Exploring Universal Basic Income A Guide to Navigating Concepts, Evidence, and Practices Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov Editors Exploring Universal Basic Income Exploring Universal Basic Income A Guide to Navigating Concepts, Evidence, and Practices Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov Editors © 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 22 21 20 19 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: Gentilini, Ugo, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov, eds. 2020. Exploring Universal Basic Income: A Guide to Navigating Concepts, Evidence, and Practices. Washington, DC: World Bank. doi:10.1596/978-1-4648-1458-7. 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-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to 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 (paper): 978-1-4648-1458-7 ISBN (electronic): 978-1-4648-1511-9 DOI: 10.1596/978-1-4648-1458-7 Cover image: © Triff / Shutterstock. Used with permission. Further permission required for reuse. Cover and interior design: Nita Congress. Library of Congress Control Number: 2019954076 Praise for the Book “The debate around universal basic income (UBI) in India has catalyzed a rethink about social protection, poverty reduction, and the role of the state in development. These are issues that go beyond just simple economics. Sociologists, political scientists, and philosophers have weighed in, each bringing a different and often provocative perspec- tive about the role of UBI in social contracts that bind our societies. Gentilini et al. offer a unifying platform from which this diverse spectrum of views can be discussed and debated in an informed and intelligent way. That it can bring together people from such diverse backgrounds on a highly divisive topic is a credit to the authors. Their frame- work, analysis, and empirical rigor will serve as a wonderful guide for policy makers and practitioners grappling with questions on how to build a modern and state-of-the-art social protection system. Scholars and policy makers in India and elsewhere have been waiting for such an analysis.” Junaid Kamal Ahmad, Country Director, India, World Bank Group “Discussion of a universal basic income (UBI) can be clouded by theology, with any- thing other than a ‘pure’ UBI cast into outer darkness. The most welcome feature of this wide-ranging volume is its treatment of the design problem as multidimensional, recognizing that objectives differ, and so do constraints, notably concerning fiscal, insti- tutional, and political capacity. While addressing these multiple aspects, the book also includes empirical analysis of UBI compared with other designs and discussion of the operational tasks necessary for successful delivery.” Nicholas Barr, Professor of Public Economics, London School of Economics “Public discourse on the potential role for a universal basic income (UBI) in addressing var- ious socioeconomic challenges has soared over the last decade. However, the discussion has often been plagued by lack of clarity on what is meant by a UBI and the challenges it is intended to address. This impressive book provides a transparent and comprehen- sive framework to inform the debate. It sets out the defining features of a UBI, the various socioeconomic issues it may help address, and the pros and cons of a UBI in various eco- nomic and political settings. It makes clear that the attraction of a UBI, or some partial variant, will depend on country-specific social, economic, and political preferences, as well as the underlying administrative and fiscal contexts. The detailed empirical analysis helps to bring these issues out into the light for much-needed scrutiny. It also helps to hit home the too-often neglected importance of considering both the tax and transfer sides of the debate to avoid drawing misleading policy conclusions. The passionate debate will no doubt continue, but this book increases the likelihood that it will now be complemented by a healthy dose of reason.” David Coady, Division Chief, Expenditure Policy Division, International Monetary Fund “The idea of universal basic income—giving cash unconditionally to everyone—has been hotly debated, mainly in developed countries. This book sheds much-needed light on that debate by providing the first dispassionate analysis of UBI in developing countries. All of the issues—poverty impact, fiscal sustainability, labor market outcomes, political economy—are elucidated with evidence. Policy makers may adopt or reject UBI, but after reading this book, they will do so with clear-eyed reasoning.” Shantayanan Devarajan, Professor of Practice of Development, Georgetown University “Universal basic income has been gaining traction as a potential solution to poverty and technological unemployment. This book is the ultimate guide for anyone inter- ested in universal basic income at the global level. The authors leave no stone unturned, examining the economics and politics of universal basic income, as well as policy imple- mentation issues across the world.” Ioana Marinescu, Assistant Professor of Economics, University of Pennsylvania “Today’s UBI debates are powerful because of the deep questions they raise—what is the nature of the social contract? What can and cannot a state with limited capacity realisti- cally do? This book provides a helpful framework for navigating the issues and grounding the debate in data. It should be a standard reference.” Paul Niehaus, Associate Professor of Economics, University of San Diego, and co-founder GiveDirectly “This is the first time the World Bank has taken up the case for a basic income in a con- structive manner, and should be welcomed by advocates and critics alike. As someone who has advocated moving in that direction for many years and been involved in pilots in several countries, I firmly believe it will be an anchor of 21st century income distri- bution. It is not a panacea, but giving people basic economic security is something that should unite us all.” Guy Standing, author of Basic Income: A Guide for the Open-Minded (Yale University Press, 2017) “Universal basic income (UBI) is one of those potentially transformational ideas in both developing and advanced countries, although for very different reasons. This terrific and timely volume is a comprehensive guide to the conceptual and implementation issues relating to UBI. A must-read.” Arvind Subramanian, former Chief Economic Adviser to the Government of India, and Visiting Lecturer in Public Policy, Harvard University “Universal basic income (UBI) is far more than a thought experiment—it’s a policy idea worthy of the in-depth consideration provided in this book. At UNICEF, we examine and share evidence regarding the potential and design considerations of universal child benefits: these can be seen as a subset of UBI and, therefore, help contribute to our common understanding of such interventions. We’re pleased to see these themes dis- cussed and hope the many other lessons this book provides also help to shape thinking about strengthened forms of social protection for children and young people.” Alexandra Yuster, Associate Director, Programme Division, and Chief of Social Policy, UNICEF Contents Preface xiii Acknowledgments xv About the Authors xvii Abbreviations xxi Overview: Exploring Universal Basic Income 1 Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov References 14 Chapter 1. The Idea of Universal Basic Income 17 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov A Glimpse at History 18 Anatomy of UBI 20 Universality 24 Conditionality 30 Transfer Modality 34 Choices on Other Parameters 38 Three Stylized Narratives on UBI 43 Social Protection 43 Automation and Labor Market Disruptions 47 Resource Dividends 52 Conclusions 57 Notes 59 References 62 vii viii Contents Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 73 Ugo Gentilini and Margaret Grosh Types of Programs 74 Benefit-Based Programs Not Based on Work 76 Child Allowances and Social Pensions 76 Guaranteed Minimum Income 77 Other Poverty-Based Cash Transfers 78 Tax-Based Programs Not Based on Work 79 Benefit-Based Programs Based on Work 81 Job Guarantee Programs 81 Temporary Public Works Programs 83 Wage Subsidy Programs 85 Tax-Based Programs Based on Work 86 Conclusions 88 Notes 92 References 93 Chapter 3. Universal Basic Income and Work 99 Francesca Bastagli The Arguments 100 The Nature of the Evidence 101 Participation in Paid Work and Financial Work Incentives 102 Conditions of Paid Work 108 Valuation and Distribution of Unpaid Work 109 Formal and Informal Work 112 Conclusions 113 Work Incentives and Participation in Paid Work 114 Conditions of Work 115 Valuation and Distribution of Unpaid Work 115 Formal and Informal Work 116 Notes 116 References 117  Contents ix Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 123 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan Data 125 Methodology 129 Results 131 Poverty and Inequality in the Baseline Scenario 132 Replacing Social Assistance Programs with a Budget-Neutral UBI 132 Replacing Baseline Social Assistance Programs with a More Generous UBI 140 Conclusions 147 Notes 150 References 151 Chapter 5. Financing a Universal Basic Income: A Primer 153 Teresa Ter-Minassian Assessing the Fiscal Space for a UBI 154 Assessing the Scope for Financing a UBI through Expenditure Measures 158 Expenditure Savings 158 Subsidies Reform 159 Reforms of Public Wages and Employment 162 Assessing the Scope for Financing a UBI through Revenue-Raising Measures 164 Taxes on Personal Income 167 Taxes on Corporate Income 170 Taxes on Wealth 170 Taxes on Consumption 172 Other Potential Revenue Sources 173 Conclusions 175 Notes 177 References 179 Chapter 6. The Political Economy of Universal Basic Income 183 Jurgen De Wispelaere and Ruslan Yemtsov A Stylized Model of Political Economy 185 Constituencies, Ideologies, and Coalitions 191 Political Transformations of the UBI 198 Conclusions 203 Notes 205 References 209 x Contents Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 217 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini Delivery Chain 217 Phase 1: Assessing 217 Phase 2: Decision Making 221 Phase 3: Implementation 221 Surrounding Ecosystem: Foundational Elements 224 ID Systems 224 Civil Registration Systems 228 Interoperability and Data Protection Frameworks 229 Payment Service Provision 230 Summary 232 Conclusions 232 Notes 233 References 233 Appendixes A UBI-Related Pilots 237 B UBI-Related Proposals 247 C Quantitative Evidence on Conditional and Unconditional Transfers 261 D Microsimulations Data and Methodology 271 E Annotated Bibliography 281 Boxes 1.1 Negative Income Tax Pilots 24 1.2 Welfare Continuity and Poverty Dynamics 29 1.3 Evidence from Systematic Reviews on the Impact of Conditions in Cash Transfer Programs 32 1.4 Do Cash Transfers Affect Mobility? 41 1.5 Rights Architecture and Select Legislative Provisions 45 1.6 UBI Proposals in India 48 1.7 Jobs and Societies 50 2.1 Balancing Wage Subsidy Compliance and Generosity 87 4.1 Baseline Welfare State Typologies 126 6.1 Mapping Ideological Attitudes on Basic Income across the Political Spectrum in Finland 196 6.2 The Politics Surrounding the Introduction of Universal Child Allowances in the United Kingdom 199 7.1 UBI within an Integrated Social Registry 219  Contents xi Figures O.1 UBI within a Social Assistance Cube 3 O.2 Basic Framework for Navigating UBI Decision Making 6 O.3 Poverty Reduction Effects of a UBI and Baseline Cash-Based Programs 7 O.4 Distributional Effects of a Budget-Neutral UBI Reform, Average across 10 Sampled Countries 8 1.1 UBI within a Social Assistance Cube 21 1.2 Cost of a UBI by Poverty and Demographic Variables 41 1.3 Modalities for UBI Phase-in 43 1.4 Alaskan Permanent Fund Dividend Beneficiaries and Annual Percentage Change in Dividends, 1982–2017 54 1.5 Global Copper Prices and Universal Basic Income in Mongolia 56 B1.2.1 Welfare Distribution in Bangladesh 29 B1.6.1 Comparison of India’s Select UBI Proposals 49 2.1 Benefit Structures of Select Interventions 75 2.2 Benefit Incidence Net of Taxes: Simulated UBI and SNAP in the United States 75 2.3 NIT Benefit Structure Based on Friedman’s 1970 Family Assistance Plan Model 80 2.4 The Inverse U-Shaped Benefit Structure of the EITC 88 4.1 Baseline Poverty and Inequality Indicators 132 4.2 Poverty Impacts of Budget-Neutral UBI Reform 133 4.3 Progressivity and Relative Performance of a Budget-Neutral UBI Reform 135 4.4 Inequality Impacts: Reduction in Gini Coefficient 136 4.5 Winners and Losers from a Budget-Neutral UBI Reform 137 4.6 Winners and Losers across Welfare Deciles 139 4.7 Universal versus Adults-Only UBI: Impact on Children Age 0–6 140 4.8 Poverty Impacts of More Generous UBI Schemes (Spending Only) 141 4.9 Total Spending as a Percentage of GDP by UBI Scenario 142 4.10 Squared Poverty Gap Impact for Consumable Income under Baseline and Different UBI Scenarios 145 4.11 Incidence of Taxes for the Top 10 Percent under Alternative UBI and Financing Scenarios 146 5.1 General Government Gross Debt as a Percentage of GDP, 2009–17 156 5.2 Variations in General Government Gross Debt, 2017 157 5.3 Selected Countries Where Pretax Energy Subsidies Exceeded 2 Percent of GDP in 2015 160 5.4 General Government Employment as a Percentage of the Working-Age Population, 2015 163 xii Contents 5.5 Public Wage Premium as a Percentage of Average Private Wage 163 5.6 Average 2005–15 Revenue/GDP Ratios for Selected Country Groups 165 5.7 Variations by Country Group in Average 2005–15 GDP/Revenue Ratios 165 5.8 Average 2005–15 PIT/GDP Ratios 168 5.9 Average 2005–15 CIT/GDP Ratios 170 5.10 Average 2005–15 Consumption Tax/GDP Ratios 172 6.1 Three Main Components of the Model and Basic Income Variations 186 6.2 The Demand-Capacity Paradox 188 B6.1.1 Political Party Support for Basic Income in Finland, 1979–2015 196 B6.1.2 Support for Basic Income and Left-Right Ideological Position 197 7.1 UBI Delivery Chain Compared to Typical Social Assistance Program 218 7.2 Population Coverage of Social Registries by Country 220 7.3 Percentage of Populations without Official Proof of Identity 226 7.4 National and Digital ID Systems in Sub-Saharan Africa, 1960–2018 227 7.5 Birth Registration Coverage by Region 229 7.6 Percentage of Adults (15+) with an Account at a Financial Institution or with a Mobile Money Provider 231 D.1 Coverage and Incidence of Baseline Programs 278 Tables 1.1 Which Initiative Is Currently a Pure UBI? 22 1.2 Barriers to Coverage 27 1.3 Alignment of UBI Narrative with Basic Transfer 39 2.1 Program Typology 74 2.2 Summary Comparison of Intervention Features 90 B4.1.1 Illustrative Welfare State Typologies 129 5.1 Revenue Sources Scored on Different Criteria 166 C.1 Comparative Impact Evaluations 261 C.2 Systematic and Literature Reviews 265 D.1 ASPIRE Database Use 272 D.2 Selected Social Assistance Programs by Country 273 D.3 Indonesia Simulated Benefit Levels for Main Social Assistance Programs 276 D.4 Evolution of Indonesia’s Social Assistance Budget 276 Preface A t the time of writing this preface, electoral debates in India, which featured uni- versal basic income (UBI) in a prominent way, just subsided; pilot programs are rolled out in several cities in the United States and Europe; a decade-long trial is under way in rural Kenya; and the World Bank is, through this very volume, issuing its first analysis on the matter. So why all this interest on a seemingly utopian and radical proposal of “just give cash to everyone”? Interest in UBI is surely symptomatic of larger societal discomforts. The chang- ing nature of work in higher-income countries demands that social protection systems co-evolve with it. While automation, globalization, and diversification of employment bolstered efficiency and productivity gains, median income and living standards have not always risen accordingly—and in some cases, they have been stagnant for decades. Lower-income contexts, where work arrangements have not changed as dramatically, face different challenges. Among them, pervasive poverty and informality, compounded with limited government capacities and revenues, are preventing hundreds of millions of people from accessing higher-productivity activities, being protected from risks, and building human capital. With societal anxieties brewing, there are new opportunities for rethinking how to forge a more inclusive social contract, including with universal social protection at the core, and do so in ways that leapfrog past models. While new technologies are expand- ing the delivery frontier, the notion of “universal” social protection is subject to different interpretations. Specifically, universality can be attained in different ways—for example, by combining assistance and insurance programs, by combining different safety net measures, or, as in the case of UBI, achieving such goals via a single measure. Put differ- ently, a UBI is a shortcut to universality. On closer scrutiny, however, such a shortcut is less straightforward than it seems. A UBI looks alluringly simple on the surface, since it provides cash unconditionally and with no targeting involved. But its implications are complex and largely unknown. In fact, the scale of UBI makes it a systemwide intervention, not just a program. As such, it may affect, for instance, several labor market issues such as unemployment insurance, severance pay, unionization, contributory pensions, and minimum wages. With no UBI xiii xiv Preface program of national scale currently in place, most debates are shaped by informed views and inference from smaller-scale schemes rather than from hard evidence and actual practices. We should be humble about what we know and what we do not on UBI. A UBI is also less radical than it appears. Depending on how it is financed, the program could end up distributing differentiated amounts of cash to different people— and some may not receive any transfer at all (the “net payers”). In other words, the net effects of benefits and financing could make a UBI a targeted program via taxes (in addition to participation based on residency and age). Precisely because a UBI may be de facto targeted, there is a need to clarify how it differs from or complements other social assistance instruments targeted by income (e.g., guaranteed minimum income programs), categorical parameters such as age (like social pensions), or other eligibility criteria. What specific problem is UBI ultimately trying to solve? How does it perform rel- ative to existing systems? Under what circumstances is the program more or less likely to be cost-effective? It is precisely this set of quandaries—on why, whether, where, and how to consider UBI—that animated the conception and production of this book. The volume should not be interpreted as a statement for or against UBI in the abstract. Instead, it engages in the more laborious, nuanced effort of providing an organizing framework—a struc- tured thought process—to gauge the many issues that surround the appropriateness and feasibility of UBI. The framework, which integrates choices around objectives, design, implementation, performance, political economy, and financing matters into a coherent device, can help guide and inform decisions in different contexts—on UBI, as well on virtually any social assistance program. Ultimately, deliberations on UBI should be based on robust and balanced thought processes, and this book is poised to inject a much-needed dose of analytics into a debate too often prone to ideology. Michal Rutkowski Global Director Social Protection and Jobs Global Practice World Bank Group Acknowledgments T he production of this book was led by Ugo Gentilini under the strategic guidance of Michal Rutkowski, Lynne Sherburne-Benz, and Anush Bezhanyan. The task would not have been possible without the generous support of the World Bank’s Rapid Social Response (RSR) Multi-Donor Trust Fund, including with generous support from the Russian Federation, the United Kingdom, Norway, Sweden, Australia, and Denmark. The editors of the volume are grateful to the book’s peer reviewers for their thought- ful advice, including David Coady, Maura Francese, Gabriela Inchauste, Delphine Prady, and William Wiseman. They are equally thankful to several global, regional, and country colleagues for their feedback and support, including Pablo Acosta, Afrah Alawi Al-Ahmadi, Jumana Alaref, Shrayana Bhattacharya, John Blomquist, Hana Brixi, Gabriel Demom- bynes, Sascha Djumena, Roberta Gatti, Yashodhan Ghorpade, Camilla Holmemo, Andras Horvai, Keiko Inoue, Johannes Koettl, Phillippe Leite, Mattia Makovec, Matteo Morgandi, Edmundo Murrugarra, Nga Nguyet Nguyen, Pedro Olinto, Stefano Paternostro, Juul Pinx- ten, Snjezana Plevko, Aleksandra Posarac, Maria Laura Sanchez Puerta, Gevorg Sargsyan, Sandor Sipos, Carolyn Turk, Veronica Silva Villalobos, and Eric Zapatero. The book’s content benefited enormously from discussions with Junaid Kamal Ahmad, Asad Alam, Nicholas Barr, Christian Bodewig, Maurizio Bussolo, Joe Chrisp, Aline Coudouel, Shantayanan Devarajan, Simeon Djankov, Louise Fox, Marcelo Giugale, Kathy Lindert, Luke Martinelli, Rinku Murgai, Farhad Nili, Philip O’Keefe, Ian Orton, Berk Özler, Truman Packard, Robert Palacios, Aleksandra Posarac, Dhushyanth Raju, Martin Rama, Djavad Salehi-Isfahani, Indhira Santos, Nika Soon-Shiong, David Stewart, Lindsay Stirton, Ramya Sundaram, Hamid Tabatabai, Nithin Umapathi, and Alexandra Yuster. Excellent research assistance was provided by Vanessa Moreira, Boban Varghese Paul, Saksham Khosla, Devarakonda Priyanka Kanth, Rong Chen, and Yucheng Zheng. The Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) team at the World Bank, including Oleksiy Ivaschenko, Claudia Rodriguez-Alas, Marina Novikova, and Usama Zafar, kindly provided household survey and administrative data. Outstanding administrative support was rendered by Helena Makarenko, Inas Ellaham, Harifera Raobelison, Angela Rubio Martinez, and Francine Pagsibigan. xv xvi Acknowledgments Patricia  Katayama and Aziz Gokdemir managed the editorial production and printing process, and Nita Congress copyedited, designed, and laid out the book. Finally, the team received valuable feedback from participants at various consulta- tions, including events at the World Bank, the International Labour Organization, and the United States Agency for International Development, as well as country-level seminars in India, the Russian Federation, and South Africa. About the Authors FRANCESCA BASTAGLI is Head of the Equity and Social Policy Programme and Prin- cipal Research Fellow at the Overseas Development Institute (ODI). She specializes in research and policy advisory work on the design, implementation, and evaluation of social policies, with a focus on social protection policies and their poverty, inequality, and employment outcomes. She is lead author of the report “Cash Transfers: What Does the Evidence Say?”(ODI, 2016) and coauthor of the book Wealth in the UK: Distribution, Accumulation, and Policy (Oxford University Press, 2013). She has led policy analysis and contributed to policy dialogue in countries in Latin America, the Middle East and North Africa, South and East Asia, West Africa, and Europe. Prior to joining ODI, Bastagli was Research Fellow at the Centre for Analysis of Social Exclusion at the London School of Economics and previously held positions at Bocconi University and the World Bank. In 2011 she was Visiting Scholar at the International Monetary Fund’s Fiscal Affairs Depart- ment. Bastagli holds a PhD in social policy from the London School of Economics and a laurea in economics from Bocconi University. JURGEN DE WISPELAERE is a Policy Fellow with the Institute for Policy Research, Univer- sity of Bath, and a Guest Professor at Tampere University. A former occupational therapist turned policy scholar, he previously worked at the University of Tampere, McGill University, Trinity College Dublin, and University College Dublin. He has been a visiting scholar at a number of prestigious institutions, including Yale University, Oxford University, Australian National University, and, most recently, the Danish Centre for Welfare Studies, University of Southern Denmark. His main research interest is the political analysis of basic income, a topic on which he has published extensively in leading international journals as well as specialist edited volumes. He coedited Basic Income: An Anthology of Contemporary Research (Wiley-Blackwell, 2013) and is currently coediting The Routledge Handbook on Basic Income (Routledge, forthcoming). He is also completing a book on basic income experiments with Evelyn Forget (Policy Press, forthcoming). In 2006, he cofounded Basic Income Studies, an international journal on basic income research, and was its coeditor from 2006 to 2012. He has been a consultant to several government departments and nongovernmental organizations on basic income and basic income experiments. xvii xviii About the Authors UGO GENTILINI serves as Global Lead for Social Assistance with the Social Protection and Jobs Global Practice at the World Bank. He has 20 years of experience in the analyt- ics, practice, and evaluation of social protection systems, particularly in the realm of cash transfers, food assistance, price subsidies, public works, and select active labor market policies. Prior to joining the Bank, he was with the World Food Programme, where he contributed to the organization’s initial engagement in cash transfers and national safety nets. His publications encompass flagship reports, edited volumes, academic jour- nals, and operational guidelines. Gentilini holds a PhD in development economics and produces a newsletter on social protection (ugogentilini.net) reaching thousands of prac- titioners on a weekly basis. TINA GEORGE is a Senior Public Sector Specialist at the World Bank. Over the past 19 years, she has worked on public service delivery and the implementation of investment project operations in technologies for the poor and vulnerable. Her work focuses on social protection delivery systems, social registries, social safety nets, government-to-person payment systems, foundational identification systems, human-centered design, insti- tutions and governance, the impact of technology on the future of work, and public service delivery platforms serving the poor. Her most recent publication is A Sourcebook on the Foundations of Social Protection Delivery Systems (with Kathy Lindert and Inés Rodriguez-Caillava; forthcoming). She earned a PhD in technology, policy, and manage- ment at the Technische Universiteit Delft in the Netherlands, investigating the design of mobile platforms to deliver services to the poor. MARGARET GROSH is the Senior Adviser for the World Bank’s Social Protection and Jobs Global Practice. She has extensive experience with social protection both in crisis response and for improving equality of opportunity. Among her most notable publica- tions are Understanding the Poverty Impact of the Global Financial Crisis in Latin America and the Caribbean (with Maurizio Bussolo and Samuel Freije, 2014), For Protection and Promotion: The Design and Implementation of Effective Safety Nets (with Carlo del Ninno, Emil Tesliuc, and Azedine Ouerghi; 2008), and Targeting of Transfers in Developing Coun- tries: Review of Lessons and Experience (with David Coady and John Hoddinott, 2004). Grosh served as Practice Manager for the Social Protection, Labor and Jobs Global Prac- tice in Latin America and the Caribbean from 2014 to 2017; previous positions held were Lead Economist in Latin America and the Caribbean, Lead for the Anchor Safety Nets/ Service Delivery Systems Team, Leader of the Global Expert Team for Safety Nets, and Team Lead for the Living Standards Measurement Study in the Development Research Department. She began her career in operations in Latin America and the Caribbean Human Development. She has a PhD in economics from Cornell University. CHRISTINA LOWE is a researcher with the social protection team at the Overseas Devel- opment Institute, and previously worked as a consultant in the World Bank’s Social Protection and Jobs Global Practice. Her main research interest is in improving the deliv- ery of social protection programs to better serve marginalized populations, including About the Authors xix through improved linkages with other services and sectors. She has worked on social policy and program design in Sub-Saharan Africa, Latin America, South Asia, and Europe through roles at the World Food Programme, the UN Refugee Agency (UNHCR), and a range of nonprofit organizations. She holds a BA in modern languages from the Univer- sity of Cambridge, and a master’s in public policy from the John F. Kennedy School of Government at Harvard University. NORA LUSTIG is Samuel Z. Stone Professor of Latin American Economics and Director of the Commitment to Equity Institute at Tulane University. She is also a Nonresident Senior Fellow at the Brookings Institution, the Center for Global Development, and the Inter-American Dialogue. Her research is on economic development, inequality, and social policies with an emphasis on Latin America. Her most recent publication—Commitment to Equity Handbook: Estimating the Impact of Fiscal Policy on Inequality and Poverty (Brookings Institution Press, 2018)—provides a step-by-step guide to assessing the impact of taxation and social spend- ing on inequality and poverty in developing countries. Lustig is a founding member and President Emeritus of the Latin American and Caribbean Economic Association and was a co-director of the World Bank’s World Development Report 2000: Attacking Poverty. She serves on the editorial board of the Journal of Economic Inequality and is a member of the Society for the Study of Economic Inequality’s Executive Council. Lustig served on the Atkinson Com- mission on Poverty, the High-Level Expert Group on Measuring Economic Performance and Social Progress, and the G20 Eminent Persons Group on Global Financial Governance. She received her doctorate in economics from the University of California, Berkeley. EMMA MONSALVE MONTIEL is a Social Protection Specialist in the Social Protection and Jobs Global Practice at the World Bank. She joined the Bank in 2014 and worked with the Social Protection and Jobs Global Practice and the Poverty and Equity Global Practice in the Africa and Latin America and the Caribbean regions. Before joining the World Bank, she worked with the Inter-American Development Bank and the Central Bank of Colombia. She coauthored the Africa Social Protection Regional Study flagship Realizing the Full Potential of Social Safety Nets in Africa (with Kathleen Beegle and Aline Coudouel, 2018). She has served as a task team member for several World Bank proj- ects, including in Cabo Verde, Rwanda, and São Tomé and Príncipe. For the past two years, she has been providing technical and operational support to help the govern- ment of Angola define social protection policies, strategies, and social protection and jobs program priorities. She holds a master’s in applied economics from Johns Hopkins University and a bachelor’s in economics from the University of Antioquia in Colombia. SIYU QUAN is a PhD candidate at the Department of Economics, Tulane University. Her main research interests include poverty and inequality, child labor issues, and charter school reforms in New Orleans. She holds a master’s in economics and policy analysis from Tulane University and a bachelor’s in psychology from the University of Complu- tense in Spain. She is a Research Assistant at the Commitment to Equity Institute and coauthored the tax and benefit incidence study “The Impact of Taxes and Social Spending xx About the Authors on Inequality and Poverty in Spain” (with Nora Lustig, 2016). She also worked as an intern at the Inter-American Dialogue and coauthored a 2018 study on educational attainment in Latin American countries. Her PhD thesis is “Does the Closing of Sweatshops in Ban- gladesh Save Child Workers?—An Empirical Study of a U.S. Child Labor Prohibition Law.” JAMELE RIGOLINI is a Lead Economist and the Program Leader for Human Development for the Western Balkan countries at the World Bank. His areas of expertise include social protection, human development, labor markets, poverty, gender, and entrepreneurship policies. During his career at the Bank, he has worked on human development, poverty reduction, and labor market policies in East Asia and the Pacific, Latin America and the Caribbean, Africa, and Eastern Europe. He also managed several of the World Bank’s flag- ship reports for Latin America and the Caribbean, maintaining close dialogue with other international organizations and Latin American academic institutions and think tanks. Prior to joining the World Bank, he was an Assistant Professor of Economics at the Uni- versity of Warwick (United Kingdom). Rigolini holds a degree in physics from the Swiss Institute of Technology in Zürich and a PhD in economics from New York University. He has published articles in leading economics journals, including the Journal of Public Eco- nomics, the Journal of Development Economics, Economic Letters, and World Development. TERESA TER-MINASSIAN is an international economic adviser and fiscal expert. Before becoming a consultant, she had a long career with the International Monetary Fund (IMF), which she joined in 1972. At the IMF, she served in the European, Western Hemisphere, and Fiscal Affairs Departments, where she led missions to, among others, Greece, Italy, and Spain, and negotiations with Argentina, Brazil, and Portugal. She was the director of the IMF’s Fiscal Affairs Department from 2001 to 2008. Ter-Minassian has published more than 40 papers and books on fiscal issues, particularly on macrofiscal and intergov- ernmental fiscal relations areas. She holds a degree in law from the University of Rome and a PhD in economics from Harvard University. RUSLAN YEMTSOV is Human Development Program Leader for China, Mongolia, and the Republic of Korea at the World Bank. Previously, he was the Lead for Social Safety Nets Global Solutions Group with the Social Protection and Jobs Global Practice. During his career at the World Bank, he has also worked in the Middle East and North Africa as well as the Eastern Europe and Central Asia regions. He has coauthored reports, books, and research articles on poverty, inequality, labor markets, the efficiency of public spending, and the political economy of reform. This work has covered global, regional, and country issues, including poverty assessments, public expenditure reviews, and rural development. Opera- tional experience includes working on social assistance reform projects (Croatia, Morocco, the Russian Federation, Tunisia, Uzbekistan), fuel subsidy reforms (Arab Republic of Egypt, Jordan, Morocco, Tunisia), crisis response (Djibouti), local social development funds (Egypt, Georgia), and cash transfer delivery systems (Croatia, Russia, Tunisia). In addition to his research and operational work, Yemtsov has organized major training events and authored several training manuals. He holds a PhD from Moscow State University. Abbreviations ASPIRE Atlas of Social Protection Indicators of Resilience and Equity BISP Benazir Income Support Programme CCT conditional cash transfer CIT corporate income tax EITC earned income tax credit ESS European Social Survey GDP gross domestic product GMI guaranteed minimum income ID identification IMF International Monetary Fund JGP job guarantee program NIT negative income tax NREGS National Rural Employment Guarantee Scheme OECD Organisation for Economic Co-operation and Development PDS Public Distribution System PIT personal income tax PPP purchasing power parity PROGRESA Programa de Educación, Salud y Alimentación SAR special administrative region SNAP Supplemental Nutrition Assistance Program UBI universal basic income UCT unconditional cash transfer VAT value-added tax xxi OVERVIEW Exploring Universal Basic Income Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov U niversal basic income (UBI) is a hotly debated idea. In fact, few development topics elicit as much interest and controversy as UBI. There is literally a book published on the subject every month, with the concept being examined across the economics, sociology, governance, philosophical, and political science litera- ture. It is prompting both curiosity and visceral reactions from policy makers in high- and lower-income countries alike, including playing a role in political discourse and elec- tions (Banerjee, Niehaus, and Suri 2019; Hoynes and Rothstein 2019). And the growing number of experiences and pilots, with variants dating back to the 1970s, is moving UBI “from a thought experiment to a concrete policy option” (Calnitsky 2017). A UBI holds an attractive promise of change across many lines. These include cov- erage potential, fairness in social contracts, power relations in labor markets, and gender equity, among others. It may speak, for some, to the appetite for social justice generated by glaring and growing inequalities in societies (Stern 2016). From this standpoint, a UBI engenders interest as a societal ideal to which to aspire, and not merely a program (Lowrey 2018). For others, a UBI is poised to mitigate the effects of purported massive job losses from automation, streamline the chaotic plethora of state-provided schemes, or empower people by redirecting natural resource–related revenues from public coffers to citizens (Devarajan 2018; Yang 2018). Overall progress in social protection systems deserves global celebration, but in many cases, the degree of frustration with those systems is palpable. In a world riddled with fears about artificial intelligence, exhaustion over complex bureaucra- cies, and resentment toward politics, the transparency and simplicity of a UBI is alluring. These diverse rationales explain why a UBI resonates among different audiences. UBI enlists advocates from those embracing a minimalist role of the state to human 1 2 Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov rights activists. Some look at a UBI as a foundation to build stronger states; others see it as a milestone toward rolling back public action and its interference with private lib- erties—that is, a UBI could embody the “Trojan horse” evoked by Milton Friedman in the late 1960s and Jean Drèze nowadays (see, e.g., Drèze 2017; Friedman 1967). The fact that a UBI generates support from some political conservatives, libertarians, and progressives alike—and from parts of the tech industry and select trade unions—is a remarkable feature. Such a heterogeneous coalition may help advance the idea, but its practical implementation would expose the lack of coherent expectations and objectives. For instance, hopes around a UBI as a societal revolution may be tempered by prosaic forces. After all, the ultimate generators of inequities may lie elsewhere, for example, in uneven access to education and health systems, low-paying and low-productivity jobs, poorly functioning markets, corruption, regressive tax codes, unequal pay, and social discrimination, among others (Piketty 2016). From this perspective, a UBI by itself could help, but the hopes bestowed on the concept seem excessive. The prominence of ideological forces and different expectations suggests the need for a balanced and evidence-based approach (Francese and Prady 2018; Hanna and Olken 2018). This volume does not aim to provide strict prescriptions for or against a UBI, but instead a framework within which to think about it. We aim to provide a com- pass to help navigate key issues, elucidate trade-offs, and offer new data and analysis to better inform choices around the appropriateness and feasibility of a UBI in differ- ent contexts. Unlike the bulk of UBI literature, which is skewed toward high-income societies, we examine the program primarily in the context of low- and middle-income countries. We intend to provide policy makers and practitioners with a realistic sense of the entire gamut of policy considerations; offer new quantitative insights around key choices and implications; and frame the issues within a coherent, objective, and compre- hensive volume concisely capturing global knowledge on the topic. By doing so, we shed light on the possible contexts where a UBI may be more or less viable based on a range of considerations. The overall analysis is conducted within a genuine spirit of curiosity, combined with a dose of empirical inquiry and a clear-eyed view of the progress and challenges in the current state of practice. Our multidisciplinary assessment shows that a UBI presents advantages and limitations just like any other social assistance program. After all, it could be considered a variant on existing age-based categorical schemes. Yet the scale and likely systemwide effects of a UBI program are exceptional and, as such, put a high premium on analytical and operational due diligence. Currently, no country has a UBI in place, although there have been (and still are) several small-scale pilots and a few larger-scale experiences. Only two countries—Mon- golia and the Islamic Republic of Iran—had a national UBI in place for a short period of time. Subnational experiences, such as in Alaska, are providing valuable insights, but these are constrained in one or more features (e.g., frequency and adequacy of trans- fers). The large majority of UBI pilots are variants of targeted schemes. For example, the proposal by Felman et al. (2019) on a “quasi-universal basic rural income” for India is simply a variant of a traditional guaranteed minimum income program. Quasi-UBI programs constitute the vast majority of so-called UBI pilots laid out in chapter 1. The Overview: Exploring Universal Basic Income 3 reframing of different programs in UBI or quasi-UBI terms is unhelpful because (1) this confuses and polarizes the current debate by trading accuracy for public resonance; (2) it risks reinventing the wheel around key questions for which there might be a consid- erable knowledge base (e.g., are cash transfers spent wisely? Do quasi-UBI programs discourage work?); (3) it widens the gulf between the actual shape of a program and its expectations; and (4) it may not always elucidate the nuanced, distinct features that a suite of alternative social protection measures possess to pursue similar objectives. Pilot- ing at least two features of a classic UBI might still produce insightful information, as well as elicit public and policy debates. However, there are systemwide questions—around financing, inflation, linkages to pensions, relationship to minimum wages, and the polit- ical economy—that pilots cannot fully answer. A UBI is a program to be delivered in cash, unconditionally, and to everyone. A UBI is the simple combination of three complex debates (figure O.1). Its design features—all in cash, no conditions, and no targeting—challenge current practices to varying degrees. FIGURE O.1  UBI within a Social Assistance Cube Public Distribu- Productive Safety tion System in Net Programme in Supplemental Nutrition India Assistance Program Ethiopia (SNAP food stamps) in National the United States Rural Em- ployment Guarantee Scheme in India Dibao in School China feeding Guaranteed Conditional Public Needs minimum cash works based income transfers Targeting Social Categorical pensions In-kind Universal Universal basic Job ity income guarantee Vouchers al programs od m Cash er sf Unconditional Conditional Conditional an (services) (work) Tr Conditionality 4 Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov For instance, while those three core choices largely shape the identity of a UBI, propos- als differ somewhat in their parameters about how much or how often to pay, whether to include truly everyone or exclude children and noncitizens, and whether some people would be net payers for the program (the latter two considerations already compromise the no-targeting principle). The dramatic expansion in cash transfer programs globally, and the generally positive evidence underlying them, suggests that the “in cash” part of the UBI design may be relatively uncontroversial. Nonetheless, the large-scale in-kind and food-based assistance programs present in virtually every country suggest that soci- eties still opt to maintain a combination of transfer modalities, based on philosophical, political economy, and technical grounds (Alderman, Gentilini, and Yemtsov 2018). In fact, especially in low-income and fragile settings, there might be locations or periods where large-scale cash programming might be less suitable than in-kind provisions. Sim- ilarly, there are many unconditional programs coexisting with a variety of conditional ones, showing again a tension between giving recipients autonomy and taking a some- what more directive approach (Marinescu 2018; Ravallion 2018). But perhaps the aspects of UBI that most challenge current practice are reaching everyone and doing so with an equal amount of support, independent of needs (Hanna and Olken 2018). It is important to note the diversity in definitions of “coverage” and “universality,” as well as recognize the multiple pathways toward universality (Gentilini, Grosh, and Rutkowski 2019; Packard et al. 2019). For example, universality can be interpreted in terms of outcomes (e.g., all people should be guaranteed a minimum level of welfare) or of receipt (“everyone should be covered”). The social insurance and health literature defines coverage in risk terms (a payout is a promise for a payment in case a speci- fied event occurs). In social assistance terms, coverage is receipt based—people are considered covered only when transfers are actually received. Similarly, some consider a universal transfer to be one based on no other criteria than age (thus many social pension programs would be considered universal); others define universality as reach- ing everyone in society independent of age, income, or other criteria (this view would classify child grants and social pensions as categorical programs targeted by age). In the book, we use coverage in social assistance terms, and universality as applying to all society. But a UBI is not the only path toward universality, but rather one among many. Universality should be considered at the system level: universality in social protection, which lies at the core of global commitments and the rights architecture, does not neces- sarily imply universality via a single program. Whether through a UBI or social protection more broadly, universality would need to be progressive and ensure that the most in need receive support to meet their wider range of vulnerabilities and necessities. A grad- ual building of a solid platform of social assistance, whether via one program or many, should proceed from the bottom up. Focusing on the “U” of UBI, the rationale for making transfers universal rests on five main arguments. First, by not establishing eligibility criteria (besides perhaps cit- izenship or established residency and age, e.g., for those above age 18), universality circumvents the contentious issue of exclusion and inclusion errors that are inherent in needs-based targeting. Under a UBI, there would be no such errors, as everybody is Overview: Exploring Universal Basic Income 5 included by design, hence achieving substantial expansions in coverage. Second, uni- versality may eliminate any stigma affecting beneficiaries. Evidence from Europe, for example, shows that shame is among the key factors behind limited take-up of benefits by eligible beneficiaries. Third, by changing the default position of people from being potential beneficiaries (subject to eligibility verification) to guaranteed recipients, there may be fewer transaction costs involved in accessing benefits (e.g., there is no need to spend time in applying), and various economic and psychological benefits stemming from a stable source of income over time (e.g., stress reduction, empowerment, avoiding taking desperate actions out of economic hardship). Fourth, a universal transfer would be more labor compatible than most programs, as it removes the price effect of transfers (i.e., the reduction in labor supply to avoid a reduction in benefits). And finally, universal- ity may strengthen programs’ political sustainability as beneficiaries (and voters) would draw from the entire income distribution. The case against the “U” in UBI rests principally on cost, fit for purpose, and a differ- ent appreciation of the magnitude of its possible benefits. The cost of making significant transfers universal is quite high. Depending on how these are financed—a reduction in existing social protection spending, a reduction in regressive subsidies, increased taxes— there are important changes in distributional outcomes among income and age groups that may or may not be desirable. Additionally, the flat benefit structure may not be fit for all purposes. It cannot be as redistributive as a more progressive structure and thus may have muted impacts on poverty and inequality. The flat structure does not respond to large and often short-run changes of state such as catastrophic illness, loss of job, or loss of assets and livelihoods in a natural disaster, and thus may be insufficient to pro- vide income smoothing in these cases. The political economy argument that universality begets political support and increased budgets is reasonable, but not well supported in country programming (Desai and Kharas 2017). Practices can be improved in more tar- geted programs to reduce transaction costs and lower stigma. And finally, significant evidence shows that current social assistance programming has not reduced work effort. These emerging considerations point to the need for an organizing framework to guide decision-making processes. We propose a basic framework to clarify, locate, and assess the viability of a UBI (figure O.2). This is organized around four components. •• It is important to have a clear understanding of the performance of the current tax and transfer system in a given context. This can be challenging in settings with limited information, a nearly nonexistent tax base, and fragmented social pro- tection programs. •• The specific objective of the UBI among the many pursued, and design parame- ters devised accordingly. For example, if the objective is to counter the effects of automation-induced job losses, transfers should be provided for an amount ade- quate to ensure a minimum living standard. If the objective is to provide a social assistance function, transfers could be set in relation to poverty or food-insecu- rity prevention. Also, the way a UBI is introduced matters: the program could 6 Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov FIGURE O.2  Basic Framework for Navigating UBI Decision Making Initial UBI scope and design Assess options across metrics Final context context Select core UBI objective Coverage   (poverty reduction, social dividends, etc.) Delivery Incidence Define design parameters (transfer size, Weigh Performance Political frequency of Adequacy comparative of existing economy distribution, etc.) performance transfer and tax system at program Identify phase-in and system modality levels (substitution, top-up, Financing etc.) Incentives options Fiscal Comparator costs (subsidies, transfers, etc.) have radically different implications if it is substituting for select programs or provided on top of them. If the former, assessments should be made against those specific programs to be substituted. •• When these choices have been made, a UBI should be compared to the existing system in light of eight metrics. These metrics are coverage, level of progressivity, adequacy of transfers, household incentives and behavioral responses, costs, financing options, political economy, and delivery. No program would score opti- mally on all dimensions, nor utterly low on all of them. Clearly, societies may place a particular weight on some metrics as opposed to others; for example, some may favor coverage, others progressivity. Therefore, the art of decision making would hinge on an understanding of the trade-offs across the overall collection of implications that span between a UBI and the counterfactual (the existing system). •• The above considerations need to be weighted by policy makers. Such a process would involve a clear-eyed view on the scope, expected performance, and trade- offs involved. Importantly, the introduction of a UBI should be assessed not only against the possible interventions it replaces, but also at the system level—for example, how does a UBI affect the overall composition and outlook of the wider social assistance and social protection system. Our volume is designed to recognize and inform these trade-offs, with the chapter organization and content closely matching this framework. For instance, our simulations in chapter 4 offer an illustration of how to consider the various metrics of the frame- work presented in figure O.2. In particular, we compare the replacement of selected social assistance programs with a UBI. We simulate a full range of options in terms of Overview: Exploring Universal Basic Income 7 UBI generosity and financing for 10 low- and middle-income countries (Brazil, Chile, Haiti, India, Indonesia, Kazakhstan, Mozambique, Nepal, the Russian Federation, and South Africa) representing an array of contexts and diverse social protection systems. The microsimulations are based on recent representative household survey data and provide new insights into the trade-offs between coverage, poverty impact, transfer ade- quacy, and the budgetary implications of a UBI relative to the status quo. We begin with a budget-neutral scenario, whereby a UBI is simply replacing selected noncontributory social assistance programs. (In fact, we argue that a UBI should not be directly compared to or assessed as a replacement for pension insurance or other contributory programs.) We then gradually increase the generosity of the UBI transfer to the level of the full value of the poverty line, thus ensuring that, by design, poverty is eliminated. For these scenar- ios of increased generosity, we weigh financing options, contrasting increasing direct or indirect taxes combined with other fiscal policy options, such as elimination of subsidies or reallocation of public spending. Under a budget-neutral scenario, the poverty impact of targeted programs is higher than that of a UBI. With one exception (Russia), and even if imperfectly targeted, the poverty impacts of existing programs (measured in terms of the squared poverty gap, which better captures extreme poverty) are higher than the poverty impacts of a UBI (figure O.3). The difference in impact is small in absolute terms, but quite sizable in rela- tive terms. In fact, existing programs are on average about 60 percent more effective in poverty reduction than a UBI. This is because most existing programs, even if they may be only slightly progressive and miss some of the poor, tend to cover relatively more of the extremely poor population. Therefore, with a few exceptions, a budget-neutral UBI reform would take resources away from poor households that are benefiting from exist- ing programs, giving them to richer households currently not benefiting. Importantly, these findings do not account, or do so only indirectly, for other poverty-related aspects FIGURE O.3  Poverty Reduction Effects of a UBI and Baseline Cash-Based Programs Baseline Budget-neutral UBI reform –0.03 –0.01 –0.08 –0.07 –0.07 –0.07 –1.67 –1.39 –1.33 –0.71 –2.12 –0.99 –0.43 –0.19 –15.55 –12.68 –1.78 –1.14 –0.63 –0.69 0 –5 Percent –10 –15 –20 Haiti Mozam- Nepal India Indonesia Brazil Kazakh- South Chile Russian bique stan Africa Fed. Low income Lower-middle income Upper-middle income High income 8 Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov that may affect performance and are not easily observable from survey and administra- tive data—for example, transaction costs to access benefits, stigma, and leakages. A budget-neutral UBI reform leads to significant distributional impacts. While in some countries, differences in poverty impacts remain modest, on average a UBI reform would generate more winners than losers among the poorest segments of the popu- lation. On average, across our sample of 10 countries, 70 percent of the population in the two poorest deciles stands to gain from a budget-neutral UBI reform; this propor- tion increases to 92 percent moving toward the richest decile. However, across deciles, people losing from a budget-neutral UBI reform would lose substantially more than the winners would stand to gain. When measured as a percentage of each country’s average disposable income, the winners among the bottom deciles would gain about 1.7  per- cent, while 30 percent of the people would lose between 3.5 and 5.0 percent (figure O.4). Not surprisingly, when a UBI replaces regressive measures, it makes poor house- holds better off. This finding is intuitive: by being flat, a UBI would benefit those at the bottom of the distribution more than a regressive measure. The magnitude is demon- strated in the literature for energy subsidies in India (Coady and Prady 2018), and a simulated compensation for broadening regressive value-added (VAT) taxes in four Afri- can countries (Harris et al. 2018). A UBI would make virtually all households in the poorest 40 percent of the population better off (and would actually benefit most of those up to the 70th  percentile). Such a regressive-to-flat shift could establish the basis for further sequential recalibration of the distribution, including toward progressivity: the Islamic Republic of Iran, for instance, first replaced energy subsidies with a UBI, and then used affluence tests for excluding those at the top, thus putting the program on a more progressive path. The poverty effectiveness of a UBI can be enhanced by providing more generous transfers, but these can quickly become fiscally unsustainable, especially in low-income FIGURE O.4  Distributional Effects of a Budget-Neutral UBI Reform, Average across 10 Sampled Countries a. Winners as a percentage of each decile b. Gains/losses in average disposable income 100 5 80 0 60 Percent Percent –5 40 –1 0 20 –1 5 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 Overview: Exploring Universal Basic Income 9 settings. In Nepal and Mozambique, providing every citizen with a transfer equal to the average distance of the poor from the poverty line would cost 7 and 20 percent of gross domestic product (GDP), respectively; though in middle-income countries, the cost of this scenario would never exceed 8 percent of GDP. If a UBI is given for an amount suffi- cient to eradicate poverty, it would cost much more—between 36 and 48 percent of GDP in low-income Haiti, Mozambique, and Nepal; and between 8 and 22 percent of GDP in the other countries. Financing generous UBI transfers requires significant increases in taxation of the more affluent or complex public spending reforms (Ortiz et al. 2018). In most low- and middle-income countries, the richest deciles contribute overwhelmingly to overall tax revenues. Accordingly, differences in poverty impacts across taxation scenarios (direct versus indirect or lump sum) remain small overall, and the poverty impacts taking taxa- tion into consideration are only slightly smaller than the gross poverty impacts without considering the financing side. This is good news for a UBI, but mobilizing the needed resources is a challenge. Financing a UBI with meaningful poverty impacts may require a complex mix of sources. Conversion of subsidies is an option in some contexts; but with some exceptions, subsidy reforms will not cover the cost of meaningful UBI trans- fers—on top of being a formidable political economy challenge. Revenues from natural resources are a more promising fiscal outlet, but these are often highly volatile. In a con- text not included in simulations such as Alaska, for example, the annual change in UBI dividends could be up to 110 percent. Taxing the rich to finance a UBI with meaning- ful impacts on poverty would also require levels that are politically prohibitive in most countries. In India, for instance, direct taxes on the top decile would need to rise from 2.2 percent to 68.4 percent; in Brazil, from 7.2 percent to 24.5 percent; in South Africa, from 19.9 percent to 40.3 percent; and in Chile, from 5.4 percent to 38.4 percent. The only case in which additional taxation has more moderate impacts is Russia, where the incidence would need to increase from 9.0  percent to 13.2  percent. Financing a UBI with indirect taxes would put a lower—but still significant—burden on the top deciles, but would also add a heavy burden on the middle classes that pay consumption taxes. Employment-related incentives are another source of concern among policy makers. Recurrent concerns over the negative effects of a UBI on labor markets might be overstated. We review and frame global evidence on cash transfers and labor market outcomes. Clarifying this issue helps dispel misconceptions around work incentives, con- ditions of paid work and worker bargaining power, the valuation and distribution of unpaid work, and formal and informal employment. Because of the paucity of real- world experiences, we examine evidence from programs we consider informative, but that are not fully consistent with a UBI. While external validity for such considerations should be interpreted with caution, trends in evidence for large-scale programs are quite consistent and are likely to be relevant for a UBI should it be implemented. With regard to participation in paid work, fears are often exaggerated relative to existing evidence. Evaluations show that changes in livelihoods and labor market occupations occur, and that such changes per se should not be considered negative. In fact, labor market distor- tions remain relatively modest. And transfers may have positive effects on labor markets 10 Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov when recipients use them to invest in family livelihoods or in their children’s human capital. Possible inflationary risks should not be dismissed a priori, nor should they be overly magnified. Very often, UBI supporters point out that cash would lead to increased competition among market actors, thus reducing prices (perhaps with only short-term price adjustments). In other words, it is posited that suppliers of goods and services will efficiently respond to the additional, cash-induced, effective demand. The available esti- mates on multipliers, which range from 0.80 to 2.52 for every dollar injected, provides some supportive evidence in that direction. However, the experience of countries such as Australia, Kuwait, Mexico, and the Philippines present contrasting effects. We argue that context matters, and that inflationary effects should be assessed within the frame- work of analytical parameters such as overall market competitiveness and conditions (e.g., a significant injection of cash in weakly integrated markets may cause inflation), the specific market for subproducts and services, program size, and probably interven- tion duration. The political economy of UBI remains vastly underexplored. There are several cru- cial political economy threads in a UBI—for example, in support for current systems, in how to replace a portion of current programs, and in resource mobilization. All of these present a large number of stakeholders with differing interests and incentives. The pace of possible introduction also matters. For example, if countries choose to expand cate- gorical transfers (e.g., universal child grants), these could provide an area to inform a number of UBI-type questions. In the medium term, these quasi-UBI programs may help in better understanding the effects of bounded universality (including its financing) and help build more inclusive delivery platforms—all the while unlocking the potential for higher coverage. The poverty effectiveness of categorical programs would depend on whether and how much age characteristics correlate with poverty, although they will be significantly more expensive than poverty-targeted programs. A gradual adoption of a UBI does not, however, eliminate core political economy challenges. For instance, piece- meal introduction may worsen path dependency challenges. Groups that are likely to gain from the first forms of a UBI may see its further expansion as a threat and block it. Experimentation trajectories are fraught with various risks. They often reflect a political bandwagon effect—that is, expressions of “cheap” support across the political spectrum with low actual commitment to subsequent larger-scale implementation. The UBI’s overall design features suggest that it may fit certain configurations of societal welfare attitudes and preferences more than others. In many cases, program design may reflect historical, philosophical, and moral norms around if and how indi- viduals are expected to reciprocate public assistance. In some societies, for instance, the concept of work constitutes the primary lens through which the exchange of individual rights and responsibilities is interpreted—and that is reflected in the choice and design of social assistance interventions. The U.S. safety net, for instance, is overwhelmingly in kind, focused on families with children and on work (Hoynes and Rothstein 2019). At the opposite end of the income spectrum, Ethiopia places a strong emphasis on work as well, with its safety net anchored in public works. In other contexts, societal preferences Overview: Exploring Universal Basic Income 11 might differ—for example, work may share primacy with other values, such as family time or community care, which may lessen expectations of reciprocity via work. In Africa, studies have shown that the public acceptability of a universal program hinges on how well it aligns with prevailing notions of deservingness (Davis et al. 2016). When operating at full scale, implementation of a UBI might be relatively simple and streamlined—but getting to that point is easier said than done. We identify core delivery elements and processes that serve a mainstream social assistance program, and illustrate how these elements and processes should be adapted to operationalize a UBI. We discuss pragmatic issues around eligibility, outreach, registration, payments, grievance redress, and other program-level mechanisms as well as overarching or foun- dational issues related to identification, interoperability, and data protection. When viewed through an implementation lens, a UBI is based on the same elements as those supporting the delivery of other social protection programs. A UBI may offer some sim- plifications that would enable extension of coverage (broader awareness of the program, reduced beneficiary transaction costs, and no complex tests for eligibility and targeting processes). But challenges in covering the poor go beyond targeting and encompass a range of practical bottlenecks across the delivery chain—which by itself, a UBI cannot overcome (Lindert, George, and Rodriguez-Caillava, forthcoming). Moreover, working at a universal scale entails challenges of its own. Very few, if any, low- and middle-income countries may be ready to have a UBI in its full version implemented in the short term. Countries that are the closest to the feasibility frontier are those that may need a UBI the least (assuming that coverage of the poor, which is already very high, is the primary goal of a UBI introduction). So where would a UBI be more or less likely to be an appropriate option? Our anal- ysis, based on both generation of new results and extensive review of the theoretical and operational literature, points to some stylized implications for different contexts. These could be summarized as follows: •• Where social assistance provides relatively adequate benefits, substantial cov- erage, and slight to marked progressivity, policy makers could consider tackling specific bottlenecks that hamper eligibility, access, coverage, or delivery within the existing system. If a UBI is to be considered, it may have to be motivated by objectives other than a poverty-related one (e.g., automation-driven job insecu- rity, social dividends, etc.). •• Where coverage is high, but not progressive, a UBI could be considered an option, although some vulnerable (age) groups may suffer from the shift. •• Where social assistance is limited but provided progressively, a UBI would extend coverage but also flatten the distribution. If budget neutral, this means “less money for more people,” and likely “less at the bottom.” •• Where social assistance is patchy and flat or regressive, a UBI could be an option to expand coverage if financed via progressive income taxation, elimination of energy subsidies, or redistribution of windfall revenues. Most low-income 12 Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov countries may not display those financing features; but some middle-income, resource-rich countries may do so. •• For a typical low-income setting, a UBI could expand coverage, but is clearly financially daunting. Other factors, such as diversity in contexts at the subna- tional level (e.g., remote areas with little connection to markets, etc.), may also suggest the need for design flexibility (e.g., a balance of in-kind and cash trans- fers, sensible ways to account for children, etc.), thus possibly making the rigid design of a UBI less palatable. The book is structured around seven chapters. •• In chapter 1, Gentilini, Grosh, and Yemtsov clarify the definition of UBI, offer an overview of the design choices underpinning it, and discuss the correspond- ing evidence base. The discussion is then extended to key thematic areas that trigger interest in a UBI, including changes in labor markets, social protection reform, governance of natural resource wealth, and the rights agenda. Finally, the chapter reviews lessons stemming from practical experiences, including pilot trials and larger-scale schemes. •• In chapter 2, Gentilini and Grosh put UBI in perspective by comparing it to other social assistance interventions. A UBI is often confused with other measures such as a guaranteed minimum income and a negative income tax. In Italy, for instance, at the time of writing this report, the citizens income program is being presented as a UBI while actually being a slightly expanded form of a guar- anteed minimum income. Similar considerations stem from the United States, Finland, and elsewhere. The chapter thus clarifies the analytical and practical differences between various options, including benefit and tax-based measures, and both wage- and nonwage-oriented schemes. The chapter compares and reviews benefit structures and succinctly identifies comparative advantages and limitations. It thereby provides thumbnail sketches of other program options against which a UBI could be selected. •• Following the analytical foundations laid out in the first two chapters, chap- ter 3, authored by Bastagli, examines one of the most contentious quandaries and concerns among policy makers: the interface between UBI and work. The chapter critically reviews and discusses the arguments and evidence on the links between UBI and four work-related outcomes: participation and hours in paid work; the conditions of paid work (e.g., whether a UBI would strengthen bargaining power); the recognition, valuation, and distribution of unpaid work (could a UBI be considered a way of remunerating unpaid work); and formal and informal employment. •• Chapter 4 generates new analysis and insights from microsimulations. Rigo- lini, Lustig, Gentilini, Monsalve, and Quan provide evidence on the impacts, Overview: Exploring Universal Basic Income 13 costs, and distributional implications of the UBI based on simulations. They look at the poverty and inequality impact of social protection systems when income support programs are replaced with UBI schemes of various levels of gener- osity. They do so using household survey data for 10 low- and middle-income countries and provide a nuanced explanation of the factors shaping program performance. They study the spending and financing sides of UBI for six mid- dle-income countries using taxation data from Tulane University’s Commitment to Equity Data Center. •• The last three chapters (chapters 5, 6, and 7) examine real-world financial, political, and operational issues. Decisions about a UBI should be taken in con- junction with decisions about its financing. So in chapter 5, Ter-Minassian lays out alternative financing options for a UBI. The chapter provides practical considerations—a primer—to assess the fiscal space and revenue mobilization measures to finance different levels of a UBI. •• The political economy of the UBI is discussed in chapter 6. De Wispelaere and Yemtsov provide an overview of theories and experiences with political economy reforms around the idea of UBI, drawing from existing literature and initiatives that are receiving considerable public and analytical attention. The chapter specifically examines experiences and issues around political con- stituencies and coalitions that can affect whether and how a UBI might be a politically viable option. •• Chapter 7 provides a framework for thinking about how to implement a UBI, including core requirements across a stylized delivery chain. This chapter, by Lowe, George, Grosh, and Gentilini, identifies a number of functions and activ- ities that would serve a mainstream social assistance program, and illustrates how those processes should be adapted to operationalize a UBI in practice. The chapter discusses pragmatic issues around eligibility, outreach, registration, payments, grievance redress, and other programmatic mechanisms, as well as overarching and foundational issues related to identification, interoperability, and data protection. The chapters are complemented by five appendixes. These are an inventory of existing or past UBI program design features (appendix A), a structured compilation of UBI-related proposals (appendix B), a granular mapping of impacts from design choices related to conditionality (appendix C), a technical discussion of the data and methodol- ogy for the microsimulations in chapter 4 (appendix D), and an annotated bibliography (appendix E) dovetailing with the chapter-specific references. 14 Ugo Gentilini, Margaret Grosh, Jamele Rigolini, and Ruslan Yemtsov References Alderman, Harold, Ugo Gentilini, and Ruslan Yemtsov, eds. 2018. The 1.5 Billion People Question: Food, Vouchers or Cash Transfers? Washington, DC: World Bank. http://documents.worldbank. org/curated/en/398281507803030509/pdf/120368-PUB-PUBLIC-ADD-BOX-405295B- DOCDATE-11-11-2017.pdf Banerjee, Abhijit, Paul Niehaus, and Tavneet Suri. 2019. “Universal Basic Income in the Developing World.” NBER Working Paper 25598. National Bureau of Economic Research, Cambridge, MA. Calnitsky, David. 2017. “Debating Basic Income.” Catalyst 1 (3): 63–90. Coady, David, and Delphine Prady. 2018. “Universal Basic Income in Developing Countries: Issues, Options and an Illustration for India.” IMF Working Paper 18/174. International Monetary Fund, Washington, DC. https://www.imf.org/en/Publications/WP/Issues/2018/07/31/Universal-Basic- Income-in-Developing-Countries-Issues-Options-and-Illustration-for-India-46079 Davis, Benjamin, Sudhanshu Handa, Nicola Hypher, Natalia Winder Rossi, Paul Winters, and Jen- nifer Yablonski, eds. 2016. From Evidence to Action: The Story of Cash Transfers and Impact Evaluation in Sub-Saharan Africa. Oxford, UK: Oxford University Press. Desai, Ray M., and Homi Kharas. 2017. “Is a Growing Middle Class Good for the Poor? Social Policy in a Time of Globalization.” Global Economy and Development Working Paper 105. Brookings Institution, Washington, DC. https://www.brookings.edu/research/is-a-growing-middle-class- good-for-the-poor-social-policy-in-a-time-of-globalization/ Devarajan, Shantayanan. 2018 “How to Use Oil Revenues Efficiently.” Working Paper 1199. Eco- nomic Research Forum, Giza. http://erf.org.eg/wp-content/uploads/2018/05/1199_Final.pdf Drèze, Jean. 2017. “Decoding Universal Basic Income.” Indian Journal of Human Development 11 (2): 163–66. Felman, Josh, Boban Paul, M.R. Sharan, and Arvind Subramanian. 2019. “Quasi-Universal Basic Rural Income: The Way Forward.” Business Standard January 27. Francese, Maura, and Delphine Prady. 2018. “Universal Basic Income: Debate and Impact Assess- ment.” Working Paper WP/18/273. Fiscal Affairs Department, International Monetary Fund, Washington, DC. Friedman, Milton. 1967. “The Case for a Negative Income Tax.” National Review 7 (March): 239–41. Gentilini, Ugo, Margaret Grosh, and Michal Rutkowski. 2019. “Understanding Coverage: What Does Universal Social Protection Really Mean?” Blog post January 28. OECD Development Matters. https://oecd-development-matters.org/2019/01/28/understanding-coverage-what- does-universal-social-protection-really-mean/ Hanna, Rema, and Benjamin A. Olken. 2018. “Universal Basic Incomes versus Targeted Transfers: Anti-Poverty Programs in Developing Countries.” Journal of Economic Perspectives 32 (4): 201– 26. https://economics.mit.edu/files/15434 Harris, Tom, David Phillips, Ross Warwick, Maya Goldman, Jon Jellema, Karolina Goraus, and Gabriela Inchauste. 2018. “Redistribution via VAT and Cash Transfers: An Assessment in Four Low and Middle Income Countries.” IFS Working Paper W18/11. Institute for Fiscal Studies, London. https://www.ifs.org.uk/uploads/WP201811.pdf Hoynes, Hilary W., and Jesse Rothstein. 2019. “Universal Basic Income in the US and Advanced Countries.” NBER Working Paper 25538. National Bureau of Economic Research, Cambridge, MA. https://gspp.berkeley.edu/assets/uploads/research/pdf/Hoynes-Rothstein-UBI-081518.pdf Lindert, Kathy, Tina George, and Inés Rodriguez-Caillava. Forthcoming. A Sourcebook on the Foun- dations of Social Protection Delivery Systems. Washington, DC: World Bank. Overview: Exploring Universal Basic Income 15 Lowrey, Annie. 2018. Give People Money: How a Universal Basic Income Would End Poverty, Revolu- tionize Work, and Remake the World. New York: Crown. Marinescu, Ioana. 2018. “No Strings Attached: The Behavioral Effects of U.S. Unconditional Cash Transfer Programs.” NBER Working Paper 24337. National Bureau of Economic Research, Cambridge, MA. Ortiz, Isabel, Christina Behrendt, Andrés Acuña-Ulate, and Quynh Anh Nguyen. 2018. “Univer- sal Basic Income Proposals in Light of ILO Standards: Key Issues and Global Costing.” ESS Working Paper 62. International Labour Office, Geneva. https://www.ilo.org/wcmsp5/groups/ public/---ed_protect/---soc_sec/documents/publication/wcms_648602.pdf Packard, Truman, Ugo Gentilini, Margaret Grosh, Philip O’Keefe, Robert Palacios, David Robalino, and Indhira Santos. 2019. Protecting All: Risk Sharing in a Diverse and Diversifying World of Work. Washington, DC: World Bank. https://hubs.worldbank.org/docs/imagebank/pages/ docprofile.aspx?nodeid=31393804 Piketty, Thomas. 2016. “What Unequal Societies Need Is Not a ‘Basic Income’ But a Fair Wage.” The Wire. https://thewire.in/uncategorised/basic-income-fair-wage-piketty Ravallion, Martin. 2018. “Guaranteed Employment or Guaranteed Income?” CGD Working Paper 482. Center for Global Development, Washington, DC. Stern, Andy. 2016. Raising the Floor: How a Universal Basic Income Can Renew Our Economy and Rebuild the American Dream. New York: Public Affairs. Yang, Andrew. 2018. The War on Normal People: The Truth about America’s Disappearing Jobs and Why Universal Basic Income Is Our Future. New York: Hachette Books. The Idea of Universal Basic Income CHAPTER 1 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov M artin Luther King Jr set forth an idea to advance inclusive economic opportu- nity. It was an idea that fascinated Keynesians like James Tobin and libertarians like Milton Friedman. It stirred controversy in the 2017 French elections and is currently piloted in rural Kenya. It is one that charms and polarizes in equal measure, and its revolutionary simplicity is summed up in three words: cash for everyone. It is no surprise that the idea of a universal basic income (UBI) has elicited such waves of interest across time and space. In part this is precisely because it is an idea, and not merely a program. As Lowrey (2018, 191) recently put it, “…a UBI is a lesson and an ideal, not just an economic policy.” Indeed, discussions on UBI are often a proxy for broader debates around the role of the state and markets, and the distribution of power within societies: “[UBI] is not just a form of redistribution; it is a moral statement” notes Sir Tony Atkinson (Atkinson 2011, 4). At the time of writing this volume, there were 126 books on UBI, 91 of which had been produced since 2010 alone—that is, an average of one book released each month. Similar statistics emerge in newspaper quotes (Hoynes and Rothstein 2019). At least 36 policy proposals have been tabled, and 22 pilot programs fielded; and our bibliog- raphy (appendix E) lists 200 titles of direct and indirect relevance to the topic. Since 1986, a Basic Income Earth Network has brought together a rich cadre of policy makers, practitioners, and academics. In 2016, Switzerland held a referendum on adopting a UBI. Documentaries are springing up,1 and some politicians have made UBI a central platform of their campaigns.2 The United Nations Secretary-General recently called for “stronger safety nets, and eventually universal basic income” (Guterres 2018), while in India it was predicted that “…within the next two years, at least one or two [Indian] 17 18 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov states will implement universal basic income” (Indian Economic Times 2018). The inevi- tability of UBI is almost a mantra in Silicon Valley.3 But what is a UBI concretely? What parameters should a policy maker consider to com- prehend and ponder its implications? What compelling lessons are emerging from practical experiences? A more granular examination of the literature shows that the definition, ana- lytical foundations, expectations, and practical insights on UBI can all vary significantly (Bregman 2017; Calnitsky 2017; Murray 2016; Standing 2017; Stern 2016; Van Parijs and Vanderborght 2017; Widerquist et al. 2013). Such heterogeneity may help in building initial momentum and generate public interest, but it could undermine a coherent and structured policy discussion. This chapter unbundles and clarifies core layers of debates surrounding the UBI concept with the aim of providing a compass to navigate the idea. After a short historical excursus on the origins of today’s UBI debates, we lay out the main design contours of a UBI and trace key empirical debates underlying the choice of practical parameters. Universality, conditionality, transfer modality—as well as a set of six other choices—are unbundled and examined in the next section. The broad ecosystem of UBI involves different stylized narratives, which are the subject of the fourth section of this chapter. In particular, UBI is interpreted in the context of social assistance reforms, rights- based approaches, automation and labor market disruptions, and resource dividends and state-citizen accountability. Such narratives are complemented with emerging lessons from country and subnational experiences. The chapter’s final section offers a set of con- cluding reflections on the analytics and practice of UBI as a social protection instrument. A Glimpse at History Contemporary social protection systems are the result of century-long experiences on how states manage risks in the context of evolving economic, social, institutional, and technological forces. So why has UBI increasingly emerged as a way—or a strawman, according to some—to challenge and rethink such systems? In this section, we step back for a moment and consider four main phases in the evolution of social protection (Hickel 2017; Lindert 2004; Ravallion 2016; Smith 2011). The first phase, which runs up to around 1600, is characterized by a dearth of public support against destitution. The seminal work by Vives (1526), through his advice to the city of Bruges, constitutes one of the first contributions in outlining a clear ratio- nale for poverty-related transfers by the state. Societies were largely rural, workers were mostly self-employed, and agricultural production was generally organized along feudal lines. Households managed life-cycle and other risks through informal arrangements, community sharing, or similar strategies. This order was disrupted by the ‘enclosure’ process in England (or the privatization of common land) which, while increasing agri- cultural productivity, gave birth to ‘poverty’ as societal phenomenon (Hickel 2017). Vives’s thinking influenced the first British Old Poor Law of 1601, which initiated a second phase for social protection. Adopted after periods of riots and famines, the Old Poor Law formalized the provision of limited public transfers for specific shocks, such as Chapter 1. The Idea of Universal Basic Income 19 old age, widowhood, disability, illness, or unemployment. The Old Poor Law was legally enforceable, financed by local taxes on landlords, and provided a minimum guarantee that, while modest, was open to anyone in need. By the late 17th century, almost all par- ishes of England were covered by the Old Poor Law.4 The advent of the Industrial Revolution fundamentally reshaped how people lived. The emergence of landless populations, large-scale migration to cities, and heavy dependence on wage employment marked a structural shift that would have profound implications for social protection. The masses flocking to cities left parishes to finance rising support bills for children and the elderly. This led, by 1818, to a sixfold increase in the tax rate to finance the Old Poor Law compared to the mid-1700s. Strong backlash from landlords against the Old Poor Law ensued, with stricter measures such as the dis- tinction between “deserving” and “undeserving” poor emerging around this time. Significant reforms were enacted with a New Poor Law of 1834. As a result, spend- ing for poor relief was slashed from 2.5 percent to 1.0 percent of gross domestic product (GDP). This was compounded by a wider use of workhouses, which provided meager payments in kind and harsh working conditions. The thinking behind the Old and New Poor Laws spread from the United Kingdom to its overseas territories and colonies, influ- encing approaches to social protection in the United States, India, and parts of Africa (Bhattacharya 2017; Harvey 2007; Mkandawire 2016; Seekings 2013).5 It was only a cen- tury later that, with the Beveridge Report of 1942, workhouses were closed. By the end of the 19th century, the industrialization process had remodeled the societal fabric so profoundly that new ways of sharing risks were necessary. Contributory insurance schemes appeared around this time, marking the beginning of a third phase in social protection history. Although there were some antecedents, pensions can be traced back to the German Bismarckian model of 1889. This was characterized by a financing structure based primarily on employer and employee (and sometimes government) con- tributions and benefits proportional to the covered worker’s salary. Coverage gradually increased as a greater share of the workforce moved into facto- ries and firms, and labor markets formalized—that is, workers and their employers were registered and monitored to comply with various regulations, including social insurance contributions from payroll, income, and corporate taxes. Various factors, such as colo- nial approaches to welfare and pervasive informality, have prevented such risk-sharing arrangements from being firmly embedded in most low- and middle-income coun- tries (Packard et al. 2019; Seekings 2013). Currently, social insurance coverage is low across most low- and middle-income countries. In Africa, on average, 10.6 percent of the working-age population contributes to pension schemes (Guven 2019). Similarly, in Bangladesh, India, Indonesia, Nigeria, and Pakistan, which have a combined population of over 2.1 billion people, social insurance coverage is below or around 10 percent of the economically active population, with generally limited progress registered over the past decade6 (Rutkowski 2018). The global South witnessed a fourth phase of social protection—an explosion of noncontributory social assistance (and social pension) programs, which are now nearly ubiquitous worldwide (World Bank 2018a). In regions such as Sub-Saharan Africa, the 20 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov institutionalization of social assistance transfers is a relatively new phenomenon taking root over the past quarter century. This was preceded by large-scale price subsidy schemes and, in the post-colonial period, international humanitarian assistance (Ben- nett, Foley, and Krebs 2016; Devereux 2001). Today social protection systems are composed of a varied, but often complex, mix of programs. In some countries, the role of social protection was largely driven by larger economic transformations, such as with the reform of state-owned enterprises in China over the 1990s (Ang 2016; Gentilini 2015). Whether driven mostly by economic or other forces, every country has a set of contributory social insurance programs and further worker protections for those with formal sector labor contracts. In some coun- tries, these provide risk management to a large part of the population; in other countries, informality dominates and, as mentioned, social insurance provides protection to only a few, though sometimes generously. Social assistance programs are usually small relative to needs, programs to help households improve their jobs or earnings still smaller, and social services even more limited. These latter three strands of social protection are often characterized by multiple programs offered by different government agencies and/or for different client segments. There may also be significant expenditures on food, energy, fertilizer, or water subsidies that share some goals with social protection. How do these four phases relate to current debates on UBI? One common thread across the centuries is that social protection is a matter of struggle and hard-won gains by coalitions of poor, working, and middle-class populations (Desai and Kharas 2017). Eventually, social protection systems adapt to contemporary challenges, but rarely with- out major crises, societal battles, or both. Institutions and polities take time to adjust to a fast-paced, evolving society. Part of the narrative on UBI is that social protection sys- tems, especially in higher-income countries, are being outpaced by structural shifts in demographics, employment, and culture. In other words, countries have a 20th century system to deal with 21st century challenges. The appetite for change is often palpable. Reimagining social protection is an ongoing process, with hot debates around the direction and modalities for change (Barr, forthcoming; Cottam 2019; Ortiz 2018; Rutkowski 2018; Shafik 2018). But virtually every study underscores the importance of social assistance as a foundational platform for the social protection system of the future. Would a UBI be part of a fifth social protection phase? Before answering this question, we need to understand UBI better. Anatomy of UBI In principle, social assistance schemes can be codified along three features or dimen- sions: what transfer modality they provide, whether and how they are conditional, and whether and how they are targeted. The cube proposed in figure 1.1 illustrates the pro- grams that ensue from these combinations. A UBI, highlighted in gold, is the combination of three choices—that is, a transfer that is provided universally, unconditionally, and in cash. Within this framework, UBI Chapter 1. The Idea of Universal Basic Income 21 FIGURE 1.1  UBI within a Social Assistance Cube Public Distribu- Productive Safety tion System in Net Programme in Supplemental Nutrition India Assistance Program Ethiopia (SNAP food stamps) in National the United States Rural Em- ployment Guarantee Scheme in India Dibao in School China feeding Guaranteed Conditional Public Needs minimum cash works based income transfers Targeting Social Categorical pensions Universal In-kind Universal basic Job ity income guarantee Vouchers al programs od m Cash er Unconditional Conditional Conditional sf an (services) (work) Tr Conditionality proposals can still differ on a range of important parameters such as transfer level and frequency, age of eligibility, and whether citizens or all residents are covered. Outlining the three overarching design parameters provides a framework for assess- ing whether current and past experiences would classify as UBI programs. As mentioned in the previous section, a range of pilot trials and broader schemes are or have been under way (Banerjee, Niehaus, and Suri 2019). But can they be considered a UBI? Table 1.1 illustrates the characteristics of various initiatives relative to UBI typical traits. In addition to the three core variables outlined above—conditionality, transfer modality, and universality—we here examine whether they are national or local in scope. As chapter 6 further illustrates, the political economy of reform is a key issue in shaping UBI debates within social contracts. Many pilots are currently privately funded as a way to help jumpstart discussion and advance an evaluation agenda; in doing so, however, they may largely bypass the political, budgetary, and inter-institutional struggles that are likely to impinge on a full-scale UBI policy attempt (Gentilini 2019). 