Climate-Adaptive Social Protection for Use in the ­Sahel.” World Bank, Washington, ­DC. Fan, J., and I. Gijbels. 1996. Local Polynomial Modelling and Its Applications. New York: Chapman and Hall. FAO (Food and Agriculture Organization of the United N ­ ations). ­2016. Adapting Agriculture to Climate C ­ hange. FAO’s Work on Climate Change A ­ daptation. Rome: F ­ AO. ­http://www.fao​ .­org/3/a-i6273e.pdf. FAO (Food and Agriculture Organization of the United Nations). 2 ­ 019. Managing Climate Risks through Social Protection, Reducing Rural Poverty and Building Resilient Agricultural ­Livelihoods. Rome: ­FAO. FNS (Food and Nutrition ­Service). ­2014. “Disaster SNAP Guidance: Policy Guidance, Lessons Learned, and Toolkits to Operate a Successful ­D-SNAP.” ­U.S. Department of Agriculture, Washington, ­DC. ­https://fns-prod.azureedge.net/sites/default/files/D-SNAP_handbook_0.pdf. FSIN (Food Security Information Network). ­2015. “Measuring Shocks and Stressors as Part of Resilience ­Measurement.” Technical Series ­5. Resilience Measurement Technical Working Group, FSIN Secretariat, World Food ­Programme. ­http://www.fsincop.net/fileadmin/user​ _upload/fsin/docs/resources/1_FSIN​_­TechnicalSeries_5.pdf. Gentilini, U ­ . ­2019. “By When Would Universal Social Protection Be Achieved?” Blog post, January 1­7. Let’s Talk Development, World B ­ ank. h ­ ttps:// blogs.worldbank.org​ /­developmenttalk/when-would-universal-social-protection-be-achieved. Grosh, ­M., ­C. Andrews, ­R. Quintana, and ­C. ­Rodriguez-Atlas. ­2011. “Assessing Safety Net Readiness in Response to Food Price ­Volatility.” SP Discussion Paper 1118, World Bank, Washington, ­DC. ­h ttps://www.securenutrition.org/sites/default/files/resources​ /­attachment/english/WBG_Assessing%20safety%20net%20readiness_food%20price%20 volatility.pdf. ­ ., ­C. del Ninno, ­E. Tesliuc, and A ­ . ­Ouerghi. 2 ­ 008. For Protection and Promotion: The Grosh, M Design and Implementation of Effective Safety ­Nets. Washington, DC: World ­Bank. ­https:// siteresources.worldbank.org/SPLP/Resources/461653-1207162275268/For_Protection​ _­and_Promotion908.pdf. Hallegatte, ­S., ­M. Bangalore, L ­ . Bonzanigo, M ­ . Fay, ­T. Kane, U ­ . Narloch, J ­ . Rozenberg, ­D. Treguer, and ­A. ­Vogt-Schlib. ­2016. Shock Waves: Managing the Impacts of Climate Change on P ­ overty. Washington, DC: World ­Bank. Haushofer, ­J., and ­J. ­Shapiro. ­2013. “Household Response to Income Changes: Evidence from an Unconditional Cash Transfer Program in K ­ enya.” Massachusetts Institute of Technology, Cambridge, ­MA.

56 | Adaptive Social Protection

Hidrobo, ­M., ­J. Hoddinott, ­J. Kumar, and M ­ . ­Oliver. ­2018. “Social Protection, Food Security, and Asset ­Formation.” World Development 101: ­88–103. Hill, ­R., ­E. Skoufias, and ­B. ­P. ­Maher. ­2019. The Chronology of Disaster: A Review and Assessment of the Value of Acting Early on Household ­Welfare. Washington, DC: World ­Bank. ­http:// documents.worldbank.org/curated/en/796341557483493173/pdf/The-Chronology-of-a​ -Disaster-A-Review-and-Assessment-of-the-Value-of-Acting-Early-on-Household​ -Welfare.pdf. IEG (Independent Evaluation G ­ roup). 2 ­ 011. Social Safety Nets: An Evaluation of World Bank Support, ­2000–2010. Washington, DC: World ­Bank. Isik-Dikmelik, ­A . ­2012. “Do Social Benefits Respond to Crises? Evidence from Europe and Central Asia during the Global ­Crisis.” Social Protection and Labor Discussion Paper ­1219. World Bank, Washington, ­DC. ­ h ttp://documents.worldbank.org/curated​ /­en/709131468283465435/pdf/NonAsciiFileName0.pdf.23EDWS ISPA (Inter-Agency Social Protection A ­ ssessments). 2 ­ 017. Social Protection Payment Delivery Mechanisms: What Matters Guidance N ­ ote. Washington, DC: World B ­ ank. ­https://ispatools​ .org/tools/payments-what-matters.pdf. Jorgensen, S ­ ., and P ­.S ­ iegel. 2 ­ 019. “Social Protection in an Era of Increasing Uncertainty and Disruption: Social Risk Management 2 ­ .0.” Social Protection and Jobs Discussion Paper 1930, World Bank, Washington, ­DC. ­ h ttp://documents.worldbank.org/curated​ /­en/263761559643240069/pdf/Social-Protection-in-an-Era-of-Increasing-Uncertainty​ -and-Disruption-Social-Risk-Management-2-0.pdf. Kanbur, ­R. ­2009. “Systemic Crises and the Social Protection System: Three Proposals for World Bank ­Action.” Cornell University, Ithaca, ­NY. Kardan, A ­ ., S ­ . Bailey, A ­ . Solórzano, and L ­ .F ­ idalgo. 2 ­ 017. “Shock-Responsive Social Protection Systems Research Case Study: ­Mozambique.” Oxford Policy Management, Oxford, ­UK. Kaur, N., D. Steinbach, A. Agrawal, C. Manuel, S. Saigal, A. Panjiyar, C. Shakya, and A. Norton. 2017. “Building Resilience to Climate Change: MGNREGS and Climate-Induced Droughts in Sikkim.” IIED Issue Paper. International Institute for Environment and Development, London. Karlan, D. W. Parienté, C. Udry, T. Bossuroy, M. Goldstein, H. Kazianga, P. Premand, and J. Vaillant. 2017. “Promoting Productive Inclusion and Resilience in National Safety Nets: A Four-Country Evaluation in the Sahel.” Abdul Lateef Jameel Poverty Action Lab, J-PAL. Cambridge, MA. https://www.povertyactionlab.org/evaluation/promoting-productive​ -inclusion-and-resilience-national-safety-nets-four-country. Knippenberg, ­E., and ­J. ­Hoddinott. ­2017. “Shocks, Social Protection, and Resilience: Evidence from ­Ethiopia.” Ethiopia Strategy Support Program Working Paper 109, International Food Policy Research Institute, Washington, ­DC. Lindert, ­K., ­T. George, ­I. Rodriguez-Caillava, and Kenichi N ­ ishikawa. ­Forthcoming. A Sourcebook on the Foundations of Social Protection Delivery ­Systems. Washington, DC: World ­Bank. Macours, K ­ ., ­P. Premand, and ­R. ­Vakis. 2 ­ 012. “Transfers, Diversification and Household Risk Strategies: Experimental Evidence with Lessons for Climate Change A ­ daptation.” Policy Research Working Paper 6053, World Bank, Washington, ­DC. ­http://documents.worldbank. org/curated/en/275241468340175496/pdf/WPS6053.pdf. Mansur, ­A., ­J. Doyle, ­J. Gerome, and ­O. ­Ivaschenko. ­2017. “Social Protection and Humanitarian Assistance Nexus for Disaster Response: Lessons Learnt from Fiji’s Tropical Cyclone ­Winston.” Social Protection and Labor Discussion Paper ­1701. World Bank, Washington, ­DC. ­http://documents.worldbank.org/curated/en/143591490296944528/pdf/113710-NWP​ -PUBLIC-P159592-1701.pdf. Mariotti, C ­ ., M ­ . Ulrichs, and L ­.H ­ arman. 2 ­ 016. “Sustainable Escapes from Poverty through Productive Inclusion: A Policy Guide on the Role of Social ­Protection.” CPAN Policy Guide ­7. Chronic Poverty Advisory Network, ­London. Marzo, ­F., and ­H. ­Mori. ­2012. “Crisis Response in Social ­Protection.” Social Protection and Labor Discussion Paper ­1205. World Bank, Washington, ­DC. ­http://siteresources.worldbank​ .org​ / ­S OCIALPROTECTION/Resources/SP-Discussion-papers/430578-1331​ 5 085​ 52354/1205​.pdf.

Programs: Design Considerations for Building Resilience | 57

Mauritania, Commissariat a la Securite Aimentare (CSA). 2019. “Programme de filets sociaux reactifs aux chocs Elmaouna: Manuel operationnel,” Version 2.0. McCord, ­A . ­2013. “Review of the Literature on Social Protection Shock Responses and ­Readiness.” ODI ­Shockwatch. Overseas Development Institute, ­London. McCord, A ­ ., ­R. Beazley, A ­ . Solórzano, and L ­ .A ­ rtur. 2 ­ 016. “ICF Social Protection and Climate Change in Mozambique with a Focus on the Role of the PASP: Feasibility and Design ­Consultancy.” Final ­Report. Overseas Development Institute, ­London. Mexico, Government of. ­2018. “Rules of Operation: Social Inclusion Program PROSPERA (Reglas de operación de prospera programa de inclusión s­ ocial).” Government of Mexico: Mexico ­City. O’Brien, C ­ ., J ­ . Congrave, K ­ . Sharp, and N ­ .K ­ eïta. 2 ­ 018a. “Shock-Responsive Social Protection Systems Research: Case Study—Social Protection and Humanitarian Responses to Food Insecurity and Poverty in ­Mali.” Oxford Policy Management, Oxford, ­UK. O’Brien, ­C., ­R. Holmes, ­Z. Scott, and ­V. ­Barca. ­2018b. “Shock-Responsive Social Protection Systems Toolkit—Appraising the Use of Social Protection in Addressing Largescale S ­ hocks.” Oxford Policy Management, Oxford, ­UK. OPM (Oxford Policy ­Management). ­2017. Shock-Responsive Social Protection Systems Research: Literature Review, 2nd ­edition. Oxford, UK: ­OPM. PEI (Partnership for Economic I­ nclusion). ­2016. “Increasing the Income Earning Opportunities of Poor and Vulnerable P ­ eople.” World Bank, Washington, ­DC. h ­ ttps://www​ .­microfinancegateway.org/sites/default/files/announcement/pei-brochure.pdf. Roelen, ­K., ­S. Devereux, ­A. ­G. Abdulai, ­B. Martorano, ­T. Palermo, and ­L. ­P. ­Ragno. ­2017. “How to Make ‘Cash Plus’ Work: Linking Cash Transfers to Services and S ­ ectors.” Innoncenti Working Paper ­2017-10. United Nations Children’s Fund Office of Research, ­Florence. Samson, ­M. ­2015. “Exit or Developmental Impact? The Role of ‘Graduation’ in Social Protection ­Programmes.” IDS Bulletin 46: ­13–24. Solórzano, ­A. ­2016. “Can Social Protection Increase Resilience to Climate Change? A Case Study of Oportunidades in Rural ­Yucatan, Mexico.” IDS Working Paper 465, Centre for Social Protection and Institute of Development ­Studies, Brighton, UK. Steinbach D., N. Kaur, C. Manuel, S. Saigal, A. Agrawal, A. Panjiyar, and A. Barnwal. 2017a. “Building Resilience to Climate Change: MGNREGS and Drought in Jharkhand.” IIED Issue Paper, IIED London. Steinbach, D., N. Kaur, C. Manuel, S. Saigal, A. Agrawal, A. Panjiyar, and A. Barnwal. 2017b. “Building Resilience to Climate Change: MGNREGS and Cyclones in Andhra Pradesh.” IIED Issue Paper, IIED London. Steinbach, D., N. Kaur, C. Manuel, S. Saigal, A. Agrawal, and A. Panjiyar. 2017c. “Building Resilience to Climate Change: MGNREGS, Drought and Flooding in Odisha.” IIED Issue Paper, IIED, London. Subbarao, Kalanidhi, Carlo del Ninno, Colin Andrews, and Claudia Rodríguez-Alas. 2013. Public Works as a Safety Net: Design, Evidence, and Implementation. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/11882. Tenzing, ­J. ­D. ­2019. “Integrating Social Protection and Climate Change Adaptation: A ­Review.” WIREs Climate Change 11 (2): ­e626. Ulrichs, ­M., and ­R. ­Slater. ­2016. “How Can Social Protection Build Resilience? Insights from Ethiopia, Kenya and ­Uganda.” Working Paper, Building Resilience and Adaptation to Climate Extremes and Disasters (BRACED), ­London. WFP (World Food ­Programme). ­2015. “Policy on Building Resilience for Food Security and ­Nutrition.” W ­ FP/EB.A/2015/5-C. WFP, ­Rome. https://documents.wfp.org/stellent/groups​ /­public/documents/eb/wfpdoc063833.pdf?_ga=2.20959473.817428444.1582152603-7527​ 67465.1554223343. WFP (World Food Programme). ­2018. “Scaling Up for Resilient Individuals, Communities and Systems in the ­Sahel.” Fact ­Sheet. WFP, ­Rome. ­https://docs.wfp.org/api/documents/WFP​ -0000110238/download/?_ga=2.26639604.817428444.1582152603-752767465.1554223343.

58 | Adaptive Social Protection

World ­Bank. ­2013. Building Resilience to Disaster and Climate Change through Social Protection: Synthesis ­Note. Washington, DC: World ­Bank. ­http://documents.worldbank.org/curated​ /­en/187211468349778714/pdf/796210WP0Build0Box0377381B00PUBLIC0.pdf. World Bank. ­2018a. Fundamentals of Disaster Risk ­Finance. Disaster Risk Financing and ­Insurance. Washington, DC: World Bank, Open Learning ­Campus. World Bank. ­2018b. “Second Niger Adaptive Safety Net Project: Project Appraisal ­Document.” World Bank, Washington, ­DC. ­http://documents.worldbank.org/curated/en​/ 77793​ 1546830037773​/­Niger-Second-Niger-Adaptive-Safety-Net-Project.

2

Data and Information

UNDERSTANDING RISK AND HOUSEHOLD VULNERABILITY

OVERVIEW Data and information strengthening are at the core of the adaptive social protection (ASP) agenda. That is, fundamentally, there is a need to invest in a stronger understanding of risk and household vulnerability to shocks within the social protection (SP) sector in order to understand who is likely to be most at-risk to which kinds of shocks, as a basis for designing appropriate programs. This will require a multisectoral approach to the collection, sharing, and analysis of data. Concretely, the insights provided from the analyses typically undertaken by the disaster risk management (DRM) community can contribute to an improved understanding of spatial disaster risk as a function of the historical hazard incidence and the exposure and vulnerability of assets and people. Integrating these analyses with the data on household-level poverty and vulnerability routinely undertaken as the basis for safety net program design and held in SP information systems promises to provide a more complete picture of household-level vulnerability to shocks. To this end, the information systems that inform the delivery of SP programs require strengthening for ASP. In turn, these information systems themselves will become more capable of generating increasingly relevant data to refine analyses of household risk and vulnerability and programs that are more capable of building resilience. First and foremost, social registries that hold valuable information on beneficiaries and nonbeneficiaries will enable safety net programs to identify those most vulnerable to shocks when they expand into, and deepen their coverage within high-risk areas. Moreover, the closer integration of social registries with those information systems leveraged by the DRM and humanitarian sectors will further sensitize the SP information system, informing ASP programs. Critically, fostering operational linkages with early warning systems alongside tools that enable the rapid assessment of post-shock needs will be transformative in predicting household needs and assessing them after a shock. To do so, considerable investment in and coordination with actors and information systems from outside of the SP sector will be required to strengthen information for ASP.

Data and information

This chapter draws extensively from a background paper by Beazley and Barca (2018).  59

60 | Adaptive Social Protection

INVESTING IN AN IMPROVED UNDERSTANDING OF RISK AND HOUSEHOLD VULNERABILITY Information on household vulnerability to disasters is crucial for the design and implementation of ASP programs. What kinds of hazards does the country face? How frequently? Where? Which assets and population groups are exposed, and among them which are the most vulnerable? What is their capacity to cope? The analysis of disaster risk is a core pillar of operational work conducted in the DRM sector. ASP programs will need to draw on these analyses and assessments of disaster risk and to integrate them with assessments of household poverty, vulnerability to poverty, and the relative ability for households to cope with shocks. Household risk and vulnerability analyses attempt to better understand the impact of shocks on poverty. Recent studies have used weather data as objective measures of shocks and have analyzed impacts on poverty (Gao and Mills 2018; Hill and Porter 2017; Skoufias et al. 2019). Risk and vulnerability analyses carried out under the first phases of the Sahel ASP Program in Burkina Faso, Chad, Niger, and Senegal focused on developing such analyses to identify the populations most vulnerable to climatic shocks. At the same time, resilience to shocks often has been measured as the ability to not become poor or food insecure in the face of adverse shocks based on factors such as distance from the poverty line, probability of exposure to shocks, and the impact of shocks on consumption-based measures of household well-being (see, for example, Gao and Mills 2018). Such analyses provide an informed, needs-based foundation for ASP policy dialogue and program design. (For a discussion of information needs for targeting assistance ­following shocks, see, for example, del Ninno and Mills 2015.) Risk and vulnerability analysis represents a clear point of intersection and complementarity for both SP and DRM. Traditionally, the DRM sector has focused on analyzing the vulnerability of assets and physical structures (such as buildings and roads) to disasters, whereas the SP community has focused on the analysis of household poverty (often as a proxy for vulnerability to shocks) and vulnerability to poverty as an outcome from shocks. Integrated SP and DRM analyses hold the potential for generating a deeper understanding of disaster risk, household exposure, and household impacts which are often highly heterogenous. As an example, in emphasizing the vulnerability of poor households to disasters, Hallegatte et al. (2017) developed a variable for measuring disaster risk called “socioeconomic resilience” that encompasses the effects of asset losses on household income, consumption, and well-being. Based on estimates of socioeconomic resilience in 117 countries and including in the analysis how poverty and lack of capacity to cope with disasters magnify losses in well-being, the authors suggest that the effects of floods, wind storms, earthquakes, and tsunamis on well-being are equivalent to a $520 billion drop in consumption—60 ­percent more than the widely reported assessment of asset losses. The measure also accounts for increases in household resilience that accompany the introduction and expansion of ASP in a country (see box 2.1).

Data and Information: Understanding Risk and Household Vulnerability | 61

BOX 2.1

Analyzing socioeconomic resilience and the potential impact of ASP Socioeconomic resilience models the effects of asset losses on income, consumption, and well-being at the household level. It also measures the expected increases to household resilience that accompany the development (introduction and expansion) of social protection in a country. Socioeconomic resilience demonstrates the value of joining vulnerability analyses related to social protection and disaster risk management (DRM), and it may be a useful starting point for a social protection–DRM engagement looking to quantify the vulnerability and resilience co-benefits of social protection systems. Socioeconomic resilience complements and extends traditional DRM analyses. Hazard, exposure, and physical vulnerability quantify asset losses but do not allow DRM interventions or social protection systems to incorporate the fact that poor households experience and recover from a given amount of asset losses differently than do wealthy households. Nor do they decode the complex factors that determine how

likely a household is to recover from its losses without assistance. Socioeconomic resilience links directly to the co-benefits of ASP engagements. Traditional DRM analyses measure the costs of interventions relative to the benefits of avoided asset losses, and therefore cannot quantify the real benefits of ASP programs— which help households to smooth income and consumption losses, accelerate recovery, and prepare for future hazards, though they do not affect asset losses. Conversely, the design and budgeting of ASP programs will in most cases need to incorporate DRM inputs—­e specially geographical and socioeconomic heterogeneity in asset and well-being risk—in order to deliver targeted, timely assistance. Sample applications of the socioeconomic resilience model in Sri Lanka and the Philippines can be found in Walsh and Hallegatte (2019a, 2019b), including the modeled impact of ASP on well-being outcomes.

Source: Hallegatte et al. 2017.

SOCIAL REGISTRIES AT THE HEART OF THE ASP INFORMATION AGENDA Beyond foundational analyses of risk and vulnerability, the global expansion of social registries is framing much of the current discussion around the ASP information agenda (Barca 2017; Bastagli 2014; Bastagli et al. 2016; IEG 2011; Kuriakose et al. 2012). Social registries are information systems that support outreach, intake, registration, and the determination of potential eligibility for inclusion in one or more social programs. Ultimately, social registries provide a “gateway” for potential inclusion of intended populations into SP programs (figure 2.1). Social registries can serve as a gateway to one or many programs. When multiple programs use a common social registry or integrated social registry, they can play an important social policy role in helping coordinate efforts to reach intended populations. This facilitates synergies across programs aiming to deliver complementary benefits and services to common groups. Recent estimates suggest that at least 60 countries have social registries, while 18 countries have beneficiary

62 | Adaptive Social Protection

FIGURE 2.1

Social registry as a gateway for multiple programs Enroll

Provide

Manage

Enroll

Provide

Manage

Enroll

Provide

Manage

Assess needs and conditions

m

1

a gr

o

Pr

Program 2

3 Pr

og

ra

m

3

Source: Lindert et al., forthcoming.

TABLE 2.1  Beneficiary

and social registries: Terminology

TERM

DEFINITION

Beneficiary registry

• Track beneficiary/benefit(s) to support program management and implementation • A household is referred to as a “beneficiary”

Integrated beneficiary registry

• Monitor and coordinate benefits • Support identification of intended/unintended duplications across programs • Better-integrate services across programs

Social registry

• Support outreach, intake, registration, and assessment of needs and conditions to determine eligibility for social programs • Contain and maintain information on all registered households regardless of beneficiary/nonbeneficiary status • A household is referred to as a “registered household”

Integrated social registry

• Monitor and coordinate across programs

Sources: Based on Barca 2017; Leite et al. 2017.

registries (table 2.1). Social registries play a key role in providing necessary ­information to operate ASP programs, such as the information base for building ­resilience and scaling up after a shock. However, the expansion of social registries alone may not meet the information requirements for ASP. Many countries operate registries with “fixed lists” of registrants and program beneficiaries, and they generally update the lists every four to five years based on a census sweep approach (figure 2.2). Thus, many static registries operating fixed lists often comprise dated information along with partial coverage. Box 2.2 compares the coverage of social registries in four countries to demonstrate the varying population shares that are more or less easily identified and reached with assistance after a shock.

Data and Information: Understanding Risk and Household Vulnerability | 63

FIGURE 2.2

Social registries: Global coverage and registration processes Pakistan NSER Dominican Republic SIUBEN Philippines Listahanan Chile RSH Colombia SISBEN Turkey ISAS Mexico SIFODE Brazil Cadastro Unico Indonesia UDB Georgia TSA Registry Montenegro SWIS Senegal RNU Yemen, Rep., SWF Registry Malawi UBR North Macedonia CBMIS China RPHR Mauritius SRM Azerbaijan VEMTAS Sierra Leone SPRINT Mali RSU 0

20

40

60

80

100

Percentage of total population in social registry Mass registration (census sweeps) Primarily on-demand registration Both Source: Leite et al. 2017. Note: CBMIS = Cash Benefits Management Information System; ISAS = Integrated Social Assistance System; NSER = National Socio-Economic Registry; RNU = Unified National Registry; RPHR = Rural Poor Household Registry; RSH = Households Social Registry; RSU = Unified Social Registry; SIFODE = Integrated System for Development; SISBEN = Beneficiary Identification System; SIUBEN = Unique Beneficiary Identification System; SPRINT = Social Protection Registry for Integrated National Targeting; SRM = Social Registry (Mauritius); SWF = Social Welfare Funds; SWIS = Social Welfare Information System; TSA Targeted Social Assistance; UBR = Unified Beneficiary Registry; UDB = Unified Data Base; VEMTAS = Electronic Application and Appointment Subsystem.

The data collected for social registries often do not differ substantially from that required for identifying households that are vulnerable to shocks. A recent study from the World Bank (Schnitzer 2018) compares two of the most widely used approaches to target poor and vulnerable households: proxy means testing, which is designed to identify the chronic poor, and the household economy approach, which is a livelihoods analysis framework. The paper finds that proxy means testing better identifies the chronic poor, and the household economy approach better identifies households suffering from seasonal food insecurity. Most importantly, it highlights that both rely largely on the same household

64 | Adaptive Social Protection

BOX 2.2

Lesotho, Mozambique, Pakistan, and the Philippines: Social registry coverage and utility for shock response The panels in figure B2.2.1, developed by Oxford Policy Management, compare and contrast social protection databases in four of the research countries, showcasing completeness of social registry data where a registry exists (the dark blue oval) and completeness of beneficiary data for one or more noncontributory social protection program(s) (the green oval). It also introduces an arbitrarily sized

group of households potentially affected by any given shock to show how existing data can be used for vertical expansion (temporary increase of the value/duration for existing beneficiaries) of the social protection scheme, horizontal expansion (temporary expansion of the caseload), or piggybacking on the social protection database to provide an emergency response.

FIGURE B2.2.1

Coverage of selected social protection databases in four countries a. Lesotho

22%

b. Mozambique

CGP beneficiaries

NISSA social registry

5%

100%

b.

Beneficiaries across all 8% a. program c.

Lesotho population

a. c.

c. Pakistan

85%

NSER social registry

d. The Philippines

100%

20%

Pakistan population

BISP beneficiaries a. b.

Social registry data

100%

Mozambique population

75%

Listahanan social registry

c.

Beneficiary data for noncontributory social protection program(s)

100%

21%

Pantawid beneficiaries

a. b.

Philippines population

c.