22 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov TABLE 1.1  Which Initiative Is Currently a Pure UBI? Initiative Uncon- Cash- Uni- State- (year) ditional based versal provided Scope Frequency/size Coverage Full-scale program Mongolia Yes Yes Yes Yes National Tog 10,000 (US$7)/month 3 million (2010–12) 2010; Tog 21,000 (US$17)/ month until 2012 Iran, Islamic Yes Yes Yes Yes National Rls 445,000 (US$40– 97% of Rep. (2011) US$45)/person/month population (25% of median income) Variants United States Yes Yes Yes Yes State US$1,000–US$2,000/year 615,000 (Alaska) United States Yes Yes Yes Yes Tribe US$4,000–US$6,000/year 16,000 (Eastern Band (disbursements made of the Chero- every 6 months) kee Nation) Kuwait (Amiri Yes Yes Yes Yes National US$3,600/one-off 1.1 million grant) Italy Yes Yes National €780/month 5 million (Reddito di Cittadinanza) China Yes Yes Yes Region Variable annual ~707,000 (Macau SAR) (resident and payments; in 2019, nonresident P 10,000 for residents holders P 6,000 for nonresidents of Macau resident identity cards) India Yes Yes Yes State Rs 5,000/acre biannually 5.8 million (Telangana) (land-holding (Rs 10,000/year) farmers) India (Odisha) Yes Yes State 5 installments of Rs 5,000 7.5 million (small and (Rs 25,000/year) for marginal small and marginal farmers, farmers; 3 installments of landless Rs 5,000, Rs 3,000, and workers, and Rs 4,500 (Rs 12,500/year) vulnerable for landless workers; agricultural Rs 10,000/year for households) vulnerable agricultural households Pilots Kenya Yes Yes Yes Villages Long-term UBI: monthly ~21,000 (GiveDirectly) payments equivalent to US$23 (US$0.75/day) for 12 years Short-term UBI: monthly payments equivalent to US$23 (US$0.75/day) for two years Lump-sum UBI: US$500/ one-off United States Yes Yes Yes Households Variable guarantee levels 9,924 1970s (Indiana, and marginal tax rates (initial Iowa, New Jer- target) sey, North Car- olina, Seattle/ Denver) (continued) Chapter 1. The Idea of Universal Basic Income 23 TABLE 1.1  Which Initiative Is Currently a Pure UBI? (continued) Initiative Uncon- Cash- Uni- State- (year) ditional based versal provided Scope Frequency/size Coverage Canada Yes Yes Yes Households Variable guarantee levels 1,300 (Manitoba) and marginal tax rates: Can$3,800, 0.35 Can$4,800, 0.50 Can$5,800, 0.75 India Yes Yes Yes Individuals Adults: Rs 200/month ~6,000 (Madhya (later raised to Rs 300); Pradesh) children: Rs 100 (later raised to Rs 150) India Yes Yes Households Rs 1,000/month 100 (New Delhi) Namibia Yes Yes Yes Individuals US$100/month 930 (Otjivero- Omitara) Finland (Kela) Yes Yes Yes Unemployed €560/month 2,000 United States Yes Yes Households US$1,500/month 100 (Oakland, CA) United States Yes Yes Individuals US$500/month 130 (Stockton, CA) Netherlands Yes Yes Individuals €960/month 250 Korea, Rep. Yes Yes 24-year-olds US$883/year 170,000 (Gyeonggy) Spain Yes Households €100–€1,676 (US$110– 1,000 (Barcelona) US$1,850)/month The table shows that, if we consider all those criteria, only Mongolia and the Islamic Republic of Iran had a national UBI scheme in place; these experiences are discussed later in this chapter.7 The bulk of smaller-scale pilot experiences revolves around four experiments in the United States, which also pioneered the use of randomized controlled trials in the social sciences (Moffitt 2002, 2003). These were mostly negative income tax experiments devised to inform President Nixon’s Family Assistance Plan (box 1.1), the literature on which is extensive (Steensland 2007).8 In Kenya, a pilot is currently under way covering 6,000 people for 12 years, with a second treatment arm of 11,500 people for 2 years. Pilots in India took place in eight villages in Madhya Pradesh (Rs 300/adult and Rs 150/child), with a similar scheme in a tribal village. Finland is undertaking a randomized controlled trial providing 2,000 unemployed citizens with nearly US$600/ month over two years;9 Oakland, California, and 25 municipalities in the Netherlands are about to start similar trial programs. Finally, in Namibia, a pilot UBI was conducted in the Otjivero-Omitara area from 2007 to 2009, including the provision of about US$100 monthly to around 930 residents below the age of 60. This was financed by private con- tributions and implemented by nongovernmental organizations. Descriptive statistics for these pilots as well as for other initiatives are laid out in appendix A.10 Let us now turn our attention to the core design features of a UBI. We start with uni- versality, followed by conditionality, and transfer modality. The section concludes with 24 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov BOX 1.1  Negative Income Tax Pilots T he first negative income tax pilot in New Jersey and Pennsylvania lasted from 1968 until 1972 and had a sample size of 1,357 households from declining urban areas. A rural experiment conducted in Iowa and North Carolina from 1969 to 1973 included 809 low-income rural families. The third pilot, which took place in Gary, Indiana, between 1971 and 1974, was composed of 1,780 African American house- holds, 59 percent of which were headed by single females. The largest experiment, which included 4,800 families, was conducted in Seattle and Denver from 1971 to 1982. This trial not only offered recipients more generous benefit structures than the others, but also extended the duration from three to five years for a quarter of participants. The U.S. experience is mirrored by Canada’s Mincome scheme in Manitoba: running over 1975–79, it covered 1,300 households in the cities of Win- nipeg and Dauphin (Forget 2011, 2018; Munnell 1986; Pechman and Timpane 1975). a consideration of choices around other parameters important to the shape and fit of a particular UBI proposal into larger policy, but whose variations do not violate the defini- tion of a UBI. Universality The notion that social protection is universal rests on two elements: namely that every- one is covered. In many cases, debate revolves around the “everyone” aspect—that is, the rationale and modalities to cover all members of society and not just some. Yet, this assumes clarity on the meaning of “coverage,” which is a big assumption. In health insurance, for example, the goal is often to provide coverage to all, so that in the event people become ill, they receive health services. The same principle is at play for crop insurance. And for contributory pensions, unemployment, or disability insurance programs, coverage is used in an analogous way. In most periods, people covered by such insurance will benefit from a guarantee or a promise of help when needed (i.e., reaching a certain age), but not necessarily from a payout (Gentilini, Grosh, and Rutkowski 2019). For social assistance, instead, coverage is often interpreted as receiving an actual transfer. This is quite a difference and a critical issue to clarify given the implications for universal social protection. For instance, if a country has a guaranteed minimum income program that provides cash when incomes fall below a threshold, the social insurance interpretation would be that—as in the case of health or pensions—everyone is covered independent of the event occurring (i.e., income falling). Thus, coverage would be many times greater than the actual benefit roster. A guaranteed minimum income is universal in insurance terms, but it is targeted from a social assistance standpoint. Part of the social assistance community refers to programs for the elderly or chil- dren as universal. But what is really meant is that eligibility for such programs entails no Chapter 1. The Idea of Universal Basic Income 25 requirement other than age—those who do not meet the requirement are excluded— hence it is not for all. A “universal social pension,” for example, is again universal from an insurance perspective, but it is targeted from a social assistance viewpoint. There is also the question of whether coverage refers to gross or net benefit. While with a UBI everyone receives a payout, those benefits must be financed. A UBI’s flat and uniform character is necessarily altered once the financing is taken into account. When financing is via progressive income taxation, for example, some people will pay more than the benefits they receive as a UBI payout. Thus, some people are nominally beneficiaries, but de facto financiers. The program therefore ceases being universal in practice. Another important aspect of coverage is what benefit the coverage supplies. In social assistance, it is generally a supplement to income, often a rather small one. But the insurances are dimensioned in a way that relates to the size of the loss. Health insurance provides benefits that are differentiated according to the problem. Unemployment insur- ance pays more to replace wages for high-income workers than for lower-wage workers. It is meant to pay a lot during the (hopefully) short period of unemployment. Providing coverage with a minor social assistance or UBI benefit would not match the payout the same way insurance is designed to do. UBI thus represents a shortcut in achieving universality. Yet there are different ways of conceiving and defining universality in coverage. Universality should be consid- ered at the system level: universality in social protection, which lies at the core of global commitments and the rights architecture, does not necessarily imply universality via a single program. Whether through a UBI or social protection more broadly, universal- ity would need to be progressive, and ensure that the most in need receive support to meet their wider range of vulnerabilities and necessities. A gradual building of a solid platform of social assistance, whether via one program or not, should proceed from the bottom up. Pros and Cons of Universality The rationale for making transfers universal rests on five main arguments. First, by not establishing eligibility criteria (besides perhaps citizenship or established residency and age, e.g., for those above age 18), universality circumvents the contentious issue of exclusion and inclusion errors that are inherent in needs-based targeting. Under a UBI, there would be no such errors, as everybody is included by design, hence achieving sub- stantial expansions in coverage (Ravallion 2018). Second, universality may eliminate any stigma affecting beneficiaries. Evidence from Europe, for example, shows that shame is among the key factors behind limited take-up of benefits by eligible beneficiaries (Atkin- son 2015; Eurofound 2015). Third, by changing the default position of people from being potential benefi- ciaries (subject to eligibility verification) to guaranteed recipients, there may be fewer transaction costs involved in accessing benefits (e.g., there is no need to spend time in applying).11 Also, there are various economic and psychological benefits stemming 26 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov from a stable source of income over time (e.g., stress reduction, empowerment, avoid- ing taking desperate actions out of economic hardship) (Mullainathan and Shafir 2013; World Bank 2015). Fourth, a universal transfer would be more labor compatible than most programs, as it removes the price effect of transfers (i.e., the reduction in labor supply to avoid a reduction in benefits). And finally, universality may strengthen pro- grams’ political sustainability as beneficiaries (and voters) would draw from the entire income distribution. By being universal, UBI (and to some extent categorical programs such as child allowances and social pensions) may draw together the interests of the poor, the near poor, and even the middle class. Hence, the political economy of policy choice and of taxation to support the programs would change to enlarge the resource pool for social protection enough to expand coverage and maintain meaningful bene- fits12 (Desai and Kharas 2017; Kidd 2015). The case against the “U” in UBI rests principally on cost, fit for purpose, and a different appreciation for the magnitude of the possible benefits of UBI. First, the cost of making significant transfers universal is quite high. Depending on how these are financed—a reduction in existing social protection spending, a reduction in regressive subsidies, increased taxes—there are important changes in distributional outcomes among income and age groups that may or may not be desirable. We will come back to this point when discussing figure 1.2 as well as in chapter 4. Second, the flat benefit structure may not be fit for all purposes. It cannot be as redistributive as a more pro- gressive structure and thus will have muted impacts on poverty and inequality. The flat structure does not respond to large and often short-run changes of state such as cata- strophic illness, loss of job, and loss of assets and livelihoods in a natural disaster, and hence may be insufficient to provide income smoothing in these cases. The political economy argument that universality begets political support and increased budgets is not well supported in country programming. Practices can be improved in more targeted programs to reduce transaction costs and lower stigma. And finally, significant evidence shows that current social assistance programming has not reduced work effort (Baird, McKenzie, and Özler 2018; Handa et al. 2018). Importantly, the evidence on the political economy of universality in transfers is limited. While the elements of the hypothesis are intuitive, hard evidence on the magni- tude of this effect is, by nature, difficult to come by.13 In Africa, for example, studies have shown that whether a universal program is publicly acceptable—and hence represents “good politics”—hinges on how well it aligns with prevailing notions of “deservingness” (Hickey et al., forthcoming; Quarles van Ufford et al. 2016). Also, in the limited imple- mentation of full UBI to date, they have all been funded not by direct or indirect taxation, but by channeling natural resource revenues, energy subsidy reforms, or in the case of some pilots, via private sector donations. Furthermore, when financing is considered, a UBI may become a targeted intervention—one where both receipt and amount are tapered out via taxes—hence resetting the discussion as not all people would benefit equally or even benefit at all. Chapter 1. The Idea of Universal Basic Income 27 Explaining Limited Coverage Let us consider the causes of low coverage among the poorest households in current systems, as well as if and how a UBI might help tackle them. The problem of limited coverage is well documented: for example, estimates show that of the poorest quin- tile of the population in each country, an average of 45 percent receives some form of social assistance, while the share is only 18  percent in low-income countries (World Bank 2018a). While several factors contribute to low coverage, they affect the problem in dif- ferent ways. Table 1.2 summarizes four main barriers. A central issue, especially in low-income countries, is low spending and fiscal constraints (Hanna and Olken 2018). In Africa, for example, total safety net coverage is lower than the number of people in poverty everywhere except Botswana, Mauritius, Namibia, and South Africa—so even if programs were only meant to serve the poor and did so perfectly, coverage would be insufficient, often by severalfold. In countries with significant flagship programs as in Ethiopia, Kenya, and Tanzania, the poverty rate (US$1.90/day) is between 34 and 47 per- cent; yet safety nets cover only 8–13 percent of the population.14 In countries such as Madagascar or Sierra Leone, the disjuncture is much larger—in Sierra Leone, there are more than 10 times as many poor people as those served by safety nets; and in Mad- agascar, more than 20 times (Beegle, Coudouel, and Monsalve 2018). In other words, low budgets may leave large swaths of people in need uncovered almost by design (e.g., via rationing). Therefore, it is expected that a program that, for example, allocates over 4  percent of GDP (Georgia’s old-age pension) displays better targeting outcomes than a similar program costing nearly 66 times less, or 0.06 percent of GDP, such as India’s IGNOAPS scheme (Kidd and Athias 2019). Two other factors that contribute to low coverage among social protection systems are limited awareness and costs for participation. In order to be enrolled, people have to know about the program, want to be enrolled, and have low enough transaction costs to make it worthwhile. Barriers faced by the poor are numerous—lower literacy and con- nectivity to media, speaking languages other than the official language of the country, TABLE 1.2  Barriers to Coverage Factor Barrier Effect of a UBI Funding •• Lack of funding •• Would likely amplify it, probably severalfold Awareness of •• Information •• Likely to help in overcoming awareness barrier the program and thereby reduce errors of exclusion •• Stigma Costs of •• Monetary costs and time •• Enrollment costs would be one-off (no participation recertifications necessary) and probably lower •• Physical barriers •• Costs of collecting benefits likely to be the same as for targeted transfers Eligibility •• Information to set up targeting criteria •• Inclusion and exclusion errors would not apply, determination as a universal program is conceptually for •• Method-specific limitations everyone •• Data changes over time •• Possible manipulation by administrators 28 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov living in areas poorly served by state services, high transaction costs, not holding requi- site identity (ID) documents, etc.—all of which can add up to significant undercoverage among the poor even in a categorical, age-based program. For example, the simple failure to identify the age of a person in 22 African countries caused social pension pro- grams to not reach up to 30 percent of eligible elderly (Guven and Leite 2016). Social assistance programs have been increasingly diligent in tackling these issues with a host of initiatives for active outreach. For example, in Brazil, an active outreach strategy for the social registry was in initiated in 2011 with the tag line Conhecer para Incluir (to know so as to include). Outreach was intense until 2014 and included media outreach and door-to-door efforts in target areas from slums to jungles. About 1.5 million new families were added to the national social registry used for 30 programs, principal among them the Bolsa Família conditional cash transfer (CCT). Of these families, over a million were from traditional groups (e.g., indigenous or riverine populations) that are highly vulnerable and often underserved. In Argentina as recently as 2016, 1.5 million children were not receiving the universal child allowance principally because of issues with either the children or their parents not having IDs or mismatches in the linkages among them, or issues related to verifying children’s enrollment in schools.15 At its early phases of introduction, a UBI would require scaling up outreach efforts in a substantial way. It is not unlikely, however, that awareness about a UBI entitlement would likely be widespread in a short period of time. The program may require significant administrative effort around rollout, follow-up in perpetuity to keep pace with life-cycle events, and possibly changes in citizenship or residency (see chapter 7). The intensity of such administrative efforts would depend on institutional and delivery capacities, the pace at which the UBI is introduced, and whether it is a new program or a substitute or extension of other programs. Overall, a UBI may have beneficial effects in drastically reducing the various proce- dural and transaction costs beneficiaries incur when applying and waiting for eligibility determination.16 The permanent and open nature of a UBI should help overcome barriers related to information—even if they are not aware or willing in the first year, individu- als initially not enrolled could gain information or confidence over time as others in their social network begin to benefit. Similarly, a UBI should help eliminate enrollment constraints based on stigma, suppositions that people might not meet eligibility require- ments, or opportunity costs in participation. A fourth factor around low coverage of the poor by social assistance is erroneous eligibility determination. Program eligibility relies on detailed information, which can be difficult to observe and is also ever-changing. There are in fact errors of exclusion due to the nature of poverty measures and limited information (Alatas et al. 2012; Coady 2018; IMF 2017; Kidd and Athias 2019; Premand and Schnitzer 2018). Needs-based targeted programs also present challenges in terms of where to draw the eligibility threshold (box 1.2). Taken in isolation, targeting methods can present severe challenges in contexts with significant information constraints17 (Brown, Ravallion, and van de Walle 2018); yet their performance can increase remarkably in higher-capacity contexts, as in the United States (Alderman, Gentilini, and Yemtsov 2018). Chapter 1. The Idea of Universal Basic Income 29 BOX 1.2  Welfare Continuity and Poverty Dynamics A n issue often invoked in favor of a UBI—and of more universal approaches in general—is that welfare distribution does not present natural discontinu- ities suggesting where an eligibility threshold should fall (see figure B1.2.1 for such a distribution for Bangladesh in 2016). The matter is intertwined with a deeper issue of definition and measurement of poverty as a concept, which can be elu- sive and involve some arbitrariness (Brown, Ravallion, and van de Walle 2017, 2018; Knox-Vydmanov 2014; Pritchett and Kenny 2013). FIGURE B1.2.1  Welfare Distribution in Bangladesh Poverty line = US$3.20 100 Proportion of population (%) 80 60 Headcount = 51.22% 40 20 0 0 2 4 6 8 10 12 14 16 18 20 Welfare aggregate (per capita per day) SOURCE: World Bank Global Stats team using the Global Poverty Working Group Microdatabase. The discussion is also related to an evolving understanding of poverty dynamics in developing countries. Even where poverty is less prevalent than in cases like Bangladesh, there can be a concentration of similarly vulnerable people around poverty lines. For instance, in some middle-income countries, those living on US$6/ day, or just above the poverty line, face a 40 percent probability of falling back into poverty (Lopez-Calva and Ortiz-Juarez 2011). In fact, poverty is often dynamic: in Africa, one-third of the population is persistently poor, while another third moves in and out of poverty (Dang and Dalaban 2019). These issues suggest the need for predictable coverage that would protect both those in poverty and those at risk of it. The smooth distribution of poverty and the volatility of income imply that a policy that does not present sharp cliffs may be more appropriate, which some- what contrasts with approaches providing assistance more selectively to the poorest. These differential needs, however, could equally argue for a benefit struc- ture that is not uniform across the population, hence violating the UBI parameter of equal transfers. 30 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov Conditionality The use of unconditional cash transfers (UCTs) has grown remarkably and is now pres- ent in at least 101 low- and middle-income countries. Yet a significant portion of social protection programs are linked, more or less explicitly, to some form of reciprocity in participant behaviors. And, as part of the rapid rise in cash transfer programs in the last two decades, a significant share has conditions that require beneficiary households to seek basic maternal-child health care and/or education for school-age children. Such programs now exist in 63 countries worldwide.18 Conditionality is not just about child-re- lated co-responsibilities, as poverty-targeted benefits are commonly tied to job search requirements, training, or community service. In a way, even public works could be con- sidered a form of conditionality. The issue of co-responsibilities or conditionality raises a set of conceptual, practical, and empirical matters that are worth reflecting on as they cast light on the acceptability and impacts of the move to unconditionality implied in a UBI. We here review them in turn. First the concepts. While virtually every society cares about its disadvantaged members, such altruistic attitudes can translate into different preferences over welfare regimes and, in turn, taxpayer utility functions (Currie and Gahvari 2008; Reinhardt 2013). For example, under one assumption taxpayers maximize their own utility when the poor are allowed to maximize their own choice, such as by receiving unrestricted support. Under other assumptions, taxpaying voters typically exhibit a more parental form of altruism that would not necessarily favor support with “no strings attached” (Tobin 1970).19 In other words, the preference among voters for bestowing on the poor conditional or unconditional benefits may well reflect societal values, attitudes, culture, and preferences toward redistribution. These overarching philosophical and cultural considerations emerge in policy debates, sometimes with no shortage of inflammatory rhetoric.20 Accord- ing to some quarters, imposing conditions on poor households tends to evoke the specter of structural adjustment (where macro conditions were part of policy reform packages) or 19th century–style social assistance providing “punitive” and minimal support. Freeland caricatures the “productivists” championing CCTs as one of the “four horsemen of the donor apocalypse” determined to pursue an agenda whereby “nobody should be given something for nothing” (Freeland 2013, 232). The vision postulates that conditions are forced onto powerless people while depriving them of their dignity. On the other hand, there is a general understanding that societies are woven together by a set of core rights and responsibilities, and that these would need to coexist in gen- eral—for instance, “pay taxes, receive services” or “work hard, care for others, get help when needed”—as well as in specific policies. These considerations might be less com- pelling for social services, where the notion of unconditional health, education, or child care provisions tends to resonate societally. Yet cash transfers engender a different philo- sophical and political conversation, especially where, for historical antecedents and moral principles, cash is interpreted as a reward stemming from effort, even if broadly defined. Chapter 1. The Idea of Universal Basic Income 31 In some polities, the nature of the welfare discourse may be such that the polit- ical viability of an intervention may hinge on embedding some form of beneficiary co-responsibilities into the program’s design. The chief architect of Mexico’s national CCT program Prospera recollects that co-responsibilities or conditions might “contrib- ute to the acceptance of the cash transfer program by the general public, given that in some countries a program that ‘just gives money’ to poor families may not be politically acceptable” (Levy 2007, 125). From this perspective, the attachment of conditions may well reflect realpolitik instead of adherence to ideological principles. From an economic perspective, choices over conditionality generally revolve around the notion of underprovision of goods. In broad terms, the theory suggests that expected externalities of a conditional transfer would be desirable when there is a pri- vate underinvestment—say, in nutrition or health—below an optimal social (or even private) level. Conditions represent a vehicle to influence behaviors, especially when people may not be well informed or may have inconsistent preferences over time, and there are coordination failures. This is the case when there is a discrepancy between per- ceived and expected returns, for example, to nutrition or education, which somewhat challenges the assumption that “people always know best.” However, there could also be unintended behavioral effects, such as reinforcing preexisting gender disparities (Buller et al. 2018; Peterman et al. 2017). Practically speaking, a conditional transfer needs a certain institutional and adminis- trative capacity for implementation. This includes both a well-functioning supply of services, and a system for coordination and reconciliation of data across the sectors that are being “conditioned to” under the program. This is why, for instance, PROGRESA was first tested in more conducive urban areas, and then rolled out only where adequate services were available. Simultaneously, for over a decade an unconditional arm of the program oper- ated in areas poorly served by schools or clinics (Alderman, Gentilini, and Yemtsov 2018). An implementation perspective also reveals that conditions are not a binary vari- able but rather that conditionality exists on a “hard to define continuum” ranging from informal to formal conditioning (Pellerano and Barca 2017), with the practice on condi- tionality generally falling into three categories: •• “Labeled” CCTs, which use informal and nonbinding conditioning to link a scheme with certain behaviors, without explicitly requiring, monitoring, or enforcing any co-responsibility •• “Soft” CCTs, which envision formal conditions that are only gently enforced, meaning they are explicitly required and monitored to some degree, but penal- izing for noncompliance is light or nonexistent •• “Hard” CCTs, which display formal conditions that are routinely monitored and strictly enforced, with significant penalties for noncompliance Empirically, there is an emerging but still rather inconclusive body of evidence sug- gesting that CCTs lead to mild increases in the use of required services (see box 1.3 and 32 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov BOX 1.3  Evidence from Systematic Reviews on the Impact of Conditions in Cash Transfer Programs F ive reviews of cash transfers’ impact on wide-ranging outcome areas conclude that CCTs can, under certain circumstances, have a higher impact than UCTs. •• Analyzing cash transfers’ impact on education through 35 studies, Baird et al. (2014) conclude that both CCTs and UCTs improve school enrollment and attendance, with no significant difference between them. However, when categorizing transfers according to the strictness of conditionality, the researchers conclude that “hard” CCTs have substantively larger effects on enrollment and attendance than either UCTs or “soft” CCTs. But they also note that none of the programs significantly affect longer-term outcomes, mea- sured through improvements in test scores. •• Focusing on cash transfers’ impact on the use and quality of maternity care services, Hunter et al. (2017) find only one published study exploring a UCT’s impact on maternity service uptake, in which no significant effect was iden- tified. By comparison, more frequently studied CCTs appear to increase the proportion of women receiving multiple antenatal checkups (a condition of the transfer). However, the CCT did not increase uptake of other maternity care services that were not included as conditions, highlighting the relatively narrow scope of CCT impact. •• The potential—but narrow—impact of CCTs is also noted in the literature review undertaken by Taafe, Longosz, and Wilson (2016) on cash transfers’ effects on livelihoods, education, and health. They conclude that condition- ality is not always necessary to produce an impact, but may lead to stronger effects. However, conditionality may limit outcomes to those linked to the conditions, whereas UCTs have the potential to generate more widespread impact across development objectives. Furthermore, conditionality requires significant administrative and financial resources that may not be feasible in lower-income contexts. •• A rigorous, wide-ranging literature review by Bastagli et al. (2019) concludes that conditionality can improve the outcomes on which the transfer was con- ditioned, but argue that much of this impact may be achieved with the clear messaging and communication of informal or soft conditions, rather than the socially and administratively costly sanctions associated with hard CCTs. •• This conclusion is more or less mirrored in a subsequent broad literature review by Pellerano and Barca (2017), which concludes that explicit conditional- ity can enhance the outcomes of a cash transfer, but that this may also be achieved through less explicit forms of behavioral conditioning. The authors (continued) Chapter 1. The Idea of Universal Basic Income 33 BOX 1.3  Evidence from Systematic Reviews on the Impact of Conditions in Cash Transfer Programs (continued) also highlight the lack of evidence of any sustained change in behavior among CCT participants over the longer term. Two systematic reviews—both relating to health—conclude that UCTs either match or outperform CCTs. •• Exploring cash transfers’ impact on nutrition, Manley, Gitter, and Slavchevska (2012) review 24 papers on 18 programs in 11 countries and find little difference between UCT and CCT impacts, with health and education–focused CCTs having the same effect on child height for age as UCTs. However, they note that CCTs with other types of conditions, mostly related to working or saving, show strongly negative impacts on nutritional status, revealing the potentially adverse secondary effects of conditionalities. •• This resembles the findings of Siddiqi, Rajaram, and Miller (2018) in their sys- tematic review on the impact of cash transfers on newborns’ health. Looking at data from 14 studies, they find that both UCTs and health-focused CCTs tended to improve infants’ birthweight outcomes and reduce infant mortality, while CCTs conditioned on labor force participation had no impact. Three systematic reviews—two on health and one on child labor—argue that the evidence is inconclusive in determining whether conditionality increases a cash transfer’s impact. •• Considering cash transfers’ impact on child labor, de Hoop and Rosati (2014) conclude that both CCTs and UCTs reduce children’s participation in child labor and their hours worked, with more information needed to determine whether schooling conditions matter in this regard. •• Looking at cash transfers’ impacts on contraception use, Khan et al. (2016) also conclude that the available evidence of CCT versus UCT effectiveness is inconclusive due to the limited number of studies, varying outcome measures, and lack of interventions specifically for contraception. •• In a Cochrane review focused primarily on UCTs’ impact on health outcomes, Pega et al. (2017) conclude that UCTs may not significantly affect health service use but may still improve some health outcomes and health care expenditure levels. They consider the Burkina Faso, Malawi, and Zimbabwe randomized controlled trials comparing UCT and CCT impact on health and conclude that the evidence on the relative effectiveness of different types of transfers remains very uncertain. SOURCE: See appendix C. 34 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov appendix C). Policy designers should be mindful that cash transfer programs produce a vector of impacts starting with the use of required services, but also including effects on consumption—especially of more and more nutritious food; and of improved mental health, confidence, and empowerment—and sometimes on increased savings or invest- ments in livelihoods, reduced migration, etc. If an individual or family is excluded from benefits for lack of, for example, enrollment, then the family may be excluded from these other benefits (not related to education). Empirics on the forgone impacts from excluding noncompliant children or households are largely missing. Because penalties for noncompliance in some CCT models are rather rarely exercised, we conjecture that such lost potential impacts may be small, but are nonetheless worth considering in any choice around conditioning. To sum up, a UBI that renders support without conditions resembles many UCT programs found in social assistance programming around the world. But the prevalence of some sort of conditioning—even if only notional—suggests that the unconditional fea- ture of a UBI will be challenging to present practice in at least some places. The empirical evidence suggests that without conditions, there may be some mild reduction in service uptake. However, sizable impacts seem to be achieved by programs with well-implemented soft conditionalities, which are likely to be less administratively and cost demanding. Transfer Modality Economics and Beyond A survey of economists found that 84  percent of them agree with the statement that “cash payments increase the welfare of recipients to a greater degree than do trans- fers-in-kind of equal cash value” (Mankiw 2009). The result from the profession is not surprising. In fact, standard economic theory predicts that, under certain assumptions,21 cash is more “utility maximizing” than in-kind transfers. This stems from the basic fea- ture that cash is flexible and provides people with choice on how to spend it. Cash can also accomplish broader goals, such as redefining the balance of power between govern- ment and its citizens in favor of the latter. Yet there are also limitations, and this section acknowledges that cash is not appropriate in all contexts. Cash transfers have grown enormously in coverage and spreading across developing countries. In Africa, over 2010–15, an average of 14 new safety net programs were intro- duced annually, mostly cash based. Such growth could partly be explained by the solid evidence base underpinning such programs. A number of myths on cash-based social assistance have been dispelled by recent evaluation compilations: research has found that cash transfers are overwhelmingly spent wisely or on desirable consumption or productive goods (Evans and Popova 2017); the risks of labor supply disincentives have been largely debunked (Baird, McKenzie, and Özler 2018; Banerjee et al. 2017); and cash transfers have a range of impacts on critical dimensions for growth, such as cognitive development, accu- mulation of human capital, asset protection, and fostering social cohesion (Bastagli et al. 2018; Beegle, Coudouel, and Monsalve 2018). Cash can spark local economic multipliers, Chapter 1. The Idea of Universal Basic Income 35 with recent evidence showing that for every dollar injected, between US$1.27 and US$2.52 is generated in the local economy (Handa et al. 2018). While there is ongoing debate on specific research questions—such as the duration of effects over time (Blattman, Fiala, and Martinez 2018)—the empirical track record of cash transfers is unrebuttably strong. Compared to in-kind food, clothing, fertilizer, or school uniforms, moving cash around to beneficiaries has always been much simpler. The procurement, storage, and distribution problems for goods can be considerable. In contrast, cash is compact and nonperishable. With the advent of transactions through the financial sector and mobile money, e-cash payments can reduce certain types of security concerns. Even in human- itarian crises, cash transfers are now delivered using electronic or even block-chain technology, such as in support of refugees in Jordan. In fact, the savings in using cash as a transfer modality can be substantial. Evidence shows that, in four studies that com- pared equal value of transfers, between 13 and 23 percent additional households could have been reached if food transfers were in cash instead (Gentilini 2016). At the same time, cash transfers are never implemented in isolation and coexist with in-kind food programs. These reach about 1.5 billion people in low- and middle-in- come countries (Alderman, Gentilini, and Yemtsov 2018). Recent evidence is casting new light on such decades-long experiences. Two main lessons stand out. First, the fact that a program is food or cash based is not necessarily a determinant of performance in terms of coverage, targeting accuracy, and a host of other dimensions. Indeed, some countries that maintained an in-kind modality managed to improve its performance remarkably, such as with the Supplemental Nutrition Assistance Program (SNAP) in the United States and the targeted Public Distribution System in select states in India. Yet, other countries that initially maintained in-kind provisions, such as in Indonesia’s Rastra program, did not improve their performance significantly (the Rastra program eventually transitioned to vouchers). Other countries that commenced a transition to vouchers and cash transfers were able to reap the benefits from such a conversion; this was the case for Mexico’s Pro- grama de Apoyo Alimentario and, to some extent, the Arab Republic of Egypt’s schemes. For some countries, however, such as Sri Lanka, the change in modality did not translate into enhanced program performance. What seems to drive improvements is not so much the transfer modality, but factors such as political leadership at key junctures, credible evidence, a judicious use of technology, appropriate delivery processes, and ways to hold providers accountable (Alderman, Gentilini, and Yemtsov 2018). Second, unlike cash transfers, food-based programs tend to pursue a variety of intertwined functions. For example, they support farmers via procurement (agricul- tural goal), they are leveraged to manage price fluctuations with strategic storage (risk management goal), and they provide income support to low-income consumers (social assistance goal). The role of cash is more streamlined, as it largely revolves solely around the social assistance function. This multiplicity of objectives for food-based programs means more stakeholders, a thornier political economy of reform, and likely higher costs. But it also demands careful consideration of possible systemwide effects from a complete replacement of in kind with cash as would occur with a UBI. This caution is particularly compelling for the capacity of lower-income countries to handle food price 36 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov volatility (i.e., in-kind provisions are often used as a way to protect against inflation), and the possible alienation of political and financial support from key constituencies in all country contexts. Though the hypothesis that cash is an efficient and acceptable transfer modality is strong, there are some limitations to its applicability and circumstances when in-kind provision may be preferable. For instance, turning needs into effective demand is a key rationale behind the microeconomics of cash transfers. But doing so might be challeng- ing in the presence of weakly integrated or poorly competitive markets. In such contexts, price transmissions would not necessarily signal relative scarcities, and localized cash injections may result in price spikes—leaving other consumers and net buyers worse off. If there are circumstances where local markets may perform poorly, food prices may be excessively high or volatile, and private traders may not supply commodities efficiently. In those contexts, a cash transfer may lead to neither more choice nor more purchas- ing power, and in-kind food may be a more appropriate response (i.e., it ensures both availability of and access to food). For example, in parts of Kenya, cash transfers did not protect purchasing power in areas with dysfunctional markets, which affected nutrition negatively (Dietrich and Schmerzeck 2019). Sound implementation is another factor shaping preferences. In India, it has been documented that people’s preferences for cash or food depend “on a combination of pragmatism, shrewdness and deep understanding of the local circumstances” (Khera 2014, 44). In particular, the study showed that preferences hinged on the implemen- tation performance of the targeted public food distribution system. In states where the system worked poorly, people preferred cash; where the existing food distribution system worked well, larger shares of people preferred food. Gender tends to be another key factor affecting preferences. In a number of soci- eties, women tend to prefer food, which they are more likely to control, while men may prefer cash transfers. Anecdotal evidence shows that the process of intrahouse- hold decision making also counts (e.g., men and women deciding in concert how to use household resources, even in contexts where women may be physically constrained in reaching markets or face risks in accessing them, such as in refugee camps). Yet very few studies, if any, are testing the differential impacts (e.g., on nutrition) of cash versus food transfers as provided to men and women. Qualitative research is shedding light on intra-community effects of cash transfers, some of which are positive and empowering, while others may generate undesired con- sequences in terms of social relations. In Zimbabwe it was observed that, unlike food, cash transfers were not shared within the community, hence hindering informal mutual support and risk management mechanisms among members, but reinforcing targeting (MacAuslan and Riemenschneider 2011). The intra- and inter-community effects of alter- native transfers is an issue that may deserve further qualitative and quantitative study. One way in which in-kind transfers may have an advantage over cash is in terms of nutrition, such as via micronutrient fortification (Alderman, Gentilini, and Yemtsov 2018). Only a few studies have documented the comparative impact of transfers on nutrition in direct comparisons. In one of those, Langendorf et al. (2014) assessed several Chapter 1. The Idea of Universal Basic Income 37 types of cash and food combinations, including a range of different high-quality foods (e.g., lipid-based supplements and fortified cereals) as well as more traditional ones (veg- etable oils, pulses). The intervention aimed at reducing severe and moderately acute malnutrition as well as mortality rates among children. Findings showed that combining food and cash transfers reduced the incidence of malnutrition at about twice the rate of either a cash transfer or supplementary food alone. Reviews of the evidence have thus shown that the impact of cash and in-kind transfers on welfare (especially food security) varies by indicator, although they tend to be similar on average. However, implementation costs tend to be lower for cash. In gen- eral, the performance of transfers seems to reflect the interactions among a number of factors such as the profile and initial conditions of beneficiaries, the functioning of local markets, program objectives, and the implementation context (Gentilini 2016). Cash as a transfer modality is applicable in many, though not all, contexts. In the design of UBI proposals, the notion that benefits be paid in cash is not very radical and, while there are unknowns (e.g., inflationary effects), it may be the least contested part of a proposal. Inflation Possible inflationary risks are one of the most debated possible effects of a UBI, and a central concern of low-income individuals. Such risks should not be easily dismissed a priori, nor should they be overly magnified. Instead, they should be assessed within the framework of analytical parameters, contextual factors, and relevant experiences. What practical experiences can inform the discussion on such effects? In the case of large-scale, one-off transfers in high-income countries, there is no evidence of infla- tion. For example, in January 2011, Kuwait announced an Amiri grant of US$3,600 to be paid to all 1.1 million Kuwaitis on February 24 to celebrate the 50th anniversary of independence.22 Monthly data on the consumer price index reveals no significant dis- continuity in price values around the time of the transfer announcement and payment. Similarly, in 2008 and 2009, the government of Australia delivered a fiscal stimulus package with various cash bonus schemes.23 One-off cash payments were made to about 90 percent of households and 80 percent of working-age individuals, with the average bonus amounting to $A 1,600 (Hyslop 2014). This represented around 4–5 percent of individual income, and the cash bonuses collectively accounted for nearly 2  percent of GDP. Again, consumer price index analysis reveals no discernible effect on inflation attributable to the bonus. Yet these findings cannot rule out more localized cases of price increases in both countries. One possible reason for limited inflation in high-income countries is that markets tend to be more integrated than in low- and middle-income contexts. This is a salient issue: if the relevant market receiving cash is mainly local, isolated, and weakly integrated into the wider economy, then effects may differ. The presence of oligopolistic producers or, if the local market is competitive, a rising marginal cost of local production will likely translate the demand from cash transfers into higher prices. In turn, transaction costs in reaching such areas may offset the potential attractiveness of serving such high-price 38 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov markets by other suppliers. The classic literature on famines, for instance, provides ample analytical treatment of such mechanisms starting from the 1980s (Drèze and Sen 1989; Devereux 1988; Ravallion 1997). Instead, if markets are well integrated, more competition among suppliers to meet the cash-induced demand may likely result in no or little inflation. Recent evidence from Mexico illustrates these arguments. Experimental work on the price effects on cash and in-kind transfers suggests that “for typical transfer pro- grams, price effects may not be economically significant in many communities” (Cunha, De Giorgi, and Jayachandran 2017, 3). However, in less developed Mexican villages, in-kind transfers decreased the price of select commodities (those provided as part of the food basket) by 5 percentage points; cash transfers in similarly remote areas led to mild increases (1.5 percent) in overall food prices. The intensity of the cash injection matters as well. In the Philippines, in remote areas where the provision of cash transfers was significant (i.e., where village income increased on the order of 9 percent), the price of nontradable, perishable protein-rich foods increased by 6–8 percent (Filmer et al. 2018). Such an effect generated unintended impacts on the nutritional levels of nonbeneficiary children (whose stunting rates rose by 11 percentage points) with inflationary effects persisting 31 months after the program’s introduction. Similar mechanisms can be at play in other markets, including health care. For example, evidence from Indonesia shows that cash transfers increased the demand for health services by beneficiaries, but crowded out health utilization by nonbeneficiaries (Triyana 2016). In fact, the demand generated by cash transfers led to higher health costs due to the limited supply of trained health providers. This inhibited access to health by poor nonparticipants. These effects may not occur where the supply side of services is less constrained, as in Thailand. In other large-scale experiences, such as in Alaska, no studies have yet tested impacts on inflation empirically.24 In Mongolia, some minor inflation was detected, although it is poorly documented (Yeung and Howes 2015); and in the Islamic Republic of Iran, the UBI scheme was itself implemented in a highly inflationary context—a factor that wiped out three-quarters of the program’s real value between 2012 and 2018. Index-linking cash benefits to inflation can help, but above certain thresholds it becomes more effective to provide in-kind transfers. The presence of basis risk could hinder the effectiveness of index linking (i.e., the inability of programs to detect localized price spikes), and poor people often do not trust policy makers in doing so (Drèze 2017). In sum, our analytical toolbox and recent experiences show that possible effects on inflation, and their intensity, are not predetermined in the abstract. Instead, they would likely hinge on overall market conditions, the specific market for subproducts and ser- vices, the size of the program, and probably the duration of the intervention. Choices on Other Parameters The three core choices of universality, unconditionality, and cash as transfer modality define a UBI, but there are variations possible in the setting of other parameters—the level and frequency of benefits, the inclusion of children or devising the program for adults Chapter 1. The Idea of Universal Basic Income 39 only, the inclusion of citizens or residents. These design dimensions should be considered in tandem with factors discussed in other chapters, such as how the program is financed, what programs are being replaced and related distributional effects (chapters 4 and 5), specific political economy dynamics (chapter 6), and delivery capabilities (chapter 7). It is their holistic consideration that will help illuminate how a UBI would fare in a given context. Level The classic UBI proposal is to give an equal benefit to all. It is indeed worth reflecting on the implications of the equal amount benefit structure. For some social goods, equal treatment or inputs is clearly what is desirable. In a democracy, everyone should have one vote. For other social goods, societies may pursue equality-related outcomes, which may require unequal inputs (Devereux 2016). For example, governments may want to have children able to read by the end of the second grade. This may take basic education for all, plus more intensive instruction for children with learning disabilities. Similarly, for everyone to live a healthy life to age 60, it may require giving those with diabetes or asthma more intensive health care than others. So, if social policy aims to ensure ade- quate living standards for all, provision should acknowledge that the most disadvantaged would be further behind and would thus need more, not equal, support. Such differenti- ated provision or prioritization implies the need for targeting, which is one of the reasons why most countries’ social protection systems include at least some programs that focus on the poor. (However, as we discuss in this chapter, this becomes a slightly more con- tentious issue when a program focuses on the poor only.) In principle, these considerations do not rule out a UBI, but they do emphasize the need to complement it with other more tailored interventions for those worse off. Pro- viding such augmented provisions within a vision of universality seems in line with the spirit of the social protection floor, as well as the experience of universal health cover- age with progressive universalism (Cotlear et al. 2015; Gentilini 2018; Gwatkin and Ergo 2011; Jamison et al. 2013; Marmot Review 2010). The considerations around transfer levels should be aligned with the core ratio- nale for a UBI, or its objectives (table 1.3). These can vary significantly, and we explore such diverse narratives later in this chapter. As is further discussed in chapter 4, setting the benefit level of UBI would need to be determined based on its goals as well as syn- chronized within the broader design and financial and political trade-offs (Arnold 2018; Francese and Prady 2018). TABLE 1.3  Alignment of UBI Narrative with Basic Transfer Context and objective Reference basic transfer size Automation related (insurance against protracted Minimum living standard artificial intelligence–related unemployment) Natural resource dividends Variable based on revenues Social assistance Amount to lift people above poverty line, or that ensures access to a minimum set of kilocalories, or that addresses specific nutritional goals 40 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov Frequency In general, the default option for payment frequency is monthly (or as close to this as transaction costs in the payment system make practical). Such a schedule is consistent with helping people to meet their daily needs for food, shelter, and other necessities. But a few policy proposals set out a UBI with a different distribution frequency. As we discuss below, Alaska distributes annual transfers from its natural resource revenues, with the frequency matching the purpose—distribution of dividends rather than equity or con- sumption smoothing—and for a relatively small total amount (Marinescu 2018).25 The decision around frequency of transfers is important because it can affect spending patterns. Daidone et al. (2019) review a compilation of evidence from seven African countries and observe that, when transfers are more lumpy and less frequent, they are more likely to be invested in productive assets. Similarly, evidence from Sri Lanka (Gentilini 2016) indicates that when transfers are less predictable and frequent, they are more likely to be treated as “income windfalls” by beneficiaries, and hence used for nontraditional investments (e.g., to buy higher-quality foods). However, in Nigeria, receiving chunkier, less-frequent transfers made no substantial difference in the pro- portion of cash held by women and in the overall positive impact on household living conditions (Bastian, Goldstein, and Papineni 2017). Children and Adults Some variants of UBI proposals, especially those posited as responses to automation-in- duced employment or wage losses, include only adults or working-age adults, meaning, the “to all” is limited by age. The UBI simulations by Browne and Immervoll (2017) are designed in this way, as was the pilot program in Namibia. Most of the proposals listed in appendix B are for adults. Since a large share of current social assistance programming is focused on children, this may mean that households with children may be worse off under a UBI proposal than under current programming—unless a UBI is also provided to children in full or in part. Comparisons depend on the specifics of the proposal and how it would substitute for, or add to, existing programming (see chapter 4). But it is clear that the choice of a UBI’s demographic composition would have significant bearing on its fiscal envelope. For instance, in low-income countries, cost estimates for a UBI for the full population versus variants for partial coverage of children or of adults only can vary by about 10 percentage points of GDP (figure 1.2). Residents and Citizens Whether the concept of universality pertains to residents or citizens is a fraught issue. A human rights or philosophical motivation would suggest that benefits should go to all people living in a territory regardless of legal status. But the distribution of cash transfers has long evoked fears of welfare migration—attracting people from other areas both within federal countries with decentralized welfare programs and from outside national borders. While the evidence shows that the impact of transfers on mobility Chapter 1. The Idea of Universal Basic Income 41 FIGURE 1.2  Cost of a UBI by Poverty and Demographic Variables 100 79.1 80 62.3 60 % of GDP 40 28.0 22.8 23.1 19.8 19.3 20 9.0 9.6 5.2 5.1 3.5 0 Amount equal to Amount equal to Poverty gap (inter- Poverty gap (inter- national poverty lines national poverty lines national poverty lines, national poverty lines, (total population) (adults + 50% children) total population) adults only) Low-income country Lower-middle-income country Upper-middle-income country SOURCES: Ortiz et al. 2018; World Bank 2018b. hinges on design (box 1.4), more evidence on how soon after immigration and in what measure new residents join the ranks of taxpayers may be helpful in informing evi- dence-based discussion. In the United States, for example, the average immigrant makes a net positive fiscal contribution of US$259,000 in net present value (Clemens et al. BOX 1.4  Do Cash Transfers Affect Mobility? A global review of practices shows that social assistance programs can be clas- sified in three clusters: (1) social assistance that implicitly deters migration, centering on place-based programs; (2) social assistance that implicitly facilitates migration by relaxing liquidity constraints and reducing transaction costs; and (3) social assistance that is explicitly conditioned on spatial mobility. The research finds that impacts on migration generally align with the implicit or explicit goals of interventions. Under cluster 1, the likelihood of moving declined between 0.22 and 11.0 percentage points; among schemes in clusters 2 and 3, the probability to move soared between 0.32–25 and 20–55 percentage points, respectively. The analy- sis also reports spillover effects within households and communities. While social assistance seems not to determine migration decisions per se, it nonetheless enters the broader calculus of mobility decision making. As such, social protection can be an important part of public policy packages to manage mobility. More research is needed to better understand the role of social protection in structural transforma- tion—a process underpinned by domestic mobility and one whose performance may ultimately affect international migration. SOURCE: Adhikari and Gentilini 2018. 42 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov 2018). Meanwhile, most proposals envisage participation of national citizens, although there are exceptions—for example, Atkinson’s “participation income” would include res- idents26 (Atkinson 2015). The citizenship criteria presumably minimizes the impetus to migration while also circumventing the thorny issue of residency, which can be a com- plex challenge in informal settlements in low- and middle-income countries (Gentilini 2015), but may also generate societal tensions. Individuals The change in focus from household to individual has a number of implications. Giving individual and equal benefits means that large households will receive larger payments in proportion to their size. This benefit structure is intuitively obvious—every person needs to eat, to be clothed and educated, etc. But, in fact, social protection programs are rife with benefit structures that do not increase linearly with family size. There are some- times economies of scale factored into costs of living, which is a reasonable conception but full of empirical unknowns and variations and thus prone to dispute. There are often limits to the number of children for which benefits will be paid, and sometimes there are flat benefits per household irrespective of size, often for (largely unfounded) fear that benefits will increase fertility. Further, by giving benefits to individuals rather than households, one of the more challenging features of administering household-based pro- grams—tracking who belongs to what household and the concomitant IDs involved—is simplified. The system of individual IDs is sufficient to support a UBI, without build- ing a system for linking and updating household IDs and their memberships; minors, however, will still need to be linked to an adult. A focus on individuals may also help in making programs more portable, that is, able to follow people independently of where they live (box 1.4). This could be an important consideration in providing benefits to sea- sonal migrants and other mobile urban dwellers (Gentilini 2015). A focus on individuals may also have effects within households, though there is a dearth of evidence as to their magnitude. One line of thought is that by giving benefits to all—including those with traditionally weak bargaining positions within households or com- munities, such as people with disabilities or the elderly—it may be empowering. In theory, the provision of cash to all individuals in a family, as opposed to selective provision to some members and not others, may reduce tensions stemming from competing over scarce resources. Just as current cash transfer programs may often reduce intimate partner and emotional violence within the household relative to nonrecipients (Buller et al. 2018; Peter- man et al. 2017), the provision of cash to all members may further attenuate those tensions. Another view is that making all household members beneficiaries may affect household composition and size. Households form and persist on the basis of affec- tion and social norms, efficiency in task sharing, and resource pooling. By giving each individual an autonomous income that does not depend on the unit in which he or she lives, a UBI theoretically gives every individual more choice about their household mem- bership—though the magnitude of the effect may be small relative to the other factors that drive household formation. The valuation of the effect may be viewed by different Chapter 1. The Idea of Universal Basic Income 43 people as either emancipatory, if people are not bound to households by need over affect, or destructive of family values, if it were to result in more divorces, young unmar- ried women living independently, or the elderly not receiving family care. Phase-In Modality The way a UBI is introduced matters. For example, it could be phased in by replacing existing schemes, be provided on top of them, or a blend of both. Figure 1.3 illustrates those modalities. The first option, presented as figure 1.3a, embodies the standard approach to UBI. This approach would be relatively complex given the choices on what programs would be replaced by a UBI—or how far the process of replacement should go. The second, top-up, modality (figure 1.3b) would likely be simpler to devise, but also more expensive. The third option, which reflects the Yang (2018) model in the United States, involves providing a choice to people on whether to opt for a UBI in lieu of current benefits received, or keep current programs if they provide higher benefits. As shown in figure 1.3c, this proposal seeks to contain costs while enhancing the performance of public welfare agencies via a market-based mechanism. Beneficiaries would become cli- ents with an actual choice of either retaining services, or replacing them with a check.27 FIGURE 1.3  Modalities for UBI Phase-in a. Replace or substitute for select schemes b. Top up existing schemes c. Choice-based or gap filler Benefits Benefits Benefits Income Income Income NOTE: Blue bar illustrates a UBI; gray bars (full) are existing social assistance programs being preserved; gray bars (partial) are existing social assistance programs being replaced. Three Stylized Narratives on UBI Social Protection There is much to celebrate in social protection in recent decades. The remarkable surge in social assistance programming and the accompanying body of impact evaluations docu- ment the significant and diverse impacts that well-designed and -executed programs can have. There is a dramatic advance in the mechanics of implementation and increased ambition for modernity—with e-IDs, e-payments, and dynamic and extensive social reg- istries connecting clients with multiple programs. There are growing ambitions to improve the earnings of the poor with either productive inclusion programs or activation measures 44 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov (Veras Soares and Orton 2017). Some contributory pensions have been reformed to improve their sustainability and equity, and a wave of social pensions is complementing them to cover the informal sector. There are many challenges to individual programs or countries that are still working to achieve good practice in one or more dimensions of their delivery systems or programming. But two challenges pervade social protection for which UBI is proposed as a solution: coverage gaps and fragmentation. In terms of coverage, one starting point for analysis is the legislative architecture or the rights agenda. For instance, under Sustainable Development Goal 1.3, signatories are committed to “implement nationally appropriate social protection systems and mea- sures for all, including floors.” Similarly, the Social Protection Floor initiative endorsed by the UN Chief Executives Board in 2009 calls for an integrated set of social policies to provide basic income security and access to essential social services for all, paying par- ticular attention to vulnerable groups.28 Universal social protection coverage can be achieved via multiple pathways. A com- bination of contributory and noncontributory schemes could lead to a universal social protection system. It could be pursued via the combination of different programs within the same class of interventions, as with multiple social assistance programs. Or it could be realized via a single program within the social assistance family, which is the case of UBI. As to this last, however, as explained above, whether a flat transfer provides adequate social protection against different shocks or states of being (poverty, unem- ployment, illness, disability, old age) and is thereby an effective instrument to deliver the economics right referred to in the Universal Declaration of Human Rights (box 1.5) is a matter open to question. But are the concepts of UBI and Social Protection Floors compatible? A floor, as defined by International Labour Organization Recommendation No. 202, guarantees at least a basic level of income security and access to essential health care.29 In principle, this is not at odds with a UBI. As Ortiz et al. (2018, 5) put it, “a UBI would be the most radical form of the income component of a social protection floor.” Whether a UBI is compatible with the objectives of the floor or not depends on how its design is aligned with the principles of Recommendation 202: if a UBI is designed to wholly replace most of the welfare system, including programs and services for special needs, etc., then it is clearly in contradiction with the floors. However, when a UBI is meant to strengthen and enhance the progressive provision of social protection, then the floors and UBI concepts are aligned. Such alignment occurs, for example, when a UBI is set at a benefit level to ensure at least a basic level of income security, complementary social assistance benefits are preserved for those with special needs, and financing is additional and nonregressive. The political economy of social protection is another hotly debated issue. Universal provisions, some argue, could help reach the “missing middle” and reactivate alliances between the poor and the middle class for demanding broad-based social protection that is politically sustainable30 (Desai and Kharas 2017). Some countries may present a trun- cated welfare redistribution that is stronger at the extremes: the poorest of the poor may be the primary beneficiaries of social assistance, while the upper deciles of the distribu- tion would be sufficiently affluent to afford formal social insurance. In the middle there Chapter 1. The Idea of Universal Basic Income 45 BOX 1.5   Rights Architecture and Select Legislative Provisions S ocial protection is core to the human rights architecture. Article 22 of the Uni- versal Declaration of Human Rights of 1948 firmly states that “everyone, as a member of society, has the right to social security and is entitled to realization, through national effort and international co-operation and in accordance with the organization and resources of each state, of the economic, social and cultural rights indispensable for his dignity and the free development of his personality.” Article 23 envisions that “everyone who works has the right to just and favourable remuneration ensuring for himself and his family an existence worthy of human dignity, and supplemented, if necessary, by other means of social protection.” And, according to Article 25, “motherhood and childhood are entitled to special care and assistance. All children, whether born in or out of wedlock, shall enjoy the same social protection.” The Committee on Economic, Social and Cultural Rights, which monitors implemen- tation of the International Covenant on Economic, Social and Cultural Rights, has progressively developed the content of the right to social protection. Under Article 9, states have an underlying obligation to respect, protect, and fulfill, meaning that they must (1) refrain from interfering with or curtailing the enjoyment of the human right to social security, (2) protect individuals and groups against human rights abuses, and (3) take positive action to facilitate the enjoyment of social security. General Comment 19 asserts the need for sustainable social security systems enshrined in law, as well as obliges states to provide adequate and accessible services in a non- discriminatory manner. And the Optional Protocol, which entered into force in 2013, allows complaints to be received in case of violations of the rights enshrined in the covenant, including violations of the right to social security, which will contribute to international jurisprudence on economic, social, and cultural rights. SOURCES: Universal Declaration of Human Rights, https://www.un.org/en/; International Covenant on Economic, Social and Cultural Rights, https://www.ohchr.org/en/professionalinterest/pages/cescr.aspx universal-declaration-human-rights/index.html.html#at25. would be an entire class, often engaged in low-productivity informal activities, that is too poor for social insurance but possibly not eligible for social assistance. Such a group has been referred to as “the strugglers”; with incomes between US$4 and US$10/day, they constitute about 60 percent of the population in developing countries (Birdsall 2018). UBI is often posited as a vehicle to reimagine public bureaucracies by consolidating the plethora of social assistance programs and streamlining their administration. Indeed, it is hard to imagine financing a UBI without repurposing budgets from some existing programs. But which ones should be replaced? Why? How far should substitution go? Envisioning a UBI replacing programs other than those providing pure income support 46 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov is a much more radical rethinking of public policy than shifting money around between a UBI, a child allowance, or a poverty-targeted transfer. Would a UBI-inspired reform pro- cess replace disability assistance? All types of pensions? Would policy makers entertain the possibility of cutting down social services? What about social workers who provide counseling for substance addictions? Health or education services? Clearly, these questions are as technical as political or ideological. As mentioned earlier in the chapter, a UBI is equally touted as a mechanism to expand the state (Van Parijs and Vanderborght 2017) or erode it (Murray 2016). Hence, the process of pro- gram substitution is the technical manifestation of normative principles—it can to some extent reveal the true nature, purpose, and direction of reforms. At present, social assistance nearly everywhere consists of a large number of individ- ual programs. Data from some individual country inventories show astounding numbers of programs. The World Bank Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) database, which is not set up to capture such detail and focuses on the larger pro- grams in each subclass, shows an average of 21.4 social assistance programs per country, with Chile, Burkina Faso, and Pakistan having 143, 54, and 37 programs, respectively. There are several reasons countries have a multiplicity of social protection pro- grams. One reason is to address different goals—poverty prevention, income smoothing, risk pooling, formation of human capital, increasing labor incomes, etc. Another is to tailor (possibly similar) services to specific groups—for example, programs to improve employment outcomes may focus on youth transitioning to work, the long-term unem- ployed, older workers, migrants, or high-risk youth. In those cases, fragmentation could well be diversification: different measures are intended for specific goals and vulnerabil- ities, thus conceptually justifying them as separate programs. The number of programs per se may not be an adequate indicator of inefficiency so long as interventions complement each other in weaving a coherent portfolio of programs (e.g., by addressing vulnerabilities throughout the life cycle) and are administratively coordinated. Well-connected and -articulated programs may provide a comprehensive system, with wrap-around care for individuals or families with multiple or complex chal- lenges. As discussed in chapter 7, technology and administrative innovations are helping connect different parts of the system in ways that make it more coherent and unified, especially in contexts of high programmatic fragmentation (Barca 2017; Leite et al. 2017; Lindert, George, and Rodriguez-Caillava, forthcoming). However, in many cases, programs may have grown in an uncoordinated manner due to institutional or historical forces not principally aimed at efficient provision of coor- dinated bundles of goods and services. Rather, before a national flagship cash transfer program was developed, individual agencies may have developed one or more sector-spe- cific programs of scholarships, fee waivers, and the like to prevent the poor from missing that sector’s services. Similarly, different levels of government may have multiple programs. This is often noted in urban areas, where national, regional, and municipal programs may overlap in the same block (Gentilini 2015). New governments or protagonists within them may have launched successive initiatives in addition to, rather than as a substitute for, what came before. These forces can truly fragment financing, result in either duplicative Chapter 1. The Idea of Universal Basic Income 47 administrative structures or programs run with insufficiently developed administrative structures, and create a bewildering set of programs for both governments and beneficia- ries to navigate—raising transaction costs and possibly discouraging participation. Few countries have as rich a history of debating a UBI as India. The conversation is a natural extension of the long-standing cash versus in-kind dilemma, including if and how the food-based Public Distribution System, which is nearly universal,31 should be replaced by cash transfers. The UBI debate is largely a reincarnation of that decades-old quandary but is also fed by concerns over fragmentation. We here briefly chronicle the main views and issues emerging from the debate. In addition to recent proposals for a guaranteed minimum income laid out as part of the 2019 electoral campaign, concrete proposals have emerged from Bardhan (2017) and Joshi (2016). Other prominent econ- omists to express support for a UBI include Banerjee (2016), Ghatak (2016), and Ray (2016). Box 1.6 sets out some of the emerging policy proposals. While there is great diversity in the range of UBI proposals, so too is there wide-rang- ing opposition to the idea of an Indian UBI. Aiyar (2017) maintains that a UBI could lead people to “sit at home and play video games” or “get involved in undesirable activities,” “create incentives for having more babies,” and “attract millions of immigrants from Bangladesh and Nepal, mostly illegal.” Drèze (2017) cautions against a UBI based on inflation, the multiplicity of objectives pursued by in-kind transfers, and delivery dimen- sions (we return to some of these issues later in this chapter). For example, Drèze cites the “sobering experience” of delayed and failed cash transfers in the context of National Rural Employment Guarantee Scheme (NREGS) wage payments and Public Distribution System cash transfer experiments. Ghosh (2017) fears that a UBI will erode the welfare state by “moving out of essential public service delivery, essentially reneging on its con- stitutional obligation to ensure the social and economic rights of citizens.” UBI skeptics argue that there are many more urgent claims for government spending than a UBI program. Aiyar (2017) claims that “the cure to poverty lies in improving state capacity and public goods.” Opponents also contend that these more valuable government projects will be threatened by a UBI because the proposed financing mechanisms outlined in the current set of UBI proposals are neither fiscally nor politically feasible.32 Even if the fiscal calculations for UBI added up on paper, many contend that the political economy makes the proposed savings unachievable in practice (Khera 2016). Removing subsidies from the middle and upper classes is notoriously difficult to achieve, to the point that the Economic Survey (see box 1.6) virtually discounts this as a viable option (Khosla 2018). Chapter 6 presents more detail on the political economy dimensions of a UBI. Automation and Labor Market Disruptions The narrative around labor markets includes three main subdebates: automation, stag- nant and low wages, and the changing (or unchanging) nature of work. We here review all three briefly, noting first that because these proposals are most linked to diminished optimism that workers can earn sufficient independent incomes, they often focus bene- fits on adults and do not include children. The possible labor market rationale for a UBI 48 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov BOX 1.6  UBI Proposals in India T hree UBI (or quasi-UBI) proposals have been detailed for the Indian context. These proposals differ substantially in how they envision a UBI would be imple- mented in terms of the size of the proposed benefit, the estimated program cost, and the intended sources of financing. •• The Finance Ministry’s Economic Survey 2016–17. The survey (Government of India 2017) recommends providing 75 percent of the population with monthly transfers amounting to Rs 7,620 (US$120) per person per year. At an estimated cost of 4.9 percent of GDP, the survey acknowledges that the quasi-UBI scheme would be “fiscally unaffordable” unless it replaced some existing welfare pro- grams, but stops short of recommending which programs to cut. A gradualist approach of phased implementation is recommended, starting with partic- ular target populations such as women, vulnerable groups, or urban areas. Even with this tentative strategy, the survey recognizes that many adminis- trative, political, and financial challenges remain, concluding that “UBI is a powerful idea whose time even if not ripe for implementation is ripe for serious discussion.” A similar proposal by Felman et al. (2019), limited to rural popula- tions, calls for the provision of Rs 18,000 per year. •• Bardhan proposal. An early advocate of an Indian UBI, Berkeley econo- mist Pranab Bardhan first proposed a UBI in 2011, and has since updated his proposal with revised figures (Bardhan 2017). He proposes a truly universal pro- gram and advocates more generous entitlements than does the Economic Survey, amounting to Rs 10,000 per person per year (indexed to inflation). He believes an effective UBI could be delivered alongside increases in health, education, and infrastructure expenditure without replacing existing social programs, such as the Public Distribution System and the National Rural Employment Guarantee Scheme. Rather, 10 percent of GDP would need to be reclaimed from other sources. Bardhan argues that this fiscal space could be created by (1) eliminating “nonmerit” fuel, fertilizer, water, electricity, and rail subsidies, which he estimates as costing 5 percent of GDP; (2) cutting certain forgone revenues, mainly corporate tax holidays and exemptions, to release around 2 percent of GDP; and (3) raising what he terms “absurdly” low or non- existent areas of taxation (such as real estate, long-term capital gains, and inheritance) to counter rising inequality. •• Joshi proposal. Relative to the Bardhan and Economic Survey proposals, that of Oxford economist Vijay Joshi (2016) is far more conservative, proposing a (continued) Chapter 1. The Idea of Universal Basic Income 49 BOX 1.6  UBI Proposals in India (continued) universal basic income supplement set at 20 percent of the Tendulkar pov- erty line threshold—the gap between the average poor person’s income and the poverty line threshold. This would amount to Rs 3,500 per person per year, at a fiscal cost of 3.5 percent of GDP. Joshi argues that this could be paid for through a broad set of fiscal reforms. Specifically, he estimates that 10 percent of GDP would be freed up by eliminating inefficient price subsidies (including Public Distribution System food subsidies), cutting unproductive tax exemptions, taxing high agricultural incomes, and pursuing more vigorous pri- vatization programs. Joshi acknowledges the political economy challenges of his proposed fiscal reforms, but notes that such challenges are a poor reason to abandon the policy altogether. A number of other economists have offered UBI recommendations for India as well. Ghatak (2016) and Banerjee (2016) have argued for UBI benefits of around Rs 13,000 per person per year, stressing the need for basic income benefits to equal the Tendulkar poverty line threshold. Universal provision of this benefit would cost approximately 11 percent of GDP, which Ghatak argues would require raising taxes and expanding the tax base, as well as cutting the nonmerit subsidies Bardhan and Joshi suggested scrapping. Ray (2016) has taken the debate in a somewhat differ- ent direction, proposing a “universal basic share” in which a fixed fraction of India’s GDP would be permanently committed to universal income provision. Figure B1.6.1 provides an overview of the main proposals currently being discussed in India. FIGURE B1.6.1  Comparison of India’s Select UBI Proposals Rs 18,000/year 20,000 (Rural only: top 25% excluded) 15,000 Rs 13,000/year Rs 13,000/year Rs 10,000/year (sequenced) 10,000 Rs RS 7,620/year (sequenced) Rs 3,500/year 5,000 (supplement) 0 Felman et al. Banerjee Ghatak Bardhan Gov’t of India Joshi 2019 2016 2016 2017 2017 2017 50 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov is of central importance for several reasons, including because jobs are a key vehicle for poverty reduction as well as in shaping deeper societal identities. Box 1.7 lays out some considerations on the meaning and role of jobs. Technology has long been a source of anxiety, especially in Western societies. These fears are well grounded. Since the Industrial Revolution, workers have coexisted with the threat of an ever-growing machine presence in agriculture, manufacturing, and service jobs. In fact, the corporate labor share of employment declined steadily between 1975 and 2012, including in about three-quarters of advanced economies and two-thirds of developing countries. BOX 1.7  Jobs and Societies J obs are more than income. Employment can shape mental and psychological welfare: evidence shows that jobs can affect individuals’ identity, self-worth, and sense of purpose, as well as bolster civic engagement and broader social cohesion. Generating jobs that fulfill the monetary and nonmonetary dimensions of well-being for all those willing and able to work should be a core societal aim. Jobs are the premier vehicle for poverty reduction. Econometric studies show that sectoral employment intensity plays an important role in turning growth into pov- erty reduction—that is, when growth generates jobs for most of the labor force, particularly the poor. For example, in 10 of 18 Latin American countries—as well as in Germany and the United States—more than half of past poverty reduction was due to jobs (World Bank 2013). But many jobs may not provide adequate income, can be frustrating or degrading, or even generate negative externalities and forms of exploitation. Low-productiv- ity, low-quality jobs can keep people in poverty: between 54 and 63 percent of workers in Africa and Asia live on less than US$2/day. In other words, most poor people are poor workers. The gulf between aspiration and reality—between a good job as the first-best option and the lower quantity and quality of jobs available in practice—has sparked long-standing debates on the role of social protection in helping bridge the gap. A range of supply-side interventions have been devised for enhancing people’s employability and reducing their distance from labor market demand. Ensuing investments in human capital, skills, and intermediation have been and will remain key in the future. But against these efforts, labor demand has not always expanded, nor has the quality of jobs necessarily improved. This impasse has led to exploration of new avenues, including, among others, job guarantee programs and UBI itself. Chapter 2 discusses job guarantees in more detail. SOURCES: Bonnet, Vanek, and Chen 2019; Furman 2019; World Bank 2013. Chapter 1. The Idea of Universal Basic Income 51 Will this trend continue in the future? Most likely so, although the magnitude of future automation or worker-machine substitution is unclear. For instance, for a pool of countries, the share of jobs susceptible to automation range from the low single digits to about half of the population (World Bank 2018b). Job losses should be balanced against the generation of new jobs in innovative sectors that an appropriately absorbed and managed technology can spur (Acemoglu and Restrepo 2019). Whether the net job balance is positive or not, substantial shares of workers may be at risk of technology-in- duced unemployment and may not easily transition to newly generated occupations. It is in this context that the idea of a UBI is gathering steam as insurance against such risk (Standing 2017; Yang 2018). It is in response to concerns around technological change that a number of the pilot programs discussed in this book are being launched. Tests in the United States that are privately funded, including with support from the tech indus- try, tend to fall in this category (e.g., Oakland pilot by Y Combinator; see appendix A). From this perspective, views on UBI hinge on how serious the technology, automa- tion, or artificial intelligence–related threat is and, therefore, how proportionate or radical the policy response should be. Two quotes may help crystallize those competing views. [Artificial intelligence] does not call for a completely new paradigm for economic policy—for example, as advocated by proponents of replacing the existing social safety net with a UBI—but instead reinforces many of the steps we should already be taking to make sure that growth is shared more broadly. To date, in fact, the problem we have faced is not too much automation but too little automation. (Furman 2019, 317) UBI has the potential to give our troubled economy a twenty-first-century shot in the arm by transforming the technological distribution that’s been causing so much anxi- ety into a force for self-fulfillment and the common good. (Stern 2016, 171) In a way, those who do not see the automation threat as warranting a UBI response see the latter quote as a declaration of surrender to technology—that is, that society has been unable to manage technology in a way that would enhance and innovate how people work, instead of undermining the concept of labor itself. Conversely, among those who believe such a tipping point has already been passed, a UBI emerges as a soci- etal stronghold against an inevitable technological tsunami. The somewhat polarized discussion on automation compounds other threads of discontent with respect to labor markets and social protection. One thread stems from the observation that, while employment is a key conduit for poverty reduction, jobs in low- and middle-income countries may not always exert such a lifting role given their low pay and low productivity.33 Relatedly, in high-income countries, wages can be flat rel- ative to overall productivity increases. In both cases, UBI is cited as a way to supplement chronically low earnings (Standing 2017). From another perspective, prevailing social insurance models were designed assuming a single, stable, full-time employment relationship. Part-time and temporary workers are typically insured in the same way as standard workers as long as they meet the income and minimum contribution requirements. But the self-employed, those who 52 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov often switch jobs, or those combining self- and dependent employment do not easily fit into the framework of contributory social protection systems. These forms of employ- ment pose questions for social protection.34 As underscored by the Organisation for Economic Co-operation and Development, “rising numbers of nonstandard workers may also erode the effectiveness of social protection systems. If taxes and social contributions are payable only for some categories of workers, firms have an incentive to shift work to workers who are less protected and less expensive” (OECD 2018, 1). These hurdles have generated interest in simple, more uniform provision of social protection that encom- passes all workers independently of how and where they work, potentially such as a UBI. Trends in nonstandard employment (temporary and part-time employment, tri- angular agency work, and “disguised” employment) are more mixed in middle-income countries. In Latin America and the Caribbean, the prevalence of such forms of employ- ment has been generally stable over the last two decades, while less homogeneous in Eastern Europe and Central Asia (Apella and Zunino 2018). Yet in a number of developing countries, the core challenge to the Bismarckian model is not necessarily posed by the diversification of labor markets and automation, but by pervasive informality. In fact, 81 and 46 percent of workers hold informal jobs in low- and upper-middle-income countries, respectively (World Bank 2018b). Given the endemic nature of the challenge and the slow progress against it, new analytical work argues that most people would be better off with a social protection system that does not depend on their work situation. This evolving thinking has sparked new ideas on how to conceive of social assistance that ensures an equity function for large swaths of the population, complemented by subsi- dized insurance against catastrophic losses, while keeping contributory social insurance for consumption-smoothing purposes. A UBI may be among the options to form such a social assistance platform, taking the pressure off social insurance in pursuing distributional or equity goals (Packard et al. 2019; Rutkowski 2018; World Bank 2018b). Resource Dividends A thread in the UBI discourse known as the dividend model has been the subject of ana- lytical attention and practice, including at national and subnational levels. The model generally involves contexts benefiting from large resource windfalls from oil and com- modities. In a number of cases, there is limited public awareness of the level of such revenues and their use. Also, those windfalls tend to coexist with vast amounts of regres- sive subsidies (Devarajan 2018). To address this dual problem of inefficient subsidies and state-citizen accountability, policy proposals have suggested redistributing part of the oil revenues to the entire population in the form of a UBI and progressively taxing it back to finance public goods. Taxation provides an incentive for citizens to demand better services by the state, activating a virtuous cycle of more progressive assistance while reigniting citizen scrutiny of state services (Devarajan et al. 2013; Sandefur 2017). Two country examples epitomize this model. Mongolia is the only country that has ever had a full-fledged UBI: the program was introduced over the period 2010–12, before being eventually scaled back due to fiscal constraints. The Islamic Republic of Iran’s energy-related Chapter 1. The Idea of Universal Basic Income 53 subsidy reform presents an intriguing variant of a resource-rich dividend scheme. In 2011, energy subsidies were replaced by cash transfers first reaching the full population and then later, as a targeted subsidy, reaching about 90 percent of the population. The redistribution of natural resources may not necessarily be motivated by effi- ciency gains, but by social and environmental principles. The United States is illustrative of two such examples. From a social perspective, the idea is to strengthen the social fabric by engendering a sense of common purpose, with proceeds from natural resources being a form of shareholder dividend for citizens. This is a rationale in line with Thomas Paine’s seminal vision. Current models in Alaska and the Eastern Band of the Chero- kee Nation epitomize this approach (Akee et al. 2018a, 2018b; Moss 2012; Widerquist and Howard 2012). The Alaska Permanent Fund is designed to redistribute oil royalties to residents, while the Eastern Band of the Cherokee Nation is related to casino profits: since 1997, the tribal government has provided a portion of its dividends to 16,000 adult tribal members. The average annual dividend is approximately US$4,000 per capita, which is subject to federal taxation and split into two payments per year. In the next sub- sections, we will briefly review large-scale, natural resource–related experiences from Alaska, Mongolia, and the Islamic Republic of Iran. From an environmental perspective, there are several proposals for a tax-and-divi- dend carbon policy.35 If a carbon tax is proposed for its climate benefits, the question of what to do with its revenues arises. In cases where carbon tax revenues are distributed on a flat per capita basis, they could become a major vehicle for redistribution. This pro- vides a clear link between the UBI and the climate change agenda. Alaska’s Permanent Fund Dividend In 1967, the state of Alaska experienced a sudden windfall of oil wealth when North America’s largest oil reserve was discovered on state-owned land. The lease sold for a staggering US$900 million, seven times the state’s yearly budget. Led by Governor Jay Hammond, a 1976 amendment to the state’s constitution required the state to deposit at least 25 percent of each year’s natural resource revenues in an Alaska Permanent Fund. The fund revenues are invested in a savings account, with part of the interest paid annu- ally to residents as a resource dividend. Since 1982, every resident has been eligible for an annual UCT from the state. Individuals must apply each year, meet the residency criteria (be present in Alaska for the preceding year, with the intent to remain indefinitely), and have no recent serious criminal conviction. More than 90 percent of the population typically receives the divi- dend. Unlike a true UBI, the amount is neither stable nor sufficient to meet basic needs: dividends typically do not exceed 7 percent of the average annual income of recipients, with inter-year fluctuations in transfer amounts between −61 percent and +9 percent, or 19.8 percent on average (figure 1.4). The total dividend distribution has historically amounted to half of the fund’s annual interest (averaged over the previous five-year period), but this was recently reduced to allocate more of the fund’s earnings to the state’s large deficit. 54 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov FIGURE 1.4  Alaska Permanent Fund Dividend Beneficiaries and Annual Percentage Change in Dividends, 1982–2017 700,000 120 100 600,000 Annual change in dividend amount (%) 80 Numbe rof beneficiaries 500,000 60 40 400,000 20 300,000 0 200,000 –20 –40 100,000 –60 0 –80 1982 1985 1990 1995 2000 2005 2010 2015 2017 Number of beneficiaries Percentage change in annual dividend amount SOURCE: Alaska Permanent Fund dividend time series data (https://pfd.alaska.gov/). The Alaska Permanent Fund dividend has attracted significant public support across demographic, socioeconomic, and political divides. For instance, current public opinion favors raising taxes over ending the dividends (Isenberg 2017). The program has lowered poverty and inequality levels to among the nation’s lowest and stimulated the economy, generating over 7,000 jobs and US$1.1  billion in personal income, without producing inflation or reducing employment. In fact, recent evidence shows part-time work has increased by 1.8 percentage points, or 17 percent (Jones and Marinescu 2018). While the dividend may have not affected fertility rates, it has acted as a mild magnet for at least 12,000 migrants (Goldsmith 2012). The Islamic Republic of Iran’s Compensatory Cash Transfer Program The Iranian UBI scheme was born out of wider reform packages. In 2008, the government announced a set of sweeping reforms in energy and food subsidies. A compensatory means- tested UCT was considered. However, public opposition to means testing was growing. As a result, the targeting plan was abandoned, replaced instead with uniform universal cash trans- fers with the rich being discouraged from participating. In January 2010, Parliament approved the subsidy reform package and preparations were launched to publicize the changes, pre- pare the bank infrastructure, ensure universal account access, and reduce inflation ahead of the price increases (Salehi-Isfahani and Mostafavi-Dehzooei 2018; Tabatabai 2012). On December 19, 2010, the universal cash transfer was deposited in the bank accounts of household heads. At the same time, domestic energy and agricultural prices rose by up to 20 times (Reza Farzin, Guillaume, and Zytek 2011). At first, only 80 percent Chapter 1. The Idea of Universal Basic Income 55 of households registered for the scheme, a share that quickly rose to 96 percent. Efforts to exclude the rich were put in place, but coverage remains at around 92 percent of the population.36 The monthly cash transfer was set at Rls  455,000 (US$40–US$45) per person—29 percent of median per capita income, and nearly three times the amount envisaged in the subsidy reform law. The program absorbs nearly 3 percent of GDP. The combination of subsidy reforms and sanctions triggered inflation rates that eroded the transfers’ purchasing power by two-thirds of their original value by 2018. Even so, the program brought about promising results. For example, empirical studies find over- all no negative labor supply effect (on hours worked and participation); but the youth worked a bit less (mostly because they were enrolling in higher education), while service workers worked more (36 minutes/week) (Salehi-Isfahani and Mostafavi-Dehzooei 2018). Mongolia’s Two-Year Experience with UBI Mongolia began enjoying a major mining boom in the early 2000s, with growth rates approaching 9 percent of GDP over 2002–08. In the run-up to the 2008 elections, politi- cal parties competed over promises for cash transfers for the population. A new Human Development Fund was established in 2009 to “create and grow sustainable permanent resources to collect and evenly distribute them” (Yeung and Howes 2015). A key element was to universalize the preexisting categorical child grant program, entitling all citizens to a share in the nation’s mineral wealth. The first universal cash transfer budget was set at Tog 324 billion, with transfers equat- ing to Tog 120,000 (US$89) per citizen over the year. Unlike in Alaska, this amount was based on electoral promises rather than the actual resource revenues accumulating in the Human Development Fund. This left the program financing vulnerable to commodity price drops. Eventually, this was exactly what happened, with copper prices plummeting and slowly recovering (figure 1.5). After the first disbursement of Tog 70,000 (US$52) per citizen in February 2010, payments quickly fell behind schedule, prompting protests. Attempts by the government to negotiate alternative provision of social welfare services failed, and from August to December 2010, monthly transfers amounted to only Tog 10,000 (US$7), which was principally financed through government loans, since the actual mining revenue in the fund was only half the annual cash transfer expenditure. From January 2011 to June 2012, monthly transfers continued to be paid out at Tog 21,000 (US$17) per person, with the gov- ernment constantly depending on borrowed funds to top up the mining revenue shortfall. The program still managed to achieve some significant results: it reduced poverty by up to a third, lowered inequality by up to 13 percent, and provided Mongolia with the highest financial inclusion level among middle-income countries. However, the scheme came with a hefty, ultimately unaffordable, price tag. Public external debt soared from 31 percent of GDP in 2010 to 48 percent in 2012. Cash transfers may also have contrib- uted to inflation, although this remains untested. The program quickly lost public and political support and was replaced, in June 2012, with the original child-targeted transfer scheme. The Human Development Fund was replaced by a new sovereign wealth fund focused solely on savings and investment.37 56 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov FIGURE 1.5  Global Copper Prices and Universal Basic Income in Mongolia a. Copper prices 5 4 US$/pound 3 2 1 0 1/3/05 1/3/06 1/3/07 1/3/08 1/3/09 1/3/10 1/3/11 1/3/12 1/3/13 1/3/14 b. Cash transfers 3.0 6 5.5% 2.5 5 4.1% Fiscal cost (% of GDP) Coverage (millions) 2.0 4 3.4% 3.0% 1.5 3 1.0 1.5% 2 1.4% 1.0% 0.5 0.7% 1 0.0 0 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 SOURCES: Based on Yeung and Howes 2015; copper prices are from Macrotrends (https://www.macrotrends. net/1476/copper-prices-histor-ical-chart-data%E2%80%99). Conclusions This chapter sheds light on some major queries in the global UBI debate. We here group those issues around definitional matters, scope of the program, experiences to date, as well as a set of strategic and programmatic choices. In general, our take is that, at least for the moment, a UBI should be taken seriously, but not necessarily literally. What is a UBI? The debate on a UBI is often chaotic and without precise defini- tional contours. In many instances, a UBI is equated with guaranteed minimum income schemes, while others define a universal program as one that does not establish any Chapter 1. The Idea of Universal Basic Income 57 eligibility criteria other than age. We propose a definition of UBI based on three core design choices—that it is paid to all, unconditionally, and in cash—and discuss other important features such as its amount and frequency, and whether children or noncit- izens would benefit. A UBI is one particular pathway to achieve universality in social protection, although differences exist on what “universality” actually means: should uni- versality be interpreted as an outcome (e.g., “everybody should have a basic level of income”), or in terms of coverage (e.g., “everybody should be covered by social pro- tection”)? The term “coverage” itself is interpreted differently in social assistance (i.e., coverage as receipt of transfers) and social insurance (i.e., a promise of help if risks materialize). These differences have important implications for the “targeting versus uni- versality” debate. What problem would a UBI try to solve? There are three different narratives in this regard, which entail diverse designs and expectations. These are a desire to improve social protection; the labor market story (interpreting UBI as insurance against artificial intelligence–induced unemployment or diversified labor markets); and the redistribution of natural resource revenues. The objectives of UBI are not only diverse, but sometimes incompatible. For instance, a UBI linked to natural resource dividends may not fully align with rights-based narratives. Similarly, UBI is sometimes interpreted as a backbone for building stronger states; others look at it as a stepping stone to dismantling them. In a 1967 article, Friedman referred to a negative income tax—which shares a number of similarities with UBI—as “…the only practicable route so far proposed for dismantling gradually but thoroughly the jerry-built structure of government interferences with the market and with individual liberty that have been adopted in the name of welfare… The Left, if it accepts the program, will find that it has bought a Trojan Horse” (Friedman 1967). A UBI is a means, not a goal—a UBI is what countries make out of it. A UBI is often associated with being a game changer in power redistribution (Stern 2016). Such calls may resonate and amplify preexisting perceptions of unfairness and inequities that are creeping into the foundations of social contracts (Cottam 2019). There- fore, a UBI may seem to offer a crisp, tangible way of meeting the appetite for change. But the generators of inequities may lie elsewhere; for example, in uneven access to education and health systems, low-paying and low-productivity jobs, poorly function- ing markets, corruption, regressive tax codes, unequal pay, and social discrimination, among others. A UBI could help, but despite its scale, may not move the needle: follow- ing Piketty (2016), “the problem with the discussion about basic income is that in most instances it leaves the real issues unexplored and in reality expresses a concept of social justice on the cheap.” This is not to downplay the societal role of UBI, but to set expecta- tions right. In this spirit, the chapter discusses the kinds of bottlenecks (e.g., in terms of factors hindering coverage) a UBI may likely help address, as well as others that it may not—or may even amplify. What is the global experience and evidence around UBI? No country currently has such a scheme in place, and only two have done so temporarily (Mongolia and the Islamic Republic of Iran). Those experiences offer some helpful insights into core ques- tions, such as financing and inflation, while pilot programs are generating information 58 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov on one or more defining features of a UBI. Yet systemwide issues are largely left unan- swered, such as the relationship to the minimum wage, severance pay, or pensions. Given current social assistance practices, the move toward an unconditional cash- based transfer is plausible, though the extensive in-kind and conditioned programming suggests that it may not be without controversy. The leap to universality within a single social assistance program is definitely more radical. To date, attempts at universality within social protection have largely been confined to social insurance. Just as the global proliferation of cash transfers was once unimaginable, the prospect of providing cash for everyone should not be ruled out. As a radical solution, a UBI is bound to be thought-provoking. But when design and financing details are laid out, it may be less extreme than envisaged, including turning into a targeted program. And by sometimes playing a strawman role, a UBI tends to expose the limits of current social protection systems (Ravallion 2017). How to bring about improvements is perhaps the most fundamental question underlying the UBI debate—that is, should countries build upon, improve, and extend what exists in their social protection systems or should they introduce a radical new approach? And if they do, do they risk “throwing the baby out with the bathwater”? These dilemmas entail the consideration of systemwide principles and programmatic choices. At the systemwide level, there is broad demand for making the overall social protection system inclusive, progressive, and adaptive. A UBI would score high in terms of inclusiveness, as it would not differentiate among people in a polity; but the lack of differentiation is precisely what makes the effects of a UBI uncertain on progressivity. As chapter 4 shows, it is important to understand the overall distributional effects of a UBI, but also to pay particular attention to the effects on those at the bottom of the distribu- tion. A UBI would likely be a rigid instrument that may not fully adapt to a diversified set of circumstances, especially in lower-income contexts (e.g., in terms of market con- ditions, etc.). At the programmatic level, the choice would depend on the statement of the prob- lem that UBI is intended to address—for example, whether technological advances will eventually result in massive net job losses—and how well systems or a particular pro- gram (e.g., UBI versus targeted cash transfers or versus regressive energy subsidies) are working against that objective in a given context (Coady and Prady 2018; Harris et al. 2018). Assessing the appropriateness and feasibility of UBI requires understanding and working though the comparative trade-offs that any program or set of programs face in terms of coverage, progressivity, adequacy, incentives, costs, financing options, political economy, and delivery (i.e., the framework laid out in this volume’s overview, and that this chapter has helped inform). None of these parameters have an easy and predeter- mined outcome (Francese and Prady 2018; Ravallion 2018). A UBI is a seemingly simple idea that involves complex choices. It is our hope that this chapter has helped unbundle and navigate this complexity and, alongside the other chapters in this book, helps in making informed choices. Chapter 1. The Idea of Universal Basic Income 59 Notes 1. These include such films as Bootstraps (https://www.bootstrapsfilm.com/) and Free Lunch Society (http://www.freelunchsociety.net/). 