Households potentially affected by shock

Source: Barca and O’Brien 2017. Note: BISP = Benazir Income Support Program; CGP = Child Grants Program; NISSA = National Information System for Social Assistance; NSER = National Socio-Economic Registry. Figures do not represent the totality of social protection databases in each country. a = households that can be reached through vertical expansion or piggybacking (on the beneficiary databases); b = households that can be easily reached through horizontal expansion or piggybacking (on the social registry); c = households less easily reached through horizontal expansion or piggybacking (not covered by existing social protection databases).

information. As a result, small tweaks to the data collected for a social registry can enable easier and more accurate estimation not only households in chronic poverty but also those ­vulnerable to shocks and their capacity to cope. The additional data required for social registries depend on the shocks that affect a country, the predominant household livelihoods in high-risk areas, and the coping mechanisms utilized by poor and vulnerable households. In the case of seasonal droughts, for example, data on food insecurity and livelihoods can help to ­identify vulnerable households. In relation to livelihoods, high reliance

Data and Information: Understanding Risk and Household Vulnerability | 65

on agricultural activities can imply high vulnerability in regions prone to floods and droughts. In the case of seasonal floods, typhoons, or hurricanes, the additional data could relate to the location of households, such as living in coastal locations, or to the proximity to rivers, streams, or ravines as areas at risk of flooding as identified by DRM models. Social registries that contain geo-referenced household data can be more effective and precise in assessing household exposure to shocks, in coordination with hazard data from the DRM sector. When this information is overlaid with data from risk and vulnerability assessments and hazard risk mappings, the government will have a powerful tool to help determine who should be supported and where. In the Philippines, the Disaster Response Management Bureau utilized the social registry, Listahanan, to estimate the number of households that would be affected by a disaster given its location/path (Bowen 2015). Expanding the coverage of social registries in high-risk areas may be particularly effective in contexts of recurrent or seasonal shocks where household needs may be more predictable. In Mauritania, the government developed a methodology to determine the ideal coverage (number of households) the social registry should have in each commune to ensure it was capable of informing shock response to drought. The analysis of vulnerability to food insecurity from drought recommended including an additional 50,000 households (distributed proportionally across communes considering hazard vulnerability and exposure indexes). Kenya’s Hunger Safety Net Program (HSNP) is commonly cited in the literature as the prime example of collecting operationally relevant data in advance of a shock, allowing horizontal expansions to preidentified vulnerable households in times of need. The program has registered almost all of the households in the four participating high-risk counties—nearly 300,000—and preenrolled them, giving them bank accounts into which post-shock payments can be deposited. A key component of the consolidation process was the launch of a computerized management information system, the Single Registry, in 2016.

DATA QUALITY CONSIDERATIONS FOR ASP Ultimately, ASP requires a clear vision for improving data quality. Barca and O’Brien (2017) provide a useful framing to consider the five main dimensions of data quality in a shock-responsive setting: completeness, relevance, currency, accessibility, and accuracy. Given no data can perfectly cover all these dimensions, it is very context-specific whether ASP programs can rely on available data and information. In some contexts, data that are not completely relevant or up-to-date or do not cover the whole population affected by the shock can still be the best vehicle for providing timely support. In other contexts, the opposite may be true, with incomplete and outdated data undermining a response. Table 2.2 analyzes the five main dimensions of data ­quality—completeness, relevance, currency, accessibility, and accuracy (Barca and O’Brien 2017). The data quality of existing SP information systems for shock response can be determined by a number of factors. In line with the literature, “data quality” is here defined as data fit for use by users (Wang and Strong 1996). In this particular case, the users are the government (or humanitarian actors) involved in preparedness and response to shocks. Whereas for data from the SP sector, “the nature and quality of SP [social protection] databases and information systems is

66 | Adaptive Social Protection

TABLE 2.2  Implications

of the different dimensions of data quality for adaptive social protection IMPLICATIONS FOR USE OF EXISTING SOCIAL PROTECTION DATA

DIMENSION

DEFINITION

Completeness

Number of records compared with what would be perceived as a full set of records

• Depends on the overlap (if any) of households that are currently beneficiaries of a program and those affected by the shock. • Higher coverage and uniform coverage across geographic areas (regions, urban/ rural) is desirable.

IMPLICATIONS FOR DIFFERENT SHOCKS

Relevance

Contains the requisite variables for the intended purpose

• More useful if the data include variables that • Operational data are relevant for any can help predict vulnerability to shocks. shock. • More useful if the data include operational • Socioeconomic data may be more information; such as location, contacts, and relevant for certain shocks; such as bank-account details. economic, slow-onset, and recurrent. • In the case of nonbeneficiaries, inclusion of socioeconomic data can help prioritize post-shock support.

Currency

Degree to which data are up to date

• Data will never reflect the situation after a shock but the more up-to-date the better. • The status depends on the underlying approach to data collection (on-demand approaches are more flexible) and information management.

• Conflict or rapid-onset disasters may cause widespread internal displacement, split up households, and significantly change their material circumstances.

Accessibility

Refers to the ease for potential users

• Digital data (maintained/stored) can increase accessibility. • Data-sharing agreements need to be defined in advance. • While accessibility is critical, provisions for data security are important to ensure privacy of beneficiary data.

• The challenges of accessing a database are compounded in a conflict or rapid-onset natural disaster. • In conflicts, security concerns around the sharing of personal information are particularly acute.

Accuracy (integrity)

Data are considered to be accurate if free of errors and omissions; meaning, trustworthy

• Accuracy increases with processes to verify • Accuracy is relevant for all shocks. and validate existing data; such as • Accuracy is particularly problematic supervision and cross-checks with other where trust between actors is already databases. undermined; such as conflicts. • A function of the perceived trustworthiness of the institution responsible for collecting and housing the data. • In poverty-targeted programs, high errors of inclusion and exclusion affect perceived accuracy and may affect the usability of data for responses.

• Extensive coverage in regions affected by recurrent shocks. • Overlap of poverty and vulnerability depends on the shock; such as rapid-onset shocks, economic crises, pandemics, and conflicts are more likely to affect the nonpoor (who often are not covered by existing databases) than other shocks.

Source: Beazley and Barca 2018.

so varied that it is meaningless to ascribe a generic role to their use in shock responses, and inappropriate to assert that they will always be of use: such a role can only be identified with reference to the particularities of the database(s) in the country and context under review” (Barca and O’Brien 2017, 5). A major challenge for improving data quality is that high-frequency data are required to properly monitor and assess risk and vulnerability. This can be a challenge in low-capacity and dynamic contexts where new risks are unfolding (such as refugee inflows). For example, in Turkey, the government, with financial support from the European Union, responded to the Syrian refugee influx by scaling up the existing SP system. Although facing a language issue for registration and placing additional burdens on local officials, its strong delivery

Data and Information: Understanding Risk and Household Vulnerability | 67

infrastructure allowed for the inclusion of refugees for emergency social assistance. This process was undertaken with substantial support from the Turkish Red Crescent, a local nongovernmental organization with a strong field network. Rapid progress in the SP sector with the use of biometrics, identification, and e-payments is providing a game-changing backdrop to the ASP information agenda. Technology can broaden the potential reach of SP programs and cash ­transfers—​a strategy that works to raise productivity and resilience among ­beneficiaries in times of shock (box 2.3). In India, the Aadhaar Program uses biometric technology to make it easier for the poor to prove their identity and to authenticate the beneficiaries of dozens of social programs. The program, which covers 1.2 billion people, has provided government cost savings by improving efficiency, reducing transaction costs, and providing the infrastructure for reform—as in the case of the Pahal/LPG subsidy program (Mittal et al. 2017). Despite ongoing difficulties and concerns related to beneficiary privacy that remain to be addressed, the program has enjoyed a high rate of satisfaction (92 ­percent) and trust in the system and has been responsible for increasing financial inclusion (72 ­percent of adults have linked their bank accounts to Aadhaar) (Sonderegger et al. 2019). In Lebanon, electronic smart cards support 125,000 Syrian refugee households; the program was such a success that the government of Lebanon now uses smart cards for its social programs. There are other important opportunities to leverage technology for low-income, fragile, and conflict-affected areas. In recent years, for example, there has been growing interest in the use of remote sensing, geographic information system data, and cell phone data as potential sources to predict poverty across time and space and to improve program performance. Box 2.4 unpacks some of these innovations, their relevance, and their constraints going forward.

BOX 2.3

Pakistan: National Database and Registration Authority In Pakistan, the National Database and Registration Authority (NADRA) is responsible for civil registration. To date, the NADRA has issued a computerized national identity card in-country and abroad to nearly two-thirds of Pakistan’s 150 million citizens. NADRA provides other information technology solutions for identification, e-governance, and program implementation, with a key strength being biometric verification. Its project management unit provides a platform to integrate different systems. Over the years, it has facilitated and implemented cash transfer and social protection programs. Services provided to governmental and Source: Watson et al. 2017.

nongovernmental agencies outside of NADRA’s core business come at a cost. NADRA’s involvement in disaster response started with the 2005 earthquake. Five years later, the NADRA responded to flooding by supporting rapid computerized national identity card registration for the Watan card and delivered cash to internally displaced persons. NADRA employs more than 11,000 technical and management personnel. It has developed 365 multibiometric Interactive Registration Centers and deployed 189 mobile vans to register citizens living in remote areas.

68 | Adaptive Social Protection

BOX 2.4

Big data and technology leapfrogging: Power and limitations Policy makers in the poorest countries are often forced to make decisions based on limited data or overriding political interests. In situations where reliable data are missing, dated, or cost-prohibitive, an area of increased focus is the use of “big data” and the role of technology. Central to the discussion of big data is the potential for machine learning to inform policy making. Machine learning is the process of instructing ­c omputers to learn. It exists at the intersection of computer science, statistics, and linear algebra, with insights from neuroscience and other fields. Traditionally, machine learning has focused on m ­ aking predictions and creating structure out of unstructured data. A recent example includes the World Bank’s new partnership with the United Nations, the Red Cross, Microsoft, Google, and Amazon on the Famine Action Mechanism. The group will use artificial intelligence and machine learning to predict famine and mobilize early funding to mitigate its effects. In this context, big data have the potential to become a disruptive technology, changing traditional approaches to survey-based data collection and shifting ways of doing business on many levels. Country experience for ASP in this area is thus far limited, but a few important initiatives are relevant. •

In Rwanda, an individual’s past history of mobile-phone use can be applied to infer his or her socioeconomic status, and the predicted attributes of millions of individuals can, in turn, accurately reconstruct the distribution of wealth of an entire nation or infer the asset distribution of micro-regions composed of just a few

•

households (Blumenstock, Cadamuro, and On 2015). Another approach uses high-resolution satellite imageries to predict poverty (World Bank 2016). In Sri Lanka, the poverty estimation by the machine learning techniques using key information from the satellite imagery (such as building density, vegetation, road types, and roof types) tracks regional differences in poverty extremely well comparing with the estimation from the census. The predictions derived from satellite data generate much more plausible estimates of changes in subdistrict poverty rates than the traditional method of using household characteristics in the model.

The potential applications for this technology are promising. In resource-constrained environments where censuses and household surveys are rare, this approach creates an option for gathering localized and timely information at a fraction of the cost of traditional methods. Improved data can inform program performance through improved monitoring and response. Although promising, these nontraditional methods have caveats. One of the strongest concerns relates to privacy, and how individual, proprietary information can be misused in the process of machine learning. While rigid data protection procedures are an important mitigating factor at the program level, it is harder to control privacy practices in the public and private sectors. So too, the proliferation of different information sources and the risk of manipulation by governments and donors is a concern.

Source: Beazley and Barca 2018.

Ensuring data security and privacy are fundamental considerations that are only heightened in a crisis context. For example, sensitive information on a country’s citizens such as information stored for SP purposes—potentially including their ethnicity, religion, and more—could be swiftly put to wrong uses in cases of conflict and violence. Examples include the Rwandan genocide and the Holocaust (Seltzer and Anderson 2001). Crises require rapid decision-making and often involve multiple actors with little prior experience in coordinating (for example, there are no memorandums of understanding in place for data s­ haring). These situations can easily lead to

Data and Information: Understanding Risk and Household Vulnerability | 69

significant breaches of standard protocols for the secure collection, transfer, and storage of information (such as non-encrypted sharing of personal information via email), especially where country laws do not adhere to international data transfer and information privacy protocols. The reverse can also be true, with existing data privacy and security legislation constraining timely sharing of data across institutions. For example, in the Philippines, a new privacy law created barriers for sharing personal data in the national social registry, Listahanan, with external agencies.

FILLING THE GAPS: LINKING ASP TO EARLY WARNING INFORMATION AND POST-SHOCK NEEDS ASSESSMENTS Given the cross-sectoral nature of ASP, data and information gaps straddle across programs and systems within and outside the SP sector. Considerable ex ante work is required to set in place rules, system requirements, and approaches to design and scale up SP programs in response to a shock. This requires careful delineation of information priorities in assessing, deciding on, and implementing a shock response. It also requires the preagreement of mechanisms to determine a shock response, prominently as related to early warning systems and predefined triggers that initiate action, along with post-shock assessments of household needs. Early warning systems continue to play a critical role in providing and monitoring information for response and in triggering early action, especially in a context of growing climate-related risks.1 Early warning systems for drought— food security hybrid systems typically use a range of information on food production, access, and livelihood outcomes from national agencies and international assessments (such as the Famine Early Warning Systems Network and the Integrated Food Security Phase Classification) and merge the information into an assessment of the food-security status and likely risk (Wilkinson et al. 2018). More recently, forecasts have started using a growing range of climate information. Systems using probabilistic forecast information typically draw on products from international, regional, and national forecasting centers. Products from international and regional forecasting centers are most common, as these are freely available and considered reliable. Where appropriate, these are complemented with products from national hydrological and meteorological services. Countries are already linking SP responses to early warning information and developing index-based triggers for response, particularly for slow-onset shocks. In the Dominican Republic, the Vulnerability to Climate Hazards Index (IVACC) ­generates household-level information that supports the assessment of vulnerability to shocks and climate change and which can provide estimations of postshock needs (box 2.5). In Uganda, satellite data and a Normalized Difference Vegetation Index anomaly provide the basis for triggering earlier response to drought (box 2.6). Triggers for rapid response can be built using the data generated by existing early warning systems and climate forecasts (Bastagli and Harman 2015; O’Brien et al. 2018). Triggers are typically designed to release funds and initiate early actions when preestablished thresholds are met. These triggers can lead to automatic responses; this implies front-loading the decision-making process and directly linking climate forecasts to their potential consequences.

70 | Adaptive Social Protection

BOX 2.5

Dominican Republic: Leveraging the social registry for early warning information The Dominican Republic developed the Vulnerability to Climate Hazards Index (IVACC) to quantify the likelihood of a household being vulnerable to hurricanes, storms, and floods. The index uses data from the country’s national social registry, the Sistema Unico de Beneficiarios, which covered approximately 85.5 ­percent of the population in 2015. The IVACC produces a vulnerability score for all households in the social registry and can be aggregated to larger administrative areas (such as towns and

municipalities). It has a scale of 0–1, where households with values close to 0 are the least vulnerable and those with values close to 1 are the most vulnerable. The score is based on three variables: (1) the physical characteristics of the dwelling, (2) average household labor income, and (3) proximity of the house to a source of danger (such as river, waterbody, or stream). Using the index, figure B2.5.1 simulates the most vulnerable neighborhoods and the characteristics of vulnerable households in such areas.

FIGURE B2.5.1

Hurricane simulation using IVACC to identify households that may be most affected María Trinidad Sanchez river

Household head: Altagracia Martínez 0.5 to 1 km from the river 1 child (from 5 to 9 years of age) 1 adolescent (10-14 years of age) Spouse Concrete roof Cinder block walls IVACC: 0.524

Household head: María Gómez 0.5 km from the river 2 children (from 0 to 4 years of age) 1 adult Zinc roof Concrete walls

Source: SIUBEN (Sistema Único de Beneficiarios/Unified Beneficiary Identification System).

Source: UNDP-UNEP 2018.

Household head: Juan Pérez 0.5 km from the river 3 children (from 5 to 9 years of age) Spouse Zinc roof Palm-thatched walls

Data and Information: Understanding Risk and Household Vulnerability | 71

BOX 2.6

Uganda: Establishing satellite-based triggers for drought response In Ka ra moja, Uga nda, the Northern Uganda Social Action Fund has a cash-for-work program that is designed to respond to drought. The government has identified an objective and automatic satellite indicator as the trigger to scale up the number of households accessing the cash-forwork program. The Normalized Difference Vegetation Index (NDVI) was selected as the indicator to determine when to respond, as it simply and quickly identifies vegetated areas and their “condition” over time. The NDVI is observed on a 14-day basis, and an average score assigned for each district in Karamoja is calculated for each calendar month.

An NDVI anomaly is an early indicator of drought but is not by itself enough to declare a drought emergency. However, it does mean cash-for-work payments are triggered well before any late-onset indicators (for example, livestock mortality and malnutrition rates) reach emergency levels. This has the effect of mitigating these late-onset negative impacts and ultimately saves money in the long term. A perceived risk in responding early is that funds will be released incorrectly to situations that turn out not to be a disaster (for example, a payout is triggered; it then rains; and no drought occurs). However, the benefits of scaling social protection programs in response to very early warning indicators in the spirit of “no regrets” are becoming increasingly clear.

Sources: Cabot Venton et al. 2012; Government of Uganda 2016.

TABLE 2.3  Early

warning systems: Triggers for rapid action

ADVANTAGES

PREREQUISITES

• • • •

• The trigger’s effectiveness depends on correlation between the index and the household’s needs • Indexes should be easily measured, objective, transparent, independently verifiable, and available in a timely manner • If based on proxies rather than on observed measures, effectiveness also depends on the correlation between the proxy and the main indicator • To turn early warning into early response, contingency planning and predictive analysis are required so that early warning system data can lead to better decisions

Enables quick, even automatic, shock response Helps secure financing Increases transparency (actual or perceived) Can minimize moral hazard and adverse selection risks • Technological triggers can be less time consuming

Sources: Beazley and Barca 2018, based on Bailey 2012; Bastagli and Harman 2015; and Levine, Crosskey, and Abdinoor 2011.

Otherwise, they can be used to inform ex post, decision-making processes to trigger early action (Wilkinson et al. 2018). The use of early warning systems to trigger SP responses is promising, predicated on a number of prerequisites (table 2.3). Ultimately, data collected before a shock cannot provide an exact assessment of needs after a shock. As noted, in a post-shock environment, those most in need are not necessarily the beneficiaries of existing programs (Bastagli 2014; McCord 2013; O’Brien et al. 2018). Multiple programs will be needed (implemented by SP, DRM, and/or humanitarian actors) to provide response alongside a vertically expanded SP program (O’Brien et al. 2018). Horizontal expansion to new households based on prepositioned SP data or piggybacking on those data for a response can help enable timely response to a shock but will often need to be complemented by additional post-shock data collection. Post-shock data collection can play a key role in reflecting the socioeconomic and damage condition of potential beneficiaries, especially after

72 | Adaptive Social Protection

fast-onset, less predictable, and destructive disasters. Postdisaster household assessments often are employed to collect information on the level of damage and needs of affected households after natural disasters. Existing SP information systems such as social registries can be a valuable resource that can help to inform such assessments. At the same time, with sufficient planning, data from postdisaster household needs assessments can be incorporated into social registries to support their expansion and updating. Several Latin American and Caribbean countries apply postdisaster needs assessment instruments to households after disasters, informing the provision of SP–related support. These instruments are at times formalized through standardized processes with strong legal foundations, such as Chile’s Basic Emergency Sheet (Ficha Básica de Emergencia—FIBE) (box 2.7); o ­ perate as formalized processes without associated legislation, such as Jamaica’s Household Disaster Impact and Needs Assessment— JHDINA (box 2.8); or may be developed from scratch following a disaster event, as in the case of Dominica’s Vulnerability Needs Assessment following Hurricane Maria in 2017. In the Philippines, the Department of Social Welfare and Development utilizes the paper-based Disaster Assistance Family Access Card containing basic information pertaining to whether the beneficiary’s house was damaged (partially or totally), whether they are existing beneficiaries of social programs (that is, the conditional cash transfer program Pantawid), and whether they belong to a vulnerable group (such as women, children, the elderly, and the disabled). Where post-shock assessments are concerned, the trade-off between a timely versus accurate shock response must be balanced. Inclusion errors can and should be tolerated in the short term (that is, those included who do not warrant support based on predetermined eligibility criteria).

BOX 2.7

Chile: A postdisaster data collection tool to assess shock-impacted households In Chile, the FIBE collects detailed information on shock-impacted households in response to emergency. FIBE is part of the Damages and Needs Assessment System for emergencies, disasters, and catastrophes. FIBE consists of a questionnaire that is administered to impacted households and individuals in the immediate aftermath of an event. An app version of FIBE links to the Household Social Registry, replacing the paper-based questionnaire. The FIBE app is downloadable for smartphones to guarantee easy access for users, and users can enter their information into the app. The FIBE app can then verify and validate information on the registrant’s education, health, housing conditions Source: Beazley and Barca 2018. Note: FIBE = Ficha Básica de Emergencia (Basic Emergency Sheet).

and characteristics, and level of vulnerability to the shock. All validated information is transmitted to the Ministry of Interior and subsequently to the other relevant ministries, including the Ministries of Health and of Education. Based on the information, the Ministry of Interior determines the nature and magnitude of the social programs (cash transfers and/or housing subsidies) that should be triggered to support affected households. This integrated system ensures that Household Social Registry registrants who are not eligible for social programs based on their socioeconomic characteristics receive assistance in the case of disaster under FIBE’s criteria.

Data and Information: Understanding Risk and Household Vulnerability | 73

BOX 2.8

Jamaica: Household Disaster Impact and Needs Assessment Jamaica’s Household Disaster Impact and Needs Assessment (JHDINA) instrument is the main tool used to assess post-shock needs of households affected by disasters, assist in quantifying post-shock social protection needs, and inform social protection shock response by a range of government and nongovernment actors. The JHDINA is an instrument of the Humanitarian Assistance Committee of the National Disaster Risk Management Council. The committee is chaired by the Ministry of Labour and Social Security, which is responsible for social protection; also serving on the committee are representatives of Jamaica’s national DRM agency, other government agencies, and nongovernmental organizations such as Food for the Poor and the Adventist Development and Relief Agency. The JHDINA is applied by multisector teams from the committee, led by ministry social workers. Communities to be assessed are prioritized following an initial damage assessment at the national level. The JHDINA was redesigned in 2017 to help address gaps identified in the previous version. Challenges included a reliance on paper-based data

collection, limited variables to inform appropriate social protection response across a range of actors, and limited coordination in the postdisaster household assessment process. The new instrument is available in multiple formats to enable quick ­d ecision-making and integration of the multiple agencies engaged in disaster response in the ­c ountry. Variables captured by the questionnaire included disaster type, location, demographics and family composition, health of household members, receipt of social assistance, damage and losses, and immediate needs. Jamaica has not experienced a national-level disaster event since the JHDINA’s redesign. However, the instrument’s predecessor informed vertical expansion of the Programme of Advancement Through Health and Education (PATH) conditional cash transfer program and National Insurance ­pensions, and horizontal expansion to non-PATH affected households following Hurricane Dean in 2007.

Source: World Bank 2018.

Exclusion errors should be addressed expeditiously through a sound grievance redress process and coordination with other programs (in the SP system and by nongovernmental actors) to swiftly reach larger shares of affected households. As the response evolves and refocuses on longer-term recovery, more precise targeting of losses and needs helps identify those eligible for longer-term support.

CONSIDERATIONS FOR INSTITUTIONALIZING INFORMATION FOR ASP The drive toward strengthening and integrating information for ASP depends substantially on mechanisms to institutionalize and operationalize information systems. Before the onset of any shock, it is essential to invest in preparatory measures to improve data quality, based on a strong policy vision and national commitments to strengthen the adaptive capacities of SP systems. This brings into focus the importance of capacities to collect, store, manage, and share these data, as well as the institutional structures that will underpin these arrangements. In shock-prone environments, this can vary significantly across countries.

74 | Adaptive Social Protection

As a starting point, information requirements for ASP involve a more diverse range of stakeholders, compared to traditional SP programming. The breadth of data requirements implies a multisectoral engagement, with significant information often coming from outside of the SP sector. Table 2.4 provides an illustrative framework to consider the different information functions at the policy and program levels, and how this may look based on the maturity of a given information structure within a country. The development of ASP information systems can also promote institutional coordination among SP, DRM, and humanitarian actors. Ultimately, this can lead to improved targeting accuracy, reduced duplication of efforts, and potentially cost-saving (for example, administrative costs of data TABLE 2.4  The

maturity of ASP information systems and the roles of different actors in strengthening them SYSTEM MATURITY

KEY FUNCTIONS

LOW (NASCENT)

(EMERGING)

HIGH (ESTABLISHED)

Policy setting Coordination and management

Fragmented approaches to social protection, DRM, and humanitarian information systems. Social protection information often focused around a singleuser program, requiring heavy engagements of development partners.

Collaboration across social protection, DRM, and humanitarian information actors. Clear identification of stakeholders. Social protection information more streamlined, across multiple programs, with increasing engagement of government.

Clear arrangements in place within government for crosssector coordination and leadership. Social protection information spans multiple programs, with clear sector links. Role of stakeholders formalized through agreements, terms of reference.

Financing

External financing is key to allow start up. Data initiatives separately resourced.

Combination of nongovernment and government financing, with increasing synergies of financial resources.

Largely government financed. Nongovernment financing is required only when the magnitude of a shock is significant.

Increased attention to formalizing protocols, privacy, and data sharing, with implications for engaging in shock response.

Mechanisms in place, with national laws adhering to international data transfer and privacy protocols.

Frameworks, standards, General information protocols are and ethics often poorly defined, outsourced to nongovernment providers. Program delivery aspects Data collection/update modalities

Infrequent (en masse registration), partial across the country. Data collected by development partners may supplement the gap.

Open registration (dynamic inclusion). Nongovernment partners encouraged to use the same ID for postdisaster data collection.

Open registration and capable of post-shock open registration.

Citizen interface (point of contact) for registration

Limited, temporary site and heavy engagements of partners required.

Online or local office.

Online and local office, mobile team.

Data verification, quality assessment

Limited, manual cross-check among government and nongovernment partners.

Automatic cross-checks available for government-led social protection, but manual crosscheck of social protection data with nonsocial protection and nongovernment data (DRM, humanitarian database).

Automatic cross-checks among government and nongovernment partners.

Data interoperability

Limited, information based at program level.

Same ID used in multiple social protection programs.

National ID used for social protection, DRM, and humanitarian database.

Source: World Bank. Note: DRM = disaster risk management; ID = identification.