2. Ricardo Anaya, candidate for Mexico’s presidential elections in July 2018, championed a UBI. Andrew Yang, a Democratic presidential candidate for the 2020 elections in the United States, is doing the same. 3. Included among its proponents are Richard Branson, Elon Musk, and Mark Zuckerberg. See AgreeList website, “Tracking Influencers’ Opinions,” https://agreelist.org/a/basic-income. 4. In 1795, a specific variant introduced in Speenhamland, Berkshire, established the first poverty line consisting of three “gallon loaves” of bread per week per adult (plus adjustments by house- hold size). Wages were topped up to ensure that family income reached the poverty line, while the unemployed received full payment. This is often referred to as a UBI scheme, while it actu- ally was a minimum guarantee program. See chapter 2 for a discussion of such interventions. 5. In the United States, the antipoverty system was largely modeled after the Old Poor Law until the Great Depression, when the New Deal built federal institutional capacities for social pro- tection. Large-scale public works implemented in the early 1930s played a key role in moving away from local Old Poor Law relief schemes and weaving a national safety net. These devel- opments would pave the way for contemporary federal programs, such as the Supplemental Nutrition Assistance Program (SNAP; food stamps) of 1963 and the earned income tax credit of 1975. 6. This coverage is for pensions only, not for insurance related to health or work injury. For example, India’s Ayushman Bharat–PMJAY scheme covers about 500 million people, 40 per- cent of the country’s population (Blake et al. 2019). 7. Another country, Iraq, has a truly universal social assistance program—the Public Distribution System—which is close to a UBI, but provides transfers in kind (Krishnan, Olivieri, and Rama- dan 2018). 8. For an elegant treatment of the analytics of negative income tax programs, see Tondani (2009). 9. Preliminary results show that the experiment did not increase employment but did augment the well-being (health and psychological status) of the treated unemployed (Kangas et al. 2019). 10. A puzzling example is the case of the municipality of Maricá in Rio de Janeiro, Brazil. According to some accounts, the municipality extended the preexisting targeted unconditional cash transfer to all its residents. The scheme is supposedly financed by oil revenues and covers about 150,000 people. However, as part of this book’s research, we found that the scheme is not yet active and that a modified version will soon be launched (Moreira 2019). See also https://www.vox.com/ future-perfect/2019/10/30/20938236/basic-income-brazil-marica-suplicy-workers-party. 11. In New York City, for example, the recertification process for SNAP requires up to 17 different types of documents (Homonoff and Somerville 2019). 12. A related point is the possible ability of universal programs to avoid political clientelism (vote buying), since there is less room for politicians and officials to influence program participation. In the context of health provision in the Philippines, for instance, Khemani (2013, 22) finds that “…vote buying is systematically negatively correlated with a particular type of service with the following characteristics—that which is the exclusive responsibility of municipal governments, that which is pro-poor (services which the rich do not use), and that which is a relatively broad, untargeted service, not particularly amenable to narrow targeting to select citizens.” 60 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov 13. In terms of political clientelism, this may hinge on the competitiveness of the political system, the level of poverty and size of the middle class, and the specific entry points for operational manipulation by local-level officials. Evidence from Argentina’s experience, for instance, shows that where political competition is high, clientelism creates an electoral trade-off: poli- ticians may gain votes among the poor, but at the cost of support from the middle class; thus, high competition is compatible with clientelism where poverty is high, but should decrease where poverty is low or in less-competitive contexts (Weitz-Shapiro 2012). And if the param- eters cited by Khemani (2013) are considered, a UBI would be managed centrally, not locally, and may be sufficiently attractive to nonpoor populations. 14. The incidence of poverty, and hence the gap with safety net coverage, is even higher if national and relative measures of deprivation are used (Ravallion 2019). 15. More recently, to rectify the undercoverage, the Argentine government initiated nearly a million eligibility processes mostly based on improved national electronic data matching protocols and outreach efforts in communities where civil registration is incomplete. For instance, the government has linked 13 public databases and distinct ID registries for a sav- ings of US$104 million in reduced leakage and tax evasion (World Bank 2016). 16. For example, surveys of applicants for SNAP means-tested vouchers in Colorado and Illinois found that among SNAP applicants who were working, 15  percent lost pay because they missed work to visit the social services office. Furthermore, when asked which customer ser- vice improvements were most important to them, “being treated more respectfully” was the top priority for up to 17 percent of survey respondents; similarly, “getting a better explana- tion of how to apply for benefits” was the top priority for improvement for 11 to 22 percent of survey respondents, depending on the state. Also, among SNAP applicants facing emer- gencies or problems such as job loss, lack of sufficient food, loss of housing or care, and emergency medical needs, 60 percent reported that these situations could have been avoided or mitigated if they had received benefits more quickly (Isaacs, Katz, and Amin 2016). 17. These debates are not limited to social protection. For instance, for a discussion on the debate of generalized and targeted programs in education, see Evans and Yuan (2019). 18. Although this section focuses on conditional cash transfers, these are not the only conditional programs. Interventions such as school feeding, for instance, are long-standing and ample in coverage. The latest available data show that these are similar to CCTs in covering the poorest quintile (40.3 percent for CCTs, 37.1 percent for school feeding). Yet the debate on conditional in-kind transfers versus their unconditional form is not as widespread or contentious as is the case of cash. 19. See Currie and Gahvari (2008) for further details on paternalism and interdependent prefer- ences. 20. While these philosophical, political, and economic issues find their apex in CCT debates, they also permeate the discourse around certain types of public works programs and, to some extent, in-kind transfers. We examine the latter in the next section, and return to public works in the context of job guarantee programs in chapter 2. 21. According to microeconomic models, an inframarginal in-kind transfer and a cash transfer of equal value would have the same effect in bolstering household food consumption—that is, beneficiaries’ marginal propensity to consume food out of an additional income from an in-kind or cash transfer should be the same. Put another way, there is only an income effect and no price effect associated with inframarginal transfers. For a broader and thought-provok- ing reflection on the economics of giving, see Reinhardt (2013). Chapter 1. The Idea of Universal Basic Income 61 22. See The Economist (2011): https://www.economist.com/briefing/2011/03/10/throwing-money- at-the-street. For more detail, see https://www.reddit.com/r/BasicIncome/comments/2ioovd/ kuwait_gave_almost_4000_to_every_citizen_in_2011/. 23. See Economic Security Strategy Bill 2008, https://www.aph.gov.au/binaries/library/pubs/ bd/2008-09/09bd063.pdf. 24. At the time of finalizing this volume, Jones and Marinescu were preparing a paper assessing the possible inflationary effects of cash dividends in Alaska (Ioana Marinescu, personal com- munication). 25. In the United States, McGovern’s proposed “demogrant” in the 1970s was designed to give each person US$1,000 a year (about US$5,700 in 2016 dollars) to be funded by general taxes. In 2000, Ackerman and Alstot suggested that every American receive a one-off grant of US$80,000 on his or her 21st birthday, or at age 18 for those enrolling in college. This was to be funded out of a dedicated annual tax of 2 percent levied on the wealthiest Americans. 26. See Lokshin and Ravallion (2019) for a discussion on the missing market of working permits and how that compares to a UBI. 27. Other proposals for introducing a UBI have focused on linking UBI transfers to country GDP, somewhat mirroring Alaska’s resource dividend model. Such proposals have been set forward for India by Ghatak and Muralidharan (2019) and Ray (2016). 28. Similar commitments were reiterated at the Third Financing for Development Conference in Addis Ababa in July 2015. In the following year, the World Bank and the International Labour Organization issued a declaration on “Shared Mission for Universal Social Protection,” whereby the organizations set out a vision of a world where “anyone who needs social pro- tection can access it at any time,” against which an objective was set to “increase the number of countries that can provide universal social protection, supporting countries to design and implement universal and sustainable social protection systems.” 29. See the International Labour Organization website, https://www.ilo.org/dyn/normlex/en/f?p= NORMLEXPUB:12100:0::NO::P12100_ILO_CODE:R202. 30. When protracted over time, the lack of services from the state can lead to a vicious cycle of dis- illusionment whereby those, often large, sections of the population unreached or poorly served by the state are also the least likely to demand better welfare provisions (Holland 2017). 31. In some states, the Public Distribution System is entirely universal, as in Tamil Nadu and Himachal Pradesh (see Drèze and Khera 2017, table 7). 32. Looking at the UBI calculations, Drèze and Khera (2017) show that the amount created by removing nonmerit subsidies would be far lower than Bardhan (2017) and Joshi (2016) origi- nally suggested, since their initial estimates of 8–9 percent of GDP were based on 20-year-old data from the National Institute for Public Finance and Policy. Although Bardhan’s proposal used newer 2011–12 data estimating nonmerit subsidies to cost 5 percent, IMF (2017) anal- ysis suggests that this is still highly optimistic, with the actual figure likely only 2  percent. Former finance minister Chidambaram also questions the fiscal validity of Bardhan’s and Joshi’s proposals to reclaim “revenues foregone,” highlighting the net economic damage that could result from reversing policies on special economic zones and specified infrastructure investments. 33. Examining data from 28 low- and middle-income countries, Bonnet, Vanek, and Chen (2019) show that, on the one hand, in most countries the poverty rate of informal workers is between 62 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov 2 and 10 times higher than that of formal workers; on the other hand, anywhere from 50 to 98 percent of poor workers are informally employed. 34. In the case of self-employment, for example, social insurance faces challenges because of uncertainty as to who is liable for employers’ contributions (having the self-employed pay both employer and employee contributions is unrealistic for many, as self-employed earnings are typically volatile). Also, the self-employed often have fluctuating earnings, with contribu- tors struggling to pay in bad years; and unemployment insurance for the self-employed raises moral hazard issues, as it can be difficult to assess whether they are in fact involuntarily unemployed. 35. See for example https://www.econstatement.org/. 36. Other countries have been more successful in promoting voluntary deselection from social assistance. For example, India’s Give It UP campaign, implemented as part of the liquid petroleum gas–related cash program reaching 177 million people, successfully promoted the exclusion of about 10 million wealthy individuals through a mix of public initiatives geared toward recognizing the gesture (online “champions and beneficiaries” with names published in the Ministry of Petroleum website, etc.). These efforts were underpinned by communica- tions to the affluent and the middle class on “nation building” and connecting to greater social goals, as well as bringing information and data on “unfair distribution” to the forefront. The program also established exclusion criteria based on legally binding self-declaration. Savings from these policies amounted to US$332 million/year (Gelb and Mukherjee 2019). 37. The Child Money Programme now covers 80 percent of households and proxy means testing is de facto used as an affluence test. References Acemoglu, Daron, and Pascual Restrepo. 2019. “Automation and New Tasks: How Technology Displaces and Reinstates Labor.” NBER Working Paper 25684. National Bureau of Economic Research, Cambridge, MA. Ackerman, Bruce, and Anne Alstott. 2000. The Stakeholder Society. New Haven: Yale University Press. Adhikari, Samik, and Ugo Gentilini. 2018. “Should I Stay or Should I Go: Do Cash Transfers Affect Migration?” Policy Research Working Paper 8525. World Bank, Washington, DC. http:// documents.worldbank.org/curated/en/609571531402897490/pdf/WPS8525.pdf Aiyar, S. 2017. “Universal Basic Income: A Doleful, Wasteful Idea.” Indian Journal of Human Devel- opment 11 (2): 183–84. Akee, Randall, William Copeland, E. Jane Costello, John B. Holbein, and Emilia Simeonova. 2018a. “Family Income and the Intergenerational Transmission of Voting Behavior: Evidence from an Income Intervention.” NBER Working Paper 24770. National Bureau of Economic Research, Cambridge, MA. Akee, Randall, William Copeland, E. Jane Costello, and Emilia Simeonova. 2018b. “How Does Household Income Affect Child Personality Traits and Behaviors?” American Economic Review 108 (3): 775–827. Alatas, Vivi, Abhijit Banerjee, Rema Hanna, Benjamin A. Olken, and Julia Tobias. 2012. “Targeting the Poor: Evidence from a Field Experiment in Indonesia.” American Economic Review 102 (4): 1206–40. Chapter 1. The Idea of Universal Basic Income 63 Alderman, Harold, Ugo Gentilini, and Ruslan Yemtsov, eds. 2018. The 1.5 Billion People Question: Food, Vouchers or Cash Transfers? Washington, DC: World Bank. http://documents.worldbank. org/curated/en/398281507803030509/pdf/120368-PUB-PUBLIC-ADD-BOX-405295B- DOCDATE-11-11-2017.pdf Ang, Yuen Yuen. 2016. How China Escaped the Poverty Trap. Ithaca, NY: Cornell University Press. Apella, Ignacio, and Gonzalo Zunino. 2018. “Nonstandard Forms of Employment in Developing Countries: A Study for a Set of Selected Countries in Latin America and the Caribbean and Europe and Central Asia.” Policy Research Working Paper 8581. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/246711536777440870/pdf/WPS8581.pdf Arnold, Carrie. 2018. “Money for Nothing: The Truth about Universal Basic Income.” Nature 557 (7707): 626–28. Atkinson, Anthony. 2011. “Basic Income: Ethics, Statistics and Economics.” Paper based on pre- sentation made at Basic Income and Income Redistribution workshop, Centre for Research in Economic Analysis, University of Luxembourg, April 2011. https://www.nuff.ox.ac.uk/users/ atkinson/Basic_Income%20Luxembourg%20April%202011.pdf —. 2015. Inequality: What Can Be Done? Cambridge, MA: Harvard University Press. Baird, Sarah, Francisco H.G. Ferreira, Berk Özler, and Michael Woolcock. 2014. “Conditional, Unconditional and Everything in between: A Systematic Review of the Effects of Cash Trans- fer Programmes on Schooling Outcomes.” Journal of Development Effectiveness 6 (1): 1–43. Baird, Sarah, David McKenzie, and Berk Özler. 2018. “The Effects of Cash Transfers on Adult Labor Market Outcomes.” Policy Research Working Paper 8404. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/334251523556191237/pdf/WPS8404.pdf Banerjee, Abhijit. 2016. “The Best Way to Welfare.” Indian Express June 18. Banerjee, Abhijit, Rema Hanna, Gabriel E. Kreindler, and Benjamin A. Olken. 2017. “Debunking the Stereotype of the Lazy Welfare Recipient: Evidence from Cash Transfer Programs.” World Bank Research Observer 32 (2): 155–84. Banerjee, Abhijit, Paul Niehaus, and Tavneet Suri. 2019. “Universal Basic Income in the Developing World.” NBER Working Paper 25598. National Bureau of Economic Research, Cambridge, MA. Barca, Valentina. 2017. Integrating Data and Information Management for Social Protection: Social Registries and Integrated Beneficiary Registries. Canberra: Commonwealth of Australia, Depart- ment of Foreign Affairs and Trade. Bardhan, Pranab. 2017. “Universal Basic Income—Its Special Case for India.” Indian Journal of Human Development 11 (2): 141–43. Barr, Nicholas. Forthcoming. The Economics of the Welfare State. Sixth edition. Oxford, UK: Oxford University Press. Bastagli, Francesca, Jessica Hagen-Zanker, Luke Harman, Valentina Barca, Georgina Sturge, and Tanja Schmidt. 2019. “The Impact of Cash Transfers: A Review of Evidence from Low- and Middle-Income Countries.” Journal of Social Policy 48 (3): 569–94. Bastian, Guatum Gustave, Markus Goldstein, and Sreelakshmi Papineni. 2017. “Are Cash Transfers Better Chunky or Smooth? Evidence from an Impact Evaluation of a Cash Transfer Program in Northern Nigeria.” World Bank, Washington, DC. Beegle, Kathleen, Aline Coudouel, and Emma Monsalve, eds. 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 Bennett, Christina, Matthew Foley, and Hanna B. Krebs, eds. 2016. “Learning from the Past to Shape the Future: Lessons from the History of Humanitarian Action in Africa.” HPG Working Paper. Humanitarian Policy Group, Overseas Development Institute, London. 64 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov Bhattacharya, S. 2017. “Revisiting the ‘Drain,’ or Transfer from India to Britain in the Context of Global Diffusion of Capitalism.” In Agrarian and Other Histories: Essays for Binay Bhushan Chaudhuri, edited by Shubhra Chakrabarti and Utsa Patnaik. New Delhi: Tulika Books. Birdsall, Nancy. 2018. “Strugglers: This Century’s New Development Challenge.” Keynote address at Australasian Aid Society Conference. Canberra, February 13. https://www.cgdev.org/sites/ default/files/Birdsall-strugglers-development-challenge-slides.pdf Blake, J. Angell, Shankar Prinja, Anadi Gupt, Vivekanand Jha, and Stephen Jan. 2019. “The Ayush- man Bharat Pradhan Mantri Jan Arogya Yojana and the Path to Universal Health Coverage in India: Overcoming the Challenges of Stewardship and Governance.” PLOS 16 (3): e1002759. Blattman, Christopher, Nathan Fiala, and Sebastian Martinez. 2018. “The Long-Term Impacts of Grants on Poverty: 9-year Evidence from Uganda’s Youth Opportunities Program.” NBER Working Paper 24999. National Bureau of Economic Research, Cambridge, MA. Bonnet, Florence, Joann Vanek, and Martha Chen. 2019. “Women and Men in the Informal Econ- omy: A Statistical Brief.” International Labour Office, Geneva. http://www.wiego.org/sites/ default/files/publications/files/Women%20and%20Men%20in%20the%20Informal%20 Economy%20-%20A%20Statistical%20Brief%20-%20for%20web.pdf Bregman, Rutger. 2017. Utopia for Realists: How We Can Build the Ideal World. New York: Little, Brown and Company. Brown, Caitlin, Martin Ravallion, and Dominique van de Walle. 2017. “Are Poor Individuals Mainly Found in Poor Households? Evidence Using Nutrition Data for Africa.” NBER Working Paper 24047. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/ papers/w24047.pdf —. 2018. “A Poor Means Test? Econometric Targeting in Africa.” Journal of Development Eco- nomics 134 (C): 109–24. Browne, James, and Herwig Immervoll. 2017. “Mechanics of Replacing Benefit Systems with a Basic Income: Comparative Results from a Microsimulation Approach.” Journal of Economic Inequality 15 (4): 325–44. Buller, Ana Maria, Amber Peterman, Meghna Ranganathan, Alexandra Bleile, Melissa Hidrobo, and Lori Heise. 2018. “A Mixed-Method Review of Cash Transfers and Intimate Partner Violence in Low- and Middle-Income Countries.” World Bank Research Observer 33 (2): 218–58. Calnitsky, David. 2017. “Debating Basic Income.” Catalyst 1 (23): 63–90. Clemens, Michael, Cinday Huang, Jimmy Graham, and Kate Gough. 2018. “Migration Is What You Make It: Seven Policy Decisions that Turned Challenges into Opportunities.” CGD Note. Center for Global Development, Washington, DC. https://www.cgdev.org/sites/default/files/ migration-what-you-make-it-seven-policy-decisions-turned-challenges-opportunities.pdf Coady, David. 2018. “Financing Social Protection.” Presentation at Social Protection in a Chang- ing World, London School of Economics and Political Science and the International Monetary Fund workshop, London, November 2. Coady, David, and Delphine Prady. 2018. “Universal Basic Income in Developing Countries: Issues, Options and an Illustration for India.” IMF Working Paper 18/174. International Monetary Fund, Washington, DC. https://www.imf.org/en/Publications/WP/Issues/2018/07/31/Universal-Basic- Income-in-Developing-Countries-Issues-Options-and-Illustration-for-India-46079 Cotlear, Daniel, Somil Nagpal, Owen Smith, Ajay Tandon, and Rafael Cortez. 2015. Going Univer- sal: How 24 Developing Countries Are Implementing Universal Health Coverage Reforms from the Bottom Up. Washington, DC: World Bank. https://openknowledge.worldbank.org/bitstream/ handle/10986/22011/9781464806100.pdf?sequence=4 Cottam, Hilary. 2019. Radical Help: How We Can Remake the Relationships between Us and Revolu- tionise the Welfare State. London: Virago. Chapter 1. The Idea of Universal Basic Income 65 Cunha, Jesse M., Giacomo De Giorgi, and Seema Jayachandran. 2017. “The Price Effects of Cash Versus In-Kind Transfers.” Working paper. Northwestern University, Evanston, IL. http:// faculty.wcas.northwestern.edu/~sjv340/pal.pdf Currie, Janet, and Firouz Gahvari. 2008. “Transfers in Cash and In-Kind: Theory Meets the Data.” Journal of Economic Literature 46 (2): 333–83. Daidone, Silvio, Benjamin Davis, Sudhanshu Handa, and Paul Winters. 2019. “The Household and Individual-Level Productive Impacts of Cash Transfer Programs in Sub-Saharan Africa.” Amer- ican Journal of Agricultural Economics 101 (5): 1401–31. Dang, Hai-Anh H., and Andrew Dalaban. 2019. “Is Poverty in Africa Mostly Chronic or Transient? Evidence from Synthetic Panel Data.” Journal of Development Studies 55 (7): 1527–47. de Hoop, Jacobus, and Furio C. Rosati. 2014. “Cash Transfers and Child Labor.” World Bank Research Observer 29 (2): 202–34. Desai, Ray M., and Homi Kharas. 2017. “Is a Growing Middle Class Good for the Poor? Social Policy in a Time of Globalization.” Global Economy and Development Working Paper 105. Brookings Institution, Washington, DC. https://www.brookings.edu/research/is-a-growing-middle-class- good-for-the-poor-social-policy-in-a-time-of-globalization/ Devarajan, Shantayanan. 2018 “How to Use Oil Revenues Efficiently.” Working Paper 1199. Eco- nomic Research Forum, Giza. http://erf.org.eg/wp-content/uploads/2018/05/1199_Final.pdf Devarajan, Shantayanan, Marcelo Giugale, Hélène Ehrhart, Tuan Minh Le, and Huong Mai Nguyen. 2013. “The Case for Direct Transfers of Resource Revenues in Africa.” CGD Working Paper 333. Center for Global Development, Washington, DC. https://pdfs.semanticscholar.org/b5b4 /7ce0186a6db2ed60fc7627ba971a882a30b8.pdf Devereux, Stephen. 1988. “Entitlements, Availability and Famine.” Food Policy 13 (3): 270–82. —. 2001. “Transfers and Safety Nets.” In Food Security in Sub-Saharan Africa, edited by Stephen Devereux and Simon Maxwell, 267–93. London: Intermediate Technology Development Group. —. 2016. “Is Targeting Ethical?” Global Social Policy 16 (2): 166–81. Dietrich, Stephen, and George Schmerzeck. 2019. “Cash Transfers and Nutrition: The Role of Market Isolation after Weather Shocks.” Food Policy 87 (August): 101739. Drèze, Jean. 2017. “Decoding Universal Basic Income.” Indian Journal of Human Development 11 (2): 163–66. Drèze, Jean, and Reetika Khera. 2017. “Recent Social Security Initiatives in India.” World Develop- ment 98 (C): 555–72. Drèze, Jean, and Amartya Sen. 1989. Hunger and Public Action. Oxford, UK: Clarendon Press. Eurofound. 2015. “Access to Social Benefits: Reducing Non-Take-Up.” Publications Office of the European Union, Luxembourg. https://www.eurofound.europa.eu/sites/default/files/ef_ publication/field_ef_document/ef1536en.pdf Evans, David K., and Anna Popova. 2017. “Cash Transfers and Temptation Goods.” Economic Development and Cultural Change 65 (2): 189–221. Evans, David K., and Fei Yuan. 2019. “What We Learn about Girls’ Education from Inter- ventions That Do Not Focus on Girls.” CGD Working Paper 513. Center for Global Development, Washington, DC. https://www.cgdev.org/sites/default/files/what-we-learn- about-girls-education-interventions-do-not-focus-on-girls.pdf Filmer, Deon B., Jed Friedman, Eeshani Kandpal, and Junko Onishi. 2018. “Cash Transfers, Food Prices, and Nutrition Impacts on Nonbeneficiary Children.” Policy Research Working Paper 8377. World Bank, Washington, DC. 66 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov Forget, Evelyn. 2011. “The Town with No Poverty: The Health Effects of a Canadian Guaranteed Annual Income Field Experiment.” Canadian Public Policy 37 (3): 283–305. —. 2018. Basic Income for Canadians: The Key to a Healthier, Happier, More Secure Life for All. Toronto: Lorimer. Francese, Maura, and Delphine Prady. 2018. “Universal Basic Income: Debate and Impact Assess- ment.” Working Paper WP/18/273. Fiscal Affairs Department, International Monetary Fund, Washington, DC. Freeland, Nicholas. 2013. “Social Protection and the Four Horsemen of the Donor Apocalypse.” In Social Protection in Developing Countries: Reforming Systems, edited by Katja Bender, Markus Kaltenborn, and Christian Pfleiderer. New York: Routledge. Friedman, Milton. 1967. “The Case for a Negative Income Tax.” National Review 7 (March): 239–41. Furman, Jason. 2019. “Should We Be Reassured If Automation in the Future Looks Like Automa- tion in the Past?” In The Economics of Artificial Intelligence, edited by Ajay Agrawal, Joshua Gans, and Avi Goldfarb, chapter 12. Chicago: National Bureau of Economic Research. https:// www8.gsb.columbia.edu/faculty/jstiglitz/sites/jstiglitz/files/The%20Economics%20of%20 Artificial%20Intelligence%20-%20Chapter%2014_0.pdf Gelb, Alan, and Anit Mukherjee. 2019. “Fuel Subsidy Reform and Green Taxes: Can Digital Tech- nologies Improve State Capacity and Effectiveness?” CGD Policy Paper 149. Center for Global Development, Washington, DC. https://www.cgdev.org/sites/default/files/fuel-subsidy-reform- and-green-taxes-can-digital-technologies-improve-state-capacity.pdf Gentilini, Ugo. 2015. “Entering the City: Emerging Evidence and Practices with Safety Nets in Urban Areas.” Social Protection and Labor Discussion Paper 1504. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/656081467980515244/pdf/98253-NWP- P148675-Box391506B-PUBLIC-1504.pdf —. 2016. “Revisiting the ‘Cash versus Food’ Debate: New Evidence for an Old Puzzle?” World Bank Research Observer 31 (1): 135–67. http://hdl.handle.net/10986/27701 —. 2018. “What Lessons for Social Protection from Universal Health Coverage?” Background paper. World Bank, Washington, DC. —. 2019. “Making Quasi-Sense of Recent Basic Income Initiatives.” Blog post March 13. Brookings Institution. https://www.brookings.edu/blog/future-development/2019/03/13/ making-quasi-sense-of-recent-basic-income-initiatives/ Gentilini, Ugo, Margaret Grosh, and Michal Rutkowski. 2019. “Understanding Coverage: What Does Universal Social Protection Really Mean?” Blog post January 28. OECD Development Matters. https://oecd-development-matters.org/2019/01/28/understanding-coverage-what- does-universal-social-protection-really-mean/ Ghatak, Maitreesh. 2016. “The Price of Basic Income.” Indian Express July 1. Ghatak, Maitreesh, and Karthik Muralidharan. 2019. “An Inclusive Growth Dividend: Reframing the Role of Income Transfers in India’s Anti-Poverty Strategy.” CEGA Working Paper 90. Uni- versity of California, Berkeley. Ghosh, Jayati. 2017. “A Universal Basic Income in India?” Blog post February 3. International Development Economics Associates. https://www.networkideas.org/themes/economy-and- society/2017/02/a-universal-basic-income-in-india/ Goldsmith, Oliver Scott. 2012. “The Economic and Social Impacts of the Permanent Fund Divi- dend on Alaska.” In Alaska’s Permanent Fund Dividend: Examining Its Suitability as a Model, edited by Karl Widerquist and Michael Howard, chapter 4. New York: Palgrave Macmillan. Government of India 2017. “Universal Basic Income: A Conversation With and Within the Mahatma.” In Economic Survey 2016–17, 172–211. New Delhi. Chapter 1. The Idea of Universal Basic Income 67 Guterres, António. 2018. “United Nations Secretary-General Address to the General Assembly.” September 25. https://www.un.org/sg/en/content/sg/speeches/2018-09-25/address-73rd- general-assembly Guven, Melis. 2019. “Extending Pension Coverage to the Informal Sector in Africa.” Social Protec- tion and Jobs Discussion Paper 1933. World Bank, Washington, DC. Guven, Melis U., and Phillippe Leite. 2016. “Benefits and Costs of Social Pensions in Sub-Saharan Africa.” Social Protection and Labor Discussion Paper 1607. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/325281469593828257/pdf/107234-WP-add- series-PUBLIC.pdf Gwatkin, D., and A. Ergo. 2011. “Universal Health Coverage: Friend or Foe of Health Equity?” Lancet 377 (9784): 2160–61. Handa, Sudhanshu, Silvio Daidone, Amber Peterman, Benjamin Davis, Audrey Pereira, Tia Pal- ermo, and Jennifer Yablonski. 2018. “Myth-Busting? Confronting Six Common Perceptions about Unconditional Cash Transfers as a Poverty Reduction Strategy in Africa.” World Bank Research Observer 33 (2): 259–98. Hanna, Rema, and Benjamin A. Olken. 2018. “Universal Basic Incomes versus Targeted Transfers: Anti-Poverty Programs in Developing Countries.” Journal of Economic Perspectives 32 (4): 201– 26. https://economics.mit.edu/files/15434 Harris, Tom, David Phillips, Ross Warwick, Maya Goldman, Jon Jellema, Karolina Goraus, and Gabriela Inchauste. 2018. “Redistribution via VAT and Cash Transfers: An Assessment in Four Low and Middle Income Countries.” IFS Working Paper W18/11. Institute for Fiscal Studies, London. https://www.ifs.org.uk/uploads/WP201811.pdf Harvey, Philip. 2007. “U.S. Job Creation Programs in the 1930s.” Background Paper for the Human Sciences Research Council Mid-Term Review of South Africa’s Expanded Public Works Pro- gram. Hickel, Jason. 2017. The Divide: Global Inequality from Conquest to Free Markets. London: W.W. Norton & Company. Hickey, Sam, Tom Lavers, Miguel Niño-Zarazúa, and Jeremy Seekings, eds. Forthcoming. The Pol- itics of Social Protection in Eastern and Southern Africa. Oxford, UK: Oxford University Press. Holland, Alisha. 2017. Forbearance as Redistribution: The Politics of Informal Welfare in Latin Amer- ica. New York: Cambridge University Press. Homonoff, Tatiana, and Jason Somerville. 2019. “Program Recertification Costs: Evidence from SNAP.” New York University Wagner School, New York. https://wagner.nyu.edu/files/faculty/ publications/Homonoff%20%26%20Somerville%20%282019%29.pdf Hoynes, Hilary W., and Jesse Rothstein. 2019. “Universal Basic Income in the US and Advanced Countries.” NBER Working Paper 25538. National Bureau of Economic Research, Cambridge, MA. https://gspp.berkeley.edu/assets/uploads/research/pdf/Hoynes-Rothstein-UBI-081518. pdf Hunter, Benjamin M., Sean Harrison, Anayda Portela, and Debra Bick. 2017. “The Effects of Cash Transfers and Vouchers on the Use and Quality of Maternity Care Services: A Systematic Review.” PLOS ONE 12 (3): e0173068. Hyslop, Dean R. 2014. “The Distributional Effects of the Australian Cash Bonus Payments Response to the Global Financial Crisis.” University of Wellington, Wellington. IMF (International Monetary Fund). 2017. Fiscal Monitor: Tackling Inequality. Washington, DC: IMF. https://www.imf.org/en/Publications/FM/Issues/2017/10/05/fiscal-monitor-october-2017 68 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov Indian Economic Times. 2018. “1 or 2 States May Roll out Universal Income in Two Yrs: CEA Arvind Subramanian.” January 29. https://economictimes.indiatimes.com/news/economy/ policy/1-or-2-states-may-roll-out-universal-income-in-two-yrs-cea-arvind-subramanian/ articleshow/62696689.cms Isaacs, Julia, Michael Katz, and Ria Amin. 2016. “Improving the Efficiency of Benefit Delivery: Outcomes from the Work Support Strategies Evaluation.” Urban Institute, Washington, DC. Isenberg, Taylor Jo. 2017. “What a New Survey from Alaska Can Teach Us about Public Sup- port for Basic Income.” Blog post January 28. Economic Security Project. https://medium. com/economicsecproj/what-a-new-survey-from-alaska-can-teach-us-about-public-support-for- basic-income-ccd0c3c16b42 Jamison, Dean T., Lawrence H. Summers, George Alleyne, Kenneth J. Arrow, et al. 2013. “Global Health 2035: A World Converging Within a Generation.” Lancet 382: 1898–95. http://www. globalhealth2035.org/sites/default/files/report/global-health-2035.pdf Jones, Damon, and Ioana Marinescu. 2018. “The Labor Market Impacts of Universal and Permanent Cash Transfers: Evidence from the Alaska Permanent Fund.” NBER Working Paper 24312. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/papers/w24312.pdf Joshi, Vijay. 2016. “Universal Basic Income for India.” Ideas for India October 21. https://www. ideasforindia.in/topics/poverty-inequality/universal-basic-income-for-india.html Kangas, Olli, Signe Jauhiainen, Miska Simanainen, and Minna Ylikännö, eds. 2019. The Basic Income Experiment 2017–2018 in Finland. Preliminary Results. Helsinki: Ministry of Social Affairs and Health. http://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/161361/Report_ The%20Basic%20Income%20Experiment%2020172018%20in%20Finland.pdf Khan, M.E., A. Hazra, A. Kant, and M. Ali. 2016. “Conditional and Unconditional Cash Transfers to Improve Use of Contraception in Low and Middle Income Countries: A Systematic Review.” Studies in Family Planning 47 (4): 371–83. Khemani, Stuti. 2013. “Buying Votes vs. Supplying Public Services: Political Incentives to Under-in- vest in Pro-poor Policies.” Policy Research Working Paper 6339. World Bank, Washington, DC. Khera, Reetika. 2014. “Cash vs. In-kind Transfers: Indian Data Meets Theory.” Food Policy 46 (June): 116–28. —. 2016. “A Phased Approach Will Make a ‘Basic Income’ Affordable for India.” Wire Decem- ber 20, https://thewire.in/88350/basic-income-phased-approach Khosla, Saksham. 2018. India’s Universal Basic Income: Bedeviled by the Details. New Delhi: Carne- gie Endowment for International Peace. https://bit.ly/2OwVkmD Kidd, Stephen. 2015. “The Political Economy of ‘Targeting’ of Social Security Schemes.” Pathways’ Perspectives 19. Development Pathways, Kent, UK. https://www.developmentpathways. co.uk/wp-content/uploads/2015/10/Political-Economy-of-Targeting-PP19-4-1.pdf Kidd, Stephen, and Diloá Athias. 2019. “Hit and Miss: An Assessment of Targeting Effective- ness in Social Protection.” Working Paper. Development Pathways, London. https://www. developmentpathways.co.uk/wp-content/uploads/2019/03/Hit-and-Miss-March13-1.pdf Knox-Vydmanov, Charles. 2014. “Why ‘The Poor’ Don’t Exist (And What This Means for Social Pro- tection Policy).” Pathways’ Perspectives 16. Development Pathways, Kent, UK. https://www. researchgate.net/publication/279059709_Why_’The_Poor’_Don’t_Exist_And_What_This_ Means_For_Social_Protection_Policy Krishnan, Nandini, Sergio Olivieri, and Racha Ramadan. 2018. “Estimating the Welfare Costs of Reforming the Iraq Public Distribution System: A Mixed Demand Approach.” Innocenti Work- ing Paper WP-2018-07. UNICEF Office of Research. https://reliefweb.int/sites/reliefweb.int/ files/resources/WP_2018_07.pdf Chapter 1. The Idea of Universal Basic Income 69 Langendorf, C., T. Roederer, S. de Pee, D. Brown, S. Doyon, and A.-A. Mamaty. 2014. “Prevent- ing Acute Malnutrition among Young Children in Crises: A Prospective Intervention Study in Niger.” PLOS Medicine 11 (9): 2–15. Leite, Phillippe, Tina George, Changqing Sun, Theresa Jones, and Kathy Lindert. 2017. “Social Regis- tries for Social Assistance and Beyond: A Guidance Note & Assessment Tool.” Social Protection and Labor Discussion Paper 1704. World Bank, Washington, DC. http://documents.worldbank. org/curated/en/698441502095248081/pdf/117971-REVISED-PUBLIC-Discussion-paper-1704.pdf Levy, Santiago. 2007. Progress against Poverty: Sustaining Mexico’s PROGRESA-Oportunidades Pro- gram. Washington, DC: Brookings Institution Press. Lindert, Kathy, Tina George, and Inés Rodriguez-Caillava. Forthcoming. A Sourcebook on the Foun- dations of Social Protection Delivery Systems. Washington, DC: World Bank. Lindert, Peter H. 2004. Growing Public: Social Spending and Economic Growth Since the Eighteenth Century, vol. 1. New York: Cambridge University Press. Lokshin, Michael, and Martin Ravallion. 2019. “The Missing Market for Work Permits.” Policy Research Working Paper 9005. World Bank, Washington, DC. Lopez-Calva, Luis F., and Eduardo Ortiz-Juarez. 2011. “A Vulnerability Approach to the Definition of the Middle Class.” Policy Research Working Paper 5902. World Bank, Washington, DC. Lowrey, Annie. 2018. Give People Money: How a Universal Basic Income Would End Poverty, Revolu- tionize Work, and Remake the World. New York: Crown. MacAuslan, Ian, and Nils Riemenschneider. 2011. “Richer But Resented: What Do Cash Transfers Do to Social Relations?” IDS Bulletin 42 (6) : 60–66. Mankiw, Greg. 2009. “News Flash: Economists Agree.” Blog post February 14. http://gregmankiw. blogspot.com/2009/02/news-flash-economists-agree.html Manley, James, Seth Gitter, and Vanya Slavchevska. 2012. “How Effective Are Cash Transfer Pro- grammes at Improving Nutritional Status? A Rapid Evidence Assessment of Programmes’ Effects on Anthropometric Outcomes.” EPPI-Centre, Social Science Research Unit, Institute of Education, University of London, London. Marinescu, Ioana. 2018. “No Strings Attached: The Behavioral Effects of U.S. Unconditional Cash Transfer Programs.” NBER Working Paper 24337. National Bureau of Economic Research, Cambridge, MA. Marmot Review. 2010. Fair Society, Healthy Lives. London: Marmot Review. http://www. instituteofhealthequity.org/resources-reports/fair-society-healthy-lives-the-marmot-review/ fair-society-healthy-lives-full-report-pdf.pdf Mkandawire, Thandika. 2016. “Colonial Legacies and Social Welfare Regimes in Africa: An Empir- ical Exercise.” UNRISD Working Paper 2016-4. Geneva, United Nations Research Institute for Social Development. http://www.unrisd.org/80256B3C005BCCF9/(httpAuxPages)/600078C0 E3281680C1257FC00032132A/$file/Mkandawire-PDRM.pdf Moffitt, Robert. 2002. “Welfare Programs and Labor Supply.” NBER Working Paper 9168. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/papers/w9168.pdf —. 2003. “The Negative Income Tax and the Evolution of U.S. Welfare Policy.” NBER Working Paper 9751. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/ papers/w9751.pdf Moreira, Vanessa. 2019. “A Note on Maricá Municipality’s Plans for a UBI in Rio de Janeiro, Brazil.” Background paper. World Bank, Washington, DC. Moss, Todd, ed. 2012. The Governor’s Solution: How Alaska’s Oil Dividend Could Work in Iraq and Other Oil-Rich Countries. Washington, DC: Center for Global Development. https://www.cgdev. org/sites/default/files/Moss-Governors-Solution_0.pdf 70 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov Mullainathan, Sendhil, and Eldar Shafir. 2013. Scarcity: Why Having Too Little Means So Much. New York: Henry Holt and Company. Munnell, Alicia H. 1986. “Lessons from the Income Maintenance Experiments: An Overview.” In Lessons from the Income Maintenance Experiments: Proceedings of a Conference Held in September 1986. https://www.bostonfed.org/news-and-events/events/economic-research- conference-series/lessons-from-the-income-maintenance-experiments.aspx Murray, Charles. 2016. In Our Hands: A Plan to Replace the Welfare State. Washington, DC: Ameri- can Enterprise Institute Press. Organisation for Economic Co-operation and Development (OECD). 2018. The Future of Social Pro- tection: What Works for Non-Standard Workers? Paris: OECD. Ortiz, Isabel. 2018. “The Case for Universal Social Protection.” Finance & Development 55 (4): 32–34. https://www.imf.org/external/pubs/ft/fandd/2018/12/pdf/fd1218.pdf Ortiz, Isabel, Christina Behrendt, Andrés Acuña-Ulate, and Quynh Anh Nguyen. 2018. “Univer- sal Basic Income Proposals in Light of ILO Standards: Key Issues and Global Costing.” ESS Working Paper 62. International Labour Office, Geneva. https://www.ilo.org/wcmsp5/groups/ public/---ed_protect/---soc_sec/documents/publication/wcms_648602.pdf Packard, Truman, Ugo Gentilini, Margaret Grosh, Philip O’Keefe, Robert Palacios, David Robalino, and Indhira Santos. 2019. Protecting All: Risk Sharing in a Diverse and Diversifying World of Work. Washington, DC: World Bank. https://hubs.worldbank.org/docs/imagebank/pages/ docprofile.aspx?nodeid=31393804 Pellerano, Luca, and Valentina Barca. 2017. “Does One Size Fit All? The Conditions for Condi- tionality in Cash Transfers.” In What Works for Africa’s Poorest? Programmes and Policies for the Extreme Poor, edited by David Lawson, Lawrence Ado-Kofie, and David Hulme, chap- ter 13. Rugby, UK: Practical Action Publishing. https://www.developmentbookshelf.com/doi/ pdf/10.3362/9781780448435 Pechman, Joseph A., and P. Michael Timpane, eds. 1975. Work Incentives and Income Guarantees: The New Jersey Negative Income Tax Experiment. Washington, DC: Brookings Institution. Pega, F., S.Y. Liu, S. Walter, R. Pabayo, R. Saith, and S.K. Lhachimi. 2017. “Unconditional Cash Transfers for Reducing Poverty and Vulnerabilities: Effect on Use of Health Services and Health Outcomes in Low- and Middle-Income Countries.” Cochrane Database of Systematic Reviews 11: CD011135. Peterman, Amber, Anastasia Neijhoft, Sarah Cook, and Tia Palermo. 2017. “Understanding the Linkages between Social Safety Nets and Childhood Violence: A Review of Evidence from Low- and Middle-Income Countries.” Health Policy and Planning 32 (7): 1049–71. Piketty, Thomas. 2016. “What Unequal Societies Need Is Not a ‘Basic Income’ But a Fair Wage.” The Wire. https://thewire.in/uncategorised/basic-income-fair-wage-piketty Premand, Patrick, and Pascale Schnitzer. 2018. “Efficiency, Legitimacy and Impacts of Target- ing Methods: Evidence from an Experiment in Niger.” Policy Research Working Paper 8412. World Bank, Washington, DC. Pritchett, Lant, and Charles Kenny. 2013. “Promoting Millennium Development Ideals: The Risks of Defining Development Down.” CID Working Paper No. 265. Center for International Devel- opment at Harvard University, Cambridge, MA. Quarles van Ufford, Paul, Charlotte Harland, Stanfield Michelo, Gelson Tembo, Kelley Toole, and Denis Wood. 2016. “The Role of Impact Evaluation in the Evolution of Zambia’s Cash Trans- fer Program.” In From Evidence to Action: The Story of Cash Transfers and Impact Evaluation in Sub-Saharan Africa, edited by Benjamin Davis, Sudhanshu Handa, Nicola Hypher, Natalia Winder Rossi, Paul Winters, and Jennifer Yablonski, chapter 9. Oxford, UK: Oxford University Press. Chapter 1. The Idea of Universal Basic Income 71 Ravallion, Martin. 1997. “Famines and Economics.” Journal of Economic Literature XXXV (Septem- ber): 1205–42. —. 2016. The Economics of Poverty: History, Measurement and Policy. Oxford, UK: Oxford Uni- versity Press. —. 2017. “Straw Men in the Debate on Basic Income versus Targeting.” Blog post May 5. Vox. https://voxeu.org/article/arguments-against-basic-income-are-straw-men —. 2018. “Guaranteed Employment or Guaranteed Income?” CGD Working Paper 482. Center for Global Development, Washington, DC. —. 2019. “On Measuring Global Poverty.” NBER Working Paper 26211. National Bureau of Economic Research, Cambridge, MA. Ray, Debraj. 2016. “The Universal Basic Share.” Ideas for India September 29. https://www. ideasforindia.in/topics/poverty-inequality/the-universal-basic-share.html Reinhardt, Uwe. 2013. “On the Economics of Benefits In-Kind.” Course handout. Princeton Univer- sity. https://scholar.princeton.edu/sites/default/files/reinhardt/files/100-2014_benefits_in_kind.pdf Reza Farzin, Mohammad, Dominique M. Guillaume, and Roman Zytek. 2011. “Iran—The Chroni- cles of the Subsidy Reform.” Working Paper 11/167. International Monetary Fund, Washington, DC. Rutkowski, Michal. 2018. “Reimagining Social Protection.” Finance & Development 55 (4): 10–13. https://www.imf.org/external/pubs/ft/fandd/2018/12/pdf/fd1218.pdf Salehi-Isfahani, Djavad, and Mohammad H. Mostafavi-Dehzooei. 2018. “Cash Transfers and Labor Supply: Evidence from a Large-Scale Program in Iran.” Journal of Development Economics 135 (November): 349–67. Sandefur, Justin. 2017. “Disintermediating the State: Would a ‘Universal Basic Income’ Reduce Poverty More Than Targeted Programs?” Blog post March 6. Center for Global Development. https://www.cgdev.org/blog/disintermediating-state-would-universal-basic-income-reduce- poverty-more-targeted-programs Seekings, Jeremy. 2013. “The Beveridge Report, the Colonial Office, and Welfare Reform in British Colonies.” Unpublished paper. University of Cape Town. Shafik, Nemat. 2018. “A New Social Contract.” Finance & Development 55 (4): 4–8. https://www. imf.org/external/pubs/ft/fandd/2018/12/pdf/fd1218.pdf Siddiqi, Arjumand, Akshay Rajaram, and Steven P. Miller. 2018. “Do Cash Transfer Programmes Yield Better Health in the First Year of Life? A Systematic Review Linking Low-Income/Middle-In- come and High-Income Contexts.” BMJ Journals. Archives of Disease in Childhood 103: 920–26. Smith, Richard. 2011. “Social Security as a Developmental Institution? The Relative Efficacy of Poor Relief Provisions under the English Old Poor Law.” In History, Historians and Development Policy: A Necessary Dialogue, edited by C.A. Bayly, Vijayendra Rao, Simon Szreter, and Michael Woolcock. Manchester, UK: Manchester University Press. Standing, Guy. 2017. Basic Income: And How We Can Make It Happen. London: Pelican. Steensland, Brian. 2007. The Failed Welfare Revolution: America’s Struggle over Guaranteed Income Policy. Princeton: Princeton University Press. Stern, Andy. 2016. Raising the Floor: How a Universal Basic Income Can Renew Our Economy and Rebuild the American Dream. New York: Public Affairs. Taafe, Jessica E., Andrew F. Longosz, and David Wilson. 2016. “The Impact of Cash Transfers on Livelihoods, Education, Health and HIV—What’s the Evidence?” Development Policy Review 35 (5): 601–19. 72 Ugo Gentilini, Margaret Grosh, and Ruslan Yemtsov Tabatabai, Hamid. 2012. “From Price Subsidies to Basic Income: The Iran Model and Its Lessons.” In Exporting the Alaska Model: Adapting the Permanent Fund Dividend for Reform around the World, edited by Karl Widerquist and Michael Howard, 17–32. New York: Palgrave Macmillan. Tobin, James. 1970. “On Limiting the Domain of Inequality.” Journal of Law and Economics 13 (2): 263–77. Tondani, Davide. 2009. “Universal Basic Income and Negative Income Tax: Two Different Ways of Thinking Redistribution.” Journal of Socio-Economics 38 (2): 246–55. Triyana, Margaret. 2016. “Do Health Care Providers Respond to Demand-Side Incentives? Evi- dence from Indonesia.” American Economic Journal: Economic Policy 8 (4): 255–88. Van Parijs, Philippe, and Yannick Vanderborght. 2017. Basic Income: A Radical Proposal for a Free Society and a Sane Economy. Cambridge, MA: Harvard University Press. Vives, Juan Luis. 1526. De Subventione Pauperum: On the Relief of the Poor, or of Human Need. New York: Peter Lang Press. Weitz-Shapiro, Rebecca. 2012. “What Wins Votes: Why Some Politicians Opt out of Clientelism.” American Journal of Political Science 56 (3): 568–83. Widerquist, Karl, and Michael Howard, eds. 2012. Alaska’s Permanent Fund Dividend: Examining Its Suitability as a Model. New York: Palgrave Macmillan. Widerquist, Karl, José A. Noguera, Yannick Vanderborght, and Jurgen De Wispelaere, eds. 2013. Basic Income: An Anthology of Contemporary Research. Oxford, UK: Wiley-Blackwell. World Bank. 2013. World Development Report 2013: Jobs. Washington, DC: World Bank. https:// openknowledge.worldbank.org/handle/10986/11843 —. 2015. World Development Report 2015: Mind, Society, and Behavior. Washington, DC: World Bank. —. 2016. “Digital Identity: Towards Shared Principles for Public and Private Sector Coopera- tion. A Joint World Bank Group, GSMA and Secure Identity Alliance Discussion Paper.” World Bank, Washington, DC. —. 2018a. The State of Social Safety Nets 2018. Washington, DC: World Bank. https:// openknowledge.worldbank.org/bitstream/handle/10986/29115/211254.pdf?sequence=4 —. 2018b. World Development Report 2019: The Changing Nature of Work. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/30435 World Bank and International Labour Organization. 2016. “A Shared Mission for Universal Social Protection.” Concept note. https://www.ilo.org/wcmsp5/groups/public/---dgreports/---dcomm/ documents/genericdocument/wcms_378996.pdf Yang, Andrew. 2018. The War on Normal People: The Truth about America’s Disappearing Jobs and Why Universal Basic Income Is Our Future. New York: Hachette Books. Yeung, Ying, and Stephen Howes. 2015. “Resources-to-Cash: A Cautionary Tale from Mongolia.” Development Policy Centre Discussion Paper 42, Crawford School of Public Policy, Australian National University, Canberra. https://im4dc.org/wp-content/uploads/2015/09/Combined- Yeung.pdf UBI as Social Assistance: CHAPTER 2 Comparative Models and Instruments Ugo Gentilini and Margaret Grosh T he debate on universal basic income (UBI) is as hot ideologically as it is confus- ing analytically. Flagship programs and pilots are often being called a UBI while they are not, and those that clearly are not are sometimes called “quasi-UBI”— even if they only share one or two UBI properties as defined in chapter 1 (i.e., universality, unconditionality, transfers in the form of cash). Such loose standards imply that virtually any program can be considered a quasi-UBI. For example, the proposal by Felman et al. (2019) on a “quasi-universal basic rural income” for India is a large-scale guaranteed minimum income (GMI) program. Quasi-UBI programs constitute the vast majority of so-called UBI pilots laid out in chapter 1 and detailed in appendix A. The reframing of different programs in UBI or quasi-UBI terms may be unhelpful in several ways. For example, it confuses and polarizes current debate by trading accu- racy for public resonance; it risks reinventing the wheel around key questions for which there might be a considerable knowledge base (e.g., do quasi-UBI programs discourage work?); it may amplify the gulf between expectations (everyone gets cash) and actual program design (only some receive cash); and it may not elucidate the nuanced, distinct features that a suite of alternative social protection measures possess to pursue similar objectives. This chapter, therefore, is meant to help inject some analytical clarity around UBI and the universe of programs that populate the “quasi-UBI” universe. It does so by contrasting UBI features against the structure of other programmatic options—including sketching out their features as well as identifying their pros and cons to fit a particular goal and context. The chapter is organized as follows: the next section outlines four broad categories of program instruments, which are then detailed in the subsequent four sections. These 73 74 Ugo Gentilini and Margaret Grosh present specific experiences with and the relative merits of each family of measures. The chapter’s final two sections reflect on emerging implications for the debate on universal- ity, and provide conclusions, respectively. Table 2.2 at the end of the chapter presents a summary comparison of UBI and seven other instruments. Types of Programs There is a broad range of benefit structures available in social assistance. These can be classified in various ways, such as those that are universal and those that are lim- ited or targeted in some way; those that are not based on work, and those that are;1 and whether they are channeled through TABLE 2.1  Program Typology the benefit or tax system. We locate the Nonwork-based Work-based options and common programs found Benefit-based UBI, age-based Job guarantee within each category in table 2.1, with allowances, poverty- programs, targeted programs (GMI, temporary public universal transfers as defined in the various unconditional and works, wage social assistance cube (figure 1.1) shown conditional transfers) subsidies in bold. Tax-based Negative income Earned income tax We deliberately included some tax credit measures that are often classified as activation measures. Because a UBI is often debated alongside job guarantee programs (JGPs), we offer an overview of wage-based programs that gravitate around the jobs-related agenda for vulnerable populations. In particular, we examine schemes like JGPs and public works that involve the financing of a full project beyond wage payments, as well as wage subsidies and the earned income tax credit (EITC), which normally cover wage costs only. We do not discuss insurance, as UBI, with its benefit uniformity, may intrinsically not substitute for the ability of insurance to make specific payouts when (and only when) the individual, household, or worker incurs a large loss. As further detailed in the next sections, an important feature of each option is the level of support given and how it changes with income or age. Some of the variants are depicted in figure 2.1. Child allowances and social pensions offer a flat benefit paid to all those in specific age groups, as shown in figure 2.1a. When those benefits are provided based on the sole eligibility criteria of age, they are called “universal” (although they are de facto targeted by age). A UBI offers a flat benefit to all, and is found in figure 2.1b. There are several kinds of programs with benefits tightly linked to income, also shown in figure 2.1b, such as means-tested GMI programs, with benefits phasing out as income increases. Programs with more extended tapering, like the negative income tax (NIT), have broader coverage than a typical GMI but are a conceptual extension of such measures. Although with a slightly different phase-in benefit structure and limited only to earners who file for personal income taxes, programs like the EITC could also fit in this panel, although, as we discuss later in the chapter, the EITC presents a slightly dif- ferent structure. Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 75 FIGURE 2.1  Benefit Structures of Select Interventions a. Interventions linked to age b. Interventions linked to income Benefits Benefits Social UBI pensions Child grants Negative income tax Guaranteed minimum income Age Income Other poverty-targeted benefits are similar in that they are available only to fami- lies or individuals below some income threshold, but have less smooth benefit structures. They may pay a flat benefit per household, or per member of the household who meets some criteria of age or behaviors. For example, an unconditional cash transfer might pay a flat benefit to poor households; a conditional cash transfer program might pay one for each child of school age who attends school regularly. These programs may also be approximated in figure 2.1b, as they present a similar structure to GMIs. These models only trace the benefit side of programs. When their financing comes into play, the shape of a program could well be altered (Banerjee, Niehaus, and Suri 2019; Hoynes and Rothstein 2019). In fact, in several low- and middle-income countries, the poorest households can be net tax payers—that is, they pay more in direct or indirect taxes than they receive in support from the state (Lustig 2018). Similarly, with a UBI, some individuals may pay more in taxes than they receive in benefits, so the net incidence of the benefit is unlikely to be the same flat incidence implied in the figure. In this vein, figure 2.2 contrasts the net benefit incidence of the actual U.S. Supplemental Nutrition Assistance Program (SNAP), a typical means-tested FIGURE 2.2  Benefit Incidence Net of Taxes: Simulated UBI and SNAP in the United States a. Average change in disposable income b. SNAP net benefits 5 5 Thousand US$/year Thousand US$/year 4 4 3 3 2 2 1 1 0 0 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Income decile Income decile SOURCES: UBI benefit structure by Ghenis 2019, based on 10% value-added tax and other taxes; SNAP data from U.S. Treasury 2017. NOTE: For figure 2.2a, welfare changes for deciles 9 and 10 are negative. 76 Ugo Gentilini and Margaret Grosh GMI program (but provided in vouchers instead of cash) with a simulated UBI net of taxes: the net benefits of both schemes would both taper off and eventually vanish. As discussed in chapter 4, if the financing side of the UBI were from a progressive personal income tax system, then the net incidence of a UBI would look more like that of the NIT.2 In other words, a UBI financed via progressive income taxation is de facto targeted via taxes (Francese and Prady 2018; IMF 2017; Ravallion 2018). Benefit-Based Programs Not Based on Work Child Allowances and Social Pensions Social pensions and child allowances are often deemed the closest programs to a UBI. There is a fundamental conceptual difference, however, in that the age-based programs are built for categories of people not expected to work. In this way, they can be thought to cover specific life-cycle risks. When social pensions or child allowances are called “uni- versal,” they are based on age only and with no means test or on a history of earnings and contributions (as in a minimum pension provision in a contributory scheme). The benefit level for child allowances is conceptualized as being a supplement to income to help families (who are assumed to have at least one, often two, active earn- ers) avoid poverty when they increase the number of dependents. A recent review by the International Labour Office and the United Nations Children’s Fund shows that child grants are present in universal form in 21 countries, and in a “quasi” form in another 14 (ILO and UNICEF 2019). The latter are categorized as being short term, affluence tested, and coordinated schemes The benefit level for social pensions is conceptualized as a replacement to income for the elderly (former workers) who no longer work. As such, transfers tend to be higher than the benefit for other safety nets. Overall, such programs exist in about 101 coun- tries. In most Organisation for Economic Co-operation and Development countries, social pensions are not “universal,” as they are means tested. The same is true in Latin America, while Europe and Central Asia display the widest share of universal social pen- sions (World Bank 2018). The first such program was New Zealand’s scheme founded in 1898, with a rich history of proliferation throughout Africa and the Caribbean islands following the publication of the Beveridge Report in 1942 (Seekings 2013).3 The age at which social pensions are granted averages around 65 in most regions, but varies from as low as 55 to as high as 80. Often the pension age for women is five years younger than for men. Brazil, Mauritius, and South Africa have the most gener- ous programs, providing over 60 percent of the income of recipient households in the poorest quintile; but these programs are the exception, and in general benefits are sub- stantially lower. In countries such as Bangladesh, China, India, Mexico, and Turkey, the amount of the noncontributory pension represents less than 40 percent of the value of the national poverty line. Older people receiving a social pension in these countries are still poor (ILO 2018). Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 77 Categorical benefits such as child grants and social pensions present a number of positive features. These include their simplicity and transparency, including their being easily communicated to and understood by the public. Also, as other forms of cash transfers they can help support the costs of child rearing and help invest in human cap- ital. Furthermore, they do not require data collection and verification other than age, thus eliminating possibly contentious needs-based eligibility metrics. They may also exert possible political appeal. On the other hand, they present several challenges. For instance, they exclude those who do not meet age criteria, even if in need (depending on how age correlates with poverty). They can be expensive depending on the country’s demographics and may generate possible undesirable effects on fertility, depending on their design. However, as mentioned earlier in the book, the impact on fertility in low- and middle-income countries is largely unsupported by the evidence. Guaranteed Minimum Income Although the GMI’s minimum income and the UBI’s basic income may sound similar, they are at opposite poles in degree of targeting and benefit structure. Also, they are usually far different in conditionalities, though both are paid in cash. A GMI ensures a minimum to any family that falls below a given income threshold—and just enough to bring them to that minimum. In this way, a GMI tends to display limited coverage at the bottom of the income distribution, with benefits sharply tapered as income rises. This contrasts markedly to the universal and flat UBI benefit. With its steep taper, if the GMI program eligibility threshold is at the poverty line, then poverty would be eradicated at the lowest budgetary cost. This potential efficiency comes at a cost of possibly introduc- ing very significant disincentives to work. In pure form, for all households with incomes below the GMI thresholds, any income from labor or other earnings will induce an equal amount in benefit reduction—that is, a 100 percent marginal tax rate. GMI programs are administratively demanding, requiring a means to assess eligibility rather exactly and to customize benefits according to need upon entry and as need varies over time. Because the design of GMI programs may discourage work, programs rarely operate in pure form. Most GMI-type programs have income disregards or moderated withdrawal of benefits. The low eligibility threshold also implies that GMI recipient fam- ilies are less likely to contain earners, often supporting families where the adults are elderly, (partially) disabled, single mothers of young children, and/or poorly educated and living in areas with few opportunities. Low eligibility thresholds provide families that do have active earners with incentives to make a work effort sufficient to earn an income above the guarantee. And GMI programs often require some sort of activation measure for “work able” adults—registration with the public employment service or active job search; sometimes training or community service; sometimes a customized plan for the family or members in it to address barriers to work. In practice, GMI programs are concentrated in Europe and Central Asia, where 15 countries have such programs. Several other emerging economies have either introduced GMI programs (e.g., Dibao in China, and the top-up to Brazil’s Bolsa Família benefit) or 78 Ugo Gentilini and Margaret Grosh are planning to do so (Turkey). Spending on the GMI varies across countries, from signif- icant (such as in Armenia, which spends 1 percent of gross domestic product [GDP]) to moderate (e.g., Albania, Azerbaijan, Georgia, Kosovo, the former Yugoslav Republic of Macedonia, and Montenegro, which devote around 0.5 percent of GDP) to almost negli- gible (Belarus, Bulgaria, Latvia, Lithuania, and Romania). In a number of countries, those programs show a low coverage rate, including in newly adopted schemes across South- ern Europe (World Bank 2019). In other countries, coverage has declined. For example, in Poland, the number of recipients of means-tested household benefits dropped from 3.8 million in 2008 to about 2.3 million in 2013, partly because the income threshold was not indexed to inflation (World Bank 2018). To summarize, the advantages of GMI programs are that they reduce poverty at low fiscal cost, a focus on the poor may have societal resonance, and a “gap-filling” function may be appropriate in contexts of transitory shocks and economic business cycles. Among the disadvantages, it is worth highlighting the likely work disincentives that, by design, entail both high income and substitution effects. Another disadvantage is that risks of exclusion are high in contexts of high poverty prevalence and exten- sive population concentration around poverty lines. Moreover, a GMI is administratively demanding, requiring both complex initial eligibility decisions and frequent updating of benefit amounts or recertification of eligibility (often every 3, 6, or 12 months). Other Poverty-Based Cash Transfers Most cash transfer programs in developing countries do not have the capacity to observe and manage means testing to run a GMI-type program. In these cases, programs use some combination of proxies for income—assets, family structure, characteristics of earners, geography, etc. They may use data collected in the field from households or community members rather than existing governmental databases in eligibility assessments. The vast majority try to focus benefits toward the bottom end of the distribution—sometimes as tightly as on the bottom 5 or 10 percent of the population, sometimes extending to the bottom 20 or 30 percent—although a few are more inclusive. Often, eligibility decisions are made on simple in/out distinctions or with house- holds placed in multiple bands of need with different levels of benefits. Programs may have a flat benefit structure and give the same amount to every household within a given band. Also, programs may give an amount differentiated by number of family members, or provide set levels for members with different characteristics—preschool children, children in school, disabled family members, the elderly—sometimes with a flat supplement for the poorest households. Into this broad set of programs fall poverty-targeted child allowances, social pen- sions, and disability assistance as well as the prototypical conditional cash transfers and unconditional cash transfers. Some of the flagship programs may have substantial cov- erage goals, but as discussed in chapter 1, many more programs are smaller and with design parameters that yield bigger target populations than their budgets will support. Thus, coverage is incomplete, both of the total population and of the poorest quintile or Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 79 nationally defined poverty thresholds. Again as discussed in chapter 1, limited coverage is the result of other factors too, including a range of delivery issues (e.g., lack of identifi- cation, outreach, or awareness, etc.) as well as limitations inherent in the proxies utilized (Brown, Ravallion, and van de Walle 2018; Guven and Leite 2016; Kidd and Athias 2019; Özler 2017). Given the variation in designs and targeting mechanisms, incidence varies between programs, although its overall shape is progressive. Taken as a category (with the inci- dence of individual programs of different designs lumped together), unconditional cash transfers deliver about 38 percent of their benefits to the poorest quintile. Conditional cash transfers, which are more likely to be poverty targeted, deliver about 45 percent of their benefits to the poorest quintile (World Bank 2018). While programs denominated as child allowances, social pensions, or disability assistance tend to be unconditional, many other cash transfer programs have some sort of soft conditions to ensure that families seek age-appropriate health and education services for their children. Such accompanying measures have shown significant results in reduc- ing malnutrition and gender-based violence (Buller et al. 2018; Hidrobo and Roy 2019). Similarly, an increasing share of programs, both unconditional and conditional with respect to social services, provide allied nudges, services, or asset transfers designed to increase earnings in recipient families. They are usually not linked to European-style public employment services or job search requirements, but to productive inclusion ele- ments to increase entrepreneurship and business skills, savings, or credit—and thus working assets and income (Veras Soares and Orton 2017). The evidence base on such graduation programs reveals positive results, especially in the immediate years after pro- gram completion, even in fragile states (Bedoya et al. 2019). However, evidence from Ethiopia and Uganda shows that impacts tend to fade in subsequent years, documenting a convergence between control and treatment groups4 (Blattman, Fiala, and Martinez 2018; Blattman, Dercon, and Franklin 2019). Overall, like GMI schemes, other poverty-targeted programs can be cost-effective in reducing monetary poverty and engender possible political appeal due to financing and co-responsibilities. When connected explicitly or implicitly to social services, cash trans- fers could also fit more organically in broader efforts to build human capital (Bastagli et al. 2019). However, programs may require observed or proxies to income, which may be unavailable, difficult to collect regularly, and present a mixed track record of perfor- mance. Also, programs such as conditional cash transfers may be relatively rigid to scale up—unless conditions can be lifted on an exceptional basis, as in the Philippines (Gen- tilini, Laughton, and O’Brien 2018)—and may exclude vulnerable, nonpoor households. Tax-Based Programs Not Based on Work It has been claimed that the negative income tax represents “…one of the funda- mental ideas of modern analysis of welfare programs” (Moffitt 2003, 3), especially in high-income countries. As with the GMI, the NIT represents the notion of a pure form of 80 Ugo Gentilini and Margaret Grosh fine-tuning tax and transfers according to welfare. It is like a GMI with a possibly higher threshold and gentler tapering. The NIT poses the same kind of information demands as a GMI, but more extensively as it covers the full income distribution and relies even more on actual income. Like the GMI, the NIT produces a high marginal tax rate, though less than 100 percent. The iconic pro- FIGURE 2.3  NIT Benefit Structure Based posal from Milton Friedman in 1970 on Friedman’s 1970 Family Assistance (Steensland 2007) for the United States, Plan Model illustrated in figure 2.3, was 50 percent over a range around the poverty line. 12 Then at a certain break-even point, NIT benefits (thousand US$) 10 program benefits would be zero (i.e., a 8 marginal tax rate of 100 percent; in the 6 U.S. pilots discussed in chapters 1 and 4 4, such a point occurred at an income 2 level 1.8 times the poverty line). 0 When considering financing and 0 5 14 22 24 30 Annual income (thousand US$) payouts jointly, UBI and NIT may, under certain conditions, be similar in net SOURCE: Packard et al. 2019. NOTE: 1970–2016 dollar conversions are rounded. effect. If UBI were to be paid for via pro- gressive and universal personal income taxation, then NIT and UBI are analyt- ically equivalent on net (Barr, forthcoming). Reducing benefits as income rises directly through means testing as in the NIT, versus indirectly through progressive income tax rates in a UBI, is analytically the same as far as disposable income is concerned. In both cases the decision about balancing the steepness of the taper and the break-even point would be taken in the tax code. The key difference is that NIT is focused on taxing and transferring less, and UBI on taxing and transferring more. Both programs face the same dilemma in balancing minimum guaranteed support, the effective marginal tax rate on benefits, and the break-even point where members of society become a net contributor to government coffers. But perhaps a UBI could be deemed less transparent than the NIT option, chiefly because it may be less explicit or visible who is paying for whom. UBI may be superior to NIT, however, in reaching the poorest (who work mostly in informal activities) and on intertemporal risk management (because of transfer frequency). In a way, there is no NIT without a full tax system. A UBI could still be paid without a full-fledged system by, for example, taxing resource (or carbon tax) revenues and simply redistributing the resources to everyone. This would not be possible under NIT, since income and tax declaration are tied to differentiated amounts of benefits received. As such, UBI could be seen as a steppingstone toward NIT—and perhaps one more suitable for informal settings. In those circumstances, however, there are also proposals to identify proxies that are normally correlated with consumption, collect them digitally, include them in social registries, and use the over- all information to mimic NIT. Such a tapered UBI would present features somewhere in between UBI, NIT, and GMI (Majoka and Palacios 2019; Packard et al. 2019). Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 81 Benefit-Based Programs Based on Work Job Guarantee Programs JGPs, like GMI programs, are a promise that anyone who does not earn an income in the private market that meets a societally defined minimum will receive support—in this case, not through a top-up transfer but through (additional) work. JGPs are open to anyone willing to work at the defined (low) wage. They are open ended in duration and do not envisage targeting criteria or eligibility requirements other than willingness to work and possibly age and citizenship or immigration status. JGPs have two objectives: to generate productive employment and to ensure adequate income. A fundamental question is whether JGPs can really achieve their competing objectives simultaneously in full measure, and if not, to which to give the most importance. If a JGP’s main objective is to generate jobs, it should not, contrary to most analyses, be compared to programs such as a UBI; instead, it should be weighed against demand- side measures of the labor market. The more direct of these range from wage subsidies to capital injection to firms; the less direct measures include opening trade markets, “doing business” style reforms, and even adjusting fiscal and monetary policy. Relative to many such policies, however, investments in JGPs may go into workers’ pockets more directly. If the core identity of JGPs is to provide income through a job of last resort (with the quality or value of work secondary), the appropriate comparisons to alternative policies is more with social assistance programs such as cash transfers (and indeed a UBI), which entail debates around universality versus targeting, conditionality (e.g., work require- ments), and transfer modalities. JGPs would be openly self-targeted and more generous than typical social assistance. JGPs strive to create meaningful jobs, but there is usually tension between the number and caliber of jobs that can be provided. Common options for productive jobs include those in the construction and maintenance of public infrastructure and those in public services (such as child or elder care, assistance to the disabled, teachers, library assistants, commu- nity health workers, clerical workers to digitize records, etc.). These are functions with value but are often underprovided due to budget constraints in the public agencies that finance the services. JGP proponents hope that by moving social welfare spending into a jobs mode, the poor will receive support, and society will benefit from the labor used. This double ben- efit is indeed attractive, though it will not be realized in full measure. To subsidize a job will cost more than the wages paid (transfer received by) the worker—some measure of tools, supplies, or raw materials; complementary skilled labor; and management will be needed as well for productive work. Moreover, it may be difficult to provide enough useful jobs—that is, those that could be useful to society but are not provided by the private sector. To provide employment with adequate incomes, JGP proposals may suggest wages more generous than typical low-skilled jobs (proposals in the United States include a poten- tial minimum wage of US$15/hour plus benefits). In doing so, they may raise the demand for work in the guarantee program—the higher the pay, the more workers it will attract. Moreover, the wage in the guarantee program may affect wages in the private sector. The 82 Ugo Gentilini and Margaret Grosh interpretation of such an effect can differ. That firms would have to raise their worker com- pensation is a welcome feature for some: “…yes, we want to disrupt business models that can only be successful if they pay poverty-level wages without the benefits that are common in all of the developed countries” (Tcherneva 2018, 1). Yet, JGPs as a whole are subject to compet- ing narratives: part of the literature posits that since everyone would be employed, workers would be empowered (by eliminating reserve employees or the fear of unemployment). A different strand of thinking suggests that while everyone would be employed, JGPs would not strengthen workers’ bargaining position; rather, they would simply replace welfare with work requirements (Bruenig 2018). It is no wonder JGPs are sparking fascinating debates around the role of work in societies, and about the deeper relationship between employment, pov- erty, and society (Gentilini 2018a, 2018c; Ravallion 2018; World Bank 2013). In a full-fledged JGP, the state offers continuous employment to anyone. In practice, this model is almost never purely implemented. There are historical experiences in the United States during the Great Depression (Harvey 2007), and some might consider the Chilean programs in the 1980s analogous. Presently, India is the only country operating a form of JGP, although employment days per family are capped at 100/year in rural areas. The National Rural Employment Guarantee Scheme (NREGS) costs nearly 0.3 percent of GDP, with wage costs absorbing 67 percent of the budget. Almost 25 percent of rural households participate in the scheme, mostly women (55 percent) (Drèze and Khera 2017). However, there is a significant rationing of jobs—that is, not everyone who wants work gets it. Only 56 percent of applicants eventually participate (a share that declines to one-third in low-income states), mostly due to leakages (Ravallion 2019). Interestingly, and differing somewhat from the public works literature, second-round labor market effects of the NREGS have been the subject of considerable empirical scru- tiny. Berg et al. (2012) show a 5.3  percent increase in the real daily agricultural wage rate across India due to the NREGS. The authors find that the program mainly affects the wages of unskilled laborers, and the wage effects are stronger in districts where the program was first rolled out. Azam (2012) documents significant increases in public sector employment and labor force participation as a result of the NREGS, particularly for women. The paper also indicates that the NREGS helps to narrow gender wage gaps—the wages for female casual workers increase 8 percent more in NREGS districts compared to non-NREGS districts, whereas the impact on male wages is less than 1 percent. Other gender effects include psychological benefits, with a reduction in depression symptoms due to economic security and independence. Such effects are particularly strong among marginalized groups. In some states, lean-season poverty is cut by half for scheduled caste and tribal households. The NREGS also revived institutions of local democracy, such as gram panchayats (village councils) and gram sabhas (village assemblies). Imbert and Papp (2015) find a 4.7 percent increase in the daily wages of casual labor- ers, a 1.2 percent increase in public employment, and a 1.5 percent decrease in private sector employment in NREGS early adoption districts, compared to those that received the program later. These wage and employment effects are concentrated in seven “star states” where the NREGS is well implemented.5 Additionally, the NREGS decreased the likelihood of migration by 8–11 percentage points (Adhikari and Gentilini 2018). Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 83 Muralidharan, Niehaus, and Sukhtankar (2017) provide labor market evidence based on a large-scale randomized experiment of biometric smart cards in Andhra Pradesh aimed at improving NREGS implementation. Although it only finds weak increases in pri- vate sector employment, the paper documents significant decline in days without paid work by 7.1 percent in treated areas. In addition, the authors find a 6.1 percent increase in private sector wages for unskilled labor in the month of June, and a 5.8 percent increase in reported reservation wages.6 Importantly, regardless of actual participation in the public works, all NREGS-registered households have benefited from these wage gains. Temporary Public Works Programs Over the past decades, public works have been one of the most popular social assistance interventions worldwide, from Afghanistan to the United States. According to Eurostat data, in 2015 over 1.1 million people participated in direct job creation public works pro- grams in the European Union, for the equivalent of around US$9.5 billion in spending (Gentilini 2018c). Currently, in Sub-Saharan Africa alone, there are 70 public works pro- grams in 29 countries (Beegle, Coudouel, and Monsalve 2018). There are five models of temporary public works programs in addition to JGPs. While not exhaustive and with the categories not mutually exclusive, this typology may provide a way to chart issues and trade-offs emerging from cross-country experiences. •• Safety net approach: temporary income support and consumption smoothing. Under this category, projects offer short-term income support, typically as a response to some form of acute and temporary crisis, including natural disasters and economic shocks. The income transfer function tends to dominate other objectives, with assets/services in some instances consisting of light activities (e.g., due to fears of dependency, political economy, societal views on co-respon- sibilities, etc.). The share of wages compared to total expenditures is typically between 60 and 80 percent, with wages themselves set below the market wage for unskilled labor to avoid market disruption and encourage the poorest to par- ticipate. As such, these programs tend to offer basic consumption smoothing for the poor, particularly on a self-targeting basis.7 Latvia’s Workplace with Stipend program is an example of such an approach. In the context of a severe eco- nomic downturn, the scheme generated 190,000 jobs with a duration between two weeks and six months for light labor-intensive activities (e.g., public space cleaning). The scheme was rapidly scaled up from 16,000 jobs (December 2009) to 186,000 (January 2010). The low wage (80 percent of minimum wage) resulted in about 96 percent of beneficiaries being in the bottom 40 percent of the income distribution, with the program being largely oversubscribed (Azam, Ferré, and Ajwad 2012; Gentilini 2015). •• Asset provision: infrastructure created and services rendered. The primary objective of this model is the provision of assets and services rather than income transfer. Within large infrastructure projects, there is an attempt to intensify the amount 84 Ugo Gentilini and Margaret Grosh of labor utilized to maximize employment—that is, expenditures shift the factor intensity from capital to labor. For instance, projects by the International Labour Organization’s Employment Intensive Investment Programme are classic exam- ples of this intervention model. Programs may resemble some area-based interventions involving technical lines, especially in ministries of transport, and urban and rural development. Another variant takes a more services-oriented perspective. For example, the Community Employment Programme in Ireland was launched in 1994 in response to high long-term unemployment rates.8 •• Social cohesion and peacebuilding: pursuing social externalities. In some cases, public works are designed with an implicit objective of promoting social cohe- sion and stability, especially in times of political turmoil. Public works can be popular ways for governments in postconflict countries to gain legitimacy and promote positive perceptions by providing income to large numbers of people quickly while rebuilding community assets. Many demobilization, disarmament, and rehabilitation–related employment programs in fragile and postconflict contexts may fall into this category—one example being the Youth Employment and Empowerment Programme in Sierra Leone. In Sri Lanka, participants in a public works program assisting more than 250,000 returnees noted that the program meetings were the first community-level gathering they had attended after having arrived from internally displaced person camps. By many accounts, community meetings, shared meals, teamwork, and the involvement of elders and children as indirect program beneficiaries promoted a sense of belonging among the newly resettled families. Relatedly, effects on the social fabric of communities and their empowerment was documented in public works pro- grams in Zambia and Peru. Other externalities (often unintended) have been observed in terms of climate mitigation and carbon benefits (e.g., Ethiopia’s Managing Environmental Resources to Enable Transition to More Sustainable Livelihoods [MERET]9 and Productive Safety Net Programme programs). •• Provision of services. This approach does not necessarily entail heavy labor-in- tensive activities; instead, it provides temporary jobs in realms like child care, old-age assistance, service at social centers (e.g., soup kitchens, orphanages), and other services. Programs such as Kinofelis in Greece and part of South Afri- ca’s Expanded Public Works Programs have been moving in this direction. •• Enhancing employability: increasing the likelihood of job market entry via public works–plus. This model includes public works programs combined with other active labor market policies. Their overall likelihood of enhancing employability by imparting skills rests on several assumptions: unemployment includes a fric- tional element, with supply-side constraints representing a key bottleneck; there are reasonable prospects of economic recovery and/or the imminent expansion of labor demand; there is matching capacity to identify and tailor skills to ben- eficiaries’ profiles; trainings are effectively delivered and post-training feedback Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 85 loops are established. One example is El Salvador’s Programa de Apoyo Tempo- ral al Ingreso targeting high-violence urban areas (Gentilini 2015). The performance of public works overall has been widely documented (GIZ 2019; McCord 2012; Subbarao et al. 2013). With the exception of the NREGS in India and the Pro- ductive Safety Net Programme in Ethiopia, public works are not being assessed against their full potential—for example, there could be benefits that persist beyond the duration of the program (e.g., in terms of assets created) that are not currently captured by empirical studies. In terms of provision of income, public works programs are often self-targeted by the requirement to work for low(ish) wages. Since almost all programs have budgets far more lim- ited than the number of workers willing to work for such wages, they usually also have other targeting devices—by rationing the number of days each worker can work, by age or geogra- phy, by some sort of assessment of poverty, or even by lotteries, such as in the Central African Republic (Alik-Lagrange and Bance 2019). Such rationing makes these programs targeted, while manual labor requirements could imply missing disabled or higher-skilled workers. In general, the progressivity of benefits in public works programs is less than for poverty-targeted cash transfer programs. In 9 out of 10 programs with results available, fewer than 40  percent of the beneficiaries were in the poorest quintile (World Bank 2018). Wages paid are not all net income gain to the household, as often participation in the public works displaces some labor from other activities. Subbarao et al. (2013) show net wage gains of about half of gross for Bangladesh and India; about 80 percent for Ethi- opia, Liberia, and Niger; and even higher for Sierra Leone. Wage Subsidy Programs Wage subsidies are direct transfers to employers/firms or individual workers to cover wages in full or in part (Almeida, Orr, and Robalino 2014; Bördo ˝s, Csillag, and Scharle 2015; Gentilini 2018b; Kluve et al. 2016). The main aim is to incentivize existing firms to either increase employment or retain employees who might otherwise be laid off for economic reasons. In those contexts, wage subsidies can represent a risk discount to compensate employers for the potential lower productivity or perceived risks inherent in hiring people with the above profiles. Wage subsidies present a range of possible benefits. These can be clustered around the following four: •• Revealed information. The period of subsidized work can act as a screening device, providing direct information on productivity. •• Direct work experience and skills acquisition. Subsidized employment can promote skill formation through on-the-job learning, leading to increased productivity and subsequent improvement in employment prospects over the longer term. •• Employment probability. Awareness of eligibility might change the (eligible) work- ers’ market perception of success rates and increase job-search efforts, which may also increase employment probability. 86 Ugo Gentilini and Margaret Grosh •• Quality alignment. By influencing certain work or career paths, people can target more suitable opportunities in their subsequent job searches. This “job ladder effect” can improve the quality of future job matches. (In contrast, if workers accept less suitable jobs in the absence of the subsidy, this can create a trap and harm their career paths in terms of future employment prospects or earnings.) Clearly, wage subsidies present a gamut of risks and limitations. A core one is dead- weight: in such a case, the subsidy may support a share of eligible workers who would have been hired in any case, regardless of whether subsidies were offered; this is also known as windfall wastage. Another risk is substitution: while hiring subsidized workers, firms may lay off ineligible workers who have similar characteristics and can be substituted by eligible workers (i.e., no rise in overall employment, but an internal firm reshuffling). Displacement and stigma are two other risks. In the case of displacement, increases in employment among firms absorbing subsidized labor might generate job losses among firms not benefiting from such a cost advantage. Stigma can occur when firms view the targeted subsidy as an indication of an employee’s low productivity and, contrary to intent, they avoid hiring from the group of those eligible. Alternatively, the targeted workers themselves may feel that eligibility is stigmatizing and may try to con- ceal their eligibility status. Establishing conditions is one way of minimizing the risks and limiting unintended behavioral responses by firms (hence hampering the effectiveness of hiring subsidies). But there is a trade-off between the additional costs incurred due to deadweight and substitution effects and reduced effectiveness due to low take-up. Indeed, conditions might reduce risks, but also increase the administrative burden and compliance costs for firms, reducing the potential benefits of the subsidy for employers. The extent of these costs is difficult to quantify, as different types of employers might weight them differ- ently. The devising of conditions is a fundamental parameter that, alongside other key considerations such as subsidy generosity (size), can significantly tilt overall program performance,10 especially in terms of firms’ take-up rates (box 2.1). Tax-Based Programs Based on Work The EITC is perceived as occupying a middle ground between a tax and transfer status—a tax program in administrative terms, but largely a poverty-oriented social assistance pro- gram in conceptual justification and economic effects. It may be thought of as the real-life approximation to NIT. Contrary to the pure NIT and to the UBI (and to many other social assistance programs), it deliberately ties benefits to labor force participation and earnings. Introduced in 1975, the United States’ EITC program was a pioneer, engineered to encourage work among lower-income individuals. A handful of other high-income coun- tries today have similar programs, including Australia, Belgium, Canada, Finland, France, Ireland, the Netherlands, New Zealand, Sweden, and the United Kingdom. These present some design variants: for example, in the United States, tax credits are granted once a Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 87 BOX 2.1  Balancing Wage Subsidy Compliance and Generosity T o make a program more attractive, stricter compliance rules may need to be counterbalanced by more generous subsidies. There are at least three main forms of conditions: penalties on dismissal; extension of postsubsidy contract or contractual conversion; and training, mentoring, and skills development. •• Penalties on dismissal. Germany’s Immediate Action Program for Reducing Youth Unemployment provided subsidies covering either 40 or 60 percent of the worker’s wage. Strict conditions were imposed on employers, obliging firms to pay back half the subsidy for dismissals. Impact evaluations found signifi- cant positive results three years after program completion. In contrast, Austria’s Eingliederungsbeihilfe, while offering similarly generous subsidies, did not envi- sion strict nondismissal clauses. Although evaluations found the program to be effective, its impacts were tempered by substantial deadweight of 60 percent. •• Extension of postsubsidy contract and/or contractual conversion. France’s Generational Contract program offers lump-sum payments upon hiring youth on permanent contracts, with the obligation to keep (or hire) older employees and assign an older “mentor” to new youth hires. •• Training, mentoring, and skills development. The U.K. New Deal for Young People program paid a flat-rate hiring subsidy to employers (equivalent to about 40 percent of the initial wage), who were obliged to offer at least one day of training per week, for which they also received a flat-rate reimbursement. SOURCE: Gentilini 2018b. year when the annual income tax filing is done. In the United Kingdom, the credit comes in the paycheck, with workers receiving their money monthly. While Australia, Canada, and the United States have no minimum hours worked to qualify for the EITC, mini- mums are set in the United Kingdom (16), Ireland (19–20), and New Zealand (20–30). The U.S. experience is one of the most widely discussed and studied, so it is worth understanding its design and impacts. In 2016, the U.S. EITC covered 26.4 million fami- lies, at a cost of about 0.35 percent of GDP—on par with spending of other major safety net programs (e.g., SNAP). Under the program, workers receive a tax credit equal to a (flat) percentage of their earnings up to a maximum credit amount; both the rate and absolute level depend on the number of children in the family. The credit remains at its maximum value until earnings (or income) reach a plateau, at which point they phase out at a rate about half of the phase-in stage.11 The “hill-shaped” structure of the EITC is illustrated in figure 2.4. In the U.S. EITC, most average credits accrue to households in the bottom 40 percent of the income distribution, and none in the top 40 percent. The credit does not reach those who 88 Ugo Gentilini and Margaret Grosh FIGURE 2.4  The Inverse U-Shaped Benefit Structure of the EITC 6,318 - 5,616 - Three children 21 .0 Credit amount (US$) 6% -in ph Two children e as as eo ph ut % 40 21 One child % .0 6% 45 3,400 - No children 15. 