Data and Information: Understanding Risk and Household Vulnerability | 75

c­ ollection, recurring costs of data management, and private costs to citizens) (Pelham, Clay, and Braunholz 2011). There is increasing evidence on countries that have been able to reap such benefits in one way or another. In Turkey, refugees benefiting from the Emergency Social Safety Net are being enrolled in SP programs (social assistance programs and employment services) through the country’s existing information management platform. In Mali, development partners have invested in establishing a solid information system for Jigisèmèjiri, with the aim of improving coordination of targeting and sharing the costs of data collection and analysis with other interventions (government and nongovernment) (O’Brien et al. 2018). In the Philippines, the Cash Working Group has piloted a process for humanitarian agencies to coordinate with the Listahanan, digitizing the information on their beneficiaries by submitting it to the National Household Targeting Office for verification and/or inclusion into the database. This process was intended to help digitize the Disaster Assistance Family Access Cards ­process of tracking which households were receiving which benefits from humanitarian agencies (Bowen 2015). A key implication is the need to anchor the information strengthening agenda around a government vision and a set of institutional commitments. While a combination of national and international expertise often will be needed to undertake different information functions, efforts need to be harnessed in a central way. In Mexico, for instance, a parliamentary act requires the ministries involved in disaster response to share a common database and information system (World Bank 2013). The country’s social registry, Sistema de Focalización para el Desarrollo, supports targeting of social assistance. It is shared by the Ministry of Social Development with other ministries/DRM actors and is layered with any additional postdisaster data collection and information on who is receiving emergency programs. Similarly, SP actors can incorporate tools used by the DRM sector in formulating their programs, such as contingency planning, early warning systems, and postdisaster needs assessments. Over time the quality of administrative and staffing structures also will become important to institutionalize. As information systems become more relevant for ASP, mechanisms will need to evolve to allow for their decentralized implementation while maintaining centralized control functions (for example, verifying and validating data). This will require a network of staff at various levels to support implementation. These networks will allow the government to dispatch staff horizontally (for example, across different departments or ministries), vertically (for example, through higher tiers of government), or through collaboration with nonstate actors (for example, humanitarian actors and voluntary groups) to address the gap of beneficiary caseload and the capacity of the government’s departments that manage the relevant information systems. The challenges of financing and costing information systems for ASP should not be overlooked. Progressing from nascent to emerging or mature information systems takes time and continuous effort, and entails many different types of financing and costs (Leite et al. 2017). The first set of costs include “back-office” costs to develop information technology system capabilities and “front-office” costs such as administrative costs for outreach, intake, and registration. Costs for individuals and households should also be acknowledged, such as for their travel to citizen interface points and for participating in the intake and registration interview.

76 | Adaptive Social Protection

Lastly, it will be critical to ensure that this information can be easily communicated to and interpreted by decision makers involved in ASP policy and implementation. Potentially, important lessons can be introduced to SP programs from crisis-related information systems. For example, the Integrated Food Security Phase Classification is a common global scale for classifying the severity and magnitude of food insecurity and malnutrition. The scale is a “big-picture” classification focusing on providing information that is consistently required by stakeholders around the world for strategic decision-making. It focuses on actionable, easily interpretable information for strategic decision-making in a manner that is simple and accessible (FAO 2016).

NOTE 1. The United Nations International Strategy for Disaster Reduction defines early warning systems as “an integrated system of hazard monitoring, forecasting and prediction, disaster risk assessment, communication and preparedness activities systems and processes that enables individuals, communities, governments, businesses and others to take timely action to reduce disaster risks in advance of hazardous events” (www.unisdr.org/we​ /­inform/terminology).

REFERENCES Bailey, R. 2012. Famine Early Warning and Early Action: The Cost of Delay. London: Chatham House. Barca, V. 2017. Integrating Data and Information Management for Social Protection: Social Registries and Integrated Beneficiary Registries. Canberra: Commonwealth of Australia, Department of Foreign Affairs and Trade. https://dfat.gov.au/about-us/publications​ /­Documents/integrating-data-information-management-social-protection-full.pdf. Barca, V., and C. O’Brien. 2017. “Factors Affecting the Usefulness of Existing Social Protection Databases in Disaster Preparedness and Response.” Policy Brief. Oxford Policy Management, Oxford, UK. https://assets.publishing.service.gov.uk/media/5a942c50ed915d57d4d0ef98​ /­Policy-Brief-Factors-affecting-usefulness-existing-social-protection-databases.pdf. Bastagli, F. 2014. “Responding to a Crisis: The Design and Delivery of Social Protection.” ODI Working Paper, Overseas Development Institute, London. Bastagli, F., J. Hagen-Zanker, L. Harman, V. Barca, G. Sturge, and T. Schmidt. 2016. “Cash Transfers: What Does the Evidence Say? A Rigorous Review of Programme Impact and of the Role of Design and Implementation Features.” Overseas Development Institute, London. Bastagli, F., and L. Harman. 2015. “The Role of Index-Based Triggers in Social Protection Shock Response.” Overseas Development Institute, London. Beazley, R., and V. Barca. 2018. “The Role of Data and Information for Adaptive Social Protection.” Background paper for the World Bank, Oxford Policy Management, Oxford, UK. Blumenstock, J., G. Cadamuro, and R. On. 2015. “Predicting Poverty and Wealth from Mobile Phone Metadata.” Science 350 (6264): 1073–76. Bowen, T. 2015. “Social Protection and Disaster Risk Management in the Philippines: The Case of Typhoon Yolanda (Haiyan).” Policy Research Working Paper 7482, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/681881468181128752/pdf​ /­WPS7482.pdf. Cabot Venton, C., C. Fitzgibbon, T. Shitarek, L. Coulter, and O. Dooley. 2012. “The Economics of Early Response and Disaster Resilience: Lessons from Kenya and Ethiopia: Economics of Resilience Final Report.” Department for International Development, London.

Data and Information: Understanding Risk and Household Vulnerability | 77

del Ninno, C., and B. Mills, eds. 2015. Safety Nets in Africa: Effective Mechanisms to Reach the Poor and Most Vulnerable. Washington, DC: World Bank. http://documents.worldbank.org​ /­curated/en/869311468009642720/pdf/Safety-nets-in-Africa-effective-mechanisms-to​ -reach-the-poor-and-most-vulnerable.pdf. FAO (Food and Agriculture Organization of the United Nations). 2016. Resilience Index Measurement and Analysis–II (RIMA-II). Rome: FAO. Gao, J., and B. F. Mills. 2018. “Weather Shocks, Coping Strategies, and Consumption Dynamics in Rural Ethiopia.” World Development 101 (January): 268–83. Government of Uganda. 2016. “Third Northern Uganda Social Action Fund (NUSAF3) Project— Disaster Risk Financing Sub-Component Handbook.” Government of Uganda, Kampala. Hallegatte, S., A. Vogt-Schilb, M. Bangalore, and J. Rozenberg. 2017. Unbreakable: Building the Resilience of the Poor in the Face of Natural Disasters. Washington, DC: World Bank. https:// www.gfdrr.org/sites/default/files/publication/Unbreakable_FullBook_Web-3.pdf. Hill, R., and C. Porter. 2017. “Vulnerability to Drought and Food Price Shocks: Evidence from Ethiopia.” World Development 96: 65–77. IEG (Independent Evaluation Group). 2011. Social Safety Nets: An Evaluation of World Bank Support, 2000–2010. Washington, DC: World Bank. Kuriakose, A., R. Heltberg, W. Wiseman, C. Costella, R. Cipryk, and S. Cornelius. 2012. “ClimateResponsive Social Protection.” Social Protection and Labor Discussion Paper 1210, World Bank, Washington, DC. https://siteresources.worldbank.org/SOCIALPROTECTION​ /­Resources/SP-Discussion-papers/430578-1331508552354/1210.pdf. Leite, P., T. George, C. Sun, T. Jones, and K. Lindert. 2017. “Social Registries for Social Assistance and Beyond: A Guidance Note and Assessment Tool.” Social Protection and Labor Working Paper 1704, World Bank, Washington, DC. http://documents.worldbank.org/curated​ /­en/698441502095248081/pdf/117971-REVISED-PUBLIC-Discussion-paper-1704.pdf. Levine, S., A. Crosskey, and M. Abdinoor. 2011. “System Failure? Revisiting the Problems of Timely Response to Crises in the Horn of Africa.” Humanitarian Practice Network Paper 11, Overseas Development Institute, London. Lindert, K., T. George, I. Rodriguez-Caillava, and Kenichi Nishikawa. Forthcoming. A Sourcebook on the Foundations of Social Protection Delivery Systems. Washington, DC: World Bank. McCord, A. 2013. “Review of the Literature on Social Protection Shock Responses and Readiness.” ODI Shockwatch. Overseas Development Institute, London. Mittal, Neeraj, Anit Mukherjee, and Alan Gelb. 2017. “Fuel Subsidy Reform in Developing Countries: Direct Benefit Transfer of LPG Cooking Gas Subsidy in India.” CGD Policy Paper. Center for Global Development, Washington, DC. O’Brien, C., Z. Scott, G. Smith, V. Barca, A. Kardan, R. Holmes, C. Watson, and J. Congrave. 2018. “Shock-Responsive Social Protection Systems Research: Synthesis Report.” Oxford Policy Management, Oxford, UK. https://www.opml.co.uk/files/Publications/a0408-shock​ -responsive-social-protection-systems/srsp-synthesis-report.pdf?noredirect=1. Pelham, L., E. Clay, and T. Braunholz. 2011. “Natural Disasters: What Is the Role for Social Safety Nets?” SP Discussion Paper 1102, World Bank, Washington, DC. https://www.gfdrr.org​ /­sites/default/files/documents/Social%20Safety%20Nets.pdf. Schnitzer, P. 2018. “How to Target Households in Adaptive Social Protection Systems? Evidence from Humanitarian and Development Approaches in Niger.” Journal of Development Studies 55 (Supplement 1): 75–90. Seltzer, W., and M. Anderson. 2001. “The Dark Side of Numbers: The Role of Population Data Systems in Human Rights Abuses.” Social Research 68 (2): 481–513. Skoufias, E., Y. Kawasoe, E. Strobl, and P. Acosta. 2019. “Identifying the Vulnerable to Poverty from Natural Disasters. The Case of Typhoons in the Philippines.” Policy Research Working Paper 8857, World Bank, Washington, DC. http://documents.worldbank.org/curated​ /­en/326941558453867995/pdf/Identifying-the-Vulnerable-to-Poverty-from-Natural​ -Disasters-The-Case-of-Typhoons-in-the-Philippines.pdf. Sonderegger, Petra, Swetha Totapally, Jasper Gosselt, Adityendra Suman, James Goh, Devvart Poddar, Surya AV, and Mahesh Vyas. 2019. “State of Aadhaar, 2019.” Harvard Dataverse. https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/M9RWZN.

78 | Adaptive Social Protection

UNDP-UNEP (United Nations Development Programme–United Nations Environment Programme). 2018. Vulnerability to Climate Hazards Index: Lessons Learned and Systematization of the Design Process and Application of the IVACC Index—Dominican Republic. Poverty-Environment Initiative. Panama City: Panama. Walsh, B., and S. Hallegatte. 2019a. “Measuring Natural Risks in the Philippines: Socioeconomic Resilience and Wellbeing Losses.” Policy Research Working Paper 8723, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/482401548966120315/pdf​ /­WPS8723.pdf. Walsh, B., and S. Hallegatte. 2019b. “Socioeconomic Resilience in Sri Lanka: Natural Disaster Poverty and Wellbeing Impact Assessment.” Policy Research Working Paper 9015, World Bank, Washington, DC. http://documents.worldbank.org/curated​ /­e n/173611568643337991​/­p df/Socioeconomic​-Resilience-in-Sri-Lanka-Natural​ -Disaster​-Poverty-and​-Wellbeing-Impact-Assessment.pdf. Wang, R., and D. Strong. 1996. “Beyond Accuracy: What Data Quality Means to Data Consumers.” Journal of Management Information Systems 12 (4): 5–33. Watson, C., T. Lone, U. Qazi, G. Smith, and F. Fashid. 2017. “Shock-Responsive Social Protection Systems Research: Case Study: Pakistan.” Oxford Policy Management, Oxford, UK. Wilkinson, E., L. Weingartner, R. Choularton, M. Bailey, M. Todd, D. Kniveton, and C. Cabot Venton. 2018. “Forecasting Hazards, Averting Disasters: Implementing Forecast-Based Early Action at Scale.” Overseas Development Institute, London. World Bank. 2013. Building Resilience to Disaster and Climate Change through Social Protection: Synthesis Note. Washington, DC: World Bank. http://documents.worldbank.org/curated​ /­en/187211468349778714/pdf/796210WP0Build0Box0377381B00PUBLIC0.pdf. World Bank. 2016. Big Data Innovation Challenge: Pioneering Approaches to Data-Driven Development. Washington, DC: World Bank. http://documents.worldbank.org/curated​ /­en/396861470905612761/pdf/107751-REVISED-PUBLIC-BigData-Publication-e-version​ -FINAL.pdf. World Bank. 2018. Post-Disaster Damage Assessment and Targeting Mechanisms in Jamaica. Washington, DC: World Bank.

3

Finance

APPLYING A DISASTER RISK FINANCING APPROACH

OVERVIEW Adaptive social protection implies different financing burdens for governments to ­address. Resilience-building interventions tend to generate outcomes among beneficiaries over the long term, meaning that sustainable-financing commitments to support such initiatives are ­critical. More generally, adaptive social protection (ASP) will benefit from increased fiscal space for the extension of social protection access to vulnerable households and the strengthening of the overall social protection ­system. At the same time, for ASP programs to become more responsive to shocks, risk financing strategies will need to be developed with appropriate risk financing instruments prepositioned and linked to responsive safety net p ­ rograms. In doing so, funding will be made more readily available for quick ­disbursement through shock-responsive safety net programs—a key determinant in the timeliness of response, and by extension its effectiveness in protecting the well-being of affected ­households. Emerging evidence highlights three core lessons for applying a disaster risk financing approach to ASP. First, investment is needed to understand the potential cost of response, leveraging a variety of data sources including historical hazard data to shed light on the anticipated contingent liability of using a safety net to respond to s­ hocks. Second, building from these costing models, appropriate funding should be ­preplanned. However, no single financial instrument should cover the entire contingent liability created through the development of a shock-­ responsive safety net ­program. Rather, a risk financing strategy that layers multiple instruments according to the frequency and severity of the shock will be ­required. Third, robust payment mechanisms that are capable of absorbing the funding made available after a shock and delivering it to households are critical to enable the disbursement of these risk financing instruments to households. Building from the payment conduits used for safety net programs, the requisite rules and capacity among payment service providers should be established ex ­ante. Doing so after the shock will often cause severe delays in disbursement and d ­ elivery.

Finance

This chapter draws extensively from a background paper by Maher, Fitzgibbon, and Solórzano (­ 2018).  79

80 | Adaptive Social Protection

DISASTER RISK FINANCING: FROM REACTIVE TO PROACTIVE RESOURCE MOBILIZATION The prevailing model for financing disaster response poses challenges to responding in a timely manner, at the expense of the well-being of those who have been affected and are in urgent need of s­ upport. Historically, the humanitarian response system has been the first responder to covariate shocks, especially in poorer countries, with national governments and international partners mobilizing resources after the shocks hit to meet estimated ­needs. Today, this system is buckling under the weight of growing ­ 016). Indeed, an “ad hoc demand, as illustrated in f­ igure ­3.1 (see also IASC 2 postdisaster funding model…is too slow, leads to a fragmented and underfunded response, and encourages underinvestment in risk reduction and preparedness, thereby increasing the economic and human costs of catastrophes” (Clarke and Hill ­2016, 10). For poorer countries, this can have significant impacts where spending is already constrained, borrowing is more costly, and reallocations can affect the continuity of basic government ­functions. In such instances, these countries often turn to global humanitarian agencies and development partners for support, which leads not only to delays but recurring, large funding gaps relative to needs on the ­ground. In such cases, funding can come late and can be insufficient (Clarke and Dercon ­2016). Disaster risk financing is part of a global shift in thinking from seeing disasters as unpredictable humanitarian crises to predictable events that can be planned for and managed to minimize their i­mpact. This involves moving from a reactive approach that addresses the impact of shocks once they happen to a more proactive approach, putting in place the required systems and financing to respond to shocks before they take ­place. A proactive approach highlights the need for governments to develop risk financing strategies that enable funding to flow in the event of a shock and, thus, to enable a faster response to ­disasters.

FIGURE ­3.1

Humanitarian response: Trend in requirements and funding gaps, 2007–17 25

Billion $

20 15 10 5 0 2007

2009

2011 Funding

Source: UN OCHA ­2018.

2013 Needs

2015

2017

Finance: Applying a Disaster Risk Financing Approach | 81

This chapter documents emerging lessons on applying these same principles to the financing of ASP, s­ pecifically. In doing so, the need for and value in proactively planning for how to finance shock-responsive social protection for timely support becomes c­ lear. Indeed, this highlights a strong synergy: when disaster risk financing strategies are established, a shock-responsive safety net represents a preprepared mechanism whereby risk financing instruments can disburse directly to affected h ­ ouseholds. Conversely, the availability of the kinds of risk financing instruments outlined in this chapter will be instrumental in enabling fast response through a safety n ­ et. In that way, they are both critical components to building the resilience of poor and vulnerable households. The disaster risk financing approach outlined here enables the delivery of shock-responsive programs to support short-term coping after a shock. This approach can be complemented by safety net interventions that also support a household’s preparedness and longer-term adaptation for a holistic approach to resilience building along the pre- and post-shock continuum. Long-term investments in preparedness and adaptation could help to reduce future household needs, and by extension, lower response costs over time—but more evidence is needed. The kinds of interventions discussed in chapter 1 to support household adaptation over the longer term can be costly at scale (for example, in contexts of widespread chronic and severe poverty in high-risk a­ reas). Yet, initial evidence indicates that more expensive investment scenarios are broadly offset by the avoided cost of humanitarian response (Cabot Venton 2018; Wilkinson et a­ l. ­2018). In Bangladesh, the Chars Livelihoods Program focused on building an annual contingency budget into its project design for disaster response, but the need for this contingency fund decreased over time because of the program’s specific focus on reducing vulnerabilities and supporting the adaptation of poor households living in the chars to regular flooding (ADB ­2018). Where resources are limited, more evidence is needed on the cost-effectiveness of and trade-offs between resilience-building interventions in support of adaptation and risk reduction at the household level over the long term versus the cost of ex post shock response to support short-term ­coping.

WHY IS A DISASTER RISK FINANCING APPROACH IMPORTANT FOR ASP? A disaster risk financing approach in support of an early response can lower the overall cost of a disaster, relative to a scenario in which the response arrives ­late. In Ethiopia in 2016, the timeliness of funding that was made available to respond to the drought created savings of over $6 ­million. The longer-term economic cost of responding later could have been ­$47.9 million (Cabot Venton and Sida ­2017). Similar estimates including Kenya and Somalia alongside Ethiopia also suggest that an early response saves ­$1.6 billion over 15 years in comparison to a late ­response. When avoided losses are incorporated, an early response saves ­$2.5 billion, or an average of $163 million per year (Cabot Venton ­2018). Cost savings are realized when faster assistance dissuades affected people from engaging in negative coping ­strategies. As outlined previously in this report, negative coping strategies can have long-term, intergenerational, and irrevocable impacts on household ­welfare. In Ethiopia, the cost of a drought to poor households increased exponentially over time: $0–$50 for a 4-month delay versus $1,300 for a 6–9-month delay (Clarke and Vargas Hill ­2013). Acting early is

82 | Adaptive Social Protection

critical to prevent the use of negative coping mechanisms, and the availability of financing is a key factor determining the ability to do s­ o. Where risk financing encourages predictable assistance among households, the reduced uncertainty in the face of potential disasters also can enable households to invest in preparedness and a­ daptation. As noted in chapter 1, in cases where households understand that they are protected by predictable, shock-­ responsive programs, evidence indicates they adopt higher-yield, higher-risk livelihood i­nvestments. This can enable beneficiaries to grow their household income and ultimately leads to a pathway of a more resilient ­state. In Mexico, municipalities participating in the drought index insurance program Component for the Attention of Natural Disasters (CADENA)—with similar properties to a shock-responsive safety net—increased expenditure per capita by 27 percent and income per capita by 38 percent, which is about $284–$378 in additional income (Clarke and Hill ­2016). Preplanning and prepositioning financing (with clear rules for its use) also can reduce uncertainty in government budgets around the role of social protection in responding to s­ hocks. In Uganda, a $10 million contingent line of credit finances drought response through the social protection s­ystem. The ­preestablished line of credit means the shock-responsive safety net does not introduce any budget uncertainty to the Ministry of Finance, Planning, and Economic ­Development. It also precludes more expensive ex post financing options such as budget reallocations, which divert resources from high-yield investments (Clarke and Dercon ­2016). Financial planning for shock-responsive social protection also can increase country ownership and government l­ eadership. While for now this is less tangible than the other benefits, it holds the promise of being the most ­transformative. Using ASP programs for disaster response places governments in the driving seat, with the ownership and (importantly) the responsibility to deliver emergency resources to their c­ itizens. This is a paradigm shift from the status quo, which relies predominantly on nongovernment actors to deliver said r­ esources. Empowering governments to invest in their systems and capacities to manage shock response ultimately leads to a more sustainable shock-response system in the ­country. It also can serve to strengthen the social compact between citizens and their ­governments.

A CLOSER EXAMINATION OF THE DISASTER RISK FINANCING APPROACH FOR ASP This section outlines the key lessons for applying a disaster risk financing approach for ­ASP. Three emerging lessons are (1) estimate the cost of response before a shock, (2) preplan the funding required to ensure timely response, and ­ echanisms. Each (3) establish and link the plan to effective disbursement m ­lesson is examined below and illustrated with reference to relevant country examples.

Lesson 1: Estimate the cost of response before the shock Financial planning for shocks before they occur requires a very different set of skills and information than planning responses after the ­event. Pre-shock planning requires the data and analysis to understand the magnitude, frequency, and

Finance: Applying a Disaster Risk Financing Approach | 83

impact of potential shocks as well as a decision on the scale and duration of ­support. Without this, it is impossible to calculate the costs of responding with social protection and its financial ­feasibility. The first good-practice principle involves estimating in advance the impact of a shock on a target population and the resulting l­osses. Developing a realistic estimation of the scale and range of required funding involves establishing, based on existing data sources from the disaster risk management, humanitarian, and social protection sectors, the following: • • • •

What major shocks are likely to affect the target population? How frequently do these shocks occur? For each risk, can its scale and magnitude be quantified? Which areas of the country or parts of the population are most likely to be affected? • What response is required to enable the identified populations to cope in or overcome the immediate impacts? • How long will this assistance be required? Natural disaster shocks have well established technical definitions, quantification metrics, and historical ­data. Droughts and hurricanes draw on meteorological indicators around rainfall and wind s­ peed. An earthquake’s strength, or magnitude, is measured using the Richter s­ cale. Information on past shocks (including the cost of response) exists in publicly available databases (O’Brien et ­al. ­2018c). This is not the case for anthropogenic hazards such as conflict where the potential impact of any given risk is much more complex to predict and ­quantify. However, these metrics and data are not always sufficient to understand how a target population will be impacted by the same level of s­ hock. Nor do they provide an acceptable level of accuracy when assessing impact in specific geographical areas (subnational) given the complex relationship between vulnerability and r­ isk. Some hazards (floods) can only be defined locally in relation to long-term ­norms. Understanding the risk of a subpopulation involves calibrating standard definitions or measurements to the local ­context. These definitions serve to trigger payment mechanisms and other response i­ nterventions. A household’s vulnerability to shocks can be difficult to assess ex a­ nte. Typical factors in the calculation include the following: • Poverty and vulnerability to poverty—levels and sources of income, assets, and savings, among others • Livelihood issues—food security (rise in food prices) and reliance on livelihood activities highly affected by shocks, such as rain-fed subsistence agriculture (drought) • Location—proximity to coastal areas, waterways (floods and hurricanes), and conflict areas • Infrastructure—construction standards, disaster-proof roads, bridges, and wells, among others • Timing—the point at which a shock hits can vary impact, such as just before or just after a harvest • Preparedness systems—community early warning strategies Previous shocks can provide proxies for gaps in information, using postdisaster needs assessments and f­ unding appeals. In Kenya, the Hunger Safety Net Program (HSNP) preassessed the additional proportion of the population to

84 | Adaptive Social Protection

receive cash transfers for different levels of d ­ rought. The HSNP used an existing quantification metric (twice yearly post-rains assessment) and past historical data and information (a maximum of 77 percent of the population needed support for high-magnitude drought, with 50 percent of households in the affected areas in need of food aid) to assess the proportion of the population in need of humanitarian support (a maximum of 75 ­percent). Assessments during previous high-magnitude drought had never put needs above 77 percent of a population and, on average, the affected areas had identified 50 percent of households in need of food aid in drought y­ ears. Hence, the guideline for shock response adopted these ­rates. When designing a shock-responsive safety net program of the kind highlighted in chapter 1, key questions can help to determine its potential c­ ost, as outlined in chapter 2. Key questions to ask in designing the component include: • When should the social protection program respond? To what shocks? How are the shocks defined and measured? Using what data or indicators? Before or only after the shock? • Where should the shock response happen? What is the geographic coverage of the expanded transfers? At a regional, district, or ward level, for example? Should the geographic coverage depend on the shock? • Who should benefit from the shock response? Existing beneficiaries, other members of the population, or both? • What should be the value of any additional transfers? Should there be a standard transfer amount or should it vary according to the shock and the needs? Should existing beneficiaries receive the same, more, or less than nonroutine beneficiaries? • How long should beneficiaries receive a scaled-up benefit? Should payments or transfers be a one-off or continue for several months after the disaster trigger has been hit (such as until the rains arrive or the floods subside)? The answers to these questions can inform a matrix to support decision-­ making. Tables ­3.1 and ­3.2 provide examples of decision support matrices for drought in Uganda and ­Kenya, respectively. This in turn can be used to develop a costing model which automatically calculates the long-term cost implications of ­permutations. In order to assist policy makers, it is useful to link the variables in the framework to a financial model that calculates the cost impact of changes to any ­parameter. A model can show the cost implication of changes in a matrix, and thus facilitate decision-making. The financial model should calculate the cost of TABLE ­3.1

Uganda: A framework to scale up the NUSAF cash-for-work program

LOCATION

By district

HOUSEHOLDS COVERED BY LABOR-INTENSIVE PUBLIC WORKS PROGRAM

DAILY WAGE FOR PUBLIC WORKS

NUMBER OF DAYS PER MONTH

DURATION OF WORKS AND PAYMENT

No drought

Routine NUSAF households (currently 4 percent of households)

U Sh 4,000

15

4 months

Drought

Routine NUSAF households (currently 4 percent of households)

U Sh 4,000

15

4 months

Additional households to cover a maximum of 15–20 percent of households in each district

U Sh 4,000

15

4 months

PRIMARY TRIGGER

DROUGHT CONDITIONS

NDVI anomaly value ≥ ­−0.02 NDVI anomaly value < ­−0.02

Source: Government of Uganda ­2016. Note: NDVI = Normalized Difference Vegetation Index; NUSAF = Northern Uganda Social Action F ­ und; U Sh = Uganda shilling.

Finance: Applying a Disaster Risk Financing Approach | 85

TABLE ­3.2

Kenya: A framework to scale up the Hunger Safety Net Program

LOCATION

Subcounty

TRIGGER OF VCI

DROUGHTPHASE EQUIVALENT

MAXIMUM COVERAGE OF HOUSEHOLDS TO RECEIVE CASH TRANSFER

AMOUNT OF TRANSFER

FREQUENCY OF TRANSFER

DURATION OF TRANSFER

≥ 50 and 35–50

Wet or no drought

1. Normal

Routine HSNP households

Standard payment (K Sh 5,100)

Every 2 months

Ongoing

20–35

Moderate drought

­2. Alert

Routine HSNP households

Standard payment

Every 2 months

Ongoing

10–20

Severe drought

­3. Alarm

Routine HSNP households

Standard payment

Every 2 months

Ongoing

Households beyond Emergency routine, up to 50 percent payment coverage in each (K Sh 2,550) sublocation

Every month

For each month VCI is at severe drought status

Routine HSNP households

Every 2 months

Ongoing

Every month

For each month VCI is at extreme drought status

< 10

Extreme drought

­4. Emergency

Standard payment

Households beyond Emergency routine, up to 50 percent payment coverage in each (K Sh 2,550) sublocation Source: Fitzgibbon ­2016. Note: HSNP = Hunger Safety Net Program; K Sh = Kenya shilling; VCI = Vegetation Condition Index.

FIGURE ­3.2

Kenya: Modeling the cost of responding to drought 45 40 35

Million $

30 25 20 15 10 5

Extreme payout

Severe payout

14 20

12

13 20

10

11

20

20

09

20

20

20 08

20 07

20 06

20 05

20 04

03 20

02 20

20

01

0

Average annual payout

Source: Maher, Fitzgibbon, and Solórzano ­2018. Note: Annual scalability costs are totaled across the four Hunger Safety Net Program ­counties.

a one-off response and also the long-term cost of scaling a shock-responsive approach over 10–20 ­years. Figure ­3.2 shows the result of modeling Kenya’s ­HSNP. The bars illustrate how much the approach would have cost over the last 14 years using an agreed drought-response t­ rigger. The model shows the cost in very severe years and nondrought years and provides an average over all ­years. This enables policy makers to see that scaled-up payments would be required

86 | Adaptive Social Protection

almost every year in response to severe drought while a scale up that only triggers in response to extreme drought would be far less frequent (and hence less ­expensive). This information is valuable to policy makers in making decisions about what scale and level of shock-responsiveness is financially feasible in the long ­term. Risk analysis can inform the trigger point(s) at which a safety net program responds or modifies its ­response. Programs can be designed to trigger an earlier response, or even before a disaster occurs, by linking responses to early warning systems (as discussed in chapter 2). When preagreed thresholds are breached, programmatic responses are triggered as are the requisite, prepositioned financial ­resources.

Lesson 2: Preplan the financing required to ensure timely response Ensuring funds are available when needed is critical for effective risk f­ inancing. Once a clearer picture of the potential costs of responding through social protection emerges, governments are far better placed to examine their risk financing ­options. Postdisaster response, recovery, and reconstruction phases require different financial instruments and disbursement ­rates/durations. The immediate response operation requires the most instantaneous funding, whereas recovery and reconstruction take time to get under w ­ ay. These later costs can be much larger than initial relief ­assistance. For example, monsoons or earthquakes destroy infrastructure that is costly to r­ eplace. That said, droughts may result in larger immediate support costs through food aid or cash payments during the event, in the relief phase, which may persist for a longer period of ­time. Funding constraints can slow the response even if systems are in p ­ lace. In Lesotho, “the fact that the mechanism for the top-up [vertical expansion] was designed at the time of the crisis—not an integral part of the Child Grant Program—meant that it was slow to start and disburse and had to hunt for ­funds.” The program “only started providing top-ups [vertical expansions] in June 2016, 6 months after the declaration of the drought emergency” (and only for two quarterly payments), with delays partly due to the program waiting for the scheduled quarterly payment date (Kardan, O’Brien, and Masasa ­2017, 26). Humanitarian response generally involves waiting until a disaster has hit before securing financing for the response, recovery, and reconstruction ­phases. As noted, these responses typically undertake a needs/loss assessment and use the results to launch a global funding appeal; few funding appeals are ever fully ­funded. Consequently, funds for emergency response arrive piecemeal and often months after the ­crisis. A disaster risk financing strategy that layers instruments according to the frequency and severity of the risk can be developed based on the potential costs for shock-responsive social p ­ rotection. No single financial instrument can or should cover all risk financing r­ equirements. Risk-layering considers how to meet the financial cost of response using a menu of financial instruments (figure ­3.3). Each instrument has its own terms and c­ onditions and, therefore, advantages and disadvantages (see table ­3.3). When assessing how to finance contingent liabilities using a safety net, assessing which instruments are the most appropriate,

Finance: Applying a Disaster Risk Financing Approach | 87

FIGURE ­3.3

Disaster risk financing: Visualizing risk layering Hazard type Low frequency/ high severity

Financing instrument Market-based instruments Risk transfer for assets such as property insurance or agricultural insurance and risk transfer for budget management like parametric insurance, catastrophe bonds/swaps Contingent financing

High frequency/ low severity

Financial instruments that provide liquidity immediately after a shock Budgetary instruments Reserve funds specifically designated for financing disaster-related expenditures, general contingency budgets, or diverted spending from other programs

Source: Financial Protection Forum ­2018.

TABLE ­3.3

Disaster risk financing: A comparison of instruments

TYPE

ADVANTAGES

DISADVANTAGES

BEST SUITED

Ex ante Contingency/ reserve funds

• Can be cheap, particularly for frequent • Requires fiscal discipline shocks • High opportunity cost of funds, given • Fast to disburse high rates of return on other government • Allows implementers to plan ahead for investments their use • Can be hard to justify given the • Approach has been used in many ­opportunity cost contexts; thus, experience is available for countries to build upon

Low risk layer such as frequent low-level events (annual flooding, localized drought, conflict)

Contingent credit

• Can be cheap, particularly for midfrequency shocks • Fast, when conditions for disbursement are met • Allows implementers to plan • Can incentivize proactive actions to reduce risk (for example, policy actions in disaster risk reduction and disaster risk management)

• Has conditionality • Opportunity cost of loan • Adds to country’s debt burden, must be repaid • Current low (but growing) uptake of Cat DDOs as some countries prefer investment projects guaranteed resources over contingent instruments

Mid-risk layer such as higher-magnitude, less frequent events whose damages exhaust the resources of national contingencies (widespread flooding, hurricanes)

Market-based risk transfer instruments

• Can be cheap, particularly for extreme shocks • Can be fast • Allows implementers to plan • Supports fiscal discipline • Risk diversification

• • • • •

High-risk layer such as extreme, less frequent events, less than every 5–10 years (severe droughts, hurricanes, earthquakes)

Humanitarian assistance

• Flexible—can respond to need • Doesn’t have to be repaid

• Can be slow to be mobilized • Can be unreliable • Undermines preplanning

Only as a last resort

Other ex post instruments

• Approach has been used in many contexts; thus, experience is available for countries to build upon

• Can be slow • Can have negative impact on long-term development/investment programs • Can be expensive

Only as a last resort

Can be expensive for frequent shocks Can be vulnerable to criticism and “regret” Can miss need Need a level playing field to negotiate Trade-off between the cost of premiums and the frequency or scale of the pay-out

Ex post

Source: Maher, Fitzgibbon, and Solórzano ­2018. Note: Cat DDO = Catastrophe Deferred Drawdown Option.

88 | Adaptive Social Protection

adequate, and cost-effective alongside the frequency and severity of the anticipated shocks is ­critical. In most cases, multiple financial instruments will be required to meet the financial cost of the ­response. There are three main financial instruments for designing a risk financing strategy for ­ASP. Budgetary instruments, contingent credit, and market-based instruments are discussed in detail b ­ elow. They can be directly linked to shock-responsive safety net ­programs. Looking at each instrument, in turn: • Budgetary instruments (contingency/reserve ­funds). These are relatively cheap and immediately available following a covariate shock, allowing ministries, departments, and agencies to develop realistic contingency ­plans. These funds may be included in shock-responsive ASP financing in the form of flexible budget lines or contingency budget lines, allocating funds for shocks and avoiding bureaucratic delays such as verification, administrative accounting, and disbursement scheduling (box 3­ .1). Of critical importance are the rules under which such funds d ­ isburse. In many countries contingency funds exist; however, the conditions under which they disburse are vaguely written and often ­b road. In such instances, it can be a challenge to ensure sufficient funds are available when a shock occurs, especially when shocks occur later in the budgetary cycle, as the funds can be ­depleted. The main disadvantage is the opportunity cost when there is no disaster because of competing needs for ­funds. In addition, when a shock hits, the sums rarely cover all costs; thus, they need to be complemented with other financial instruments (World Bank ­2017). Budgetary reallocations, though these are not an ex ante instrument, also fall under this category, but in contrast are used in the aftermath of a shock, usually in the absence of financial planning (discussed further ­below). • Contingent credit (mainly ex ante loan a ­ greements). Such credit provides immediate liquidity in the aftermath of a covariate shock, often at highly concessional terms (long duration with low interest r­ ates). Ex ante loan agreements typically are offered by multilateral development banks and international financial institutions (box 3­ .2). These agreements have the potential to assure financing beyond a government’s own reserve funds

BOX ­3.1

Mexico: The Natural Disasters Fund In Mexico, the federal government established the Natural Disasters Fund (Fondo de Desastres Naturales—FONDEN) in 1996 as a mechanism to finance the postdisaster recovery and reconstruction of Mexico’s public assets and low-income ­housing. The fund consists of three primary financial accounts: the FONDEN Program for Reconstruction, the FONDEN Trust, and the Revolving ­Fund. Source: World Bank ­2017.

Collectively, these instruments assist the government in its efforts to respond quickly to natural disasters by providing funding for emergency relief, rehabilitation, and r­ econstruction. These instruments are continuously updated to enhance their efficiency and ­effectiveness. FONDEN has a mandatory annual budget allocation of at least ­0.4 percent of total ­expenditure.

Finance: Applying a Disaster Risk Financing Approach | 89

(World Bank ­2017). The main drawbacks are (1) the funds are often provided as budget support, so there is no guarantee that they will be used to finance ASP programs; and (2) it is still a loan and so adds to the country’s debt (O’Brien et ­al. ­2018a). Contingent credit is most cost-effective for high-­ impact, low-frequency ­shocks.

BOX ­3.2

Contingent credit instrument: Catastrophe Deferred Drawdown Option losses from small and medium events; (2) attracting donor support to capitalize a fund; (3) pooling ­country-specific disaster risks into one diversified portfolio; (4) accessing international reinsurance markets on competitive terms, diversifying risk across multiple countries with different risk profiles; and (5) building a more robust foundation of risk information and management (World Bank ­2017). • Cat DDOs incentivize proactive action to reduce For example, the Caribbean Catastrophe Risk ­risk. Governments must demonstrate the Insurance Facility has 17 members (primarily small capacity to manage natural risks to be ­eligible. Caribbean island s­ tates). It allows member govern• Cat DDOs may be drawn upon declaration of a ments to purchase insurance coverage to finance state of emergency in the borrower’s territory, as immediate postdisaster recovery ­needs. The facility a result of a natural ­disaster. acts as a risk aggregator by enabling participating • Cat DDOs act as a fiscal buffer that reduces countries to pool their country specific risks into one, disaster ­impact. better-diversified ­p ortfolio. This diversification • Cat DDOs can be used to back up existing should result in a substantial reduction in premium insurance ­pools. cost of 45–50 ­percent. Claims payments are based on Cat DDOs have been effective liquidity instru- parametric triggers, which means they are indexments, providing countries with needed cash in the based insurance instruments that pay claims based on immediate aftermath of a ­disaster. The World Bank the occurrence of a predefined event, (such as hurrihas approved 16 Cat DDOs for a total value of $ ­ 3.1 canes, earthquakes, flooding) rather than on an assess­billion. It also, in 2018, expanded the instrument to ment of actual losses on the ­ground. This measurement, countries with low per capita incomes that lack the made remotely by an independent agency, allows for financial ability to borrow from the International transparent, low-settlement costs and quick-­ Bank for Reconstruction and Development (that is, disbursing ­contracts. International Development Association ­countries). Insured countries pay an annual premium comKenya is the first African country to develop a mensurate with their own specific risk e­ xposure. disaster risk financing strategy and the first Parametric insurance products are priced for each International Development Association country to country based on their individual risk ­profile. Annual receive Cat DDO financing ($220 m ­ illion). Cat DDOs premiums typically vary from $200,000 to $4 million, are under preparation in Benin, Madagascar, and for coverage ranging from $10 million to $50 m ­ illion. ­Malawi. Catastrophe risk pools are emerging as useful The Caribbean Catastrophe Risk Insurance Facility mechanisms to support country access to cost-­ has made 38 payouts between 2007 and 2018 totaling effective risk transfer ­solutions. Regional risk pools $135 million, including payouts to 10 Caribbean councan facilitate (1) building regional reserves to finance tries following the 2017 hurricane ­season.

The World Bank developed the Catastrophe Deferred Drawdown Option (Cat DDO) in 2008 as a development policy loan for countries that have the financial ability to borrow from the International Bank for Reconst r uct ion a nd Development (t hat is, International Bank for Reconstruction and Development ­countries).

Sources: CCRIF SPC ­n.d.; Ghesquière et ­al. ­n.d.

90 | Adaptive Social Protection

FIGURE ­3.4

Ethiopia: PSNP integrates ex post humanitarian assistance within a risk financing strategy Bad years

Average year HRD

10 Federal contingency caseload and HRD

Millions needing assistance in PSNP woredas

9 8

HRD

Households with transitory needs

7 6 5 4 3 2

Chronically poor/ foodinsecure households

Expanded core caseload under PSNP 4

Previous core PSNP caseload

1 0

Start of PSNP 4

Time

Source: Woldemichael ­2018. Note: HRD = Humanitarian Requirement ­Document. PSNP = Productive Safety Net ­Programme. Federal contingency budget can also be used in non-PSNP woredas within PSNP ­regions.

• Other common ex post financing i­nstruments. Some governments fund part of the response ex post, usually through budget ­reallocations. This was the case in Fiji after Tropical Cyclone Winston in ­2016. The government financed relief and recovery activities through the social protection system by reallocating budgeted resources from lower-priority ­expenditures. Countries also may have scope for borrowing (depending on market access and existing levels of domestic and external debt) as well as tax increases or private sector investments, but these ex post measures are not cheap and come with other ­drawbacks. Humanitarian assistance remains a prevalent ex post instrument for response, as discussed at the outset of this chapter. When developing a comprehensive risk-layering strategy, humanitarian assistance can be integrated alongside other instruments, especially for shocks of low frequency, limited predictability, and high i­ntensity. In Ethiopia, if needs exceed 8 ­million people in Productive Safety Net Programme (PSNP) woredas ­(districts), a combination of contingency budgets and humanitarian assistance is utilized to cover up to an additional 2 million p ­ eople. When needs surpass 10 million people, the additional people are reached through the humanitarian system (figure ­3.4). It is advisable to consider a suite of options rather than to rely on a single financing instrument, as each type of funding has advantages and disadvantages, relative to the frequency, severity, and predictability of the ­disaster. This is the core principle of risk layering, the core considerations of which are outlined in table ­3.3. Examples of risk layering from five countries are depicted in table ­3.4.

Finance: Applying a Disaster Risk Financing Approach | 91

TABLE ­3.4  Ethiopia, Kenya, Mexico, the Philippines, and Uganda: Financial layering for adaptive social protection programs COUNTRY

RESERVE FUND

Ethiopia

l

Kenya

l

Mexico

l

Philippines

l

Uganda

CONTINGENT CREDIT

RISK TRANSFER

BUDGET REALLOCATION

l

DONOR FINANCE

l l

l

l l

l

Source: Maher ­2018.

Lesson 3: Establish and link the financing to effective disbursement mechanisms Establishing effective disbursement mechanisms (that is, how funding reaches beneficiaries) and linking disaster risk financing instruments to them is as important as securing funds in the first p ­ lace. Having funds available in-country is of limited benefit if they cannot be timely transferred to the relevant institutions and to the shock-affected c­ ommunities. One reason there has been so much interest in shock responsive social protection is that this is very often the only government service already regularly transferring cash (or other in-kind benefits) to large numbers of ­households. A key factor affecting the disbursement of funds to affected households is the existence of effective safety net payment s­ ystems. Countries can select different approaches in the use of social protection payment mechanisms for the delivery of cash-based emergency ­responses. These may include manual systems or electronic transfer to bank accounts or new technologies such as mobile phone ­payments. Scaling up manual systems tends to increase costs in line with the number of people assisted; it also can prove difficult in the aftermath of some disasters where roads and other key infrastructure prevent the physical distribution of ­cash. E-payment systems are emerging as the preferred option to deliver aid, as they have the advantages of speed, accuracy in targeting, and flexibility, even in challenging ­environments. Establishing such payment systems can be very costly and delay assistance if the infrastructure is only set up during or in response to a c­ risis. The registration of households for bank accounts can take months, even ­years. The adoption of e-payment systems takes time and faces challenges, including the coverage of agents and vendors, their financial liquidity, and unreliability of the ­t echnology. Nonetheless, e-payment systems are increasingly being introduced to channel private payments and remittances as well as social ­ ayments. Depending on the institutional arrangements, payprotection p ments to the affected population can be conducted by the implementing agency, decentralized to a local government, or outsourced to a financial ­service provider (which may be a private or state-owned company) (O’Brien et ­al. ­2018b). Once established, automatic payment systems provide a faster and much more efficient mechanism to disburse cash for both regular and emergency ­programs. Kenya’s HSNP is one of the best examples of putting a payment

l

HUMANITARIAN RESOURCES

l

92 | Adaptive Social Protection

mechanism in place in advance of any ­disaster. Likewise, in Ecuador, the robust payment system of the social protection system allowed a timely response to the 2016 ­earthquake. Expanding e-payment systems also has the positive side-effect of substantially increasing financial inclusion (which, as noted in chapter 1, is a key component for enhancing a household’s capacity to prepare and become more ­resilient). In Kenya, the coverage of households with bank accounts in the four poorest counties rose from being negligible to over 90 percent as part of the HSNP second ­phase. Core factors determining the effectiveness of safety net payment mechanisms are explored below, including their implication for disbursing postdisaster financing to ­beneficiaries. • Release of ­funds. Delays in the disbursement of upstream funding can cause delays to the beneficiaries receiving t­ ransfers. This is a common challenge for routine social protection delivery, and countries have tackled it in a variety of ways, including: minimizing the number of accounts for transferring resources; automating transfer procedures where possible; ensuring timely approval of budgets and monitoring availability of funds; defining a strategy for liquidity and cash management, as well as contingency planning for delays; (where feasible) utilizing the country’s single treasury account ­system; exploring the feasibility of classifying social protection expenditures as personnel emoluments rather than general expenses in the national budget (Barca et al. ­2017).1 • Absorptive capacity within the disbursement ­mechanism. In shock-­ affected contexts, standard delays are inherently exacerbated, partly as tried and tested financing arrangements for routine programming need to be complemented by new ad hoc processes, depending on the country’s ­strategy. Both Lesotho and Mali, for instance, have faced challenges in releasing the funds pledged to interventions after disasters ­occurred. In the Philippines, too, despite the existence of contingency financing mechanisms for disaster response, there were administrative delays in releasing the emergency funding allocated to the department overseeing disaster response after Typhoon ­Haiyan. Therefore, robust processes need to be put in place for anticipating and releasing funds (O’Brien et al. 2018b), in line with the contingency planning highlighted in chapter 2. • Transfer of funds to the local l­ evel. The delays in upstream funding can be further exacerbated if processes for ensuring liquidity at the local level are not in ­place. In practice, this will depend on the overarching payment modality of any existing social protection programs in the country and the selected payment mechanism for shock response (if not piggybacking on existing ­systems). E-systems pose some advantages but may be disrupted by shocks ­themselves. Evidence from existing experiences maintaining the routine delivery schedule in the aftermath of a shock has shown that liquidity can be a major constraint for payment agents—stressing, again, the need for preparedness measures and contingency plans to ensure liquidity is rapidly available down to the local l­evel. In terms of over-the-counter payments to beneficiaries, this may include plans for moving currency from headquarters and regional hubs to provincial and district distribution networks (such as local automated teller machines or pay ­agents). ­ • Reconciliation. In the reconciliation process of routine social protection ­payments, the amount paid to the payment provider (whether a private

Finance: Applying a Disaster Risk Financing Approach | 93

contractor or a government counterpart) is reconciled with the amount ­disbursed by the provider to ­beneficiaries. If data management is electronic (for example, using a program management information system), a reconciliation statistics report is drafted and analyzed to identify to whom transfers were paid and whether any inconsistencies emerge (Barca et al. ­2017). Robust processes for postdisaster reconciliation of payments can better ensure limited leakage and c­ orruption. Moreover, when funding comes from separate budget envelopes, such as from development and humanitarian partners, the underlying financial procedures outlining the source and flow of funds, including reconciliation requirements, are often ­different. This poses ­challenges for a timely response due to differing donor accountability ­requirements. In such instances, establishing common processes with key nongovernmental partners in advance of a shock will be important to encourage alignment and ultimately expedite the processing of payments to beneficiaries after a ­shock.

NOTE 1. These are prioritized government expenses that are honored and p ­ redictable.

REFERENCES ADB (Asian Development ­Bank). ­2018. “Strengthening Resilience through Social Protection ­Programs.” Guidance ­Note. ADB, ­Manila. Barca, Valentina, Thea Westphal, and Veronica Wodsak. 2 ­ 017. “Administration of Non-Contributory Social Protection: Delivery Systems—Manual for a Leadership and Transformation Curriculum on Building and Managing Social Protection Floors in A ­ frica.” Transform (Transformation Curriculum on Building and Managing Social Protection Floors in Africa). Cabot Venton, ­C. ­2018. “Economics of Resilience to Drought in Ethiopia, Kenya and Somalia: Executive S ­ ummary.” Center for Resilience, U ­ .S. Agency for International Development, Washington, ­DC. Cabot Venton, ­C., and ­L. S ­ ida. ­2017. “The Value for Money of Multi-Year Humanitarian Funding: Emerging ­Findings.” Department for International Development, ­London. CCRIF ­SPC. ­n.d. 2017–18 Annual R ­ eport. Grand Cayman: CCRIF ­SPC. ­https://www.ccrif.org​ /­sites/default/files/publications/CCRIF_Annual_Report_2017_2018_0.pdf. Clarke, ­D., and ­S. ­Dercon. ­2016. Dull Disasters? How Planning Ahead Will Make a ­Difference. Oxford, UK: Oxford University ­Press. Clarke, ­D., and ­R. ­Hill. ­2016. “Disaster Risk Financing as a Tool for ­Development.” ­Report. World Bank and Disaster Risk Financing and Insurance Program, Washington, D ­ C. ­https://www​ .gfdrr.org/sites/default/files/publication/DisasterRisk.pdf. Clarke, D ­ ., and R ­ . Vargas H ­ ill. 2 ­ 013. “Cost-Benefit Analysis of the African Risk Capacity ­Facility.” IFPRI Discussion Paper ­01292. International Food Policy Research Institution, Washington, ­DC. Financial Protection ­Forum. ­2018. “Disaster Risk Finance: A Primer, Core Principles and Operational ­Framework.” World Bank, Washington, ­DC. ­https://financialprotectionforum​ .­o rg/publication/disaster-risk-finance-a-primercore-principles-and-operational​ -framework. Fitzgibbon, ­C. ­2016. “Shock-Responsive Social Protection in Practice: Kenya’s Experience in Scaling up Cash ­Transfers.” Humanitarian Policy Group, Overseas Development Institute, ­London. ­https://odihpn.org/blog/shock-responsive-social-protection-in-practice-kenyas​ -experience-in-scaling-up-cash-transfers/.