98 % % 34 510 - 7.65% 7.65% 6,670 8,340 10,000 14,040 15,010 18,340 39,617 45,007 48,340 Income amount (US$) SOURCE: Maag 2017. cannot or do not work, and benefits are extremely limited to those without children (97 per- cent of recipients have at least one child). Reviews of empirical evidence suggest that the credit increased employment rates of single mothers—one of the groups with higher elastic- ity of labor force participation—between 2 and 10 percentage points. The literature, however, is less settled on how much the EITC affects work intensity, or hours of work. The main advantages of an EITC include the following: it is politically attractive as it rewards work, it is relatively “off the radar” for attacks, incidence is progressive, and work- ing through the tax authority makes it easier to administer and may be less stigmatizing. Conversely, its main limitations involve the following: it only reaches taxpayers, who tend to be formal wage earners. This would largely leave out informal workers, the self-em- ployed, those not filing taxes, etc., which would be a very large share of the poor population in developing countries. Working through the tax authority makes it more difficult to link to other programs and activation. It requires credible and efficient tax administration, tax tri- bunals, and a dispute resolution system. And when paid annually, as in the United States, an EITC is of less use in meeting recurrent living expenses such as food and shelter. Conclusions There are many objectives and features embedded in social assistance. These closely mirror the framework laid out in this book’s overview: for example, social assistance can be intended to cover everyone who needs support; to provide adequate transfers to households or individuals; to be manageable to finance and administer; to garner suf- ficient societal consensus to maintain political support; and to create as many positive and as few negative side effects as possible. There are tensions among these objectives, Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 89 and no program scores highly on all of them. The trade-offs and difficult balance across dimensions account for part of why social assistance is so varied in different contexts, and why it remains an ongoing area of fervent debate. The idea of a UBI is a possible new entry point to build social protection systems. Yet what exactly a UBI is remains contested. For example, Italy’s recent Citizens’ Income, initially hailed as a UBI, is instead a GMI program, and one conditioned on work. The broad application of the term “UBI” to a wide array of different schemes may not only increase semantic confusion, but also obscure the actual comparative advantages—and limitations—that specific options bring to the table. In a way, the misalignment between what a program is called and what it is in practice may reduce the space for informed policy discussions around how to best achieve intended objectives with the instruments at hand. By reviewing the core parameters of a suite of social assistance interventions, this chapter aims to elucidate the differences between a UBI and other possible instruments. We discuss the traits of different programmatic choices with the view of better inform- ing the debate through precision in definition, specificity in objectives, and clarity on the relative merits of different programs in pursuing them. The discussion around alternative social assistance programs is largely one around charting the benefits curve—that is, the reach and steepness of the benefit structure across the income distribution. Such steepness ranges from being flat (in UBI and cate- gorical programs) to being quite sharp in GMI models. It is extensive in reach (or coverage) for UBI- and categorical NIT-type programs, and less so for other interventions. The exact contours of the benefit structure would hinge upon program objectives and contextual variables such as the funding envelope. Some of the reviewed programs operate via the tax code, such as the NIT and the EITC, thus presenting features that may not be easily applicable to contexts with large informality. However, even in programs in lower-income countries that are managed by social protection authorities, such as a prospective UBI, taxation is too often not consid- ered as part of the debate. As we discuss in chapters 4 and 5, if financing includes direct or indirect taxation instead of—for example—subsidy reforms or fiscal consolidation, some citizens will be net receivers and other net payers of a UBI. In other words, a UBI could become a targeted scheme via taxes. Targeting is therefore not necessarily an alien feature in a UBI—both in tax terms and categorically (a UBI may be limited to adults, for instance). Societal preferences and attitudes toward redistribution may also influence whether programs should be based on co-responsibilities in general, and on work in particular. The emergence of JGPs extends the debate on the UBI from one of universality in income to one on universality in jobs. This, in turn, involves a host of other issues that span across the supply and demand sides of labor markets—how to generate good jobs— whereas the UBI is only a slice of the bigger pie (pertaining to the supply side). Chapter 3 discusses more extensively the relationship between UBI and employment. As noted in chapter 1, one of the core challenges of UBI will be to manage expec- tations. To this effect, it will be critical to anchor UBI debates to a clear definition of program design, a crisp articulation of objectives, and how the UBI is expected to per- form better than alternatives relative to those objectives. TABLE 2.2  Summary Comparison of Intervention Features Select features Overall context/ Uni- Payment Work- Tax- Assistance No. of Program objectives versal frequency based based unit countriesa Pros Cons Direct resources No Monthly No No Household 71 •• Can be cost-effective in •• Requires observed/proxy to Poverty- toward the bottom of reducing monetary poverty income targeted the distribution •• Can have work disincentives •• Relatively rapid scalability unconditional cash transfer •• Generally small scale •• Excludes vulnerable/nonpoor Direct resources No Monthly No No Household 62 •• Can be cost-effective in •• Requires observed/proxy to Conditional toward the bottom of reducing monetary poverty income cash transfer the distribution and •• Relatively rigid to scale up •• Possible political appeal build human capital •• Excludes vulnerable/nonpoor Similar to GMI, but via No Yearly No Yes Tax unit 0 •• De facto combines tax and •• Requires functioning tax system Negative tax system and likely benefits systems •• Paid annually income tax of higher coverage •• Incentivizes formalization •• No country experience Support certain No Monthly No No Household Child •• Simplicity and transparency •• Excludes those who do not categories of grants: 35 meet age criteria, even if in •• Does not require data people identified (21 universal need (depends on how age collection and verification Categorical via age (or disability, and 14 quasi- correlates with poverty) other than age transfer orphanhood, etc.) universal) •• Can be expensive depending •• Eliminates possibly contentious on demographics Social needs-based eligibility metrics pensions: 101 •• In rare cases, may have •• Possible political appeal behavioral effects (e.g., fertility) Provides flat support Yes Monthly No No Individual 2 •• Simplicity and transparency •• The worse-off receive no to the whole more support than others •• No data collection and population verification for eligibility •• An adequate benefit amount required other than possibly would be fiscally expensive age or citizenship (but UBI •• May be socially and still needs identification, politically polarizing in some payments, etc.) societies •• Possible social cohesion •• Limited country experience •• Relatively incentives- to understand systemwide compatible effects (continued) TABLE 2.2  Summary Comparison of Intervention Features (continued) Select features Overall context/ Uni- Payment Work- Tax- Assistance No. of Program objectives versal frequency based based unit countriesa Pros Cons Provides jobs at the Yes Monthly Yes No Household 1 •• Appealing where preferences •• Administrative complexity minimum wage to for co-responsibilities are high •• Possible dead-end, low- anyone willing to •• Can enhance labor market quality jobs work opportunities for the poorest •• Unclear how to deal with bad (by eliminating fear of workers or employers JGP unemployment) •• Excludes children/elderly/ •• Generate assets and services disabled •• Possible learning on the job •• Politically popular (creating jobs and co-responsibility) Engage beneficiaries No Monthly Yes No Household 79 •• Rapid scalability of temporary •• Small effects over time in temporary work- jobs •• Competes with other related activities •• Tackling chronic labor livelihood activities demand in permanent model Public works •• Administratively demanding •• Various benefits associated (if permanent) with work •• Trade-offs in objectives (income, employment, and assets) Incentivizes more No Yearly Yes Yes Tax unit 7 •• Keeps and augments •• Annual payments work among low- incentives to work (in formal •• Excludes informal workers income people in sector) EITC and nonworkers formal-sector jobs; •• De facto combines tax and delivered via tax •• Separate from other social benefits systems system services Reducing hiring No Monthly Yes No Household 23 •• Keeps labor market •• Possible deadweight, costs (often for attachment substitution, displacement youth); insurance Wage •• Provides direct work •• Complex administration against uncertain subsidies experience (balancing generosity productivity conditions, monitoring of •• Learning on the job and compliance) training a. Preliminary, based on recent and current experience. 92 Ugo Gentilini and Margaret Grosh Notes 1. We here include four options related to wage employment. Programs for self-employment tend to provide one-off payments, which are conceptually and technically different. 2. Short-term age-based grants are paid for a limited period of the life course (e.g., ages 0–2 in Belarus, and 0–3 in Ukraine); in other cases, programs that exclude high-income households are known as affluence-tested schemes, which cover the large majority of the population including middle-class households (e.g., Mongolia); finally, a coordinated mixed scheme com- bines social insurance and tax-financed provision. Among the 14 quasi-universal child grants, four countries (e.g., Japan) combine social insurance and noncontributory targeted/means- tested schemes (ILO and UNICEF 2019). 3. For a contemporary political economy analysis of how social protection spread across Africa, see Hickey et al. (forthcoming). 4. Similar trajectories are observed for the long-term effects of conditional cash transfers in Malawi (Baird, McIntosh, and Özler 2019). 5. These star states are Andhra Pradesh, Chhattisgarh, Himachal Pradesh, Madhya Pradesh, Rajasthan, Tamil Nadu, and Uttarkhand. 6. However, Zimmermann (2013) finds no significant impact of the NREGS on public sector employment or private sector wages, comparing Phase 2 to Phase 3 districts. 7. A similar large-scale program is the Ethiopia Productive Safety Net Programme, which oper- ates for six months of the year. While maintaining a safety net spirit, the program’s permanent nature and longer-term duration make it distinct from other crisis-response, short-term design options. 8. The program’s objectives were to provide temporary opportunities as well as training and apprenticeships for long-term unemployed and socially excluded persons. Projects were also used to reintegrate former drug addicts into the labor market. In 1998, the scheme had 43,000 participants engaged in two part-time activities: (1) an integration option offers ongo- ing employment for people over age 25 who have received social welfare payments for over a year (or those over age 18 on disability); and (2) an option for up to six years of work was devised for those over age 35 who have been on social welfare for three years or more. Evalu- ations show that if the public works are considered as supply-side measures to move workers into regular employment, then the Community Employment Programme had been relatively unsuccessful; on the other hand, if they are considered demand-side measures designed to create aggregate employment, the program can be considered a success. 9. MERET is a government-led public works program adopting a community-based watershed management approach; it provided important lessons informing Productive Safety Net Pro- gramme design in the mid-2000s. 10. There are several critical design parameters to consider in wage subsidy planning. For instance, an important choice is around the payee, or whether the wage subsidy is to be paid to the employer or the worker (vouchers). The target profile of beneficiaries would entail various choices (e.g., incumbents or new hires). As to subsidy size and structure, the benefit should be large enough to make it attractive for the employer to hire the worker. The target group and related objectives matter for design. For subsidies that aim to promote all youth employment, it is sensible to define the subsidy proportional to wages. However, if the goal is integration of disadvantaged youth, then setting a maximum threshold for the subsidy or defining it as Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 93 a fixed amount can be effective, since this will tilt the employer’s incentives toward hiring the lowest-wage workers (i.e., it may offset a larger proportion of their earnings). Moreover, it might be appropriate to reduce the subsidy over time since productivity will increase with experience, and hence the need to subsidize wages might decline alongside the gap between minimum wage and workers’ productivity. The duration of the subsidy must be carefully cal- ibrated to the objectives: short-term subsidies (six months or less) are appropriate in case of unknown productivity of the worker. Medium-term and longer subsidy periods (from nine months to two years) can permit young workers to develop necessary skills, and as a result increase their productivity. Subsidization beyond this learning-by-doing period tends to be cost ineffective as it leads to deadweight losses. Long-term subsidies are rare and are only used for target groups with multiple disadvantages (e.g., low-skilled, long-term unemployed, and health impaired). 11. For example, during the climbing phase-in stage, a person with one child receives US$0.34 in EITC benefits for every additional dollar of earnings. After the plateau, the credit declines for each dollar of earnings at a stable rate (almost 16 percent), until it tapers out entirely. In 2017, the maximum credit was US$3,400 for a family with one child, US$5,616 for those with two children, and US$6,318 for those with three or more children. In contrast, child- less workers could receive no more than US$510. Also, the income cap for childless workers is much lower. They receive no EITC if their income exceeds US$15,010 (US$20,600 if mar- ried), while workers with three or more children may have incomes as high as US$39,617 (US$45,207 if married) before they lose the benefit. In most cases, EITC recipients do not owe federal income taxes, so they receive the credit as a lump sum in the spring when they file their annual tax returns. This annual payment is in tension with the objective of supporting the day-to-day expenses of poor households. There are state-level EITCs in 23 states. Since the information needed to calculate the state credit is supplied on the federal tax form, adminis- tration at the state level is relatively easy. References Adhikari, Samik, and Ugo Gentilini. 2018. “Should I Stay or Should I Go: Do Cash Transfers Affect Migration?” Policy Research Working Paper 8525. World Bank, Washington, DC. http:// documents.worldbank.org/curated/en/609571531402897490/pdf/WPS8525.pdf Alik-Lagrange, Arthur, and P. Bance. 2019. “The Impact of Public Works in Fragile, Conflict and Vio- lence-Affected Settings: Preliminary Evidence from the Londo Program in the Central African Republic.” Presentation. World Bank, Washington, DC. Almeida, Rita, Larry Orr, and David Robalino. 2014. “Wage Subsidies in Developing Countries as a Tool to Build Human Capital: Design and Implementation Issues.” IZA Journal of Labor Policy 3 (12). Azam, Mehtabul. 2012. “The Impact of Indian Job Guarantee Scheme on Labor Market Outcomes: Evidence from a Natural Experiment.” IZA Working Paper 6548. Institute for the Study of Labor, Bonn. http://ftp.iza.org/dp6548.pdf Azam, Mehtabul, Céline Ferré, and Mohamed Ihsan Ajwad. 2012. “Did Latvia’s Public Works Pro- gram Mitigate the Impact of the 2008–2010 Crisis?” Policy Research Working Paper 6144. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/705321468265794265/ text/WPS6144.txt Baird, Sarah, Craig McIntosh, and Berk Özler. 2019. “When the Money Runs Out: Do Cash Transfers Have Sustained Effects on Human Capital Accumulation?” Journal of Development Economics 140: 169–85. 94 Ugo Gentilini and Margaret Grosh Banerjee, Abhijit, Paul Niehaus, and Tavneet Suri. 2019. “Universal Basic Income in the Developing World.” NBER Working Paper 25598. National Bureau of Economic Research, Cambridge, MA. Barr, Nicholas. Forthcoming. The Economics of the Welfare State. Sixth edition. Oxford, UK: Oxford University Press. Bastagli, Francesca, Jessica Hagen-Zanker, Luke Harman, Valentina Barca, Georgina Sturge, and Tanja Schmidt. 2019. “The Impact of Cash Transfers: A Review of Evidence from Low- and Middle-Income Countries.” Journal of Social Policy 48 (3): 569–94. Bedoya, Guadalupe, Aidan Coville, Johannes Haushofer, Mohammad Isaqzadeh, and Jeremy Shapiro. 2019. “No Household Left Behind: Afghanistan Targeting the Ultra-Poor Impact Evaluation.” Policy Research Working Paper 8877. World Bank, Washington, DC. http:// documents.worldbank.org/curated/en/855831560172245349/pdf/No-Household-Left- Behind-Afghanistan-Targeting-the-Ultra-Poor-Impact-Evaluation.pdf Beegle, Kathleen, Aline Coudouel, and Emma Monsalve, eds. 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 Berg, Erlend, Sambit Bhattacharyya, Rajasekhar Durgam, and Manjula Ramachandra. 2012. “Can Rural Public Works Affect Agricultural Wages? Evidence from India.” CSAE Working Paper 2012-05. University of Oxford, Oxford, UK. http://conference.iza.org/conference_files/ worldb2012/bhattacharyya_s8177.pdf Blattman, Christopher, Stefan Dercon, and Simon Franklin. 2019. “Impacts of Industrial and Entrepre- neurial Jobs on Youth: 5-year Experimental Evidence on Factory Job Offers and Cash Grants in Ethiopia.” NBER Working Paper 25788. National Bureau of Economic Research, Cambridge, MA. Blattman, Christopher, Nathan Fiala, and Sebastian Martinez. 2018. “The Long-Term Impacts of Grants on Poverty: 9-year Evidence from Uganda’s Youth Opportunities Program.” NBER Working Paper 24999. National Bureau of Economic Research, Cambridge, MA. Bördo˝s, Katalin, Márton Csillag, and Ágota Scharle. 2015. “What Works in Wage Subsidies for Young People: A Review of Issues, Theory, Policy and Evidence.” Employment Working Paper 199. International Labour Office, Geneva. https://www.ilo.org/wcmsp5/groups/public/---ed_ emp/documents/publication/wcms_466538.pdf Brown, Caitlin, Martin Ravallion, and Dominique van de Walle. 2018. “A Poor Means Test? Econo- metric Targeting in Africa.” Journal of Development Economics 134 (C): 109–24. Bruenig, Matt. 2018. “Some Notes on Federal Job Guarantee Proposals.” Blog post March 22. People’s Policy Project. https://www.peoplespolicyproject.org/2018/03/22/some-notes-on- federal-job-guarantee-proposals/ Buller, Ana Maria, Amber Peterman, Meghna Ranganathan, Alexandra Bleile, Melissa Hidrobo, and Lori Heise. 2018. “A Mixed-Method Review of Cash Transfers and Intimate Partner Violence in Low- and Middle-Income Countries.” World Bank Research Observer 33 (2): 218–58. Drèze, Jean, and Reetika Khera. 2017. “Recent Social Security Initiatives in India.” World Develop- ment 98 (C): 555–72. Felman, Josh, Boban Paul, M.R. Sharan, and Arvind Subramanian. 2019. “Quasi-Universal Basic Rural Income: The Way Forward.” Business Standard January 27. Francese, Maura, and Delphine Prady. 2018. “Universal Basic Income: Debate and Impact Assess- ment.” Working Paper WP/18/273. Fiscal Affairs Department, International Monetary Fund, Washington, DC. Gentilini, Ugo. 2015. “Entering the City: Emerging Evidence and Practices with Safety Nets in Urban Areas.” Social Protection and Labor Discussion Paper 1504. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/656081467980515244/pdf/98253-NWP- P148675-Box391506B-PUBLIC-1504.pdf Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 95 —. 2018a. “Battle of the Titans: GMI vs PWs vs WS vs JGPs vs EITC vs NIT vs UBI.” Background paper. World Bank, Washington, DC. —. 2018b. “Do Wage Subsidies Work? A Synthesis of Issues and Evidence.” Background paper. World Bank, Washington, DC. —. 2018c. “What Lessons for Social Protection Universal Health Coverage?” Blog post, August 22. Let’s Talk Development, World Bank. https://blogs.worldbank.org/developmenttalk/what- lessons-social-protection-universal-health-coverage Gentilini, Ugo, Sarah Laughton, and Clare O’Brien. 2018. “Lessons on Better Connecting Humanitar- ian Assistance and Social Protection.” Social Protection and Labor Discussion Paper 1802. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/946401542689917993/ pdf/Human-itarian-Capital-Lessons-on-Better-Connecting-Humanitarian-Assistance-and- Social-Protection.pdf Ghenis, Max. 2019. “A Budget-Neutral Version of Andrew Yang’s Freedom Dividend.” Blog post June 24. UBI Center. https://medium.com/ubicenter/a-revenue-neutral-version-of-andrew- yangs-freedom-dividend-d7d517dbeeea GIZ. 2019. “Do Public Works Programmes Work? A Systematic Review of the Evidence in Africa and the MENA Region.” GIZ, Bonn. https://socialprotection.org/sites/default/files/publications_ files/GIZ%20-%20Policy%20Brief%20-%20Do%20Public%20Works%20Programmes%20 Work.pdf Guven, Melis U., and Phillippe Leite. 2016. “Benefits and Costs of Social Pensions in Sub-Saharan Africa.” Social Protection and Labor Discussion Paper 1607. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/325281469593828257/pdf/107234-WP-add- series-PUBLIC.pdf Harvey, Philip. 2007. “U.S. Job Creation Programs in the 1930s.” Background Paper for the Human Sciences Research Council Mid-Term Review of South Africa’s Expanded Public Works Program. Hickey, Sam, Tom Lavers, Miguel Niño-Zarazúa, and Jeremy Seekings, eds. Forthcoming. The Pol- itics of Social Protection in Eastern and Southern Africa. Oxford, UK: Oxford University Press. Hidrobo, Melissa, and Shalini Roy. 2019. “Cash Transfers and Intimate Partner Violence.” Blog post March 29. VoxDev. https://voxdev.org/topic/public-economics/cash-transfers-and-intimate- partner-violence Hoynes, Hilary W., and Jesse Rothstein. 2019. “Universal Basic Income in the US and Advanced Countries.” NBER Working Paper 25538. National Bureau of Economic Research, Cambridge, MA. https://gspp.berkeley.edu/assets/uploads/research/pdf/Hoynes-Rothstein-UBI-081518.pdf ILO (International Labour Office). 2018. “Social Protection for Older Persons: Policy Trends and Statistics 2017–19.” Social Protection Policy Paper 17. ILO, Geneva. ILO and UNICEF (International Labour Office and United Nations Children’s Fund). 2019. Towards Universal Social Protection for Children: Achieving SDG 1.3. UNICEF: New York. https://www. social-protection.org/gimi/gess/RessourcePDF.action?id=55472 Imbert, Clément, and John Papp. 2015. “Labor Market Effects of Social Programs: Evidence from India’s Employment Guarantee.” American Economic Journal: Applied Economics 7 (2): 233–63. IMF (International Monetary Fund). 2017. Fiscal Monitor: Tackling Inequality. Washington, DC: IMF. https://www.imf.org/en/Publications/FM/Issues/2017/10/05/fiscal-monitor-october-2017 Kidd, Stephen, and Diloá Athias. 2019. “Hit and Miss: An Assessment of Targeting Effective- ness in Social Protection.” Working Paper. Development Pathways, London. https://www. developmentpathways.co.uk/wp-content/uploads/2019/03/Hit-and-Miss-March13-1.pdf Kluve, Jochen, Susana Puerto, David Robalino, Jose Manuel Romero, Friederike Rother, Jonathan Stöterau, Felix Weidenkaff, and Marc Witte. 2016. “Do Youth Employment Programs Improve 96 Ugo Gentilini and Margaret Grosh Labor Market Outcomes? A Systematic Review.” IZA Discussion Paper 10263. Institute for the Study of Labor, Bonn. http://ftp.iza.org/dp10263.pdf Lustig, Nora, ed. 2018. Commitment to Equity Handbook: Estimating the Impact of Fiscal Policy on Inequality and Poverty. Washington, DC: Brookings Institution Press. Maag, Elaine. 2017. “Refundable Credits: The Earned Income Tax Credit and the Child Tax Credit.” Tax Policy Center, Washington, DC. https://www.urban.org/sites/default/files/ publication/89171/2001197-refundable-credits-the-earned-income-tax-credit-and-the-child- tax-credit_0.pdf Majoka, Zaineb, and Robert Palacios. 2019. “Targeting versus Universality: Is There a Middle Ground?” Unpublished. World Bank, Washington, DC. McCord, Anna Gabriele. 2012. Public Works and Social Protection in Sub-Saharan Africa: Do Public Works Work for the Poor? New York: United Nations University Press. Moffitt, Rober. 2003. “The Negative Income Tax and the Evolution of U.S. Welfare Policy.” NBER Working Paper 9751. National Bureau of Economic Research, Cambridge, MA. https://www. nber.org/papers/w9751.pdf Muralidharan, Karthik, Paul Niehaus, and Sandip Sukhtankar. 2017. “General Equilibrium Effects of (Improving) Public Employment Programs: Experimental Evidence from India.” NBER Working Paper 23838. National Bureau of Economic Research, Cambridge, MA. Özler, Berk. 2017. “Fact Checking Universal Basic Income: Can We Transfer Our Way Out of Poverty?” Blog post February 27. Development Impact. https://blogs.worldbank.org/impactevaluations/ fact-checking-universal-basic-income-can-we-transfer-our-way-out-poverty Packard, Truman, Ugo Gentilini, Margaret Grosh, Philip O’Keefe, Robert Palacios, David Robalino, and Indhira Santos. 2019. Protecting All: Risk Sharing in a Diverse and Diversifying World of Work. Washington, DC: World Bank. https://hubs.worldbank.org/docs/imagebank/pages/ docprofile.aspx?nodeid=31393804 Ravallion, Martin. 2018. “Guaranteed Employment or Guaranteed Income?” CGD Working Paper 482. Center for Global Development, Washington, DC. http://dx.doi.org/10.2139/ssrn.3208865 —. 2019. “Is a Decentralized Right-to-Work Policy Feasible?” NBER Working Paper 25687. National Bureau of Economic Research, Cambridge, MA. Seekings, Jeremy. 2013. “The Beveridge Report, the Colonial Office, and Welfare Reform in British Colonies.” Unpublished. University of Cape Town. Steensland, Brian. 2007. The Failed Welfare Revolution: America’s Struggle over Guaranteed Income Policy. Princeton: Princeton University Press. 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. http://documents.worldbank.org/curated/en/289871468339050938/pdf/Public-works-as-a- safety-net-design-evidence-and-implementation.pdf Tcherneva, Pavlina. 2018. “The Job Guarantee and the Economics of Fear.” One Pager 55. Levy Economics Institute, New York. http://www.levyinstitute.org/pubs/op_55.pdf U.S. Treasury. 2017. “The Distribution and Evolution of the Social Safety Net and Social Insur- ance Benefits, 1990 to 2014.” Issue Brief Four. Washington, DC. https://www.treasury.gov/ resource-center/economic-policy/Documents/The%20Economic%20Security%20of%20 American%20Households-%20the%20Safety%20Net.pdf Veras Soares, Fábio, and Ian Orton. 2017. “Graduation: An Overview.” Policy in Focus 14 (2): 7–10. https://www.ipc-undp.org/pub/eng/PIF39_Debating_Graduation.pdf Chapter 2. UBI as Social Assistance: Comparative Models and Instruments 97 World Bank. 2013. World Development Report 2013: Jobs. Washington, DC: World Bank. https:// openknowledge.worldbank.org/handle/10986/11843 —. 2018. The State of Social Safety Nets 2018. Washington, DC: World Bank. https:// openknowledge.worldbank.org/bitstream/handle/10986/29115/211254.pdf?sequence=4 —. 2019. “A Quantitative Evaluation of the Greek Social Solidarity Income.” World Bank, Washington, DC. http://documents.worldbank.org/curated/en/882751548273358885/ pdf/133962-WP-P160622-Evaluation-of-the-SSI-Program-Jan-2019.pdf Zimmermann, Laura. 2013. “Why Guarantee Employment? Evidence from a Large Indian Pub- lic-Works Program.” University of Michigan, Ann Arbor, MI. http://www-personal.umich. edu/~lvzimmer/Zimmermann_NREGS_current_draft.pdf Universal Basic Income and Work CHAPTER 3 Francesca Bastagli T he recent rise in interest in universal basic income (UBI) reflects a growing con- cern that the economy and existing welfare policies are failing to deliver for parts of the population. In the world of work, the rise in nonstandard employment and work insecurity, high informality, and the threat of technological unemployment highlight the limitations of existing policies. Other challenges and shortcomings are not new. For instance, assumptions about gendered roles commonly underpinning social protection schemes have historically resulted in higher exclusion from social protection and weaker provision for women compared to men, in many cases reinforcing gender inequalities in the world of work. One of the main achievements of this recent renewed interest in UBI is that it has thrown open questions that, in some circles at least, had been considered closed. These range from fundamental questions about the nature and value of work, to the role of the regulation of labor relations, to the specifics of how a UBI could address the limitations and unintended effects of existing social protection schemes. In terms of conceptualiza- tion, the UBI debate questions the near-exclusive focus (in both research and policy) on paid work and financial incentives, and brings back to the fore the issues of type and quality of work. In terms of policy design and implementation choices, the debate ques- tions how alternative options—for instance, concerning targeting and conditionality and treatment unit (individual or household)—influence work outcomes. Against this backdrop, this chapter explores key questions the UBI debate raises with respect to what are commonly labeled “behavioral” individual- and household-level work outcomes: Would a UBI lead to reductions in people’s participation in paid work? By providing workers with an exit option and strengthened bargaining power, could it lead to 99 100 Francesca Bastagli improved conditions of paid work? Could a UBI free up time to take on work that is unpaid but valued by individuals and society? By weakening incentives to formalization, would a UBI contribute to a rise in informal work? Based on a review of both theoretical arguments and available evidence, the chapter aims to contribute to this lively and welcome debate. More specifically, the chapter explores claims about how the defining features of a UBI—its universal, unconditional, individual nature and periodic payments (Van Parijs 2004)—influence work outcomes. It pays special attention to the role played by varia- tions in proposed UBI schemes, for instance, concerning the level of the basic income and a scheme’s articulation within broader social and labor market policy. By avoiding targeting and conditionality, by essentially divorcing eligibility for social protection trans- fers from any labor contribution, a UBI could influence work outcomes differently than alternative cash transfer schemes. Similarly, a UBI’s payments to individuals rather than households are expected to make a difference. Such issues are explored here, with refer- ence to both theory and practice. As this chapter highlights, and to no surprise, variations in the details of program design and a scheme’s positioning within wider social and labor policy are key determinants of policy outcomes. The Arguments Four broad sets of work-related outcomes are covered here: •• Participation in paid work and financial work incentives. UBI critics warn that the unearned income from a UBI would lead people to work less. Compared to transfers with work conditions and job search and training requirements, a UBI would additionally weaken people’s participation in paid work. At the same time, compared to a means-tested cash transfer, the universal nature of the UBI could weaken the work disincentive generated by a means test, especially when the latter displays a high marginal tax rate or benefit withdrawal rate. Concerns about people reducing their time in paid work as a result of additional unearned income imply this is a negative outcome. As highlighted by the UBI debate, and discussed further below, such a result could in fact be linked to increased indi- vidual and societal welfare. •• Conditions of paid work. UBI advocates argue that guaranteeing everyone an unconditional income floor potentially enables workers to turn down insecure, low-paid, exploitative work or demand improved work conditions by granting them an exit option from such work or employment relations. Conversely, con- cerns are raised that a UBI could act as a subsidy to low wages and make low pay more acceptable. A UBI could promote casual work and job insecurity by increasing the supply of labor for insecure jobs. •• Valuation and distribution of unpaid work. Much of the discussion on cash trans- fers and work incentives focuses on paid work and the concern that a UBI could Chapter 3. Universal Basic Income and Work 101 reduce incentives to take up or spend more time in paid employment. However, this misses the consideration that a UBI could free up time to take up work that is unpaid, including work that may be valuable to society or the individual, but is not recognized as such in terms of monetary compensation in the labor market. In the case of domestic and care work, still disproportionately carried out by women and girls, advocates argue that a UBI could support its redistribution between the sexes. At the same time, critics warn a UBI would reinforce the gen- dered division of work, for instance, by increasing the incentive for women to reduce their participation in paid work given their relatively weaker attachment to the labor market compared to men. •• Formal and informal work. The universal and unconditional nature of a UBI weak- ens the link between labor market status and social protection entitlements. As such, it could weaken the incentive to move toward formal employment. In the case of means-tested and conditional noncontributory cash transfers, beneficia- ries’ perceptions that they could lose their transfer entitlement by working may generate an incentive to stay in or take up informal work, avoiding “visible” work. The universal and unconditional nature of a UBI could weaken concerns that formal work could lead to loss of benefit eligibility. The Nature of the Evidence The limited implementation of full UBI schemes to date means there is limited direct empirical evidence of the implications of a UBI for the outcomes of interest here. Although this poses a challenge, this chapter maintains there is much to be learned of direct relevance to the UBI debate from the experience of other cash transfers, includ- ing basic income–type experiments, negative income tax schemes, as well as targeted and conditional cash transfers. The chapter reviews available evidence arising from the implementation of UBI schemes, such as the Alaska Permanent Fund Dividend and the Islamic Republic of Iran’s Price Subsidy Reform; from negative income tax schemes, including the 1960s–80s experiments in the United States and Canada; and from tar- geted and conditional cash transfers worldwide (see also chapter 4). Throughout, in recognition of the variety of UBI proposals in practice and of the degrees of targeting and types of conditionality adopted, special attention is paid to identifying how differences in such design and implementation parameters influence work outcomes. Two qualifying points need to be made with regard to the evidence and how it is interpreted. First, as mentioned above, the majority of experiments and pilots that are labeled as “basic income” generally do not comply even with the fundamental core parameters of a UBI outlined above (e.g., unconditional, no means test, individual). For example, the majority of basic income pilots and experiments have elements of tar- geting. The negative income tax experiments in Canada and the United States in the 1960s–80s had an income-related eligibility threshold (Hum and Simpson 1993). They 102 Francesca Bastagli also mostly focused on particular family structures (e.g., families with two children) and particular ages. In the contemporary Ontario pilot, only those earning less than the full value of the basic income payments are eligible for the scheme (Young 2018). The recent Finnish partial basic income experiment (January 2017–December 2018) paid a transfer solely to the unemployed, specifically those who were in receipt of unemployment ben- efits prior to the experiment (Kangas et al. 2019). Second, conversely, cash transfers that are not considered to share much with a UBI may actually have some commonalities as a result of design and implementation details. A case in point is provided by means-tested conditional and unconditional cash transfers across low- and middle-income countries. In some cases, even when a transfer has a clear income-related eligibility threshold, delays or failure to recertify and update information over time lead to a delinking of changes in beneficiaries’ circumstances with transfer receipt, in practice weakening the targeting element of the scheme. This is not a systematic review. However, every effort is made to include the results from relevant rigorous studies, including ones that estimate policy impact via counter- factual analysis relying on experimental or quasi-experimental approaches as well as qualitative studies relying on smaller samples of respondents (e.g., to capture informa- tion on process and perceptions on outcomes of interest that may not be covered by counterfactual analyses). This chapter mostly focuses on evidence from micro-, individual-, and house- hold-level analysis of the impact of cash transfers. For the limited number of studies available on UBI schemes (mostly the Alaska Permanent Fund Dividend), it reports any available evidence on aggregate measures such as state-level wage rates and number of employed. Since the universal nature of a UBI poses a challenge to identifying individu- al-level effects via counterfactual analysis, available studies rely on aggregate measures. Finally, the chapter covers schemes in low-, middle-, and high-income countries. Variations in the labor markets (e.g., share of unemployed, self-employed) and finan- cial markets across these countries are significant and have implications for the barriers and opportunities faced by individuals, which in turn shape the effects of cash trans- fers. Throughout, attention is paid to context and how this influences the impact of cash transfers on work-related outcomes. Participation in Paid Work and Financial Work Incentives Much of the debate on UBI and work is framed in terms of work incentives, with a focus on paid work and financial incentives. Cash transfers paid directly to individuals or households are commonly criticized for giving rise to adverse incentive effects, contrib- uting to unemployment and slack economic growth (e.g., on Europe, see Immervoll and O’Donoghue 2002). Standard economic theory predicts that the additional unearned income via a cash transfer will negatively affect people’s participation in paid work Chapter 3. Universal Basic Income and Work 103 (extensive margin) and number of hours worked (intensive margin) via a direct income effect. The additional unearned income would permit recipients to spend the extra income on activities other than paid work. Furthermore, elements of targeting and con- ditionality in a cash transfer may affect labor supply via a substitution effect. In the case of means testing, for example, the income or asset eligibility threshold may generate an incentive for people to maintain their incomes/assets below such a level in order to qual- ify for a transfer—for instance, by reducing their work effort (Atkinson 1995). Compared with targeted schemes, a UBI paid to all could in principle strengthen the income effect by extending coverage to those not receiving means-tested transfers. In principle, a UBI additionally weakens the link to the labor market provided by conditional- ities on work participation. Indeed, the introduction of conditionality in the form of work requirements for cash transfer recipients and related in-work benefit reforms in countries such as the United Kingdom and the United States was motivated by work incentive and labor supply concerns. Policy developments in these countries in the 1990s, and again more recently, focused on increasing participation in paid work among cash transfer recipients and claimants through the introduction of work requirements and welfare-to- work reforms (e.g., see Blundell and Macurdy 1999; Brewer 2003; Moffitt 2002). At the same time, a UBI paid to everyone would weaken the work disincentive effect generated by means testing, potentially supporting an increase in work participation. A cash transfer’s effect on participation in paid work is expected to vary depend- ing on individual characteristics and by population subgroup. The income effect may be strong for those with weak labor market attachment and/or low earning potential. In couples, and in the case of cash transfers means tested at the household level, individu- als whose partners earn relatively high wages may face high disincentives as additional work effort could disqualify them from transfers means tested at the household level. Critically, cash transfer design details—such as whether transfers are paid at the individ- ual or household level, as just mentioned, and the level of the transfer—are expected to matter, with a generous transfer potentially associated with higher work disincentive effects. Concerns about the potential negative impact of UBIs/cash transfers on participa- tion in paid work are commonly based on a number of assumptions. Three of these, discussed briefly here, provide examples of why we may observe departures from what standard economic theory predicts in empirical investigations of the effects of cash transfers on participation in paid work. First, there is typically an underlying assumption of functioning labor and financial/credit markets. In practice, however, contexts are often ones in which markets for financial services and labor are difficult to access or do not function well. Particularly in such contexts, we could expect that the regular additional income of a cash transfer would help tackle barriers to work and productive investments that affect people’s work opportunities and ability to work (Banerjee et al. 2015; Bastagli et al. 2016). Second, the focus on financial work incentives tends to omit that there are of course other dimensions to cash that will be equally or more relevant for decisions of whether to seek or stay in paid work—among them, nonfinancial rewards. Third, a common assumption is that policy implementation follows on directly from policy 104 Francesca Bastagli design; for instance, a transfer displaying a high marginal tax rate by design will lead to one in practice. However, policy implementation may depart from policy design, and such departures may play an important role in shaping impact in practice. Results from studies of two UBI schemes implemented to date—Alaska’s Perma- nent Fund Dividend and the Islamic Republic of Iran’s cash transfer—yield evidence on the labor supply effects of such transfers. They find that overall, the regular payment of a universal cash transfer does not lead to a significant impact on participation in paid work. There is some evidence for Alaska of an increase in part-time work associated with the dividend. For population subgroups for which a reduction in participation in paid work was recorded (specifically, Iranian youth), this was linked to a shift in time use toward other valued activities. Overall, the available evidence suggests that an uncondi- tional income floor generated no general significant disincentive to work. For the Alaska Permanent Fund Dividend, all available studies report effects on aggregate employment measures (e.g., state-level wage rates, number of employed and unemployed, early retirement rates).1 Jones and Marinescu (2018) use data from states that match Alaska in terms of outcomes of interest to study the impact of the dividend on employment-to-population ratios. They find the employment rates in the state match those of the control states. However, the share of Alaskans working part time increased by 2 percentage points, indicating that the dividend led to an intensive margin response: people worked less hours given the additional income. Earlier studies look at compara- tive trends in aggregate state-level wage rates, the number of employed and unemployed, and early retirement (Goldsmith 2010, 2012). They find no or limited reported reduc- tions in time spent in paid employment and no evidence of impact on early retirement, as labor force participation has continued to trend upward despite the growth of num- bers of people age 65 and over (Goldsmith 2012). The researchers attribute these findings for Alaska’s UBI to two factors: the compar- atively low value of the transfer and the potential for the macro effects of the dividend to have countered the scheme’s income effects. The potential income effect is balanced by the macro effect, whereby any decrease in the labor supply would increase the equi- librium wage rate; in turn, the dividend increases consumption, which stimulates labor demand, thus increasing wages and employment (Goldsmith 2012; Jones and Marinescu 2018). A study of the Islamic Republic of Iran’s universal cash transfer (Salehi-Isfahani and Mostafavi-Dehzooei 2017) examines its impact on labor force participation and hours of work by Iranian men and women. It finds that the UBI does not result in a negative labor supply effect for either hours worked or the probability of participation in market work, either for all workers or those in the bottom 40 percent of the income distribu- tion. However, they do find a negative effect on the number of hours worked for workers 20–29 years old. The authors reflect that this is likely due to the weak attachment of Ira- nian youth to the labor market and the option for many to enroll in tertiary and graduate education. The U.S. and Canadian guaranteed income experiments of the 1960s–80s explic- itly tested the labor supply effects of negative income taxes.2 Overall, the experiments Chapter 3. Universal Basic Income and Work 105 find evidence of no effects or moderate reductions in work participation in some cases. Where the latter were recorded, reduction rates were higher for women. According to Burtless’s review of the results of studies from the four U.S. experiments, the only consistently negative and statistically significant result arises from the Seattle-Den- ver experiment, with the highest average generosity of transfer (Burtless 1986).3 The Manitoba Basic Annual Income Experiment (Mincome) in Canada recorded a modest reduction in number of hours worked—1  percent for men, 3  percent for wives, and 5 percent for unmarried women—which is statistically insignificant when time effects are controlled for (Hum and Simpson 1993). In sum, the negative income tax experiments find few adverse effects and no evidence of high numbers of workers reducing their work effort; this suggests that an unconditional income floor delivered as a negative income tax generated no or a moder- ate disincentive to work. Where reductions in work effort were recorded, these may have been the result of people’s improved balancing of work and home lives/paid and unpaid work and a shift in time use toward other valued activities (e.g., as reflected in results for women’s reductions in participation in paid work), leading potentially to increased indi- vidual and societal welfare (Widerquist 2005). These studies do not allow us to draw conclusions about the extent to which the absence of a clear and significant effect on participation in paid work is the result of the universal and unconditional nature of the UBI and negative income tax. Empiri- cal investigations into the labor supply effects of targeted and/or conditional transfers provide some indication of the role of such features in influencing work participation outcomes. A common approach to empirical investigations of a cash transfer’s potential work (dis)incentive effect relies on measures such as the participation tax rate (measures the proportion of gross earnings lost through tax and/or benefit withdrawal) and the marginal effective tax rate (measures the propor­ tion of each additional unit of earned income lost to tax and/or benefit withdrawal). Such measures provide an indication of a transfer’s potential work incentive effects. High participation tax rates and marginal effective tax rates describe circumstances in which people face little financial incentive to take on paid work or work longer hours (Atkinson 1995; Immervoll and O’Donoghue 2002; Martinelli 2017). One of the potential advantages of a UBI is that it avoids the disincentive effects associated with means testing and benefit withdrawal as captured in marginal tax rate and participation tax rate measures. In both the United Kingdom and the United States, studies indicating the high marginal tax rates generated by means-tested schemes and related concerns for potential negative labor supply effects contributed to the introduc- tion of work requirements and time limits such as those introduced in the 1990s to the U.S. Aid to Families with Dependent Children program, the country’s main means- tested cash transfer paid mostly to single-mother families (Moffitt 2002). Now known as the Temporary Assistance for Needy Families, the scheme requires recipients to work a minimum number of hours per week and includes benefit reduction penalties if these requirements are not met, including termination from the program. 106 Francesca Bastagli The expansion of targeted (mostly means-tested or proxy means–tested) cash transfers over the last two decades in countries worldwide was accompanied by a flurry of impact evaluations examining work and labor supply effects. Based on recent sys- tematic reviews, the evidence indicates that, on the whole, such noncontributory social assistance schemes lead to no or limited effects on the adult labor supply. With respect to work participation (extensive margin) of working-age adults, the evidence indicates that cash transfers have either no impact or limited positive impact, with cash recipients more likely to be working. On number of hours worked (intensive margin), the evidence is mixed, with examples of both increases in numbers of hours worked and decreases as a result of cash transfer receipt. Where a reduction in work is observed, it is in rela- tion to reductions in casual wage labor among those of working age, in paid work by women with care responsibilities, and—in the case of social pensions provided to the elderly—an associated reduction in the elderly working for pay (e.g., Baird, McKenzie, and Özler 2018; Banerjee et al. 2015; Bastagli et al. 2016; Bosch and Manacorda 2012; Owusu-Addo, Renzaho, and Smith 2018). Bastagli et al. (2016) review the evidence spanning 15 years (2000–15), from 165 low- and middle-income countries and find that cash transfers (noncontributory, targeted) have either no effect or a positive effect on adult labor force participation. Out of eight studies reporting on cash transfer impact on work participation for adults of working age, four find statistically significant impacts, three being increases and one a decrease.4 In terms of the intensity of adult work, again, half the studies find the cash transfers reviewed to have no significant effect. Among those that do, three studies find increases and three find decreases. Among those with decreases, one was the result of a social pen- sion in Brazil allowing elderly individuals to reduce time in paid work,5 another was only significant among those who had not yet received a second transfer that was due,6 and the third finds a reduction in hours worked in casual wage labor.7 Similar findings emerge from other recent reviews. Banerjee et al. (2015) analyze data from seven randomized controlled trials of government cash transfer programs (non- contributory, targeted). Across the seven programs reviewed (based on results from 21 studies, covering 17 conditional or unconditional cash transfer programs that do not have explicit work requirements for the poor in six countries), they find no systematic evidence that cash transfer programs discourage work. Also, they find no observable impacts of cash transfer programs on either the propensity to work or the overall number of hours worked, for either men or women. Baird, McKenzie, and Özler (2018) find that, overall, cash transfers that are made without an explicit employment focus (such as conditional and unconditional cash transfers and remittances) tend to result in little to no change in adult labor. The main exceptions are transfers to the elderly and some refugees, who reduce work. In contrast, transfers made for job search assistance or business start-up tend to increase adult labor supply and earnings, with the likely main channels being the alleviation of liquidity and risk constraints. Owusu-Addo, Renzaho, and Smith (2018) review 53 studies on conditional and unconditional cash transfers. Seven studies report the impact of cash transfers on adult labor force participation, with only one of them showing a significant effect. The Zambian Child Grant Programme resulted in a significant Chapter 3. Universal Basic Income and Work 107 decrease in adult labor force participation in wage labor outside the household (9 percent- age points). This was primarily driven by a shift from agricultural wage labor to family agricultural business. The effect was stronger for households with females within the working-age group compared with households with males in this age group (Daidone, Davis, Dewbre, and Covarrubias 2014). In their review of cash transfers in Latin America and the Caribbean, Bosch and Manacorda (2012) find that noncontributory cash transfers have no large significant effects on participation and overall employment (both the inten- sive and extensive margins), other than possibly among the elderly. The apparent absence of a significant large effect of means testing of cash trans- fers, even when they generate a high marginal effective tax rate on paper (e.g., if based on a clear income threshold and a unit increase in income by design leading to loss of eligibility), is in many cases attributed to the implementation of targeting in practice not leading to high marginal effective tax rates. Reliance on proxy means testing in many of the Latin America and the Caribbean conditional cash transfers may also weaken this poten- tial effect by weakening the link between program eligibility, earnings, and current income levels. In many of the conditional cash transfers covered in the systematic reviews summa- rized above, in practice participation in a program and the level of transfers are not affected by people’s work decisions or household income level. For example, even though in theory, according to program regulation, beneficiary eligibility status is reexamined at regular inter- vals, such verifications may not happen in practice or may take place after several years. For example, in Mexico’s PROGRESA (Programa de Educación, Salud y Alimentación) pro- gram, the eligibility status of households was supposed to be reviewed within three years after a household’s entry into the program. In fact, more than five years elapsed before any effort was made to revise the list of beneficiaries (Skoufias and Di Maro 2008). A recurrent theme in discussions of the reasons behind the observed labor supply effects of cash transfers concerns the level/value of the cash transfer and the duration of the transfer payment. Observed limited or lack of transfer effects on labor supply are commonly explained in terms of the low value of the transfer (e.g., Bastagli et al. 2016). Studies comparing work participation effects of transfers with different values find mixed evidence: either no differential effects (e.g., Novella et al. 2012 find that despite large differences in transfer sizes in Honduras’s Programa de Asignación Familiar, Mex- ico’s PROGRESA, and Nicaragua’s Red de Protección Social any—limited—change in labor supply is not correlated with the size of the transfer) or limited evidence that higher transfer amounts are associated with a negative impact on the number of hours worked (e.g., as outlined by the U.S. negative income tax experiment above; Burtless 1986). The duration of the transfer may also matter, as behavioral adjustments may begin to take place only once a program is institutionalized and payments regularly made in a predictable fashion over time. In the studies reviewed above, the experimental or short-term nature of the transfers covered is commonly mentioned as another potential reason for the limited or absent income effect of the cash transfer. Available evidence suggests, however, that transfer duration and permanency need not be associated with modified effects on work participation. A longitudinal study by Ardington, Case, and Hosegood (2009) of South Africa’s old-age pension is a case in point, with a large (paying 108 Francesca Bastagli more than twice the median per capita income) and stable transfer resulting in increased work participation among prime-aged adults in recipient households. Conditions of Paid Work In addition to influencing participation in paid work, cash transfers may affect the con- ditions of paid work and type of paid work carried out. The additional income accrued through a cash transfer may help tackle the barriers to better work—for instance, through productive investments. It may also support people in refusing insecure work and low pay, and poor working conditions or employment relationships, and in demanding improved terms of work. Such potential is especially strong with a UBI, some argue, by guaranteeing everyone an unconditional income floor. A UBI could “prevent conditions of exploitable dependency and vulnerability to abuse” (Birnbaum and De Wispelaere 2016), by provid- ing them with an alternative source of income to paid work with no work conditionality requirement. By offering an option to exit the labor force, a UBI would strengthen work- ers’ negotiating power and voice (Calnitsky 2018; Widerquist et al. 2013). A different view argues that a basic income mostly facilitates the maintenance of poor work conditions and low pay and acts as a subsidy to employers. According to this perspective, a UBI would enable employers to reduce wages, since workers effectively receive a top-up, and weaken the bargaining position of some workers. If a UBI increased the supply of labor for low-paid and insecure work, this would be expected to have an adverse impact on remuneration and terms of employment. The UBI would make it easier for people to be hired into exploitative positions—casual or low-paid work—and would in effect be a subsidy for such types of work (Gray 2017; Harrop and Tait 2017). Again, policy design is expected to make a difference. For example, the value of the transfer and wider labor regulation are expected to influence whether a UBI offers an exit option in practice. A basic income that is high enough to enable people to refuse low- pay or insecure work may achieve just that; a basic income at a lower level could have the opposite effect, making adverse work conditions including low pay more acceptable. Labor regulations on working time and guaranteeing a minimum wage could help ensure basic pay and work conditions are not rolled back and address the risk that a UBI could increase casual, insecure, and low-paid work (Gray 2017). What does the evidence show? A study of the impact on wages (wage rates offered on advertised job vacancies and actual wages on new hires) of Manitoba’s three-year Mincome guaranteed income scheme finds that the guaranteed income led to a consid- erable increase in wages. Calnitsky (2018) finds the improved work outcomes were the result of the guaranteed income providing workers with a threat of exit from the labor market and, importantly, are contingent on a few crucial policy details: the absence of conditionality in the form of work requirements and the fact that the scheme did not replace existing welfare provisions and/or labor regulations. Evaluations of the impact of targeted and unconditional cash transfers (to poor households with children) on workers engaged in agricultural labor find that such Chapter 3. Universal Basic Income and Work 109 schemes either have no impact on the type of work carried out or, when a significant effect is observed, lead to a reduction in paid casual and occasional work and shifts in the type of work considered to reflect improvements in work conditions and to be wel- fare enhancing (Bastagli et al. 2016). In the case of Lesotho’s Child Grant Programme, an unconditional cash transfer targeted to poor and vulnerable households with children, Daidone, Davis, Dewbre, and Covarrubias (2014) find that the transfer reduces the intensity of adult participation in paid occasional and irregular work—particularly piecework labor, which is generally con- sidered a negative coping mechanism in times of hardship. In a later study, Dewbre et al. (2015) find additional resources provided by the Child Grant Programme and the Linking Food Security to Social Protection Programme led to an increase in own-crop activities for the labor unconstrained and a decrease in temporary wage work. A shift from occa- sional agricultural wage labor to own-farm labor is considered welfare enhancing. A study of the impact of Zambia’s Child Grant Programme on type of employment finds a reduction in agricultural wage labor (Daidone et al. 2014). Agricultural wage labor is generally considered the labor activity of last resort, and when liquidity constrained, households may be obliged to overly depend on it. This is accompanied by evidence of a small increase in permanent non-agricultural wage employment for females and an increase in time spent working on family nonfarm enterprises as well as on own-farm work. For Pakistan’s Benazir Income Support Programme (BISP), Cheema et al. (2014) find that the transfer is associated with a clear and significant reduction in casual labor among working-age adults and an associated increase in the proportion of men engaged in agricultural activities, including caring for livestock. Male members are redirecting their labor toward two types of agriculture: sharecropping and own-agriculture, which includes own-cultivation and livestock breeding, as the BISP enabled some households to purchase small livestock. In his study of the Madhya Pradesh and Tribal Village unconditional cash transfer pilots in India, Standing (2013) finds that cash transfer receipt enabled changes in the terms of labor market engagement. Access to transfers enabled recipients to move out of the most exploitative forms of employment and into potentially more sustainable work, and promoted movement out of casual wage labor (where households were sometimes trapped in bonded labor or caught in interlocked markets for labor, land, and credit) and into own-account activities. Skoufias and Di Maro (2008) find that conditional cash transfers in Mexico enable beneficiaries, at least initially, to move from low-paid family business jobs to salaried jobs. Valuation and Distribution of Unpaid Work By paying an income to individuals independently of people’s participation in paid work, a UBI could free up time for people to take on work of low or no monetary value that is, however, valued by individuals and society in other ways. This would help recognize 110 Francesca Bastagli unpaid work and could in principle facilitate its redistribution, for instance, by allowing individuals engaged in paid work to spend more time in unpaid work. In the specific case of unpaid care work in contexts where care services are not publicly provided, the additional income of a UBI could also be spent on paying for such services, thus freeing up time for people engaged in unpaid domestic work to take on paid work (Leibbrandt et al. 2013). As women are commonly the primary unpaid domestic and care work providers (ILO 2018), this could have implications for women’s participation in paid work. A UBI’s distinguishing features—specifically its payment to individuals (as opposed to households) and its unconditional nature—would be critical in this respect. Targeted cash transfers that define eligibility based on household income are expected to generate incentives for secondary earners in a dual-earning couple to reduce their participation in paid work to help secure eligibility. With regard to conditionality, a UBI, by remain- ing neutral regarding the activities in which individuals engage, could avoid potential negative effects arising from imposing conditions either in terms of care work (and potentially reinforcing the existing uneven distribution of unpaid work) or participation in paid work (and associated penalties for differentiating between eligible and non-eligi- ble in terms of labor market status). In the case of conditional cash transfers with behavioral requirements set in terms of children’s regular school attendance and health care visits, adults’ time and task allo- cation may be influenced both directly and indirectly. As is common practice in such conditional cash transfers, when they are specifically paid to women in households with children, schemes may reinforce women’s status as mothers and primary care provid- ers (Cookson 2018; Molyneux 2009). By making mothers explicitly responsible for child care, such conditional cash transfers can reinforce women’s caregiving role, reproducing one of the underlying causes of economic gender inequality (ECLAC 2013), and working against the redistribution of unpaid work. Conditionalities of this sort can also affect time and task allocation indirectly, for example, as children alter their time in paid and unpaid work as they increase school attendance, leading to an increase in unpaid domestic workload for adults. On the other hand, a reduction in remunerated child labor outside the household may lead to increased adult participation in paid work to compensate for the associated income loss. Additionally, there may be indirect effects as child and adult health improves and caring responsibilities for the sick or infirm are reduced (Barrientos and Villa 2015; Kabeer, Piza, and Taylor 2012). On the gendered distribution of labor, feminist advocates for basic income highlight its potential to correct the paid-work bias of contemporary social protection schemes and increase women’s economic autonomy and power by providing a floor of economic security for everyone and not specifying the activities in which they engage—thus help- ing to recognize the unpaid work largely done by women without reinforcing existing inequalities between men and women (McLean 2016). On the other hand, UBI critics have argued that such schemes would do nothing to directly challenge gendered divi- sions—and may well reinforce them, “especially to the extent that unconditional cash Chapter 3. Universal Basic Income and Work 111 transfers increase the incentive for women in particular to reduce their labor market par- ticipation, given their relatively weaker attachment to the labor force as a group relative to men, and the central role that this plays in broader inequalities such as income gaps and poverty risks” (McKay 2001). Such arguments underscore the need for basic income schemes to be accompanied by wider efforts to address inequalities in the gendered dis- tribution of work in support of its potential for addressing these differences. Empirical investigations into the effects of unconditional cash transfers on par- ticipation in paid work provide some insights into the potential for cash transfers to differentially affect men and women, reflecting variations in time use, distribution of unpaid work, and labor market attachment. Two studies of Germany’s Kindergeld (paid to all households with a child below the age of 17) reveal the scheme is associated with a significant reduction in number of hours worked (weekly working hours) among married mothers (Hener 2016; Tamm 2009) and no effect on fathers’ labor supply (Hener 2016). Among single mothers too, Kindergeld is associated with a reduction in number of hours worked (smaller effect compared with married mothers) (Tamm 2009). For both groups of women, the univer- sal child grant does not have an effect on the extensive margin of work participation. Descriptive evidence provided by the authors suggests mothers are spending additional hours on child care in the home. Studies of labor supply responses of married women with children to means-tested cash transfers explore whether and how these vary depending on their partners’ earn- ings and time spent in paid work—reflecting issues of labor market attachment and policy design, especially regarding means testing. In a study of the U.S. earned income tax credit (EITC), Eissa and Hoynes (2004) find the scheme leads to a decline in labor force participation by married women (by just over 1 percentage point), while the labor market participation of married men increases (by about 0.2 percentage points). They argue the EITC effectively subsidizes married mothers to stay at home, because the EITC is based on family earnings as opposed to individual earnings. A study of the U.K. work- ing families tax credit (Francesconi, Rainer, and van der Klaauw 2007) finds that, for women with a partner who did not work or worked fewer than 16 hours per week, the scheme increased the probability of working 16 or more hours per week by 3 percentage points, while increasing the full-time employment rate by 2 percentage points. Based on the disaggregated analysis by income and time spent in work by women’s partners, they find no statistically significant response among women with wealthier partners. They also find no labor supply response among men. These results contrast somewhat with results on the impact of cash transfers on single women with children. Both for the U.S. EITC and the U.K. working families tax credit, studies find an increase in participation in paid work among single women with children. Eissa and Liebman (1996) find that single women with children increase their relative labor force participation by up to 2.8 percentage points (and observe no change in number of working hours of single women with children and already in the labor force). Francesconi, Rainer, and van der Klaauw (2007) find the working families tax credit reform leads to a substantial increase in the employment rate (of about 5 percentage 112 Francesca Bastagli points) among single mothers. Their study highlights the potential role of increased utili- zation of child care (including because of the child care credit component of the working families tax credit reform) in accompanying the increase in women’s employment. This finding is echoed in a study by Eyal and Woolard (2011) of South Africa’s Child Support Grant, which led to significantly higher labor force participation among mothers in their 20s—thanks, in part, to utilization of the transfer to pay for daycare/schooling. Two studies of conditional cash transfers paid to women, with behavioral require- ments on children’s regular school attendance and health care visits, and their effect on the type of work carried out by women find that women spend more time on domestic work as a result of the cash transfer.8 Rubio-Codina (2009) finds Mexico’s Oportunidades increases women’s time spent on domestic work; she shows this arises from the reallo- cation of time spent on unpaid housework between children and mothers as a result of the cash transfer (as women increase their participation in these activities so as to substi- tute for the work children were performing prior to the intervention). Ospina (2010) finds that for Colombia’s Familias en Acción an increase in hours spent on domestic labor by women as a result of the scheme was matched by a decrease in time spent on it by men, who increased hours spent on paid work. The role of conditionality in reinforcing a traditional patriarchal/maternal model of care, strengthening women’s maternal responsibilities, and displaying little or no recog- nition for women’s paid work has also been explored (Molyneux and Thomson 2011). Even after a comparatively brief period of PROGRESA implementation in Mexico, Parker and Skoufias (2000) find that mothers in PROGRESA increased their time allocation for child care as a result of the time demands on women associated with satisfying program obligations. Studies of the impact of conditional cash transfers on child labor find mixed results and examples of increases or low/no impact on girls’ involvement in unpaid housework. A study on Pakistan’s BISP (Cheema et al. 2014) finds a significant decrease in child labor participation (including housework) for boys, but not for girls. The authors argue that this is because girls are more engaged in household chores and because it is hard to shift cultural norms, which are unlikely to be affected, at least in the short term, by cash trans- fers. A 2015 study of Colombia’s Familias en Acción finds that the program increases the leisure time of boys while reducing their paid work, but reduces the leisure time of girls while increasing their domestic labor (Canavire-Bacarreza and Ospina 2015). Formal and Informal Work Theoretically, means-tested and conditional cash transfers could generate an incentive for people to remain in or move to informal work as a result of eligibility rules that imply formal work could disqualify them from receiving future transfers. It is also com- monly expected that such incentives could be affected by the value and type of transfers for which informal workers are eligible and how these compare with the conditions of formal work (Banerjee et al. 2015; Bosch and Manacorda 2012; Levy 2008). Chapter 3. Universal Basic Income and Work 113 In principle, as a universal and unconditional transfer, a UBI could weaken people’s concerns that formal work could lead to loss of benefit eligibility. At the same time, by potentially weakening links of social protection entitlements to formal employment, it could create incentives for informality. The available evidence for cash transfers in low- and middle-income countries indi- cates that the targeting and conditionality elements of conditional cash transfers can generate an informality incentive. For transfer schemes that explicitly target people in informal work and exclude formal sector workers, the evidence highlights such examples. Bosch and Manacorda (2012) show noncontributory schemes in Mexico and Colombia, accessible only to those who are not in formal employment, lead to a significant sub- stitution effect induced by the eligibility criteria. A study of Argentina’s universal child allowance, which explicitly targets informal workers with children, finds a large disin- centive to labor market formalization for program beneficiaries (Garganta and Gasparini 2015).9 In the case of noncontributory social assistance transfers that are not restricted to workers in the informal economy, studies point to limited or no impact of a cash transfer on informality. For Brazil’s Bolsa Família, paid to anyone with a declared income below the income eligibility threshold, Holanda Barbosa and Corseuil (2013) find it had no effect on the proportion of working hours households dedicated to informal activities. For Mexi- co’s Oportunidades, Azuara and Marinescu (2011) find no effect on informal employment. In practice, factors that influence a cash transfer’s impact on participation in formal or informal work include policy implementation details; the supply of formal jobs; and the conditions of formal work, including job security and social protection (e.g., see Angel-Urdinola, Haimovich, and Robayo 2009; Garganta and Gasparini 2015). Much like the case of income means testing, in the case of a scheme targeting informal workers (or the unemployed), weak policy monitoring in practice may mean participants do not fear losing the transfer by accepting a formal job. Moreover, the low supply of formal jobs for typical cash transfer beneficiaries may be so low as to lead to insignificant program effects. Finally, the advantages of formal over informal work—for instance, in terms of job security and wages—may lead workers to accept offers in formal work even if this implies discontinuation of benefit receipt (Garganta and Gasparini 2015). Conclusions Recent UBI debates and experiments have drawn attention to fundamental questions about the nature and value of work and to the role and objectives of social and labor market policy. The renewed interest in UBI has also (re)ignited important discussions on the specifics of whether and how basic income schemes could address the limitations and unintended effects of existing social protection policies and on the role of cash trans- fer policy design and implementation details in shaping work outcomes. This chapter contributes to the debate in two ways: first, by considering policy implications for the type and quality of work carried out, in addition to common priority 114 Francesca Bastagli concerns for participation in paid work and financial work incentives; and second, by paying close attention to the role of both cash transfer policy design details and the wider policy context. A summary of the chapter’s key findings, including implications for policy design and the wider UBI debate, follows. Work Incentives and Participation in Paid Work The available evidence indicates there is limited or no impact of UBI-type schemes on aggregate measures of participation in paid work. Breakdowns by population subgroup indicate that effects vary, with examples of schemes leading to reduced participation in paid work for some and increased participation for others. Reductions in “work effort” are observed in some schemes among women in a couple with children and among the elderly; and a reduction in participation in paid work among youth is associated with the Islamic Republic of Iran’s universal cash transfer. In the majority of cases reviewed, these results are discussed in terms of welfare-enhancing shifts as individuals take up valued activities (by the individual or society) such as care and further education. The specific case of care work and implications for gender inequality in the world of work is discussed further below. In terms of basic income/cash transfer policy design, the absence of a clear nega- tive effect on work participation is in some cases discussed as the result of the low level of the transfer. Another explanation is provided by the effect the cash transfer has on tackling barriers to participation in paid work, such as costs associated with travel and care of family members. The evaluations of Alaska’s UBI also point to the potential role of a “macro” effect countering schemes’ income effect through consumption increases, which stimulate labor demand and increase wages and employment. Returning to cash transfer policy design issues, the evidence from means-tested unconditional and conditional cash transfers in low- and middle-income countries— including ones generating a high marginal tax rate on paper (e.g., income-tested, with clear eligibility cutoff and no benefit taper)—provides some insights into the relative roles of the income and substitution effect of cash transfer components. The evidence highlights the absence of a clear negative effect on work participation associated with means tests in conditional cash transfers in low- and middle-income countries. In the case of income-tested transfers, this result is discussed as being linked to the weak imple- mentation of the means test in practice, leading to a weak or no association between changes in people’s incomes and transfer receipt in practice. This poses a challenge to drawing conclusions with regard to the impact of means testing, including income tar- geting, on work effort and to the potential advantages of a UBI over alternative targeted cash transfers. Attaching work requirements to transfer schemes is associated with higher partici- pation in paid work in some schemes, but can work against the objectives of promoting improved work conditions and quality. The absence of work conditionalities that require workers to accept job offers appears critical to enabling basic income–type schemes to contribute to improved wages and work conditions in the experience of some schemes. Chapter 3. Universal Basic Income and Work 115 The inclusion of work requirements also raises questions about their potential role in screening out vulnerable groups from participating in a transfer scheme. Conditionalities set in terms of human capital accumulation, on the other hand, such as regular school attendance and health care visits among children, have been linked to reduced partici- pation in paid work among women, and may reinforce gendered divisions of labor. This is discussed further below. Conditions of Work The available evidence reviewed in this chapter shows that additional cash through basic income–type and other cash transfers can lead to processes of emancipation by address- ing constraints to better work and offering workers an exit strategy by strengthening their bargaining positions. How does a UBI influence these potential processes compared with alternative cash transfer schemes? The theoretical debate is clear. Targeted and conditional trans- fers with work requirements risk acting as a subsidy to low-pay and “bad” work. A UBI would provide an exit strategy and options to workers if paid at an adequate level and accompanied by adequate regulation, such as minimum wage legislation. One basic income study indicates that indeed, continued support and regulation from wider social protection and labor policy are critical for the scheme to contribute to higher wages by increasing workers’ bargaining power. The experience of targeted unconditional and conditional cash transfers in low- and middle-income countries provides examples of how regular additional income can lead to shifts in the type of work carried out that are welfare enhancing. Examples include clear and significant reductions in casual labor and decreases in temporary wage work as a result of cash transfer receipt. In all cases reviewed, the schemes do not include work-related conditionalities. Valuation and Distribution of Unpaid Work The theoretical literature points to the risk for additional unearned unconditional cash to reinforce gendered divisions of labor, especially for some groups, such as among women in a couple with weaker labor market attachment as secondary earners. The evidence indicates that this can indeed be the case, with a number of basic income–type schemes leading to reductions in work participation among women with children in a couple. At the same time, the evidence for basic income–type schemes also highlights examples of related increases in women’s participation in paid work when they enable women to tackle barriers to employment participation, for instance, by affording child care. In this regard, the potential advantage of a UBI over alternative schemes that are paid to households or to specific individuals is that, by being paid to everyone individ- ually, it avoids targeting the main breadwinner or the female “head of household.” This feature, together with the unconditional nature of a UBI, promises to avoid the risks of reinforcing the woman’s care-provider role brought on by cash transfers with human capital accumulation conditionalities that target women, as highlighted by some studies on conditional cash transfers. 116 Francesca Bastagli Both the theoretical and empirical literature appear to converge on the point that for a basic income scheme to at a minimum not aggravate the unequal distribution of unpaid work across the sexes, and to possibly address this inequality, transfers need to be accompanied by wider efforts to tackle existing gendered norms and inequalities. Formal and Informal Work Studies of the experience of cash transfers targeted to informal workers, and exclud- ing formal workers, highlight how these can lead to disincentives to formalization. In principle, the universal and unconditional nature of a UBI would reduce this effect. How- ever, precisely these same features risk creating a disincentive to formalize by delinking social protection eligibility from labor market status. The expansion of social assistance schemes, aiming to reach both formal and informal workers, can weaken the incentive to formalization. Once again, a basic income’s positioning within the wider social and labor market policy landscape seems to matter. If a UBI replaces formal worker contributory schemes or is accompanied by a reduction in job security in formal employment, then there may be a risk of weakening incentives to formalize. Its situation within a wider social protection system that includes contributory and broader job security provisions holds potential for reducing this risk. Notes 1. The universal nature of such schemes poses a challenge to identifying individual-level effects via counterfactual analysis. 2. Negative income tax schemes were tested in the United States in four separate experiments: New Jersey and Pennsylvania (1968–72); a rural experiment in Iowa and North Carolina (1969–73); Gary, Indiana (1971–74); and Seattle and Denver (1971–82). The Canadian gov- ernment initiated the Manitoba Basic Annual Income Experiment/Mincome in 1975 (Burtless 1986; Widerquist 2005). 3. Several commentators have voiced methodological concerns about the design of the U.S. neg- ative income tax studies. Only low-income families were included in the experiments. Many of the results are not attributable to the negative income tax per se but to the fact that the plans tested were more generous than the existing welfare programs for which the control group was eligible. This is likely to have overstated the work-effort response (Munnell 1986; Widerquist 2005). 4. Among the two studies on elderly adults, one finds a significant effect from a social pension in Mexico of reducing pensioners working for pay. The second study, Dabalen, Kilic, and Wane (2008) on Albania’s Ndihma Ekonomike social assistance scheme targeted to households with an unemployed head of household, finds a negative labor supply response among female workers; the coefficient for males is not statistically significant. 5. Kassouf and de Oliveira (2012) find that Brazil’s Benefício de Prestação Continuada (BPC) social pension led to a reduction in hours worked, but by elders (over 65 years old), suggesting that the pension enabled elderly householders to retire and reduce the time spent in active work. Chapter 3. Universal Basic Income and Work 117 6. Bazzi, Sumarto, and Suryahadi (2012) only find a statistically significant reduction in hours worked for those who received their first disbursement and were awaiting their (delayed) second transfer. 7. Daidone, Davis, Dewbre, and Covarrubias (2014) find Lesotho’s Child Grant Programme to reduce hours worked in any labor by 2.8 hours in the previous week, with much of this due to a reduction in casual wage labor. 8. According to a 2016 review by Bastagli et al., three of six studies reporting on cash transfer impact on type of work (number of hours) by women find at least one statistically significant result. Two of these find an associated increase in domestic work. A third study finds a small reduction in the proportion of women engaged in unpaid family labor. Cheema et al. (2014), with regard to Pakistan’s BISP, observe that it has resulted in a (small) reduction in the propor- tion of women engaged in unpaid family labor. 9. Argentina’s universal child allowance provides a monthly benefit to households whose mem- bers are not registered in the national social security system, meaning that they are either unemployed or working in the informal sector. References Angel-Urdinola, Diego F., Francisco Haimovich, and Monica Robayo. 2009. “Is Social Assistance Contributing to Higher Informality in Turkey?” Background paper prepared for Country Economic Memorandum—Informality: Causes, Consequences, Policies. https://mpra.ub.uni- muenchen.de/27675/1/MPRA_paper_27675.pdf Ardington, Cally, Anne Case, and Victoria Hosegood. 2009. “Labor Supply Responses to Large Social Transfers: Longitudinal Evidence from South Africa.” NBER Working Paper 13442. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/papers/w13442.pdf Atkinson, Anthony. 1995. Incomes and the Welfare State. Cambridge, UK: Cambridge University Press. Azuara, Oliver, and Ioana Marinescu. 2013. “Informality and the Expansion of Social Protection Programs: Evidence from Mexico.” Journal of Health Economics 32 (5): 938–50. Baird, Sarah, David McKenzie, and Berk Özler. 2018. “The Effects of Cash Transfers on Adult Labor Market Outcomes.” Policy Research Working Paper 8404. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/334251523556191237/pdf/WPS8404.pdf Banerjee, Abhijit, Rema Hanna, Gabriel Kreindler, and Benjamin A. Olken. 2015. “Debunking the Stereotype of the Lazy Welfare Recipient: Evidence from Cash Transfer Programs Worldwide.” Working Paper RWP15-076. John F. Kennedy School of Government, Harvard University, Cam- bridge, MA. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2703447 Barrientos, Armando, and Juan Miguel Villa. 2015. “Antipoverty Transfers and Labour Market Out- comes: Regression Discontinuity Design Findings.” Journal of Development Studies 51 (9). https://www.tandfonline.com/doi/full/10.1080/00220388.2015.1010157?scroll=top&needA ccess=true Bastagli, Francesca, Jessica Hagen-Zanker, Luke Harman, Valentina Barca, Georgina Sturge, Tanja Schmidt, and Luca Pellerano. 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. https://www.odi.org/publications/10505-cash-transfers-what- does-evidence-say-rigorous-review-impacts-and-role-design-and-implementation 118 Francesca Bastagli Bazzi, Samuel, Sudarno Sumarto, and Asep Suryahadi. 2012. “Evaluating Indonesia’s Uncondi- tional Cash Transfer Programme, 2005–6.” 3ie Evaluation Report. International Initiative for Impact Evaluation (3ie), New Delhi. https://pdfs.semanticscholar.org/0216/7e1d3522d86bc7 d4c7b11a59cdc9fdcb6543.pdf Birnbaum, Simon, and Jurgen De Wispelaere. 2016. “Basic Income in the Capitalist Economy: The Mirage of ‘Exit’ from Employment.” Basic Income Studies 11(1): 61–74. Blundell, Richard, and Thomas Macurdy. 1999. “Labor Supply: A Review of Alternative Approaches.” In Handbook of Labor Economics, vol. 3A, edited by Orley Ashenfelter and David Card, chapter 27. Amsterdam: North Holland. Bosch, Mariano, and Marco Manacorda. 2012. “Social Policies and Labor Market Outcomes in Latin America and the Caribbean: A Review of the Existing Evidence.” CEP Occasional Paper No. 32. Centre for Economic Performance, London. http://cep.lse.ac.uk/pubs/download/ occasional/op032.pdf Brewer, Mike. 2003. “The New Tax Credits.” Briefing Note. Institute for Fiscal Studies, London. https://www.ifs.org.uk/bns/bn35.pdf Burtless, Gary. 1986. “The Work Response to a Guaranteed Income: A Survey of Experimental Evi- dence.” In Lessons from the Income Maintenance Experiments: Proceedings of a Conference Held in September 1986. https://www.bostonfed.org/news-and-events/events/economic-research- conference-series/lessons-from-the-income-maintenance-experiments.aspx Calnitsky, David. 2018. “The Employer Response to the Guaranteed Annual Income.” Socio-Eco- nomic Review 25. Canavire-Bacarreza, Gustavo J., and Monica Ospina. 2015. “Intrahousehold Time Allocation: An Impact Evaluation of Conditional Cash Transfer Programs.” CIEF Working Paper No. 15–17. Center for Research in Economics and Finance. https://papers.ssrn.com/sol3/papers. cfm?abstract_id=2645096 Cheema, Iftikhar, Maham Farhat, Simon Hunt, Sarah Javeed, Luca Pellerano, and Sean O’Leary. 2014. “Benazir Income Support Programme: First Follow-up Impact Evaluation Report.” Oxford Policy Management, Oxford, UK. http://bisp.gov.pk/wp-content/uploads/2016/10/ BISPFirstImpactEvaluationReport.pdf Cookson, Tara Patricia. 2018. Unjust Conditions: Women’s Work and the Hidden Cost of Cash Transfer Programmes. Berkeley: University of California Press. Dabalen, Andrew, Talip Kilic, and Waly Wane. 2008. “Social Transfers, Labor Supply and Poverty Reduction: The Case of Albania.” Policy Research Working Paper 4783. World Bank, Washing- ton, DC. http://documents.worldbank.org/curated/en/246771468191366177/pdf/WPS4783. pdf Daidone, Silvio, Benjamin Davis, Joshua Dewbre, and Katia Covarrubias. 2014. “Lesotho’s Child Grant Programme: 24-Month Impact Report on Productive Activities and Labour Alloca- tion. Lesotho Country Case Study Report.” Food and Agriculture Organization of the United Nations, Rome. http://www.fao.org/3/a-i4186e.pdf Daidone, Silvio, Benjamin Davis, Joshua Dewbre, Mario González-Flores, Sudhanshu Handa, David Seidenfeld, and Gelson Tembo. 2014. “Zambia’s Child Grant Programme: 24-Month Impact Report on Productive Activities and Labour Allocation.” Food and Agriculture Organization of the United Nations, Rome. http://www.fao.org/3/a-i3692e.pdf Dewbre, J., Silvio Daidone, Benjamin Davis, Borja Miguelez, Ousmane Niang, and Luca Pellerano. 2015. “Lesotho Child Grant Programme and Linking Food Security to Social Protection Pro- gramme.” Food and Agriculture Organization of the United Nations, Rome. http://www.fao. org/3/a-i4694e.pdf Chapter 3. Universal Basic Income and Work 119 ECLAC (Economic Commission for Latin America and the Caribbean). 2013. Gender Equality Observatory of Latin America and the Caribbean. Annual Report 2012: A Look at Grants, Sup- port and Burden for Women. Santiago: United Nations. https://repositorio.cepal.org/bitstream/ handle/11362/35445/S2013107_en.pdf?sequence=1&isAllowed=y Eissa, Nada, and Hilary Williamson Hoynes. 2004. “Taxes and the Labor Market Participation of Married Couples: The Earned Income Tax Credit.” Journal of Public Economics 88 (2004): 1931– 58. https://gspp.berkeley.edu/assets/uploads/research/pdf/Eissa-Hoynes-JPUBE-2004.pdf Eissa, Nada, and Jeffrey B. Liebman. 1996. “Labor Supply Response to the Earned Income Tax Credit.” Quarterly Journal of Economics 111 (2): 605–37. Eyal, Katherine, and Ingrid Woolard. 2011. “Throwing the Book at the CSG.” Working Paper No. 53. Southern Africa Labour and Development Research Unit, University of Cape Town, Cape Town. http://opensaldru.uct.ac.za/bitstream/handle/11090/64/2011_53.pdf?sequence=1 Francesconi, Marco, Helmut Rainer, and Wilbert van der Klaauw. 2007. “The Effects of In-Work Benefit Reform in Britain on Couples: Theory and Evidence.” IZA Discussion Paper No. 2980. Institute for the Study of Labor, Bonn. http://hdl.handle.net/10419/34490 Garganta, Santiago, and Leonardo Gasparini. 2015. “The Impact of a Social Program on Labor Informality: The Case of AUH in Argentina.” Journal of Development Economics 115 (C): 99–110. Goldsmith, Oliver Scott. 2010. “The Alaska Permanent Fund Dividend: A Case Study in Imple- mentation of a Basic Income Guarantee.” Working paper. Institute of Social and Economic Research, University of Alaska, Anchorage. —. 2012. “The Economic and Social Impacts of the Permanent Fund Dividend on Alaska.” In Alaska’s Permanent Fund Dividend: Examining Its Suitability as a Model, edited by Karl Wider- quist and Michael Howard, chapter 4. New York: Palgrave Macmillan. Gray, Anne. 2017. “Behavioural Effects of a Citizen’s Income on Wages, Job Security and Labour Supply.” Citizen’s Basic Income Trust. http://citizensincome.org/research-analysis/behavioural- effects-of-a-citizens-income-on-wages-job-security-and-labour-supply/ Harrop, Andrew, and Cameron Tait. 2017. “Universal Basic Income and the Future of Work.” Fabian Society. https://www.tuc.org.uk/sites/default/files/UBI.pdf Hener, Timo. 2016. “Unconditional Child Benefits, Mothers’ Labour Supply and Family Well-Being: Evidence from a Policy Reform.” CESifo Economic Studies 62 (4): 624–49. Holanda Barbosa, Ana Luiza Neves de, and Carlos Henrique L. Corseuil. 2013. “Bolsa Família, Occupational Choice and Informality in Brazil.” Working Paper 118, International Policy Centre for Inclusive Growth. Hum, Derek, and Wayne Simpson. 1993. “Economic Response to a Guaranteed Annual Income: Experience from Canada and the United States.” Journal of Labor Economics 11 (1). https:// home.cc.umanitoba.ca/~simpson/JOLE1993.pdf ILO (International Labour Office). 2018. World Employment Social Outlook, Trends for Women 2018. Geneva: ILO. Immervoll, Herwig, and Cathal O’Donoghue. 2002. “Welfare Benefits and Work Incentives: An Analysis of the Distribution of Net Replacement Rates in Europe Using EUROMOD, a Multi-Country Microsimulation Model.” EUROMOD Working Paper No. EM4/01. University of Essex, Institute for Social and Economic Research, Colchester, UK. https://www.iser.essex. ac.uk/research/publications/working-papers/euromod/em4-01.pdf Jones, Damon, and Ioana Marinescu. 2018. “The Labor Market Impacts of Universal and Per- manent Cash Transfers: Evidence from the Alaska Permanent Fund.” NBER Working Paper 24312. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/ papers/w24312.pdf 120 Francesca Bastagli Kabeer, Naila, Caio Piza, and Linnet Taylor. 2012. What Are the Economic Impacts of Conditional Cash Transfer Programmes? A Systematic Review of the Evidence. London: EPPI-Centre, Social Science Research Unit, Institute of Education, University of London. https://assets.publishing. service.gov.uk/media/57a08a6840f0b649740005a4/CCTprogrammes2012Kabeer.pdf Kangas, Olli, Signe Jauhiainen, Miska Simanainen, and Minna Ylikännö, eds. 2019. The Basic Income Experiment 2017–2018 in Finland. Preliminary Results. Helsinki: Ministry of Social Affairs and Health. http://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/161361/Report_ The%20Basic%20Income%20Experiment%2020172018%20in%20Finland.pdf Kassouf, Ana Lucia, and Pedro de Oliveira. 2012. “Impact Evaluation of the Brazilian Non-Contrib- utory Pension Program Benefício De Prestação Continuada (BPC) on Family Welfare.” Working Paper 2012-12. Partnership for Economic Policy. Leibbrandt, Murray, Kezia Lilenstein, Callie Shenker, and Ingrid Woolard. 2013. “The Influence of Social Transfers on Labour Supply: A South African and International Review.” Working Paper No. 112. Southern Africa Labour and Development Research Unit, University of Cape Town, Cape Town. http://www.opensaldru.uct.ac.za/bitstream/handle/11090/670/2013_112. pdf?sequence=1 Levy, Santiago. 2008. Good Intentions, Bad Outcomes: Social Policy, Informality and Economic Growth in Mexico. Washington, DC: Brookings Institution Press. Martinelli, Luke. 2017. “Assessing the Case for a Universal Basic Income in the UK.” IPR Policy Brief. Institute for Policy Research, University of Bath, Bath. https://www.bath.ac.uk/publications/ assessing-the-case-for-a-universal-basic-income-in-the-uk/attachments/basic_income_policy_ brief.pdf McKay, Ailsa. 2001. “Rethinking Work and Income Maintenance Policy: Promoting Gender Equal- ity through a Citizens’ Basic Income.” Feminist Economics 7 (1): 97–118. McLean, Caitlin. 2016. “…and Justice for All? Basic Income and the Principles of Gender Equity.” Institute for Public Policy Research, London. https://www.ippr.org/juncture/and-justice-for-all- basic-income-and-the-principles-of-gender-equity Moffitt, Robert. 2002. “Welfare Programs and Labor Supply.” NBER Working Paper 9168. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/papers/w9168.pdf Molyneux, Maxine. 2009. “Conditional Cash Transfers: A ‘Pathway to Women’s Empowerment’?” Pathways Working Paper 5. Institute of Development Studies, Brighton. https://assets. publishing.service.gov.uk/media/57a08b4eed915d3cfd000c52/PathwaysWP5-website.pdf Molyneux, Maxine, and Marilyn Thomson. 2011. “CCT Programmes and Women’s Empowerment in Peru, Bolivia and Ecuador.” CARE Policy Paper. https://insights.careinternational.org.uk/ media/k2/attachments/cct-progs-and-womens-emp-peru-bolivia-ecuador.pdf Munnell, Alicia H. 1986. “Lessons from the Income Maintenance Experiments: An Overview.” In Lessons from the Income Maintenance Experiments: Proceedings of a Conference Held in September 1986. https://www.bostonfed.org/news-and-events/events/economic-research- conference-series/lessons-from-the-income-maintenance-experiments.aspx Novella, Rafael, Laura Ripani, Guillermo Cruces, and María Laura Alzúa. 2012. “Conditional Cash Transfers, Female Bargaining Power and Parental Labour Supply.” IDB Working Paper No. 368. Inter-American Development Bank, Washington, DC. https://pdfs.semanticscholar.org/af7d/ e490c3d6243ea65a2335909d0e42ce8dfc7a.pdf Ospina, Monica P. 2010. “The Indirect Effects of Conditional Cash Transfer Programs: An Empirical Analysis of Familias en Acción.” Dissertation, Georgia State University, Atlanta. Owusu-Addo, Ebenezer, Andre M.N. Renzaho, and Ben J. Smith. 2018. “The Impact of Cash Transfers on Social Determinants of Health and Health Inequalities in Sub-Saharan Africa: A Systematic Review.” Health Policy and Planning 33 (5): 675–96. Chapter 3. Universal Basic Income and Work 121 Parker, Susan W., and Emmanuel Skoufias. 2000. “The Impact of PROGRESA on Work, Leisure and Time Allocation.” Final report. International Food Policy Research Institute, Washington, DC. http://www.ifpri.org/publication/impact-progresa-work-leisure-and-time-allocation Rubio-Codina, Marta. 2009. “Intrahousehold Time Allocation in Rural Mexico: Evidence from a Randomized Experiment.” Institute for Fiscal Studies, London. Salehi-Isfahani, Djavad, and Mohammad H. Mostafavi-Dehzooei. 2017. “Cash Transfers and Labor Supply: Evidence from a Large-Scale Program in Iran.” Working Paper No. 1090. Economic Research Forum, Giza, Egypt. https://erf.org.eg/wp-content/uploads/2017/05/1090.pdf Skoufias, Emmanuel, and Vicenzo Di Maro. 2008. “Conditional Cash Transfers, Adult Work Incen- tives and Poverty.” Journal of Development Studies 44: 935–60. Standing, Guy. 2013. “India’s Experiment in Basic Income Grants.” Global Dialogue: Newsletter for the International Sociological Association 3 (5). Tamm, Marcus. 2009. “Child Benefit Reform and Labour Market Participation.” Ruhr Economic Papers No. 97. Rheinisch-Westfälisches Institut für Wirtschaftsforschung, Essen. http://hdl. handle.net/10419/29906 Van Parijs, Philippe. 2004. “Basic Income: A Simple and Powerful Idea for the Twenty-First Cen- tury.” Politics & Society 32 (1): 7–39. Widerquist, Karl. 2005. “A Failure to Communicate: What (If Anything) Can We Learn from the Negative Income Tax Experiments?” Journal of Socio-Economics 34 (1): 49–81. Widerquist, Karl, José A. Noguera, Yannick Vanderborght, and Jurgen De Wispelaere, eds. 2013. Basic Income: An Anthology of Contemporary Research. Oxford, UK: Wiley-Blackwell. Young, Charlie. 2018. Realising Basic Income Experiments in the UK—A Typology and Toolkit of Basic Income Design and Delivery. London: RSA Action and Research Centre. https://www.thersa. org/globalassets/pdfs/reports/realising-basic-income.pdf Comparative Effects of Universal Basic CHAPTER 4 Income: Emerging Issues and Estimates Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan D ebates on universal basic income (UBI) are often polarizing. As other chapters examine in more detail, diversity in objectives, perspectives, and expectations can shape discussions in ways that make UBI somewhat elusive. Injecting more rigor and clarity in UBI debates is a priority. This chapter pursues two core objectives to this end. First, we provide a practical application of the decision-mak- ing parameters laid out in the overview of this volume. These parameters—coverage, financing, delivery, etc.—could underpin virtually any social assistance program, but are particularly compelling for a UBI given its large scale and deep implications. This chap- ter provides more concrete examples and granularity around the trade-offs between, for example, coverage, adequacy, costs, and financing options. Second, drawing on existing data on social protection performance, we set out an illustrative typology of country contexts. Microsimulations are run to provide further insights into the com- parative impacts and distributional effects of a UBI replacing part of a country’s social assistance portfolio. By laying out and quantifying those trade-offs, the analysis aims to move the UBI debate from one fueled by principles to a more evidence-based and contextual discus- sion grounded in analytics. The chapter is not intended to provide ultimate evidence on the effects of a UBI, as general equilibrium and dynamic simulations may better fulfill that function. Instead, we use the simulations as a logical way to think about the issues— as an analytical framework for approaching, unbundling, and understanding UBI within a wider system of social protection. The analysis complements an emerging literature that evaluates the cost and impli- cations of UBI. Among others, Browne and Immervoll (2017) use EUROMOD data to 123 124 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan simulate the effects of existing means-tested cash transfers versus a UBI. Hoynes and Rothstein (2019) review the distributional and behavioral effects of a UBI in the United States. Based on household survey data from Indonesia and Peru, Hanna and Olken (2018) use receiver operating characteristic curves and other methods to estimate trade- offs between targeting errors (exclusion and inclusion) and transfer adequacy among a UBI and flagship targeted programs. Based on Living Standards Measurement Study data, Brown, Ravallion, and van de Walle (2016) predict the performance of various proxy means–testing methods against a UBI in nine African countries. The International Monetary Fund examines the distributional effects of a UBI in a handful of high- and middle-income countries (IMF 2017), while Coady and Prady (2018) do so for India. Ortiz et al. (2018) and World Bank (2018) set out estimates for the cost of a UBI for a large number of low- and middle-income countries. We conduct a harmonized, comparable analysis for 10 countries ranging from low-income Haiti, Mozambique, and Nepal; lower-middle-income India and Indonesia; upper-middle-income Brazil, Kazakhstan, and South Africa; to high-income Chile and the Russian Federation.1 The objective is not to analyze the country-specific details of a UBI reform, but rather to gain a better understanding of how the poverty and distri- butional implications of UBI reforms may vary depending on country level of income, the existing structure of the welfare state, and the taxation structure. While not assess- ing them in detail, the chapter also points to some political economy challenges such reforms may entail, including the identification of winners and losers who may oppose the reform. The simulations estimate first-round effects and do not contemplate behav- ioral responses (e.g., labor market responses that could stem from the UBI transfer or changes in marginal taxation). More discussion of financing and political economy mat- ters is presented in chapters 5 and 6, respectively. The selected countries present substantially different outlooks and performance in social protection systems. We group them in four broad clusters: 1. Low coverage of the poorest deciles with progressive spending: Haiti, Kazakhstan 2. Low coverage with flat or regressive spending: Mozambique, Nepal 3. High coverage with strongly progressive spending: Brazil, Indonesia 4. High coverage with slightly progressive, flat, or regressive spending: Chile, India, Russia, South Africa To be sure, within each category there remain substantial differences; and for some countries, the nature of coverage and spending may not allow for clear-cut categoriza- tion. Overall, however, the classification will help to better identify challenges that relate to specific features of the welfare state, or of the programs a UBI would replace. The chapter unfolds as follows. The next two sections discuss the data and meth- odology. The subsequent section presents the results, including the impacts of a UBI of various levels of generosity on poverty, inequality, and distributional effects by age and Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 125 income deciles. For a subset of six countries for which there is information on the inci- dence of taxes (Brazil, Chile, India, Indonesia, Russia, and South Africa), we examine the distributional implications of more generous UBI reforms that involve more spending, which requires taking into account the higher burden of taxation.2 Conclusions are pre- sented in the chapter’s final section. Data The main sources of information for this chapter are the household surveys and admin- istrative data housed in the Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE); and, for the subset of countries for which such information is available, the tax incidence results in the Commitment to Equity Data Center on Fiscal Redistribution, which includes information on the incidence of taxes and transfers by decile from fiscal incidence analyses that used a common methodological framework (Lustig 2018). The household surveys include information on social assistance programs disaggregated by program, welfare indicator (disposable income, expenditure, or consumption), and household demographic characteristics. The ASPIRE database collects program-level information on social protection programs including spending, number of beneficia- ries, and program design features. Administrative data were available for 9 out of 10 countries (Haiti does not have administrative data in ASPIRE, and for Russia, we used World Bank staff estimates; see table D.1 in appendix D). This database is used to val- idate/compare amounts spent on social assistance programs from administrative data by country, with the total benefit amounts retrieved from household surveys. The World Development Indicators database is used for some of the indicators reported in this study, such as the country income groups of countries, gross domestic product (GDP), and purchasing power parity (PPP). Data on energy subsidies are from the International Monetary Fund. The choice of countries and years was driven by various criteria, the first of which was quality of the data. Our preference was to include countries in which the household survey captured the largest cash transfer programs. To avoid excessive discrepancies between survey and administrative data, we selected countries and years for which the administrative and household survey totals (both in terms of beneficiaries and amounts spent) were similar.3 In the cases of India, Indonesia, and Mozambique, cash transfers had to be imputed (see appendix D for details). To capture a large enough variance in social protection approaches, we also aimed at achieving a balance between low-, middle-, and high-income countries, as well as a regional balance. Our simulations look at the poverty and distributional impacts of replacing selected cash-based social assistance programs with a UBI. We looked to include countries with distinct welfare systems regarding their cash transfer programs that would be replaced, countries with poverty-targeted cash transfer programs with limited coverage and small transfers, countries with relatively generous poverty-targeted cash transfer programs, and countries with categorical coverage (box 4.1). 126 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan BOX 4.1  Baseline Welfare State Typologies T his chapter relates some of its findings to features of country welfare systems. In each country, the set of programs we consider is only a subset of all its social assistance/cash transfer programs (many of which are not well captured in household surveys); hence our description of the welfare systems may differ from a broader characterization that includes all programs. We here review countries’ typologies as characterized by the years and programs we consider, and how they may differ from typologies that would consider larger sets of programs. A detailed list of programs considered in the analysis is provided in appendix D. Haiti (2012), Mozambique (2014), and Nepal (2010). Haiti, Mozambique, and Nepal have overall limited social assistance coverage; and the household surveys only capture a subset of this. In Haiti, the survey captures the scholarships program (less than 0.1 percent of GDP), and in Mozambique the Basic Social Subsidy (PSSB) program (of which around 0.1 percent of GDP is captured in the survey, against current spending of 0.6 percent of GDP in the social protection sector). Nepal has a large portfolio of social protection programs, which include public sector pen- sions, social security allowances, scholarships, health subsidies, and public works. Most programs are categorical, and only around 0.9 percent of GDP is spent on social assistance programs specifically intended to assist the poor and vulnerable. The survey captures old-age, single women, disability, and endangered ethnicities pension and child grants—around 0.4 percent of GDP. India (2012). Over the past 15 years, there has been a major expansion and diver- sification of social protection programs in India. While the bedrock of social assistance, the Public Distribution System (PDS), has been in place since 1941, the Right to Food legislation of 2001 provided new impetus to social assistance provi- sions. For instance, half of the country’s children age 6–14 benefit from the national school meals program; 29 percent of rural households participate in the National Rural Employment Guarantee Scheme (NREGS), which provides 100 days of work at the minimum wage to anyone interested in applying; 52 percent of the population access PDS food subsidies; and between 19 and 22 percent of senior citizens above the age of 60 receive a form of social pension. The analysis is based on the PDS, since we do not consider public works in any country. Indonesia (2014). Indonesia has made significant headway in introducing and enhancing social assistance. In 2012, for instance, general subsidies represented 20 percent of the national budget. Since then, several rounds of reduction in energy subsidies have been achieved, with their share of the budget cut in half. Part of the savings has been reallocated to targeted social assistance programs, which reached 0.7 percent of GDP in 2018. A national registry of poor and vulnerable (continued) Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 127 BOX 4.1  Baseline Welfare State Typologies (continued) households, the Unified Database, was put in place for all implementing agencies to adopt. The cash transfer program for poor and at-risk students has expanded to around 18 million students since 2012. Similarly, the Program Keluarga Harapan (PKH, a conditional cash transfer program) has expanded from just under 2 million house- holds in 2012 to approximately 10 million in 2018. Overall, between 2014 (the year of our analysis) and 2018 the social assistance budget has increased by 50 percent in real terms (see appendix D). Targeted programs particularly benefited from the expansion—for instance, the PKH budget tripled in real terms between 2014 and 2018. These developments will only have strengthened the findings from our analy- sis, which for 2014 classifies Indonesia as a high-coverage/high-incidence country. Brazil (2015). Spending on social assistance accounted for 1.5 percent of GDP in 2015. This is only a small share of the country’s total social protection spending, which accounted for 13.8 percent of GDP that same year, with social insurance pensions absorbing the bulk of spending (11.1 percent of GDP). Brazil’s overall alloca- tion to social assistance is in line with the global average for developing countries, but lower than regional (1.6 percent) and BRICS (Brazil, Russia, India, China, and South Africa) (1.9 percent) averages. Two programs account for 75 percent of fed- eral spending on social assistance: the Benefício de Prestação Continuada (BPC) social pension and the Bolsa Família conditional cash transfer. Although a wide array of social programs is financed from the federal budget, the BPC, which is targeted to poor elderly and disabled persons, is the largest program, absorbing 0.69 percent of GDP (nearly half of all social assistance spending) in 2015. Brazil’s social safety net also includes numerous smaller programs that have become more diverse over time. Kazakhstan (2015). The country has inherited a Soviet-style social assistance system and undertook several reforms over the transition period, gradually reducing in-kind untargeted subsidies and privileges and introducing means-tested trans- fers (including conditional cash transfers). The overall level of social assistance spending hovered around 1 percent of GDP, but means-tested programs remain small compared to disability and survivor allowances and compensation for the removed subsidies and in-kind categorical programs. The overall incidence of social assistance is progressive, since most of the categories receiving assistance tend to be in the bottom quintiles. About one-third of the population is covered, and coverage in the bottom quintile is over 50 percent. The survey captures only about half of the spending on social assistance due to underreporting of amounts received in categorical benefits, but captures the number of beneficiaries rela- tively well, as recorded in administrative data. (continued) 128 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan BOX 4.1  Baseline Welfare State Typologies (continued) South Africa (2014). South Africa has a very generous social assistance system. According to ASPIRE, in 2015 the country spent around 3.3 percent of GDP on cash transfers alone. The poverty rate, according to the national poverty line, is around 55 percent of the population, and cash transfer programs cover a large share of the poor population—91, 85, and 74 percent of the population in the three poorest quintiles live in households that receive benefits. Some benefits spread to the rich- est quintile, but coverage decreases significantly (22 percent of the population in the richest quintile live in households that receive benefits). The survey captures most beneficiaries (in fact, when weights are used, the survey registers slightly more beneficiaries than administrative data—see appendix D). Chile (2013). According to the ASPIRE database, in 2015 Chile spent around 3.5 per- cent of GDP on more than 150 social programs, ranging from social pensions to other cash transfers, housing, scholarships, school feeding, other in-kind transfers, social care, and employment programs, among others. The programs we consider are a subset of the cash transfer programs and include selected social pension, dis- ability, and family support programs. Overall, they represent 1.1 percent of GDP, of which 0.8 percent is captured in the survey (see appendix D). Because not all the programs we consider have a clear poverty focus, coverage is not highly progres- sive—45 percent of individuals in the seventh income decile still live in households that receive some of the benefits. Note that the largest poverty-targeted cash transfer program, Aporte Familiar Permanente (0.1 percent of GDP), started in 2014 and is thus not included in the simulations, which are based on 2013 data. Russia (2016). According to official statistics, Russia currently spends 3.2 percent of GDP on social assistance. However, some of the spending is statistical misclassifi- cation—for example, wage subsidies are counted as social assistance. Moving to internationally harmonized data, the overall level of spending is 2.8 percent of GDP, which comes close to the survey estimate of 2.4 percent of GDP in social assistance received. The amount budgeted for social assistance programs has increased in real terms over the last 10 years, but its impact on poverty remains weak. This is largely due to high fragmentation. A stocktaking exercise found 800 programs financed and legislated at the federal level, plus over 10,000 regional programs (there are 85 regions in Russia, so each has on average of over 120 different social assistance benefits). Only a handful of programs are poverty targeted, while most are targeted to specific groups or categories (e.g., veterans, artists, civil servants, sportsmen); and everyone within the same group is entitled to the same benefit, regardless of actual need. Hence, while 85 percent of the poor are covered by social assistance, they receive a disproportionately small share of it, with 80 per- cent of the budget going to the nonpoor. (continued) Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 129 BOX 4.1  Baseline Welfare State Typologies (continued) Based on the social programs we have considered in each country, we can group countries into four broad welfare state modalities (table B4.1.1): (1) low coverage with progressive spending, (2) low coverage with flat or regressive spending, (3) high coverage with progressive spending, and (4) high coverage with slightly pro- gressive, flat, or regressive spending. TABLE B4.1.1  Illustrative Welfare State Typologies Progressive incidence Slightly progressive/flat/regressive incidence Low coverage Haiti, Kazakhstan Mozambique, Nepal High coverage Brazil, Indonesia India, South Africa, Russian Fed., Chile NOTE: The classification is based on the programs selected for the analysis and may change if other pro- grams are considered. Incidence refers to benefits. To be sure, within each category there remain substantial differences, and for some countries the nature of coverage and spending may not allow for such a clear-cut categorization, but overall the classification will help to better identify challenges that relate to specific features of the welfare state, or of the programs a UBI would replace. Methodology Our analysis is based on simulations of the first-round effects on poverty and inequality that result from changing the baseline social assistance system (baseline scenario) with alternative UBI scenarios, both without and with financing in the cases for which the UBI option implies higher levels of government spending. The cash transfer programs included in the baseline cover noncontributory programs only: that is, means-tested conditional and unconditional cash transfers, cash transfers based on categorical target- ing (e.g., people with disabilities), and noncontributory pensions. Contributory pensions and unemployment compensation programs are not included in the analysis because, by definition, these programs have an insurance component and reforming them involves complexities (e.g., financing for the transition period) that are beyond the scope of this study.4 We also do not consider public works. The UBI simulations proceed as follows. First, we select baseline cash transfer pro- grams to be replaced by a UBI. Second, using the ASPIRE household survey database, we subtract from each household’s income/expenditure the total amount for the selected cash benefits household members currently receive under the programs chosen to be replaced. Third, we simulate new values of disposable income (or consumption, depend- ing on the survey) by adding cash benefits under a UBI scheme, and divide the new disposable income totals for each household by the number of members to generate 130 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan values in per capita terms. The simulations only estimate first-round effects, and do not contemplate behavioral responses or general equilibrium effects. Among others, we assume away any behavioral responses concerning labor supply, an assumption that may be appropriate to assess impacts in the short run, but may not hold in the longer run. The typical characteristic of a UBI is that the size of the transfer is the same for all beneficiaries. We consider accordingly a cash transfer that is given to each member of the household irrespective of age. Our simulation decision may be in contrast with several UBI policy proposals that recommend distributing smaller amounts to children. While this remains a common policy option, we do not want our comparison to be influ- enced by demographics across countries; hence the choice of a flat, universal transfer. We have also generated results for a UBI that is given to adults only. We find that children and poorer households (who tend to have more children) are negatively affected by such a choice, but overall, the main policy findings would not be affected. We consider four alternative UBI scenarios defined by increasing levels of gener- osity. The main scenario is a budget-neutral UBI reform where the total cost of the UBI program is equivalent to the cost of the programs it replaces.5 In most countries, the size of the benefit going to the poor will be smaller than under the current programs. We then consider scenarios of increasing generosity (see below for details); to balance the budget, we need to either increase direct or indirect taxes, or reduce subsidies. Using the incidence results from the Commitment to Equity Data Center for a subset of six coun- tries for which such information is available, we simulate the net impact of combining UBI transfers with various financing mechanisms. We consider three financing schemes: (1) a proportional increase in direct taxes, (2) a proportional increase in indirect taxes, and (3) a lump-sum increase in direct taxes for the richest decile. See appendix D for a more detailed discussion of the methodology. In our simulations, we also consider two poverty measures: the poverty headcount and the squared poverty gap. There is a clear rationale in using two measures: the pov- erty headcount is widely used in policy circles but fails to capture the impacts of poverty reforms among the extremely poor. To give an example, assume that social assistance covers the extremely poor well, but coverage is not as good among households whose income or consumption lies close to the poverty line (this is the case, for instance, in South Africa). A budget-neutral UBI reform may show greater poverty reduction when measured with the poverty headcount index (because with the UBI all households close to the pov- erty line would now receive a transfer—and hence jump over the poverty line); but the reform would come at the expense of higher extreme poverty, because the budget would be taken away from the extremely poor to be redistributed among a greater number of people. The squared poverty gap measure, by giving a greater weight to the welfare of the extremely poor, is more likely to capture such an increase in extreme poverty. The choice of poverty lines also deserves some clarification. In our simulations, we consider the World Bank income class international poverty lines; these vary by country income levels, since in wealthier countries, higher international poverty lines are more appropriate. As described by Jolliffe and Prydz (2016), each income class–specific pov- erty line is chosen as the median of the national poverty lines of the countries in that Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 131 income class. Specifically, there are three income class–specific poverty lines: US$1.90 a day in 2011 PPP for low-income countries (Haiti, Mozambique, and Nepal), US$3.20 a day in PPP for lower-middle-income countries (India and Indonesia), and US$5.50 a day for upper-middle-income countries (Brazil, Kazakhstan, and South Africa).6 The World Bank has not generated poverty lines for high-income countries. Thus, for Chile and Russia, we computed a poverty line based on the methodology proposed by Rav- allion and Chen (2017). Using their formula, we compute the poverty line for Chile and Russia as a function of inequality-adjusted mean and intercept for the lower boundary of income (consumption per capita) in high-income countries. The result is a poverty line of US$11.66/day in 2011 PPP. We set the poverty line for Chile and Russia at US$11/day, which lies between our estimate and the US$10/day lower-bound national poverty line reported for Estonia and Poland, two high-income countries (Ravallion and Chen 2017). Note that these poverty lines are fairly recent, and may differ from poverty lines that may have been used for poverty monitoring in many reports; therefore, some of the pov- erty numbers may differ from existing reports. The choice of programs a UBI would replace, and the way we interpret the results, also deserves some explanation. First, the choice of programs is not dictated by the belief that a UBI should replace specific programs, but by our intention to explore the implications on poverty and inequality when certain programs are replaced. So, for example, the choice of including social pensions in the programs a UBI would replace is not dictated by a belief that a UBI should replace social pensions, but by our interest in exploring what happens if a UBI replaces social pensions, which often represent a sig- nificant proportion of the social assistance budget. Second, when we characterize social assistance systems and refer to them as “poverty-targeted” or “categorical,” we do not intend to make any claim about the social protection system of each country, but only about the programs our analysis is taking into consideration. The simulations that follow should not be viewed as actual estimates for the countries included in the chapter—any country-specific proposal would require analyses that are better tailored to the specific context. Rather, the objective is to reach a better understanding of how varying contexts affect the impacts on poverty and inequality of varying UBI schemes. Results The objectives of simulating the UBI scenarios (without and, whenever possible, with financing) are twofold: first, to look at the poverty and inequality impacts of various UBI scenarios according to their generosity; and second, to assess winners and losers from UBI reforms. Accordingly, we begin by comparing the poverty and inequality impacts of a budget-neutral UBI reform, where selected social assistance programs are replaced by a UBI. We then look at winners and losers from the reform along the income ladder and by demographic categories. Next, we look at the poverty and inequality impacts of more generous UBI schemes, keeping the sources of financing as exogenous. The way these more generous UBI schemes are financed, however, may significantly affect their 132 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan impacts on the poor. We conclude by looking at the poverty and inequality impacts of these more generous UBI schemes when the financing side is taken into consideration. Poverty and Inequality in the Baseline Scenario Figure 4.1 presents the baseline scenario poverty and inequality indicators, which include the incidence of the cash transfer programs that existed in the year of the analysis. Note that our estimated poverty impacts depend significantly on the choice of poverty line: in countries where the poverty line represents a lower proportion of average income (e.g., Kazakhstan, Russia), with all else being equal, the poverty impacts of a UBI reform will be lower. Accordingly, in interpreting the results, excessive attention should not be paid to the differences between the absolute impact on poverty of each scenario across countries (which depends significantly on the choice of poverty line), but rather on the differential impact of each scenario within each country (which depends much less on the poverty line). In other words, the analysis does not focus on comparing differences in impacts between, say, Russia and Brazil, but rather, on how a UBI reform would affect poverty and inequality within each of these countries. Replacing Social Assistance Programs with a Budget-Neutral UBI Figure 4.2 shows the poverty impacts of a budget-neutral reform that replaces the (selected) social assistance programs with a uniform income transfer given to all individuals, children FIGURE 4.1  Baseline Poverty and Inequality Indicators a. Poverty headcount b. Squared poverty gap c. Gini coe cient Low income Haiti 23.4 3.4 41.4 Mozambique 58.0 13.5 55.7 Nepal 13.5 0.9 32.8 Lower-middle income India 59.6 7.6 34.4 Indonesia 40.3 4.2 40.5 Upper-middle income Brazil 21.0 4.4 51.1 Kazakhstan 6.6 0.2 26.5 South Africa 56.3 17.5 63.0 High income Chile 35.3 5.6 47.1 Russian Fed. 7.3 0.8 35.6 0 10 20 30 40 50 60 0 5 10 15 20 0 10 20 30 40 50 60 70 Percent Percent Percent NOTE: Indicators are for the year of the analysis. Poverty and inequality indicators may differ from official sta- tistics due to differences in the deflator and welfare aggregate used to compute poverty. World Bank income class international poverty lines are used (2011 PPP): Haiti, Mozambique, and Nepal: US$1.90/day; India and Indo- nesia: US$3.20/day; Brazil, Kazakhstan, and South Africa: US$5.50/day; Chile and Russia: US$11.00/day. Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 133 FIGURE 4.2  Poverty Impacts of Budget-Neutral UBI Reform a. Poverty headcount 0.0 0.0 –0.05 –0.12 –0.52 –0.46 –2.29 –2.20 –1.84 –1.78 –2.41 –1.27 –2.87 –1.96 –4.58 –4.56 –3.25 –2.46 –2.93 –2.89 0 –1 Percentage points –2 –3 –4 –5 b. Squared poverty gap –0.03 –0.01 –0.08 –0.07 –0.07 –0.07 –1.67 –1.39 –1.33 –0.71 –2.12 –0.99 –0.43 –0.19 –15.55 –12.68 –1.78 –1.14 –0.63 –0.69 0 –5 Percentage points –10 –15 –20 Haiti Mozam- Nepal India Indonesia Brazil Kazakh- South Chile Russian bique stan Africa Fed. Low income Lower-middle income Upper-middle income High income Baseline Budget-neutral UBI reform NOTE: Baseline programs are the programs we considered in each survey to be replaced by a UBI. The impacts show the percentage point difference with respect to disposable income without transfers. included. With the exception of Mozambique, the baseline allocation of social assistance programs is more effective in reducing the poverty headcount ratio than a budget-neutral UBI reform. The reason is simple. A budget-neutral UBI reform yields higher coverage at the expense of lower per capita benefits (since the same resources are now spread across all individuals). Because in most countries the poor are more likely to receive a transfer, the lower benefits effect dominates: that is, under a budget-neutral UBI reform, there are more poor individuals who fail to cross the poverty line because they receive lower ben- efits than individuals who now manage to cross the poverty line because previously they did not receive a transfer. In Mozambique, on the other hand, the higher coverage effect (more households are able to cross the poverty line) dominates the lower benefits effect, so a UBI delivers higher poverty impacts. In quite a few countries, differences between the baseline and UBI scenarios remain relatively small; we discuss this issue below. 134 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan Note that the impacts of both baseline programs and a budget-neutral UBI reform are very low in low-income countries (sometimes close to imperceptible), as in Haiti and Mozambique, because cash transfer programs remain extremely limited. Because of the very limited presence of cash transfers, it will be difficult to assess the relative perfor- mance of baseline programs versus a UBI reform, since it hinges on a very narrow group of beneficiaries. The analysis in these countries will gain relevance when we explore the impacts and costs of more generous UBI schemes. While in some countries a budget-neutral UBI reform may deliver similar pov- erty reduction impacts as baseline programs, in all countries but Russia the extremely poor would be penalized. Figure 4.2b shows the same reform, now measured using the squared poverty gap as opposed to the poverty headcount. The squared poverty gap puts a much stronger emphasis on the welfare of the extremely poor, because—due to the squared distance from the poverty line—the poorer a person is, the more that contributes to the aggregate squared poverty gap. In all countries (apart from Russia), the baseline system delivers larger impacts on extreme poverty reduction than the budget-neutral UBI reform. The reason is simple: in all countries, extremely poor households are more likely to receive some transfers, and a budget-neutral UBI reform would take away some resources from the extremely poor to redistribute them to less-poor individuals. Russia seems to be going the opposite way: when measured using the squared poverty gap, the UBI scheme seems to deliver even better poverty impacts than when measured using the poverty headcount—signaling that a UBI reform would likely make many extremely poor better off. The reason, again, is quite straightforward. Russia is the only country where the poor receive a substantially lower share of cash transfers than the rich (see the targeting incidence in appendix D), which likely reflects the result of cat- egorical targeting where poor individuals who are young and single or couples without children are left out of the social assistance system. Thus, a budget-neutral UBI reform would give many extremely poor households larger transfers.7 These differences in relative performance across countries suggest that the pro- gressivity of baseline programs matters. Figure 4.3 shows that there is a very strong relationship between how much of existing resources go to the poorest decile, and how much baseline programs deliver better poverty impacts (measured in terms of the squared poverty gap). Referring to our categorization of welfare state modalities in box 4.1, in countries where the existing welfare state is progressive, a budget-neutral UBI reform would lead to significantly lower poverty impacts than the baseline system, while differences remain smaller in countries with flat or regressive welfare states. In contrast, level of coverage appears to have less of an influence on the relative performance of the baseline system. In Brazil and Indonesia, for instance, 17 percent and 18 percent, respectively, of the cash transfers we consider are distributed to the poorest decile; coverage of cash transfer programs decreases significantly along the income ladder (also see appendix D). Both features indicate a fairly progressive social assistance system. Accordingly, the baseline programs deliver a reduction in the squared poverty gap that is 116 percent (Brazil) and 86  percent (Indonesia) higher than a budget-neutral UBI reform. In Haiti, where over Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 135 FIGURE 4.3  Progressivity and Relative Performance of a Budget-Neutral UBI Reform 200 Baseline programs Haiti perform better Relative performance of baseline 150 Kazakhstan mix of programs (%) Brazil 100 Indonesia 50 Chile South Africa India Mozambique 0 Nepal Russian Fed. UBI performs better –50 0 5 10 15 20 25 30 35 Baseline benefits going to poorest decile (%) NOTE: Baseline programs are those we considered in each survey to be replaced by a UBI. The relative per- formance of baseline programs measures the additional decrease in squared poverty gap of the baseline programs with respect to a budget-neutral UBI reform, measured in percentage terms. A positive performance implies that the baseline programs deliver higher reductions in poverty. 30 percent of the benefits go to the bottom 10 percent, the allocation of the baseline programs results in a reduction of the squared poverty gap that is about 170  percent higher than a budget-neutral UBI reform. At the other end of the spectrum, in Russia, the poorest decile receives less than 10 percent of the benefits, and coverage of social programs is high and fairly flat across deciles. In such a situation, a budget-neutral UBI reform would increase the poverty-reducing impact of social assistance, because more resources would go to the poor. Inequality Impacts We conclude the section by looking at the impact on inequality (measured in Gini points) of baseline programs and the budget-neutral UBI reform (figure 4.4). In the low-income countries (Haiti, Mozambique, and Nepal), both the baseline system and the budget-neu- tral UBI reform have limited effects on inequality due to the very limited resources (and coverage) of the baseline transfer programs. In the remaining seven countries, the bud- get-neutral UBI scenario results in lower reductions in inequality than the baseline. This is not surprising because the switch from the baseline to the budget-neutral UBI reform implies going from a system that is often progressive in absolute terms (e.g., the size of the transfer declines with income) to a system that is neutral in absolute terms (e.g., the size of the transfer is the same for every member of the population irrespective of income). As shown by Enami, Lustig, and Aranda (2018), if spending as a proportion of prefiscal income is kept constant, a less progressive transfer will result in a smaller reduction in inequality. 136 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan FIGURE 4.4  Inequality Impacts: Reduction in Gini Coefficient Baseline Budget-neutral UBI reform 0.0 0.0 –0.1 –0.1 –0.2 –0.2 –1.5 –1.1 –1.3 –0.7 –1.5 –0.8 –1.2 –0.7 –7.4 –6.0 –1.7 –1.1 –1.8 –1.6 0 –1 Percentage points –2 –3 –4 –5 –6 –7 –8 Haiti Mozam- Nepal India Indonesia Brazil Kazakh- South Chile Russian bique stan Africa Fed. Low income Lower-middle income Upper-middle income High income NOTE: The impacts show the percentage point difference with respect to disposable income without transfers. Winners and Losers Any reform has winners and losers; and understanding winners and losers from UBI reforms is important not only from a political economy perspective (because losers could derail reforms), but also because having too many losers within the vulnerable popula- tion may defeat the very purpose of the reform. Figure 4.5 presents the proportion of winners and losers (and the magnitude of the change in income or consumption) for five population categories: the overall population, children age 0–6, the elderly (age 65 and up), and people in the bottom and top welfare deciles. Several facts emerge. In low-income countries that have both low coverage and low levels of spending, a budget-neutral move toward a UBI would minimally affect the entire population. In Haiti and Mozambique, most of the population (including the poor), would find itself within 1  percent of previous income/consumption levels with a UBI reform. To be sure, previous beneficiaries may be hurt, but coverage is so low to begin with that gains from the reform would be too diluted to observe meaningful changes. In countries with generous social pensions (such as South Africa), any reform toward a UBI would significantly hurt the elderly: more than two-thirds of the elderly population in South Africa would lose from a budget-neutral UBI reform that would replace social pensions. From a policy perspective, this implies that any UBI reform in countries with generous social pensions may need to consider topping up the UBI trans- fer with an additional transfer for the elderly. Given that social pensions often represent a significant share of social spending, such a constraint would, however, substantially increase the cost of a UBI reform. Similarly, in countries where the baseline social assistance spending that we take into consideration is relatively progressive or covers a large part of the poor population (Brazil, Indonesia), many poor households would lose from a budget-neutral UBI reform. In Brazil, more than 60 percent of the population in the poorest welfare decile would lose Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 137 FIGURE 4.5  Winners and Losers from a Budget-Neutral UBI Reform a. Haiti b. Mozambique 100 80 % of individuals 60 40 20 0 c. Nepal d. India 100 80 % of individuals 60 40 20 0 e. Indonesia f. Brazil 100 80 % of individuals 60 40 20 0 g. Kazakhstan h. South Africa 100 80 % of individuals 60 40 20 0 i. Chile j. Russian Fed. 100 80 % of individuals 60 40 20 0 Overall Age 0–6 Age 65+ D1 D10 Overall Age 0–6 Age 65+ D1 D10 Gain 10+ Gain 5–10 Gain 1–5 Within 1 Lose 1–5 Lose 5–10 Lose 10+ 138 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan from a UBI reform, and close to half the poor population would lose more than 10 percent of their income with respect to the baseline system of transfers. In Indonesia, more than one-third of the population in the poorest decile would lose more than 10 percent of their income. In contrast, in South Africa (where more than 90 percent of individuals in the poorest decile benefit from cash transfers), most people in the poorest decile would ben- efit from a UBI reform. This is because people in the poorest decile tend to benefit from child grants as opposed to social pensions, which are less generous; hence a UBI reform that replaces social pensions would distribute larger amounts to many poor households. If we look at the total poor population (not only the poorest decile) and exclude low-income countries where budgets remain very small, our simulations indicate that a significant number of the poor population would suffer losses from a UBI reform; the proportion would range between 26 percent and 53 percent. The notable exception is Russia, where only 16 percent of the poor population would suffer losses—again, the result of a social assistance system based on categorical targeting rather than means- tested transfers. The simulations show that in most countries the richest decile would gain, albeit relatively little, from a budget-neutral UBI reform. A few individuals from the richest decile may lose from the reform—but only a very small percentage would lose more than 10 percent of their income or consumption. Most of the distributional impacts from a budget-neutral reform would therefore especially occur among the poor and, in some countries, the middle deciles, but would only minimally affect the richer population. Regarding the middle deciles, there is strong heterogeneity across countries. Figure 4.6 shows, for selected countries, the proportion of winners from a budget-neu- tral UBI reform across income/consumption deciles (figure 4.6a), and how much people would win/lose from a reform (figure 4.6b). In low-income Haiti, Mozambique, and Nepal almost everyone stands to win from a budget-neutral UBI reform, simply because there are very few beneficiaries covered by existing safety nets; as shown in figure 4.5, how- ever, gains in these countries are very small. On the other side of the spectrum, in India, Brazil, and Indonesia, half of the poorest population (the first decile) stands to lose from a budget-neutral UBI reform. Overall, a majority of the population would win from a budget-neutral UBI reform, and the proportion of winners increases along the income/consumption ladder. On aver- age, across our sample of 10 countries, 70 percent of the population in the poorest decile stands to win from a budget-neutral UBI reform; the proportion increases to 92 percent moving toward the richest decile. Such a high proportion of winners across deciles seems to contrast with the finding that a budget-neutral UBI reform would, in most cases, lead to lower poverty reduction compared to the existing programs under consideration. Figure 4.6b shows that merely looking at winners and losers is, however, not enough; it is also important to consider the magnitude of the gains and losses. Across deciles, people losing from a budget-neu- tral UBI reform would lose substantially more than the winners would stand to win. When measured as a percentage of each country’s average disposable income, within the first decile, losers would lose on average more than double what winners would win Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 139 FIGURE 4.6  Winners and Losers across Welfare Deciles a. Percentage of winners from a b. Gains by winners (positive)/ budget-neutral UBI reform losses by losers (negative) Percentage of average disposable income 100 5 80 0 Percentage of population 60 –5 40 –10 20 –15 0 –20 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 Decile Decile Brazil Mozambique Russian Fed. Average 10 countries NOTE: Gains are measured among winners from the reform only, and losses among losers only. Gains and losses are measured as a percentage of each country’s average disposable income (see appendix D for definitions). (3.5 percent versus 1.6 percent); within the second decile, losers would lose almost three times what winners would win (4.8 percent versus 1.7 percent). The simulations also show that in most countries the majority of children would benefit from a budget-neutral UBI reform. Again, these results need to be interpreted with some caveats. Existing child-focused social assistance often targets poor children; thus, the children who may benefit from the UBI reform may not be those who are in the greatest need of assistance. Figure 4.7 confirms that this would be the case in many countries (all middle- and high-income ones except Russia)—especially those where a UBI would replace large cash transfers covering (extremely) poor children, such as Brazil and South Africa. Figure 4.7 also shows that in all countries, distributing a budget-neutral UBI only to adults would affect child poverty even further—sometimes significantly, such as (again) in Brazil and South Africa. The main reason is the differential fertility rates across the income ladder. Poorer households tend to have more children per adult; thus, at the household level, the per capita transfer would be smaller if children did not receive it. Overall, the findings show that a budget-neutral UBI reform has deep distributional impacts. Who wins and who loses from it depends very much on the type of programs the UBI is replacing, and on how well existing resources are geared toward the poorer population. In countries with generous social pensions, the elderly—and households with an elderly member—may lose significantly from a reform, but at the same time other population groups—such as children—may benefit from it. On the other hand, if a flat cash transfer replaces child grants and is only distributed to adults, poor chil- dren may lose from the reform. Furthermore, in countries where most of the resources already go to the poorest, many poor households may suffer from the reform, while 140 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan FIGURE 4.7  Universal versus Adults-Only UBI: Impact on Children Age 0–6 Baseline Budget-neutral UBI reform: all Budget-neutral UBI reform: adults only 0.0 0.0 0.0 –0.1 –0.1 –0.1 –0.1 –0.1 –0.1 –1.9 –1.8 –1.4 –1.3 –0.8 –0.7 –3.3 –1.8 –1.1 –0.7 –0.3–0.2 –19.0 –16.1 –12.0 –1.8–1.7–1.1 –1.8 –1.8 –1.1 0 Percentage points –5 –10 –15 –20 –25 Haiti Mozam- Nepal India Indonesia Brazil Kazakh- South Chile Russian bique stan Africa Fed. Low income Lower-middle income Upper-middle income High income NOTE: The impacts show, among children, the percentage point difference with respect to disposable income without transfers. richer households may only minimally benefit from it. The only countries where a UBI reform may have limited distributional impacts are those where both the coverage and incidence of cash transfers are already spread across population and income groups, such as Russia. When this is the case, a budget-neutral UBI reform may in fact increase the equity of the system both across and within population groups. Replacing Baseline Social Assistance Programs with a More Generous UBI Impact on Poverty In most countries, cash transfers disproportionately cover the poor; therefore, a bud- get-neutral UBI reform leads to lower poverty impacts than the existing mix of programs. But what about more generous UBIs? In this section, besides the baseline scenario, we consider three scenarios with increasing levels of generosity: the equivalent benefits, poverty gap, and poverty line UBI scenarios. In the equivalent benefits scenario, the size of the UBI transfer equals the beneficiary-weighted average of baseline cash benefits among the existing pool of beneficiaries; in most cases, and unless every member of the relevant population is a beneficiary of all relevant programs, the size of the UBI and the total budget will be higher than under the budget-neutral case. In the poverty gap scenario, the size of the UBI equals the average distance from the poverty line among the poor; in all the con- templated cases, the poverty gap scenario delivers higher transfers than the equivalent benefits scenario, except in Kazakhstan.8 Our last, and most generous, scenario is the poverty line scenario, where the size of the UBI equals the corresponding poverty line. Note that the poverty line scenario is the only one where a UBI transfer can completely Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 141 eradicate poverty, since under the other scenarios some extremely poor households may not receive a high enough transfer to fully overcome poverty. Figure 4.8 looks at the scenarios’ poverty impacts. There is a direct relation between the generosity of the UBI transfer and its impacts on poverty and extreme poverty. The relation is strongest in low-income countries (in our sample, Haiti, Mozambique, and Nepal), where the baseline level of transfers makes only a minimal dent on poverty and extreme poverty due to relatively low coverage. To observe a meaningful impact of a UBI in these countries, the transfer should be at least the average level of transfers received by the baseline beneficiaries (the equivalent benefits scenario), or, possibly, the average distance of the poor from the poverty line (the poverty gap scenario). In middle- and high-income countries, to attain impacts on poverty (as measured by the squared poverty gap) that equal those of the baseline transfers, the generosity of transfers must be increased at least to the average level currently received by the exist- ing pool of beneficiaries (the equivalent benefits scenario).9 But in some countries (India, FIGURE 4.8  Poverty Impacts of More Generous UBI Schemes (Spending Only) a. Poverty headcount 0 –10 Percentage point –20 reduction –30 –40 –50 –60 –70 b. Squared poverty gap 0 –5 –10 Percentage point –15 reduction –20 –25 –30 –35 Haiti Mozam- Nepal India Indonesia Brazil Kazakh- South Chile Russian bique stan Africa Fed. Low income Lower-middle income Upper-middle income High income Baseline Budget-neutral UBI reform Equivalent benefits Poverty gap Poverty line NOTE: The impacts show the percentage point difference with respect to disposable income without transfers. 142 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan Indonesia) even such a generosity level would not suffice. And in all countries, the sce- narios that deliver significantly higher poverty impacts are the poverty gap and poverty line scenarios. But how much would these programs cost? Would countries be able to afford such levels of spending? Before responding to these questions, note that the poverty line scenario yields a relatively small improvement in poverty reduction for the extremely poor—as measured by the squared poverty gap—than the poverty gap scenario. This is an important result since, as we discuss below, the poverty line scenario is often too expensive and may not be a viable option. Fiscal Cost Figure 4.9 looks at the cost of the various scenarios, taking into consideration the cost of the transfers but not the cost of program administration. It also compares the cost of each scenario with baseline spending on social programs and with existing spending on energy subsidies (from 2015 International Monetary Fund statistics). A first observation is that the cost of UBI rapidly increases with the generosity level. To distribute to the entire population the average levels of transfers currently received by the beneficiary population (the equivalent benefits scenario), in low-income countries— which tend to face limited coverage of social programs—the cost would increase from 0.4–0.7 percent of GDP to 2.5 percent (Nepal) and more than 4 percent (Haiti, Mozam- bique) of GDP. But to make a meaningful dent on poverty, the generosity of the UBI would need to be at least equal to the average distance of the poor from the poverty line—the poverty gap scenario. The costs of such a transfer would be much higher, ranging from 7 percent (Nepal) to more than 20 percent of GDP (Mozambique). In middle- and high-in- come countries, the poverty gap scenario’s cost remains lower, reaching a maximum of 8 percent of GDP in South Africa. This is because in middle- and high-income countries, the poverty line represents a lower proportion of average income; thus the transfer, in FIGURE 4.9  Total Spending as a Percentage of GDP by UBI Scenario 50 40 Cost (% GDP) 30 20 10 0 Haiti Mozam- Nepal India Indonesia Brazil Kazakh- South Chile Russian bique stan Africa Fed. Low income Lower-middle income High income Equivalent benefits Poverty gap Poverty line Baseline transfers Baseline transfers and energy subsidies SOURCE: Energy subsidies: International Monetary Fund 2015. Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 143 relative terms, becomes lower. Finally, eradicating poverty through a UBI that distributes the equivalent of the poverty line to everyone in Haiti, Mozambique, and Nepal would cost 45 percent, 48 percent, and 36 percent of GDP, respectively. In middle- and high-in- come countries, again, the poverty line scenario would cost less, between 11  percent (Indonesia) and 22 percent of GDP (India). Nonetheless, these are extremely high financ- ing needs; hence the poverty line UBI scenario is not likely to represent a viable option in most low- and middle-income countries. Overall, even for scenarios that may not fully eradicate poverty such as the equiv- alent benefits and poverty gap scenarios, the additional financing needs appear to be relatively high. Nevertheless, to assess feasibility, what really matters is the additional cost of a UBI reform, taking into consideration the fiscal resources that reform would free up. In addition to substituting selected cash transfers, a prime candidate for replace- ment would be subsidies—in particular, energy subsidies. These represent significant amounts in many countries, and are often regressive in nature. Therefore, replacing energy subsidies with a UBI may represent a politically feasible move toward more equi- table redistribution, without sacrificing the universal nature of the subsidy. In figure 4.9, we compare the cost of the various UBI scenarios with the cost of the baseline programs they would replace, and the amount that countries currently spend on energy subsidies. In most cases, replacing energy subsidies with a UBI may not lead to meaningful impacts on poverty since subsidies remain below the financing needs of generous UBI transfers. There seem to be a few exceptions, however. In Mozambique, India, Indonesia, and Russia, replacing selected social assistance programs and subsidies with a UBI would allow for financing the equivalent benefits scenario—which, in most cases, would still deliver relatively low poverty impacts. Only in Russia would the combi- nation of replacing existing programs and subsidies with a UBI allow financing of either the equivalent benefits or poverty gap scenario. In most countries, a UBI reform that would have a meaningful impact on poverty would therefore need to be financed out of additional taxation. But under additional taxation, the net impacts on poverty and inequality may change significantly, because everyone would not only receive a flat transfer, but would also be taxed to finance the UBI scheme.10 The net impacts on poverty and inequality of more generous UBI schemes would depend therefore on the specific choice of taxation instruments, which we explore below. Financing the Gap through Higher Taxation Our results show that the costs of the poverty line scenario are too high to represent a viable policy option. We thus focus here on the equivalent benefits and poverty gap sce- narios and look at the distributional implications when they are financed through various forms of taxation. As detailed in appendix D, the scenarios use consumable income (in contrast to disposable income, as used in the previous sections) as the welfare indica- tor and rely on the fiscal incidence results by decile available from the Commitment to Equity Data Center.11 144 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan Figure 4.10 presents the change in percentage points of the squared poverty gap for the baseline system and the three UBI scenarios: budget neutral, equivalent bene- fits, and poverty gap. The last two are not budget neutral, and we present the poverty impacts for three financing scenarios: (1) a proportional increase in direct taxes, (2) a proportional increase in indirect taxes, and (3) a lump-sum increase in direct taxes for the richest decile. Cases that are left blank reflect scenarios that were not feasible to compute or that would not be meaningful. For instance, in the case of India, covering the financing gap of the poverty gap scenario with a lump-sum increase of direct taxes of the top 10 percent would render the per capita disposable income of this decile below that of the ninth decile. In Indonesia, the survey’s highest income is below the threshold for direct taxes (Jellema, Wai-Poi, and Afkar 2017) so the scenario in which the financing gap is covered by a proportional increase in direct taxes cannot be simulated. Note that we show the poverty impact for the poverty gap scenario without financing not because we think this is a viable option: if spending increases, someone will need to pay for it either through higher taxes, borrowing (which could result in higher taxes on future gen- erations), or grants from abroad. We show the without financing option for illustrative purposes to demonstrate that, in general, the with financing scenarios feature poverty impacts that are very similar to the without financing. It is also important to stress that we are not advocating any specific financing scenario. The purpose of the analysis is to illustrate how prototypical alternatives in the policy realm may affect UBI poverty results. With this in mind, the results—perhaps unsurprisingly—show that, even after the impact of higher taxes is taken into consideration, the poverty gap scenario results in the largest impact on poverty. Furthermore, for either of the nonneutral UBI scenarios, financing the budgetary gap with a proportional increase in indirect taxes is worse from the poverty impact point of view than financing it with a proportional increase in direct taxes. This is because in low- and middle-income countries, large parts of the population do not pay direct taxes either by law or due to informality in labor markets; but most are likely to pay some consumption taxes. Figure 4.10 shows that although the scenario in which the financing gap is covered with a lump-sum increase of the top 10 percent’s direct taxes will not affect the incomes of the poor by definition, the difference in pov- erty impact between this scenario and the one financed by a proportional increase in direct taxes is extremely small. Again, this is because the incidence of direct taxes for the poor in most of the countries we cover is very small to begin with, and a proportional increase of this incidence would not affect the income of the bottom deciles. Given the above results, it may appear that an attractive policy option would be to select a poverty gap UBI scheme and cover the financing gap with a proportional increase in direct taxes. However, as shown in figure 4.11, in most cases the burden of direct taxes on the top 10 percent would need to increase substantially under all taxation scenarios— often by politically unrealistic proportions. In India, direct taxes on the top 10 percent would need to rise from 2.2  percent to 68.4  percent; in Brazil, from 7.2  percent to 24.5 percent; in South Africa, from 19.9 percent to 40.3 percent; in Chile, from 5.4 per- cent to 38.4 percent. The only case in which this option has more moderate impacts is Russia, where the incidence would have to increase from 9.0 percent to 13.2 percent. Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 145 FIGURE 4.10  Squared Poverty Gap Impact for Consumable Income under Baseline and Different UBI Scenarios a. India 0 –0.1 Percentage points –0.5 –0.5 –0.5 –0.4 –0.5 –2 –4 –5.5 –6 –8 –7.4 b. Brazil Percentage points 0 0.0 –1 –1.1 –2 –2.3 –3 –2.6 –2.6 –2.6 –4 –4.4 –4.3 –4.0 –4.4 –5 c. Chile 0 0.0 Percentage points –0.5 –0.9 –1.2 –1.2 –1.2 –2 –4 –5.3 –4.4 –5.3 –5.3 –6 d. Indonesia 0 Percentage points –1 –1.0 –1.6 –1.6 –1.4 –1.6 –2 –3 –4 –3.9 –4.6 –5 –4.6 e. South Africa 0 Percentage points –5 –10 –8.8 –11.5 –13.2 –15 –14.2 –14.2 –14.3 –20 –21.6 –23.8 –23.8 –23.8 –25 f. Russian Fed. 0 Percentage points –0.2 –0.3 –0.5 –0.5 –0.6 –0.6 –0.6 –0.7 –0.7 –0.7 –0.7 –1.0 UBI equivalent benefit UBI poverty gap Baseline UBI Without Proportional Proportional Lump-sum tax consumable budget financing increase in increase in on richest decile income neutral direct tax indirect tax 146 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan FIGURE 4.11  Incidence of Taxes for the Top 10 Percent under Alternative UBI and Financing Scenarios a. India 0 Tax incidence of the –2.2 –4.0 –4.8 –7.2 –7.8 –20 richest decile –40 –33.4 –60 –68.4 –80 –83.3 –100 b. Brazil Tax incidence of the 0 richest decile –10 –7.2 –15.8 –13.0 –17.1 –18.5 –20 –21.4 –30 –24.5 –30.0 c. Chile 0 Tax incidence of the –10 richest decile –5.4 –10.0 –10.8 –11.2 –12.0 –20 –22.2 –30 –40 –38.4 –50 –40.9 d. Indonesia 0 Tax incidence of the –10 –4.8 richest decile –5.5 –6.7 –20 –17.0 –30 –40 –47.4 –50 e. South Africa 0 Tax incidence of the –10 richest decile –20 –12.6 –19.9 –15.7 –22.9 –30 –26.0 –27.3 –40 –40.3 –50 –44.5 f. Russian Fed. Tax incidence of the 0 richest decile –10 –9.0 –7.3 –8.8 –9.9 –11.4 –15.6 –13.2 –20 –20.8 Direct tax Indirect tax Baseline UBI equivalent benefit UBI equivalent benefit UBI poverty gap UBI poverty gap proportional lump-sum tax on proportional lump-sum tax on increase richest decile increase richest decile Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 147 The impacts of the various financing scenarios on the middle deciles remain more ambiguous, and depend very much on which/how much taxes the middle class currently pays in each country. Simulations of changes in the taxation burden of various financ- ing scenarios on the middle three deciles (deciles 4–6) show, however, that financing a UBI with indirect taxes will put a much heavier burden on the middle classes than financing it with direct taxes—which is again a reflection of the middle classes paying taxes on consumption, while the incidence of direct taxes remains often small even on the middle classes. Financing the poverty gap scenario with direct taxes would increase the tax burden on the middle three deciles between 0.2 percent (Chile) to a maximum of 3 percent (Brazil); in contrast, financing the poverty gap scenario with indirect taxes would increase the tax burden on the middle three deciles from 3 percent (Russia) to as much as 20.4 percent (India). Conclusions This chapter presents an overview of the issues and implications that a possible introduc- tion of a UBI might entail. These include a variety of trade-offs across parameters such as coverage, adequacy, and costs, as well as generating significant distributional effects. Emerging estimates from microsimulations are discussed within a broader framework to help navigate and understand UBI relative to existing, cash-based social assistance programs. Several policy-relevant lessons emerge from the analysis. To begin with, we sim- ulate the impacts of a budget-neutral UBI that replaces selected cash transfer/subsidy programs in each country. We find that a UBI is less effective at reducing poverty than existing programs. The difference in impacts is small in absolute terms, but in relative terms is quite sizable. In fact, existing programs are on average about 60 percent more effective in poverty reduction than a UBI. This is because most existing programs, even if they may be only slightly progressive and miss some of the poor, tend to cover relatively more of the extremely poor population. With a few exceptions, a budget-neutral UBI reform would take resources away from poor households that are benefiting from exist- ing programs, and give to richer households currently not benefiting. Importantly, these findings do not account—or do so only indirectly—for other poverty-related aspects that may affect performance and that are not easily observable from survey and administra- tive data, such as transaction costs to access benefits, stigma, leakages, etc. A second relevant finding is that a budget-neutral UBI reform leads to significant distributional impacts. While, in some countries, differences in poverty impacts remain modest, on average a UBI reform would generate more winners than losers among the poorest segments of the population. However, the amount of gains by the winners is lower (about half among the poorest decile) than the loss of the losers, raising a chal- lenging dilemma for policy making. Who wins and who loses depends very much on the programs a UBI would replace. Often, a large spending item is social pensions, and a budget-neutral UBI reform that would replace social pensions would affect significantly 148 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan many elderly people. Similarly, if the UBI would replace programs that cover children, or the UBI transfers were only given to adults, then children—in particular poor ones— would be penalized by the reform. To avoid penalizing groups that are in need of social assistance, it is critical to go beyond the analysis of poverty impacts and consider the distributional implications in thinking through any reform. Such an analysis may also provide an understanding of who may favor or oppose the reform. The differences in poverty and distributional impacts from a budget-neutral UBI reform that would replace social assistance programs appear to be relatively smaller in countries where coverage of social programs is already quite high and transfers are spread across the entire population. In countries where many social programs are concentrated among the poor, a budget-neutral UBI reform would lead to significant dif- ferences in poverty impacts and would see many people—especially poor ones—losing from the reform. In other words, the less existing programs are poverty targeted, the more a UBI reform may be a viable instrument. To make a significant dent in poverty, however, in most countries the generosity levels of a UBI transfer would need to be higher than a budget-neutral reform would allow. We consider various scenarios, from distributing to each individual the average transfer received by beneficiaries of baseline programs, to distributing an amount equal to the poverty line—which is the minimal amount necessary for a UBI to fully eradicate poverty. These more generous transfers have, of course, larger impacts on poverty and inequality (although if the same amount of resources were distributed only among the poor and vulnerable populations, poverty impacts would be even larger). Nevertheless, they are also more expensive. Overall, we find that providing a UBI with a generosity level that has a meaningful impact on poverty is financially prohibitive in low-income coun- tries, but may be a relatively more viable option in some middle- and high-income countries. Giving every citizen a transfer equal to the average distance of the poor from the pov- erty line would cost 7 and 20 percent of GDP in Nepal and Mozambique, respectively; fully eradicating poverty with a UBI in the same countries would cost 36 and 48 percent of GDP. In middle- and high-income countries, however, giving every citizen a transfer equal to the average distance of the poor from the poverty line never surpasses 8 percent of GDP; and eradicating poverty by distributing to every citizen the equivalent of the pov- erty line would cost between 8 and 22 percent of GDP (although the amounts are highly sensitive to the choice of poverty line). To assess whether these costs are fiscally sustainable, the sources of financing must be examined. A potentially relevant source of financing may be subsidies—espe- cially untargeted energy subsidies, which tend to be regressive in nature. We find that in order to finance a UBI with meaningful impacts on poverty, replacing subsidies will not be sufficient, and most countries will need to increase taxes. Only a handful of countries with relatively large energy subsidies may be able to finance a UBI with meaningful poverty impacts by replacing subsidies (e.g., Kazakhstan, Russia). Under additional taxation, the net impacts on poverty and inequality may change significantly, because everyone would not only receive a flat transfer but would also be taxed to finance the UBI scheme. We find that the poverty impacts taking into consideration Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 149 taxation are only slightly smaller than the gross poverty impacts that do not look at the financ- ing side. While the findings may appear counterintuitive at first sight, they are consistent with the fact that in many low- and middle-income countries, the richest deciles contrib- ute the most toward taxation (in particular when informality is widespread), while taxation minimally affects the income of the poor. We find that financing the budgetary gap with a proportional increase in indirect taxes is slightly worse from the poverty impact point of view than financing it with a proportional increase in direct taxes. This is, again, because in low- and middle-income countries large parts of the population do not pay direct taxes. The fact that generous UBI programs continue to have meaningful impacts on pov- erty, even when considering taxation, suggests that they might be viable policy options. To assess their feasibility, however, it is necessary to look at the impacts they would have on the richest segments of the population, which in most cases will have to con- tribute most of the financing needs. Their impacts are not trivial, since on the one hand the richer segments are taxed; but on the other hand, they receive a transfer as well. We find that in most cases, to finance UBI levels that have a meaningful impact on poverty, the burden of taxation on the top 10  percent would need to increase substantially—often by politically unrealistic proportions. In India, for instance, the direct taxes on the top 10 percent would need to rise from 2.2 percent to 68.4 percent; in Brazil, from 7.2 per- cent to 24.5 percent; in South Africa, from 19.9 percent to 40.3 percent; and in Chile, from 5.4 percent to 38.4 percent. The only case among our sample countries in which this option seems within feasible realms is Russia, where the incidence would have to increase from 9.0 percent to 13.2 percent. To end where we began, we conclude by summarizing and locating the main find- ings by country contexts: •• Where social assistance provides relatively adequate benefits, substantial cov- erage, and slight progressivity, policy makers could consider tackling specific bottlenecks that hamper eligibility, access, coverage, or delivery within the existing system. If a UBI is to be considered, it may have to be motivated by objectives other than a poverty-related one (e.g., automation-driven job insecu- rity, social dividends, etc.). •• Where coverage is high, but not progressive, a UBI could be considered an option, although some vulnerable (age) groups may suffer from the shift. •• Where social assistance is limited, but provided progressively, a UBI would extend coverage but also flatten the distribution. If budget neutral, this means less for more, and likely less at the bottom. •• Where social assistance is patchy and flat or regressive, a UBI could be an option to expand coverage if financed via progressive income taxation, elimination of energy subsidies, or redistribution of windfall revenues. Most low-income countries may not display those financing features; but some middle-income, resource-rich countries may. 150 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan •• For a typical low-income setting, a UBI could expand coverage but is clearly financially daunting. Other factors, such as diversity in contexts at the subna- tional level, may suggest the need for design flexibility (e.g., a balance of in-kind and cash transfers, sensible ways to account for children, etc.)—possibly making the rigid design of a UBI less palatable. Notes 1. The reference years and data sources are discussed in appendix D. The classification of coun- tries by category of gross national income per capita in 2011 purchasing power parity is based on World Bank thresholds; see https://datahelpdesk.worldbank.org/knowledgebase/ articles/906519-world-bank-country-and-lending-groups. Russia is classified as a high-income country because that is how it was classified in 2015, the data collection year of the survey for which the analysis presented here was conducted. 2. The incidence of taxes is obtained from the Commitment to Equity Institute’s Data Center on Fiscal Redistribution. For a description of the methodology that underpins the tax incidence analysis, see Lustig (2018) and appendix D. 3. The data sources by country are as follows: Brazil, Higgins and Pereira (2017); Chile, Martinez-Agu- ilar and Ortiz-Juarez (2016); India, Kundu and Cabrera (2018); Indonesia, Afkar, Jellema, and Wai-Poi (2015); Russia, Malytsin and Popova (2016); and South Africa, Inchauste et al. (2016). 4. For a discussion of contributory social security systems and the challenges of reforming them see, for example, Barr and Diamond (2008). 5. For internal consistency, we consider the costs as captured by the household survey, as opposed to the administrative data. 6. For more detail, see Jolliffe and Prydz (2015, 2016, 2017). 7. This is consistent with findings by Lopez-Calva et al. (2017). 8. Kazakhstan is the only country where the equivalent benefits scenario is more expensive than the poverty gap scenario because of a combination of relatively generous benefits with respect to a relatively low international poverty line. 9. To be sure, our simulations are highly sensitive to the choice of poverty line, which vary by country. If two countries have the same poverty line, but one is wealthier than the other, as a percentage of GDP, transfers will appear to be lower in the wealthier country. While using international poverty lines is the most appropriate option for a comparative study, more detailed country analyses should pay attention to country-specific characteristics in determin- ing the level of benefits. Accordingly, the impacts in Kazakhstan and Russia remain relatively modest because of the choice of using the international poverty line, which represents a rela- tively low proportion of each country’s average income. 10. Replacing subsidies to finance a more generous UBI would also have differential effects across income deciles. However, given that subsidies may not be sufficient to finance a UBI with meaningful impacts on poverty, we focus the analysis on higher taxation alone. 11. Consumable income is equal to disposable income minus indirect taxes plus indirect subsidies. Chapter 4. Comparative Effects of Universal Basic Income: Emerging Issues and Estimates 151 References Afkar, Rythia, Jon Jellema, and Matthew Wai-Poi. 2015. “CEQ Master Workbook: Indonesia. Ver- sion: February 26, 2015.” Commitment to Equity Data Center on Fiscal Redistribution, Tulane University, New Orleans. Barr, Nicholas, and Peter Diamond. 2008. Reforming Pensions: Principles and Policy Choices. New York: Oxford University Press. Brown, Caitlin, Martin Ravallion, and Dominique van de Walle. 2016. “A Poor Means Test? Econo- metric Targeting in Africa.” NBER Working Paper 22919. National Bureau of Economic Research, Cambridge, MA. https://www.nber.org/papers/w22919.pdf Browne, James, and Herwig Immervoll. 2017. “Mechanics of Replacing Benefit Systems with a Basic Income: Comparative Results from a Microsimulation Approach.” Journal of Economic Inequality 15 (4): 325–44. Coady, David, and Delphine Prady. 2018. “Universal Basic Income in Developing Countries: Issues, Options and an Illustration for India.” IMF Working Paper 18/174. International Monetary Fund, Washington, DC. https://www.imf.org/en/Publications/WP/Issues/2018/07/31/Universal-Basic- Income-in-Developing-Countries-Issues-Options-and-Illustration-for-India-46079 Enami, Ali, Nora Lustig, and Rodrigo Aranda. 2018. “Analytic Foundations: Measuring the Redis- tributive Impact of Taxes and Transfers.” In Commitment to Equity Handbook: Estimating the Impact of Fiscal Policy on Inequality and Poverty, edited by Nora Lustig, chapter 2. Washington, DC: Brookings Institution Press. Hanna, Rema, and Benjamin A. Olken. 2018. “Universal Basic Incomes versus Targeted Transfers: Anti-Poverty Programs in Developing Countries.” Journal of Economic Perspectives 32 (4): 201– 26. https://economics.mit.edu/files/15434 Higgins, Sean, and Claudiney Pereira. 2014. “The Effects of Brazil’s Taxation and Social Spending on the Distribution of Household Income.” In The Redistributive Impact of Taxes and Social Spending in Latin America, edited by Nora Lustig, Carola Pessino, and John Scott, special issue. Public Finance Review 42 (3): 346–67. Hoynes, Hilary W., and Jesse Rothstein. 2019. “Universal Basic Income in the US and Advanced Countries.” NBER Working Paper 25538. National Bureau of Economic Research, Cambridge, MA. https://gspp.berkeley.edu/assets/uploads/research/pdf/Hoynes-Rothstein-UBI-081518. pdf IMF (International Monetary Fund). 2017. Fiscal Monitor: Tackling Inequality. Washington, DC: IMF. https://www.imf.org/en/Publications/FM/Issues/2017/10/05/fiscal-monitor-october-2017 Inchauste, Gabriela, Nora Lustig, Mashekwa Maboshe, Catriona Purfield, Ingrid Woolard, and Precious Zikhali. 2016. “CEQ Master Workbook: South Africa. Version: March 6, 2016.” Com- mitment to Equity Data Center on Fiscal Redistribution, Tulane University, New Orleans. Jellema, Jon, Matthew Wai-Poi, and Rythia Afkar. 2017. “The Distributional Impact of Fiscal Policy in Indonesia.” In The Distributional Impact of Taxes and Transfers: Evidence from Eight Low- and Middle-Income Countries, edited by Gabriela Inchauste and Nora Lustig. Washington, DC: World Bank. http://commitmentoequity.org/wp-content/uploads/2018/04/The-Distributional- Impact-of-Taxes-and-Transfers_Inchauste-Lustig.pdf Jolliffe, Dean, and Espen Beer Prydz. 2015. “Global Poverty Goals and Prices: How Purchasing Power Parity Matters.” Policy Research Working Paper 7256. World Bank, Washington, DC. https://openknowledge.worldbank.org/bitstream/handle/10986/21988/Global0poverty0powe r0parity0matters.pdf?sequence=1&isAllowed=y —. 2016. “Estimating International Poverty Lines from Comparable National Thresholds.” Policy Research Working Paper 7606. World Bank, Washington, DC. http://documents. 152 Jamele Rigolini, Nora Lustig, Ugo Gentilini, Emma Monsalve, and Siyu Quan worldbank.org/curated/en/837051468184454513/Estimating-international-poverty-lines- from-comparable-national-thresholds —. 2017. “Societal Poverty: A Relative and Relevant Measure.” Policy Research Work- ing Paper 8073. World Bank, Washington, DC. http://documents.worldbank.org/curated/ en/133671495562984832/pdf/WPS8073.pdf Kundu, Sridhar, and Maynor Cabrera. 2018. “CEQ Master Workbook: India (2011–2012). Prelim- inary version: April 12, 2018.” Commitment to Equity Data Center on Fiscal Redistribution, Tulane University, New Orleans. Lopez-Calva, Luis Felipe, Nora Lustig, Mikhail Matytsin, and Daria Popova. 2017. “Who Benefits from Fiscal Redistribution in the Russian Federation?” In The Distributional Impact of Taxes and Transfers: Evidence from Eight Low- and Middle-Income Countries, edited by Gabriela Inchauste and Nora Lustig. Washington, DC: World Bank. http://documents.worldbank.org/curated/ en/947831504161332955/pdf/119229-PUB-PUBLIC-pubdate-8-24-17.pdf Lustig, Nora, ed. 2018. Commitment to Equity Handbook: Estimating the Impact of Fiscal Policy on Inequality and Poverty. Washington, DC: Brookings Institution Press. Malytsin, Mikhail, and Daria Popova. 2016. “CEQ Master Workbook: Russia. Version: March 17, 2016.” Commitment to Equity Data Center on Fiscal Redistribution, Tulane University, New Orleans. Martinez-Aguilar, Sandra, and Eduardo Ortiz-Juarez. 2016. “CEQ Master Workbook: Chile. Version: October 7, 2016.” Commitment to Equity Data Center on Fiscal Redistribution, Tulane Uni- versity, New Orleans. Ortiz, Isabel, Christina Behrendt, Andrés Acuña-Ulate, and Quynh Anh Nguyen. 2018. “Univer- sal Basic Income Proposals in Light of ILO Standards: Key Issues and Global Costing.” ESS Working Paper 62. International Labour Office, Geneva. https://www.ilo.org/wcmsp5/groups/ public/---ed_protect/---soc_sec/documents/publication/wcms_648602.pdf Ravallion, Martin and Shaohua Chen. 2017. “Welfare-Consistent Global Poverty Measures.” Policy Research Working Paper 8170. World Bank, Washington, DC. https://openknowledge. worldbank.org/handle/10986/27977 —. 2018. World Development Report 2019: The Changing Nature of Work. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/30435 Financing a Universal Basic Income: CHAPTER 5 A Primer Teresa Ter-Minassian T his chapter attempts to develop a general qualitative methodological framework to analyze financing options for a universal basic income (UBI), highlighting les- sons from the literature on how such options may be expected to affect public finances, income distribution, and poverty in different types of countries.1 Such a framework could be used to guide analysis of a UBI and financing options in individual countries that may be considering the introduction of such an instrument. The chapter expands the range of financing options explored in the microsimulations discussed in chapter 4. It focuses on options to finance a nonbudget-neutral UBI—that is, one that would complement or replace existing social safety net mechanisms, but increase the generosity of protection. This focus is justified by the fact that, as suggested by the above-mentioned simulations, a budget-neutral UBI is likely to end up reducing the degree of protection to the poor (especially the extremely poor) in most countries. The chapter concentrates on emerging market and low-income countries, because (1) informational and institutional constraints on targeting effectiveness are likely to be more significant in such countries, and (2) existing studies on the effects of UBI introduc- tion mainly focus on advanced economies.2 In the next section, the chapter outlines a step-by-step framework for assessing the fiscal space to accommodate the cost of a proposed UBI without raising additional reve- nues or cutting other spending—that is, through a reduced surplus or through borrowing. This assessment would involve estimating both short-term financing possibilities and their longer-term impact on debt sustainability, taking due account of risks. It would also involve verifying the consistency of the new levels of the budget balance and the public debt with any existing fiscal rules. 153 154 Teresa Ter-Minassian It should be stressed that a decision to use available fiscal space to finance a UBI, rather than other potential revenue or expenditure policies (e.g., cuts in especially dis- tortive taxes or increases in education, health, or infrastructure spending), should be made by governments on the basis of a careful analysis of the growth and distributional implications of the alternative policies, and of their alignment with societal preferences. If, as is often likely to be the case, there appears to be little or no fiscal space to accommodate a proposed UBI with unchanged policies, countries would need to scale back its generosity, and/or consider a range of expenditure-saving and/or revenue-raising measures to finance it. The chapter’s second section looks at the scope and impact of various types of expenditure savings that could be pursued to fund, wholly or partly, the additional fiscal cost of a UBI. It notes the usefulness of international benchmarking (using appropriate comparators) in identifying allocative and technical inefficiencies in both broad spending areas and specific expenditure programs. It also stresses the importance of sound bud- getary institutions and processes in creating sustainable fiscal space for a UBI through gains in spending efficiency. In particular, the chapter focuses on two areas of spending where reforms could yield significant savings in some countries to fund a proposed UBI: energy subsidies and the public wage bill. The third section analyzes the likely effects of alternative options to finance a UBI through revenue mobilization measures. It notes that the choice of revenue instruments inevitably involves trade-offs between different objectives, including efficiency, redistri- butional power, administrative and compliance ease, and political feasibility, since no potential revenue instrument fares well in all these dimensions. While the trade-offs are likely to vary by country, reflecting a range of economic and non-economic factors, some instruments and policies score better than others in a number of dimensions. The chap- ter discusses, on a tax-by-tax basis, the main considerations that should be taken into account by policy makers in emerging market and low-income countries in assessing these trade-offs. The final section summarizes the chapter’s main conclusions. Assessing the Fiscal Space for a UBI It is, of course, possible to envisage a UBI calibrated ex ante to replace a set of existing noncontributory social protection programs, at an unchanged fiscal cost for the budget. Under such a scenario, the analysis of first-round effects of the substitution would con- centrate on the distribution of the new transfer among different groups of households (different deciles of income, poor versus nonpoor, different age groups, urban versus rural, etc.), compared with that of the replaced transfers. This analysis could be con- ducted through microsimulations based on household surveys, such as those discussed in chapter 4. To the extent that the household groups had different propensities to consume or different income-leisure preferences, there would be second-round macro- economic and fiscal implications, the assessment of which would require the use of empirically calibrated dynamic general equilibrium models. Chapter 5. Financing a Universal Basic Income: A Primer 155 In principle, it cannot be excluded that a budget-neutral replacement of targeted social transfers with a UBI could turn out to be pro-poor under certain circumstances, particularly when the targeting mainly reflects objectives other than poverty alleviation. However, this is unlikely to be the case in practice in most instances. In fact, the microsimulations of this scenario presented in chapter 4 suggest that existing systems of social protection in most of the countries analyzed tend to be more cost-effective than a budget-neutral UBI in reducing poverty headcounts and the squared poverty gap (which is a better indicator of impact on the extremely poor than the pov- erty headcount). Moreover, in many instances, the welfare gains that would be obtained by some groups of poor from the shift to a UBI are estimated to fall short of the losses that would be experienced by other groups. Therefore, in practice, policy makers considering the introduction of a UBI would likely look to increase the generosity of the social protection system, with attendant higher fiscal costs. The simulations in chapter 4 modeled the effects of three alternatives for the level of the UBI: one equivalent to the average benefit provided to the recipients of current targeted transfers, one equivalent to the average poverty gap, and one sufficient to raise the entire population above the poverty line. A policy to increase the generosity of current social assistance mechanisms requires careful assessment of the existing fiscal space to accommodate the additional budgetary cost without endangering financing access and debt sustainability and—in the event that the space does not exist or is insufficient—of the options for creating (or expanding) it through expenditure savings and/or the mobilization of additional revenues. The assessment of fiscal space for a UBI costlier than existing social protection, on unchanged revenue and other expenditure policies, essentially involves the following steps: •• Calculation of the additional fiscal cost, in terms of the primary and overall fiscal balances, and of gross borrowing requirements •• Analysis of possible sources of financing of the additional cost in the short to medium term •• Evaluation of the UBI’s impact on medium- to long-term debt sustainability •• Evaluation of the consistency of the resulting fiscal balances with any existing numerical rules The calculation of the first-round additional annual cost in terms of the primary balance is relatively straightforward in the case of a fixed-amount UBI applicable to each resident (or citizen) in the country.3 Its complexity increases if other eligibility require- ments (such as age) are stipulated or if the amount varies across categories of recipients, as such features involve additional (e.g., demographic) informational needs. Project- ing the cost beyond the first year also requires making demographic projections and assumptions about adjustments of the monetary amount of the UBI over time (e.g., to keep its level constant in real terms). 156 Teresa Ter-Minassian Estimating the impact on the overall balance is more complex, as this requires assessing the effects of the change in the primary balance on the debt stock, and on its cost over the short to medium term. In particular, this assessment should reflect the structure of the debt (composition and maturity profile) and the likely elasticity of inter- est rates to changes in the debt stock.4 These effects may initially be quite limited for countries with strong policy credibility and relatively low debt levels, but may be signif- icant for countries with prolonged histories of fiscal deficits, gross debt levels above the average of comparable countries, and/or relatively low liquid government assets. They would also be affected by the cyclical performance of the economy. The impact of changes in the overall balance on a country’s gross borrowing requirements would be stronger the shorter the average maturity of its debt. Careful assessment of the prospects for mobilizing needed financing in the context of prevailing market conditions would be essential, particularly if the additional cost of the UBI was estimated to increase gross borrowing requirements to levels considered risky in light of the country’s level of development and other relevant characteristics (e.g., its historical record of financial stress).5 For low-income countries, this assessment should, of course, take into account the scope for mobilizing additional concessional financing. Assessment of the UBI’s financing prospects should be conducted over a multiyear horizon, particularly if it is to be introduced for an indefinite time period. This assess- ment should take into account the likelihood of changes in market conditions (and, for low-income countries, in aid flows) over that horizon—for example, by using stochastic fan charts (Celasun, Debrun, and Ostry 2007; Ostry et al. 2010). The analysis of financing prospects over the short to medium term should be comple- mented by a careful evaluation of the impact of the UBI’s cost on public debt sustainability over the medium to longer term. Traditionally, public debt has been considered sustain- able if its trajectory under the projected primary balances, cost of the debt, and gross domestic product (GDP) growth rate converges to a stable or declining level. More recently, however, debt sustainability analyses have placed increased emphasis on the level of debt, rather than just its direction of change, as well as on risks to debt projections stemming from macroeconomic and other shocks (such as natural disasters and the realization of contingent liabilities), and/or from opti- FIGURE 5.1  General Government Gross mism of the underlying assumptions. Debt as a Percentage of GDP, 2009–17 On average, public debt levels have 50 risen significantly in both emerging Emerging market economies, average market and low-income countries over 40 the last decade (figure 5.1), approach- ing 50 percent of GDP, although there is Percent 30 Low-income countries, average 20 substantial variation in country perfor- mance in this area (figure 5.2). This trend 10 suggests that the scope to increase debt 0 levels without significant risks for fiscal 2009 2010 2011 2012 2013 2014 2015 2016 2017 sustainability is likely to be severely con- SOURCE: IMF 2018. NOTE: Figure does not include oil producers strained in the majority of countries. Chapter 5. Financing a Universal Basic Income: A Primer 157 Several recent papers by the Inter- FIGURE 5.2  Variations in General national Monetary Fund (IMF) provide a Government Gross Debt, 2017 detailed discussion of the considerations 140 Low-income countries that should guide a debt sustainability Emerging market 120 economies analysis for countries with access to 100 financial markets and for low-income Percent 80 countries (see in particular IMF 2013 60 and 2017b). This assessment should 40 entail comparison of an appropriately 20 0 constructed baseline debt sustainability Avg. High Low Avg. High Low analysis with an alternative that would SOURCE: IMF 2018. include the impact of the proposed UBI NOTE: Figure does not include oil producers. on the primary and overall fiscal bal- ances, as well as on other relevant macroeconomic variables, such as the growth rate and the average cost of the public debt. If the additional cost of the UBI significantly increases the likelihood of the debt becoming unsustainable over the projected period, policy makers should look at scaling back the proposed amount of the benefit and/or exploring complementary revenue mobilization measures or savings in other expendi- tures. Even if debt sustainability analysis does not identify significant risks, the fiscal space for a UBI may be constrained by existing national or supranational numerical fiscal rules—especially if the country belongs to a common currency area such as the euro area. The number of countries that have adopted one or more such rules has grown rapidly in recent decades (Schaechter et al. 2012). Fiscal rules vary widely in coverage (levels of government, inclusion of state- owned enterprises, etc.), base (budget balances, public debt, and/or expenditures), and features (flexibility, enforcement and revision mechanisms, etc.) as well as in their effectiveness (Corbacho and Ter-Minassian 2013; Ter-Minassian 2010). Most “second generation”–type rules strive to ensure a degree of flexibility—for example, through cyclical adjustment or escape clause provisions. Nevertheless, for countries that have deficit or debt levels close to the applicable rules’ limits, the cost of a nonbudget-neutral UBI may turn out to be unaffordable without accompanying revenue increases or cuts in other spending. It should be stressed that, even if the debt sustainability analysis and the exist- ing rules signal availability of fiscal space to accommodate a UBI involving additional fiscal costs, the decision to use the space to finance the UBI, rather than other potential revenue or expenditure policies (e.g., cuts in especially distortive taxes or increases in education, health, or infrastructure spending) should be made by governments based on careful analysis of the growth and distributional implications of the alternative policies and their consistency with societal preferences. 