94 | Adaptive Social Protection

Ghesquière, ­F., ­O. Mahul, ­M. Forni, and ­R. ­Gartley. ­n.d. “Caribbean Catastrophe Risk Insurance ­Facility.” International Aid + Trade I­ AT03-13/3. World Bank, Washington, D ­ C. h ­ ttp:// siteresources.worldbank.org/PROJECTS/Resources/Catastrophicriskinsurancefacility​.­pdf. IASC (Inter-Agency Standing ­Committee). 2 ­ 016. “Too Important to Fail—Addressing the Humanitarian Financing Gap: High-Level Panel on Humanitarian Financing Report to the ­Secretary-General.” United Nations Office for the Coordination of Humanitarian Affairs, ­Geneva. Kardan, A ­ ., ­C. O’Brien, and M ­ . ­Masasa. 2 ­ 017. “Shock-Responsive Social Protection Systems Research Case Study: ­Lesotho.” Oxford Policy Management, Oxford, U ­ K. Maher, ­B. ­2018. “Financing Adaptive Social Protection Financial Tools and Approaches for Adaptive Safety ­Nets.” Presentation at World Bank South–South Learning Forum, ­Frankfurt. ­http:// pubdocs.worldbank.org/en/827871520537561050/SSLF18-Financing-ASP-Framing.pdf. Maher B ­ ., ­C. Fitzgibbon, and A ­ . ­Solórzano. 2 ­ 018. “Emerging Lessons in Financing Adaptive Social ­Protection.” Background Paper for the World Bank, Oxford Policy Management, ­London. O’Brien, ­C., J ­ . Congrave, K ­ . Sharp, and ­N. K ­ eïta. 2 ­ 018a. “Shock-Responsive Social Protection Systems Research: Case Study—Social Protection and Humanitarian Responses to Food Insecurity and Poverty in ­Mali.” Oxford Policy Management, Oxford, ­UK. O’Brien, ­C., ­R. Holmes, ­Z. Scott, and ­V. ­Barca. ­2018b. “Shock-Responsive Social Protection Systems Toolkit—Appraising the Use of Social Protection in Addressing Largescale S ­ hocks.” Oxford Policy Management, Oxford, ­UK. O’Brien, ­C., ­Z. Scott, ­G. Smith, ­V. Barca, ­A . Kardan, ­R. Holmes, ­C. Watson, and ­J. ­Congrave. ­2018c. “Shock-Responsive Social Protection Systems Research: Synthesis R ­ eport.” Oxford ­ K. h ­ ttps://www.opml.co.uk/files/Publications/a0408​ Policy Management, Oxford, U -shock-responsive-social-protection-systems/srsp-synthesis-report.pdf?noredirect=1. Uganda, Government of. 2 ­ 016. “Third Northern Uganda Social Action Fund (NUSAF3) Project— Disaster Risk Financing Sub-Component ­Handbook.” Government of Uganda, ­Kampala. UN OCHA (United Nations Office for the Coordination of Humanitarian A ­ ffairs). ­2018. Global Humanitarian Overview ­2018. New York: UN ­OCHA. ­https://www.unocha.org/sites​ /­unocha/files/GHO2018.PDF. Wilkinson, ­E., ­L. Weingartner, ­R. Choularton, ­M. Bailey, ­M. Todd, ­D. Kniveton, and ­C. Cabot ­Venton. 2 ­ 018. “Forecasting Hazards, Averting Disasters: Implementing Forecast-Based Early Action at ­Scale.” Overseas Development Institute, ­London. Woldemichael, ­B. ­2018. “New Ways of Working: Linkages between Humanitarian Assistance and the Productive Safety Net Program in ­Ethiopia.” Presentation at World Bank South– ­ rankfurt. ­http://pubdocs.worldbank.org/en/304641520538162620​ South Learning Forum, F /­SSLF18-Humanitarian-Assistance-Ethiopia.pdf. World ­Bank. ­2017. Sovereign Climate and Disaster Risk Pools: World Bank Technical Contribution to the ­G20. Washington, DC: World ­Bank.

4

Institutional Arrangements and Partnerships MULTISECTORAL COORDINATION AND HUMANITARIAN LINKAGES

OVERVIEW Adaptive social protection (ASP) is an inherently multisectoral undertaking, requiring coordination across a broad range of actors who are engaged in building the resilience of vulnerable populations, and whose expertise and resources will be critical to the advancement of ASP in any ­country. This includes, most prominently, those in the disaster risk management (DRM) sector as well as those operationalizing climate change adaptation ­initiatives. To promote the required coordination, ASP needs to have coherence and complementarity across the policies and strategies of these s­ ectors. Concretely, social protection strategies and frameworks can provide the foundation for pursuing objectives related to A ­ SP. However, an exclusive focus on articulating the function of social protection in building resilience within its own sectoral strategies can result in overlapping mandates, competition for resources, and a limited impact on resilience b ­ uilding. Clear policy commitments that align the overall objectives of these different sectors along with the specific mandates, roles, and responsibilities of the actors involved will help to provide the policy framework to encourage increased ­coordination. At the same time, policy commitments will need to be backed by appropriate financial resources and investments in the required capacity for those tasked with delivery to be c­ redible. Moreover, as the role of safety net programs in response to previously humanitarian crises and shocks increases, the closer integration of social protection actors and programs within the humanitarian system, particularly in settings of limited national capacity and/or particularly severe shocks, increasingly comes to the ­fore. Evolutions in the humanitarian system enshrined in the World Humanitarian Summit’s Grand Bargain (2016) promote the expanded use of cash transfers and, in doing so, a commitment to “increase SP [social protection] programs and strengthen national and local systems to build resilience in fragile ­contexts.” Operationally, unbundling a clear delineation of roles and responsibilities for the delivery of resilience-building programs between government and humanitarian actors, based on the specific context and relative comparative advantages, will help to create actionable, strategic partnerships for the advancement of ASP across the government–­ humanitarian ­divide.

Institutional arrangements and partnerships

This chapter draws extensively from background papers by Bailey (2018) and Kardan (2018).  95

96 | Adaptive Social Protection

MULTISECTORAL COORDINATION FOR ASP A defining feature of ASP is the many actors within government that may be involved in its ­implementation. The inherent multidisciplinary and interagency nature of resilience building across the three capacities of preparedness, ­coping, and adaptation requires diversified expertise and coordination among ­actors. Indeed, the number of potential actors and complementary programs aligned to ASP objectives calls for institutional arrangements to anchor the planning, management, and delivery of this a­ ssistance. In practice, the development of ASP in many countries has shifted attention from a singular focus on national social protection systems, the policies that guide them, and the organizations that deliver social protection programs to a wider focus inclusive of the policies, organizations, and programs involved in DRM and climate change ­adaptation. Strong government leadership is necessary to ensure the coordination of the often disconnected actors, based on a clear articulation of respective roles and ­responsibilities. Concretely, governments lead by setting resilience-related objectives in policies and strategies, inclusive of social protection, DRM, and climate change adaptation, among other ­sectors. Policy commitments can work to instill the necessary budgetary allocations for national ministerial structures to translate objectives into outcomes among vulnerable h ­ ouseholds. In practice, government leadership also includes establishing the standards and procedures to guide the integration of nongovernmental organizations (NGOs) and humanitarian actors in ASP ­implementation. This chapter begins with an overview of the key considerations around the requisite institutional arrangements among national actors before moving to focus more squarely on the creation of strategic partnerships with nongovernmental ­actors.

Establishing policy coherence for adaptive social protection National social protection policies and strategies can provide the foundation for A ­ SP. Most countries have social protection policies and strategies that set out the government’s ­vision. The extent to which these policies and strategies are rooted in legislation varies (see, for example, Beegle, Coudouel, and Monsalve ­2018). The functions of social protection often are equity, which provides protection against deprivation; resilience, which is insurance against shocks; and opportunity, which seeks to promote human capital and access to income earning opportunities (World Bank 2012; see also Devereux and Sabates-Wheeler 2004, who set a similar framework of protection, prevention, promotion, and ­transformation). Articulated in this manner, these policies and strategies provide a foundation for the aims of ­ASP. The strategies of other sectors such as—prominently—DRM also may ­support the advancement of A ­ SP. For example, the ASP objective of building the resilience of poor and vulnerable households to shocks overlaps with the objectives set out in national DRM policies and s­ trategies. This is particularly true given the ongoing shift from response to disaster preparedness within the DRM ­community.1 In Kenya, ASP emanated out of the government’s resolve to address poverty and vulnerability as a cause of and outcome from drought emergencies; a framework developed by the government (Ending Drought Emergencies) laid the policy foundation for the Hunger Safety Net Program

Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages | 97

which scales and retracts vertically and horizontally for drought e­ mergencies. Where the political ­appetite for social protection is low, national DRM policies, and the DRM sector more broadly, can provide a foundation for introducing ­ASP. This suggests a government’s commitment to ASP can come from sectors other than social protection itself (boxes 4 ­ .1 and ­4.2). Beyond the DRM sector, ASP commitments are found in climate change adaptation policies aiming to promote the livelihoods of the poorest ­households. For example, social protection has been singled out as a key component to climate-change adaptation (Stern 2006) and thus countries seeking to adapt to climate change may opt to invest in ­ASP. Ultimately, policy coherence within individual sectors provides the basis for coordination across ­sectors. A starting point to improve cross-sectoral coordination is to improve policy coherence within each ­sector. A more internally consistent sector provides clearer opportunities for cross-sectoral collaboration (O’Brien et a­ l. ­2018). Sectoral policy frameworks provide the space to articulate roles and responsibilities within and across sectors and enable the establishment of appropriate platforms for ­coordination. A key challenge of coordinating social protection and other sectors is that each sector often is itself multisectoral and in need of coordination (OPM ­2017). Covariate shocks themselves can create momentum for policy reform in support of A ­ SP. For example, the evolution in Japan’s legal framework for DRM can be traced to specific shocks (figure 4 ­ .1). The Basic Act was enacted in 1961 after major typhoons caused extensive damage in a southwestern part of Japan in ­1959. The Act has since been amended several times, with at least 10 specific amendments in the last 5 years, following the East Japan Great

BOX ­4.1

Pacific Island countries: Disaster risk management (DRM) institutions can help to drive the ASP agenda The commitment to ASP objectives may be located in a DRM policy and driven forward by a DRM agency, particularly in regions or countries where exposure to disasters is a key determinant of ­poverty. In the Pacific Region, there appears to be an appetite for this type of institutional arrangement for A ­ SP. All recognize the importance of DRM and have put in place DRM legislation and plans since the ­mid-1990s. By contrast, no Pacific Island countries have legislation governing social protection and few have agencies explicitly tasked with managing social protection ­programs. As a result, some countries have expressed interest in embedding elements of the ASP agenda in Natural

Disaster Management Acts and having the equivalent National Disaster Management Office in charge of coordinating a ­response.a This is particularly the case for such countries as Vanuatu, the Solomon Islands, and Samoa which are reluctant to allocate funding for regular social safety net programs (due to financing constraints or concerns around the impact on informal support networks) but clearly recognize the importance of having the delivery systems in place required to target transfers following large-scale ­disasters. While this type of institutional arrangement has not yet been fully tested, it could prove to be suitable in regions where social protection institutions have limited capacity or are ­nonexistent.

a. These priorities were formally set out during the inaugural Pacific Shock-Responsive Social Protection Conference, which was held in Nadi, Fiji, in ­2018.

98 | Adaptive Social Protection

BOX ­4.2

The Philippines: Social protection embedded in disaster risk management (DRM) frameworks The Philippines has a well-developed social protection system and a comprehensive legislative and institutional arrangement governing D ­ RM. The social protection sector uses a common definition of social protection that is understood by all actors and is spearheaded by the Department of Social Welfare and Development ­(DSWD). The respective policies of each sector explicitly recognize the role of social protection in disaster ­response. The DRM sector has a well-established and strong institutional framework articulated through the Disaster Risk Reduction and Management Act of ­2 010. The National Disaster Risk Reduction and Management Framework of 2011 and the National Disaster Response Plan, developed in 2015 based on the lessons learned from Typhoon Haiyan, elaborate the structures, roles, and responsibilities at different levels of government and across different ­departments. Within this framework, the DSWD is the lead on disaster ­response. The social welfare and development officers at the regional, provincial, and municipal levels are part of the action teams responsible for supporting disaster ­response. Beyond immediate response, DSWD also plays an important role

during the rehabilitation and recovery phases following a disaster, with interventions and services that contribute to longer-term r­ ecovery. At the same time, the DSWD is the co-lead of the international humanitarian cluster system designed to coordinate government and humanitarian actors, and is the ­government’s lead agency for the protection, food, and shelter ­clusters. As such, social protection and DRM have become more closely interlinked over time in the Philippines through an iterative process that has captured learning from each emergency ­response. Central to this policy coherence and coordination is high-level political commitment, strong legislative and regulatory backing, clear roles and responsibilities, and well-established coordination s­ tructures. Having the DSWD as the lead agency for social protection as well as emergency response has provided a seamless link between the DRM, humanitarian, and social protection ­actors. However, this has not been without its challenges, with capacity constraints straining the DSWD’s ability to effectively manage all its obligations during large shocks such as Typhoon Yolanda in ­2013.

Sources: Bowen 2015; Smith et ­al. ­2017.

Earthquake in 2011, to refine the roles and responsibilities across actors and ultimately the effectiveness of ­D RM. In Ethiopia, the government launched the Productive Safety Net Programme (PSNP) in 2005 in response to p ­ ersistent drought in rural areas and associated food insecurity, which had plagued the country over an extended period of ­time.

From policy coherence on paper to coordination in practice Multisectoral coordination for ASP requires clear institutional guidelines for program ­delivery. National social protection systems already include a range of service providers to support program i­mplementation. For example, the data that inform program participation can be generated by government systems or contracted o ­ ut. Payments are often made by banks or telephone companies (for e-payments) through a contractual arrangement with the ­government. Additional services, such as livelihood support, may be provided by local ­government institutions or by ­NGOs. NGOs or communities may operate a ­program’s grievance redress ­mechanism.

Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages | 99

FIGURE ­4.1

Japan: History of postdisaster DRM reforms and ASP legislation Background 1946 Earthquake

Evolution of legislation (summary)

Necessity of swift response, defining the role of national/local governments

1947

Disaster Relief Act

Necessity of comprehensive DRM system (both physical and social)

1961

Disaster Countermeasures Basic Act

Necessity of consistent and swift approach to severe disasters

1962

Act on Special Financial Support to Deal with the Designated Disaster of Extreme Severity

1964 Earthquake

Fire insurance (private) could not cover exceptional large-scale disasters; necessity of public intervention

1966

Act on Earthquake Insurance

1967 Heavy rain

Discussions: Difficulty to support individuals versus heavier damage to vulnerable people (took 6 years) → As condolence grant

1973

Act on Provision of Disaster Condolence Grant (cash transfer)

Necessity of institutional arrangement and collaboration after severe disasters → HQs in national and local governments

1995

(Amendment) Disaster Countermeasures Basic Act

Civic movement complaining about the necessity of safety net (firstly to target by income, but eliminated at amendment 2007)

1999 2004

(Amendment) Act on Support for Reconstructing Livelihoods of Disaster Victims

Various lessons, i.e., necessity to support people requiring support for evacuation and at a shelter, breakdown of local governments

2013

(Amendment) Disaster Countermeasures Basic Act

1959 Typhoon

1995 Earthquake

2011 Earthquake tsunami

Source: Kawasoe, ­forthcoming. Note: DRM = disaster risk ­management.

The choice of the ministry responsible for coordinating social protection tends to determine the level of government o ­ versight. Social protection ­programs are often housed in central ministries (Ministry of Finance, Office of the Prime Minister), social ministries (Ministry of Social Welfare), and other sectoral m ­ inistries. Management by central ministries situates social protection close to the locus of power and authority, which can result in secured (and possibly increased) financing and advancement of the agenda but oversight by staff who may be less supportive to the program’s ­aims than are dedicated social protection staff. Management by a social ministry can result in staff delivering the program or advancing the agenda who are supportive of social protection objectives but may not have the direct access to the resources of central ministries (Beegle, Coudouel, and Monsalve ­2018). The overall effectiveness of ASP rests on government-led coordination among and delivery modalities within programs. A number of factors can contribute to more effective programmatic coordination in responding to shocks, including the existence of and support through a single central agency that coordinates response and works closely with other national agencies and

100 | Adaptive Social Protection

subnational authorities; the presence of dedicated, multisectoral technical working groups; existence of DRM strategies and response plans with clearly identified roles and responsibilities; and advanced planning for coordination of implementation during crises and its incorporation within the program’s standard operating procedures (operational manual) (OPM 2 ­ 017). Importantly, coordination across programs (and even across sectors) requires strong, sustained government leadership through oversight agencies. The introduction of national standards or a coordination framework can help to facilitate the required ­coordination. Because ASP spans a number of sectors and involves a number of ministries and/or NGOs, each of these organizations will have its own management systems as well as incentives and a­ ccountabilities. While these differences may lead to fragmentation (and even competition), the creation of national standards or frameworks may be able to bring coherence and consistency across ASP programs, while also drawing on the strengths of different ­organizations.2 Examples of the application of harmonized approaches or frameworks for various aspects of ASP design and delivery exist across a range of ­countries. For example, common transfer guidelines were developed in the Philippines and ­Lesotho. Such frameworks aim to introduce similar accountability across services as well as flows of information, both to the government and ­citizens. Box ­4.3 describes the framework that Ethiopia has put in place to create a national, shock-responsive safety ­net. Determining the roles and functions of each ASP program (or providers of each function within a program) in advance of any shock is v­ ital. Such an approach is critical given the large number of organizations involved in delivering different aspects of ASP programming and the overriding need for speed in response to most s­ hocks. In Pakistan, the Citizen Damage Compensation Program established partnerships with private sector commercial banks through a memorandum of understanding implemented as part of disaster preparedness plans (World Bank ­2013). The Federal Disaster Response Action Plan was developed following the Citizen Damage Compensation Program’s response to flooding in 2010, defining the cashbased response model for future shocks as well as the roles and responsibilities of the respective agencies critical for implementation, including the national and provincial disaster management authorities, the Benazir Income Support Program, the National Database and Registration Authority, the Ministry of Finance, and commercial ­banks. In Niger, the government implemented a national contingency plan for responding to weather-related shocks that includes ensuring access to food through social assistance, protection of household assets, and developing early warning indicators (Bastagli 2 ­ 014). Beyond establishing roles in advance, ensuring stakeholders’ active participation in joint planning can strengthen the knowledge and skills of each ­partner. Such participation also can contribute to creating a common understanding of the ­problems. National standardized guidelines for program implementation can facilitate vertical coordination from national to local ­actors. The degree of vertical coordination within a government from national to regional to local actors is determined by the broader government structure and the extent of its decentralization (see ­below). In Lesotho, for example, social protection is managed centrally, while emergency is the responsibility of district DRM ­teams. Regardless of the precise government system, vertical coordination from central to local levels is strengthened by clear roles and responsibilities for each

Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages | 101

BOX ­4.3

Ethiopia: Framework of a national, shock-responsive safety net In Ethiopia, the government needed an instrument with the capacity to respond to chronic and transitory food insecurity in rural areas of the c­ ountry. It turned to social protection and DRM policies to create a framework for a scalable safety net ­program. In 2018, the government of Ethiopia recognized that bringing the PSNP and Humanitarian Food Assistance (HFA) together into a common framework could help putting a scalable safety net system into ­practice. It operates based on the following guidelines: (1) the rural safety net is a government-led framework that allows three delivery modalities: the government, the United Nations (World Food Programme), and NGOs; (2) these three delivery modalities share a set of common procedures and common monitoring and evaluation mechanisms; (3) the procedures distinguish a “normal year” from an “emergency” situation for safety net transfers to be used (that is, a distinction which is required to trigger humanitarian assistance); and (4) financing includes developmental and humanitarian ­funds. Within this system: •

The National Disaster Risk Management Commission is responsible for conducting needs assessments and the allocation of resources for all transitory food ­needs. This includes humanitarian resources pledged in response to the Humanitarian Appeal and the geographic allocation of any resources channeled through the PSNP federal contingency ­budget. Through

•

•

this process, any transitory needs of core PSNP clients will be considered in the needs assessment and contingency planning and met through humanitarian ­financing. All implementing stakeholders use a common set of core procedures and structures to deliver both core PSNP transfers and ­HFA. This includes (1) procedures for targeting, (2) software to keep track of beneficiary lists and transfers, and (3) using common delivery structures at the district and community ­level. All transfer values will be set for “normal” periods and increase to the Sphere standards when an emergency is ­identified.a All resources channeled through the government are managed by the ministry with the appropriate mandate and ­skills. Therefore, (1) all cash transfer resources (whether for the PSNP or HFA) channeled through the government flow through the Ministry of Finance, while cash payments are the responsibility of District Finance Offices; and (2) all food transfer resources (whether for the PSNP or HFA) channeled through the government flow through the National Disaster Risk Management Commission’s Logistics Department with the relevant disaster management desk at the district level responsible for overseeing the appropriate storage and distribution of food at the subdistrict ­level.

a. The Sphere movement was started in 1997 by a group of humanitarian professionals aiming to improve the quality of humanitarian work during disaster ­response. With this goal in mind, they framed a Humanitarian Charter and identified a set of humanitarian standards to be applied in humanitarian ­response. The Sphere Handbook outlines common principles and universal minimum standards in humanitarian ­response.

level of government as well as clear lines of accountability and ­resourcing. (For a general discussion of vertical coordination for social protection, see Beegle, Coudouel, and Monsalve 2018; Barca et al. ­2017.)

Investing in the capacity to deliver ASP programs For national actors to deliver ASP, significant investments in capacity are required at both the national and local l­ evels. The extent to which the organizations responsible for program delivery have an effective local presence influences their ability to deliver program ­benefits. ASP programs require

102 | Adaptive Social Protection

front-line ASP staff to carry out a range of tasks, including coordination with other sectors to ensure that households receive complementary s­ ervices. For example, this includes the delivery of training to households or the provision of information or “accompanying measures” to strengthen ­nutrition. In Peru, the Haku Winay Program utilizes Yachachiq (trained local productive community leaders) to provide technical assistance, capacity building, and accompaniment of households to implement the program and build resilience to food ­insecurity. In Brazil, the Social Assistance Center (Centro de Referencia de Assistencia Social—CRAS) provides social assistance at the municipal level; households that are beneficiaries of Bolsa Familia and registered in the social registry are assisted by the Social Assistance Center and can be included in the other services, policies, and programs delivered by the Brazilian Unified Social Assistance ­System (WWP ­2019). The degree of governmental decentralization affects the central–local relationship and the capacity to respond to shocks with ­ASP. In a devolved, decentralized setting where the state transfers authority for decision-making and management from central government to autonomous units of subnational, local government, any reliance on local financing, variations in needs and resources are likely to lead to inconsistencies in service provision between areas unless equalization mechanisms are put in place (Barca et al. 2 ­ 017). These may be mitigated if they are understood and reflected in the legislative and policy frameworks of the country and where the relationships of the ­different actors at different tiers of government are clearly d ­ efined. In Ecuador, DRM has been decentralized to the local level; yet, the national guidelines stipulate that when a shock exceeds capacities at the local level, higher levels of government provide ­support. This was the case in 2016, when severe capacity constraints at the local level triggered a centralized response to a major ­earthquake (Beazley ­2017). When responding to shocks, the delivery capacity of ASP programs also needs to expand and ­contract. Most simply, to expand coverage in response to shocks, social protection programs will require more staff, vehicles, and equipment, for example, to serve more people often across a greater geographic ­area. Often, government systems and resources are required to respond at a time when these systems themselves may be under strain and further constrained (O’Brien et a­ l. ­2018). Identifying surge capacity to support the response processes is c­ ritical. This surge capacity may be obtained, for instance, horizontally across ­departments and ministries, vertically through higher tiers of government, or collaboratively with nonstate ­actors. The relative roles and responsibilities at each level of government, as well as how staff access information and resources, have an impact on the effectiveness of ASP p ­ rogramming. Semi-autonomous or autonomous central government agencies often have the mandate, and resources, to deliver services through field ­offices. In contrast, delivery through devolved national structures requires vertical coordination that may demand significant investments in systems and procedures in advance of any shock, as well as coordination among decision makers at different levels of g­ overnment. In El Salvador, Indonesia, and Mozambique, strengthening subnational DRM committees has been identified as a key strategy for strengthening the DRM systems (Harkey ­2014). In the Philippines, local government units lead disaster response with reserve funds (5 percent of their revenue), although national resources supplement this local-level financing in the case of large d ­ isasters. In Pakistan, in 2011, the federal structure devolved the

Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages | 103

responsibility for social protection and DRM to provincial governments; however, the Benazir Income Support Program (national cash transfer program) remained federal because it was established before the d ­ evolution.

CREATING STRATEGIC PARTNERSHIPS ACROSS THE GOVERNMENT-HUMANITARIAN DIVIDE At the same time, given the nature of covariate shocks, ASP highlights the importance of deepening linkages with humanitarian a­ ctors. In lower-capacity environments and contexts where there is no functioning, legitimate state with access to populations affected by a shock, the international humanitarian system often leads the provision of a­ ssistance. When states lack the capacity or willingness to ensure that the basic needs of its population are met, international humanitarian assistance often complements and sometimes substitutes for national and local c­ apacity. When protracted crises that result from intractable political instability and conflict are the norm, the leadership role of the humanitarian system often becomes longer term in n ­ ature. Some of the primary features of the international humanitarian systems are summarized in table ­4.1 along with their implications for ASP, which are explored in further detail, in ­turn.

TABLE ­4.1

International humanitarian system: Features and implications for adaptive social protection

FEATURE

CHARACTERISTICS

IMPLICATIONS

Policy commitments and the Grand Bargain

High-level policy support for building resilience and increasing the role of social protection

Opportunities to advance aims of ASP but need to be translated into more concrete and strategic actions

Bifurcation of humanitarian/ development and rise of resilience building

Humanitarian and development assistance often underpinned by different financing channels, coordination structures, mandates, and principles

Divide between humanitarian and development systems may remain an obstacle; need for specificity on “resilience building”

Humanitarian financing

Very little direct funding goes to national governments; significant flows to fragile and conflict settings, and year-on-year to the same places

Limited potential for humanitarian financing to be channeled to governments for national safety nets; scope to fund NGOs operating within national frameworks for ASP in some countries

Shares of humanitarian financing go to some areas and populations supported by national safety nets Humanitarian principles

Humanitarian assistance is guided mainly by the four principles of humanity, impartiality, independence, and neutrality Differing views on flexibility of principles exist, but they are not incompatible with working with governments

Humanitarian principles should inform the response function of ASP to shocks; can be referenced to advocate for a principled engagement around ASP with governments by humanitarian agencies

Coordination

Established mechanisms for coordination (see the cluster system) but varying coordination approaches because of differing levels of national involvement in those mechanisms

Increasing shift to cash transfers

Cash transfers increasingly accepted as mainstream Offers an entry point for engagement of national tool of humanitarian response, but programs often ASP programs with humanitarian system fragmented and still represent only a small share of total assistance

Source: Bailey 2 ­ 018.

Need for engagement of ASP at various levels of humanitarian operational and strategic coordination and for bilateral engagement with major donors and aid agencies

104 | Adaptive Social Protection

The Grand Bargain: High-level commitment in support of government–humanitarian integration The 2016 World Humanitarian Summit, and the resulting Grand Bargain, created high-level policy support to strengthen linkages with social p ­ rotection. More than 50 donors and aid agencies committed to the Grand Bargain, which is a series of 10 commitments to improve assistance to crisis-affected p ­ opulations. Among others, the donors commit to increasing the routine use of cash and to enhancing the engagement between humanitarian and development ­actors. This includes a commitment to “increase social protection programmes and strengthen national and local systems and coping mechanisms in order to build resilience in fragile contexts” (Grand Bargain 2016, ­14). However, the challenge remains to move from endorsements to concrete ­actions. In parallel, increased policy attention to building household resilience within the humanitarian community is highly favorable to advancing the aims of A ­ SP. The increasing focus on building resilience within the humanitarian community (see also appendix A) aims to promote adaptation and to respond to the interaction between chronic poverty and acute n ­ eeds. In Somalia, for example, efforts are under way to improve the continuity of activities across preparedness, relief, rehabilitation, and ­development. This includes providing a ­minimum level of social protection in response to chronic and seasonal food insecurity (Bailey ­2018).

“Humanitarian principles”: What are they and what are their implications for ASP? The provision of humanitarian assistance is informed by a set of p ­ rinciples. Primarily, these principles include humanity (alleviating suffering wherever it is found), impartiality (assisting solely on the basis of need), neutrality (not taking sides in a conflict), and independence (operating freely from political and military agendas). In some instances, commitment to humanitarian principles can create tensions with supporting governments to assist and protect people in their territory if the government is not working in the interest of its people, in opposition to the principles of neutrality and impartiality (FAO 2016; Harvey ­2009). In other instances, humanitarian principles can be invoked by humanitarian actors to justify not engaging with governments when there may be many possible—and necessary—relations with states and governments in a given context (Harvey 2009; Levine and Mosel ­2014). Humanitarian principles should play a role in guiding the design and implementation of ­ASP. Engaging with humanitarian actors necessitates speaking the language of humanitarian principles and understanding how these inform ­actions. The principles exist to help ensure that organizations can access people affected by crisis and support those most in need, which is ultimately at the heart of the ASP resilience-building o ­ bjectives. At the same time, the humanitarian community will need to continue aligning with, leveraging, and complementing efforts to support national systems where appropriate and consistent with humanitarian principles (World Bank ­2016). While tensions exist, it is possible to remain committed to principles underpinning both humanitarian and ­development-oriented ASP initiatives (Harvey ­2009). That is, the humanitarian principles should provide a basis for deepening linkages and coordination as opposed to a ­roadblock.

Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages | 105

Coordination mechanisms: Streamlining government and humanitarian interventions Coordination is a perennial challenge with the humanitarian system ­itself. Recent estimates reveal that international humanitarian assistance involves some 4,480 actors drawing from international and national NGOs, civil society, and others (Development Initiatives 2 ­ 018). Numerous organizations provide similar and ­different types of aid in the same locations and thus must coordinate to minimize gaps and overlap in meeting needs (Clarke and Campbell ­2018). To address these issues, the United Nations system has a well-­institutionalized, country-level “cluster” coordination system articulated by sectors and an array of institutional governing ­arrangements. Clusters are groups of humanitarian organizations, both UN and non-UN, in each of the main sectors of humanitarian action (such as water, health, and ­logistics). They are designated by the InterAgency Standing Committee and have clear responsibilities for the coordination of aligned actors3 (figure 4 ­ .2).

FIGURE ­4.2

United Nations cluster system: Overview Health WHO Food security WFP & FAO

Logistics WFP

Emergency telecommunications WFP

Protection UNHCR

Shelter IFRC/UNHCR ry

n tio iga Mit

Early recovery UNDP

Camp coordination and camp management IOM/UNHCR

Water, sanitation, and hygiene UNICEF

Pr ep are dn ess

Re co ve

Prevention

Humanitarian and emergency relief coordinator

Education UNICEF & Save the Children

Reconstruc tion

Nutrition UNICEF

se on sp Re

Disaster

Source: United Nations Office for the Coordination of Humanitarian Affairs S ­ ervices. Note: ECHO = European Civil Protection and Humanitarian Aid Operations; FAO = Food and Agriculture Organization of the United Nations; IFRC = International Federation of Red Cross and Red Crescent Societies; IOM = International Organization for Migration; UNDP = United Nations Development Programme; UNHCR = United Nations High Commissioner for Refugees; UNICEF = United Nations Children’s Fund; USAID = United States Agency for International Development; WFP = World Food Programme; WHO = World Health Organization.

106 | Adaptive Social Protection

Where ASP intersects with the humanitarian system, these humanitarian coordination mechanisms will be instrumental in enhancing p ­ artnerships. Coordination of international humanitarian interventions at the country level is carried out through Humanitarian Country Teams (comprised of senior leadership from international organizations and sometimes donors) and ­clusters. The cluster system serves as the mechanism for sector-based planning and implementation discussions with state and nonstate ­actors. Clusters are tasked with developing strategies for their sectors, which in practice tends to be done by aligning the autonomous programs of different aid agencies (Clarke and Campbell 2 ­ 018). In the case of disaster response, for example, in some cases, coordination is government-led, with clusters or similar working groups led by government sectoral agencies, as in Mozambique (Kardan et ­al. ­2017). In other cases, coordination is a hybrid of international and government leadership, with clusters being co-led by government departments, as noted earlier in the case of the Philippines (Smith et a­ l. ­2017). However, the risk of having parallel systems and lack of coordination ­remains. Coordination can lead to and be enhanced by the sharing of data among social protection and humanitarian actors—and can elicit improved knowledge, reduce duplication of efforts, and lead to cost s­ avings. In Turkey, refugees were enrolled into the Emergency Social Safety Net Program through the country’s existing information management ­platform. In the Philippines, the Department of Social Welfare and Development set up a system for coordinating data on typhoon-affected households assisted by aid agencies with its own ­database. This coordination led to the enrollment of an additional 20,000 households into a government cash transfer program (Smith et ­al. ­2017). This process also helped track which households were receiving which benefits from humanitarian agencies (Bowen ­2015). In the Hurricane Irma and Maria responses in the Caribbean, similar attempts were made through bi-lateral relationships between government departments and aid ­agencies. However, there were trade-offs between speed and using social protection ­systems. In Dominica, UNICEF and the World Food Programme worked together to support the delivery of cash transfers through the Ministry of Social Services to provide scaled-up support to beneficiary households and horizontal expansion to affected nonbeneficiary households in the aftermath of Hurricane M ­ aria. In Dominica and in Antigua and Barbuda, the International Federation of Red Cross and Red Crescent Societies supported the distribution of cash transfers using prepaid debit cards (already procured as part of regional preparedness measures) and the assistance was t­ imelier. In the British Virgin Islands, two NGOs worked with the Ministry of Health and Social Development to undertake an emergency cash transfer program and to design data collection tools to inform future registration in social protection ­programs. Even so, implementation faced ­delays.

Cash transfers: An entry point for bridging the government-humanitarian divide The humanitarian sector’s pivot toward cash transfers over the past decade has helped bridge the “humanitarian–social protection ­divide.” Cash transfers were virtually nonexistent in the humanitarian sector before 2005, whereas cash and vouchers together accounted for ­$2.8 billion (10 percent) of humanitarian assistance in 2016 (CaLP 2 ­ 018). The fact that social protection programs often

Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages | 107

transfer money to populations has led to numerous calls to utilize these systems to a greater extent in times of emergencies and seasonal shocks (see, for example, Bailey and Harvey 2015; Grand Bargain 2016; World Bank 2 ­ 016). There are cases where humanitarian cash transfers have either paved the way for social protection programs or strengthened the foundations and improved a country’s ability to deliver ASP (box ­4.4).

Moving beyond dichotomies: Unbundling specific roles and responsibilities to create strategic partnerships Conceptually, government and humanitarian actors are often viewed simplistically in “either-or” ­terms. In addition to the entry points for ASP highlighted above, moving beyond a broad commitment to coordination and increased partnerships requires a nuanced ­approach. It outlines each actor’s specific contributions in terms of delivering assistance to the poor and vulnerable to strengthen their preparedness, coping, and adaptation ­capacities.

BOX ­4.4

Cash transfers: A vehicle for ASP development across the government–humanitarian divide Humanitarian cash transfers can encourage change and reform in the humanitarian system, provide an opening for the development of ASP systems, and result in improved coordination among humanitarian and government a­ ctors. The recent Panel on Humanitarian Cash Transfers and other high-level initiatives, most notably the World Humanitarian Summit, established unequivocally that cash transfers are of strategic importance in improving humanitarian ­response. As of 2015, approximately 7 percent of international humanitarian assistance is channeled toward either cash transfers or vouchers (ODI 2 ­ 015). The Summit established a target for the percentage of humanitarian resources to be provided in cash by 2030, ensuring a responsibility among all actors to align and work ­together. This shift is one of the most fundamental reforms of the humanitarian system over the past ­decades. There are multiple cases where humanitarian cash transfers have either paved the way for social protection programs or strengthened the foundations and improved a country’s ability to deliver ­ASP. •

In the Democratic Republic of Congo, the introduction and expansion of cash-based responses has been the biggest evolution in

Sources: CaLP 2018; ODI ­2015.

•

humanitarian assistance in the last decade (ODI ­2015). The evidence base built through these initiatives helped establish the case for a national safety net system, which is being financed through the World Bank (Lisungi Safety Nets System Project) and will assist the country in responding to future ­crises. In Nepal, following the earthquake in 2015, the government and the humanitarian actors made great efforts to coordinate humanitarian cash transfers to ensure uniform benefit amounts and coverage of all ­affected. UNICEF financed a vertical and horizontal expansion of the Social Security Allowance, a cash transfer, in the affected areas (CaLP ­2018). Following this, Nepal has engaged in a broad dialogue on ASP and started to establish the building blocks that would promote mobilization of cash transfer programs for disaster response—a robust database of beneficiaries and vulnerable households, electronic payment systems, and institutional coordination across social protection and disaster response ­agencies. As a result, Nepal is expected to be better equipped to respond to future ­disasters.

108 | Adaptive Social Protection

Identifying the precise roles and responsibilities of government and humanitarian actors can help establish actionable, operational partnerships for the delivery of ­ASP. A framework laid out by Seyfert et ­al. (2019) attempts to facilitate the identification of workable pathways for progress among national and humanitarian actors (figure 4 ­ .3). Instead of falling back on the “either–or” choice, the framework lays out four strategic options (parallel systems, alignment, piggybacking, and national-led ­systems). It also discusses how collaborations may emerge around select programmatic “functions” and the “degrees” of possible connection between national and humanitarian actors within a given ­function. While a work-in-progress, such a granular analytical approach holds the potential to move beyond strategic dialogue and strategies in support of coordination; that is, coordination toward an operationally relevant delineation of roles and responsibilities based on relative comparative advantages in differing country ­contexts. The exact arrangements will differ across countries and contexts, with no predetermined division of ­labor. Indeed, the creation of actionable, strategic partnerships for ASP requires clear institutional guidelines for d ­ elivery. National social protection systems already include a range of service providers to support program implementation along the delivery c­ hain. For example, the data that inform program participation can be generated by government systems or contracted ­out. The number of organizations may increase as ASP continues to e­ volve. The additional capacity that is needed to identify, enroll, and pay an increasing number of people can be contracted in or o ­ ut. This notion of contracting out extends to ASP and humanitarian actors, who may have greater capacity or expertise at key points along the delivery chain in certain contexts (table ­4.2).

FIGURE ­4.3

Framework: Mix of delivery approaches across government and humanitarian actors Parallel systems

Alignment

Financing Legal and policy framework Setting eligibility criteria and qualifying conditions Setting transfer type, level, frequency, duration Governance and coordination Outreach Registration Enrollment Payment Case management Complaints and appeals Protection VAM/M&E Information management Source: Seyfert et ­al. ­2019. Note: VAM/M&E = Vulnerability analysis and mapping/monitoring and e ­ valuation.

Piggybacking

National-led systems

Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages | 109

TABLE ­4.2  Kenya and Ethiopia: Features that influence the use of government systems for shock response versus humanitarian actors or NGOs FEATURE

TYPES

INTERACTION WITH OTHER FEATURES

EXAMPLES

Source of financing

• Development partner humanitarian funds • Development partner development funds • Domestic government financing

Managing agency: Higher probability of development funds being channeled through government departments and humanitarian funds through ­WFP/ NGOs. When humanitarian funds are channeled through a government agency these tend to be through a disaster management agency rather than a sectoral ­ministry.

• Humanitarian funds: {{ ECHO financing of cash transfers in Ethiopia and Kenya (largely through NGOs/WFP) {{ USAID support for food aid in Ethiopia and Kenya (through NGOs/WFP) {{ European Union financing for the scale-up of HSNP transfers in Kenya • Development funds: {{ Development partner contributions to the federal contingency budget of the PSNP in ­Ethiopia {{ UK Department for International Development financing for the scale-up of HSNP transfers in Kenya • Domestic government financing: {{ Ethiopia government budget contributions to food assistance coordinated and implemented by the disaster management agency {{ Kenya government budget for emergency cash transfers managed by department of special programs, targeting coordinated by the National Disaster Management Authority

• National government led system • Ad hoc individual agency led

Strong coordination by government: • Well-financed government led system: Ethiopia, regarding both geographic Ethiopia through the Humanitarian Requirement resource allocation through needs Document leading to a unified response plan assessment and management of implemented by various ­actors. response • Individual agency led: HSNP scale-up in Kenya, which uses a separate Weak coordination by government: system of triggers to the standard needs Kenya, coordination of early warning ­assessment. is fairly strong, but there is little or no

Needs assessment/ response planning

Coordinating institution: Humanitarian funds are more likely to be channeled through agencies that are coordinated by a disaster management agency and development funds through a sectoral ­ministry.

ability to influence actions of implementing agencies and therefore avoid duplications and minimize gaps; also, each implementing agency tends to use its own procedures Managing agency or execution

• Government • Outside of government: {{ NGO {{ WFP

In combination with existence or absence of strong coordinating body can affect use of standard procedures (whether needs assessment or implementing procedures) and thus harmonization of ­response.

• Government: Management of cash and food response through the PSNP • NGO: Joint Emergency Operations, financed by USAID, in Ethiopia operates within the government framework • WFP: Food distribution in Ethiopia operates within the government framework • WFP-government combination: Cash transfers in Ethiopia (until recently, has used an ad hoc government funding channel) • WFP-NGO combination: Food distributions in Kenya continued

110 | Adaptive Social Protection

TABLE 4.2, continued FEATURE

TYPES

INTERACTION WITH OTHER FEATURES

EXAMPLES

• Nationally owned and set: Ethiopia emergency response and PSNP scale-up • Ad hoc, individual agency-defined procedures: {{ In Kenya, most of the emergency response, although WFP has some standardized procedures {{ In Ethiopia, emergency cash transfers have less standardized procedures than emergency ­food.

Implementing • Nationally owned, procedures standardized procedures for targeting, benefit levels, ­etc. • Ad hoc, individual agency-defined procedures

Source: Based on Sandford 2 ­ 018. Note: ECHO = European Civil Protection and Humanitarian Aid Operations; HSNP = Hunger Safety Net Program; NGO = nongovernmental organizations; PSNP = Productive Safety Net Programme; USAID = United States Agency for International Development; WFP = World Food P ­ rogramme.

NOTES 1. Sendai Framework, ­https://www.unisdr.org/we/coordinate/sendai-framework. 2. National standards (often articulated in the form of guidelines or operational manuals) typically describe the procedures to be followed in implementing program targeting, ­registration, payments, case management, grievance mechanisms, and e­ xit. 3. The Inter-Agency Standing Committee is a coordination forum of the UN system, bringing together the executive heads of 18 UN and non-UN organizations to ensure coherence of preparedness and response efforts, to formulate policy, and to agree on priorities for strengthened humanitarian action, ­https://interagencystandingcommittee.org​/­the-inter​ -agency-standing-committee.

REFERENCES Bailey, S ­. 2 ­ 018. “Institutions for Adaptive Social Protection: External Linkages and the Humanitarian ­Sector.” Background ­Paper. Oxford Policy Management, Oxford, ­UK. Bailey, ­S., and ­P. ­Harvey. ­2015. “State of Evidence on Humanitarian Cash ­Transfers.” Background Note for the High Level Panel on Humanitarian Cash ­Transfers. Overseas Development Institute, ­London. Barca, Valentina, Thea Westphal, and Veronica Wodsak. 2 ­ 017. “Administration of NonContributory Social Protection: Delivery Systems—Manual for a Leadership and Transformation Curriculum on Building and Managing Social Protection Floors in ­Africa.” Transform (Transformation Curriculum on Building and Managing Social Protection Floors in Africa). Bastagli, F ­.2 ­ 014. “Responding to a Crisis: The Design and Delivery of Social P ­ rotection.” ODI Working Paper, Overseas Development Institute, ­London. Beazley, ­R. ­2017. “Study on Shock-Responsive Social Protection in Latin America and the Caribbean: Ecuador Case ­Study.” Oxford Policy Management, Oxford, ­UK. Beegle, ­K., ­A. Coudouel, and E ­ . ­Monsalve. ­2018. Realizing the Full Potential of Social Safety Nets in ­Africa. Washington, DC: World ­Bank. ­http://documents.worldbank.org/curated​ /­en/657581531930611436/pdf/128594-PUB-PUBLIC.pdf. Bowen, ­T. ­2015. “Social Protection and Disaster Risk Management in the Philippines: The Case of Typhoon Yolanda ­(Haiyan).” Policy Research Working Paper 7482, World Bank, Washington, ­DC. ­http://documents.worldbank.org/curated/en/681881468181128752/pdf​ /­WPS7482.pdf. CaLP (Cash Learning ­Partnership). ­2018. The State of the World’s Cash Report: Cash Transfer Programming in Humanitarian ­Aid. Oxford, ­UK. ­http://www.cashlearning.org/downloads​ /­calp-sowc-report-web.pdf.

Institutional Arrangements and Partnerships: Multisectoral Coordination and Humanitarian Linkages | 111

Clarke, P. K., and ­L. ­Campbell. ­2018. “Coordination in Theory, Coordination in Practice: The Case of the ­Clusters.” Disasters 42 ­(4) 655–73. Development ­Initiatives. ­2018. “Global Humanitarian Assistance Report ­2018.” Development Initiatives, Bristol, ­UK. ­ rotection.” IDS Working Devereux, ­S., and ­R. ­Sabates-Wheeler. ­2004. “Transformative Social P Paper 232, Institute of Development Studies, Brighton, ­UK. FAO (Food and Agriculture Organization of the United Nations). 2016. “The World Humanitarian ­Summit.” Position Paper, FAO, ­Rome. Grand ­Bargain. ­2016. “The Grand Bargain: A Shared Commitment to Better Serve People in ­Need.” World Humanitarian Summit, ­Istanbul. h ­ ttps://reliefweb.int/sites/reliefweb.int​ /­files/resources/Grand_Bargain_final_22_May_FINAL-2.pdf. Harkey, J ­ . ­2014. “Experiences of National Governments in Expanding Their Role in Humanitarian Preparedness and ­Response.” Feinstein International Center, Tufts University, Medford, ­MA. Harvey, ­P. ­2009. “Towards Good Humanitarian Government: The Role of the Affected State in Disaster ­Response.” Humanitarian Policy Group Brief ­37. Overseas Development Institute, ­London. Kardan, ­A . ­2018. “Institutions for Adaptive Social Protection ­Systems.” Background Paper, Oxford Policy Management, Oxford, ­UK. Kardan, A ­ ., S ­ . Bailey, A ­ . Solórzano, and ­L. F ­ idalgo. 2 ­ 017. “Shock-Responsive Social Protection Systems Research Case Study: ­Mozambique.” Oxford Policy Management, Oxford, ­UK. Kawasoe, ­Y. ­Forthcoming. “Social Protection Disaster Risk Management Case Study for ­Japan.” World Bank, Washington, ­DC. Levine, ­S., and ­I. ­Mosel. ­2014. “Supporting Resilience in Difficult Places: A Critical Look at Applying the ‘Resilience’ Concept in Countries Where Crises Are the ­Norm.” Report Commissioned by Humanitarian Policy ­Group. Overseas Development Institute, ­London. O’Brien, ­C., ­J. Congrave, ­K. Sharp, and ­N. ­Keïta. ­2018. “Shock-Responsive Social Protection Systems Research: Case Study—Social Protection and Humanitarian Responses to Food Insecurity and Poverty in ­Mali.” Oxford Policy Management, Oxford, ­UK. ODI (Overseas Development ­Institute). 2 ­ 015. “Doing Cash Differently: How Cash Transfers Can Transform Humanitarian ­Aid.” Report of the High Level Panel on Humanitarian ­Cash. ODI, ­London. OPM. ­2017. Shock-Responsive Social Protection Systems Research: Literature Review, 2nd ­edition. Oxford, UK: ­OPM. Sandford, ­J. ­2018. “Responding to Shocks: Humanitarian Response through National ­Systems.” ­Unpublished. World Bank, Washington, ­DC. Seyfert, ­K., ­V. Barca, ­U. Gentilini, ­M. Luthria, and ­S. ­Abbady. ­2019. “Unbundled: A Framework for Connecting Safety Nets and Humanitarian Assistance in Refugee S ­ ettings.” Social Protection and Labor Discussion Paper 1935, World Bank, Washington, ­DC. ­https:// openknowledge.worldbank.org/ bitstream/ handle/10986/32467/Unbundled-A​ -Framework-for-Connecting-Safety-Nets-and-Humanitarian-Assistance-in-Refugee​ -Settings.pdf?sequence=1&isAllowed=y. Smith, ­G., ­Z. Scott, ­E. Luna, and ­T. ­Lone. ­2017. “Shock-Responsive Social Protection Systems ­ hilippines.” Oxford Policy Research Case Study: Post-Haiyan Cash Transfers in the P Management, Oxford, ­UK. Stern, ­N. ­2006. The Economics of Climate Change: The Stern ­Review. Cambridge, UK: Cambridge University ­Press. World ­Bank. ­2012. “Resilience, Equity, and Opportunity: The World Bank’s Social Protection and Labor Strategy ­2 012–2022.” World Bank, Washington, ­D C. h ­ ttp://documents​ .worldbank.org/curated/en/443791468157506768/pdf/732350BR0CODE200doc0​ version0REVISED.pdf. World Bank. ­2013. Building Resilience to Disaster and Climate Change through Social Protection: Synthesis ­Note. Washington, DC: World ­Bank. ­http://documents.worldbank.org/curated​ /­en/187211468349778714/pdf/796210WP0Build0Box0377381B00PUBLIC0.pdf.

112 | Adaptive Social Protection

World Bank. ­2016. “Cash Transfers in Humanitarian Contexts: Strategic ­Note.” World Bank, Washington, ­DC. ­http://documents.worldbank.org/curated/en/697681467995447727​ /­pdf/106449-WP-IASC-Humanitarian-Cash-PUBLIC.pdf. WWP (World without ­Poverty). ­2019. “Social ­Assistance.” Brazil Learning Initiative for a World without ­Poverty. ­https://wwp.org.br/en/social-policy/social-assistance/.

APPENDIX A

Resilience as It Relates to Adaptive Social Protection This appendix outlines in greater detail the definition of resilience, the resilience capacities, and their relationship to poverty and vulnerability. Drawing from the expansive literature, resilience is defined in this report as: The ability for households to prepare for, cope with, and adapt to shocks in a manner that protects their well-being, ensuring that they do not fall into poverty or destitution. This definition suggests that a more resilient person will possess three interlinked capacities that help to minimize and resist a shock’s negative impacts. The higher the person’s capacity to prepare, cope, and adapt, the lesser the implied impact of a shock and the likelihood of a faster “bounce back,” recovering to pre-shock levels of well-being. With this definition, the generalized vulnerability of the poor and vulnerable to covariate shocks can be ascribed to a deficit in terms of the capacity to prepare, cope, and adapt. Poor households tend to be particularly vulnerable to the impacts of covariate shocks. By 2030, an estimated 325 million poor people will live in the 49 countries most prone to hazards, with the majority in South Asia and Sub-Saharan Africa (Shepard et ­al. 2013) (box A.1). At the same time, by 2030, it is estimated that man-made disasters will lead to half of the world’s poorest people residing in very poor countries that are fragile or affected by conflict (OECD 2018). For a poor household that is hit with a shock, factors such as having limited to no savings, access to finance, access to formal insurance, or safety nets can combine and contribute to this excessive vulnerability and a generally limited capacity to cope with the impacts from shocks (see, for example, figure A.1; Dercon 2005; Hallegatte et ­al. 2016). To protect short-term well-being and consumption after a shock, poorer households may instead turn to negative coping strategies such as removing children from school to work for extra household income, availing high-interest loans, and selling productive assets and forced migration (del Ninno, Pierre, and Coll-Black 2016; Hallegatte et ­al. 2016; Skoufias 2005). However, such short-term coping strategies often work to the household’s ­longer-term detriment. In this way, at the aggregate level, shocks work to undermine poverty reduction efforts and can cause a country to hemorrhage human capital, also to its long-term disadvantage: international experience overwhelmingly shows that poverty is both a driver and consequence of disasters (UNISDR 2015). While poor households are especially vulnerable to covariate shocks, the near- and nonpoor similarly struggle to cope with such impacts, in some cases becoming at-risk of impoverishing losses. Globally, an estimated 26 million people fall into poverty every year because of natural disasters—especially frequent floods and drought (Hallegatte et ­al. 2017). Across Africa, figure A.1 describes the  113

114 | Adaptive Social Protection

BOX A.1

Unbreakable: The poor suffer disproportionately from natural hazards A recent study by the World Bank identified five ­factors that account for the increased vulnerability of the poor to disasters: • Overexposure to disasters. Poor households are overexposed to disasters (floods, droughts, and high temperatures). Informal settlements, for instance, tend to be located in hazard-prone areas (such as on hillsides, close to riverbanks, or near open drains and sewers), making them highly vulnerable to frequent, low-intensity events (such as recurrent floods). The issue of overexposure is expected to worsen, given ­population trends and climate change. • Higher degree of vulnerability. Poorer populations have a higher degree of vulnerability than nonpoor populations. When the poor are impacted by a disaster, the share of their wealth lost is two to three times that of the nonpoor, mainly due to the nature and vulnerability of their assets and livelihoods. The poor tend to have fewer or no savings to protect against crises and are twice as likely to live in fragile areas. • Less able to cope and recover. Poor households are less able to cope and recover from a disaster. The impact of disasters on well-being and ­quality of life is larger on poorer households, who receive less support from financial

•

•

instruments, social protection schemes, and other private financing. The poor have few buffers against the effects of disasters and deploy coping strategies that further-erode well-being. Repeated losses, destruction of assets, or forced displacement degrade economic activity and well-being. Permanent impacts on health and education. Over the longer term, disasters lead to permanent impacts on the health and education of poorer populations, as households need to make difficult coping decisions, such as withdrawing children from school or cutting health expenditures. In turn, these impacts reinforce the intergenerational cycle of poverty, with children being the most aggrieved victims. In Guatemala, for instance, Hurricane Stan (2005) increased the probability of child labor in affected areas by more than 7 percent. Negative impacts on saving and investment behavior. From a behavioral standpoint, disasters can impact saving and investment behavior of poorer households, reinforcing poverty. The impact can be present before a disaster hits, such as smallholders planting low-risk crops to protect their yields in case of bad weather.