158 Teresa Ter-Minassian Assessing the Scope for Financing a UBI through Expenditure Measures Expenditure Savings This section discusses options to finance the cost of a UBI that cannot be accommo- dated through the use of existing fiscal space wholly or partly through expenditure savings. It is widely recognized that both the level and composition of public expen- ditures vary across countries even more than those of revenues, reflecting a host of economic, historical, sociopolitical, and institutional factors. Notable among such factors are the country’s level of economic development, its demographic trends, political power balances, and societal preferences regarding government’s allocative and redistributive roles. These preferences are in turn influenced by citizen perceptions of the effectiveness and efficiency of government spending. Institutional factors—such as the distribution of spending responsibilities among different levels of government and the quality of budget processes—also play an important role in shaping the level and composition of public spending and the effectiveness of different spending programs. Given the variety of country choices regarding the level and composition of public expenditures, the scope for financing a UBI through savings in other spending is likely to be highly country specific. This section looks at the main considerations that could guide an assessment of such scope. After a brief review of some analytical tools that can be used to assess the level and composition of pre-UBI spending and the cost-effectiveness of its main categories, detailed information is provided on how to assess the scope for savings in two important expenditure areas: subsidies and the public wage bill.6 International benchmarking is a useful tool for initial identification of potential candidates for expenditure savings in a particular country. A country’s performance can be compared with that of appropriately selected other countries in three main dimen- sions:7 •• Levels of chief economic and functional categories of government expenditures relative to GDP •• Relevant indicators of outcomes in each area (standardized test scores and schooling completion rates for education spending, life expectancy and infant and maternal mortality rates for health spending, crime rates for spending on citizen security, etc.) •• Input mix for selected spending categories (public employment and public wage rates, teacher/pupil ratios and school equipment per pupil, hospital beds versus primary clinics, etc.) The benchmarking tools can be more or less sophisticated, ranging from simple scatter diagrams to data envelopment analysis and stochastic frontier estimations. Chapter 5. Financing a Universal Basic Income: A Primer 159 International benchmarking is extensively used in spending reviews, whether comprehensive or sectoral.8 It helps assess whether weaknesses in public spending performance are systemic, affecting most categories of spending, or specific to some only. Systemic underperformances in a country are usually a symptom of serious insti- tutional weaknesses—particularly in the budget process—such as a lack of planning or medium-term budgeting capacities, unrealistic optimism in forecasting, poor control of budget execution, and inadequate accounting and transparency practices. This last effect is the availability and reliability of budgetary information, and consequently the politi- cal accountability of budget policy makers. Careful diagnosis and public recognition of such institutional weaknesses can help mobilize the broad social and political consensus needed to carry out effective reforms in the relevant areas. This benchmarking can also help identify concrete actions to improve the effective- ness and efficiency of specific categories of spending, through both reallocation of funds to better-performing programs, and changes in programs’ input mixes. Governments can promote such changes by requesting spending units (ministries and other agencies) to identify savings (the so-called efficiency dividends) equivalent to a given proportion of their respective budgets, to be used for reallocation to priority or new spending initia- tives—such as a UBI. Of course, effectiveness and efficiency considerations should be complemented by distributional ones. The impact of any proposed substantial reallocations of budgetary spending on income distribution should be analyzed, particularly regarding level and depth of poverty (poverty headcount and gap), using available household survey information. Subsidies Reform Many countries at different levels of development devote substantial budgetary resources to various types of subsidies. Consumer subsidies are more common, and costlier in terms of budgetary resources, than subsidies to producers. However, producer subsidies also can give rise to serious allocative distortions, with longer-term adverse effects on productivity and growth. In countries with significant subsidies, policy makers should carefully assess the scope for eliminating them—or at least for substantially cutting them back—because such measures would yield gains in terms of efficiency and horizontal equity, as well as budgetary savings. The main types of consumer subsidies relate to food, energy, and water usage. The reform of food subsidies has many dimensions of a sociopolitical as well as fiscal nature. While the trend worldwide has been toward replacing generalized price subsidies for some essential foodstuffs with targeted in-kind interventions or with cash transfers, reform paths have varied significantly across countries, reflecting demographic and social characteristics as well as institutional capacities (Alderman, Gentilini, and Yemtsov 2018). This section looks at energy and water subsidies, where the scope for significant budgetary savings, as well as efficiency gains, is clearest. Energy subsidies can be measured on a pre- or post-tax basis. A pretax subsidy is defined as the gap between the supply cost of an energy product and the price paid for 160 Teresa Ter-Minassian it by consumers (households or enterprises). For products that are internationally traded (such as fuels), the supply cost is taken to be their international price plus transportation and distribution costs. For products that are not internationally traded (as is generally the case for electricity), the supply cost includes generation, transmission, and distribu- tion costs. Post-tax subsidies include the pretax ones plus an adjustment for the revenue for- gone by not subjecting the energy products to standard indirect taxes, and to corrective taxes accounting for their adverse externalities (environmental and road damage, acci- dents, and related health costs). The IMF estimates that total post-tax subsidies were equivalent to 6.5 percent of global GDP in 2015 (Coady et al. 2015), with a very wide variance across countries. A recent Organisation for Economic Co-operation and Devel- opment (OECD) study on energy taxes (discussed further in the next section) confirms that only limited progress has been made in recent years in increasing corrective taxa- tion on the uses of energy products in major advanced and emerging market economies. Pretax subsidies (estimated by the IMF to be equivalent in total to 0.4 percent of global GDP in 2015; Coady et al. 2015) vary significantly across countries, as well as over time, reflecting changes in international oil prices and exchange rate developments. While nonexistent or very low (generally under 0.1 percent of GDP) in advanced econo- mies, they are substantial in a number FIGURE 5.3  Selected Countries Where of emerging market and low-income Pretax Energy Subsidies Exceeded countries; in a few of these (mostly 2 Percent of GDP in 2015 oil producers), they are estimated to exceed 10 percent of GDP (figure 5.3). Trinidad &Tobago The fiscal cost of pretax subsidies Venezuela, RB is not always made explicit in govern- Bahrain ment budgets. In numerous countries, Egypt, Arab Rep. national oil companies bear the finan- Iran, Islamic Rep. cial burden of below-supply-cost Lebanon Libya consumer prices; over time, this can Saudi Arabia lead to de-capitalization and underin- Congo, Rep. vestment. Mozambique Energy subsidies have a range of Zambia well-recognized costs in terms of envi- Zimbabwe ronmental and health externalities; Kyrgyz Republic distortions in a country’s productive Ukraine structure (because they incentivize Uzbekistan energy inefficiency by firms and house- Tajikistan holds); balance of payments (through Turkmenistan higher energy imports or low energy Algeria exports); and preemption of fiscal space 0 4 8 12 16 20 that could be used for public investment Percent in infrastructure or education, health, or SOURCE: IMF online energy subsidies database. other priority spending. Chapter 5. Financing a Universal Basic Income: A Primer 161 Distributional effects vary across countries and products. Some energy products (such as kerosene, liquid petroleum gas, fuels used in public transport, and electricity) have a significant direct weight in the consumption basket of the poor. More generally, increases in fuel prices may be passed onto the prices of other goods, such as basic foodstuffs, that also weigh heavily in the consumption of lower-income groups. Empiri- cal studies have shown that, in general, middle- to upper-income recipients receive the bulk of energy subsidies.9 Frequently, this constitutes a major political obstacle to their reform, since middle- to upper-income voters tend to have greater political voice and clout than the poor. A further obstacle is the potential impact of substantial discrete adjustments in energy prices on the rate of inflation, which is a concern of some policy makers. International experiences with energy subsidy reforms suggest the following as important ingredients for success:10 •• Smoothing the impact of the initial price adjustments through an appropriate, preferably preannounced, phase-in •• Linking subsequent adjustments to developments in international prices (con- verted to local currency) through a formula •• Introducing simultaneous compensation mechanisms (e.g., cash transfers) for the most vulnerable groups of the population; the budgetary cost of such com- pensation should be taken into account in the calculations of the fiscal space created by subsidy reform •• Minimizing second-round inflationary impacts through an appropriate mone- tary stance •• Conducting an information campaign to sensitize the public to the benefits of the reform Introduction of a UBI may facilitate reform of energy subsidies, insofar as it would allow a partial compensation of the nonpoor affected by the reform, who are likely to be its most vocal and politically effective opponents. This option seems to be relevant in only a limited number of countries, however. The simulations in chapter 4 suggest that, in most cases, a UBI that could be funded through the elimination of energy subsidies would not significantly reduce poverty. In India, Indonesia, and Mozambique, replacing both subsidies and selected social assis- tance programs would allow financing of an equivalent-benefit UBI—which would still deliver only relatively low poverty impacts. Only in the Russian Federation would the combination of replacing existing programs and subsidies with a UBI eliminate the pov- erty gap. Water subsidies can be defined as the gap between water supply cost, including depreciation and maintenance of existing infrastructure, and the price charged to users by water utilities. Frequently, these utilities are not adequately compensated for this 162 Teresa Ter-Minassian gap by the budgets of the national or relevant local governments—a fact that, as for oil companies, leads to undercapitalization, poor maintenance of infrastructure, and underinvestment, with adverse effects on the population’s access to good-quality water, especially in rural areas. As a result, the true size of water subsidies is often significantly underestimated in general government accounts. Estimates by IMF (2015) point to an overall (on- and off-budget) cost of water sub- sidies on the order of 0.6 percent of global GDP in 2012, but with large variation across countries. Subsidies generally were found to be very limited in advanced economies, but to range between 0.4 and 1.8 percent of GDP in other regions. Seven countries, some of which are already under high water stress, were estimated to provide subsidies in excess of 5 percent of GDP. Water subsidies incentivize overconsumption of an increasingly scarce resource, and therefore have substantial adverse implications for efficiency and sustainable growth. They are also inequitable, as they disproportionately benefit higher-income groups. The above-mentioned IMF study estimates the share of these benefits going to the richest income quintile to be nearly three times that going to the lowest quintile in a sample of low-income countries. The gap was found to be smaller in emerging market economies, but still sizable. Phasing out water subsidies could provide significant savings, which could be used in part to fund compensatory cash transfers to the poor. However, since water supply is frequently a responsibility of local governments, the savings would largely accrue to their budgets, rather than to the national one. This fact makes water subsidy reform a less appropriate candidate to finance a UBI than reform of energy subsidies. Reforms of Public Wages and Employment Wage bills worldwide absorb a large share of government budgets. In 2015 they repre- sented on average an estimated 27 percent of public expenditures in emerging market and low-income countries, and 24.5 percent in advanced ones—albeit with significant variance around these means (IMF 2016). Intercountry differences reflect, in addition to the level of development, structural factors such as societal preferences about public versus private provision of public services and demographic characteristics such as the shares of schoolchildren and older people in the population. Economic and electoral cycles also affect the evolution of government wage bills, as do the institutional factors discussed below. Given their weight in government spending, government wage bills are obvious candidates for scrutiny in any search for expenditure savings. This requires a coun- try-specific analysis of the drivers of the evolution of their two determinants: public employment and the average compensation of employees (wages and fringe benefits). International benchmarking, in line with the caveats noted above, can help identify inef- ficiencies and related options for policy reforms in this area. The shares of government employment within the total working-age population vary significantly across countries (figure 5.4). The average is substantially higher in Chapter 5. Financing a Universal Basic Income: A Primer 163 advanced than in emerging market FIGURE 5.4  General Government and, especially, low-income countries, Employment as a Percentage of the albeit with large variations around the Working-Age Population, 2015 means. This reflects not only the gen- 18 erally higher level of provision of public 16 services in advanced economies, but 14 also the often greater rigidities in legis- 12 Mean = 11.6 lation governing public service in those Percent 10 countries (e.g., civil servant tenure and 8 Mean = 8.3 limitations on geographic or functional 6 n = 33 mobility). Further, there is evidence of 4 n = 33 Mean = 3.7 ratchet effects in the behavior of public 2 n = 11 employment in advanced economies 0 Advanced Emerging Low-income during economic cycles. In contrast, economies market economies developing employment in developing countries 10th–90th percentile range countries appears to be more closely correlated SOURCE: IMF 2016. with electoral cycles. NOTE: Figure does not include oil producers. Available evidence suggests that, on average, government wages tend to be higher than those for comparable FIGURE 5.5  Public Wage Premium as a jobs in the private sector. The premium Percentage of Average Private Wage tends to be higher in developing coun- 14 n = 23 tries than in advanced economies, 12 n = 37 12.8 partly reflecting the more limited job 10 n = 86 11.7 opportunities offered by the private 10.1 Percent 8 sector in those countries (figure 5.5). It 6 n = 26 also tends to fluctuate during economic 4 5.4 and political cycles. Large premiums are 2 often associated with the strong bar- 0 All Advanced Emerging Low-income gaining power of public sector unions, economies market developing economies countries especially in sectors that provide critical public services. SOURCE: IMF 2016. NOTE: Figure does not include oil producers. International experience shows that structural reforms of government wages are fraught with technical, and especially political, difficulties. For this reason, governments under financial constraints—for example, those associated with fiscal adjustment programs—often resort to such blunt instruments as temporary across- the-board hiring and wage freezes. Such policies do not address the root causes of the excessively large wage bills, and therefore are not sustainable over the medium term. They can also give rise to bottlenecks and inefficiencies in the provision of essential public services, such as health and education. For these reasons, they should be avoided; in any case, they cannot be considered for funding permanent spending programs such as a UBI. 164 Teresa Ter-Minassian Depending on the circumstances of the individual country, structural reforms may include the following: •• Changes in the legal regime for the civil service, aiming to increase flexibility in adjusting the workforce to changing needs and budgetary priorities, such as increasing the share of position-based (as opposed to tenured) employment; making it easier to dismiss nonperforming employees; and facilitating geo- graphic and functional redeployment of civil servants, including teachers and health workers •• Instituting transparent systems for comparing compensation of public and pri- vate employees, and using these as a basis for periodic review of civil servant remuneration scales •• Linking public employees’ wage and career progression more closely to individ- ual performance Such reforms are unlikely to yield substantial savings in the short run, but could do so over the medium term, on a more sustainable basis. They could also provide gains in terms of efficiency and equity. Policy makers considering initiating or intensifying such reforms should strive to obtain broad popular support for them by explaining their bene- fits to the population at large. They should also ensure that the various strands of reform policies are well coordinated in substance and timing. Realism about their feasibility, their distributional impact, and the extent and time profile of their expected budgetary payoffs is essential in assessing their suitability as sources of financing for a proposed UBI. Assessing the Scope for Financing a UBI through Revenue-Raising Measures In contrast to the assessment of fiscal space on unchanged policies, which can be based on a broadly accepted standardized framework, an assessment of the scope to finance a UBI (or the additional cost of it) through revenue-raising measures involves a number of country-specific judgments about the level and composition of tax and nontax revenues before and after the proposed measures. The vast literature on these subjects can help inform such judgments by shedding light on the likely macroeconomic, allocative, and distributional effects of alternative revenue mobilization options, as well as on their administrative feasibility. Ultimately, however, a country’s policy makers have the responsibility for deciding on the trade-offs such effects frequently imply—and their decisions are likely to be heavily influenced by political economy considerations. There are no hard-and-fast rules to assess the appropriateness of the level of a coun- try’s overall revenues, but a number of economic, social, and institutional considerations Chapter 5. Financing a Universal Basic Income: A Primer 165 should play a role in such an assessment. Typically, the size of government—and conse- quently the ratio of revenues to GDP—can be expected to rise with the level of economic development (the so-called Wagner’s law), and indeed the revenue/GDP ratio is on aver- age significantly higher in advanced than in emerging market economies and, even more so, low-income countries (figure 5.6). FIGURE 5.6  Average 2005–15 Revenue/GDP Ratios for Selected Country Groups 40 30 Percent 20 10 0 Total revenues Tax revenues Advanced economies Emerging market economies Low-income countries Oil producers SOURCE: IMF online revenues database. The data displayed in the figure reflect both economic factors (notably, the greater complexity of the economy that facilitates tapping different tax bases) and institutional factors (especially, a greater tax administration capacity). It also shows that oil producers tend to have higher revenue/GDP ratios, but lower average tax burdens, than is typical for countries of comparable levels of development. This reflects the substantial nontax revenues (mainly royalties) that these countries receive from foreign and national oil companies, and the associated disincentive to domestic revenue mobilization efforts. Nevertheless, there is substantial variance in the ratios of both total and tax reve- nues to GDP within each group of countries (figure 5.7). These intercountry differences FIGURE 5.7  Variations by Country Group in Average 2005–15 GDP/Revenue Ratios 80 Oil producers 70 Advanced economies Low-income countries 60 Emerging market economies 50 Percent 40 30 20 10 0 Median High Low Median High Low Median High Low Median High Low SOURCE: IMF online revenues database. 166 Teresa Ter-Minassian reflect a country’s stage of development as well as a number of factors, including the following: •• Societal preferences regarding the size of government and the redistributive role of the state •• The structure of its economy (in particular, the relative weight of harder-to-tax sectors, such as agriculture, and small-scale retail and services) •• The degree of informality of its labor and other markets •• Its degree of financial openness (which constrains its scope for taxing mobile financial capital) •• The effectiveness of its tax administration and, more generally, the country’s culture of tax compliance Several of these factors affect not only the trend of revenue ratios, but also their fluctuations around the trend. Policy makers considering additional domestic revenue mobilization efforts to finance a UBI need to carefully consider the role the above-listed factors could play in constraining the effectiveness of their efforts. Not only the level but also the composition of revenues matters, as different reve- nue sources have different economic and distributional effects, ease of administration, and distribution across levels of government, and frequently give rise to different politi- cal economy dynamics. Table 5.1 summarizes how different revenue sources tend to fare in relation to various macroeconomic, allocative efficiency, distributional, administra- tive, and political economy criteria that are likely to be viewed as relevant by tax policy makers. It should be emphasized that the scores represent broad characterizations, TABLE 5.1  Revenue Sources Scored on Different Criteria Administrative Redistributive acceptability Sensitivity to Compliance friendliness potential potential Revenue Visibility Political Growth cycle costs costs Revenue source Personal income taxes V L M/H H H M/H H M/L Corporate income taxes M L H L M/H H M M General consumption taxes H M M L M M L M Excise taxes M/L M M/L L L L L M Property taxes M/L H L M/H H M H L Social security contributions M L M/H L L L M M/L “Green” taxes L M/H L L/M H M H M/L User fees M M/H L L M L/M H M Royalties M/H L H L M M M H NOTE: H = high; M = medium; L = low; V = varying. Chapter 5. Financing a Universal Basic Income: A Primer 167 suggested by theory and international experience, which, however, as discussed further below, may not be necessarily applicable to individual countries’ circumstances. It is clear from the table that no revenue source scores well in relation to all the criteria; policy makers considering different revenue mobilization options face trade-offs among them. These issues are discussed in more detail in the next subsections, with par- ticular emphasis on considerations relevant to emerging market and low-income countries. Taxes on Personal Income A policy maker considering options to finance a UBI would likely focus first and foremost on the personal income tax (PIT), given that PITs typically have a progressive rate struc- ture and therefore, in principle, high redistributive potential. Indeed, the PIT is generally the backbone of the tax system in advanced economies. Revenues from the PIT average the equivalent of 8 percent of GDP in those countries, and account for the largest share of tax revenues in these countries. The PIT also accounts for most of the redistributive impact of tax systems in OECD countries,11 despite the fact that (1) the progressivity of the rate structure has been significantly reduced in recent decades, (2) many countries have moved toward a dual PIT system through more favorable treat- ment of capital income and capital gains, and (3) they maintain a number of deductions that predominantly benefit the higher deciles of the income distribution. In contrast, in emerging market and low-income countries, both the redistribu- tive and the revenue-raising capacities of the PIT tend to be seriously undermined by a number of factors: •• High threshold levels (sometimes multiples of median per capita income), which concentrate the tax burden on a relatively small proportion of the population •• Frequent exemption of capital income and capital gains •• A proliferation of deductions and special treatments (tax expenditures) •• A high degree of labor market informality •• Lower capacities of the tax administration, which enables tax avoidance and outright evasion Some of these factors interact, compounding the respective adverse effects. For instance, checking eligibility for deductions and exemptions absorbs resources of already weak tax administrations, reducing their enforcement capacities. Some deductions and special treatments also have efficiency costs, promoting excessive resort to debt or distorting the allocation of savings. A high degree of informality erodes the tax base, requiring higher rates, which in turn further incentivize informality, with an adverse impact on firm size and productivity. The impact of these factors on PIT revenues is evidenced by figure 5.8, which shows that the average PIT/GDP ratio in emerging market economies is less than one-third (and that 168 Teresa Ter-Minassian FIGURE 5.8  Average 2005–15 PIT/GDP in low-income countries about one-quar- Ratios ter) that in advanced economies. With regard to redistributive Advanced economies power, a recent study of several Latin American countries (Corbacho, Cibils, Emerging market economies and Lora 2013) shows that, while the Kakwani index, which measures the Low-income countries progressivity of the PIT, has relatively high values, the Reynolds-Smolensky Oil producers index, which measures the actual redis- 0 2 4 6 8 10 tribution effected by the tax, has very Percent low or even negative values in those SOURCE: IMF online revenues database. countries. An analysis of the scope for financing a UBI (or part thereof) through increases in the PIT in a particular country needs to start from a diagnostic of the main specific obstacles to the revenue-raising capacity of the tax in that country. This can provide initial guidance in identifying feasible policy options to raise the needed addi- tional revenue from the tax. In general, steps to broaden the tax base are preferable to increases in rates, because they are more efficiency-friendly,12 improve horizontal equity, and facilitate the administration of the tax; but changes in tax schedules may also be appropriate in some circumstances, as detailed below. A comparison of the income threshold for the PIT with the median per capita income can provide insights into the scope for reducing the threshold. An analysis of tax expenditures under the PIT can point to likely candidates for elimination or for scaling back.13 In line with the IMF Manual on Fiscal Transparency’s recommendation to compile estimates of tax expenditures and include them in budget documentation (IMF 2007), many countries do publish such estimates.14 Assessment of the potential revenue gains from repealing some deductions should ideally be accompanied by an analysis of the distributional effects of such a move. How- ever, in many countries, this may be constrained by a lack of sufficiently detailed data on the income distribution of the beneficiaries of tax expenditures. An alternative approach could be to impose a cap on the total amount of deductions a taxpayer can claim, spec- ified as a declining percentage of his or her pretax income. The introduction of such a cap would by design enhance the progressivity of the PIT, compared to the preexisting system. The scope for broadening the tax base to some harder-to-assess incomes such as fringe benefits, homeowners’ imputed rents, and income from self-employment depends on the state of the tax administration in a given country. Presumptive assess- ment methods can be resorted to, but inevitably entail costs in terms of horizontal equity, as taxpayers with different levels of actual income may be subject to the same tax burden. Also, setting up such systems may involve significant administrative costs and delays. Chapter 5. Financing a Universal Basic Income: A Primer 169 The taxation of personal incomes from capital highlights the trade-offs between the objectives of revenue mobilization, redistribution, and economic efficiency. There are several efficiency considerations that argue for a lower tax burden on capital income than on labor income: •• The greater cross-border mobility of capital than labor, which enhances the scope for tax competition among countries seeking to attract foreign capital, although the attractiveness of tax havens is likely to have been reduced by the OECD-sponsored recent advances in automatic exchange of information among tax administrations •• The fact that entrepreneurial income is taxed under the corporate income tax (CIT), and then again under the PIT when distributed as dividends, except in the (not common) case of full integration of the two taxes •• The fact that taxing the inflation-related component of capital income is tanta- mount to taxing the capital stock itself; thus, a lower rate of capital taxation is an approximation to taxing only the real return on capital •• A relatively low tax burden on capital income is also sometimes justified by the objective of promoting domestic savings, although the available empiri- cal evidence on the weight of tax considerations in savings decisions remains inconclusive At the same time, distributional considerations argue against discriminating in favor of capital incomes, since the share of capital in total incomes of taxpayers tends to rise with the level of income. Different countries weigh the two types of considerations differently, but the growing worldwide trend toward dual PITs suggests that the bal- ance is tilted in favor of the efficiency arguments. Nevertheless, countries that currently exempt certain types of capital incomes, or tax them at very low rates, may have scope for mobilizing additional revenue through the introduction of a comprehensive dual PIT, with the same rate for all types of capital income, or through a moderate increase in its rate. Similar considerations apply to the taxation of capital gains.15 In summary, there are a number of steps that can be taken in emerging market and low-income countries to improve the revenue performance (and simultaneously the redistributive power) of the PIT, but in most cases they should realistically be expected to cover only a fraction of the cost of a UBI, if the latter is projected to exceed 1–2 per- cent of GDP. The illustrative microsimulations presented in chapter 4 focus on increases in PIT rates (an across-the-board surcharge and an increase concentrated on the upper decile) for simplicity and cross-country comparability, but country-specific analyses should pref- erably include modeling options to broaden the PIT base. Those simulations suggest that, since in most emerging market and developing countries the lower-income deciles are not taxable under current PIT rate schedules, funding a UBI with increases in the 170 Teresa Ter-Minassian latter would be quite effective in reducing poverty. However, financing a UBI sufficient to eliminate the poverty gap would require increases in the PIT’s burden on the upper deciles too large to be politically feasible in all the countries in the simulation sample, with the possible exception of Russia. Taxes on Corporate Income A second potential revenue source for the UBI is the corporate income tax. Differences in the revenue performance of the CIT across groups of countries are smaller than those in the PIT (figure 5.9). This is the case because many emerging market and low-in- come countries are natural resource revenue producers that can extract significant rents (in the form of profit taxes as well as royalties) from the (predominantly foreign) companies exploiting such resources. Revenues from these taxes are, how- FIGURE 5.9  Average 2005–15 CIT/GDP Ratios ever, highly volatile, reflecting changes in international commodity prices; they Advanced economies are therefore hardly suitable candidates to finance an entitlement program such Emerging market as the UBI that aims to provide a steady economies stream of income to the population.16 Low-income countries CITs have trended downward over the last decades in advanced econo- Oil producers mies, reflecting policy makers’ concerns over their potential adverse impact on 0 1 2 3 4 5 6 Percent investment, and especially international tax competition. The revenue loss has SOURCE: IMF online revenues database. stemmed from both reductions in the CIT rate and the erosion of the tax base as a result of a proliferation of tax incentives, and of tax planning practices such as thin capitalization and international transfer pricing. While recent initiatives promoted by international organizations to reverse the erosion of the CIT base—such as the OECD’s Base Erosion and Profit Shifting (BEPS) initiative17—can be expected to support CIT rev- enues over the medium term, international competition is likely to continue putting downward pressure on CIT rates worldwide for the foreseeable future. In summary, although policy makers should be encouraged to intensify efforts to broaden the CIT base, including by closely scrutinizing the benefits of any incentives under the tax and by implementing the anti-avoidance and evasion measures rec- ommended by international institutions, it would not seem realistic over the short to medium term to rely on increases in the CIT to fund a UBI. Taxes on Wealth It is a well-known fact that wealth is even more unequally distributed than incomes. For instance, in the OECD, the bottom 40 percent is estimated to own only 3 percent Chapter 5. Financing a Universal Basic Income: A Primer 171 of household wealth (in contrast to 20  percent of income), while the top decile owns 50 percent of such wealth (25 percent of income). While comparable data are lacking for non-OECD countries, wealth inequality is unlikely to be smaller on average in those countries. Moreover, there is clear evidence that wealth inequality has been rising in many countries in recent decades. There is thus a strong case for taxing wealth on distributional grounds, and for policy makers seeking revenue sources to fund a UBI to explore options in this area. The case is reinforced by the fact that some taxes on wealth, notably recurrent taxes on immovable properties, have low efficiency costs and are therefore considered relatively growth-friendly (Eyraud 2015; OECD 2010). However, despite these advantages, taxes on property account for relatively low shares of total tax revenues (less than 6 percent in advanced economies, and less than 3 percent in emerging ones). Taxes on a more comprehensive measure of wealth (per- sonal net worth) are even less common and productive in terms of revenue. Only six OECD countries and very few non-OECD ones (four in Latin America) levy such taxes, and only Switzerland obtains significant revenues from them (Benitez and Velayos 2018). The poor revenue performance of property taxes can be explained by a number of factors. •• Difficulty of administration. A well-functioning property tax requires setting up and regularly updating property cadasters, a task that is resource intensive, albeit now facilitated by technological advances such as Google Maps (Bonet, Munoz, and Pineda Mannheim 2014; Norregaard 2015). Valuation of properties is especially difficult in less developed countries, with high degrees of infor- mality (unregistered properties, frequently transacted on a cash basis). The difficulties are further compounded by the fact that property taxes are generally assigned to local authorities, whose administrative capacities, especially in small rural municipalities, are weaker than at the national level. •• Visibility and compliance costs. Typically, recurrent property taxes are paid in annual or semiannual installments, which makes them very visible and can give rise to liquidity constraints. Such constraints may also arise with nonrecurring wealth taxes such as those on gifts and inheritance. •• Taxpayer perceptions. Taxpayers may perceive property valuations as unfair, especially in the absence of objective, market price–based indicators of prop- erty value. These characteristics contribute to make property taxes particularly unpopular with both taxpayers and politicians. For these reasons, and especially given their assignment to local governments, property taxes do not appear to be good candidates for funding a UBI, at least not in the foreseeable future. Given the cross-border mobility of financial wealth, the scope for mobilizing sig- nificant resources through other forms of wealth taxation, such as taxes on net worth or 172 Teresa Ter-Minassian financial assets, would appear to hinge on developing a broad-based international con- sensus on the desirability of such taxes. In the absence of international coordination, which is currently nonexistent, cross-border competition is likely to severely constrain individual countries’ maneuvering room in this area. Taxes on Consumption General taxes on consumption, particularly the value-added tax (VAT), are the workhorse of taxation systems, especially in emerging market and low-income countries. In emerg- ing market economies, their yield (at over 6.5  percent of GDP) is comparable to the average in advanced ones, and accounts for a third of tax revenues. In low-income coun- tries (at 4.8 percent of GDP) it accounts for over 30 percent of tax revenues. Excise taxes also play a significant role in revenue mobilization, accounting on average for about one- tenth of total tax revenues in all three groups of countries (figure 5.10). The broad worldwide reliance on consumption taxes reflects the fact that they are relatively efficiency- and growth-friendly, easier to administer than income or wealth taxes, and generally less visible than other revenue sources. Excises, by being levied on goods that have adverse side effects on health (such as tobacco and alcoholic or soda bev- erages), often serve complementary policy objectives in addition to revenue generation. Consumption-based taxes are, however, inferior to income and wealth taxes from a distributional standpoint. A broad-based single-rate VAT is regressive with respect to current income, because consumption tends to decline as a proportion of income as the latter rises. However, some empirical studies have found it to be proportional or even slightly progressive in relation to consumption and to lifetime incomes. Concern with the potential distributional impact of a broad-based, single-rate VAT has prompted a number of countries to exempt, or subject to reduced rates, items such as some or most foods, and educational and health goods and services. Empirical studies of the distributional effects of such tax expenditures suggest that, while these goods may represent a larger proportion of the consumption of poorer households, the lost revenue disproportionately benefits the middle and upper quintiles of the income distribution FIGURE 5.10  Average 2005–15 Consumption Tax/GDP Ratios 8 6 Percent 4 2 0 Advanced economies Emerging market economies Low-income countries Oil producers VAT Excise tax SOURCE: IMF online revenues database. Chapter 5. Financing a Universal Basic Income: A Primer 173 (Corbacho, Cibils, and Lora 2013; Keen 2015; Lustig 2017). Thus, while these tax expen- ditures (which in some countries are equivalent to a significant fraction of GDP) are progressive, they redistribute in favor of the middle and upper classes. Moreover, exten- sive exemptions and multiple rates substantially complicate VAT administration. The empirical evidence on the distributional impact of excises is mixed, reflecting varying patterns of consumption of the taxed products in different countries. Excises on tobacco are generally found to be regressive, while those on alcohol are regressive in some countries and mildly progressive in others. It is clear from the considerations above that policy makers considering financing a UBI with increases in taxes on consumption would need to carefully weigh their potential advantages and disadvantages. If a country already has a broad-based VAT with a moder- ate single rate, a relatively small increase in that rate could yield significant revenue with limited costs in terms of efficiency and ease of administration and compliance. If the VAT base is eroded by significant exemptions, or if there are several different rates, a reduc- tion of the exemptions or a simplification of the rate structure would yield gains not only in terms of revenue, but also with regard to efficiency and administration. In both cases, the distributional effects of the proposed changes in the VAT base and rate(s) should be assessed, together with those in the UBI, through microsimulations based on available household expenditure surveys and standard assumptions on the pass-through of the changes onto consumer prices.18 A recent paper (Harris et al. 2018) presents some such simulations, which indi- cate that broadening the VAT base and using the revenue to fund a UBI would boost the consumption of the least-well-off households and reduce extreme poverty and inequality—even if only 75 percent of the additional revenue was used for this purpose. Chapter  4 also provides examples of such simulations, which suggest that in some cases the combination of a more generous UBI with a proportional increase in indirect tax rates can reduce poverty and inequality, albeit of course less so than a comparable com- bination of UBI and increased PIT. Other Potential Revenue Sources Environmental Taxes Environmental taxes are another potential source of revenue to finance a UBI. These taxes are more prevalent in advanced than in developing economies (many of which, as discussed in the next section, still subsidize the consumption of pollutants such as fuels) and can take different forms, the most common being excises on fuels used for transport and heating, and on electricity.19 A recent OECD study, covering 42 advanced and large emerging market economies that together account for about 80 percent of global energy use and the related carbon emissions, finds that the coverage and magnitude of energy taxes have been rising in recent years, but on average they remain relatively low from an environmental stand- point, covering only a fraction of the environmental costs of the energy use (OECD 2018). 174 Teresa Ter-Minassian These taxes vary widely, however, across countries, sectors, and products (OECD 2018). They tend to be larger in higher per capita income countries, with some notable exceptions such as the United States. They are typically very low on (highly polluting) coal, which remains the key input for electricity generation in many countries, and higher on fuels used in transportation. Even among the latter, there are variations across products, with diesel tending to be more lightly taxed than gasoline, despite its higher polluting effects. These differences, which distort energy consumption patterns, have tended to be reduced in more recent years. Environmental taxes have obvious benefits in terms of efficiency, since they charge for environmental externalities, and also have significant revenue-raising potential. Espe- cially in their excise form, they are easy to administer and entail low compliance costs. Recent empirical studies (e.g., Flues and Lutz 2015) also do not find evidence of a signif- icant adverse impact on countries’ international competitiveness. The main obstacle to raising environmental taxes to levels closer to covering the environmental costs of energy use has traditionally been concern over their distribu- tional impact. However, such an impact can be expected to vary across products and countries, reflecting varying patterns of energy consumption. Moreover, mitigating steps can be taken to reduce any adverse effects of the taxes on poorer income classes. Recent OECD research (Flues and Thomas 2015; Flues and van Dender 2017) using microsimulation models suggests that, in 21 advanced (mainly European) economies, transport taxes tend to be progressive in terms of consumption in the first half or more of the distribution—which is not surprising, since many poor households do not own a vehicle. The taxes’ progressivity is likely to be even greater in low-income countries, where car ownership tends to be concentrated in the upper income classes. Taxes on heating fuels are generally found to be somewhat regressive, as are taxes on electricity consumption. The latter, however, can be made less regressive by exempting electricity consumption below a minimum threshold. As with other taxes, any adverse distributional effects should be weighed against the distributional benefits of the expenditures they finance. Policy makers considering financing a UBI with the introduction or increase in environmental taxes would need to analyze the combined impact of the two measures on the distribution of incomes and consumption (Parry 2015). If the net effect is distributionally positive—as is likely to be the case especially in emerging market and low-income countries—environmental taxes would seem to have much to recommend them. Nontax Revenues Unfortunately, there appears to be only limited potential to use nontax revenues to finance a UBI. As indicated above, the volatility of royalties from natural resource rev- enues makes them unsuitable to fund a benefit aiming to provide a stable and secure source of income to the population. User fees for public services are recommendable on efficiency grounds, but tend to be highly visible and therefore resisted by citizens—especially when levied for nationally Chapter 5. Financing a Universal Basic Income: A Primer 175 provided services, where their link to the accessibility and quality of the service is fre- quently unclear. For this reason, user fees tend to be more utilized by local governments, which makes them unsuitable to finance a nationwide UBI. In principle, a fairly steady stream of revenues could be obtained through better utili- zation of the commercial assets owned by governments, including state-owned enterprises, real estate, and some natural resources.20 Setting up the required governance structures is, however, a task that is likely to require substantial time and political commitment. Also, many such assets belong to subnational levels of government (e.g., cities). The scope for funding a UBI through this type of revenue is thus likely to be highly country specific. Foreign aid could in theory represent an additional funding source for a UBI in low-income countries. Indeed, some donors might find the prospect of consolidating a number of different aid programs into one supporting a unified cash transfer mech- anism attractive, with lower administrative costs and risks of leakages (e.g., through corruption). Others might find it a difficult political “sell” at home—funding a program that also provides benefits to high-income recipients. In all cases, in the current envi- ronment of budget consolidation in traditional donor countries, it may be difficult for low-income countries to secure significant increases in the total amount of foreign aid they receive. Therefore, foreign aid may not offer much scope to increase the generosity of the social safety net through the introduction of a UBI. Conclusions This chapter emphasizes that decisions about a UBI should be taken in conjunction with decisions about its financing, as alternative financing options can be expected to have quite different macroeconomic, fiscal, and distributional effects that could reinforce or offset those of the UBI. Those effects would reflect a range of economic, demographic, social, and institutional factors that are likely to be highly country specific. Drawing on relevant theoretical literature and international experiences, the chap- ter discusses considerations and criteria that can guide the analysis of the effects of various financing options in individual emerging market or low-income countries. It is crucial to extend this analysis to a sufficiently long time horizon, given the fact that a UBI is intended to provide households a permanent, stable, and predictable source of revenue. Policy makers should resist the likely temptation to use a temporary fiscal space created by easy financing conditions to fund a program that would need to continue if and when those conditions disappeared. The chapter first looked at how to assess available fiscal space to fund a non- budget-neutral UBI on current tax and expenditure policies, taking into account both short-term financing constraints and longer-term fiscal sustainability, and concludes that such an option is unlikely to be viable in most circumstances—especially given the fact that (as suggested by the microsimulations in chapter 4) in many countries the fiscal cost of a nonbudget-neutral UBI that would adequately protect the poor is likely to exceed several points of GDP. 176 Teresa Ter-Minassian Accordingly, the chapter then focused on possible measures to augment the fiscal space for a UBI through savings in other expenditures and increases in revenues. On the expenditure side, two important candidates for savings are subsidies and government wages. Despite reforms in a number of countries to reduce energy subsidies in recent years, they remain significant (in excess of 2  percent of GDP) in several middle- and low-income countries, especially oil producers. They have a range of well-recognized costs in terms of environmental and health externalities, distortions in the productive structure of a country, balance of payments, and preemption of fiscal space that could be used for productive and social spending. Although some fuel products weigh significantly in the consumption basket of the poor, the bulk of energy subsidies benefits middle- and upper-income groups, a fact that complicates the political economy of their reform. The introduction of an appropriately calibrated UBI may facilitate such a reform, insofar as it would allow partial compensa- tion of these groups, in addition to full compensation of the affected poor. The chapter highlights a range of options for reform in employment and wage policies for the civil service that—although unlikely to yield substantial savings in the short run—could do so over the medium term on a more sustainable basis. Realism about their feasibility, their distributional impact, and the extent and time profile of their expected budgetary payoffs would be essential in assessing the suitability of such reforms as sources of financing for a proposed UBI in individual countries. With regard to revenue measures, the chapter’s main takeaways can be summa- rized as follows: •• In assessing the appropriateness and scope of revenue-raising measures to fund a UBI, the focus should be on the impact of any proposed tax package on the progressivity of the entire tax system. •• In general, steps to broaden the tax base should be preferred to increases in rates, because they are more efficiency-friendly, improve horizontal equity, and facilitate tax administration; but changes in tax schedules may also be appropri- ate in some circumstances. •• In terms of the PIT, there is frequently scope for raising revenue in an efficiency- and distribution-friendly way by lowering a too-high exemption threshold, eliminating or reducing deductions and exclusions from the tax base that ben- efit primarily the upper-income classes, and taxing capital income and capital gains at a moderate rate that does not encourage capital flight. This scope is likely to be significantly constrained in many emerging market and low-income countries by the high degrees of informality prevailing in those economies, and by the foreseeable resistance to sizable increases in the tax burden by upper-in- come groups, which wield considerable political clout in those countries. •• The CIT does not currently seem a good source of funding for a UBI, given its Chapter 5. Financing a Universal Basic Income: A Primer 177 volatility, especially in countries dependent on natural resource revenues, and the still high degree of international tax competition. •• Taxes on property and wealth are desirable from a distributional standpoint, but difficult to administer and highly unpopular politically. Therefore, they tend to produce limited revenue, especially in non-advanced countries. Also, they are frequently assigned to subnational governments, a fact that makes them unsuit- able to fund a nationwide UBI. •• Consumption taxes are recommendable on efficiency grounds, and typically have substantial revenue mobilization potential, but they entail distributional costs in many instances. Their increase may be more distribution-friendly if coupled with an increase in the generosity of the social safety net through the introduction of a UBI. •• “Green” taxes are a thus far underutilized instrument that could provide a useful revenue source for a UBI. •• The scope to raise various types of nontax revenues to fund a proposed UBI appears limited for the foreseeable future. In conclusion, financing a nonbudget-neutral UBI in a fiscally sustainable way would represent a substantial, and for low-income countries with high poverty incidence daunting, challenge. Even in countries with lower poverty incidence, and thus a more manageable fiscal cost, sustainable financing of a UBI would require the simultaneous use of several of the policy options discussed in this chapter—in particular both signif- icant savings in other spending programs and increases in income and consumption taxes. This in turn would significantly complicate the political economy of introducing a UBI—the subject of the next chapter—as it would require securing the support of multi- ple groups of stakeholders. Notes 1. Helpful comments by the book’s editors and reviewers, and research assistance by D.P. Kanth, are gratefully acknowledged. 2. Specifically, the Organisation for Economic Co-operation and Development analyzed the impact of the introduction of a variant of the UBI in four European Union countries (OECD 2017). Ensor et al. (2017) analyze the distributional implications of a budget-neutral UBI for the United States. Also, the International Monetary Fund’s October 2017 Fiscal Monitor dis- cusses the results of microsimulations of the fiscal and distributional first-round effects of a UBI calibrated at 25 percent of the median income for eight advanced and middle-income countries (IMF 2017a). 3. The only complexity in this case may be the availability of up-to-date information on the number of residents (or citizens) in the country. And, note as indicated earlier, that estimating 178 Teresa Ter-Minassian second-round effects would require modeling the behavioral responses of households to the additional income, and their macroeconomic and fiscal implications. 4. Regarding debt composition, countries with high shares of variable rate, foreign exchange– denominated, or foreign-held debt would tend to be more vulnerable to changes in domestic interest rates or in risk premiums for their debt. For emerging market economies, the Inter- national Monetary Fund considers a share of the public debt denominated in foreign currency as signaling moderate risk if it is above 20 percent, and high risk if it is above 60 percent. The corresponding ratios for the share of the public debt held by nonresidents are 15 percent and 45 percent, respectively (IMF 2013). 5. The International Monetary Fund considers countries with public gross financing require- ments equivalent to more than 10  percent (emerging market economies) or 15  percent (advanced economies) of gross domestic product as having the potential for being finan- cially stressed. 6. Other chapters of this book discuss considerations in choosing between a UBI and other types of cash transfers to households; see chapter 2. 7. The relevance of such benchmarking hinges crucially on the choice of the comparator coun- tries. These should indeed be comparable to the country in question, not only in terms of level of development, but also in terms of size; geographic, demographic, and social characteris- tics; and extent of expenditure decentralization. 8. For comprehensive discussions of the theory and international practice of spending reviews, see Cangiano, Curristine, and Lazare (2013) and Robinson (2013). 9. Coady, Flamini, and Sears (2015) estimate that, on average in developing countries, the share of energy subsidies captured by the richest quintile of the population is more than six times larger than that accruing to the lowest quintile. 10. Clements et al. (2013) provide a comprehensive analysis of international experience with energy subsidy reforms, including 22 successful—and unsuccessful—case studies. 11. It is estimated that taxes account for about one-quarter of the difference in the Gini coefficients for market and disposable incomes on average in the OECD, with the other three-quarters accounted for by cash transfers (Brys et al. 2016). 12. The deadweight loss entailed by a tax is estimated to rise more than proportionally to the increase in its marginal rate. 13. For a discussion of the challenges in identifying and measuring tax expenditures, see Villela, Lemgruber, and Jorrat (2009). 14. See also Peláez Longinotti (2018) for estimates of tax expenditures in Latin America. The study estimates that total tax expenditures in the region average about 3.5 percent of GDP, of which 0.6 percentage points are accounted for by the PIT. There is substantial variation around these averages. In some countries, tax expenditures under the PIT result in revenue losses in excess of 1 percent of GDP. 15. See Brys et al. (2016) for a fuller discussion of the design of capital income taxes for inclusive growth. 16. For instance, the residents’ resource dividend in Alaska has fluctuated significantly since its inception, and some experiments with UBI-type transfers in natural resource revenue–produc- ing countries like the Islamic Republic of Iran and Mongolia have fallen victim to downturns in the prices of those commodities. Chapter 5. Financing a Universal Basic Income: A Primer 179 17. Extensive documentation on the BEPS initiative can be found on the OECD website (www.oecd.org). 18. The IMF conducted simulations of the combined distributional impact of the introduction of a UBI equivalent to 25 percent of median per capita income and of a revenue-equivalent flat tax on consumption for eight advanced and emerging market economies. It found this impact to be progressive in countries like South Africa, characterized by high degrees of pre-UBI inequality (IMF 2017a). 19. Other types of environmental charges are taxes on the purchase or ownership of motor vehi- cles, and levies on effluent discharges by firms or on urban traffic patterns responsible for congestion. These are mostly levied at the subnational level, and therefore are not suitable candidates for financing a nationwide UBI. 20. Detter and Fölster (2015, 2017) make a strong case for professional management of gov- ernment assets at the national and local levels, and provide a number of relevant country examples in this respect. References Alderman, Harold, Ugo Gentilini, and Ruslan Yemtsov, eds. 2018. The 1.5 Billion People Question: Food, Vouchers or Cash Transfers? Washington, DC: World Bank. http://documents.worldbank. org/curated/en/398281507803030509/pdf/120368-PUB-PUBLIC-ADD-BOX-405295B- DOCDATE-11-11-2017.pdf Benitez, J.C., and F. Velayos. 2018. “Impuestos a la Riqueza o al Patrimonio de las Personas Físicas con especial mención a América Latina y El Caribe.” CIAT Working Paper 1/2018. Inter-Amer- ican Center of Tax Administrations, Panama. Bonet, J., A. Munoz, and C. Pineda Mannheim, eds. 2014. El potencial oculto: Factores determinan- tes y oportunidades del impuesto a la propiedad inmobiliaria en America Latina. Washington, DC: Inter-American Development Bank. Brys, Bert, Sarah Perret, Alastair Thomas, and Pierce O’Reilly. 2016. “Tax Design for Inclusive Economic Growth.” OECD Taxation Working Paper No. 26. Organisation for Economic Co-op- eration and Development, Paris. Cangiano, M., Teresa Curristine, and Michel Lazare. 2013. Public Financial Management and Its Emerging Architecture. Washington, DC: International Monetary Fund. Celasun, Oya, Xavier Debrun, and Jonathan D. Ostry. 2007. “Primary Surplus Behavior and Risks to Fiscal Sustainability in Emerging Market Countries: A ‘Fan Chart’ Approach.” IMF Staff Papers 53 (3): 401–25. https://www.imf.org/External/Pubs/FT/staffp/2006/04/pdf/celasun.pdf Clements, Benedict, David Coady, Stefania Fabrizio, Sanjeev Gupta, Trevor Serge Coleridge Alleyne, and Carlo A. Sdralevich, eds. 2013. Energy Subsidy Reform. Washington, DC: International Monetary Fund. Coady, David, Valentina Flamini, and Louis Sears. 2015. “The Unequal Benefits of Fuel Subsidies Revisited: Evidence for Developing Countries.” IMF Working Paper WP/15/250. International Monetary Fund, Washington, DC. https://www.imf.org/external/pubs/ft/wp/2015/wp15250. pdf Coady, David, Ian Parry, Louis Sears, and Baoping Shang. 2015. “How Large Are Global Energy Subsidies?” IMF Working Paper WP/15/105. International Monetary Fund, Washington, DC. https://www.imf.org/external/pubs/ft/wp/2015/wp15105.pdf 180 Teresa Ter-Minassian Corbacho, Ana, Vicente Fretes Cibils, and Eduardo Lora. 2013. More than Revenue: Taxation as a Development Tool. New York: Palgrave Macmillan. Corbacho, Ana, and Teresa Ter-Minassian. 2013. “Public Financial Management Requirements for Effective Implementation of Fiscal Rules.” In The International Handbook of Public Financial Management, edited by Richard Allen, Richard Hemming, and Barry H. Potter, chapter 2. New York: Palgrave Macmillan. Detter, Dag, and Stefan Fölster. 2015. The Public Wealth of Nations: How Management of Public Assets Can Boost or Bust Economic Growth. London: Palgrave Macmillan. —. 2017. The Public Wealth of Cities: How to Unlock Hidden Assets to Boost Growth and Prosper- ity. Washington, DC: Brookings Institution Press. Ensor, Will, Anderson Fraley, Matt Jensen, and Amy Xu. 2017. “A Budget-Neutral Universal Basic Income.” AEI Economics Working Paper 2017-03. American Enterprise Institute, Washington, DC. https://www.aei.org/publication/a-budget-neutral-universal-basic-income/ Eyraud, Luc. 2015. “The Wealth of Nations: Stylized Facts and Options for Taxation.” In Inequality and Fiscal Policy, edited by Benedict Clements, Ruud de Mooij, Sanjeev Gupta, and Michael Keen, chapter 7. Washington, DC: International Monetary Fund. Flues, Florens, and Benjamin Johannes Lutz. 2015. “Competitiveness Impact of the German Electricity Tax.” OECD Environment Working Paper No. 88. Organisation for Economic Co-op- eration and Development, Paris. http://www.oecd.org/officialdocuments/publicdisplaydocum entpdf/?cote=ENV/WKP(2015)9&docLanguage=En Flues, Florens, and Alastair Thomas. 2015. “The Distributional Effects of Energy Taxes.” OECD Taxation Working Paper No. 23. Organisation for Economic Co-operation and Development, Paris. https://www.oecd-ilibrary.org/docserver/5js1qwkqqrbv-en.pdf?expires=1566366396& id=id&accname=guest&checksum=FFB5F3B60E0D31AD0EEEE1F4191C1F7D Flues, Florens, and Kurt van Dender. 2017. “The Impact of Energy Taxes on the Affordability of Domestic Energy.” OECD Taxation Working Paper No. 30. Organisation for Economic Co-op- eration and Development, Paris. https://www.oecd-ilibrary.org/docserver/08705547-en.pdf? expires=1566366669&id=id&accname=guest&checksum=DEA5C346C88B9F69F23A63 A962B2AD75 Harris, Tom, David Phillips, Ross Warwick, Maya Goldman, Jon Jellema, Karolina Goraus, and Gabriela Inchauste. 2018. “Redistribution via VAT and Cash Transfers: An Assessment in Four Low and Middle Income Countries.” IFS Working Paper W18/11. Institute for Fiscal Studies, London. https://www.ifs.org.uk/uploads/WP201811.pdf IMF (International Monetary Fund). 2007. Manual on Fiscal Transparency. Washington, DC: IMF. https://www.imf.org/external/np/pp/2007/eng/051507m.pdf —. 2013. “Staff Guidance Note for Public Debt Sustainability Analysis in Market-Access Coun- tries.” IMF, Washington, DC. https://www.imf.org/external/np/pp/eng/2013/050913.pdf —. 2015. “Is the Glass Half Empty or Half Full? Issues in Managing Water Challenges and Policy Instruments.” Staff Discussion Note SDN/15/11. International Monetary Fund, Washing- ton, DC. https://www.imf.org/external/pubs/ft/sdn/2015/sdn1511.pdf —. 2016. “Managing Government Compensation and Employment—Institutions, Policies, and Reform Challenges.” Policy Paper. International Monetary Fund, Washington, DC. https:// www.imf.org/external/np/pp/eng/2016/040816a.pdf —. 2017a. Fiscal Monitor: Tackling Inequality. Washington, DC: IMF. https://www.imf.org/en/ Publications/FM/Issues/2017/10/05/fiscal-monitor-october-2017 Chapter 5. Financing a Universal Basic Income: A Primer 181 —. 2017b. “Review of the Debt Sustainability Framework in Low-Income Countries: Proposed Reforms.” Policy Paper. International Monetary Fund, Washington, DC. https://www.imf.org/ en/Publications/Policy-Papers/Issues/2017/10/02/pp082217LIC-DSF —. 2018. Fiscal Monitor: Managing Public Wealth. Washington, DC: IMF. https://www.imf.org/ en/Publications/FM/Issues/2018/10/04/fiscal-monitor-october-2018 Keen, Michael. 2015. “Targeting and Indirect Tax Design.” In Inequality and Fiscal Policy, edited by Benedict Clements, Ruud de Mooij, Sanjeev Gupta, and Michael Keen, chapter 12. Washing- ton, DC: International Monetary Fund. Lustig, Nora. 2017. “Fiscal Policy, Income Redistribution and Poverty Reduction in Low and Middle Income Countries.” Working Paper No. 54. Commitment to Equity Institute, Tulane University, New Orleans. http://www.commitmentoequity.org/wp-content/uploads/2017/08/ CEQ_WP54_Lustig_June2017.pdf Norregaard, John. 2015. “Taxing Immovable Property: Revenue Potential and Implementation Challenges.” In Inequality and Fiscal Policy, edited by Benedict Clements, Ruud de Mooij, San- jeev Gupta, and Michael Keen, chapter 11. Washington, DC: International Monetary Fund. OECD (Organisation for Economic Co-operation and Development). 2010. “Tax Policy Reform and Economic Growth.” Tax Policy Studies No. 20. OECD, Paris. —. 2017. “Basic Income as a Policy Option: Can It Add Up?” Policy Brief on the Future of Work. OECD, Paris. https://www.oecd.org/els/emp/Basic-Income-Policy-Option-2017.pdf —. 2018. Taxing Energy Use 2018. Paris: OECD. Ostry, Jonathan D., Atish R. Ghosh, Jun I. Kim, and Mahvash S. Qureshi. 2010. “Fiscal Space.” IMF Staff Position Note 10/11. International Monetary Fund, Washington, DC. https://www.imf. org/external/pubs/ft/spn/2010/spn1011.pdf Parry, Ian. 2015. “Carbon Tax Burdens on Low-Income Households: A Reason for Delaying Climate Policy?” In Inequality and Fiscal Policy, edited by Benedict Clements, Ruud de Mooij, Sanjeev Gupta, and Michael Keen, chapter 13. Washington, DC: International Monetary Fund. Peláez Longinotti, Fernando. 2018. “Overview of Tax Expenditures in Latin America: Main Statis- tics of the CIAT Database.” Inter-American Center of Tax Administrations, Panama. https:// www.ciat.org/Biblioteca/DocumentosdeTrabajo/2018/WP_05_2018_pelaez.pdf Robinson, Marc. 2013. “Spending Review.” Paper prepared for the 34th Annual Meeting of OECD Senior Budget Officials. GOV/PGC/SBO (2013)6. Organisation for Economic Co-operation and Development, Paris. Schaechter, Andrea, Tidiane Kinda, Nina Budina, and Anke Weber. 2012. “Fiscal Rules in Response to the Crisis—Toward the ‘Next-Generation’ Rules. A New Dataset.” IMF Working Paper WP/12/187. International Monetary Fund, Washington, DC. https://www.imf.org/external/ pubs/ft/wp/2012/wp12187.pdf Ter-Minassian, Teresa. 2010. “Preconditions for a Successful Introduction of Structural Fiscal Bal- ance-Based Rules in Latin America and the Caribbean: A Framework Paper.” IDB Discussion Paper No. 157. Inter-American Development Bank, Washington, DC. Villela, L., A. Lemgruber, and M. Jorrat. 2010. “Tax Expenditure Budgets.” IDB Working Paper No. 179. Inter-American Development Bank, Washington, DC. The Political Economy of Universal CHAPTER 6 Basic Income Jurgen De Wispelaere and Ruslan Yemtsov T his chapter discusses the political feasibility of a universal basic income (UBI) from a broad political economy perspective. In recent years, political interest in UBI has grown exponentially, with local, regional, and national decision makers initiating UBI experiments.1 Elsewhere, governments or political actors have com- missioned high-level reports examining the case for UBI.2 Chapter 1 provides details of these developments in countries as diverse as Finland, India, the Islamic Republic of Iran, Kenya, Mongolia, and the United States. Political decisions are typically grounded in economic interests. The net gains or losses from the introduction of UBI will depend on its financing and the performance of programs it replaces. Financing a meaningful level of UBI requires mobilizing additional funds; how progressive the tax regime is determines the distribution of gains and losses from the UBI.3 If a larger share of taxes is paid by the rich, the absolute majority of the population is likely to gain from introduction of a UBI. In a democratic system where voters follow their interests, one can predict that policy makers who institute a UBI are likely to enjoy increased political support; such a mechanism is called a “feedback loop.” However, this simple view does not fully reflect reality. There is a growing body of economic theory that seeks to predict (using economic modeling tools) political decision making on policy choices (Commander 2012; Khemani 2017; World Bank 2008, 2016, 2017). Political economy analysis seeks to identify political factors that lead to instituting and implementing a given policy. Such factors may relate to perceptions and values, the institutional structure of the decision-making process, the political power or influence of possible winners and losers, the credibility of proposed policy changes, and the effective- ness of communication. The specific challenge for a UBI is that determining a technically 183 184 Jurgen De Wispelaere and Ruslan Yemtsov sound UBI proposal needs engagement with political philosophy and public preferences. Political institutions play a role in shaping societal preferences—that is, whether society believes it is in everyone’s interest to institute a particular policy (Khemani 2017). Chapter 4’s policy simulations adopt a particular view as to the technical sound- ness of a UBI proposal. It uses effects on poverty and income distribution (progressivity of combined effects of transfers and financing) in making comparisons with the current system as the criterion of whether a particular version of a UBI is better or worse than the current system. This principle helps in discriminating among various alternatives. But citizens need not share the goals of using redistribution via a UBI to reduce poverty and inequality (Khemani 2017). In the absence of a shared understanding of the goals for redistribution, there is a real risk that any such policy would be captured by the prevailing political elites regardless of its technical soundness, or that sound proposals would be modified to serve narrow political objectives. This chapter examines these factors and risks by simultaneously looking at demand- and supply-side determinants of politics. We start with an exploration of the foundations of demand for a policy by introducing the concept of constituency, which is rooted in both economic interests and attitudinal factors. In the policy process, constituencies are represented by coalitions of decision makers responsible for instituting policies. In the policy process model we adopt here, the political feasibility of a policy proposal (a UBI being one such proposal) depends on cementing its constituency (producing a robust policy demand); this demand can lead to policy change if there is an enacting robust coalition among decision makers willing and able to act on this demand (supply). The enactment produces sustainable policy change if there is capacity for selecting the most appropriate form and pathway for implementation. This simple framework mimics the actual process of policy debate, adoption, and implementation (De Wispelaere and Mar- tinelli 2017), similar to the operational view of the political economy developed by Fritz, Levy, and Ort (2014).4 One of the main features of the political process is path dependency. The literature on the dynamics of policy development (e.g., Mahoney and Thelen 2015; Pierson 2000; Steinmo 2010) suggests that political choices are limited by decisions made in the past. Such path dependency hinges both on the consolidation of vested interests wanting the continuation of policies from which they benefit, and the formation of preferences and beliefs (reflected in political ideologies) that motivate actions. Path dependency influ- ences all social policy proposals, but it is clearly a major obstacle for the UBI, as it departs from instituted social protection policies in a radical way. Another factor influencing the politics of social protection are attitudes toward who deserves to receive support from the state. “Deservingness” in social policy influences what society considers fair (van Oorschot 2000, 2006). A UBI goes against beliefs that social aid should be based on reciprocity, and that support is conditional on social action by a recipient that makes him or her “earn” the right for such support. The idea of a UBI, with its principle of unconditionality, means that assistance is “unearned.” These views were expressed in a 1969 address to the nation by then–U.S. president Richard Nixon: Chapter 6. The Political Economy of Universal Basic Income 185 Under the guaranteed income proposal, everyone would be assured a minimum income, regardless of how much he was capable of earning, regardless of what his need was, regardless of whether or not he was willing to work. Now, during the presidential campaign last year, I opposed such a plan. I oppose it now and I will continue to oppose it, and this is the reason: A guaranteed income would undermine the incentives to work…a guaranteed income establishes rights without any responsibilities. There is no reason why one person should be taxed so that another can choose to live idly. (quoted in Caputo 2012, 269) The majority of policy makers nowadays agree. By and large—but with some nota- ble exceptions discussed in this chapter—major parties have not incorporated UBI into their political platforms. The reaction against “free money” is strong across both devel- oping and developed countries, and is independent of self-identification on a left-right spectrum.5 However, the challenges faced by traditional social policy instruments of risk management keeps resurrecting the idea of radical reform toward redistribution, includ- ing through UBI. Hence, the political feasibility of UBI cannot be ruled out and needs to be examined carefully. The chapter is based on models of the political process depicting democratic political regimes. Many developing countries have different polities, often labeled “autocracies.” Closer inspection shows that the political process in an autocracy is not diametrically opposite that in a democracy (Acemoglu and Robinson 2001; Commander 2012), and we offer in conclusion some considerations on the political economy of UBI in low- and middle-income countries. A Stylized Model of Political Economy This section outlines a stylized model for exploring the political economy of UBI drawing from De Wispelaere and Martinelli (2017). The model consists of three main compo- nents and their interrelations: constituencies, capabilities, and coalitions (figure 6.1). A constituency is a concept most closely associated with economic interests, or the extent to which a group is a net beneficiary of a public policy. However, a constituency is distinct from the concept of a beneficiary (De Wispelaere 2015). A beneficiary is a policy category indicating those who are directly benefiting from a policy being instituted (e.g., net recipients of a UBI).6 The extent to which individuals benefit from a UBI will depend on design parameters such as the level of the UBI, the funding mechanism, and—cru- cially—what implications a UBI has on the existence of, or eligibility for, other social programs. Chapter 4 shows that both the size of benefits and their incidence vary greatly between income groups depending on the economic and social context. A constituency, by contrast, is a political category that is both narrower and broader than the set of beneficiaries. It is narrower because not all beneficiaries are will- ing and able to expend political capital to pressure decision makers, creating an effective demand for a policy that will trigger benefits (or costs). It is broader because not all who 186 Jurgen De Wispelaere and Ruslan Yemtsov FIGURE 6.1  Three Main Components of the Model and Basic Income Variations COALITIONS •• Preferences, risk profiles, •• Fiscal constraints normative values of voters •• Administrative capability •• Potential beneficiaries and •• Existing policies losers from proposed reforms •• Institutional legacies and their power resources •• Ideological cleavages Political agency CONSTITUENCIES CAPABILITIES BASIC INCOME Demand for UBI Supply of UBI •• Objectives •• Design •• Implementation may support a policy proposal in the relevant way are net recipients; some who would be net contributors to a proposed transfer scheme may end up supporting it, perhaps for reasons of solidarity or social justice. Key determinants for effective demand include, among others, voter preferences and normative values, the number of potential benefi- ciaries and losers from proposed reforms and their power resources, and the prospects of forming meaningful coalitions of support and opposition. The demand side is not only directly related to objective forces, but is also influ- enced by expectations—for example, the strength of prevailing perceptions of labor market dysfunction and the failure of existing welfare provisions to provide adequate income security. Different groups of actors may view the policy problem in very different terms, and this will color the extent to which support and opposition coalesce (see De Wispelaere 2015; De Wispelaere and Martinelli 2017). Coalitions designate the political agents involved in the process to legislate for and implement a policy (or oppose its adoption and implementation). Typically, this will involve the governing party or coalition of parties. Support for a given policy proposal may differ considerably within a ruling coalition of parties or even within a single party, which can affect both the aggregate level of support at any given time and its robustness over time.7 It depends crucially on the beliefs of policy makers about how other politi- cal actors are behaving, as Khemani (2017) explains. Furthermore, implementation may Chapter 6. The Political Economy of Universal Basic Income 187 require additional political support.8 Finally, the governing coalition—even with a consol- idated position on a policy proposal—may need support from key stakeholders such as trade unions, business organizations, and various civil society associations. Coalitions are responsive to two different sets of political feasibility factors: one set related to the demand for a policy proposal (or constituencies), the other set related to the capacity to implement or supply it. Generally speaking, the supply side relates to policies that political agents can feasibly offer to address voter demands, taking into account constraints arising from fiscal capacities, bureaucratic capabilities, sunk costs of previous policies, and institutional legacies attributable to the vested interests that these cultivate (see Pierson 2003, 2004). Policy feedback effects can reinforce an existing tra- jectory, leading to stability (see Bonoli and Palier 2000; Mettler and SoRelle 2014). While this path dependency does not preclude significant reform, it limits the extent to which we observe dramatic shifts in policy and explains why welfare reforms are usually of a piecemeal nature. The demand and supply determinants mutually influence each other. Structural features in labor markets and social protection regimes facilitate or disrupt the creation of robust constituencies demanding social policy reform. Similarly, constituencies will put pressure to change. Coalitions in turn respond to both demand- and supply-side fac- tors, while simultaneously trying to influence each in an attempt to further their own goals (figure 6.1). Few considerations are needed to apply this model beyond typical democracies. In an autocracy, political voices are more likely to be distorted because of narrow political participation, low expectations, limited information on government policies, patronage, and the salient role of noneconomic issues such as ethnicity or religion (Roemer 1998; van de Walle 2014). Thus policy makers seem to have greater freedom of choice. But the social contract in such settings is mostly founded on intragroup solidarity rather than on the government-led provision of benefits and services. Support for the deserving is predominantly provided through private solidarity networks shaped by kinship (Hill and Verwimp 2017), making redistribution through UBI almost alien. Because they face the threat of popular uprisings or divisions in ruling coalitions, autocratic regimes also have incentives to secure support from and stability for the majority of the population, including through extending social transfer programs (Lavers and Hickey 2016). Nondemocratic polities tend to be those with high inequality in both income and wealth distributions. To lower risks of upheaval or social turmoil, such gov- ernments may choose to offer citizens some element of redistribution (Acemoglu and Robinson 2001). The general model of the political process presented here is therefore also appli- cable to autocratic regimes, with two possible modifications: feedback loops will be less complex or weaker than in democratic systems, and the role of policy champions/lead- ers/epistemic communities with genuine access to the governing institution will be a key variable. The particular relationship between constituencies, capabilities, and coalitions influences the political feasibility of a specific policy proposal for a UBI (figure 6.2). To 188 Jurgen De Wispelaere and Ruslan Yemtsov FIGURE 6.2  The Demand-Capacity Paradox High Demand Intermittent/weak support Strong support for basic income for basic income D EM A N D -C A Low Capacity High Capacity PA C IT Y PA RA D O X No support Intermittent/weak support for basic income for basic income Low Demand begin with, where strong demands for UBI combine with robust capabilities for imple- mentation, we should expect coalitions to enact legislation and move on to practical rollout of the policy (upper right quadrant of the figure). In this context, coalitions would have good reasons to accommodate the demand for UBI and little reason to resist such a policy shift. Purely ideological objections to UBI would come at a political and electoral cost. There is no country so far that would squarely fit into such a description. A plausi- ble context for this scenario is where windfall gains (e.g., from natural resources) could be universally distributed to all citizens or long-term residents in the form of a social dividend (see Cummine 2016). As the case of the Alaska Permanent Fund Dividend illus- trates, such social dividends create a strong demand, are relatively easy to administer, and meet with comparatively less resistance—as in principle, everyone is a winner (Wid- erquist and Howard 2012a, 2012b). The demand is likely to be stronger in cases where resource rents are combined with minimal social protection (hence vested interest in continuing current policies or allocating additional resources to them is weak), although implementation problems due to a lack of infrastructure need to be overcome, as a short- lived UBI experience in Mongolia, described in chapter 1, suggests.9 Conversely, in a scenario where little demand exists and major policy or institu- tional barriers impede its implementation, we would expect to see little political interest in instituting a UBI (lower left quadrant of figure 6.2). In this scenario, there may exist disparate calls for UBI by advocacy groups, but a genuine window of opportunity is not Chapter 6. The Political Economy of Universal Basic Income 189 likely to open. Under these circumstances, most political agents would view moving toward the implementation of a UBI as an exercise in futility—producing no real bene- fits—or even prohibitively costly in political terms. Low demand and limited capacity for UBI is a typical situation in most develop- ing countries. The demand is low because of extremely limited fiscal resources and many claims from other priorities, taking policy makers’ attention away from UBI pro- posals and reducing public interest in generous universal social transfers. Capacity is low because existing administrative systems have limited outreach or cannot mobilize suffi- cient fiscal resources to make UBI meaningful. In such contexts, calls for UBI arise from time to time from international (e.g., recently by UN Secretary-General António Guterres; see Yamamori 2018) or local propo- nents. There are experiments that are often financed by donors (mostly private), driven by politics in their home countries, rather than local demands.10 Advocates among policy makers in developing countries may use this idea as part of their electoral campaigns, but are seldom willing to assemble the three parts of the political process to implement such proposals on a nationwide scale. For example, the experiment on UBI in Namibia was operating in a squatter community for two years with positive results that were widely publicized (CPI 2016). However, even after a leading advocate for UBI in the country was appointed by the newly elected president in 2015 as a minister of poverty eradication and social welfare, the original plan for scaling up UBI was swept away and replaced by a program intended to alleviate poverty through economic growth. This program, the Harambee Prosper- ity Plan, focuses on graduation approaches through productive inclusion, and includes the creation of a food bank and grants for young people conditional on participation in this food bank and other community activities (which is in opposition to UBI principles) (Coelho 2016). Periodic media reports that the ministry is still developing plans for a UBI covering specific demographic groups are not commented on by officials (Bott 2017). In South Asia, following nongovernmental organization activity in neighboring India piloting various versions of a UBI, Sri Lanka’s advocates for a UBI have launched a number of experimental schemes, along with an advocacy campaign targeted to mem- bers of the Parliament and Cabinet. In a public statement, Sri Lanka’s minister of finance assured that the government would study the idea of unconditional UBI in depth (Rice 2018). The deputy director general of the Department of Planning of the Sri Lankan gov- ernment also expressed interest in the scheme and requested academics to evaluate its results. No further political commitments were offered. The UBI experiment in Kenya described in appendix A is supported by an interna- tional nongovernmental organization, GiveDirectly, and has a sizable reach (over 20,000 recipients over 12 years).11 So far, however, country officials have not commented on the pilot. Instead, the government is stepping up its efforts to expand social pensions—pos- sibly the only politically viable alternative to the UBI and a competing program in terms of fiscal resources (Douillard 2017). Moving to a different diagonal of figure 6.2, we see that many developed and some middle-income countries experience stronger and more pronounced demands for UBI, 190 Jurgen De Wispelaere and Ruslan Yemtsov but at the same time have weak experience and capacity to implement it (upper left quadrant of the figure). Developed countries have stronger capacity to implement a UBI, but the demand for it as a redistributive policy is limited by preferences, beliefs, and ide- ologies, all reflected in institutions (lower right quadrant). At the extremes, strong demand for UBI is met with rigid structural barriers (or lack of capacity)—or conversely, high capabilities to institute a UBI run against low demand (De Wispelaere and Martinelli 2017). These situations are depicted as a demand-capac- ity paradox, a term coined by Parolin and Siöland (2018). The scenario of high demand but low capacity is represented in southern European welfare states (see Noguera 2019 on Spain). Some transition economies of Eastern Europe have recently begun debating a UBI, including Bulgaria, Hungary, Kazakhstan, and the Russian Federation.12 The idea has achieved some traction in Latin America (especially in Brazil and Mexico, as docu- mented in Caputo 2012), and has become prominent in India (as detailed in chapter 1 of this book). The opposite scenario of high capacity but low demand is exemplified in the developed high-income countries, especially northern European states. The idea of a UBI entered the policy-making arena some time ago in Australia, Belgium, Canada, Finland, France, the Netherlands, Switzerland, and the United States, but failed to gain any trac- tion in terms of real policy reform (Caputo 2012). The fiscal and institutional capacity to deliver a UBI especially exists in the Nordic countries, but their populations and policy makers are generally skeptical about its virtues; moreover, a sufficient degree of redistri- bution already exists that makes a UBI largely redundant. Select cases illustrate how, in the presence of the demand-capacity paradox, a UBI features in policy debates. For example, in 2017–18, the Russian media vehemently debated the idea of a UBI following an opinion poll in which 62 percent of respondents declared themselves to be in favor of it, suggesting some demand exists. There are small- scale privately financed experiments in Moscow reflecting an interest in the idea in some circles. One of the leading economic and social policy think tanks in Russia organized an international conference on UBI in 2018 with prominent policy advisors expressing their views.13 All in all, the majority of experts expressed strong reservations about the desir- ability of a UBI in Russia’s current political and governance context. The reluctance of Russian academics to support a UBI as a policy idea may seem surprising given the inefficiency of the country’s existing social assistance system; as illustrated in chapter 4 of this book, Russia is an exceptional case, where UBI can lead to improved welfare outcomes for the poor compared to the status quo. Reasons given for this reluctance include a lack of trust in institutions to implement a UBI, suggest- ing low capacity as the limiting factor. Another factor is strong path dependency: the public may entertain an idea of a UBI in a certain context, but opinion polls in Russian regions suggest very strong support for the existing merit-based system of redistribution and opposition to a pro-poor shift (Yemtsov et al. 2019). Disillusionment with egalitar- ian forms of social policy, inherited from communism, tarnish UBI proposals to provide everyone with the same transfer. In all of these scenarios, a prudent coalition would offer only a weak endorsement for UBI. Such weak endorsement can take the form of offering to experiment with the Chapter 6. The Political Economy of Universal Basic Income 191 policy without making a commitment to actual policy implementation; this is discussed later in this chapter. Constituencies, Ideologies, and Coalitions The political feasibility of a UBI depends on the existence of a significant constituency that produces a strong political demand for the policy. As already discussed, constitu- ency is ultimately built on net gains from a policy proposal compared to the status quo. Such gains primarily take an economic form. Chapter 4 shows that for many examples of country contexts in UBI simulation and financing scenarios, there are a considerable number of losers.14 The size of their losses depends on the incidence of transfers the UBI is replacing and on the taxes. In Brazil, Chile, India, Kazakhstan, Russia, or South Africa, at least 20 percent of the population will be worse off if some of the current social assistance transfers are replaced with a UBI. When the UBI is set at a level higher than current transfers and financed by taxes, it is mostly the upper deciles that are the losers. Their losses are sufficient to be a mobilizing factor in actively opposing the introduction of such policies. But there are also groups that will benefit from the UBI, and those can span the poor (especially those currently excluded from social welfare) and the middle classes. Hence, building a robust constituency in support of a UBI is pos- sible in principle. The next question to be addressed in assessing the scale of a potential constituency for a UBI proposal is assessment of attitudes to the idea. Several national and interna- tional polls collect information on attitudes toward various social policy proposals, and the concept of a UBI has recently been included in such polls. The results show that the public across countries is rather favorable to the idea, and that this support is increasing. Today, 48  percent of Americans support it, according to a new Northeastern Univer- sity/Gallup survey of more than 3,000 U.S. adults (Nova 2018). In Canada, a recent poll shows that 57 percent of respondents support a minimal UBI proposal.15 In the United Kingdom, 49 percent of respondents supported the introduction of a UBI calibrated to minimum needs; this support drops to 30  percent, however, when financing through taxes is considered (Ipsos MORI 2017). In 2016, Dalia Research conducted a large survey of 28 European countries, with 64 percent of the respondents adopting a favorable view about a UBI proposal, and 24 percent opposing it (Holmes 2017; for further discussion, see Van Parijs and Vanderborght 2017). The latest wave of the European Social Survey (ESS) also included a question about support for UBI, also with overall favorable responses: 47  percent in favor; 8  percent strongly in favor (see Fitzgerald 2017; Vlandas 2018). There was significant variation across countries surveyed, with Norway, Sweden, and Switzerland being least support- ive and Eastern Europe most in favor.16 UBI support in all countries exhibits a strong age distribution: the young are most favorably inclined, while those at retirement age and the working-age population are less enthusiastic. There is a significant income gradi- ent, with those at the lower end of the income distribution favoring a UBI compared to 192 Jurgen De Wispelaere and Ruslan Yemtsov those higher up on the income scale. In addition, there is important variation by source of income, with those receiving unemployment benefits (65 percent) and social bene- fits (66 percent) expressing the most support, and those with income from investments and/or savings expressing the least (37 percent).17 Labor market position is also strongly associated with UBI support. Specifically, in the ESS, those who have experienced unem- ployment in the last five years (65 percent) are vastly more supportive of UBI than those who have not (56 percent).18 Among those in employment, the ESS finds mild variation depending on current employment status (employee versus self-employed) and type of employment contract (permanent versus temporary). Strong support from those who are currently unemployed, or have experienced unemployment recently, is perhaps unsurprising.19 The unemployed have everything to gain from a UBI scheme that offers an economic floor without delay or gaps in payment, provides long-term income security, and avoids stigmatizing conditionality. But in most countries, the unemployed are a relatively small and heterogeneous category without much political clout, and form a weak basis for building a UBI constituency. A potential expanded constituency might also include those in precarious and nonstandard employ- ment, often referred to as the “precariat.”20 Precarious employment captures a variety of nonstandard contracts, differing extensively in terms of pay or working conditions as well as—importantly—access to social benefits (see Kalleberg 2018). Those working on temporary contracts have different needs and views than those with part-time contracts; these are expressed in the levels of support for a UBI among different groups of precar- ious workers.21 In terms of building a robust political constituency demanding UBI, the inclusion of different groups of precarious workers will depend on what policy alterna- tives are available to promote the job or economic security of, for instance, temporary workers compared to part-time workers.22 In the UBI debate, labor market risk is closely associated with the increasing automation of routinized labor (Reich 2015; Stern 2016; Walker 2016); this involves high-skilled outsiders unable to find permanent employment (Häusermann, Kurer, and Schwander 2015; Wren and Rehm 2014). This avenue may be less promising than UBI advocates suggest. A study of attitudes toward unconditionality among various groups of labor market outsiders confirms a strong gap between support for UBI for those who face labor market disadvantage compared to those who face labor market risk. Chrisp and De Wispelaere (2017) use 2008 European Value Studies data to examine attitudes toward conditionality among a cluster of outsiders in 18 European countries. In their study, unemployment was strongly associated with an aversion toward conditionality, but this association is absent with those in one of the at-risk categories. This finding suggests increased labor market risk as such does not translate into stronger support for UBI, in contrast with the actual experience of labor market disadvantage. In the UBI debate, the three different labor market positions outlined above are assumed to have a strong preference for UBI. However, insights from the ESS and related research do not bear this out. Heterogeneity within and across these labor market groups makes it difficult to assume that they will converge around a clearly demarcated UBI pro- posal. Chapter 6. The Political Economy of Universal Basic Income 193 Further, realistic differentiated UBI models based on different levels of UBI, vari- ation in tax rates, and whether UBI replaces other social support schemes show that even in developed countries demand for UBI drops significantly (Ipsos MORI 2017; Pulkka 2018). And expanding the search for a UBI constituency to include the middle classes runs into a paradox. Social history and comparative political economy research has demonstrated that the middle classes have a critical role to play in siding with the working classes to install social protection programs and increase social spending (Bald- win 1990; Korpi and Palme 1998). But over time, the middle classes tend to capture a progressively larger share of social spending by supporting contribution-based and earnings-related social insurance programs over means-tested and tax-financed social assistance—which is contrary to the underlying financing logic of UBI (Desai and Kharas 2017). This “Matthew effect”23 in social protection bodes ill for the prospects of building an expanded UBI constituency including the middle classes.24 The challenge of identifying a constituency for UBI in low- and middle-income countries appears even more daunting. Polls in developing countries that assess atti- tudes toward redistribution via transfers are very rare, but when they are done, they show a much lower level of support than in developed countries. A recent example from Indian states shows support for cash transfers as opposed to almost any alterna- tive public policy is notoriously low (Khemani, Habyarimana, and Nooruddin 2019). As shown by Bossuroy and Coudouel (2018), many societal groups across the African conti- nent express deeply entrenched prejudices against redistribution by the state and share a fear of the dependency and laziness any unearned income is perceived to generate. UBI experiments in Kenya and Namibia have so far not led to any clear political constit- uency for UBI in these contexts. Advocates of UBI often depict themselves as humanists embracing universal values radically different from the world of actual politics.25 However, for a UBI to become a real- ity, it must enter the world of realpolitik and be mapped to existing political forces and ideologies (De Wispelaere 2015). Advocates of UBI attack many aspects of social transfers—their targeting, condi- tionalities, work requirements, limited generosity, time-bound nature—all of which are antithetical to the idea of a UBI. They seldom analyze the origins of these design fea- tures, which most of the time are introduced to respond to political considerations. This chapter offers a more impartial perspective on the UBI as a specific form of social pro- tection instrument, and hence is primarily focused on politics of implementation and not on ideological underpinnings. However, in order to understand demand for UBI, it is necessary to see how different ideologies react to a UBI. Mapping the UBI on ideologies is tricky—in fact, any ideological faction has both proponents and adversaries of UBI and there is not a single ideology that can be counted as the “main” defender. Nevertheless, some clear patterns have started to emerge (see Torry 2016). Right-wing parties (pro-market “liberals” or religious “conservatives”) in general have negative attitudes to redistribution. They also see “conditional” forms of social transfer programs as desirable (see, e.g., Lindert and Vincensini 2010). More liberal 194 Jurgen De Wispelaere and Ruslan Yemtsov political ideologies have a more positive attitude to the UBI. This divide typically emerges across countries and over time. For example, in the most recent Gallup poll in the United States, 65 percent of those supporting Democrats want to see a UBI. In comparison, just 28 percent of Republicans support a UBI. Despite a preference for narrow redistribution, UBI is becoming acceptable to con- servative constituencies as a response to new threats to the traditional world of work and family. Recently, several proposals for a UBI have originated among neo-liberals or libertarians as a response to automation and increased unemployment, and these pro- posals are coupled with the dismantling of the existing welfare state. For example, in the United States, Charles Murray (2016) advocates for a UBI to be financed by getting rid of Social Security, Medicare, Medicaid, food stamps, Supplemental Security Income, housing subsidies, welfare for single women, and every other kind of welfare or social service program, as well as agricultural subsidies and corporate welfare. Another strand of support in favor of a UBI among right-wing parties is as a response to political pres- sure to distribute natural resource rents or to overcome popular opposition to powerful industrial interests.26 Social democrats or representatives of organized labor have rather skeptical or mixed views on the UBI.27 On the one hand, they see UBI as development of a “universal- ist” welfare state (which they support), closing the last remaining loopholes that prevent full inclusion in solidarity arrangements. On the other hand, as cogently expressed by Birnbaum (2012), social democratic ideologies are strongly rooted in the value of work and contributory principles. Trade unions also see threats from a UBI to their political influence, stemming primarily from “opting out” because of the guaranteed income offered by a UBI and its potential to exert downward pressure on wages. These parties are concerned about a broader dismantling of time-tested welfare programs, institutions, and principles. The liberal/conservative proposals for a UBI exacerbate this worry. In a way, social democratic and left-wing parties see a UBI as a Trojan horse threatening the existing welfare state, which they see as their main achievement. However, once modified and adapted to serve the specific purposes of easing job transitions or accumulation of human capital, UBI becomes an acceptable policy propo- sition for this bloc. In some circumstances, socialists can even rally around the idea of a UBI (as happened in the last presidential campaign in France28), building on the dissatis- faction of voters with the current state of welfare regimes. The most consistent support for the UBI comes from the green parties. The ecol- ogist view of the objective of UBI is to allow a transition to new societal values that demand reduced consumption and economic activity levels, and a greater role of the state. This school of thought, which is often referred to as postproductivist, connects UBI to green movements. The increasing political influence of green parties across Europe is providing the main stimulus for the UBI as a policy project. Going further to the left, Marxist scholars and political figures have a contradictory attitude toward the UBI. On the one hand, it is seen as a promise for the liberation of workers from their need to sell their labor (Wright 2010). Van der Veen and Van Parijs in 1986 published a classic article on UBI as the “capitalist road to communism,” and many Chapter 6. The Political Economy of Universal Basic Income 195 leftist ideologists continue to express this view (Goodman 2017). On the other hand, UBI is regarded within this tradition as a highly suspicious plot of the oligarchy to buy social peace and continue to exploit humanity, creating a fake sense of “social justice on the cheap” (e.g., Piketty 2016). Hence, depending on the context and other political issues at stake, they can take different attitudes toward the UBI—from supporting it to criticizing it. Extreme right and nationalist parties do not have deeply rooted ideologies based on theories of welfare. Rather, their opportunistic aim to amass more power and influ- ence and radicalize the political discourse can feed on controversies around the UBI,29 especially if it is being connected to citizenship issues. Political ideologies on the UBI and their changes over time have been well studied in the case of Finland (box 6.1). The long history of debates around a UBI in that coun- try produces many useful lessons in understanding the political factors driving changes in reform coalitions. The political evolution of the UBI debate in Finland shows an interesting phenome- non: while left-wing parties are historically the driving force to push UBI on the agenda, right-wing parties are the king makers. Left-wing political actors (including the center-left Greens) may, in order to build a UBI coalition, pragmatically adopt a UBI model that is moderately conservative and appeals to the political center. As Gibson and Goodin (1999) have put it, the key for success in such heterogeneous coalitions is to adopt deci- sions behind a “veil of vagueness,” where support remains broad. Across-party-lines support for a UBI idea may lead to a coalition to institute new policies. Box 6.2 discusses the introduction of universal child allowances in the United Kingdom to show how support across the political spectrum can be mobilized even when there are persistent differences in viewpoints regarding the objectives of a pro- posed policy change. It is evident from this overview that all parts of the political spectrum and ideolo- gies may become favorable to the UBI idea or its modification, adapting it for specific objectives. This is apparent in the European opinion polls, which suggest the ideological position of an individual respondent does not explain individual-level support for UBI (Fitzgerald 2017; Vlandas 2018). This broad appeal across the spectrum of political views offers some interesting parallels to other ideas. An example of a social protection idea having such broad appeal to different parts of the political spectrum (with different objectives) is the history of conditional cash transfers. As documented by Lindert and Vincensini (2010), the conditional cash trans- fer in Brazil (Bolsa Família) appealed to both sides of the political spectrum. This broad constituency made a coalition of support sufficiently stable to enable legislation and successful implementation and scale-up, with positive feedback loops, rewarding policy makers advancing the scale-up. Similar dynamics have been documented recently in poverty targeting programs in Africa. For example, in Kenya and Zambia, pressure from members of parliament was instrumental in expanding small-scale pilot initiatives (Pruce and Hickey 2017).30 Political shock related to subsidy reform may also change the prospects for building a pro-UBI coalition in a developing country. The bulk of the political economy literature 196 Jurgen De Wispelaere and Ruslan Yemtsov BOX 6.1  Mapping Ideological Attitudes on Basic Income across the Political Spectrum in Finland I n a recent study, Stirton et al. (2017) model what they call “the latent disposi- tion” toward basic income among the main Finnish political parties as expressed in political data such as manifestos and parliamentary debates over eight elec- tion cycles (1979–2015), and drawing on the results of a 2015 poll of all 200 elected parliamentarians. Figure B6.1.1 charts each party’s support for basic income, with the Left Alliance, the Greens, and the Centre Party in favor, and the other parties skeptical to varying degrees; historical analysis confirms the point (Halmetoja, De Wispelaere, and Perkiö 2018). FIGURE B6.1.1  Political Party Support for Basic Income in Finland, 1979–2015 Centre Party Christian Democrats 5.5 0 −5.5 Greens Left Alliance 5.5 Ideal point estimate 0 −5.5 National Coalition Party Social Democratic Party 5.5 0 −5.5 Swedish People’s Party True Finns 5.5 0 −5.5 1990 2000 2010 1990 2000 2010 NOTE: The vertical axis shows aggregate level of support for a UBI of each party relative to each other, across eight election cycles. The division is robust over time, with the degree of favorable disposition or opposi- tion polarizing across parties rather than converging. For example, the Greens are becoming more in favor of this idea after being positive initially; Christian Dem- ocrats and Social Democrats are becoming less predisposed after initially being mildly skeptical. The Centre Party is a surprising exception, with its support weak- ening just as the country enters a basic income experimentation period. It is worth noting that the 2015–19 government coalition, responsible for initiating the basic income experiment, contains only one party in favor (the Centre Party (continued) Chapter 6. The Political Economy of Universal Basic Income 197 BOX 6.1  Mapping Ideological Attitudes on Basic Income across the Political Spectrum in Finland (continued) of Prime Minister Juha Sipilä), with the two other coalition partners (the True Finns and the National Coalition Party) resisting the basic income idea. The presence of mixed support within a political coalition explains both the limited scope of the basic income experiment—which is focused almost entirely on labor market participation rates among unemployed recipients of basic unemployment secu- rity—and the current policy shift away from unconditional income support toward activation. Ideological orientation (left versus right) is strongly associated with political parties’ support for basic income (Stirton et al. 2017). At the party level, we clearly find that ideological position matters, but it does so in a complex and paradoxical manner, as figure B6.1.2 shows. Scoring each political party for each election cycle based on ideological orien- tation, we find that, overall, left-wing political parties are more likely to adopt a position in favor of basic income. Paradoxically they are more likely to support it when their party program has shifted to the right. Right-wing opposition also reduces when the party shifts to the right. FIGURE B6.1.2  Support for Basic Income and Left-Right Ideological Position Centre Party Christian Democrats 2 0 −2 Disposition toward basic income Greens Left Alliance 2 0 −2 National Coalition Party Social Democratic Party 2 0 −2 Swedish People’s Party True Finns 2 0 −2 −50 −25 0 25 −50 −25 0 25 Comparative Manifesto Project left/right score NOTE: As in figure B6.1.1, the vertical axis shows relative party aggregate support or opposition for a UBI; the horizontal axis shows the party’s overall policy orientation, from conservative to socialist. = CMP • score over time; — = association between UBI support and policy orientation for all parties; — = associa- tion between UBI support and policy orientation for each party. 198 Jurgen De Wispelaere and Ruslan Yemtsov on developing countries covers the experiences of subsidy reforms (see Inchauste and Victor 2017). The political economy of the UBI is intricately linked to subsidy reform, because removing subsidies is often considered a source for UBI financing. As Com- mander (2012) notes, compensating large swathes of the population for losses incurred in removing subsidies has become a doxology in policy advice for such reforms. Chap- ter 1 describes the case of the Islamic Republic of Iran, with its almost universal cash transfer to compensate for food and energy subsidy removal (see Tabatabai 2012 for an in-depth political economy analysis of the Iranian case). Jordan has adopted compen- sation schemes for 80  percent of its population while removing petroleum subsidies (Atamanov, Jellema, and Serajuddin 2015). In all those cases, empirical analysis demon- strates that the poor and middle class are better off after the reform. However, compensatory mechanisms in subsidy reforms are very far from UBI principles. Near-universal cash transfers used as a vehicle to overcome public opposition to subsidy removal lack a genuine constituency to ensure sustainability. The architects of subsidy reforms call for “temporary” compensations, intended to overcome transitional effects in adjusting to the new relative prices. They do not carry the connotations of per- manent new engagement between the state and citizens—and more often than not tend to become extremely politicized and abandoned with changes in ruling coalitions. UBI may thus become a topic for discussion, or even an element of the political discourse, in the context of subsidy reform, but its sustainability is highly problematic. Constituency around a novel idea of UBI can be built, even in the context of mid- dle-income countries. But it is important to remember a fundamental purpose of such a policy idea: it is a means, not a goal. It is crucial to assess whether the political process that such constituencies will motivate will lead to implementation of a UBI that will be conducive to poverty reduction and shared prosperity—or not. Political Transformations of the UBI Over the long run, the success of a policy idea can be measured with several clear parameters (Marsh and McConnell 2010). The most basic measure is a firmly legislated UBI in the statute books. But legislative presence is a necessary but insufficient condition for success. Three additional measures allow assessment of a new policy idea to have a lasting impact: effectiveness, robustness, and resilience. A new social program is effective when it achieves its stated objectives.31 In the specific case of UBI, effectiveness also relates to the absence of negative effects, such as a large-scale reduction in labor market participation, unsustainable fiscal burdens, an increase in inflation, or pressure on minimum wage levels. The effectiveness of a UBI can be assessed in various ways, including through a social experiment or pilots, or sim- ulations and evidence reviews as done in this book.32 Robustness and resilience are criteria that describe the political stability of a new program over time (resilience),33 or the ability of a policy to deliver on its intended goals under a variety of changing circumstances (robustness). Robustness is challenged by Chapter 6. The Political Economy of Universal Basic Income 199 BOX 6.2  The Politics Surrounding the Introduction of Universal Child Allowances in the United Kingdom T he policy process that led to the introduction of family allowances in the United Kingdom in 1945 is a good example of how a multiparty coalition with diverging interests can emerge to support a specific policy instrument. The initial idea came from a persistent campaign, begun with a pamphlet in 1911, by Eleanore Rathborne, a women’s and children’s rights political advocate. Evidence was building in sup- port of the claim that under current labor market conditions, families with many children remained in poverty. It also became clear that poverty traps were inher- ent in poverty benefits design: an unemployed worker receiving benefits would see a drop in family income by accepting minimum wages. It was also observed that tax child allowances were highly regressive, benefiting only the well-to-do paying taxes above the eligibility limit. Several policy concerns converged after the end of World War II to increase sup- port for universal allowances: more women had become members of trade unions during the war years; there were concerns about falling fertility and the reduced health status of young men unfit for military service, reflecting poverty and poor nutrition; and finally there were concerns about inflation due to the abolishment of price and wage controls and a fear of spiraling prices. As advocated by John May- nard Keynes, family allowances were meant to act as a brake on excessive wage demands by trade unions. By 1945, when William Beveridge advanced his policy proposal for the new wel- fare state, the motives for providing direct support to all families with children were recognized. The proposal prescribed universal child allowances paid to mothers in accordance with the number of children, and not dependent on means testing or work conditionalities. Interestingly, political opposition to the proposal came from both Conservatives and trade unions, motivated by fundamentally the same concern: that paying the child allowances to mothers (as advocated by Rathborne) would shift the family power balance and undermine men’s dominance. Conservative opposition to family allowances was a factor in the party’s defeat in the 1945 elections, and the Family Allowances Act was soon passed by Parliament. An interesting aspect of the process is that there was little debate about the objec- tives of the new transfers, a conscious decision on Rathborne’s part. This allowed the proposal to appeal to a broad swath of members of Parliament, as each could find something to like in the allowance (poverty relief, suppression of wage infla- tion, reduction of disincentives for work, or a simple desire to appeal to voters). SOURCE: Torry 2012. 200 Jurgen De Wispelaere and Ruslan Yemtsov policy drift, the (deliberate) failure to update a particular policy in line with changing cir- cumstances (Hacker 2004, 2005). The Islamic Republic of Iran’s subsidy reform program from 2014 onward can be viewed as a type of policy drift, with the government having continued energy price reform while refusing to increase the real value of compensatory cash transfers accordingly (Salehi-Isfahani 2014). The Finnish experiment, which the Sipilä government refused to expand and extend when it came to an end in December 2018, even before the results of its evaluation had been published, is another example of a deliberate refusal to update.34 Another process that challenges robustness is conversion, whereby an exist- ing policy is redirected toward new goals or purposes to fit the interests of new actors (Streeck and Thelen 2005). A UBI scheme initially aimed at addressing poverty can sub- sequently be repurposed as a labor market activation policy simply by making sure that the surrounding social support architecture retains a strict monitoring and employment incentives regime.35 Resilience refers to the ability of a policy to resist direct pressure for program change, resulting either in its radical abolishment/replacement or in gradual changes that significantly alter key characteristics of the policy.36 In 2010, in the first weeks of coming to power, the newly elected British coalition government canceled the Child Trust Fund scheme that paid out a stakeholder grant to each newborn U.K. national in a special sav- ings account accessible upon reaching adulthood—despite the fact that this was viewed as an important social achievement.37 More recently, in June 2018, the new provincial government of Ontario suddenly decided to cancel the ongoing UBI trial, despite its having indicated the contrary during the elections. Another major concern is presented by the mechanism of layering—a process whereby new rules are grafted onto an existing policy, leading to policy change over time (Mahoney and Thelen 2010; Streeck and Thelen 2005). Postconflict countries offer examples of how pro-poor programs once introduced are marginalized by pressures to maintain and even expand entitlements for veterans and victims of war, depleting polit- ical and fiscal oxygen for pro-poor transfers (see, e.g., World Bank 2008 on Bosnia and Herzegovina). Another example of layering discussed in the literature is the complex dynamics of competition between cash-based targeted social safety nets and an in-kind rice distribution scheme in Indonesia (McCarthy and Sumarto 2018). Given the lack of political feasibility of introducing UBI in a policy big bang moment, the proponents of this idea offer some variations to radical reform with three more “feasible” trajectories. These, however, as we will show, face difficulties in leading to a sustainable policy change. The first trajectory is to institute a partial UBI at a low level and build up incre- mentally toward a more generous UBI.38 Introducing a UBI at a low level minimizes the financial burdens. Such a partial UBI is more likely to fit the institutional structure of countries that have already put in place basic social protection schemes.39 This scenario merely requires universalizing the most basic layer of social protection—retaining addi- tional social policies not replaced by the UBI but in some cases reducing the payment in proportion to the UBI (Van Parijs and Vanderborght 2017). Chapter 6. The Political Economy of Universal Basic Income 201 This pathway faces a number of risks. First, the complex layer of social protec- tion schemes—some of which will be abolished entirely; others retained partially, but with reduced payouts; still others kept unchanged after instituting a partial UBI—car- ries the risk that some vulnerable individuals and households will lose out.40 This effect would be exacerbated if the partial UBI replaces policies that serve a gateway function.41 Second, the interaction between a partial UBI and other support policies that remain means tested or conditional is likely to hollow out many of the effects expected from a UBI. Conditionality dominates unconditionality where individuals make use of other forms of public support to supplement their partial UBI. This leads to a third problem: if introducing a partial UBI does not produce the anticipated effects, or these do not mate- rialize early enough, both the size and the stability of the relevant constituency—and, subsequently, an enacting coalition—could be seriously undermined (see Groot 2004 for a discussion of this point). Finally, even if a partial UBI at a low level is properly imple- mented, ensuring that the level increases over time is far from automatic—particularly once we take into account competition for social spending from other programs (Desai and Kharas 2017). Gourevitch and Stanczyk (2018) have examined this challenge. Meaningful UBI requires heavily redistributing wealth from those at the apex of the income distribution toward the rest of society (requiring a dramatic increase in the incidence of taxes on the rich, as discussed in chapter 4, which one can expect will be resisted at all costs). The reality of politics in many countries suggests that the rich have disproportional influence on the political process. Hence, the UBI promise is illusory. Calnitsky (2017) responds to this rather pessimistic view by defending the achievability of UBI, passing from a modest—but still emancipatory—program to a more ambitious one. He notes some pos- itive political feedback effects that may sustain progress toward a more meaningful UBI. The debate continues, but so far both countries that have introduced some version of UBI—the Islamic Republic of Iran and Mongolia—have seen its real value collapse, rather than increase, over time. A second proposal envisages a cognate version of UBI that compromises on uncon- ditionality by introducing a soft participation requirement.42 This pathway tries to inoculate UBI from concerns about providing income support without any reciprocal condition. The dynamics of this trajectory are to first identify a series of appropriate activities that are broader than labor market participation or education but that still elicit wide public support as a form of social participation.43 The aim of such a modifica- tion is to overcome opposition to UBI, and foster consistency with many existing social assistance programs built with an “activation” perspective. In a critical second step, the proponents of this pathway maintain that bureaucratic hurdles related to the monitor- ing of an expanding range of social activities in practice will rapidly be perceived as too burdensome and costly, paving the way for moving toward a fully unconditional UBI (see Van Parijs and Vanderborght 2017). This two-step path to a UBI through participation income is hard to implement polit- ically. The first stage requires agreement on an operational set of criteria for determining when an individual meets the social participation requirement, which is a nontrivial 202 Jurgen De Wispelaere and Ruslan Yemtsov challenge.44 Even if the first stage is implemented as planned, it remains uncertain whether subsequent policy development involves relaxing conditions and embracing a fully unconditional UBI—or instead reverting back to earlier conditions. A third trajectory opts for implementing a UBI sequentially for different target populations.45 This strategy proceeds with a number of easily identifiable steps, each progressing toward more universal coverage by including new social groups. In a typical scenario, the starting point would be by instituting a universal child benefit—or expand- ing an existing child benefit scheme by abolishing any means testing—or, moving to the other end of the life cycle, instituting a basic pension to be granted to all citizens upon reaching retirement age. It is assumed that gradually the program will gain popularity and be extended to other family members—for example, integrating students, the dis- abled, the self-employed, and so on. Again, several difficulties may arise. First, the assumption that distinct target popu- lations can be neatly divided into discrete programs that can be transformed into a UBI is not likely to hold across the whole population. Conversely, programs that cover distinct target populations vary considerably—for example, in payment levels—which makes it more difficult to turn each into a uniform UBI. Second, the problem with sequential approaches is that they create significant path dependencies (see Pierson 2003, 2004). From an economic perspective, pensioners would worry about a reduction of their ben- efit when shared more widely, either by reducing the actual universal benefit or by increased taxation. In addition, a deservingness norm is likely to block expanding the benefit to less-deserving recipients. The sequential approach assumes a smooth transi- tion from more- to less-deserving target populations, which is unrealistic and contradicts actual experience in implementing such programs (e.g., child allowances).46 A factor leading to the lack of resilience of UBI proposals is that many expressions of political support for the UBI are “cheap”: they come with no real political commit- ment, which means they are both easy to express and an unreliable indicator of current or future willingness to expend political capital in support of UBI. A good example is political actors endorsing UBI when in the opposition, only to fail to act on their earlier endorsement once in government.47 A UBI coalition is highly instable when much of the support for a UBI is cheap.48 The option to pilot or experiment with UBI on a small scale offers a unique political solution: it avoids the pitfalls of comprehensive policy commitment and allows engage- ment with the idea while pushing the real decision further along the time horizon. Political agents conducting UBI experiments can claim credit for engaging with a policy innovation while avoiding the risk of being held responsible for a policy fiasco.49 UBI experiments also afford decision makers considerable leeway in terms of deciding how much resources (time, finances, administrative support, and political capital) to commit to such a venture: this partly explains the considerable variation in experimental design across ongoing and planned pilots. Importantly, support for a UBI experiment does not in any way imply commitment to instituting a UBI. In Namibia, years of experiment- ing with a UBI—and even the presence of strong advocates in the government—have not resulted in any action in its scale-up. In Finland, the government is already moving Chapter 6. The Political Economy of Universal Basic Income 203 toward implementing a new set of social security reforms, which changes the political context within which the UBI experiment was conducted, with its first evaluation results published in early 2019.50 Small pilot schemes do not generate systematic opposition, because those opposed do not think it is worthwhile to spend political resources; as soon as pilots are used to change the policy agenda, that calculation changes and real opposition materializes. A variant of this mechanism is found in the Swiss referendum; the initiative to vote on a UBI proposal rapidly gained traction, while the actual referendum mobilized opposition that did not exist before. Support for UBI, as for many other popular ideas, displays a bandwagon effect (see Nadeau, Cloutier, and Guay 1993). As media attention balloons, so does policy attention. And, as some local, regional, or national governments start taking UBI more seriously, decision makers are pulled (and at times pushed) into taking a closer look and to either emulate or at least learn from what is going on elsewhere (De Wispelaere 2016a). Opponents of UBI need not spend any political resources on resisting UBI proposals until these have entered the policy or political process. UBI can be opposed largely because of its association with a particular political party, organization, or platform. In this sce- nario, toxic support from one faction prevents UBI from garnering support among factions that might otherwise have considered supporting it. A good example of such a dynamic is Finland. The traditional supporters of UBI—the Green League and the Left Alliance—were both in opposition at the time when a ruling coalition initiated a UBI experiment and had to criticize a proposal they had historically campaigned for (see box 6.1).51 Conclusions This chapter has discussed the political economy of UBI. The combination of variation in demand for a UBI and variation in capability to implement a UBI requires a detailed analysis of the policy and institutional characteristics of a country before deciding on the feasibility of specific UBI models. The demand-capacity paradox—that is, a UBI is feasible where it is not really demanded and needed; and where it is needed, it is not feasible—reminds us that where a UBI is feasible and not demanded, there may be a good reason to not have a UBI at all. What matters is whether a basic income can positively influence the political dynamics of redistribution and whether its introduction can lead to better social welfare outcomes. This chapter points out that even if the answer is “yes,” politics introduce an addi- tional layer of complexity that could make a welfare-enhancing UBI infeasible. Different political constituencies can potentially support only one form of UBI, reducing common support and the chances of implementing a fully fledged UBI in any setting. Only the unemployed and those in precarious employment seem to emerge as the most stable constituency for a UBI idea, and they represent a minority among voters. Across polit- ical parties, left-wing programmatic ideology and ecological platforms support the UBI idea. While left-wing and green parties are historically the driving force to push UBI on 204 Jurgen De Wispelaere and Ruslan Yemtsov the agenda, in the actual political process around UBI, the right-wing parties are the king makers. As a result, we observe across different contexts and countries that UBI pro- posals become conservative rather than revolutionary to appeal to the political center: a modest level of support, links to the conditionalities and activation requirements of the existing welfare state, citizenship requirements, and so on. In today’s highly fragmented political context, UBI is poised to be increasingly used for propagandist and political campaigns with little commitment, resulting in a prolifera- tion of pilots and schemes presented as a UBI, but deviating from it in essence. Moreover, when combined with weak capacity to implement redistributive policies where basic public goods such as law and order, property rights, and public health are not being deliv- ered, a UBI may not be advisable. The risk in such a setting is that a UBI would be used to buy votes and fuel the politics of vote buying at the continued expense of public goods. Then what is the reason for outsiders pushing for it? It would seem these reasons should be communicated to citizens and political leaders alike, so political institutions in the country can then debate the reasons and decide whether and how to use the means of a UBI for their goals. In countries that lack the capacity to provide their population with meaningful social protection, the UBI idea can garner demand and be transformative, but its actual emergence hinges on a unique circumstance and resource endowments that could help build capacity while maintaining commitment. Having at their disposal political econ- omy analysis tools that assess perceptions, constituencies, coalitions, and their dynamics can help policy makers and policy analysts make more informed choices when such cir- cumstances arise. To assess the political feasibility of a UBI, policy advisers can consider the main elements discussed in this chapter. First, it is essential to have a good assessment of the economic consequences of various UBI proposals (including taxation) and their likely effects on poverty and inequality. This analysis would reveal whether a UBI is a techni- cally sound economic policy for a country—and if so, what is the rationale for it, and what is the evidence. Second, it is important to know how well the public understands the proposals, and what the attitude is toward various forms of UBI. Such opinion polls, especially when overlaid with other political participation data, serve as an important reality check. Finally, it is necessary to understand the ideologies of the main politi- cal parties and groups and to collect as much information as possible on the primary arguments for and against the UBI to use to motivate these stakeholders (through press analysis or more systematic data analysis as presented in this chapter). Political institutions determine the extent to which any technically sound UBI policy is adopted. The political economy problem of the UBI is the same as for any other “reform”—to understand the obstacles that are causing political institutions to not allow technically sound policies to be pursued, and based on this understanding, to provide ideas to reform leaders on how to overcome these obstacles. Even where UBI as a policy tool remains largely outside of feasible social action in both developed and developing countries, its presence in the debate is likely to have positive side effects, including in sharpening the focus on inclusion and leaving no one Chapter 6. The Political Economy of Universal Basic Income 205 behind in existing social protection programs. Policy analysts need to be careful to pre- vent proposals for cash transfers to occupy the debate space at the expense of building the institutions needed for economies to grow. Notes 1. Experiments in the Netherlands (several municipalities) and Spain (Barcelona) are ongoing, while a pilot in Ontario was suddenly canceled in July 2018. Meanwhile, preparations by local authorities in Edinburgh, Fife, Glasgow, and North Ayrshire, with the collaboration of the Scot- tish government, are quite advanced; and policy makers in several other countries, including Italy and Portugal, are exploring options. Private, nongovernmental agents have also initiated or are operating basic income experiments, such as GiveDirectly in Kenya or Ycombinator in California. Such nongovernmental pilot studies can feed into the policy process at a later stage, but are distinct from those initiated by or run in close collaboration with government agencies. 2. Examples include France, Mexico, and the United Kingdom among developed countries and India, Kazakhstan, Namibia, and Sri Lanka among middle-income countries. 3. Most models to date (e.g., Atkinson 1996; Van Parijs 1995) have combined a UBI with a flat tax, splitting the distribution in the middle and producing a comparable number of net win- ners and losers. Recent work described by Francese and Prady (2018) shows that under a progressive tax regime, all but the top three deciles are net winners in UBI scenarios simulated with data from Organisation for Economic Co-operation and Development countries. Chap- ters 3 and 5 of this book present detailed discussions of financing options. 4. This approach is developed further by Khemani (2017), who uses a problem-driven approach to coalition building and implementation by pulling out three distinct strands of explanations of political constraints to undertaking reforms: (1) explanations centered around problems of credible commitment to a policy change by decision makers, (2) around problems of norms or beliefs, and (3) around preferences for public goods in the broad sense of determining public attitudes toward reform proposals. 5. See Bossuroy and Coudouel (2018) for a thorough review of the generally hostile political attitude of ruling parties in Africa to the idea of even targeted safety nets, let alone uncon- ditional universal grants. Caputo (2012) documents a history of debates around the UBI in Australia, Canada, Germany, Ireland, Japan, the Netherlands, Spain, the United Kingdom, and the United States. Among these countries, none has come close to a UBI as a serious policy proposal, but it has become part of the policy debate everywhere. In Mexico, basic income has taken on the form of a universal basic pension, which was successfully implemented; but expanding it to other groups is out of the question. Brazil has followed a similar path. Only in Finland and subnational entities such as Catalonia, Ontario, Quebec, and Scotland (at the local level) has a UBI entered the realm of real policy making. See Chrisp (2017) for more on left-right attitudes. For a recent study emphasizing how basic income produces political con- flict along class lines, see Gourevitch and Stanczyk (2018). 6. In most cases, UBI is a tax and transfer policy: net recipients have an income below a thresh- old point such that their tax liability does not contribute to basic income, while those above this threshold contribute to funding the basic income scheme. In this scenario, the UBI is a redistributive scheme: even when all are recipients of a basic income, they are not all benefi- ciaries. See Van Parijs and Vanderborght (2017) for further discussion. 206 Jurgen De Wispelaere and Ruslan Yemtsov 7. This is the case with the Finnish government in rolling out the first nationwide basic income experiment, where only one of three coalition partners favors basic income (De Wispelaere, Halmetoja, and Pulkka 2018; Stirton et al. 2017). 8. The Finnish case is again illustrative: the two main advocates of basic income—the Green League and the Left Alliance—were in opposition (De Wispelaere, Halmetoja, and Pulkka 2018; Stirton et al. 2017). 9. It is worth noting that the level of the social dividend is constrained by the size of the sover- eign wealth fund and its performance. Note also that the share of a sovereign wealth fund paid out as a social dividend remains a political decision. The Alaska Permanent Fund, for instance, pays out most of its returns in the form of a (regressive) tax rebate and only a small part in the form of a universal dividend. 10. Private philanthropists who want to make their giving easier, as in the success of GiveDirectly (https://www.givedirectly.org/basic-income). 11. Source: GiveDirectly website, https://www.givedirectly.org/basic-income. 12. See, for instance, Coelho (2018) on Hungary. As discussed in chapter 4, Kazakhstan and Russia are characterized both by the significant coverage of their social assistance schemes and their institutional complexity, reflecting higher-than-average capacity. 13. See https://isp.hse.ru/en/announcements/227307109.html. 14. Only in cases where the financing comes from windfall revenue (such as a natural resource rent) can the UBI be presented as a policy that benefits everyone. 15. Source: Angus Reid Institute, http://angusreid.org/wp-content/uploads/2016/08/2016.08.10_ Basic_Income_PressReleaseTables.pdf. Note that the different surveys cited in this section adopt different definitions of basic income, including variable levels of detail which will affect the comparability of the results. Introducing less detail runs the risk that respondents fail to fully appreciate the distinctiveness of the basic income proposal (Pulkka 2018). But introduc- ing too many details primes respondents to focus on very specific models. 16. Interestingly, with the exception of Finland, support for basic income in the Nordic countries is comparatively weak. The Finnish exception could be explained by the media attention associated with the basic income experiment. Another explanation is that in countries like Norway and Sweden, the existing welfare state functions well enough to depress any demand for basic income. Andersson and Kangas (2005), comparing basic income support in Finland versus Sweden, find support for this hypothesis; also see Lee (2018). 17. Vlandas (2018). Those receiving income from self-employment and farming reported com- paratively low levels of support, which is an interesting finding given that those two groups are often singled out as key beneficiaries in concrete proposals at both the national and Euro- pean Union–wide level. See, e.g., Van Parijs and Vanderborght (2017) on the EU dividend. 18. See Vlandas (2018). A recent survey in Finland confirms this finding; the unemployed are the most supportive group, compared to those in full or part-time employment and those outside of the labor market (Pulkka 2018). 19. Recent unemployment is widely considered to have a scarring impact that affects future employment opportunities; see Gangl (2004, 2006). 20. Standing (2011) employs this term to indicate a very broad and heterogeneous social category, an “emerging class” in his parlance. Precarious employment is more narrowly defined. Chapter 6. The Political Economy of Universal Basic Income 207 21. Vlandas (2018) argues that ESS variation in support for a basic income between regular and nonstandard employment is much less than expected from an insider/outsider theory per- spective. The large variation in nonstandard contracts across different countries, including access to benefits, likely underestimates important variation in levels of support among dif- ferent types of precarious workers. Country-level analysis offers important additional insight. Pulkka (2018) finds variation between support for basic income between full-time (48 per- cent) and part-time (61 percent) employees. 22. The primary concern regarding precarious employment is typically job tenure or job loss. The extent to which precarious workers worry about job loss can be captured as a function of the probability of losing one’s current job (job insecurity), one’s ability to find another job (labor market insecurity), and the availability of income support during an extended unemployment spell (income insecurity) (Anderson and Pontussen 2007). Precariousness in employment does not merely refer to uncertainty or instability of job tenure but also comprises a wide range of qualitative aspects, including wage rates, nature of tasks, opportunity to use and upgrade skills, or authority relations and employee participation. Taking this variety into account means we need to be careful about imputing similar interests in implementing a basic income policy for precarious workers who are, in effect, differently situated. 23. The Matthew effect, a term coined by sociologist Robert K. Merton, alludes to the Gospel of Matthew quote, “For to every one who has will more be given, and he will have abundance; but from him who has not, even what he has will be taken away.” In social policy, the Matthew effect refers to the phenomenon that the middle classes tend to be the main beneficiaries of social benefits and services, even if these are primarily targeted at the poor. 24. Even when middle classes express support for basic income in survey polling, this support may fragment once design and implementation details are put in full view (De Wispelaere 2016a); for a contrasting view, see Torry (2016) and Van Parijs and Vanderborght (2017). 25. For instance, Philippe Van Parijs insists on the superiority of a UBI over conventional means- tested welfare state programs, claiming that it will lead to the liberation of an individual from being forced to sell his or her labor power in the market to satisfy his or her most basic needs. 26. As for instance happened among U.K. Tories as a response to public concerns with increased fracking (UBIE 2017), to appease the opposition and make it more acceptable for the con- cerned communities (as a form of sharing profits from shale gas exploitation). However, their member of Parliament and former minister Nick Boles condemned it as “dangerous non- sense” (Mortimer 2017). 27. We actually know very little about trade union views; research here is very much lacking and information is anecdotal at best. 28. Benoît Hamon, the Socialist Party candidate for president in 2017, built his platform around basic income; however, it did not pass into the second round. 29. This is demonstrated by the very opportunistic use of the term “UBI” by the ruling populist coalition in Italy to evoke a concept that is very different, i.e., a guaranteed minimum income. 30. The UBI can positively influence all aspects of social policies and of a social protection pro- gram. An especially interesting case is that of Brazil, where in 2004, President Lula da Silva signed the UBI principle into law. Widely heralded by UBI advocates as a major victory, the law has, at least until now, achieved relatively little. Some researchers blame this on competi- tion for resources and political capital with the flagship conditional cash transfer Bolsa Família program, which was introduced at the same time. But there is a clear positive effect of UBI principles on the conditional cash transfer as it has became effective in Brazil. First, unlike in 208 Jurgen De Wispelaere and Ruslan Yemtsov most Latin American countries that strongly enforce the conditions of cash transfers, in Brazil the transfer includes an unconditional “guaranteed” part as an entitlement. The conditions do not trigger penalties or sanctions automatically, but failure to comply leads to increased attention from social workers and a search for remedial action; it takes years before benefits are affected (Fizsbein and Schady 2009). Second, the program has put a strong emphasis on inclusion, leading the government to launch the “Brazil without Misery” plan in 2011 which aimed to increase participation in and access to social programs by the extremely poor. 31. Congruence between formal policy goals and street-level operational goals is a prerequisite for effective welfare administration (Meyers, Riccucci, and Lurie 2001). 32. Examples of such performance indicators—e.g., impact on poverty and inequality, cost-bene- fit ratio—are detailed in chapter 4 using illustrations from country simulations. 33. See de Beus (2013) and De Wispelaere and Morales (2016). Stability is a key feature of any policy with long-term aspirations but is of specific interest to basic income, as the security of one’s future income stream is a critical feature of the intervention (Standing 2002). 34. De Wispelaere, Halmetoja, and Pulkka (2018) argue that the timing of the upcoming national election, weak support among Sipilä’s coalition partners, and the focus on basic income as a labor market activation instrument explain why expanding or extending the experiment was never really in the offing. 35. An important complication in experimenting with basic income is that while the basic income itself is unconditional, many recipients require additional support that remains highly con- ditional. The combined effect in terms of incentives and behavior is very far from the ideal model of basic income. 36. It is often argued that a basic income functions as a “third rail” in politics by virtue of being universal (Calnitsky 2016). The Alaska Permanent Fund seems to have weathered significant political pressure, attesting to its robustness as well as its resilience. This is largely due to the particular funding mechanism of a sovereign wealth fund (Bryan and Castillo 2012). More empirical research is required to understand other factors affecting a policy’s resilience. 37. As a one-time basic capital grant, the U.K. Child Trust Fund scheme was not a strict basic income (paid out in regular installments), but shares with basic income a focus on individ- ual, universal, and unconditional eligibility. See LeGrand (2006) for discussion. Two political factors that are used to explain the lack of opposition and hence the ease with which the U.K. government could abolish the Child Trust Fund are the lack of universality—the scheme was rolled out cohort by cohort—and the fact that those currently in the scheme would only achieve their benefit upon reaching adulthood in the future. This lack of perceived benefits and strong interests in support of the policy made it less resilient. 38. See Van Parijs and Vanderborght (2017). Simulations in chapter 4 use this as a starting point. 39. One explanation for why the most likely basic income model in the Finnish context is pitched at €560 per month (the amount of the basic income experiment) is that it matches what unemployed Finns receive from basic unemployment security payments. See Halmetoja, De Wispelaere, and Perkiö (2018) for a discussion of the institutional fit of a partial basic income in Finland. 40. See chapter 4 for a detailed analysis of winners and losers in several developing country con- texts. Also see Martinelli (2017) for a discussion of this problem based on microsimulations in the U.K. context. Chapter 6. The Political Economy of Universal Basic Income 209 41. The evolution of single-gateway policy structures varies cross-nationally. See Clasen et al. (2001). 42. Atkinson (1996) proposes a participation income. 43. Prime examples include care work or volunteering (Hiilamo and Komp 2018); but see Zelleke (2018) for a critical reflection from a care perspective. 44. See De Wispelaere and Stirton (2007) for a discussion of the implementation challenges of participation income. 45. See Frankel and Mulvale (2014) for a discussion of this strategy in Manitoba. 46. For example, Spain had a universal birth grant, cheque bebé, which was introduced in 2007 and discontinued in 2010 amidst the austerity measures intended to stabilize public finances. Discontinuing the policies, the government argued that a significant portion of the public was against the universal character of the grant, which was providing a sum of €2,500 even to bet- ter-off families that did not need it. 47. De Wispelaere (2016b) discusses the example of the Green Party in Ireland, which was a long- time supporter but never brought up basic income once in its terms as part of the coalition government during the 30th Dáil Éireann (2007–11). 48. See Steensland (2006) for a detailed analysis of the difficulties in obtaining a robust political coalition around a guaranteed income in the 1970s in the United States. 49. See Van Parijs and Vanderborght (2017) for a skeptical view of basic income experiments. 50. Kalliomaa-Puha, Tuovinen, and Kangas (2016) and Kangas, Simanainen, and Honkanen (2017) illustrate some of the hurdles faced by the Finnish basic income experiment. De Wis- pelaere, Halmetoja, and Pulkka (2018) discuss the political nature of experimental design choices in the Finnish context. 51. The experimental design was based on models advocated by the Green League and the Left Alliance, but nevertheless both offered harsh criticism to the experiment as proposed by the Sipilä government (De Wispelaere, Halmetoja, and Pulkka 2018). See De Wispelaere (2016a) for other examples of “toxic” support. References Acemoglu, Daron, and James A. Robinson. 2001. “A Theory of Political Transitions.” American Eco- nomic Review 91 (4): 938–63. https://economics.mit.edu/files/4121 Anderson, Christopher, and Jonas Pontussen. 2007. “Workers, Worries and Welfare States: Social Protection and Job Insecurity in 15 OECD Countries.” European Journal of Political Research 46 (2): 211–35. Andersson, Jan-Otto, and Olli Kangas. 2005. “Universalism in the Age of Workfare. Attitudes to Basic Income in Sweden and Finland.” In Normative Foundations of the Welfare State: The Nordic Experience, edited by Nanna Kildal and Stein Kuhnle. London: Routledge. Atamanov, Aziz, Jon Jellema, and Umar Serajuddin. 2015. “Energy Subsidies Reform in Jordan: Welfare Implications of Different Scenarios.” Policy Research Working Paper No. 7313. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/595761467998814982/ pdf/WPS7313.pdf 210 Jurgen De Wispelaere and Ruslan Yemtsov Atkinson, Anthony. 1996. “The Case for a Participation Income.” Political Quarterly 67 (1): 67–70. Baldwin, Peter. 1990. The Politics of Social Solidarity: Class Bases of the European Welfare State, 1875–1975. Cambridge, UK: Cambridge University Press. Birnbaum, Simon. 2012. Basic Income Reconsidered: Social Justice, Liberalism and the Demands of Equality. New York: Palgrave Macmillan. Bonoli, Giuliano, and Bruno Palier. 2000. “How Do Welfare States Change? Institutions and Their Impact on the Politics of Welfare State Reform in Western Europe.” European Review 8 (3): 333–52. Bryan, James B., and Sarah Lamarche Castillo. 2012. “Politics, Preservation of Natural Resource Wealth, and the Funding of a Basic Income Guarantee.” In Alaska’s Permanent Fund Dividend: Examining Its Suitability as a Model, edited by Karl Widerquist and Michael Howard. New York: Palgrave Macmillan. Bossuroy, Thomas, and Aline Coudouel. 2018. “Recognizing and Leveraging Politics to Expand and Sustain Social Safety Nets.” In Realizing the Full Potential of Social Safety Nets in Africa, edited by Kathleen Beegle, Aline Coudouel, and Emma Monsalve. Africa Development Forum series. Washington, DC: World Bank. Bott, Claire. 2017. “Namibia May Be Considering Basic Income Grants.” Basic Income Earth Network. News November 10. https://basicincome.org/news/2017/11/namibia-may- considering-basic-income-grants/ Calnitsky, David. 2016. “‘More Normal Than Welfare’: The Mincome Experiment, Stigma, and Community Experience.” Canadian Review of Sociology 53 (1): 26–71. —. 2017. “Debating Basic Income.” Catalyst 1 (23): 63–90. Caputo, Richard, ed. 2012. Basic Income Guarantee and Politics: International Experiences and Per- spectives on the Viability of Income Guarantees. New York: Palgrave Macmillan. Chrisp, Joe. 2017. “Basic Income: Beyond Left and Right?” Juncture 23 (4): 266–70. Chrisp, Joe, and Jurgen De Wispelaere. 2017. “Basic Income as Social Insurance? Labour Market Risk in the ‘New’ Welfare State.” Paper presented at the 17th BIEN Congress, Lisbon, Septem- ber 25–27. Clasen, J., G. Duncan, T. Eardley, M. Evans, P. Ughetto, W. van Oorschot, and S. Wright. 2001. “Towards ‘Single Gateways’? A Cross-National Review of the Changing Roles of Employment Offices in Seven Countries.” Zeitschrift für Ausländisches und Internationales Arbeits-und Sozi- alrecht 15 (1): 43–63. Coelho, Andrew. 2016. “On Why Basic Income Has Not Yet Been Deployed: Now It’s Namibia.” Basic Income Earth Network. Editorial May 11. https://basicincome.org/news/2016/05/ on-why-basic-income-has-not-yet-been-deployed-now-its-namibia/ —. 2018. “Hungary: Basic Income Related Activity in Hungary.” Basic Income Earth Network. News November 1. https://basicincome.org/news/2018/11/hungary-basic-income-related- activity-in-hungary/ CPI (Centre for Public Impact). 2016. “Basic Impact Grant (BIG) in Namibia.” Case study. https:// www.centreforpublicimpact.org/case-study/basic-income-grant-big-namibia/ Commander, Simon. 2012. “A Guide to the Political Economy of Reforming Energy Subsidies.” IZA Policy Paper 52. Institute for the Study of Labor, Bonn. Cummine, Angela. 2016. Citizens’ Wealth: Why (and How) Sovereign Funds Should Be Managed by the People for the People. New Haven: Yale University Press. Chapter 6. The Political Economy of Universal Basic Income 211 de Beus, Jos. 2013. “The Stability of Basic Income.” In Basic Income: An Anthology of Contemporary Research, edited by Karl Widerquist, José A. Noguera, Yannick Vanderborght, and Jurgen De Wispelaere. Oxford, UK: Wiley-Blackwell. Desai, Ray M., and Homi Kharas. 2017. “Is a Growing Middle Class Good for the Poor? Social Policy in a Time of Globalization.” Global Economy and Development Working Paper 105. Brookings Institution, Washington, DC. https://www.brookings.edu/research/is-a-growing-middle-class- good-for-the-poor-social-policy-in-a-time-of-globalization/ De Wispelaere, Jurgen. 2015. An Income of One’s Own? The Political Analysis of Universal Basic Income. Tampere, Finland: UTA Press. https://trepo.tuni.fi/bitstream/handle/10024/98162/978- 951-44-9989-0.pdf —. 2016a. “Basic Income in Our Time: Improving Political Prospects through Policy Learn- ing?” Journal of Social Policy 45 (4): 617–34. —. 2016b. “The Struggle for Strategy: On the Politics of the Basic Income Proposal.” Politics 36 (2): 131–41. De Wispelaere, Jurgen, Antti Halmetoja, and Ville-Veikko Pulkka. 2018. “The Rise (and Fall) of the Basic Income Experiment in Finland.” CESifo Forum 19 (3): 15–19. De Wispelaere, Jurgen, and Luke Martinelli. 2017. “A New Universalism? Varieties of Basic Income and Welfare State Reform.” ESPAnet Conference Proceedings, Lisbon, September 14–16, 2017. De Wispelaere, Jurgen, and Leticia Morales. 2016. “The Stability of Basic Income: A Constitutional Solution for a Political Problem?” Journal of Public Policy 36 (4): 521–45. De Wispelaere, Jurgen, and Lindsay Stirton. 2007. “The Public Administration Case against Partic- ipation Income.” Social Service Review 81(3): 523–49. Douillard, Austin. 2017. “New Study Published on Results of Basic Income Pilot in Kenya.” Basic Income Earth Network. News March 27. https://basicincome.org/news/2017/03/us-kenya- new-study-published-results-basic-income-pilot-kenya/ Fitzgerald, Rory. 2017. “Survey Reveals Young People More Likely to Support Universal Basic Income, But It’s Not a Left-Right Thing.” The Conversation November 17. https://theconversation.com/ survey-reveals-young-people-more-likely-to-support-universal-basic-income-but-its-not-a-left- right-thing-87554 Fizsbein, Ariel, and Norbert Schady. 2009. Conditional Cash Transfers: Reducing Present and Future Poverty. Washington, DC: World Bank. https://siteresources.worldbank.org/INTCCT/ Resources/5757608-1234228266004/PRR-CCT_web_noembargo.pdf Francese, Maura, and Delphine Prady. 2018. “Universal Basic Income: Debate and Impact Assess- ment.” Working Paper WP/18/273. Fiscal Affairs Department, International Monetary Fund, Washington, DC. Frankel, Sid, and James P. Mulvale. 2014. “Support and Inclusion for All Manitobans: Steps toward a Basic Income Scheme.” Manitoba Law Journal 37 (2): 425–64. Fritz, Verena, Brian Levy, and Rachel Ort, eds. 2014. Problem-Driven Political Economy Analy- sis: The World Bank’s Experience. Directions in Development. Washington, DC: World Bank. https://openknowledge.worldbank.org/bitstream/handle/10986/16389/9781464801211. pdf?sequence=1&isAllowed=y Gangl, Markus. 2004. “Welfare States and the Scar Effects of Unemployment: A Comparative Anal- ysis of the United States and West Germany.” American Journal of Sociology 109 (6): 1319–64. —. 2006. “Scar Effects of Unemployment: An Assessment of Institutional Complementari- ties.” American Sociological Review 71 (6): 986–1013. 212 Jurgen De Wispelaere and Ruslan Yemtsov Gibson, D., and R.E. Goodin. 1999. “The Veil of Vagueness: A Model of Institutional Design.” In Organizing Political Institutions: Essays for Johan P. Olsen, edited by M. Egeberg and P. Lægreid. Oslo: Scandinavian University Press. Goodman, Peter S. 2017. “Capitalism Has a Problem. Is Free Money the Answer?” New York Times November 15. https://www.nytimes.com/2017/11/15/business/dealbook/universal-basic- income.html Gourevitch, Alex, and Lucas Stanczyk. 2018. “The Basic Income Illusion.” Catalyst 1 (4): 1–22. Groot, Loek. 2004. Basic Income, Unemployment and Compensatory Justice. Dordrecht: Kluwer. Hacker, Jacob. 2004. “Privatizing Risk Without Privatizing the Welfare State: The Hidden Politics of Social Policy Retrenchment in the United States.” American Political Science Review 98 (2): 243–60. —. 2005. “Policy Drift: The Hidden Politics of US Welfare State Retrenchment.” In Beyond Con- tinuity: Institutional Change in Advanced Political Economies, edited by Wolfgang Streeck and Kathleen Thelen. New York: Oxford University Press. Halmetoja, Antti, Jurgen De Wispelaere, and Johanna Perkiö. 2018. “A Policy Comet in Moomin- land? Basic Income in the Finnish Welfare State.” Social Policy and Society 13: 1–12. Häusermann, Silja, Thomas Kurer, and Hanna Schwander. 2015. “High-Skilled Outsiders? Labor Market Vulnerability, Education and Welfare State Preferences.” Socio-Economic Review 13 (2): 235–58. Hiilamo, Heikki, and Kathrin Komp. 2018. “The Case for a Participation Income: Acknowledging and Valuing the Diversity of Social Participation.” Political Quarterly 89 (2): 1–6. Hill, Ruth, and Philip Verwimp. 2017. “Managing Risk and Conflict.” In Accelerating Poverty Reduc- tion in Africa, edited by Kathleen Beegle and Luc Christiaensen, chapter 4. Washington, DC: World Bank. Holmes, Anisa. 2017. “31% of Europeans Want Basic Income as Soon as Possible.” Blog post May 3. Dalia. https://daliaresearch.com/31-of-europeans-want-basic-income-as-soon-as-possible/ Inchauste, Gabriela, and David G. Victor, eds. 2017. The Political Economy of Energy Sub- sidy Reform. Washington, DC: World Bank. http://documents.worldbank.org/curated/ en/745311489054655283/pdf/113262-PUB-PUBLIC-PUBDATE-3-22-17.pdf Ipsos MORI. 2017. “Half of UK Adults Would Support Universal Basic Income in Principle.” Poll- ing commissioned by the Institute for Policy Research, University of Bath. https://www.ipsos. com/ipsos-mori/en-uk/half-uk-adults-would-support-universal-basic-income-principle Kalleberg, Arne. 2018. Precarious Lives: Job Insecurity and Well-Being in Rich Democracies. Cam- bridge, UK: Polity Press. Kalliomaa-Puha, Laura, Anna-Kaisa Tuovinen, and Olli Kangas. 2016. “The Basic Income Experi- ment in Finland.” Journal of Social Security Law 23: 75–88. Kangas, Olli, Miska Simanainen, and Pertti Honkanen. 2017. “Basic Income in the Finnish Con- text.” Intereconomics 52 (2): 87–91. Khemani, Stuti. 2017. “Political Economy of Reform.” Policy Research Working Paper No. 8224. World Bank, Washington, DC. Khemani, Stuti, James Habyarimana, and Irfan Nooruddin. 2019. “What Do Poor People Think about Direct Cash Transfers?” Blog post April 8. Brookings Institution. https://www.brookings.edu/ blog/future-development/2019/04/08/what-do-poor-people-think-about-direct-cash-transfers/ Korpi, Walter, and Joakim Palme. 1998. “The Paradox of Redistribution and Strategies of Equality: Welfare State Institutions, Inequality, and Poverty in the Western Countries.” American Socio- logical Review 63 (5): 661–87. Chapter 6. The Political Economy of Universal Basic Income 213 Lavers, Samuel, and Tom Hickey. 2016. “Conceptualising the Politics of Social Protection Expan- sion in Low Income Countries: The Intersection of Transnational Ideas and Domestic Politics.” International Journal of Social Welfare 25 (4): 388–98. Lee, Soomi. 2018. “Attitudes toward Universal Basic Income and Welfare State in Europe: A Research Note.” Basic Income Studies 13 (1): 101–9. LeGrand, Julian. 2006. “Implementing Stakeholder Grants: The British Case.” In Redesigning Redis- tribution: Basic Income and Stakeholder Grants as Cornerstones for an Egalitarian Capitalism, edited by Erik Olin Wright. New York: Verso. Lindert, Kathy, and Vanina Vincensini. 2010. “Social Policy, Perceptions and the Press: An Analysis of the Media’s Treatment of Conditional Cash Transfers in Brazil.” Social Protection Dis- cussion Paper No. 1008. World Bank, Washington, DC. http://siteresources.worldbank.org/ SOCIALPROTECTION/Resources/SP-Discussion-papers/Safety-Nets-DP/1008.pdf Mahoney, John, and Kathleen Thelen. 2010. “A Theory of Gradual Institutional Change.” In Explain- ing Institutional Change: Ambiguity, Agency and Power, edited by John Mahoney and Kathleen Thelen. New York: Cambridge University Press. —. 2015. “Comparative-Historical Analysis in Contemporary Political Science.” In Advances in Comparative-Historical Analysis, edited by John Mahoney and Kathleen Thelen, 3–36. Cam- bridge, UK: Cambridge University Press. Marsh, David, and Allan McConnell. 2010. “Towards a Framework for Establishing Policy Success.” Public Administration 88 (2): 564–83. Martinelli, Luke. 2017. “Assessing the Case for a Universal Basic Income in the UK.” IPR Policy Brief. Institute for Policy Research, University of Bath, Bath. https://www.bath.ac.uk/publications/ assessing-the-case-for-a-universal-basic-income-in-the-uk/attachments/basic_income_policy_ brief.pdf McCarthy, John, and Mulyadi Sumarto. 2018. “Distributional Politics and Social Protection in Indo- nesia: Dilemma of Layering, Nesting and Social Fit in Jokowi’s Poverty Policy.” Journal of Southeast Asian Economies 35 (2): 223–36. Mettler, Suzanne, and Mallory SoRelle. 2014. “Policy Feedback Theory.” In Theories of the Policy Process, edited by Paul A. Sabatier and Christopher M. Weible. Boulder: Westview Press. Meyers, Marcia K., Norma M. Riccucci, and Irene Lurie. 2001. “Achieving Goal Congruence in Com- plex Environments: The Case of Welfare Reform.” Journal of Public Administration Research and Theory 11 (2): 165–201. Mortimer, Caroline. 2017. “Tory MP Condemns Universal Basic Income as ‘Dangerous Nonsense.’” Independent December 29. https://www.independent.co.uk/news/uk/politics/universal-basic- income-tory-mp-nick-boles-conservative-poverty-unemployment-a8133461.html Murray, Charles. 2016. In Our Hands: A Plan to Replace the Welfare State. Washington, DC: Ameri- can Enterprise Institute Press. Nadeau, Richard, Edouard Cloutier, and J.-H. Guay. 1993. “New Evidence about the Existence of a Bandwagon Effect in the Opinion Formation Process.” International Political Science Review 14 (2): 203–13. Noguera, José A. 2019. “The Political Debate on Basic Income and Welfare Reform in Spain.” Social Policy and Society 18 (2): 289–99. Nova, Annie. 2018. “Personal Finance: More Americans Now Support a Universal Basic Income.” CNBC February 26. https://www.cnbc.com/2018/02/26/roughly-half-of-americans-now- support-universal-basic-income.html Parolin, Zachary, and Linus Siöland. 2018. “Support for a Universal Basic Income: A Demand-Capac- ity Paradox?” Paper presented at the Foundation of International Social Security Conference, Sigtuna, Sweden. 214 Jurgen De Wispelaere and Ruslan Yemtsov Pierson, Paul. 2000. “Increasing Return, Path Dependence, and the Study of Politics.” American Political Science Review 4 (2): 251–67. —. 2003. “When Effect Becomes Cause: Policy Feedback and Political Change.” World Politics 45 (4): 595–628. —. 2004. Politics in Time: History, Institutions, and Social Analysis. Princeton: Princeton Uni- versity Press. Piketty, Thomas. 2016. “What Unequal Societies Need Is Not a ‘Basic Income’ But a Fair Wage.” The Wire. https://thewire.in/uncategorised/basic-income-fair-wage-piketty Pruce, Kate, and Sam Hickey. 2017. “The Politics of Promoting Social Protection in Zambia.” ESID Working Paper 75. Effective States and Inclusive Development Research Centre, University of Manchester, Manchester, UK. http://www.effective-states.org/wp-content/uploads/working_ papers/final-pdfs/esid_wp_75_pruce_hickey.pdf Pulkka, Ville-Veikko. 2018. “Finland Shares Unconditional Money, But the Public View Remains Polarised.” Blog post January 23. Institute for Policy Research, University of Bath. http://blogs. bath.ac.uk/iprblog/2018/01/23/finland-shares-unconditional-money-but-the-public-view- remains-polarised/ Reich, Robert. 2015. Saving Capitalism: For the Many, Not the Few. New York: Knopff. Rice, Faun. 2018. “Sri Lanka: Basic Income Debates and Initiatives.” Basic Income Earth Network. News August 14. https://basicincome.org/news/2018/08/sri-lanka-basic-income-debates-and-initiatives/ Roemer, John. 1998. “Why the Poor Do Not Expropriate the Rich: An Old Argument in New Garb.” Journal of Public Economics 70 (3): 399–424. Salehi-Isfahani, Djavad. 2014. “Iran’s Subsidy Reform: From Promise to Disappointment.” Policy Perspective 13. Economic Research Forum, Giza, Egypt. http://erf.org.eg/publications/irans- subsidy-reform-from-promise-to-disappointment/ Standing, Guy. 2002. Beyond the New Paternalism: Basic Security as Equality. New York: Verso. —. 2011. The Precariat: The New Dangerous Class. London: Bloomsbury. Steensland, Brian. 2006. “Cultural Categories and the American Welfare State: The Case of Guar- anteed Income Policy.” American Journal of Sociology 111 (5): 1273–326. Steinmo, S. 2010. The Evolution of Modern States: Sweden, Japan, and the United States. Cambridge, UK: Cambridge University Press. Stern, Andy. 2016. Raising the Floor: How a Universal Basic Income Can Renew Our Economy and Rebuild the American Dream. New York: Public Affairs. Stirton, Lindsay, Jurgen De Wispelaere, Johanna Perkiö, and Joe Chrisp. 2017. “Modelling Political Parties’ Support for Basic Income in Finland, 1979–2016.” Paper presented at the 17th BIEN Congress, Lisbon, September 25–27. Streeck, Wolfgang, and Kathleen Thelen. 2005. “Institutional Change in Advanced Political Econ- omies.” In Beyond Continuity: Institutional Change in Advanced Political Economies, edited by Wolfgang Streeck and Kathleen Thelen. New York: Oxford University Press. Tabatabai, Hamid. 2012. “Iran: A Bumpy Road towards Basic Income.” In Basic Income Guarantee and Politics: International Experiences and Perspectives on the Viability of Income Guarantees, edited by Richard Caputo, 285–300. New York: Palgrave Macmillan. Torry, Malcolm. 2012. “The United Kingdom: Only for Children?” In Basic Income Guarantee and Politics: International Experiences and Perspectives on the Viability of Income Guarantees, edited by Richard Caputo. New York: Palgrave Macmillan. —. 2016. The Feasibility of Citizen’s Income. New York: Palgrave Macmillan. Chapter 6. The Political Economy of Universal Basic Income 215 UBIE (Unconditional Basic Income Europe). 2017. “Basic Income after the Elections in France and the UK.” Interview July 26. https://www.ubie.org/basic-income-after-the-elections-in-france- and-the-uk/ van der Veen, Robert J., and Philippe Van Parijs. 1986. “A Capitalist Road to Communism.” Theory and Society 15 (5): 635–55. van de Walle, Dominique. 2014. “The Democratization of Clientelism in Sub-Saharan Africa.” In Clientelism, Social Policy, and the Quality of Democracy, edited by Diego Abente Brun and Larry Diamond, 230–52. Baltimore: Johns Hopkins University Press. van Oorschot, Wim. 2000. “Who Should Get What, and Why? On Deservingness Criteria and the Conditionality of Solidarity among the Public.” Policy and Politics 28 (1): 33–48. —. 2006. “Making the Difference in Social Europe: Deservingness Perceptions among Citi- zens of European Welfare States.” Journal of European Social Policy 16 (1): 23–42. Van Parijs, Philippe. 1995. Real Freedom for All: What (If Anything) Could Justify Capitalism? Oxford, UK: Oxford University Press. Van Parijs, Philippe, and Yannick Vanderborght. 2017. Basic Income: A Radical Proposal for a Free Society and a Sane Economy. Cambridge, MA: Harvard University Press. Vlandas, Tim. 2018. “The Politics of Universal Basic Income (UBI).” Blog post March 7. LSE Blog, New European Trade Unions Forum. http://blogs.lse.ac.uk/netuf/2018/03/07/the-politics-of- universal-basic-income-ubi/ Walker, Mark. 2016. Free Money for All: A Basic Income Guarantee Solution for the 21st Century. New York: Palgrave Macmillan. Widerquist, Karl, and Michael Howard, eds. 2012a. Alaska’s Permanent Fund Dividend: Examining Its Suitability as a Model. New York: Palgrave Macmillan. —. 2012b. Exporting the Alaska Model: Adapting the Permanent Fund Dividend for Reform around the World. New York: Palgrave Macmillan. World Bank. 2008. “The Political Economy of Policy Reform: Issues and Implications for Policy Dialogue and Development Operations.” Social Development Department Report 44288-GLB, World Bank, Washington, DC. —. 2016. Making Politics Work for Development: Harnessing Transparency and Citizen Engage- ment. Policy Research Report. Washington, DC: World Bank. http://documents.worldbank.org/ curated/en/268021467831470443/pdf/106337-PUB-REVISED-PUBLIC.pdf —. 2017. World Development Report 2017: Governance and the Law. Washington, DC: World Bank. https://www.worldbank.org/en/publication/wdr2017 Wren, Anne, and Philip Rehm. 2014. “The End of the Consensus? Labour Market Developments and the Politics of Retrenchment.” Socio-Economic Review 12 (2): 409–35. Wright, Erik Olin. 2010. Envisioning Real Utopias. London: Verso. Yamamori, Toru. 2018. “UN Secretary-General Endorses UBI.” Basic Income Earth Network. News September 26. https://basicincome.org/news/2018/09/un-secretary-general-endorses-ubi/ Yemtsov, Ruslan, Aleksandra Posarac, Maria Nagernyak, and Irina Albegova. 2019. “Towards More Effective Social Assistance in Russia: An Update of the System Performance Considering New National Target of Halving Poverty by 2024.” Report No: AUS0000898. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/391591564381077495/pdf/Strengthening- Social-Protection-Towards-more-effective-social-assistance-in-Russia-An-update-of-the- system-performance-considering-new-national-target-of-halving-poverty-by-2024.pdf Zelleke, Almaz. 2018. “Work, Leisure, and Care: A Gender Perspective on the Participation Income.” Political Quarterly 89 (2): 273–79. What Does It Take to Deliver a Universal CHAPTER 7 Basic Income in Practice? Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini T his chapter explores the practical administrative steps needed to successfully implement a universal basic income (UBI) scheme. A UBI is distinct from a typical social protection program, because of both its universal nature and its provision of payments as a right, rather than as a discretionary benefit. In practice, however, a UBI is still a variant of a social protection scheme, and its implementation would require many of the same systems as in a standard social protection delivery chain. This chap- ter discusses each phase of the delivery chain in keeping with the framework described by Lindert, George, and Rodriguez-Caillava (forthcoming), showing how for a UBI some steps are simpler and some more demanding (figure 7.1). The chapter also considers key foundational elements of the surrounding ecosystem for a social protection agency’s successful operation of a UBI—namely, identification (ID) systems, civil registration sys- tems, interoperability and data protection frameworks, and payment service provision. Delivery Chain Phase 1: Assessing Delivering a UBI needs a database of all individuals who are entitled to the program and payment information for each person. Developing this database constitutes the first phase of the delivery chain. A country’s approach to completing this phase will vary greatly according to the existing state of its government systems, as well as its preferred strategies for outreach 217 218 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini FIGURE 7.1  UBI Delivery Chain Compared to Typical Social Assistance Program Outreach Wider-reaching but simpler, rights-based message Must register many more individuals, but only once Intake and Registration Phase 1: Assess per person and only basic information Assessment of Needs Automated screening based on very limited and Conditions exclusionary criteria Determine Eligibility SKIP (universal right to payment) and Enrollment Phase 2: Decide Usually no variation in payment amount Set Benefit Level (possible adult/child or urban/rural distinction) Much larger scale increases importance of financial Payments Delivery inclusion and e-payment capacity Phase 3: Implement No conditionality monitoring and fewer eligibility Monitoring complaints but solid monitoring needed to ensure quality, consistency, and legitimacy of payments and registration. In countries with advanced tax systems or large social registries for existing social programs, the government may already have ID and payment informa- tion for much of the population. Typically, this information will be registered at the household level, meaning that data will still need to be converted to or newly collected at the individual level. Some existing registries may already include entries and payment information for all or almost all individuals within a certain age category. For example, old-age social pensions have proliferated in the past two decades, now covering on aver- age more than half of the population over age 60 in the Sub-Saharan African region and offering universal old-age coverage in countries such as Bolivia, Mauritius, and Namibia (World Bank 2018b). Notwithstanding, in most low- and middle-income countries, the majority of the population is not yet covered by tax databases and social registries, necessitating a substantial outreach and registration effort to build a populationwide reg- istry of individuals and their payment information. Outreach A UBI outreach campaign would need to be wide reaching, using a range of communi- cation strategies to inform the entire population about its right to a basic income, the plan for implementation, and instructions for registration. As for all social assistance programs, outreach will need to be rigorous in reaching the hard to serve. The poor- est and most in need typically live in more remote areas, have less access to means of communication, are more likely to be illiterate, may not speak the official language of the country, and may have little trust in government, among other challenges. As dis- cussed in chapter 1, this outreach is important even for universal programs. If the UBI is being introduced as a substitute for any existing programs, special care must be taken to ensure that past program beneficiaries are informed about and supported through the transition. Aside from these concerns, the message of a UBI information campaign Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 219 should be relatively simple compared to the intensive and repeated outreach needed to explain targeted programs with complex assessments and intricate eligibility criteria. Intake and Registration The UBI database would need to cover the entire population but would require much less information about each registrant than targeted schemes, which often use detailed socio- economic data to determine eligibility, customize benefit levels, and/or offer allied services. The system would only need to register each individual once, and no continuous updates to reassess eligibility based on socioeconomic and household changes would be needed. Limited updates would need to be made over time, however, both in maintaining the roster of claimants (registering births, deaths, or newly qualified residents or citizens) and in the mechanics of payment (a claimant may move or change his or her bank account, minors will reach the age of majority and be paid individually rather than via their parents). Imple- mentation might therefore involve a major push for a couple of years to register the entire population, and then move to a less intensive stage of ongoing maintenance. The social registry is the main system for collecting and processing registrant infor- mation (box 7.1). Unlike the registries of many existing social assistance programs, a UBI registry would need to contain indi- vidual- rather than household-level files. The primary information needed from BOX 7.1  UBI within an Integrated each individual would be his or her Social Registry identifying data, payment information, W and proof of residency or citizenship hile a UBI’s registry would need (if relevant). If the UBI entitlements to cover more individuals than begin from birth, the government will other programs, it would need rela- also need to establish a mechanism for tively simple functionality and limited registering and making payments to data collection for each registrant. To minors, such as linking minors’ files to improve overall efficiency and avoid a designated caregiver. There may also repetition in data collection, the reg- be a procedure to allow payments to be istry used for the UBI may well be an made to a caregiver for people with a integrated social registry, serving mul- severe disability. tiple social protection programs. Many Countries may enlist various of these other schemes would likely approaches to collect information for UBI continue to exist even after the intro- registration. En masse census sweeps duction of a UBI. Thus, the integrated (using door-to-door or mobile registra- social registry used for the UBI might tion units) can be effective in reaching still gather and process detailed infor- people who are otherwise unaware mation to enable other programs to of the program, but they are highly determine eligibility or benefit levels, resource intensive. Another option is connect to allied services, or link indi- to provide registration on demand, viduals within the same household. whereby people can submit application 220 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini forms online, by mail, and/or in person, either during particular registration windows or on a continuous basis. Information collected from registrants through either en masse sweeps or on-demand methods can be supplemented by linking to other administrative systems, such as the ID system, the civil registry, or immigration or citizenship systems. Some countries may choose to develop a UBI registry by building on the infor- mation and registration processes already used for existing social registries, which are present in at least 60 low- and middle-income countries (Socialprotection.org 2017). The information in a few of these registries is already quite expansive, covering a vast majority of households (figure 7.2). For example, Chile, the Dominican Republic, Paki- stan, and the Philippines all have large social registries covering 75 percent or more of their population (Leite et al. 2017). In many other countries, social registry coverage is at 50 percent or less of the population. Countries in the early stage of developing a social registry, such as Sierra Leone and Mali, cover less than 5  percent of the population, demanding enormous efforts to expand the registry for UBI implementation. While building on existing registries may be the preferred approach in some con- texts, other countries may choose to approach universal registration from a blank slate. As an example, India has laid the groundwork for a potential UBI by creating a new ID FIGURE 7.2  Population Coverage of Social Registries by Country Mali RSU Sierra Leone SPRINT Azerbaijan VEMTAS Mauritius SRM Macedonia CBMIS Yemen, Rep., SWF Senegal RNU Montenegro SWIS Georgia TSA Indonesia UDB Brazil Cadastro Unico Mexico SIFODE Turkey ISAS Colombia SISBEN Chile RSH Philippines Listahanan 0 10 20 30 40 50 60 70 80 % of population SOURCE: Leite et al. 2017. Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 221 database, Aadhaar, from scratch. This database now covers 1.2 billion people (virtually all the country’s adults and a growing number of children), and forms the basis for social pro- gram delivery (OECD 2018). Assessment of Needs and Conditions In the standard social protection delivery chain, the registration stage is followed by an assessment process, which determines registrants’ eligibility for the program based on the information collected. The assessment step will be far less demanding for a UBI than for targeted schemes, consisting only of a simple, automated screening based on any exclusionary criteria the UBI policy has set (e.g., screening out children or noncitizens). The elimination of complex targeting procedures has been one of the most common arguments made in favor of the UBI, with advocates often forecasting large administrative savings from cutting assessments of needs and conditions (e.g., Jhabvala and Standing 2010; Kidd 2016). While some savings will certainly be realized, because the program will cover so many more people, it is not clear that these will be sufficient for a UBI’s total administrative costs to be lower than those for targeted programs. There is limited recent literature estimating the administrative costs of targeting (Devereux et al. 2017), but ear- lier estimates suggest costs of around 0.6–9.0  percent of total program costs (Baulch 2002; Grosh et al. 2008), or 3–8 percent of the value of transfers (Van der Berg, Siebrits, and Lekezwa 2010). While low relative to overall program costs, targeting procedures can account for up to about 75 percent of administrative costs (Grosh et al. 2008), though they can also be as low as 18 percent (Tesliuc et al. 2014). This large range indicates that the actual costs of targeting—and therefore the potential savings from its elimination—vary greatly depending on the particular targeting approach, program, and delivery context. Phase 2: Decision Making The second phase in the social protection delivery chain consists of decision making about enrollment and benefit levels, both of which are straightforward processes for a UBI. As a universal program, there is no discretion to select the most in need from among eligible registrants, as sometimes happens based on budget availability in tar- geted schemes. Furthermore, there is little to no variation in benefit levels among scheme recipients, with the only potential difference being a lower benefit value for children rel- ative to adults, or slight geographic adjustments based on the cost of living in different areas (De Wispelaere and Stirton 2004). Phase 3: Implementation Payments A UBI would generally follow the same payment administration and provision processes as any other cash transfer, with the only major differences being the larger number of claimants and the use of the individual as the assistance unit with UBI (versus the 222 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini household unit as used for many other cash transfers). Both of these factors dramati- cally expand the number of payments administered, increasing the appeal of a digital payment system (in which money is transferred electronically into recipients’ financial institution or mobile money account, rather than manually distributed in cash or check form at local payment points). Government capacity for e-payments, as well as the state of the payments infrastructure (detailed later in this chapter), is currently inadequate for a UBI in most low- and middle-income countries. Thus, countries interested in a UBI might first improve payment structures before rolling out a UBI and/or rely on a mix of manual and digital methods to deliver a UBI scheme. Currently, 74  percent of people receiving government payments in upper-mid- dle-income countries are paid electronically via a financial institution or mobile account; 15 percent receive payments in cash and 12 percent through some other method such as checks or vouchers (Demirgüç-Kunt et al. 2018). In lower-middle-income countries also, digital payments are the norm, with just over half (55 percent) of people receiving gov- ernment payments into an account, compared to 27 percent in cash and 18 percent by another method. But in low-income countries, cash continues to be the most common method for government payments (reported by 43 percent of recipients), although digi- tal payments made into an account follows closely behind (39 percent). Shifting from a manual to a digital system for government payments can offer many advantages, including potential increases in cost efficiency and reductions in leak- ages. In Brazil, switching from a manual to a digital payment system issued by the state-owned bank helped cut administrative costs from 15 percent to 3 percent of the disbursed benefit (Lindert et al. 2007); the leakage of pension payment funds in a trial in Andhra Pradesh, India, dropped by 47 percent when payments were made via bio- metric smart cards rather than cash handouts (Muralidharan, Niehaus, and Sukhtankar 2016). Switching from manual to electronic payments can also bring important benefits to recipients, such as improved financial inclusion and program accessibility. For exam- ple, in Thailand, 14 percent of adults with a bank or mobile money account opened their first account to receive a government transfer (Demirgüç-Kunt et al. 2018). In Niger, beneficiaries of a social assistance program saved approximately 3.5 hours of travel and wait time per mobile money transfer, relative to the manual cash equivalent (Aker et al. 2016). Because of their lower marginal cost (for both government and recipients) and greater flexibility, digital payments can be paid more frequently and more easily stag- gered over the month, reducing security risks and supporting markets to better cope with potential surges in demand following disbursements. While digital payment systems offer significant benefits, they also bring challenges. They typically take much longer to set up—at least 12–18 months, compared to only 3–12 months for a manual payment system (ISPA 2016). They require reliable physical infrastructure and can easily be undermined by electricity or mobile network outages. Nationwide financial infrastructure is also needed, but banks often do not find it cost-ef- fective to extend services across the population, prompting risks of financial exclusion. Lower levels of literacy, numeracy, and technology access further heighten exclusion risks among disadvantaged groups. Depending on the partnership arrangement reached, Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 223 bank fees for service provision can make e-payment mechanisms less cost-effective for the government and/or the recipient. Within digital payment systems, government transfers into a unique, preauthen- ticated bank account are considered the gold standard, but require substantial upfront investment. One of the historical challenges has been that Know Your Customer (KYC) banking regulations require authentication of a person’s identity before a bank account can be opened, and many people in developing countries have lacked official ID docu- mentation. Yet the case of India demonstrates that these challenges can be overcome. As part of a push to increase banking access across the population, a rapid electronic authentication process was used, enabling ID to be verified and KYC requirements to be met anywhere with mobile phone connectivity in a matter of minutes. Through this ini- tiative, India increased the share of adults with a bank account from 35 percent in 2011 to 80 percent in 2017 (Demirgüç-Kunt et al. 2018). A less common digital option for government payments is mobile money pay- ment delivery. Of the 60 million unbanked adults worldwide who receive government transfers in cash, two-thirds have a mobile phone (Demirgüç-Kunt et al. 2018). Mobile phone ownership, however, is not sufficient to enable successful government payments through this mechanism. An appropriate regulatory environment must be developed, since existing banking regulations typically do not cover mobile money providers (ISPA 2016). Nationwide network connectivity and mobile data coverage are needed, along with a reliable electricity supply to power mobile phones, cell towers, and payment pro- vider operating systems. Even Kenya, a leader in mobile money infrastructure, could not make a mobile money–based solution for a social assistance scheme work from 2010 to 2012 due to network connectivity issues (Zimmerman, Bohling, and Parker 2014). The program administrators instead opted to transfer payments into accounts with debit cards issued by a financial institution. Monitoring As an unconditional entitlement, a UBI would not need to monitor recipients’ compli- ance with any of the conditions stipulated by conditional cash transfer or job search/ labor activation schemes. However, monitoring systems would still be required to ensure that complete and on-time payments were made to all intended recipients and not paid to any duplicate, fraudulent, or deceased recipients. Ceasing payments upon death may actually be a fairly challenging task, if an indi- vidual is only required to register once for a UBI program, and payments are thereafter transferred digitally into their account. Ideally, the UBI registry would be linked to an up-to-date death register in the civil registration system, automatically flagging deceased recipients in the UBI system. But, as discussed below, many countries have extremely limited civil registration systems and deaths routinely go unrecorded. Other approaches would therefore be needed to identify deceased beneficiaries. Where biometric authenti- cation is required for payments, proof of life would automatically be provided each time a person accesses his or her account. A more burdensome alternative would be to rely 224 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini on periodic in-person recertification for continued participation in the UBI scheme. The required frequency of this recertification might be greater for those above a certain age— for example, UBI recipients under 65 years of age might be asked to present themselves in person every five years, while those over 65 might have to do so every three years, and those over 75 annually. As in all social protection programs, recertification requirements present important trade-offs. Tighter demands may reduce the risk of leakage, but they also heighten the potential for exclusion errors and increase the administrative burden. Another important function of the monitoring system would be to address any issues, complaints, or appeals. Programs with simpler eligibility requirements generally have lower rates of errors, fraud, and corruption (van Stolk and Tesliuc 2010), suggesting that a UBI scheme would have a fairly small proportion of grievances in relative terms. On the other hand, given its universal scope and permanent duration, a UBI program will still have significant numbers of complaints in absolute terms, requiring a solid grievance redress mechanism. This mechanism could build on existing social protection grievance redress systems, which are in place in many countries but have historically been underused and/or underperforming, due to both demand- and supply-side prob- lems (Barca, Notosusanto, and Emmett 2012). International best practice recommends that there be multiple channels for resolving complaints, including independent chan- nels such as ombudsmen, audit institutions, or third-party complaint handlers (Barca 2016). To reduce costs and improve accessibility for citizens, there should be grievance redress mechanisms at the point of service delivery, where information and transaction costs are lowest. Surrounding Ecosystem: Foundational Elements Besides the systems created and managed directly by the relevant social protection agency, several additional systems should be in place to create the enabling architecture for UBI delivery. To accurately register and pay recipients, each individual must be iden- tifiable in a credible manner, requiring a strong ID system. Ideally, this should go hand in hand with a robust civil registration system to ensure that continuous changes in the population (e.g., births and deaths) are recorded. These ID systems need to communi- cate with multiple other administrative systems (such as the UBI registry and payrolls), meaning that well-considered interoperability and data protection frameworks are also needed. In addition, the provision of payment services must be sufficiently developed so the government can make a substantial proportion of UBI payments electronically. ID Systems Identification is the process of ensuring that an individual has a unique identifier estab- lishing who he or she is (Leite et al. 2017), thus enabling verification of a registrant’s or recipient’s identity, ensuring nonduplication, and linking the social registry with other Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 225 administrative information systems (such as the tax or civil registration system). The unique identifier can take the form a foundational ID (a unique government-recognized ID credential) or a functional ID (assigned to the individual by a government agency for a specific transaction or service, such as a voter card, health or insurance records, or a driver’s license). While in past decades, ID systems often existed in manual, paper-based forms, countries today are increasingly relying on digital technologies in their ID systems. Modern ID systems use digital technologies to enable more efficient and transpar- ent administration and service delivery, increased security, reduced fraud and payment leakages, more accurate statistics for planning purposes, and greater capacity to respond to disasters and epidemics (World Bank 2017). But digital ID platforms also pose a number of risks and challenges, including inadvertent exclusion of hard-to-reach popu- lations, onerous mandates that deter individuals from accessing services, and increased rent-seeking involving registration or certificates (World Bank 2016). In the worst case, identifying information may be used for nefarious purposes, such as to track or discrim- inate against marginalized groups. There are thus serious legal and regulatory concerns about the types, extent, and use of information collected. Ensuring trusted data exchange and data security is a major technological challenge, as is the extension of digital access to remote areas. Government procurement of digital platforms for ID systems can also be challenging, since technology solutions are often tied to specific vendors. A UBI requires a government to be able to find and pay each member of the pop- ulation; this is clearly far easier when the government has a single, accurate population database and a means of quickly and credibly verifying each person’s identity. Provided that the above risks are appropriately managed, a foundational ID system using digital technologies could provide an optimal ID ecosystem for UBI implementation. Yet this configuration is out of reach for many countries at present. Where digital technologies are less advanced, a government may need to rely on certain manual elements for issu- ing or authenticating a person’s identity. And where a foundational ID system has yet to be created, a government may instead expand upon existing functional ID databases (e.g., electoral rolls or earlier social programs’ registries) to find and verify payments to individuals (Atick et al. 2014). Globally, around 1  billion people lack access to a government-recognized ID.1 Half (50  percent) of these people live in Sub-Saharan Africa, and approximately one- third (32 percent) live in South Asia. Nearly two-thirds of unidentified people worldwide reside in lower-middle-income countries, with the largest absolute numbers living in India (162 million), Nigeria (140 million), and Pakistan (77 million). Just over a quarter of unidentified people live in low-income countries, with the largest absolute number in Ethiopia, where 69 million people lack access to official ID. There is wide variation in the national coverage of ID systems in developing countries (figure 7.3). In 10 countries, more than half of the population lacks govern- ment-recognized IDs—this includes 77 percent of Somalia’s population, 72 percent of Nigeria’s, and 70 percent of Eritrea’s. By contrast, less than 1 percent of the population remains unidentified in many countries, including several lower-middle-income coun- tries such as El Salvador, Georgia, and Mongolia. 226 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini FIGURE 7.3  Percentage of Populations without Official Proof of Identity Somalia Nigeria Eritrea Ethiopia Angola Zambia South Sudan Chad Equatorial Guinea Yemen, Rep. Uganda Guinea-Bissau Tanzania Libya Swaziland Djibouti Liberia Zimbabwe Cameroon Côte d’Ivoire Mauritania Congo, Dem. Rep. Mozambique Pakistan Sudan Botswana Afghanistan Gabon Burkina Faso Bangladesh Myanmar Micronesia, Fed. Sts. Madagascar Niger Kiribati Senegal Central African Republic Burundi South Africa Tunisia Nepal Morocco Guinea Comoros Vanuatu Lesotho Mali Bhutan Gambia, The Timor-Leste Malawi Azerbaijan Papua New Guinea Samoa Tuvalu Sierra Leone Rwanda Paraguay Togo Bolivia Kenya Jordan Lao PDR Benin Guatemala Tajikistan Philippines Ghana Kazakhstan Maldives Grenada Nicaragua Cambodia Lebanon Venezuela, RB Haiti Palestine Namibia Congo, Rep. India Belize Algeria Syrian Arab Republic Panama Solomon Islands Indonesia Nauru Turkmenistan Korea, Dem. People’s Rep. Costa Rica Jamaica Brazil Suriname Dominican Rep. Cuba Cabo Verde Uzbekistan Iran, Islamic Rep. Vietnam Guyana Mauritius Dominica Fiji Serbia Tonga Ukraine Mexico Honduras São Tomé and Príncipe Kosovo China Kyrgyz Republic Egypt, Arab Rep. Marshall Islands St. Lucia Croatia Turkey Malaysia Sri Lanka El Salvador Colombia Iraq Armenia Albania Mongolia Macedonia, FYR 0 10 20 30 40 50 60 70 80 90 Percent SOURCE: World Bank Identification for Development (ID4D) Global Dataset 2018. Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 227 Marginalized groups are particularly likely to lack access to ID systems. Data on 99 countries show a large gender gap in the unidentified population in low-income coun- tries, with over 45 percent of women lacking a national ID compared to 30 percent of men (World Bank 2018a). In the same low-income countries, nearly half (45 percent) of those in the poorest income quintile lack a national ID, whereas only a quarter (28 per- cent) of those in the richest quintile lack official ID. Reaching these marginalized groups often requires significantly more resources than the average cost of ID provision (Palacios 2014). For example, Peru increased national ID coverage to around 95 percent with special outreach campaigns. While the standard cost of this was US$10, the cost was double (US$22) in the coastal region, qua- druple (US$42) in the mountains, and twice over again (US$80) in the jungle. The feasibility of bringing large numbers of unregistered people into the ID system is continually increasing, thanks to improvements in the mobility and affordability of digital technologies. In recent decades, many countries have either shifted from paper- based to electronic ID systems, or leapfrogged the paper-based stage altogether to build new national ID systems from scratch using digital technologies (World Bank 2016). Cur- rently, 161 countries have ID systems based on digital technologies.2 This includes more than three-quarters of developing countries (World Bank 2016), with almost every coun- try in Africa and Asia having either introduced an e-ID or intending to do so in the near future. Figure 7.4 shows the rapid growth in national and digital ID systems in Sub-Saha- ran Africa over the last decade. Digital technologies are useful for many aspects of the ID process (World Bank 2017), including registration, issuance of credentials, and authentication of IDs: •• For registration, many countries are now collecting biometric data (such as finger- prints, iris, face scan) alongside traditional biographic data (name, date of birth, gender, etc.). The biometric data greatly facilitate the process of de-duplication, FIGURE 7.4  National and Digital ID Systems in Sub-Saharan Africa, 1960–2018 50 40 Numbe of countries 30 20 National ID systems Digital ID systems 10 0 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 SOURCE: World Bank Identification for Development (ID4D) Global Dataset 2018. 