Source: Hallegatte et ­al. 2017.

prevalence of “transitory poverty” where two poor households in five are moving into or out of poverty as income fluctuates and they become exposed to shocks (Beegle, Coudouel, and Monsalve 2018). Concretely, between 2006 and 2011, 45 percent of poor households in Senegal escaped poverty, but 40 percent of nonpoor households fell into it, with households affected by a natural disaster having been found to be 25 percent more likely to fall in poverty during the period (Dang, Lanjouw, and Swinkels 2014). Similarly, in Asia, Walsh and Hallegatte (2019) estimate that almost 0.5 million Filipinos per year are vulnerable to poverty due to natural disasters. Indeed, Skoufias et ­al. (2019) also capture the prevalence of this vulnerability to poverty across the Philippines, relative to chronic poverty, as depicted in map A.1. Turning to a different type of shock, it also has been estimated that globally 20 million people sink into poverty for each percentage point decline in the gross domestic product growth rate that results from an economic downturn (Ötker-Robe and Podpiera 2013). Similarly, Laborde, Lakatos, and Martin (2019) find that the 2010–11 food price spike tipped

Resilience as It Relates to Adaptive Social Protection | 115

FIGURE A.1

77.3 24.2 17.2 28.0 11.6 18.1 16.2 7.3 6.5

Congo, Dem. Rep. Madagascar Malawi Mozambique Rwanda Burkina Faso Zambia Togo Sierra Leone Average Uganda Senegal Ethiopia Tanzania Chad Ghana Eswatini Côte d´Ivoire Cameroon Nigeria Botswana Mauritania

39.0 15.1 99.4 53.5 14.0 27.4 1.3 22.7 23.3 182.2 2.3 4.1 0

20

40

60

80

Total population (millions)

Africa: Chronic and transient poverty

100

Percent Chronically poor

Downwardly mobile

Upwardly mobile

Never poor

Source: Dang and Dabalen 2017, as cited in Beegle, Coudouel, and Monsalve 2018. Note: Poverty statistics are from the latest household survey year for each country. “Chronically poor” are households that were poor in both periods of the analysis; “downwardly mobile” are households that fell into poverty in the second period; “upwardly mobile” are those that were poor in the first period but not in the second; and “never poor” are households that were nonpoor in both periods.

MAP A.1

The Philippines: Vulnerability to poverty because of typhoons a. Chronic poverty rate

b. Nonpoor vulnerable to poverty rate (based on 24.9% threshold) ≤ 0.10 ≤ 0.20 ≤ 0.30 ≤ 0.45 ≤ 0.69

Source: Skoufias et ­al. 2019.

≤ 0.16 ≤ 0.24 ≤ 0.29 ≤ 0.32 ≤ 0.37

116 | Adaptive Social Protection

8.3 million people into poverty, representing nearly 1 percent of the world’s poor. The Ebola crisis increased rural poverty 12 percentage points, from 70 percent in the first half of 2014 just before the Ebola crisis to 82 percent in the first half of 2016 following the Ebola crisis (de la Fuente, Jacoby, and Lawin 2019). These select data points indicate the extent of vulnerability to poverty as a direct outcome of covariate shocks. For this reason, sustainable poverty reduction requires an understanding of both current and future poverty dynamics in the context of vulnerability to poverty among nonpoor households (Diwakar et ­al. 2019). The notion of future poverty highlights the fact that poverty is a backward-looking concept, where a poor household may have been affected adversely by shocks experienced in the recent or more distant past (Skoufias et ­al. 2019). However, the conceptual distinction between poverty and vulnerability to poverty rests on the fact that vulnerability to poverty is not about the present but rather about the future (Gallardo 2018). Notably also, women, children, the disabled, and the elderly may not be poor or vulnerable to poverty but belong to vulnerable groups identified as often disproportionately vulnerable to the impacts of shocks (see, for example, Holmes 2019). The concept of “resilience” to shocks has become widely and ubiquitously applied throughout development and humanitarian institutions in the last decade. This has led to a proliferation of definitions among these actors and across the literature (for a brief sample, see box A.2; for a more detailed synopsis, see Béné, Devereux, and Sabates-Wheeler 2012; IFPRI 2013; De Weijer 2013). In summary, however, resilience can be broadly defined as “the ability of countries, communities and households to manage change, by maintaining or transforming living standards in the face of shocks or stresses without compromising their long-term prospects” (Alfani et al. 2015, 4). The Multi-Agency Resilience Measurement Technical Working Group of the Food Security Information Network (FSIN), led by the World Food Programme and the Food and Agriculture Organization of the United Nations, highlights that the concept of resilience has become compelling in development and humanitarian circles in recent years because it highlights the positive capacity to prepare for and respond to shocks in a way that prevents individuals, households, and communities from suffering long-term adverse consequences (FSIN 2015). A definition of resilience is offered here as the basis for considering the role of adaptive social protection in building it.1 Firstly, as noted in the definitions from box A.2, resilience applies to different levels of analysis; for example, the resilience of a society, system, community, household, and individual to shocks. Where the definition used here is concerned, the resilience of the household is the focus specifically. Several definitions of resilience also define resilience as an outcome relative to a benchmark. As the FSIN notes, for conceptual clarity and eventually measurement, it is useful to define resilience as a capacity that prevents households from falling below a “normatively defined level for a given development outcome” including food security, nutrition, well-being, and poverty, for example (FSIN 2014, 7). Accordingly, resilience is defined here as: The ability for a household to prepare for, cope with, and adapt to shocks in a manner that protects their well-being: ensuring that they do not fall into poverty or become trapped in poverty as a result of the impacts. This definition and the benchmarking of resilience to poverty highlight the previously outlined vulnerability of the poor to shocks as well as the vulnerability of the nonpoor to poverty because of shocks. Accordingly, the definition reflects the necessity of achieving and protecting the long-term development

Resilience as It Relates to Adaptive Social Protection | 117

BOX A.2

Resilience: A brief sample of prominent definitions in the literature and among development and humanitarian institutions The constituent parts of these multiple definitions of resilience can be grouped in the following ways: • • •

•

Resilience as an ability, capacity Unit of analysis: system, country, community, society, household, individual Mix of specific abilities/capacities of the analysis unit: to resist, absorb, adapt, accommodate, recover, anticipate, prevent, withstand, manage, overcome, maintain, transform, recover from Outcomes: without compromising their longterm prospects, do not have long-lasting adverse development consequences; advance the rights of every child; ensuring the preservation, restoration, or improvement of its essential basic structures and functions

United Nations International Strategy for Disaster Reduction: The ability of a system, community, or society exposed to hazards to resist, absorb, accommodate to, and recover from the effects of a hazard in a timely and efficient manner (UNISDR, 24). Intergovernmental Panel on Climate Change: The ability of a system and its component parts to anticipate, absorb, accommodate, or recover from the effects of a hazardous event in a timely and efficient

manner, including through ensuring the preservation, restoration, or improvement of its essential basic structures and functions (IPCC 2012, 563). UK Department for International Development: The ability of countries, communities, and households to manage change, by maintaining or transforming living standards in the face of shocks or stresses—such as earthquakes, drought or violent conflict—without compromising their long-term prospects (DFID 2011, 6). Building Resilience and Adaptation to Climate Extremes and Disasters (BRACED): The ability to anticipate, avoid, plan for, cope with, recover from, and adapt to (climate related) shocks and stresses (Bahadur et al. 2015a, 11). Resilience Measurement Technical Working Group: The capacity to ensure that shocks and stressors do not have long-lasting adverse development consequences (WFP 2015). UNICEF: The ability of children, households, communities, and systems to anticipate, prevent, withstand, manage, and overcome cumulative stresses and shocks in ways which advance the rights of every child, with special attention to the most vulnerable and disadvantaged children (UNICEF 2017, 3).

goals of poverty reduction as a driver and a consequence of shocks. In that sense, the definition is directly aligned with the Sustainable Development Goals where social protection (Target 1.3) and building resilience of the poor and vulnerable (Target 1.5) are both contributory targets to the achievement of the goal of ending poverty (Goal 1). Where this definition of resilience is concerned, resilience and vulnerability to poverty can be simplistically seen as “two sides of the same coin.” Jorgensen and Siegel (2019) note an implied2 symmetry between reducing household ­vulnerability to poverty and increasing household resilience to poverty, where vulnerability to poverty is related to the exposure and susceptibility of households, and the potential negative impacts and resilience to poverty is related to the ability of households to prevent/resist/recover from their negative impacts over time. The “SRM 2.0 model” for vulnerability and resilience to poverty is outlined briefly in box A.3, wherein all households are separated into four categories based on whether they are expected to be poor in the future. This concept of vulnerability and resilience to poverty has also been linked explicitly to the mean and variance of the welfare measure used. Skoufias et ­al. (2019) have developed similar concepts of vulnerability and resilience as being

118 | Adaptive Social Protection

BOX A.3

Social risk management: Vulnerability and resilience to poverty In social risk management frameworks, a household is defined as vulnerable to poverty if the probability of its future well-being falling below a socially accepted norm (or “benchmark”) is high. This definition identifies “at-risk” individuals/households as being “at-risk of falling into poverty.” A household can be vulnerable to poverty because of a specific hazard/risk and/or because of multiple ­hazards/risks, and vulnerability to poverty can be for one or more period(s) into the future. It is possible to divide all of society into four groups of households that TABLE BA.3.1  Poor

are either poor/nonpoor or vulnerable/resilient to poverty at a given point in time. If a household’s current income is below the poverty line the household is income poor; and if the expected income is less than the poverty line, the household is considered asset poor. Income poverty is an outcome. Asset poverty is ­forward-looking and shows lack of opportunity to be nonpoor. Increasing household wealth through asset accumulation (and improved risk management) is critical to moving from vulnerability-to-poverty to resilience-to-poverty.

and nonpoor households grouped as vulnerable or resistant to poverty

HOUSEHOLD POOR TODAY

HOUSEHOLD NONPOOR TODAY

Expected to be poor in future

Not expected to be poor in future, depends on variance in expected income

Expected to be poor in future, depends on variance in expected income

Not expected to be poor in future

Chronic poor

Transient poor

Transient poor

Sustainably nonpoor

Vulnerable to poverty

Exit poverty? Vulnerable/resilient to poverty?

Enter poverty? Vulnerable/resilient to poverty?

Resilient to poverty

Outcome depends on good/bad luck

Outcome depends on good/bad luck

Household Group 1

Household Group 2

Household Group 3

Household Group 4

Income poor

Income poor

Income nonpoor

Income nonpoor

Asset poor

Asset nonpoor

Asset poor

Asset nonpoor

Source: Jorgensen and Siegel 2019.

the product of both the mean and variance in welfare, and Hoddinott (2014) has done so using food consumption, asking the questions: is the household poor today, and can it be expected to be poor tomorrow based on its variance in welfare resulting from a shock? In that sense, a household is resilient if the answer to both questions is no, and a household’s resilience can be built through interventions that increase mean resilience in excess of the poverty line and/or reduce variance in welfare to ensure it does not fall beneath the poverty line (box A.4). Broadly speaking, the literature further disaggregates household resilience as being the product of a set of interlinked household capacities. As with the myriad definitions of resilience itself, definitions of these individual capacities, as well as the respective emphases applied to each of them, vary across institutions and within the literature. Nevertheless, taken together they are most often summarized in terms of “absorptive, adaptive, anticipatory, and transformative” capacities (most prominently as found in Bahadur et ­al. 2015a; Béné et ­al. 2012; FSIN 2015). To briefly summarize, together, these capacities imply that a household is resilient when it can adequately prepare for a shock (anticipatory), adapt and

Resilience as It Relates to Adaptive Social Protection | 119

BOX A.4

Welfare mean versus welfare variance: Conceptualizing resilience to poverty •

The welfare of a household measured by per capita expenditure is assumed to be characterized by two key parameters: the mean value of household per capita expenditure (Is the household poor?) and the variance of household per capita expenditure (Is the household vulnerable to poverty?). Thus, households are assumed to differ with respect to the mean level of their per capita expenditure and the variance of their per capita expenditure around that mean. Figure BA.4.1 summarizes these two dimensions of welfare, mean and variance, for 10 hypothetical households (households A–J). The mean consumption expenditure of a household (or the average value of consumption expenditures, for example, associated with many different shocks or states of the world over time) is depicted by the blue square. Households differ with respect to the mean level of their expenditure and the variance of their expenditure around that mean, with some households having a low (or high) mean level of consumption and a low (or high) variance of consumption.

•

•

•

Poor households. Households A, D, G, and I are, on average, poor households and their vulnerability is “poverty induced,” meaning that it is determined primarily by low endowments of assets and human capital that are the primary determinants of their low mean value of welfare. Nonpoor households. Households B, C, E, F, H, and J are, on average, nonpoor households as their mean welfare is above the poverty line. However, some of these households are vulnerable to poverty while others are not. Not vulnerable to poverty. Households B, E, and H, for example, have variability in their consumption but the variance line never crosses the poverty line. Vulnerable to poverty. Households C, F, and J may end up below the poverty line under some circumstances, as depicted by the fact that the variance of their consumption around the mean crosses the poverty line. For households C, F, and J, vulnerability to poverty is “risk induced.”

FIGURE BA.4.1

Vulnerability to poverty characterized by the mean and variance of welfare Poverty line

A

B C

Households

D E F G H I J

Mean and variance of welfare

Source: Skoufias et ­al. 2019. Note: Hoddinott (2014) applies a similar concept using the food consumption score as the threshold.

120 | Adaptive Social Protection

reduce its risk to that shock over time (adaptive), cope with its impacts should it materialize (absorptive), and potentially even completely remove its exposure and vulnerability to the shock through a structural transformation in its ­livelihood, assets, and/or location (transformative). “Transformative” capacity also may be viewed as a constituent part of the household’s adaptive capacity (Davies et ­al. 2009). These definitions, as found in the literature, are expanded upon in greater detail in box A.5. Drawing from the capacities identified in the literature, a household’s resilience to a shock can be seen as the product of its capacity to prepare, cope, and adapt. The higher the household’s capacity to prepare, cope, and adapt, the lesser the implied impact and the likelihood of a faster bounce-back (Schipper and Langston 2015), recovering to pre-shock levels of well-being. For greater precision, taking each interlinked capacity in turn, a more resilient household can do the following: • Prepare for a shock to minimize and mitigate its impact on well-being.3 Primarily, the degree of preparedness is related to the household’s access to information that improves the understanding of the factors that drive its exposure and vulnerability to shocks, as well as the actions taken to minimize these factors (Bahadur et ­al. 2015a). This preparedness directly enhances the

BOX A.5

Household resilience “capacities” in the literature Drawing primarily on the absorptive, adaptive, and transformative framework from Béné et ­al. (2012) and the anticipatory, absorptive, and adaptive framework from Bahadur et ­al. (2015a), each of which have been widely adopted in the social protection and resilience literature: •

•

Anticipatory capacity is denoted as the ability for the household to be better prepared for the eventuality of a specific shock by proactively acting to avoid or reduce exposure or by minimizing vulnerability to it. The ability to anticipate a shock, to understand the associated risks, and to prepare is instrumental to the other capacities—informing decisions that can assist with short-term coping and long-term adaptation. Absorptive capacities allow households or systems to absorb and cope with climate-­related shocks and stresses during and after their occurrence. It enables households to reduce the immediate negative impact on livelihoods and

Sources: Based on Bahadur et ­al. 2015a; Béné et ­al. 2012; FSIN 2015.

•

•

basic needs. In other words, they are the various (coping) strategies by which individuals and/or households moderate or buffer the impacts of shocks on their livelihoods and basic needs. Adaptive capacity is the ability of social systems to adapt to multiple, long-term, and future climate risks and to learn and adjust after a disaster. It capacitates deliberate and planned decisions to achieve a desired state even when conditions have changed or are about to change. Adaptive capacity is realized through “income stability and asset accumulation and retention over time” as a result of proactive and informed choices about alternative livelihood strategies based on an understanding of changing conditions. Transformative capacity refers to strategies that aim at altering permanently and drastically the household structure or functioning to ensure the long-term “survival” of the individual member/household.

Resilience as It Relates to Adaptive Social Protection | 121

capacity to cope should the shock materialize and better enables adaptation in the longer term. • Cope with a shock’s direct impacts to minimize their impact on well-being.4 That is, a more resilient household will have more assets to draw upon to resist the negative impact on its well-being—to the greatest extent possible— and to bounce back to their pre-shock state in a timely fashion (Bahadur et ­al. 2015a). To do so, a more resilient household may draw upon a range of skills, resources, and assets to face and manage adverse conditions, emergencies, or disasters, including own savings and private (insurance) and public (social protection) resources. • Adapt in a manner that reduces both the exposure and vulnerability over the longer term and before, during, and after a shock occurs. A more resilient household is capable of acting on information pertaining to its exposure and vulnerability, making longer-term investments and undertaking strategies that reduce both factors. This includes diversifying or adjusting livelihood portfolios away from sources of income that are especially vulnerable to the impacts of a shock; building the household’s asset base, including productive, financial, and human capital-related assets to enable it to make these adjustments; and/or, leveraging such assets to relocate away from an area of spatially concentrated risk. Indeed, the ultimate expression of adaptation may be the household’s ability to reduce exposure altogether through relocation and planned migration when in situ adjustments to livelihood and assets portfolios fail and where remaining in place would lead to chronic vulnerability and even maladaptation.5 This definition of resilience implies that providing a pathway toward achieving a more resilient state is critical for poor and vulnerable households. A key point, as Frankenberger et ­al. note, is that it is important to emphasize that resilience is not synonymous with coping capacity. Whereas coping capacity typically refers to the ability of households to return to their previous state in the wake of disaster, resilience programming must focus on strengthening the adaptive capacity of vulnerable households…this entails taking incremental steps to reduce their exposure…so that they can eventually escape poverty and continually improve their wellbeing. (Frankenberger et ­al. 2012, 2)

As the FSIN (2015, 7) also notes, it is important to distinguish between resilience capacities that allow households to maintain normative levels of wellbeing above the threshold (the ‘resilient state’) and resilience capacities that reflect growth toward and in excess of the normative threshold (‘resilient pathway’). The concept of “bouncing forward” or “bouncing back better” is critical when considering resilience to poverty. Indeed, Tanner, Bahadur, and Moench (2017) and Shelton (2013) note the increasingly popular idea of “bouncing forward” or “bouncing back better.” This is in line with the concept of “building back better” after disasters (Priority 4 of the Sendai Framework on Disaster Risk Reduction) wherein the conditions underlying vulnerability to the previous disaster are altered in its aftermath to reduce future risk. Essentially, these concepts emphasize that the household’s state prior to the shock may not have been an optimal state, instead being one characterized by vulnerability. This suggests that bouncing back to that state alone will be a less than optimal outcome, underscoring the need for resilience to consider the capacity to adapt in the long term, reducing

122 | Adaptive Social Protection

vulnerability and improving well-being (Schipper and Langston 2015). Without doing so, households may experience recurring cycles of vulnerability. The concept of resilience can be useful in fostering multisectoral collaboration around a broadly shared, if differently defined, objective. While, as discussed, many definitions of resilience exist, the literature frequently notes that this also has served to create a basis for collaboration and cooperation among institutions and agencies. Broadly speaking, in most cases, differences in definitions of resilience among institutions are slight and semantic rather than significant and substantive. Indeed, Bahadur et ­al. (2015b) note that the application of the term across disciplines and institutions is helping to break down sectoral silos; while Béné et ­al. 2012 similarly state that precisely because the term is used loosely in a large number of disciplines and organizations, it can be a powerful integrating concept across different sectors. Indeed, in the original and pioneering concept of adaptive social protection by researchers at the Institute of Development Studies, resilience (and vulnerability) served as the conceptual cornerstone for linking the three traditionally separate sectors of social protection, disaster risk management, and climate change adaptation around a common objective (see box A.5). As noted in chapter 3, resilience also can be seen as a potentially unifying concept and policy instrument that uses humanitarian and development approaches to address the chronic vulnerability of populations exposed to recurrent shocks (FSIN 2015, similarly articulated in FAO 2016).

NOTES 1. While an already poor individual or household cannot “fall” into poverty (although the poor individual or household may fall deeper into poverty), the poor individual or household cannot be considered resilient according to this definition. 2. This symmetry is contested but is defendable. It is contested in the literature where one can be characterized as being both resilient and vulnerable at the same time. This holds in instances where one can return to her/his prior state after a shock, but that prior state may be one characterized by vulnerability. The symmetry is defendable in instances where a benchmark for resilience is set, such as poverty in this case. Here, returning to a previously vulnerable state (that is, remaining poor) cannot be defined as being resilient. 3. This capacity is also referred to as “anticipatory” capacity in the BRACED 3As framework (Bahadur et ­al. 2015a). The term “preparedness” is used here to more explicitly reflect the meaning of the capacity as used in this report as well as to make a clear linkage to the emphasis on preparedness reflected in the disaster risk reduction and disaster risk management communities. 4. This capacity is also referred to as “absorptive” capacity in the BRACED 3As framework (Bahadur et ­al. 2015a; similar in Béné et ­al. 2012). The term “coping” is chosen here because of its widespread use in the social protection community and its interchangeability with the term “absorptive.” 5. Defined as a failure to adjust adequately or appropriately to a shock.

REFERENCES Alfani, Federica, Andrew Dabalen, Peter Fisker, and Vasco Molini. 2015. “Can We Measure Resilience? A Proposed Method and Evidence from Countries in the Sahel.” Policy Research Working Paper 7170. World Bank Group, Washington, DC. https://openknowledge​ .­worldbank.org/handle/10986/21387.

Resilience as It Relates to Adaptive Social Protection | 123

Bahadur, A., K. Peters, E. Wilkinson, F. Pichon, K. Gray, and T. Tanner. 2015a. “The 3As: Tracking Resilience across BRACED.” Working Paper, Building Resilience and Adaptation to Climate Extremes and Disasters (BRACED), London. Bahadur, A., E. Wilkinson, E. Lovell, and T. Tanner. 2015b. “Resilience in the SDGs: Developing an Indicator for Target 1.5 That Is Fit for Purpose.” ODI Briefing Note. Overseas Development Institute, London. Beegle, K., A. Coudouel, and E. Monsalve. 2018. Realizing the Full Potential of Social Safety Nets in Africa. Washington, DC: World Bank. http://documents.worldbank.org/curated​ /­en/657581531930611436/pdf/128594-PUB-PUBLIC.pdf. Béné, C., S. Devereux, and R. Sabates-Wheeler. 2012. “Shocks and Social Protection in the Horn of Africa: Analysis from the Productive Safety Net Programme in Ethiopia.” IDS Working Paper 395, Institute of Development Studies, Brighton, UK. Béné, C., R. Wood, A. Newsham, and M. Davies. 2012. “Resilience: New Utopia or New Tyranny? Reflection about the Potentials and Limits of the Concept of Resilience in Relation to Vulnerability Reduction Programmes.” IDS Working Paper 405, Institute of Development Studies, Brighton, UK. Dang, H.-A., and A. Dabalen. 2017. “Is Poverty in Africa Mostly Chronic or Transient? Evidence from Synthetic Panel Data.” Policy Research Working Paper 8033, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/172891492703250779/pdf/WPS8033​ .­pdf. Dang, H.-A., P. Lanjouw, and R. Swinkels. 2014. “Who Remained in Poverty, Who Moved up, and Who Fell Down? An Investigation of Poverty Dynamics in Senegal in the Late 2000s.” Policy Research Working Paper 7141, World Bank, Washington, DC. http://documents.worldbank​ .org/curated/en/862121468296672365/pdf/WPS7141.pdf. Davies, M., B. Guenther, J. Leavy, T. Mitchell, and T. Tanner. 2009. “Adaptive Social Protection: Synergies for Poverty Reduction.” IDS Bulletin 39 (4): 105–12. De la Fuente, A., H. Jacoby, and K. Lawin. 2019. “Impact of the West African Ebola Epidemic on Agricultural Production and Rural Welfare Evidence from Liberia.” Policy Research Working Paper 8880, World Bank, Washington, DC. http://documents.worldbank.org​ /­curated/en/423511560254844269/pdf/Impact-of-the-West-African-Ebola-Epidemic-on​ -Agricultural-Production-and-Rural-Welfare-Evidence-from-Liberia.pdf. del Ninno, C., F. Pierre, and S. Coll-Black. 2016. Social Protection Programs for Africa’s Drylands. Washington, DC: World Bank. http://documents.worldbank .org/curated​ /­en/736221471343475745/pdf/107854-PUB-PUBLIC-PUBDATE-8-9-16.pdf. De Weijer, F. 2013. “Resilience: A Trojan Horse for a New Way of Thinking?” ECDPM Discussion Paper 139, European Centre for Development Policy Management, Maastricht. Dercon, S. 2005. “Risk, Poverty and Vulnerability in Africa.” Journal of African Economies 14 (4): 483–88. DFID (Department for International Development). 2011. “Defining Disaster Resilience: A DFID Approach Paper.” DFID, London. Diwakar, V., J. Albert, J. Vizamos, and A. Shepherd. 2019. “Resilience, Near Poverty and Vulnerability Dynamics, Evidence from Uganda and the Philippines.” U.S. Agency for International Development, Washington, DC. https://www.agrilinks.org/sites/default/files​ /­usaid-report-nearpoor_clean_march_508.pdf. FAO (Food and Agriculture Organization of the United Nations). 2016. Adapting Agriculture to Climate Change. FAO’s Work on Climate Change Adaptation. Rome: FAO. http://www.fao​ .­org/3/a-i6273e.pdf. Frankenberger, T., T. Spangler, S. Nelson, M. Langworthy. 2012. “Enhancing Resilience to Food Insecurity amid Protracted Crisis.” United Nations High-Level Expert Forum, Rome. http:// www.fao.org/fileadmin/templates/cfs_high_level_forum/documents/Enhancing​ _­Resilience​_­FoodInsecurity-TANGO.pdf. FSIN (Food Security Information Network). 2014. “Resilience Measurement Principles: Toward an Agenda for Measurement Design.” Technical Series 1. Resilience Measurement Technical Working Group, FSIN Secretariat, World Food Programme. http://www.fsincop.net​ /­fileadmin/user_upload/fsin/docs/resources/FSIN_29jan_WEB_medium%20res.pdf.

124 | Adaptive Social Protection

FSIN (Food Security Information Network). 2015. “Measuring Shocks and Stressors as Part of Resilience Measurement.” Technical Series 5. Resilience Measurement Technical Working Group, FSIN Secretariat, World Food Programme. http://www.fsincop.net/fileadmin/user​ _upload/fsin/docs/resources/1_FSIN​_­TechnicalSeries_5.pdf. Gallardo, M. 2018. “Identifying Vulnerability to Poverty: A Critical Survey.” Journal of Economic Surveys 32 (4): 1074–105. Hallegatte, S., M. Bangalore, L. Bonzanigo, M. Fay, T. Kane, U. Narloch, J. Rozenberg, D. Treguer, and A. Vogt-Schlib. 2016. Shock Waves: Managing the Impacts of Climate Change on Poverty. Washington, DC: World Bank. Hallegatte, S., A. Vogt-Schilb, M. Bangalore, and J. Rozenberg. 2017. Unbreakable: Building the Resilience of the Poor in the Face of Natural Disasters. Washington, DC: World Bank. https:// www.gfdrr.org/sites/default/files/publication/Unbreakable_FullBook_Web-3.pdf. Hoddinott, J. 2014. “Understanding Resilience for Food and Nutrition Security.” Building Resilience for Food and Nutrition Security 2020 Conference Paper 8. International Food Policy Research Institute, Washington, DC. http://ebrary.ifpri.org/utils/getfile/collection​ /­p15738coll2/id/128165/filename/128376.pdf. Holmes, R. 2019. “Promoting Gender Equality and Women’s Empowerment in Shock Sensitive Social Protection.” ODI Working Paper 549, Overseas Development Institute, London. IFPRI (International Food Policy and Research Institute). 2013. “Definitions of Resilience: 1996–Present.” Building Resilience for Food & Nutrition Security. IFPRI, Washington, DC. http://www.2020resilience.ifpri.info/files/2013/08/resiliencedefinitions.pdf. IPCC (International Panel on Climate Change). 2012. “Glossary of Terms.” In Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. A Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change (IPCC), edited by C. B. Field, V. Barros, T. F. Stocker, D. Qin, D. J. Dokken, K. L. Ebi, M. D. Mastrandrea, K. J. Mach, G.-K. Plattner, S. K. Allen, M. Tignor, and P. M. Midgley, 555–64. Cambridge, UK and New York, NY: Cambridge University Press. Jorgensen, S., and P. Siegel. 2019. “Social Protection in an Era of Increasing Uncertainty and Disruption: Social Risk Management 2.0.” Social Protection and Jobs Discussion Paper 1930, World Bank, Washington, DC. http://documents.worldbank.org/curated​ /­en/263761559643240069/pdf/Social-Protection-in-an-Era-of-Increasing-Uncertainty​ -and​-Disruption-Social-Risk-Management-2-0.pdf. Laborde, D., C. Lakatos, and W. Martin. 2019. “Poverty Impact of Food Price Shocks and Policies.” Policy Research Working Paper 8724, World Bank, Washington, DC. http://documents​ .­worldbank.org/curated/en/863311549375011898/pdf/WPS8724.pdf. OECD (Organisation for Economic Co-operation and Development). 2018. “States of Fragility 2018, Highlights.” OECD, Paris. Ötker-Robe, I. and A. M. Podpiera. 2013. “The Social Impact of Financial Crises: Evidence from the Global Financial Crisis.” Policy Research Working Paper 6703, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/498911468180867209/pdf/WPS6703​ .­pdf. Schipper, E., and L. Langston. 2015. “A Comparative Overview of Resilience Measurement Frameworks.” ODI Working Paper 422, Overseas Development Institute, London. Shelton, P. 2013. “Bouncing Back Better: Defining Resilience at the 2013 Global Hunger Index Launch.” International Food Policy and Research Institute, Washington, DC. Shepherd, Andrew, Tom Mitchell, Kirsty Lewis, Amanda Lenhardt, Lindsey Jones, Lucy Scott, and Robert Muir-Wood. 2013. “The Geography of Poverty, Disasters and Climate Extremes in 2030.” Research Reports and Studies, October 2013. ODI, London. Skoufias, E. 2005. “PROGRESA and Its Impacts on the Welfare of Rural Households in Mexico.” Research Reports 139. International Food Policy Research Institute, Washington, DC. Skoufias, E., Y. Kawasoe. E. Strobl, and P. Acosta. 2019. “Identifying the Vulnerable to Poverty from Natural Disasters. The Case of Typhoons in the Philippines.” Policy Research Working Paper 8857, World Bank, Washington, DC. http://documents.worldbank.org/curated​

Resilience as It Relates to Adaptive Social Protection | 125

/­en/326941558453867995/pdf/Identifying-the-Vulnerable-to-Poverty-from-Natural​ -Disasters-The-Case-of-Typhoons-in-the-Philippines.pdf. Tanner, T., A. Bahadur, and M. Moench. 2017. “Challenges for Resilience Policy and Practice.” Working Paper 519. Overseas Development Institute, London. https://www.odi.org/sites​ /­odi​.org.uk/files/resource-documents/11733.pdf. UNICEF (United Nations Children’s Fund). 2017. “Resilience, Humanitarian Assistance and Social Protection for Children in Europe and Central Asia.” Social Protection Regional Issue Brief 2, UNICEF Europe and Central Asia Regional Office. https://www.unicef.org/eca​ /­media/2671/file/Social_Protection2.pdf. UNISDR (United Nations International Strategy for Disaster Reduction). 2009. 2009 UNISDR Terminology on Disaster Risk Reduction. Geneva: UNISDR. UNISDR (United Nations International Strategy for Disaster Reduction). 2015. Making Development Sustainable: The Future of Disaster Risk Management: Global Assessment Report on Disaster Risk Reduction. Geneva: UNISDR. Walsh, B., and S. Hallegatte. 2019. “Measuring Natural Risks in the Philippines: Socioeconomic Resilience and Wellbeing Losses.” Policy Research Working Paper 8723, World Bank, Washington, DC. http://documents.worldbank.org/curated/en/482401548966120315/pdf​ /­WPS8723.pdf. WFP (World Food Programme). 2015. “Policy on Building Resilience for Food Security and Nutrition.” WFP/EB.A/2015/5-C. WFP, Rome. https://documents.wfp.org/stellent/groups​ /­p ublic/documents/eb/wfpdoc063833.pdf?_ga=2.20959473.817428444.1582152603​ -752767465.1554223343.

APPENDIX B

Considerations for Shock Response along the Social Protection Delivery Chain The social protection delivery chain can help to think through pertinent questions and challenges that need to be addressed when establishing contingency plans, along with the investment in these core delivery processes that may be needed to support shock response. Despite heterogeneity in program composition and maturity of social protection systems across countries, most rely on common phases of delivery to ensure that programs provide the right amount/ composition of benefits and services, to the right persons, at the right time. This delivery chain is centered on four implementation phases, captured in the World Bank Sourcebook on the Foundations of Social Protection Delivery Systems (Lindert et al., forthcoming), namely: assess, enroll, provide, and manage. Applying the delivery chain to evaluate whether social protection programs are supporting adaptive social protection (ASP) objectives can help establish whether the essential implementation features required to ensure efficient and effective delivery of ASP interventions are functioning in those programs. Four common phases apply to the delivery chain for social protection programs and services, with associated business processes for each phase. The assess phase of the delivery chain seeks to determine potential eligibility and includes business processes for outreach, intake and registration, and assessment of needs and conditions among potentially eligible households and individuals. This is followed by an enroll phase which is concerned with making decisions on enrollment and onboarding into programs, and the determination of what benefits and services would be provided to enrolled beneficiaries. The provide phase follows, where implementation begins in earnest with the provision of benefits and services, including payment transactions and services provision etc. Finally, the recurring phase of manage follows, where beneficiary monitoring processes are addressed, including addressing grievances and compliance, exit notifications and case outcomes. The defined operational processes to support program implementation also help to guide program management at all stages of the delivery chain and is a critical tool for supporting the beneficiary monitoring and management processes (figure B.1).* This appendix draws extensively from a background paper by Smith (2018).

â&#x20AC;&#x192;127

128â&#x20AC;&#x192;|â&#x20AC;&#x192;Adaptive Social Protection

FIGURE B.1

Social protection delivery chain Assess

Enroll

Outreach

Intake and registration

Assess needs and conditions

1

2

3

Eligibility Determine and benefits enrollment and service decisions package

4

5

Provide

Notification and onboarding

Benefits and/or services

6

7

Manage

Beneficiaries, Exit decisions, notifications, grievances, compliance case outcomes

Recurring cycle

8

9

Periodic reassessment

Source: Lindert et al., forthcoming.

ASSESS Outreach: How will I let affected households know that they may be eligible for support? The objective of the Outreach process is to inform beneficiaries of the social protection program, create awareness, and encourage potential beneficiaries to apply. Outreach activities usually use a variety of communication mechanisms, including public messaging through the media (electronic and print); local information sessions through community structures; word of mouth; visual information campaigns; staff outreach; and digital communications (for example, social media). Outreach activities are necessary to encourage potential beneficiaries to apply, and to inform existing beneficiaries on regular interactions, such as payment dates and times (Smith 2018). Post-shock considerations. Shocks can disrupt communication channels, undermining outreach activities. Therefore, the communication mechanisms may need to change in a post-shock setting. For instance, where digital communication is interrupted, outreach may need to focus on word-of-mouth or staff outreach campaigns. Also, the type of information to be provided following a shock depends on the type of response to be implemented. In the case of vertical expansions, beneficiaries will already be aware of the program, but may not be aware of the additional payment(s) to be provided. In the case of a horizontal expansion or emergency program/piggybacking, the information provided would need to be more comprehensive, informing potential beneficiaries on the objective of the program, who can apply, how to apply, what kind of assistance they may expect and when, etc. (Smith 2018). In the case of Turkey, the horizontal expansion to include the Syrian refugees undertook innovative outreach processes in order to inform those refugees they may be eligible for support through the Emergency Social Safety Net Program.

Considerations for Shock Response along the Social Protection Delivery Chainâ&#x20AC;&#x192;|â&#x20AC;&#x192;129

Intake and registration: How should I gather information on the disaster-affected households to assess their needs and potential eligibility for support? The objective of the intake and registration process is to collect information on potential beneficiaries (individuals, households, or communities), to be able to assess their eligibility for the social protection program. Intake and registration are completed either on demand, where potential beneficiaries register themselves (typically at social welfare offices), or all together, in which case survey teams visit the population to collect household data. Surveying all households may have been used to establish a social registry containing information on social protection and nonsocial protection beneficiary households. Where a social registry exists, it may not be necessary to complete separate intake and registration processes for each social protection program, but rather the social registry serves as the basis for intake and registration for multiple programs (Smith 2018), as depicted in figure B.2. Some social registries have near universal coverage (Chile, Colombia, the Dominican Republic, Pakistan, and the Philippines). Others cover between one third and one half of the population (Brazil, Georgia, Indonesia, Mexico, Montenegro, and Turkey). Others operate on a much smaller scale, either because they have been implemented in specific geographic areas before expanding to national coverage (such as China, Djibouti, Mali, Senegal, and the Republic of Yemen) or because of strict eligibility verification rules that can discourage applicants (Azerbaijan) (Leite et al. 2017). Post-shock considerations. In the case of a vertical expansion of an existing program, the registration of beneficiaries is already completed via the existing, regular program. However, depending on program objectives, it may be necessary to leverage additional information, in the event that only some of the existing beneficiaries are to be targeted (for example, only those living in a

FIGURE B.2

Social registry intake and registration Enroll

Provide

Manage

Enroll

Provide

Manage

Enroll

Provide

Manage

Assess needs and conditions

m

1

a gr

o

Pr

Program 2

3 Pr

og

ra

m

Source: Lindert et al., forthcoming.

3

130 | Adaptive Social Protection

shock-affected area), prioritizing categorical vulnerabilities (such as women, children, the disabled, and the elderly). For horizontal expansions or piggybacking, it will depend on the extent of the existing information contained in a social registry, as well as how up to date that information is. Where up to date and complete, the social registry will already contain the information on most of the households in the affected areas. Where ­otherwise, and even in those cases, additional information will need to be gathered through postdisaster surveys or on-demand registration (see c­ hapter 2 and below related to assessment). Depending on the impacts of the disaster, it may be necessary to change the location and number of registration offices, including through temporary offices to facilitate such on-­demand registration.

Assess needs and conditions: Among the affected, who are most likely to require support? The objective of this step is to assess needs of applicants according to eligibility criteria and screening tools, to inform the determination of eligibility for benefits and the composition of the benefits and service package. The assessment of needs and conditions for regular social protection benefits is based on adherence to criteria-based information such as income/consumption, poverty status, age, gender, or geography. These are often relatively static and evolve infrequently. For postdisaster social protection benefits and services, assessment of needs and conditions often requires assessment of postdisaster impacts and needs on affected households to inform eligibility determination for the post-shock benefits and services that may be offered. Post-shock considerations. As noted, in case of vertical expansions, the eligibility criteria for regular programs are retained in the event of disaster, which can lead to exclusion errors as beneficiaries or households that need assistance may not be among the regular beneficiaries of the program. However, by virtue of already being in the program, and having been selected as eligible in advance, those existing beneficiaries can be reached quickly if financing is available (see chapter 3). By contrast, horizontal expansions or piggybacking can allow for changes in eligibility criteria of the program (using damage to the household as a proxy for eligibility, for example) and most likely the reduction of exclusion errors. Such examples are discussed extensively in chapter 2, including the case of Chile where the Basic Emergency Sheet (Ficha Básica de Emergencia—FIBE) assessment tool is used to assess post-shock household needs, linked to the national social registry (Registro Social de Hogares—RSH). Emergency programs, which are solely deployed after shocks, may combine indicators related to chronic poverty or categorical criteria with shock impacts and needs to assess needs and conditions.

ENROLL Eligibility and enrollment decisions: Based on intake, registration, and assessment, who should be enrolled into the program? As part of the eligibility and enrollment decision process, program ­implementers decide which of the potential beneficiaries who completed intake and registration are included in the program as beneficiaries.

Considerations for Shock Response along the Social Protection Delivery Chain | 131

Both applicants who are included in the program as beneficiaries, and applicants who are not included in the program, are informed of the decision. For beneficiaries, the enrollment process then includes the collection of additional documents (if applicable), as well as registration for the programs in which they are being enrolled, and preparation of the delivery of benefits and services packages, for instance, through establishing a payment mechanism and/or setting up a bank account. Beneficiaries are then provided with a beneficiary identification (ID), or other means, to verify their status as a beneficiary. Post-shock considerations. In cases where shocks such as fast-onset disasters are disruptive, beneficiaries may lose their means of ID and documentation. Similarly, the forcibly displaced will most often lack ID. Depending on the type of ID required, processes may need to be adapted. For instance, where ID is based on beneficiary IDs, these may need to be reissued following a disaster by program staff. Where ID is based on national IDs, the process to replace these documents is outside the jurisdiction of the social protection program institution, such that assistance from the issuing authority would be required. In postdisaster contexts, biometric data are typically most easily verifiable. For the case of horizontal expansion, processes need to be adapted to enable enrollment of new beneficiaries, ensuring accessible and convenient locations and times for enrollment; mechanisms to relax or waive documentation requirements or facilitate referrals to have document processes expedited for beneficiaries who do not possess them.

Determine benefits and service package: Once enrolled, what kind of benefits will they receive? As part of the Determination of the Benefits and Service Package, the beneficiary list is finalized prior to each payment, the amount of the transfer is set, and provision points are determined. The determination of benefits and services packages will seek to ensure that the benefits and services provided are appropriate in light of need and will depend on resource availability and political feasibility. For regular social protection programs, benefits and service packages are often determined so as to smooth consumption, to reduce chronic poverty, to provide assistance in the event of unemployment, or to ensure an adequate pension. Post-shock considerations. Post-shock benefits can have a range of objectives from smoothing income and consumption to addressing lost livelihoods and replacing lost assets. Typically, benefits and service packages are designed to address chronic, rather than transitory, multifaceted post-shock needs. Therefore, benefits and service packages may need to be modified to address the needs generated in postdisaster situations. For instance, where benefits are provided in cash, it is important to ensure that the amount provided is enough to purchase an appropriate amount of food for the household, and where benefits are provided in-kind, the amount should suffice to meet nutritional requirements. Moreover, coordination with other actors that are providing post-shock benefits is critical to ensure alignment of benefit packages. If transfer values and durations provided by different actors vary widely, this may lead to perceptions of inequity and have further implications for beneficiaries’ expectations of transfers provided by regular programs once emergency ­support ends.

132â&#x20AC;&#x192;|â&#x20AC;&#x192;Adaptive Social Protection

Notification and onboarding: What is the best way to let the selected beneficiaries know they will receive support? As part of the Notification and Onboarding step, beneficiaries receive notification that they have been selected as a beneficiary. Beneficiaries are informed of the benefits and service package they will receive, the timing and duration of payments and services, and the steps required to avail themselves of benefits and services. Post-shock considerations. As a result of a destructive or disruptive shock, such as a fast-onset disaster that has generated significant displacement, it may be difficult to notify beneficiaries of their selection as into a program. As with the outreach stage, alternative mechanisms will need to be created to inform beneficiaries of the decision. Ideally notification (and in some cases even payment or initiation of public works activities, for example) could be delivered when the beneficiary applies for entry into the program, where possible. This was successfully undertaken through the utilization of one stop shops in Pakistan as part of the Citizenâ&#x20AC;&#x2122;s Damage Compensation Program, where applicants passed from the assessment to payment phase in as little as 13 minutes in one location.

PROVIDE Benefits and/or services: How will I physically deliver this assistance to the selected beneficiaries in the post-shock setting? The benefits and/or services process focuses on the provision of payments to beneficiaries. This requires providing the correct payment amount to the right people with the right frequency and at the right time (Grosh et al. 2008). Payment modalities include physical cash, transfers to bank accounts, and mobile money. For electronic transfers, such as bank transfers or mobile money, this involves two steps: the provision of funds to the bank account/mobile money account, and the collection of funds by the beneficiary. Post-shock considerations. In the event of destructive disasters, payment and other benefit delivery processes can be disrupted. For instance, pay points and offices of banks or payment service providers can be destroyed or rendered inaccessible, and a lack of connectivity can affect mobile money and automated teller machine (ATM) networks. Therefore, in the event of disasters, it is important to understand what the payment and benefit delivery infrastructure is; what markets exist; what is convenient and accessible for beneficiaries; and what is quick, timely, transparent, less subject to fraud and error, and can be reconciled with confidence. For instance, it can be useful to temporarily switch to manual cash payments where electronic payment delivery is not feasible. Moreover, temporary pay points can be set up to ensure accessibility, as was the case in the Philippines. In the cases of vertical and horizontal expansion after a shock, additional transactions will increase the workload of staff and the payment service provider; and additional financial resources at the local level will be required for distribution. Frequency of payments may also need to be revisited in situations of crisis to ensure that they reach beneficiaries quickly and to reduce the administrative burden in these complex payment environments.

Considerations for Shock Response along the Social Protection Delivery Chainâ&#x20AC;&#x192;|â&#x20AC;&#x192;133

MANAGE Beneficiary monitoring, grievance redress, and compliance monitoring: Is the program functioning effectively, responding to needs, and do I need to make any adjustments? The objective of this step is to conduct beneficiary monitoring and ensure that grievances made by beneficiaries and nonbeneficiaries are addressed. This includes, for instance, monitoring payment receipt, ensuring that beneficiaries meet conditionalities, such as health and education conditions, recording grievances and complaints,1 addressing administrative issues related to delivery, and facilitating case management processes. To ensure that these processes are implemented effectively, it is important that the processes are supported by adequate human resources, and that systems are in place to record these processes. Post-shock considerations. Following a destructive disaster, damaged offices, a breakdown of communication channels, and limited staff (who themselves may be among the affected) can create challenges in maintaining program operations. Further, postdisaster contexts may be associated with additional demands on the system. Additional complexities include changed program conditionalities, the inclusion of new beneficiaries, and introduction of temporary modifications. These challenges can be addressed by temporarily deploying staff from other regions, ensuring the availability of alternative data management options, and backing up systems (for example, using the cloud). Where program conditionalities are in place, these can be waived temporarily; however, in such cases, this must be communicated clearly to program staff, beneficiaries, and partners. As noted previously in the discussion around timeliness versus accuracy, functioning grievance redress mechanisms will be critical to addressing exclusion errors, a priority during shock response.

Exit decisions, notifications, and closing cases: When should the program be wound down and how is it best to inform beneficiaries? As part of the exit decisions, notifications, and case outcomes process, it is assessed whether beneficiaries have met the conditions to graduate from the program and are notified that they will exit. Where there are changes in regulations and exit conditions, these would need to be communicated to beneficiaries. After beneficiaries exit the program, cases are closed. Post-shock considerations. Programs that undergo vertical and horizontal expansion need to be scaled down after the emergency. Scaling programs down following an emergency can be difficult. Different approaches include scaling down the program after a certain amount of time or number of transactions or following a change in household vulnerability indicators after the assistance was provided, for example. While scaling down following a vertical expansion requires a termination of additional payments, scaling down following a horizontal expansion can be more challenging as it involves reducing the caseload. In some cases, however, beneficiaries who have been newly included in an existing social protection program as part of the horizontal expansion may qualify for continued assistance; where this is the case, these decisions need to be communicated carefully to ensure there is clarity on which beneficiaries will continue to receive benefits and which will exit the program. In each case, social

134 | Adaptive Social Protection

protection case workers should connect beneficiaries to additional, longer-term (non–shock specific) social protection programs for which they may be eligible. Similarly, stand-alone emergency programs should have a defined temporality that is well communicated to beneficiaries so they know for how long they will receive the temporary assistance.

NOTE 1. Examples of grievances include: Beneficiaries did not receive information about the program, were misinformed, did not receive the information they needed, were not notified, were treated poorly, were incorrectly classified in a certain category, did not receive payments, and received an incorrect amount.

REFERENCES Grosh, M., C. del Ninno, E. Tesliuc, and A. Ouerghi. 2008. For Protection and Promotion: The Design and Implementation of Effective Safety Nets. Washington, DC: World Bank. https://siteresources.worldbank.org/SPLP/Resources/461653-1207162275268/For _Protection_and_Promotion908.pdf. Leite, P., T. George, C. Sun, T. Jones, and K. Lindert. 2017. “Social Registries for Social Assistance and Beyond: A Guidance Note and Assessment Tool.” Social Protection and Labor Working Paper 1704, World Bank, Washington, DC. http://documents.worldbank.org/curated​ /­en/698441502095248081/pdf/117971-REVISED-PUBLIC-Discussion-paper-1704.pdf. Lindert, K., T. George, I. Rodriguez-Caillava, and Kenichi Nishikawa. Forthcoming. A Sourcebook on the Foundations of Social Protection Delivery Systems. Washington, DC: World Bank. Smith, G. 2018. “Responding to Shocks: Considerations along the Delivery Chain.” Oxford Policy Management, Oxford, UK.

ECO-AUDIT

Environmental Benefits Statement The World Bank Group is committed to reducing its environmental footprint. In support of this commitment, we leverage electronic publishing options and printon-demand technology, which is located in regional hubs worldwide. Together, these initiatives enable print runs to be lowered and shipping distances decreased, resulting in reduced paper consumption, chemical use, greenhouse gas emissions, and waste. We follow the recommended standards for paper use set by the Green Press Initiative. The majority of our books are printed on Forest Stewardship Council (FSC)–certified paper, with nearly all containing 50–100 percent recycled content. The recycled fiber in our book paper is either unbleached or bleached using totally chlorine-free (TCF), processed chlorine–free (PCF), or enhanced elemental ­chlorine–free (EECF) processes. More information about the Bank’s environmental philosophy can be found at http://www.worldbank.org/corporateresponsibility.

A

daptive social protection (ASP) helps to build the resilience of poor and vulnerable households to the impacts of large, covariate shocks, such as natural disasters, economic crises, pandemics, conflict, and forced displacement. Through the provision of transfers and services directly to these households, ASP supports their capacity to prepare for, cope with, and adapt to the shocks they faceâ&#x20AC;&#x201D;before, during, and after these shocks occur. Over the long term, by supporting these three capacities, ASP can provide a pathway to a more resilient state for households that may otherwise lack the resources to move out of chronically vulnerable situations. Adaptive Social Protection: Building Resilience to Shocks outlines an organizing framework for the design and implementation of ASP, providing insights into the ways in which social protection systems can be made more capable of building household resilience. By way of its four building blocksâ&#x20AC;&#x201D;programs, information, finance, and institutional arrangements and partnershipsâ&#x20AC;&#x201D;the framework highlights both the elements of existing social protection systems that are the cornerstones for building household resilience, as well as the additional investments that are central to enhancing their ability to generate these outcomes. In this report, the ASP framework and its building blocks have been elaborated primarily in relation to natural disasters and associated climate change. Nevertheless, many of the priorities identified within each building block are also pertinent to the design and implementation of ASP across other types of shocks, providing a foundation for a structured approach to the advancement of this rapidly evolving and complex agenda.

ISBN 978-1-4648-1575-1

SKU 211575