228 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini particularly in countries without reliable civil registration documents to verify the uniqueness of biographic information. •• For the issuance of ID credentials, countries are moving beyond purely paper-based IDs to provide digital ID cards (e.g., cards with bar codes or more advanced chip- based smart cards), mobile IDs (e.g., SIM cards with digital ID certificates), or entirely virtual IDs (e.g., India’s Aadhaar program does not provide any physical cre- dentials; instead, each person’s biometric ID is stored on a server). As of early 2017, 82 percent of all countries issuing official ID cards have implemented programs that use smart cards or plastic cards and biometrics (Acuity Market Intelligence 2017). •• For authentication of IDs, various digital technologies are also growing in cover- age, including biometric readers, personal identification number (PIN) codes, digital signatures, and smartphone applications. Experts expect to see as many as 600 million devices with biometric authentication by 2021 (Smith 2016). By far the most famous country example of a newly developed digital ID system is India, which has dominated the headlines for its ambitious Aadhaar program. Through this initiative, more than 1 billion Indians have been biometrically enrolled in just six years. India achieved this mass identification at an extremely low cost, spending US$1.16 per person for enrollment and registration (Atick 2014). The India example is at the very lowest end of what a foundational ID system is typically expected to cost per capita. One estimate puts the average cost of enrollment and registration for a foundational ID system at approximately US$3–US$6 per person, plus an additional 15–25 percent per year for maintenance, software, and data updat- ing (Atick 2014). Card production and distribution—which was not part of the cardless Aadhaar initiative—may cost an additional US$1–US$5 per person (and an additional US$0.50 for digital certificates), plus US$0.05–US$0.10 per card per year for mainte- nance. Civil Registration Systems Even as countries make major advances with their ID systems, an important role remains for civil registration systems—the official systems for the universal, continu- ous, permanent, and compulsory recording of vital events, such as births and deaths (UN 1998). For example, countries with the most sophisticated biometric ID systems still benefit from a strong civil registration system because it records biographic information at birth, whereas most biometric data cannot be collected from newborn infants. Civil registration systems thus establish a legal identity at birth and provide the initial basis for enrollment in a foundational ID system. Civil registration coverage in many countries is very limited (figure 7.5). Globally, the births of around one-quarter of children under the age of five have never been record- ed.3 Birth registration rates are especially low in Sub-Saharan Africa (43  percent) and South Asia (60 percent), with Ethiopia’s and Somalia’s rates dipping as low as 3 percent. Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 229 FIGURE 7.5  Birth Registration Coverage by Region Sub-Saharan Africa 43 South Asia 60 East Asia and Pacific 84 Middle East and North Africa 92 Latin America and the Caribbean 95 Eastern Europe and Central Asia 99 Western Europe 100 North America 100 World 71 0 20 40 60 80 100 % of children under age 5 whose births are registered SOURCE: UNICEF global databases, 2017. Death registration is even more limited, with 81 countries collecting data of very low quality or not registering deaths at all—including all low-income countries and two-thirds of lower-middle-income countries.4 Interoperability and Data Protection Frameworks Some social assistance programs need extensive interoperability with other systems. This might be the case, for example, to check beneficiary eligibility against databases with socioeconomic information, and to verify conditionality adherence against health or schooling records. As a universal and unconditional scheme, a UBI draws on far fewer systems, perhaps limited to the ID, civil registration, and social registry systems described in this chapter. Yet even between these systems, there will be a need for interoperability frameworks and a unique ID or alternative identifier to link correspond- ing entries across databases. Given the vast number of identifiable files in a UBI database, strong data protection frameworks adhering to international data transfer and information privacy protocols are critical.5 According to the United Nations Conference on Trade and Development, only 58 percent of countries globally have data protection laws in place. A further 10 per- cent have legislation in draft form, and 21 percent of countries have no data protection legislation at all.6 A data protection framework should also include a number of practical elements, including memorandums of understanding and data sharing protocols, strong access restrictions for different levels of staff, and network security (Leite 2018). Arrange- ments for database storage vary, with some social programs storing their own data, and others outsourcing the security and hosting of databases to a third party (Barca 2017). For example, the servers for Kenya’s Hunger Safety Net Programme are physically and logically secured at the program level; Pakistan’s Benazir Income Support Programme 230 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini database is hosted by the National Database and Registration Authority; and South Afri- ca’s integrated beneficiary operations management system, SOCPEN, is hosted by the South African State Information Technology Agency. Payment Service Provision Another foundational element is the availability and uptake of banking or mobile money services; this strongly determines the feasibility of regular government payments to the entire population. Recent years have seen significant improvements in the global availability and uptake of banking and mobile money services, with 1.2  billion people having gained access to some form of financial account since 2011 (Demirgüç-Kunt et al. 2018). Accord- ing to Global Findex, the world’s most comprehensive database on financial inclusion, 69 percent of adults worldwide now have an account with a financial institution (a bank, credit union, or cooperative or microfinance institution) or a mobile money provider (Demirgüç-Kunt et al. 2018). In high-income countries, account access is nearly univer- sal, at 94  percent, and in upper-middle-income countries the rate is now 73  percent. However, account ownership is still only 58 percent in lower-middle-income countries, and 35 percent in low-income countries. The vast majority of account owners (98  percent globally) have an account at a financial institution (Demirgüç-Kunt et al. 2018). Mobile money is far less widely used (6 percent of account owners globally), although it is growing in availability and is now present in 90 countries—including three-quarters of low- and lower-middle-income countries (GSMA 2017). The global number of mobile money accounts increased by 25 percent in 2017 relative to the previous year, primarily driven by fast growth in Africa and Asia. There are now 12 countries (all in Sub-Saharan Africa) where more people have accounts with mobile money providers than with a financial institution (Demirgüç-Kunt et al. 2018). But these countries remain the exception; in about half (52 percent) of all countries, less than 5 percent of account owners have mobile money accounts or no data are available. As figure 7.6 shows, there is substantial deviation across and within regions in account ownership rates (Demirgüç-Kunt et al. 2018). As a region, Sub-Saharan Africa has the lowest level at 43 percent, followed by the Middle East and North Africa (48 per- cent) and Latin America and the Caribbean (54 percent). Within these regions, countries exhibit great variation. For example, in Sub-Saharan Africa, more than 80 percent of the population in Kenya and Mauritius has an account, compared to less than 20  percent of the population in the Central African Republic, Madagascar, Niger, and South Sudan. Among low- and middle-income countries, the Islamic Republic of Iran and Mongolia (both of which have already implemented universal cash transfer schemes) have the high- est financial inclusion levels, surpassing 90 percent; other countries in their respective regions have account ownership levels below 25 percent, including Iraq and Cambodia. This leaves around 1.7 billion adults worldwide still lacking access to an account, with women, low-income households, and people with lower education levels Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 231 FIGURE 7.6  Percentage of Adults (15+) with an Account at a Financial Institution or with a Mobile Money Provider Iran, Islamic Rep. Mongolia Mauritius Croatia Malaysia Thailand Kenya Belarus Namibia China India Macedonia, FYR Russian Federation Sri Lanka Venezuela, RB Bulgaria Serbia Brazil South Africa Turkey Montenegro Costa Rica Libya Ukraine Georgia Uganda Bosnia and Herzegovina Kazakhstan Gabon Romania High-income mean Ghana Dominican Republic Zimbabwe Bolivia Kosovo Ecuador Botswana Bangladesh Rwanda Indonesia Argentina Paraguay Armenia Tajikistan Tanzania Panama Zambia Colombia Lesotho Nepal Honduras Upper-middle-income mean Togo Lebanon Guatemala Moldova Burkina Faso Algeria Peru Jordan Senegal Mozambique Turkmenistan Albania Kyrgyz Republic Nigeria Benin Uzbekistan Mexico Tunisia Liberia Mali Cameroon Lower-middle-income mean Philippines Malawi Egypt, Arab Rep. Haiti Nicaragua Vietnam El Salvador Lao PDR Morocco Azerbaijan Congo, Rep. Myanmar Congo, Dem. Rep. West Bank and Gaza Guinea Iraq Chad Cambodia Pakistan Mauritania Low-income mean Sierra Leone Madagascar Niger Afghanistan Central African Republic South Sudan 0 20 40 60 80 100 Percent SOURCE: Demirgüç-Kunt et al. 2018. 232 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini disproportionately represented (Demirgüç-Kunt et al. 2018). In 61 countries, fewer than half of adults have an account; in 24 of these countries, account ownership drops below one-third. In such contexts, UBI implementation would be highly challenging in the near future. But the rapid surge in account access in certain middle-income countries and the evolution of supportive technologies provide cause for some optimism. Given the six-year leap from 35 to 80 percent account ownership in a country as vast and populous as India, widespread e-payments within five years seems feasible—with a concerted effort—in many middle-income countries. For low-income countries, the average timeline would be stretched, but Uganda’s progress from 20 to 59 percent ownership in six years suggests that market and technology growth alongside financial inclusion initiatives can produce rapid progress toward e-payment readiness even in less-developed settings. Summary Compared to generating political consensus or finding financing for a UBI, the imple- mentation challenges may be smaller. The ability to implement critically depends on having or creating an enabling environment of widespread foundational ID and wide- spread financial inclusion. Many countries are as yet far from universal coverage in this regard, but the goals are already high on development agendas and there are numerous successful experiences. Further, the lure of receiving a UBI payment may help mobi- lize citizens to obtain IDs and bank accounts, once the regulatory and delivery systems are developed and ready for rollout. Thus, if a country wants to implement a UBI and gives development of the underlying systems full and determined political and techni- cal support, they should be able to make progress in the time frame of a single political administration. Conclusions As discussed in previous chapters, a UBI looks simple in theory, but could prove more challenging in practice. In this chapter, we have shown that the implementation of a UBI would still require many of the same processes and systems as in the delivery chain for social protection programs. Depending on the availability of universal ID and pay- ment systems, substantial outreach using a range of communications methods may be required to register the population. A UBI may use an integrated social registry covering multiple social protection programs with individual- instead of household-level files. Reg- istration may be through a one-time census sweep, through on-demand methods, or by linking to other administrative databases through an interoperability framework for data sharing across government. This would need to be conducted with the appropriate data protection and privacy protocols in place. Under a UBI, a simple automated screening may take the place of complex assess- ments of needs and conditions. There is little to no variation in benefit levels, with Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 233 potentially only a lower benefit value for children, or slight adjustments over time based on the cost of living. A UBI program would utilize bulk payments administration and transactions with the individual as the unit of assistance, rather than the household head. Improvements in payments infrastructure and public financial management systems would be necessary to enable these transactions seamlessly to financial or nonfinancial accounts. Monitoring systems would be critical to ensure complete, on-time payments to intended recipients and to check for error, fraud, duplication, or corruption, as well as to address grievances, complaints, and appeals. Notes 1. The data in this and the following paragraph are from the Identification for Development (ID4D) Global Dataset 2018, https://datacatalog.worldbank.org/dataset/identification-development- global-dataset. 2. ID4D Global Dataset 2018. 3. United Nations Children’s Fund (UNICEF), Birth Registration (database) 2017, https://data. unicef.org/topic/child-protection/birth-registration/. 4. World Health Organization, Global Health Observatory Data 2018, http://www.who.int/gho/ mortality_burden_disease/registered_deaths/text/en/. 5. Examples of data protection frameworks include the Council of Europe’s Convention for the Protection of Individuals with regard to Automatic Processing of Personal Data, the United Nations Guidelines for the Regulation of Computerized Personal Data Files, and the Organ- isation for Economic Co-operation and Development (OECD) Guidelines on the Protection of Privacy and Transborder Flows of Personal Data. These standards are practiced beyond Europe and OECD geographies; for example, they are common across Latin America. 6. United Nations Conference on Trade and Development, Data Protection and Privacy Legislation Worldwide (database) 2018, http://unctad.org/en/Pages/DTL/STI_and_ICTs/ICT4D-Legislation/ eCom-Data-Protection-Laws.aspx. No information is available on the remaining countries. References Acuity Market Intelligence. 2017. The Global National eID Industry Report: 2017 Edition. Louisville, CO: Acuity Market Intelligence. http://www.acuity-mi.com/GNeID_Report_2017.php Aker, Jenny C., Rachid Boumnijel, Amanda McClelland, and Niall Tierney. 2016. “Payment Mechanisms and Anti-Poverty Programs: Evidence from a Mobile Money Cash Transfer Experiment in Niger.” Working Paper. Fletcher School and Department of Economics, Tufts University, Medford, MA. Atick, Joseph. 2014. “Digital Identity: The Essential Guide.” ID4Africa. http://www.id4africa.com/ main/files/Digital_Identity_The_Essential_Guide.pdf Atick, Joseph, Alan Harold Gelb, Seda Pahlavooni, Elena Gasol Ramos, and Ziad Safdar. 2014. Dig- ital Identity Toolkit: A Guide for Stakeholders in Africa. Washington, DC: World Bank. http:// documents.worldbank.org/curated/en/147961468203357928/pdf/912490WP0Digit00Box38 5330B00PUBLIC0.pdf 234 Christina Lowe, Margaret Grosh, Tina George, and Ugo Gentilini Barca, Valentina. 2016. “Grievance Mechanisms for Social Protection Programmes: Stumbling Blocks and Best Practice.” One-Pager 320. International Policy Centre for Inclusive Growth, Brasilia. http://www.ipc-undp.org/pub/eng/OP320_Grievance_mechanisms_for_social_ protection_programmes_stumbling_blocks_and_best_practice.pdf —. 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. Barca, Valentina, Smita Notosusanto, and Bethan Emmett. 2012. Review of, and Recommendations for, Grievance Mechanisms for Social Protection Programmes. Oxford, UK: Oxford Policy Management. Baulch, Bob. 2002. “Poverty Monitoring and Targeting Using ROC Curves: Examples from Viet- nam.” Working Paper 161. Institute of Development Studies, Brighton. De Wispelaere, Jurgen, and Lindsay Stirton. 2004. “The Many Faces of Universal Basic Income.” Political Quarterly 75 (3): 266–74. Demirgüç-Kunt, Asli, Leora Klapper, Dorothe Singer, Saniya Ansar, and Jake Hess. 2018. The Global Findex Database 2017: Measuring Financial Inclusion and the Fintech Revolution. Washington, DC: World Bank. https://openknowledge.worldbank.org/handle/10986/29510 Devereux, Stephen, Edoardo Masset, Rachel Sabates-Wheeler, Michael Samson, Althea-Maria Rivas, and Dolf te Lintelo. 2017. “The Targeting Effectiveness of Social Transfers.” Journal of Development Effectiveness 9 (2): 162–211. Grosh, Margaret, Carlo Del Ninno, Emil Tesliuc, and Azedine Ouerghi. 2008. From Protection to Promotion: The Design and Implementation of Effective Safety Nets. Washington, DC: World Bank. GSMA. 2017. 2017 State of the Industry Report on Mobile Money. London: GSMA. https://www. gsma.com/mobilefordevelopment/wp-content/uploads/2018/05/GSMA_2017_State_of_the_ Industry_Report_on_Mobile_Money_Full_Report.pdf ISPA (Inter-Agency Social Protection Assessment Partnership). 2016. Social Protection Payment Delivery Mechanisms Tool. Washington, DC: World Bank. Jhabvala, Renana, and Guy Standing. 2010. “Targeting to the ‘Poor’: Clogged Pipes and Bureau- cratic Blinkers.” Economic & Political Weekly 45: 26–27. Kidd, Stephen. 2016. “Universal Provision Is More Effective than Poverty Targeting.” Blog post February 9. Agence Française de Développement ID4D. https://ideas4development.org/en/ social-protection-universal-provision-is-more-effective-than-poverty-targeting/ Leite, Phillippe. 2018. “Social Registry Information System for Social Assistance.” Core course. World Bank, Washington, DC. http://pubdocs.worldbank.org/en/897101464039057078/ SPLCC-2016-SNCC-D4S1-Leite-Social-Registry.pdf Leite, Phillippe, Tina George, Changqing Sun, Theresa Jones, and Kathy Lindert. 2017. “Social Registries for Social Assistance and Beyond: A Guidance Note & Assessment Tool.” Social Pro- tection and Labor Discussion Paper 1704. World Bank, Washington, DC. http://documents. worldbank.org/curated/en/698441502095248081/pdf/117971-REVISED-PUBLIC-Discussion- paper-1704.pdf Lindert, Kathy, Tina George, and Inés Rodriguez-Caillava. Forthcoming. A Sourcebook on the Foun- dations of Social Protection Delivery Systems. Washington, DC: World Bank. Lindert, Kathy, Anja Linder, Jason Hobbs, and Benedicte de la Briére. 2007. “The Nuts and Bolts of Brazil’s Bolsa Família Program: Implementing Conditional Cash Transfers in a Decentralized Context.” Social Protection Working Paper No. 0709. World Bank, Washington, DC. Muralidharan, Karthik, Paul Niehaus, and Sandip Sukhtankar. 2016. “Building State Capacity: Evi- dence from Biometric Smartcards in India.” American Economic Review 106 (10): 2895–929. Chapter 7. What Does It Take to Deliver a Universal Basic Income in Practice? 235 OECD (Organisation for Economic Co-operation and Development). 2018. “Case Study: Aadhaar– India.” In Embracing Innovation in Government: Global Trends 2018. Paris: OECD. https://www. oecd.org/gov/innovative-government/India-case-study-UAE-report-2018.pdf Palacios, Robert. 2014. “Social Protection Delivery Systems with a Focus on Identification.” World Bank, Washington, DC. Smith, Sam. 2016. “Voice and Facial Recognition to Be Used in Over 600 million Mobile Devices by 2021.” Juniper Research, London. https://www.juniperresearch.com/press/press-releases/ voice-and-facial-recognition-to-be-used-in-over-60 Socialprotection.org. 2017. “Information Systems for the Social Protection Sector: Social Regis- tries and Beyond.” Webinar presentation by Oxford Policy Management, the World Bank Group, and Australia Department of Foreign Affairs and Trade. http://socialprotection.org/ sites/default/files/publications_files/Information%20systems%20for.pdf Tesliuc, Emil, Lucian Pop, Margaret Grosh, and Ruslan Yemtsov. 2014. Income Support for the Poor- est: A Review of Experience in Eastern Europe and Central Asia. Washington, DC: World Bank. UN (United Nations). 1998. Handbook on Civil Registration and Vital Statistics System. ST/ESAISTAT/ SER.F/73. New York: UN. https://unstats.un.org/unsd/publication/SeriesF/SeriesF_73E.pdf Van der Berg, Servaasr, Krige Siebrits, and Bongisa Lekezwa. 2010. “Efficiency and Equity Effects of Social Grants in South Africa.” Working Paper 15. Stellenbosch University, Department of Economics. van Stolk, Christian, and Emil Tesliuc. 2010. “Toolkit on Tackling Error, Fraud and Corruption in Social Protection Programs.” SP Discussion Paper No. 1002. World Bank, Washington, DC. http://siteresources.worldbank.org/SOCIALPROTECTION/Resources/SP-Discussion-papers/ Safety-Nets-DP/1002.pdf World Bank. 2016. “Spotlight 4. Enabling Digital Development: Digital Identity.” In World Develop- ment Report 2016: Digital Dividends. Washington, DC: World Bank. http://pubdocs.worldbank. org/en/822821519686607466/9781464806711-WDR2016-Spot4-Rev-Oct2017.pdf —. 2017. Technical Standards for Digital Identity: Draft for Discussion. World Bank: Washington, DC. http://pubdocs.worldbank.org/en/579151515518705630/ID4D-Technical-Standards-for- Digital-Identity.pdf —. 2018a. Global ID Coverage by the Numbers: Insights from the ID4D-Findex Survey. Washington, DC: World Bank. http://pubdocs.worldbank.org/en/953621531854471275/ID4D-FINDEX- Note-Release2018.pdf —. 2018b. The State of Social Safety Nets 2018. Washington, DC: World Bank. https:// openknowledge.worldbank.org/bitstream/handle/10986/29115/211254.pdf?sequence=4 Zimmerman, Jamie, Kristy Bohling, and Sarah Rotman Parker. 2014. “Electronic G2P Payments: Evidence from Four Lower-Income Countries.” Focus Note 93/2014. Consultative Group to Assist the Poor, Washington, DC. Appendix A: UBI-Related Pilots Characteristic Description Maricá, Brazil – Renda Básica de Cidadania: Municipal Government of Maricá Type of intervention Universal basic income Benefit level (absolute terms) 10 mumbucas per month initially (R$10, or ~US$3); increased to 20 mumbucas (R$20, ~US$6) in June 2017 Benefit level (% of average Not specified income/living wage) Frequency Monthly Method of transfer Paid electronically through a virtual social currency (mumbuca) Targeting criteria None Coverage All citizens (roughly 150,000) since beginning of 2016; previously, 14,000 of the poorest families (a third of the population) Duration Not specified; continuous: 2015–present Source of funds Government (financed by town’s oil revenues) Comments Based on background research conducted for this volume, program seems not to have been implemented in practice, nor are there plans for a UBI in Maricá for the future Link https://bit.ly/2QkP2na Ontario, Canada (Thunder Bay, Lindsay, and Hamilton): Government of Ontario Type of intervention Unconditional cash transfer Benefit level (absolute terms) •• Single participants: up to Can$16,989/year; couples: up to Can$24,000/year; participants with disabilities eligible for another Can$6,000/year, although they will then not receive state disability support, as payments replace Ontario Works and Ontario Disability Support Program; this can work out to more money •• Disbursement reduced by Can$0.50 for each dollar earned Benefit level (% of average 75% of Statistics Canada’s low-income measure income/living wage) Frequency Monthly Method of transfer Bank transfer Targeting criteria Participants selected randomly—low-income people (age 18–64) living on less than Can$34,000 individually, or Can$48,000 as a couple; this includes those who are working, in school, or living on financial assistance Coverage Up to 4,000 individuals in three cities Duration 2 years: April 2017–March 2019; program was originally supposed to end after 3 years in April 2020 237 238 Appendix A: UBI-Related Pilots Characteristic Description Source of funds Government of Ontario Comments •• During implementation, the Canada Child Benefit and the Ontario Child Benefit were maintained as guaranteed security incomes for children •• Budget: Can$50 million/year •• Pilot canceled July 2018 Link https://bit.ly/2JWq5vh; https://bit.ly/2BCCTXj; https://bit.ly/2qCyV7p Dauphin, Manitoba, Canada – Mincome: Canadian Government/Provincial Government Type of intervention Universal basic income Benefit level (absolute terms) •• Everyone given same base amount; cutoff varied depending on family size and location: Can$3,800–Can$5,800/year •• Can$0.50 subtracted from every dollar earned from other income sources Benefit level (% of average 60% of Statistics Canada’s low-income cutoff income/living wage) Frequency Monthly Method of transfer Checks Targeting criteria Universal Coverage Dauphin (total population: 10,000; rural municipality population: 2,500) Duration 5 years: 1974–79 Source of funds 75% federal funds, 25% provincial (cost: Can$17 million) Comments Program replaced existing social benefit schemes Link https://en.wikipedia.org/wiki/Mincome Finland: Kela (Finland’s Social Security Agency) Type of intervention Unconditional basic income Benefit level (absolute terms) €560/month Benefit level (% of average Not specified income/living wage) Frequency Monthly Method of transfer Not specified Targeting criteria Unemployed, age 25–58 Coverage 2,000 randomly selected households Duration 2 years: 2017–2019 Source of funds Government Comments •• Replaced existing social benefit schemes •• Budget: €20 million/year •• Government to discontinue payments beyond January 2019 Link http://julkaisut.valtioneuvosto.fi/bitstream/handle/10024/161361/Report_ The%20Basic%20Income%20Experiment%2020172018%20in%20Finland.pdf Hong Kong SAR, China: Hong Kong SAR Government Type of intervention Tax rebate/unconditional cash transfer Benefit level (absolute terms) HK$6,000/year Benefit level (% of average Not specified income/living wage) Frequency Annual Method of transfer Bank transfer/checks via mail Targeting criteria Adults with ID card (not necessary to be residing in Hong Kong SAR) Coverage Nationwide (4 million people registered in November 2011) Appendix A: UBI-Related Pilots 239 Characteristic Description Duration Not initially, but not renewed after first year: August 2011–June 2012 Source of funds Government Comments In addition to existing social programs Link https://bit.ly/2Rm9iqa; https://bit.ly/2Qr0wpo Madhya Pradesh, India – Madhya Pradesh Unconditional Cash Transfers Project: UNICEF/SEWA Type of intervention Universal basic income Benefit level (absolute terms) Adults: Rs 200; children: Rs 100; after a year, amounts raised to Rs 300 and Rs 150, respectively; in year-long tribal pilot, amounts were Rs 300 and Rs 150 Benefit level (% of average Roughly 30% of expenditure for a family (of five) at the poverty line in the income/living wage) state Frequency Monthly Method of transfer Directly into bank account (for women in villages where SEWA operated, funds were transferred to SEWA cooperative account) Targeting criteria Everyone in target villages (mothers/designated guardians receive monthly funds for children under 18) Coverage 9 villages (including 1 tribal village); 6,000 participants Duration 12–17 months: June 2011–December 2012 Source of funds UNICEF Comments Program was in addition to existing social programs Link https://bit.ly/2zM9qIs Islamic Republic of Iran: Iranian Government Type of intervention Universal basic income Benefit level (absolute terms) US$40–US$45/month Benefit level (% of average 29% of median household income/15% of minimum wage income/living wage) Frequency Monthly Method of transfer Bank deposits Targeting criteria Citizens based in country who applied for monthly payment Coverage Nationwide; 96% of the population received payments in 2011 Duration Not specified; continuous: December 2010–present Source of funds Government funds Comments •• Scheme replaced bread and energy subsidies •• Monthly payments (cash subsidies) paid to household head, not individuals •• Transfers amounted to 6.5% of gross domestic product in 2011, first full year of implementation •• Media campaign conducted to encourage wealthy people to forgo their payments •• Proposal passed by Parliament in 2016 to cut payments to 24 million citizens Link https://theforum.erf.org.eg/2017/11/19/energy-subsidies-universal-basic- income-lessons-iran/ Kenya: GiveDirectly Type of intervention Universal basic income Benefit level (absolute terms) 1. US$23 monthly (40 villages for 12 years) 2. US$23 monthly (80 villages for 2 years) 3. One lump sum at start equal in net present value as group 2 (70 villages) 4. Control group—no transfers (100 villages) 240 Appendix A: UBI-Related Pilots Characteristic Description Benefit level (% of average Roughly half average income in rural Kenya income/living wage) Frequency Monthly Method of transfer Mobile-based transfer Targeting criteria Resident of a treatment village, must own mobile phone Coverage 21,000 receive some type of cash transfer over 12 years (5,000 receive long- term basic income) Duration 2–12 years: 2011–present Source of funds Crowd funded Comments •• Program in addition to existing social programs •• Budget: US$25 million •• Money transferred using M-Pesa Link https://bit.ly/22KiRwF Macau SAR, China – Wealth Partaking Scheme: Government of Macau SAR, China Type of intervention Social wealth fund/universal basic income Benefit level (absolute terms) •• Since 2014, permanent residents (including those under age 18) receive P 9,000/year (US$1,300); nonpermanent residents receive P 5,400 (US$670) •• Between 2008 and 2013, permanent residents received annual amounts between P 5,000 and P 8,000; nonpermanent residents received a little over half of that Benefit level (% of average Not specified income/living wage) Frequency Annual Method of transfer Direct bank transfer/mailed check Targeting criteria None Coverage All residents/resident ID card holders; in 2017, there were 638,600 permanent residents and 62,000 nonpermanent residents entitled to benefits Duration Not specified; supposed to be continuous: 2008–present Source of funds Government Comments •• In addition to scheme, Macao SAR, China, government has injected annual capital into all qualified Provident Fund individual accounts since 2010 •• Program funded by profits from city’s casino taxes •• Total budget for 2017: P 6,080 million (US$757 million); this is renewed each year through legislation •• Initially a one-off policy, but has evolved into a long-term one Link https://bit.ly/2QqbNq6 Mongolia – Human Development Fund: Mongolian Government Type of intervention Social wealth fund/unconditional cash transfer Benefit level (absolute terms) •• Initial: Tog 120,000/person (US$89) •• Tog 70,000/person (US$52) distributed February 2010; Tog 10,000/person (US$7) distributed August–December 2010 •• Tog 21,000/person (US$17) distributed January 2011–June 2012 Benefit level (% of average Not specified income/living wage) Frequency Monthly Method of transfer Bank transfer Targeting criteria None Coverage Nationwide Appendix A: UBI-Related Pilots 241 Characteristic Description Duration Not specified initially; program ended after 28 months: February 2010–June 2012 Source of funds Mining income Comments •• Program replaced universal cash transfer for children program (Human Development Fund cost was three times that of Mongolia Development Fund; latter funded the child grant); after Human Development Fund program ended, child grants were restarted •• Annual cost: Tog 324 billion Link https://www.ebrd.com/downloads/research/economics/workingpapers/ wp0138.pdf Otjivero-Omitara Village, Namibia – Basic Income Grant Experiment Type of intervention Universal basic income Benefit level (absolute terms) •• N$100/month (~US$12) for adults until age 60; after project ended, monthly allowance of N$80 was paid to all participants •• Money for children and youths up to age 21 provided to a person designated as their primary caregiver; by default, this is the mother Benefit level (% of average Not specified income/living wage) Frequency Monthly Method of transfer Cash dispensed at designated payout points; after July 2008, through direct bank transfers Targeting criteria Everyone age 18–60 Coverage 930 people Duration 2 years: January 2008–December 2009 Source of funds Experiment financed by donations from individuals, churches, organizations, and donors, organized through Basic Income Grant Coalition Namibia Comments Complemented other programs, including universal old-age pension (then N$500/month) Link https://bit.ly/2IwvTvX; https://bit.ly/2Qsb8V6; https://bit.ly/1Swpygj Utrecht, Netherlands – Weten Wat Werkt (Know What Works) Type of intervention Unconditional basic income Benefit level (absolute terms) •• Group 1: conditional benefits for people who live alone: €972; couples: €1,390 (according to the old workfare regime) •• Group 2: unconditional benefits for people who live alone: €972; couples: €1,390 •• Group 3: same as Group 1, only an extra €150 at month’s end if they choose to do volunteer work •• Group 4: same as Group 3 but receive €50 upfront, which will need to be returned if they do not do volunteer work •• Group 5: receive unconditional benefits, no bonus for volunteering, allowed to earn from other jobs •• Group 6: expressed interest in receiving basic income, but will continue to receive only standard benefits Benefit level (% of average Not specified income/living wage) Frequency Monthly Method of transfer Not specified Targeting criteria City residents Coverage 250 citizens Duration 2 years: start date to be determined Source of funds Government 242 Appendix A: UBI-Related Pilots Characteristic Description Comments •• Not yet voted on by City Council, but will probably be approved •• Delays in obtaining permission from national-level authorities due to experiment’s unconditional aspect; authorities say it goes against 2015 Participant Act which requires citizens to actively seek jobs/participate in the labor market Link https://bit.ly/2maGKmj; https://bit.ly/2NYdLRz Barcelona, Spain – B-Mincome: City Council of Barcelona Type of intervention Universal basic income Benefit level (absolute terms) •• Between €100 (US$110) and €1,676 (US$1,850) per month for a household (not an individual) depending on household composition •• 10 treatment groups differ according to whether the basic income is accompanied by an additional program and whether the Municipal Inclusion Support (SMI) is means tested Benefit level (% of average Not specified income/living wage) Frequency Monthly Method of transfer Cash payment Targeting criteria •• Household has to have at least one member age 25–60, reside in the Besos District until September 2019, and be a current beneficiary of the city’s municipal social services •• Some treatment groups receive means-tested payments Coverage 1,000 households in the Besos District, the city’s poorest area; overall random, stratified sample of 2,000 households Duration 2 years: October 2017–December 2019 Source of funds Funded by the City Council and a grant provided by the European Commission’s Urban Innovative Actions program Comments •• Any municipal social services support recipients receive was supposed to be deducted from the basic income transfer •• Participants are encouraged or obliged to participate in support programs related to employment, social enterprise, housing, or community action •• Participation in the experiment is voluntary •• Budget: €13 million (US$14.3 million) Link https://basicincome.org/news/2017/08/barcelona-spain-design-minimum- income-experiment-finalized/ Busibi Village, Uganda: Eight Type of intervention Universal basic income Benefit level (absolute terms) U Sh 60,000 (US$18.25) for adults; U Sh 30,000 (US$9.13) for children, paid to mother/caretaker Benefit level (% of average 30% of income of lower-income families in Uganda income/living wage) Frequency Monthly Method of transfer Via mobile phones Targeting criteria No targeting (universal) Coverage Everyone in the village (56 adults, 88 children) Duration 2 years: January 2017–December 2018 Source of funds Eight (nonprofit) Comments Program was in addition to existing social programs Link http://eight.world/ Appendix A: UBI-Related Pilots 243 Characteristic Description Two U.S. States (unspecified as yet): Y Combinator Research Type of intervention Unconditional basic income Benefit level (absolute terms) US$1,000/month; control group members receive US$50/month Benefit level (% of average Not specified income/living wage) Frequency Monthly Method of transfer Direct deposit to a GoBank account (reloadable debit card with no overdraft fees) Targeting criteria All individuals age 21–40 whose total household income in year prior to enrollment did not exceed area median income for their county of residence Coverage •• 3,000 participants (1,000 in the two treatment groups; 2,000 in control group) •• Sample no more than 1% of people in a census tract Duration 3 years for one treatment group; 5 for the other; proposed start date: mid-2019 Source of funds Y Combinator Comments Program is in addition to existing social benefit programs (efforts under way to have beneficiaries waived from being exempted from existing benefit schemes due to enrollment in program) Link https://static1.squarespace.com/static/599c23b2e6f2e1aeb8d35ec6/t/59c3188 c4c326da3497c355f/1505958039366/YCR-Basic-Income-Proposal.pdf Stockton, California, United States: Stockton Economic Empowerment Demonstration (SEED) (Government of Stockton) Type of intervention Basic income pilot Benefit level (absolute terms) US$500 Benefit level (% of average Not specified income/living wage) Frequency Monthly Method of transfer Not specified Targeting criteria Stockton residents can qualify if they are at least 18 years old and reside in a neighborhood with a median income of US$46,033 or less; individuals who earn more than US$46,033 can still be eligible as long as their neighborhood fits the criteria Coverage 100 resident families Duration 18 months: February 2019–August 2020 Source of funds Fully funded by private donations: US$1.2 million in philanthropic funding (US$1 million from the Economic Security Project) Comments Program is in addition to existing social programs Link https://cnnmon.ie/2ugs7Ad; https://bit.ly/2hPgxG2; https://nyti.ms/2Lf82Bh Great Smoky Mountains, North Carolina, United States: Eastern Band of Cherokee Nations Type of intervention Social wealth fund/universal basic income Benefit level (absolute terms) 1996: US$595/person; 2001: US$6,000/person; 2006: US$9,000/person; 2012: US$12,000/person Benefit level (% of average Not specified income/living wage) Frequency Twice a year Method of transfer Cash payment Targeting criteria All tribe members Coverage Initially 15,000 tribe members Duration Continuous: 1996–present Source of funds Casino profits 244 Appendix A: UBI-Related Pilots Characteristic Description Comments •• Program is in addition to existing social programs •• Children receive supplements from birth; funds paid into trust fund to age 18 •• Tribal Council recently passed legislation staggering minors’ fund payouts: members will now receive US$25,000 at age 18, US$25,000 at age 21, and the remainder at age 25 Link https://nyti.ms/2P5IkBi; https://bit.ly/2zyfIvX Alaska, United States – Alaska Permanent Fund: Alaska Type of intervention Social wealth fund Benefit level (absolute terms) Up to US$2,072/person (reduced to US$1,022/person in 2016 to reflect lower commodity prices; US$1,100/person in 2017) Benefit level (% of average Varied between 1.7% of personal income in 1984 to 6.4% in 2000 income/living wage) Frequency Annual (each October) Method of transfer Check/bank deposit Targeting criteria None Coverage All state residents (distributed to 600,000 state residents in 2017; individuals sentenced for a serious crime or incarcerated in the previous year are ineligible) Duration Continuous; 1982–present Source of funds State government/sovereign fund Comments •• Program in addition to federal social assistance •• US$60.1 billion state fund established in 1976 to collect revenue from Alaska’s oil and mineral leases Link https://bit.ly/2uruizi Gary, Indiana, United States: Gary Income Maintenance Experiment Type of intervention Negative income tax (NIT) Benefit level (absolute terms) •• Guaranteed annual income levels: US$3,300 and US$4,300 •• Participants assigned randomly to one of four NIT plans. In two plans, wage and nonwage income was subject to a 40% tax rate; in the remaining two, income was taxed at a 60% rate. Two plans offered basic income supports, scaled according to family size, that were equal to slightly more than the poverty level; the other two offered basic supports, also scaled to family size, that were one-quarter less. •• NIT rate: 40% and 60% Benefit level (% of average 75% of poverty line for one group, and 100% for another; social services income/living wage) counseling and day care subsidies of 35%, 60%, and 80% provided Frequency Not specified Method of transfer Not specified Targeting criteria African American households in low-income neighborhoods, head age 18–58 with at least one dependent and income below 240% of poverty line Coverage 1,780 households (60% female-headed households; later, 125 households with incomes greater than 240% of poverty line were added) Duration 3 years: 1971–74 Source of funds U.S. federal funding (U.S. Department of Health, Education and Welfare) Comments •• Budget (1973): US$20.3 million •• All federal, state, and social security income tax liabilities were fully reimbursed for income up to the break-even point; earned income above break-even point was taxed according to federal, state, and social security tax tables •• Control families received no benefits except a small payment for their continued participation Appendix A: UBI-Related Pilots 245 Characteristic Description Link https://bit.ly/2O2aO2A; https://bit.ly/2nmUVD1; https://bit.ly/2zMpES6; https:// bit.ly/2Rhy0Yy Seattle and Denver, United States: Seattle/Denver Income Maintenance Experiments Type of intervention Negative income tax (NIT) Benefit level (absolute terms) •• Absolute benefit level not specified •• Guaranteed annual income levels: US$3,800, US$4,800, and US$5,600 •• NIT rate: 50%, 70%, and 80% Benefit level (% of average 95%, 126%, and 146% of poverty rate (1971 poverty line: US$4,000) income/living wage) Frequency Not specified Method of transfer Not specified Targeting criteria Families with at least one dependent and incomes below US$11,000 (single- headed) or US$13,000 (double-headed) (or 325% of poverty line overall) Coverage 4,801 families (2,758 in Denver and 2,043 in Seattle) Duration 6 years (9 years for some): 1970–76 (some until 1980) Source of funds U.S. federal funding (U.S. Department of Health, Education and Welfare) Comments •• Budget (1975): US$77.5 million •• Researchers obtained approval to extend the experiment for 20 years for a small group of subjects, but it was canceled in 1980, so a few subjects had a guaranteed income for about nine years, during some of which time they were led to believe they would receive it for 20 years Link https://bit.ly/2Rhy0Yy; https://bit.ly/2zMpES6 Iowa and North Carolina, United States: Rural Income Maintenance Experiment Type of intervention Negative income tax (NIT) Benefit level (absolute terms) •• Guaranteed annual income levels: US$1,741, US$2,612, and US$3,482 •• NIT rate: 30%, 50%, and 70% Benefit level (% of average 50%, 75%, and 100% of poverty line income/living wage) Frequency Not specified Method of transfer Not specified Targeting criteria Families with at least one dependent and incomes below 150% of poverty line Coverage 809 families (587 male-headed, 108 female-headed, and 114 older heads); sample at program end: 729 Duration 3 years: 1969–73 Source of funds U.S. federal funding (Office of Economic Opportunity) Comments None Link https://bit.ly/2Rhy0Yy; https://bit.ly/2zMpES6 Pennsylvania and New Jersey, United States (Trenton, Paterson-Passaic, Jersey City, Scranton): New Jersey Graduated Work Incentive Experiment Type of intervention Negative income tax (NIT) Benefit level (absolute terms) •• Guaranteed annual income levels: US$1,650, US$2,457, US$3,300, and US$4,125; $5,000 for family of 4 (1985 purchasing power parity) •• NIT rate: 30%, 50%, 70%, and 80% (for 70% and 80%, tax rate declines 2.5% per US$1,000 income) Benefit level (% of average 50%, 75%, 100%, and 125% of poverty line (1968 poverty line: US$3,800) income/living wage) Frequency Not specified Method of transfer Not specified 246 Appendix A: UBI-Related Pilots Characteristic Description Targeting criteria Households headed by able-bodied males age 18–58 with at least one dependent and income less than 150% of poverty line Coverage 1,357 households (725 experimental and 632 control) Duration 3 years: 1968–72 Source of funds U.S. federal funding (Office of Economic Opportunity) Comments None Link https://bit.ly/2zMpES6; https://bit.ly/2Rhy0Yy; https://bit.ly/2nmUVD1 Appendix B: UBI-Related Proposals Characteristic Description Australia: Henderson, 1975 Type of intervention Guaranteed minimum income Benefit level (absolute terms) Not specified Benefit level (% of average 106% of poverty line income for categorical (those who already qualified income/living wage) for any social security benefit or pension); 50–71% of poverty line for noncategorical beneficiaries Frequency Not specified Targeting criteria Income and age Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) Not specified Comments •• Family, not individual, used as unit to estimate categorical inclusion •• Poverty line set at benchmark income of $A 62.70 for September quarter 1973, which was around the value of the basic wage plus child endowment (an earlier version of family allowance), for a reference family of two adults with two children; adjustments then made for other household types •• Proposal to replace existing progressive taxation system with a proportional tax + a proposed 5% tax surcharge for incomes over $A 240 per week as of August 1973 Link None Canada: Parliamentary Budget Officer, 2017 Type of intervention Guaranteed minimum income Benefit level (absolute terms) Eligible individuals would receive an amount of Can$16,989; couples would receive Can$24,027, before deductions for any income earned Benefit level (% of average 75% of Statistics Canada’s LIM income/living wage) Frequency Monthly Targeting criteria People living on less than Can$34,000 individually, or Can$48,000 as a couple Coverage Universal (estimated 7.5 million recipients) Source of funds/implementer Federal government UBI funds (% of GDP) Can$76 billion in 2018–19; increase to Can$79.5 billion by 2022–23 Comments Replace programs costing Can$32.9 billion, so additional Can$43 billion needed Link https://bit.ly/2DYMdH4; https://bit.ly/2OSftAX 247 248 Appendix B: UBI-Related Proposals Characteristic Description Quebec, Canada: Expert Committee on Guaranteed Minimum Income, 2017 Type of intervention UBI Benefit level (absolute terms) 1. Can$1,637/adult; Can$737/child 2. Can$878/individual 3 & 4. Can$5,832/individual Benefit level (% of average 1 & 2: Not specified income/living wage) 3 & 4: Assistance equal to 50% of social assistance paid to a couple; current recipients of last-resort financial assistance who receive more than Can$5,832 would continue to receive amounts corresponding to difference between current benefit and universal allowance of Can$5,832 Frequency Not specified Targeting criteria 1. 0–64 years 2. 18–64 years 3. 19–64 years 4. 18 years and over Coverage Statewide Source of funds/implementer Government UBI funds (% of GDP) 1. Can$9.6 billion gross cost 2. Can$4.4 billion gross cost 3. Can$29.2 billion gross cost 4. Can$38.4 billion gross cost Comments 1, 2 & 3: Financing: Redistribution of current assistance, including family assistance, replacing almost all current support (Organisation for Economic Co-operation and Development approach) 4. Replaces part of current support Link http://www.gouv.qc.ca/EN/RevenuMinimumGaranti/Documents/ Rapportfinal_RMG_volume1ENG_V2.pdf Quebec, Canada: Provincial Government, 2017 Type of intervention Guaranteed minimum income Benefit level (absolute terms) •• Can$73/month (will increase to Can$440/month by 2023, bringing annual guaranteed income to Can$18,029) •• Single Quebecers who receive social benefits will see it increased by Can$180 per year by 2018; in 2021, they will receive about Can$540 annually Benefit level (% of average Poverty line for an individual: Can$18,000/year income/living wage) Frequency Monthly Targeting criteria Means tested Coverage Statewide (as many as 84,000 people estimated to become beneficiaries) Source of funds/implementer Government UBI funds (% of GDP) Not specified Comments Part of a larger, multiyear Can$3 billion plan including additional funding for existing programs (e.g., Can$286 million for social housing projects, Can$580 million for social benefits, Can$40 million to create more kindergarten classes in low-income areas, and Can$300,000 for mental health initiatives) Link https://bit.ly/2NYiW3X; https://tgam.ca/2CSlSoJ Appendix B: UBI-Related Proposals 249 Characteristic Description Ontario, Canada: Segal (1), 2016 Type of intervention Guaranteed minimum income Benefit level (absolute terms) Group 1: Can$1,320—75% of LIM; disabled receive additional Can$500 on top of state disability support, lower taxes on their income until their benefit from basic income is 0 Group 2: 75% of LIM—additional income is charged higher tax rate until net benefit from basic income is 0 Group 3: 75% of LIM—income taxed back at a higher rate Group 4: no change in terms of income support Benefit level (% of average 75% of LIM income/living wage) Frequency Monthly Targeting criteria Participants selected randomly from population age 18–65, with their primary residence in the chosen site for at least one year Coverage Three sites distributed across the province Source of funds/implementer Provincial government UBI funds (% of GDP) Not specified Comments Existing social safety programs (disability support and unemployment benefits) replaced for all recipients of the basic income, but kept in place for control groups and nonparticipants Link https://bit.ly/2DQk6tN Ontario, Canada: Segal (2), 2016 Type of intervention UBI Benefit level (absolute terms) Group 1: 75% of LIM; disabled receive an additional Can$500 on top of state disability support, lower taxes on their income until net benefit from basic income is 0 Group 2: 75% of LIM—additional income is charged higher tax rate until net benefit from basic income is 0 Group 3: 75% of LIM—income taxed back at a higher rate Group 4: no change in terms of income support Benefit level (% of average 75% of LIM income/living wage) Frequency Monthly Targeting criteria •• Population age 18–65, with their primary residence in the chosen site for at least one year •• By design, all adults who meet age and residency eligibility criteria for the pilot and who live in the saturation site should be able to receive top-up benefits should their income drop below the relevant threshold Coverage One “saturation site”—universal coverage Source of funds/implementer Provincial government UBI funds (% of GDP) Not specified Comments •• Benefit (which would completely replace Ontario Works and Ontario Disability Support Program) would be clawed back as a percentage of their earned income, according to a predetermined tax rate, until the net benefit received is equal to Can$0, after which their earned income would be taxed at rate prescribed by existing tax schedule •• Replaces existing unemployment benefits and disability benefits program Link https://bit.ly/2xSmGKo 250 Appendix B: UBI-Related Proposals Characteristic Description France: Hamon/French Economic Observatory (OFCE), 2017 Type of intervention Negative income tax (framed as UBI) Benefit level (absolute terms) €600/month for people without resources; payments lowered until income reaches 1.9 times minimum wage (€9.76/hour in 2017); 27.4% of total income of a taxable household to be subtracted from monthly disbursement Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria Everyone age 18–64 Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) Gross cost: €51 billion Comments For the poorest households, UBI replaces income supplement for the working poor and working tax credit; and calculation of social benefits (housing and family allowances, disabled adult allowance, scholarships, etc.) not modified, as their amounts are included in resources used to calculate UBI Link https://bit.ly/2QoEOCe; https://bit.ly/2yaHLir India: Rahul Gandhi (leader of India’s National Congress Party), 2019 Type of intervention Guaranteed minimum income Benefit level (absolute terms) Up to Rs 72,000/year (up to Rs 6,000/month, to bring households to minimum income line of Rs 12,000/month) Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria Poorest 20% of households Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) Estimated cost of Rs 3.6 lakh crore/year (~US$50 billion), roughly 1.9% of GDP Comments •• Party’s manifesto states it intends to implement program as a joint scheme of the central and state governments, funded through new revenues and rationalization of expenditure; current merit subsidy schemes intended to achieve specific objectives will be continued •• Estimated cost < 1% of GDP in Year 1; < 2% of GDP in Year 2 and thereafter Link https://manifesto.inc.in/en/nyay.html India: Felman, Paul, Sharan, and Subramanian, 2019 Type of intervention Quasi–universal basic rural income Benefit level (absolute terms) ~Rs 18,000/year or ~Rs 1,500/month (inflation-adjusted) to each household Benefit level (% of average One-third of current consumption of poorest 40% income/living wage) Frequency Monthly Targeting criteria Poorest 75% of rural households Coverage Rural areas nationwide Source of funds/implementer State and central governments UBI funds (% of GDP) 1.3% of GDP = Rs 2.64 lakh crore (2019–20 prices) Comments Central government should offer to finance half of each transfer, primarily by cutting or phasing out ineffective agricultural schemes (e.g., interest rate subsidy for crop loans, state insurance of agricultural income scheme, fertilizer subsidy); states will likely look to cut subsidies (power and water) and other wasteful schemes Link http://bit.ly/2PzVlEr Appendix B: UBI-Related Proposals 251 Characteristic Description Sikkim State, India: Sikkim Democratic Front, 2019 Type of intervention UBI Benefit level (absolute terms) Not specified Benefit level (% of average Not specified income/living wage) Frequency Not specified Targeting criteria Either all holders of Sikkim subject certificates, or all Indian citizens resident in the state Coverage Statewide Source of funds/implementer State government (using hydropower and tourism revenues) UBI funds (% of GDP) Not specified Comments Declared in Party Manifesto in March 2019, in the run-up to the assembly election, with an aim of implementing the scheme by 2022 Link http://bit.ly/2UXHuh8 India: Himanshu, 2017 Type of intervention UBI Benefit level (absolute terms) Rs 1,000/month Benefit level (% of average Around half of the poverty line income/living wage) Frequency Monthly Targeting criteria Universal coverage within groups excluded from the labor market—widows, the elderly, and the disabled Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) < 0.5% Comments Increase coverage of National Social Assistance Program, which presently provides Rs 200/month (amount has not been revised since 2006 when it was fixed, barring a couple of states) to people in target categories who are also classified as below–poverty line households Link http://www.ideasforindia.in/topics/poverty-inequality/a-proposal-for- universal-basic-services.html India: Economic Survey 2016–17, 2017 Type of intervention Quasi-UBI Benefit level (absolute terms) Rs 7,620/year (US$120) Benefit level (% of average 52.9% of Tendulkar poverty line (2015–16 prices) income/living wage) Frequency Not specified Targeting criteria Bottom 75% of income distribution Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) 4.9% Comments •• Prerequisite is that all Indians have Aadhaar identification and are financially included—initially a quasi-UBI is proposed •• Roll back social sector programs: 2.07% of GDP •• Implicit middle-class subsidies: 1.05% of GDP •• Top 10 centrally sponsored schemes: 1.38% of GDP Link https://bit.ly/2OwVkmD; https://bit.ly/2DSs45x 252 Appendix B: UBI-Related Proposals Characteristic Description India: Bardhan, 2016 Type of intervention UBI Benefit level (absolute terms) Rs 10,000/year Benefit level (% of average 83.4% of Tendulkar poverty line (2015–16 prices) income/living wage) Frequency Not specified Targeting criteria All Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) 10% Comments Inflation-indexed annual transfer of Rs 10,000—75% of India’s 2014–15 poverty line—to every Indian citizen; roll back nonmerit subsidies: 9% of GDP; eliminate corporate tax holidays and exemptions: 3% of the GDP Link https://bit.ly/2E2aeNR; https://bit.ly/2OwVkmD India: Joshi, 2016 Type of intervention UBI Benefit level (absolute terms) Rs 3,500/head of household/year (Rs 17,500/family/year) Benefit level (% of average Annual disbursement is 20% of Tendulkar poverty line, since the average income/living wage) family below the poverty line earns 80% of Tendulkar poverty line; disbursement is inflation adjusted Frequency Not specified Targeting criteria All Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) 3.5% (alternatively, 2.5% and 1.9% of GDP if transfers are paid only to 67% and 50% of the population, respectively) Comments Hints at using Aadhaar identification technology for disbursements; roll back nonmerit and food subsidies: 8.5% of GDP; savings from tax exemptions: 1.5% of GDP; privatization of public sector enterprises: 1% of GDP; taxing agricultural incomes: 0.5% of GDP; remove dysfunctional social welfare schemes: 0.5% Link https://bit.ly/2OwVkmD; https://bit.ly/2zNaDzu India: Ghatak, 2016 Type of intervention UBI Benefit level (absolute terms) Rs 13,432/year Benefit level (% of average Poverty-level income income/living wage) Frequency Not specified Targeting criteria All Coverage All adults (69% of the population) Source of funds/implementer Government UBI funds (% of GDP) 11% Comments Roll back subsidies going to the nonpoor: 9% of GDP; raise additional taxes Link https://bit.ly/2OwVkmD; https://bit.ly/2DP6teb Appendix B: UBI-Related Proposals 253 Characteristic Description India: Banerjee, 2016 Type of intervention UBI Benefit level (absolute terms) Rs 13,000/year (Rs 250/week) Benefit level (% of average Not specified income/living wage) Frequency Weekly Targeting criteria All Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) 11% Comments Predicated on use of Aadhaar identification; replaces welfare schemes such as the Public Distribution System and the National Rural Employment Guarantee Scheme Link https://bit.ly/2OwVkmD; https://bit.ly/2DR5Kte India: Ray, 2016 Type of intervention Universal basic share Benefit level (absolute terms) Rs 10,000–Rs 13,000 Benefit level (% of average 75–100% of poverty line income/living wage) Frequency Annual Targeting criteria All Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) 9–12% Comments Commit universal basic share, fixed fraction of GDP: 9–12%; this would amount to around 25% of government expenditure Link https://bit.ly/2OwVkmD; https://bit.ly/2QoUd5D; https://bit.ly/2P1cTIq India: Khera, 2016 Type of intervention UBI Benefit level (absolute terms) Pensions (for the elderly and widows)—Rs 12,000; maternity entitlements per child—Rs 6,000 Benefit level (% of average Not specified income/living wage) Frequency Monthly for pensions Targeting criteria All elderly, widows, disabled persons (approximately 10% of the population), and pregnant women (approximately 26 million children born annually) Coverage All in the identified categories Source of funds/implementer Government UBI funds (% of GDP) 1.5% Comments A phased approach to UBI, starting with pensions and maternity entitlements; replaces social security pensions (both at central and state levels) and existing maternity benefit schemes Link https://bit.ly/2RkZ3SL; https://bit.ly/2OwVkmD 254 Appendix B: UBI-Related Proposals Characteristic Description The Islamic Republic of Iran: Karrubi, 2005 Type of intervention UBI Benefit level (absolute terms) US$50/month Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria None Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) Not specified Comments Financed from oil exports, savings in the national budget, reduced consumption of gasoline, and replacement of some other transfers Link https://bit.ly/2ycuKVQ New Zealand: Rankin, 1998 Type of intervention Guaranteed minimum income Benefit level (absolute terms) •• Refundable tax credit (universal tax credit) of $NZ 123 per week, which is equal to both the age 18–24 rate of unemployment benefit and married rate of unemployment benefit •• A general means-tested benefit (GMTB) of $NZ 70/week (same level as present-day benefits) that will abate at a rate of $NZ 0.25 per dollar of gross privately sourced income; GMTB is reduced 25% for every dollar earned Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria None Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) Provides for a fund of $NZ 10 billion to cover GMTB payments and existing subsidies; implicit subsidies include corporate tax exemptions and tax avoidance as well as tax exemptions on rent accruing to mortgage-free owner-occupied homes. If all explicit and implicit subsidies could be eliminated, GMTB payments averaging just over $NZ 70/adult could be paid. Comments •• Raise income tax rate (including corporate tax) from 33% to a flat rate of 39% •• GMTB would incorporate all present means-tested benefits (excluding youth and married unemployment benefit which would be fully replaced by the universal tax credit; and excluding student allowances): domestic purposes benefit, invalid’s benefit, single adult unemployment benefit, New Zealand superannuation, family support, independent family tax credit, guaranteed minimum family income, accommodation supplement, students accommodation allowance, special benefit Link http://rankinfile.co.nz/rf98_UBIat39percent.html Scotland: RSA, 2018 Type of intervention UBI Benefit level (absolute terms) 1. Weekly individual benefits: 0–4 first child: £84.5; 0–4 additional children: £67; 5–15: £57.9; 16–24: £57.9; 25–64: £73.1; 64+: £155.6 2. Weekly individual benefits: 16–24 (from age 18): £57.9; 25–64 (until age 31): £73.1; 55–64: £73.1 3. Weekly individual benefits: 0–4 first child: £84.5; 0–4 additional children: £67; 5–15: £57.9; 16–24: £57.9; 25–64: £73.1; 64+: £155.6 4. Multiple treatments in this experiment: 500 receive only payments; 150 receive extra money for volunteering + payments; 100 are engaged with entrepreneurial landscape + payments; 175 are temporarily placed in a public job + payments; 75 get rent support + payments Appendix B: UBI-Related Proposals 255 Characteristic Description Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria 1. None 2. Youth (age 18–30) and those age 55–64; further characteristics of this cohort not specified, but employment status and gender suggested as criteria 3 & 4. All residents of selected community Coverage 1. Universal (in a midscale test site; could be a portion of a rural area/town/ city of 1,000 people) 2. Randomly selected treatment group within targeted cohort (1,000 people) 3. Universal (community of 250 recipients) 4. Universal (community of 1,000 recipients) Source of funds/implementer Government UBI funds (% of GDP) 1. Annual budget: £4.4 million 2. Annual budget: £3.61 million 3. Annual budget: £1.1 million (as high as £1.3 million) 4. Not specified Comments Payments are updates of those laid out in Creative Citizen, Creative State to suit 2016/17; pegged to payment levels for, in this case, Jobseeker’s Allowance Link https://bit.ly/2NYivXr Switzerland: Basic Income Campaign, 2016 Type of intervention UBI Benefit level (absolute terms) Sw F 2,500/month for adults; Sw F 625 for each child Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria Age-based Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) ~4% Comments 77% of voters rejected the plan in June 2016; cost estimated at Sw F 25 billion Link https://bbc.in/2xYi0CJ Pavlograd, Ukraine: Movchan (head of city), 2018 Type of intervention Basic income pilot Benefit level (absolute terms) Equivalent of €100/month Benefit level (% of average ~35% of average monthly salary in Ukraine income/living wage) Frequency Monthly Targeting criteria Adult citizens only Coverage 2,000 randomly selected Pavlograd adult citizens Source of funds/implementer Donors UBI funds (% of GDP) Not specified Comments Announced by head of city in November 2018. City government wants to run a 24-month experiment to measure effect of unconditional cash transfers on labor market, objective and subjective well-being, financial health, and changes in mental and physical health, among other social indicators. City will be responsible for pilot management but is seeking charitable donors to fund cost of transfers. Link https://basicincome.org/news/2018/12/ukraine-basic-income-experiment- has-started-being-prepared-in-ukraine/ 256 Appendix B: UBI-Related Proposals Characteristic Description United Kingdom: Standing, 2019 Type of intervention Basic Income Pilot Benefit level (absolute terms) 1. Model A: £100/week per adult; £50/week per child + additional separate benefits for people with disabilities 2. Model B: £70/week per adult; £20/week per child on top of child benefit 3. Model C: £50/week tax-free per adult 4. Model D: value of current state benefit, with existing conditions removed 5. Model E: Not specified Benefit level (% of average Not specified income/living wage) Frequency Not specified Targeting criteria 1. Legal residents 2. Legal residents (working-age adults + children) 3. Legal residents 4. Sample of existing welfare recipients 5. Homeless people Coverage 1. Whole pilot locality 2 & 3. Preferably everyone in pilot locality 4. Pilot locality 5. Four randomly selected pilot localities Source of funds/implementer Government UBI funds (% of GDP) 1, 2, 3 & 4. Up to £5 million 5. Not specified Comments 1. Replaces existing means-tested benefits, except housing benefit 2. Means-tested benefits would be left in place, and basic incomes would be added to means taken into account in their calculation, so each recipient household’s means-tested benefits would automatically be adjusted downward/upward by changes in net earnings brought about by tax code change 3. Implemented as a supplement to existing state benefits, but basic income amount would not be taken into account when determining access to means-tested benefits 4. Sample of welfare-recipient adults would have existing conditions for entitlement to existing means-tested benefits removed, so as to make benefits closer to a basic income 5. Refinement of an approach taken in London in which cash grants were given to the homeless instead of various other measures. It should not be made a national policy, but the results could be taken into account in formulating and implementing a national basic income system. Link https://www.progressiveeconomyforum.com/wp-content/uploads/2019/05/ PEF_Piloting_Basic_Income_Guy_Standing.pdf United Kingdom: New Economics Foundation, 2019 Type of intervention UBI Benefit level (absolute terms) £48.08/week in England, Northern Ireland, and Wales; £45.68/week in Scotland Benefit level (% of average Not specified income/living wage) Frequency Weekly Targeting criteria All adults previously benefiting from personal tax allowance = all adults with a U.K. national insurance number earning less than £125,000 per year Coverage Nationwide Appendix B: UBI-Related Proposals 257 Characteristic Description Source of funds/implementer Government (using additional taxes from abolishing personal tax allowance) UBI funds (% of GDP) £126.8 billion, but fiscally neutral since it abolishes previous tax benefit program Comments Proposed as replacement for existing personal tax allowance (£12,500/ year); another program component would be to restore current child benefit scheme to its real-terms 2010/11 value Link https://neweconomics.org/2019/03/nothing-personal United Kingdom: Reed and Lansley – Scheme 1, 2016 Type of intervention UBI Benefit level (absolute terms) Pensioners: £41; other adults over age 25: £61; adults under age 25: £51; children: £49 Benefit level (% of average Not specified income/living wage) Frequency Weekly Targeting criteria Age and pension status Coverage Universal Source of funds/implementer Government UBI funds (% of GDP) £0.7 billion Comments •• Existing means-tested and nonmeans-tested programs remain; UBI is taken into account when calculating qualification for means-tested benefits; child benefit is replaced; state pension paid on top of UBI •• Income tax personal allowance abolished—higher income tax rates (basic: 23%; higher: 43%; top: 48%) •• Lower earnings limit reduced to zero; national insurance contributions levied at 12% on all earnings Link https://bit.ly/1taL5GB United Kingdom: Reed and Lansley – Scheme 2, 2016 Type of intervention UBI Benefit level (absolute terms) Pensioners: £51; other adults over age 25: £71; adults under age 25: £61; children: £59 Benefit level (% of average Not specified income/living wage) Frequency Weekly Targeting criteria Age and pension status Coverage Universal Source of funds/implementer Government UBI funds (% of GDP) Just under 0.5% of GDP (£8.2 billion) Comments •• Existing means-tested and nonmeans-tested programs remain; UBI is taken into account when calculating qualification for means-tested benefits; child benefit is replaced; state pension paid on top of UBI •• Income tax personal allowance abolished—higher income tax rates (basic: 25%; higher: 45%; top: 50%) •• Lower earnings limit reduced to 0; national insurance contributions levied at 12% on all earnings Link https://bit.ly/1taL5GB 258 Appendix B: UBI-Related Proposals Characteristic Description Newark, United States: Baraka (mayor), 2019 Type of intervention UBI pilot Benefit level (absolute terms) Not specified Benefit level (% of average Not specified income/living wage) Frequency Not specified Targeting criteria Not specified Coverage Not specified Source of funds/implementer Not specified UBI funds (% of GDP) Not specified Comments In March 2019 state-of-the-city address, Newark Mayor Ras Baraka announced decision to create a task force and pilot to study whether program is possible Link http://bit.ly/2W8L2JP Chicago, U.S.: Pawar (Chicago alderman), 2018 Type of intervention Unconditional cash transfer (framed as a UBI) Benefit level (absolute terms) US$500/month Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria Not specified Coverage 1,000 families in Chicago Source of funds/implementer Unclear UBI funds (% of GDP) Not specified Comments In addition to monthly payments, program would also adjust the earned income tax credit program to “smoothen” it for the chosen 1,000 families (monthly payments instead of one annual payment) Link https://bit.ly/2Lr3FU9 United States: Yang, 2018 Type of intervention UBI Benefit level (absolute terms) US$1,000/month Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria Everyone age 18–64 Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) Not specified Comments 10% value-added tax on corporations will generate US$800 billion; citizens already receiving government benefits would choose between existing welfare and US$1,000/month payments Link https://cnb.cx/2x1BBRj; https://bit.ly/2NWUnod United States: Widerquist, 2017 Type of intervention UBI Benefit level (absolute terms) 1. US$12,000 per adult; US$6,000 per child 2. US$20,000 per adult; US$10,000 per child Appendix B: UBI-Related Proposals 259 Characteristic Description Benefit level (% of average 1. Equal to 2015 Census Bureau poverty line for person living alone income/living wage) 2. Slightly above 160% of 2015 Census Bureau poverty level Frequency Not specified Targeting criteria None Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) 1. 2.95% of GDP (US$539 billion/year) 2. Just under 10% of GDP (US$1.816 trillion/year) Comments Proposal urges policy makers to seriously consider paying for UBI at least partially with a tax increase targeted at wealthy people; US$0.50 decrease in UBI payment with each US$1 increase in recipient earnings Link https://bit.ly/2Qp39bq United States: Khanna, 2017 Type of intervention Negative income tax Benefit level (absolute terms) Childless workers’ maximum credit: US$3,000 (currently US$500); families would see maximum credit rise from US$6,318 to US$12,131, depending on their income and number of children Benefit level (% of average Not specified income/living wage) Frequency Annual Targeting criteria No credits for families of 3 or more with maximum annual income of US$75,940, or single individuals earning US$37,500 or more Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) Cost: US$1.4 trillion over 10 years Comments Bill is framed as significantly increasing earned income tax credit scheme already in place Link https://bayareane.ws/2zNaxI2; https://bit.ly/2f0UACZ United States: Nikiforos, Steinbaum, and Zezza, 2017 Type of intervention UBI Benefit level (absolute terms) 1. US$500/mo 2. US$1,000/mo Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria All adults Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) 1. US$1,495 billion 2. US$2,990 billion Comments 1. Simulations assume UBI will be implemented gradually over 4 years; two scenarios considered: in fiscally neutral scenario, tax rates increase (5% up for households in 4th quintile, 11–26% up for those in top quintile); other scenario is purely debt-inducing and does not increase taxes 2. Considered two scenarios: in fiscally neutral scenario, tax rates increase (11% up for those in middle quintile, 12% up for households in 4th quintile, 21–35% up for those in top quintile); other scenario is purely debt-inducing and does not increase taxes Link https://bit.ly/2MSc2Z0 260 Appendix B: UBI-Related Proposals Characteristic Description United States: Murray, 2016 Type of intervention UBI Benefit level (absolute terms) US$13,000/year (US$3,000 earmarked for compulsory medical insurance); minimum of US$6,500 for everyone; reduced benefits for those earning US$30,000/year or more Benefit level (% of average Not specified income/living wage) Frequency Monthly Targeting criteria Every person age 21 and older Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) Not specified Comments Replaces social security, Medicare, Medicaid, food stamps, supplemental security income, housing subsidies, welfare for single women, and every other kind of welfare and social services program, as well as agricultural subsidies and corporate welfare Link https://bit.ly/2tWK6h6; https://bit.ly/2v9l0Jt United States: Stern, 2016 Type of intervention UBI Benefit level (absolute terms) US$12,000/year for everyone, plus top-up to ensure that every senior age 65 and up receives at least US$12,000 a year in social security Benefit level (% of average Just above the poverty line income/living wage) Frequency Monthly Targeting criteria Every person age 18–64 Coverage Nationwide Source of funds/implementer Government UBI funds (% of GDP) Not specified Comments None Link https://bit.ly/2P7j0uH United States: President Nixon (Family Assistance Plan), 1969 Type of intervention Negative income tax Benefit level (absolute terms) Family of 4 receives US$1,600/year if without income (~US$10,000 in 2016); slightly lower amount until its income reaches US$3,920 Benefit level (% of average Not specified income/living wage) Frequency Annual Targeting criteria Means tested Coverage Nationwide Source of funds/implementer Federal government UBI funds (% of GDP) US$1.9 billion/year Comments Bill passed in the House but died in the Senate Link https://bit.ly/2maGKmj; https://bit.ly/1TuKJ2J; https://bit.ly/2NYERs2 NOTE: GDP = gross domestic product; LIM = low-income measure; UBI = universal basic income. Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers TABLE C.1  Comparative Impact Evaluations Conditional Transfers Outperform Unconditional Burkina Faso RCT (2008–10) Health visits (under age 6) Conditions School enrollment + 90% attendance (age 7–15) Type Hard CCT versus UCT Health: CCT led to a 49% increase in number of routine preventive visits compared to control group; UCT had no significant impact (Akresh, de Walque, and Kazianga 2012). CCT reduced the level of C-reactive protein (a biomarker for infections) by 29% relative to the control; UCT had no significant impact. At Round 2, CCT impact on probability of recent illness was nearly double that of UCT (reduction of 7.3 versus 3.9 percentage points). CCT also had some impact on arm circumference for age and height for age scores; UCT had no significant impact Outcomes (Akresh, de Walque, and Kazianga 2016). Education: At Round 3, UCT impact on school enrollment was 5.9 percentage points lower than CCT increase of 14.7 percentage points. Much of this difference related to transfers’ effects on marginal children (those less likely to be favored for schooling). For example, CCT increased girls’ enrollment by 45%; UCT did not have a significant impact (Akresh, de Walque, and Kazianga 2016). Conclusion Hard CCT outperforms UCT South Africa (Small-Scale) RCT Conditions School enrollment + 80% attendance or one sexual health clinic visit Type Hard CCT versus UCT Health: Higher proportion of clinic visits in clinic CCT group (64%) compared to other two study arms (26% in UCT group and 24% in school CCT group); there was no difference in sexual behaviors (Delany-Moretlwe and Brahmbhatt 2014) Outcomes Education: 75% of adolescents in school CCT group had missed at least one day of school, compared to 62% in UCT group and 51% in clinic CCT group (Delany-Moretlwe and Brahmbhatt 2014) Conclusion Hard CCT increases usage of “dreaded” health services Zimbabwe RCT (Jan 2010-Jan 2011) Birth certification; up-to-date vaccinations; biannual growth monitoring (under age 5) Conditions 90% school attendance (age 6–17) Two-thirds attendance at local parenting skills classes Type Soft CCT versus UCT 261 262 Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers Health: Proportion of infants with birth certificates increased by 16.4% in CCT group and 1.5% in UCT group, relative to the control; neither transfer significantly increased vaccination record completion (Robertson et al. 2013) Education: Both transfers increased the proportion of children with above 80% school attendance in roughly equal measures (7.2–7.6% on average), except for least poor quintile Outcomes where only CCT increased schooling. CCT participants had 0.69 lower odds of repeating previous school grade compared to control group, whereas UCT participants saw no reduction in grade repetition (Fenton et al. 2016). Child labor: CCT reduced children’s hours of paid work by 41% (0.31 hours per week) relative to control; UCT reduced this figure by 18% (0.15 hours per week) (Fenton et al. 2016) Conclusion Soft CCT outperforms UCT Bangladesh RCT (2012–14) Conditions Attendance at behavior change communication (BCC) sessions on nutrition and health Type Soft CCT versus UCT Mothers’ knowledge of iron deficiency improved by 9.2–11.9 percentage points with transfer + BCC; the same transfer alone had no impact. Maternal awareness of multiple-micronutrient powders was 17.5–22.1 percentage points higher with transfer + BCC than with transfer Outcomes alone. Probability that child had ever consumed multiple-micronutrient powders was 19.4 percentage points higher for cash + BCC than cash alone, and 8 percentage points higher for food + BCC than food alone (Hoddinott, Ahmed, and Roy 2018). Conclusion Soft conditional transfer outperforms unconditional transfer Brazil (Bolsa Escola) Conditions 85% school attendance Type Soft CCT versus general income Simulation predicts that UCT would not change non-attendance rate of poor children (age Outcomes 10–15), while CCT would reduce non-attendance by 5.2 percentage points (Bourguignon, Ferreira, and Leite 2003) Conclusion Soft CCT outperforms UCT Lesotho (Child Grant Programme) Conditions None Type Labeled cash transfer versus general income Compared to equivalent-size increase in general income, labeled cash transfer resulted in disproportionate increases in child-related expenditure. Increase in education expenditure Outcomes with transfer was 3.2 times amount typically seen with equivalent-size increase in general income (Pace et al. 2016). Conclusion Labeled cash transfer outperforms UCT Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers 263 No Clear Winner Malawi RCT (2008–10) Conditions School enrollment and 80% attendance Type Hard CCT versus UCT Education: At end of two-year program, UCT group’s decline in dropout rate was only 43% as large as CCT impact. CCT led to significantly improved test scores for English (0.14 SD higher), math (0.12 SD higher), and cognitive ability (0.174 SD higher); UCT had no significant effect (Baird, McIntosh, and Özler 2011). Health: At end of two-year program, UCT participants’ likelihood of pregnancy was 34% lower and marriage 48% lower than in control group; CCT had no significant impact. Difference was driven by impact on girls who were likely to drop out of school at baseline (Baird, McIntosh, and Özler 2011). Apart from probability of pregnancy, no significant differences between CCT and UCT impacts on sexual behavior or sexually transmitted infections (Baird et al. 2012). Mental health: Among baseline schoolgirls, UCT led to a 38% reduction in participants’ Outcomes psychological distress; CCT led to only a 17% reduction (Baird, de Hoop, and Özler 2013) Two years after program end, Baird, McIntosh, and Özler (2016) find the following: •• For baseline school dropouts (a narrow subset of CCT participants), CCT led to sustained improvements relative to control group, increasing school attainment by 0.6 years; reducing incidence of marriage and pregnancy by 10.7 and 4.0 percentage points, respectively; increasing age at marriage and first birth by 0.43 and 0.27 years respectively; reducing total number of live births by more than 10%. •• For baseline schoolgirls, neither CCT nor UCT sustained earlier positive impacts. UCT’s strong reduction in pregnancy and marriage rates during program were reversed immediately after program. But children of UCT beneficiaries born during program did have lower stunting (adjusted direct effect estimated at 0.523 SD higher height for age z-score). Mixed: hard CCTs have adverse consequences for some, but higher impact for children Conclusion already out of school at baseline Mexico (PROGRESA – Programa de Educación, Salud y Alimentación) School enrollment + 85% school attendance for children age 8–16 Conditions Attendance at monthly health seminars Routine preventive health checkups for all family members Type Hard CCT versus general income Outcomes No difference in way transfer is spent relative to general income (Handa et al. 2009) Conclusion Hard CCT same as UCT Slovenia (Otroski Dodatek) Conditions None Type Labeled versus general income No difference between expenditure of labeled cash transfer income and general income Outcomes (Edmonds 2002) Conclusion Labeled cash transfer same as UCT Harder Conditions Outperform Softer Brazil (Bolsa Família) 85% school attendance for children age 6–15 (75% for children age 16–17) Conditions Nutrition checkups twice/year and up-to-date vaccination records for children under 7 Pre- and postnatal monitoring Type Hard versus soft CCT School dropout rate estimated to be 1.8 percentage points lower in municipalities that strictly enforce conditionalities (threaten loss of benefits), and 1.5 percentage points higher in municipalities that support noncompliant households with visits from program officials (de Janvry, Finan, and Sadoulet 2006) Outcomes Models by Paiva et al. (2016) predict that (1) a municipality with no attendance monitoring would have a 3.3–5.6 percentage point higher dropout rate and an 8.8–16.4 percentage point lower grade progression rate, relative to a municipality with average (90%) monitoring; and (2) low monitoring (87%) would increase dropout rate by 0.3–0.5 percentage points and reduce grade progression by 0.7–1.3 percentage points, relative to high (94%) monitoring. 264 Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers Conclusion Hard outperforms soft CCT Mexico (PROGRESA – Programa de Educación, Salud y Alimentación) School enrollment and 85% attendance for children age 8–16 Conditions Attendance at monthly health seminars Routine preventive health checkups for all family members Type Hard CCT versus labeled cash transfer Children in beneficiary households that did not receive conditionality monitoring form were 7.2 percentage points less likely to enroll in school and 16 percentage points less likely to Outcomes transition from primary to secondary school relative to households that received the form (de Brauw and Hoddinott 2011) Conclusion Hard CCT outperforms labeled cash transfer Colombia (Familias en Acción) Preventive health visits (for children under age 7 born before registration) Conditions At least 80% school attendance for children age 7–18 Type Hard CCT versus labeled cash transfer Preventive health visits for children not covered by conditionality requirement were estimated Outcomes to be 50% lower than for children covered by requirement (Attanasio, Oppedisano, and Vera- Hernández 2015) Conclusion Hard CCT outperforms labeled cash transfer Ecuador (Bono de Desarrollo) Conditions School enrollment Type Soft CCT versus labeled cash transfer Education: Transfers increased probability of school enrollment by 13 percentage points among households that believed it to be conditional (conditioned households), but there was no significant impact among those that thought it was unconditional (unconditioned households (Schady and Araujo 2006) Child labor: Probability of children doing recent paid or unpaid work/household labor was Outcomes not different between conditioned and unconditioned households (Edmonds and Schady 2012), but the decrease in number of hours children worked per week was much more pronounced in conditioned than unconditioned households (5.92 versus 0.024 fewer hours). Transfers also reduced children’s probability of full-time work (by 7.8 percentage points) in conditioned households; no reduction was seen in unconditioned households (Schady and Araujo 2006). Conclusion Soft CCT outperforms labeled cash transfer Honduras (Bono 10,000) Regular health visits for children under age 6 (and pregnant/nursing mothers) if children have Conditions no older siblings School enrollment + 80% attendance of at least one 6- to 18-year-old child Type Soft CCT versus labeled cash transfer Program significantly increased school enrollment (by 8.3 percentage points), reduced child labor (by 6 percentage points), and increased likelihood of recent preventive health Outcomes service usage (by 7 percentage points) only for children directly covered by conditionality requirements (Benedetti, Ibarrarán, and McEwan 2015) Conclusion Soft CCT outperforms labeled cash transfer No Difference between Harder and Softer Conditions Paraguay (Tekoporã) Conditions School attendance (85%), regular health center visits, immunization Type Soft CCT versus labeled cash transfer Beneficiaries who were aware of health and education conditions did not have significantly Outcomes better results than beneficiary households that did not know about the conditions (Teixeira et al. 2011) Conclusion Labeled same as soft CCT Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers 265 Softer Conditions Outperform Harder Morocco RCT (2008–10) Conditions School enrollment and attendance (no more than 4 days absent per month) Type Hard CCT versus labeled cash transfer Education: CCT impact on school participation was 2 percentage points lower than labeled cash transfer (LCT) increase of 7.3 percentage points. Relative to control group rate (14.7%), LCT almost doubled re-enrollment of dropouts (to 27.2%); CCT only increased it by half this Outcomes amount (to 20.9%). Significant difference between LCT and CCT participants’ math test scores (LCT increased standardized test scores by 11% of a standard deviation in control group; CCT had no significant impact) (Benhassine et al. 2013). Cost: LCT administrative costs more than one-fourth lower than CCT’s (Benhassine et al. 2013) Conclusion LCT outperforms hard CCT Mozambique RCT (2016–17) Conditions 90% school attendance Type Information versus hard CCT Estimated effect of simple information treatment on school attendance was as much as 54% Outcomes of effect of child incentive treatment and 75% of effect of CCT (de Walque and Valente 2018) Conclusion Information more important than conditions NOTE: CCT = conditional cash transfer; RCT = randomized controlled trial; SD = standard deviation; UCT = uncon- ditional cash transfer. TABLE C.2  Systematic and Literature Reviews Baird et al. (2014) Focus area Education Type Systematic review •• Both UCTs and CCTs improve school enrollment and attendance, with no significant difference between the two groups •• Hard CCTs—programs that are explicitly conditional, monitor compliance, and penalize Findings noncompliance—have substantively larger effects than either UCTs or soft CCTs (60% improvement in odds of enrollment versus 18–25% improvement) •• None of the programs significantly affect test scores Conclusion Hard CCT better than soft CCT or UCT Hunter et al. (2017) Focus area Health: maternity service use Type Systematic review •• CCTs that included among their conditionalities uptake of antenatal care services appear to have had an impact on proportion of women receiving multiple antenatal checkups, but findings were less clear with respect to receipt of any antenatal care, or on the uptake Findings of other maternity care services in continuum including childbirth and postnatal care (not included as conditionalities) •• Only published study identified on UCTs found no difference in uptake of any maternity care service Conclusion CCT has more impact, but narrow 266 Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers Taafe, Longosz, and Wilson (2017) Focus area Multiple (livelihoods, education, health, HIV) Type Literature review •• Conditionality is not necessary to achieve impact in all cases, but may produce stronger effects •• CCTs may require more programmatic and financial resources to administer in order to effectively monitor and enforce conditions; maintaining such a program in lower-income Findings context may not be feasible •• Conditionality may limit outcomes to those related to the conditions, whereas outcomes from UCTs have potential to be widespread across development sectors Conclusion CCT may strengthen effect, but often infeasible and impact narrow Pellerano and Barca (2017) Focus area Multiple Type Literature review •• The evidence appears to point to the success of explicit conditionality in achieving its goals over the soft approach of UCTs. It is difficult to make a final judgment, however, as the effectiveness of other types of behavioral conditioning has been underinvestigated. •• It is also unclear whether explicit schemes of conditionality can produce a more Findings sustainable change of preferences and thus behavior in the long run. •• Steering the use of social transfers toward socially relevant outcomes can be, and has often been, achieved through several different mechanisms of less explicit behavioral conditioning. Conclusion Behavioral conditioning often sufficient Bastagli et al. (2016) Focus area Multiple Type Rigorous literature review There was some evidence that making transfers conditional on certain behaviors or actions can positively affect the outcomes relating to the conditions on which the transfers are conditioned. While it was not possible to disentangle which aspect of conditions was driving results in most studies, a number of studies highlight the role of people’s perceptions of whether a conditionality is in place or not and of the messaging or communication of desired Findings behaviors in facilitating intended outcomes. Such findings point to the potential for clear communication regarding the importance of service use and support in accessing relevant services to contribute to progress toward program objectives (e.g., in education and health and nutrition), beyond the implementation of additional elements of conditionality such as sanctionary responses to noncompliance associated with potentially high administrative and social costs. Conclusion Behavioral conditioning often sufficient Siddiqi, Rajaram, and Miller (2018) Focus area Health: newborn health Type Systematic review •• Both UCTs and health-focused CCTs tended to improve infants’ birth weight outcomes and Findings reduce infant mortality •• CCTs conditioned on labor force participation had no impact Conclusion UCT best Manley, Gitter, and Slavchevska (2012) Focus area Health: nutrition Type Literature review and meta-analysis Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers 267 •• Health and education–focused CCTs have same effect on child height for age as UCTs Findings •• CCTs with other types of conditions, mostly related to working or saving, show strongly negative impacts on nutritional status (a 0.32 reduction in height for age z-score) Conclusion UCT best Khan, Kant, and Ali (2016) Focus area Health: contraception use Type Systematic review Available evidence of CCT versus UCT effectiveness is inconclusive due to limited number of Findings studies, varying outcome measures, and lack of intervention specifically for contraception Conclusion Inconclusive Pega et al. (2017) Focus area Health: effects of humanitarian cash transfers Type Cochrane systematic review •• UCTs may not significantly affect health service use but may still improve some health Findings outcomes and health care expenditure levels •• Evidence on relative effectiveness of different types of transfers remains very uncertain Conclusion Inconclusive de Hoop and Rosati (2014) Focus area Child labor Type Systematic review •• Both CCTs and UCTs reduce children’s participation in child labor and their hours worked Findings •• More information needed to determine whether schooling conditions matter in this regard Conclusion Inconclusive NOTE: CCT = conditional cash transfer; UCT = unconditional cash transfer. References Akresh, Richard, Damien de Walque, and Harounan Kazianga. 2012. “Alternative Cash Transfer Delivery Mechanisms: Impacts on Routine Preventative Health Clinic Visits in Burkina Faso.” NBER Working Paper 17785. National Bureau of Economic Research, Cambridge, MA. http:// www.nber.org/papers/w17785.pdf —. 2016. “Evidence from a Randomized Evaluation of the Household Welfare Impacts of Conditional and Unconditional Cash Transfers Given to Mothers or Fathers.” Policy Research Working Paper 7730. World Bank, Washington, DC. http://documents.worldbank.org/curated/ en/944741467047531083/pdf/WPS7730.pdf Attanasio, Orazio P., Veruska Oppedisano, and Marcos Vera-Hernández. 2015. “Should Cash Trans- fers Be Conditional? Conditionality, Preventive Care, and Health Outcomes.” American Economic Journal: Applied Economics 7 (2): 35–52. https://www.aeaweb.org/articles?id=10.1257/ app.20130126 Baird, Sarah, Jacobus de Hoop, and Berk Özler. 2013. “Income Shocks and Adolescent Mental Health.” Journal of Human Resources 48 (2): 370–403. http://jhr.uwpress.org/content/48/2/370. refs Baird, Sarah, Francisco H.G. Ferreira, Berk Özler, and Michael Woolcock. 2014. “Conditional, Unconditional and Everything in Between: A Systematic Review of the Effects of Cash Trans- fer Programmes on Schooling Outcomes.” Journal of Development Effectiveness 6 (1): 1–43. https://www.tandfonline.com/doi/abs/10.1080/19439342.2014.890362 268 Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers Baird, Sarah J., Richard S. Garfein, Craig T. McIntosh, and Berk Özler. 2012. “Effect of a Cash Trans- fer Programme for Schooling on Prevalence of HIV and Herpes Simplex Type 2 in Malawi: A Cluster Randomised Trial.” Lancet 379 (2012): 1320–29. https://www.thelancet.com/journals/ lancet/article/PIIS0140-6736(11)61709-1/abstract Baird, Sarah, Craig McIntosh, and Berk Özler. 2011. “Cash or Condition? Evidence from a Cash Trans- fer Experiment.” Quarterly Journal of Economics 126: 1709–53. https://www.researchgate.net/ publication/260676894_Cash_or_Condition_Evidence_from_a_Cash_Transfer_Experiment —. 2016. “When the Money Runs Out: Do Cash Transfers Have Sustained Effects on Human Capital Accumulation?” Policy Research Working Paper 7901. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/495551480602000373/pdf/WPS7901.pdf Bastagli, Francesca, Jessica Hagen-Zanker, Luke Harman, Valentina Barca, Georgina Sturge, Tanja Schmidt, and Luca Pellerano. 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. ­https://www.odi.org/publications/10505-cash-transfers-what- does-evidence-say-rigorous-review-impacts-and-role-design-and-implementation Benedetti, Fiorella, Pablo Ibarrarán, and Patrick McEwan. 2015. “Do Education and Health Con- ditions Matter in a Large Cash Transfer? Evidence from a Honduran Experiment.” Working Paper IDB-WP-577. Inter-American Development Bank, Washington, DC. https://publications. iadb.org/bitstream/handle/11319/6816/Do-education-and-health-conditions-matter-in-a- large-cash-transfer.pdf?sequence=1 Benhassine, Najy, Florencia Devoto, Esther Duflo, Pascaline Dupas, and Victor Pouliquen. 2013. “Turning a Shove into a Nudge? A ‘Labeled Cash Transfer’ for Education.” NBER Working Paper 19227. National Bureau of Economic Research, Cambridge, MA. http://www.nber.org/ papers/w19227 Bourguignon, François, Francisco H.G. Ferreira, and Phillippe G. Leite. 2003. “Conditional Cash Trans- fers, Schooling, and Child Labor: Micro-Simulating Brazil’s Bolsa Escola Program.” World Bank Economic Review 17 (2). https://academic.oup.com/wber/article-abstract/17/2/229/1688944 de Brauw, Alan, and John Hoddinott. 2011. “Must Conditional Cash Transfer Programs Be Con- ditioned to Be Effective? The Impact of Conditioning Transfers on School Enrollment in Mexico.” Journal of Development Economics 96 (2): 359–70. https://www.sciencedirect.com/ science/article/pii/S0304387810000982 de Hoop, Jacobus, and Furio C. Rosati. 2014. “Cash Transfers and Child Labor.” World Bank Research Observer 29 (2): 202–34. https://elibrary.worldbank.org/doi/pdf/10.1093/wbro/lku003 de Janvry, Alain, Frederico Finan, and Elisabeth Sadoulet. 2006. “Evaluating Brazil’s Bolsa Escola Program: Impact on Schooling and Municipal Roles.” University of California at Berkeley, Berkeley. https://are.berkeley.edu/~esadoulet/papers/BolsaEscolaReport6-6.pdf Delany-Moretlwe, Sinead, and Heena Brahmbhatt. 2014. “Feasibility of Economic Interventions for HIV Prevention in Young People Living in Inner-City Johannesburg.” USAID Project SEARCH, Task Order No. 2, Research to Prevention. John Hopkins University, Baltimore. de Walque, Damien, and Christine Valente. 2018. “Incentivizing School Attendance in the Pres- ence of Parent-Child Information Frictions.” Policy Research Working Paper 8476. World Bank, Washington, DC. http://documents.worldbank.org/curated/en/720071529003623702/ Incentivizing-school-attendance-in-the-presence-of-parent-child-information-frictions Edmonds, Eric. 2002. “Reconsidering the Labeling Effect for Child Benefits: Evidence from a Tran- sition Economy.” Economics Letters 76 (3): 303–9. https://www.sciencedirect.com/science/ article/pii/S0165176502000861 Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers 269 Edmonds, Eric, and Norbert Schady. 2012. “Poverty Alleviation and Child Labor.” American Eco- nomic Journal: Economic Policy 4 (4): 100–24. https://pubs.aeaweb.org/doi/pdfplus/10.1257/ pol.4.4.100 Fenton, Rory, Constance Nyamukapa, Simon Gregson, Laura Robertson, Phyllis Mushati, Ranjeeta Thomas, and Jeffrey W. Eaton. 2016. “Wealth Differentials in the Impact of Conditional and Unconditional Cash Transfers on Education: Findings from a Community-Randomised Con- trolled Trial in Zimbabwe.” Psychology, Health & Medicine 21: 909–17. https://www.ncbi.nlm. nih.gov/pubmed/26899880 Handa, Sudhanshu, Amber Peterman, Benjamin Davis, and Marco Stampini. 2009. “Opening Up Pandora’s Box: The Effect of Gender Targeting and Conditionality on Household Spending Behavior in Mexico’s Progresa Program.” World Development 37 (6): 1129–42. https://www. sciencedirect.com/science/article/abs/pii/S0305750X08003057 Hoddinott, John, Akhter Ahmed, and Shalini Roy. 2018. “Randomized Control Trials Demonstrate That Nutrition-Sensitive Social Protection Interventions Increase the Use of Multiple-Micronu- trient Powders and Iron Supplements in Rural Pre-School Bangladeshi Children.” Public Health Nutrition 21: 1753–61. https://www.ncbi.nlm.nih.gov/pubmed/29467053 Hunter, Benjamin M., Sean Harrison, Anayda Portela, and Debra Bick. 2017. “The Effects of Cash Transfers and Vouchers on the Use and Quality of Maternity Care Services: A Systematic Review.” PLOS ONE 12 (3): e0173068. https://www.ncbi.nlm.nih.gov/pubmed/28328940 Khan, M.E., A. Hazra, A. Kant, and M. Ali. 2016. “Conditional and Unconditional Cash Transfers to Improve Use of Contraception in Low and Middle Income Countries: A Systematic Review.” Studies in Family Planning 47 (4): 371–83. https://onlinelibrary.wiley.com/doi/full/10.1111/ sifp.12004 Manley, James, Seth Gitter, and Vanya Slavchevska. 2012. “How Effective Are Cash Transfer Pro- grammes at Improving Nutritional Status? A Rapid Evidence Assessment of Programmes’ Effects on Anthropometric Outcomes.” EPPI-Centre, Social Science Research Unit, Institute of Education, University of London, London. Pace, Noemi, Silvio Daidone, Benjamin Davis, and Luca Pellerano. 2016. “Does Soft Conditionality Increase the Impact of Cash Transfers on Desired Outcomes? Evidence from a Randomized Control Trial in Lesotho.” Department of Economics Research Paper 33/WP/2016. Ca’ Foscari University of Venice, Venice. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2876458 Paiva, Luis Henrique, Fábio Veras Soares, Flavio Cireno, Iara Azevedo Vitelli Viana, and Ana Clara Duran. 2016. “The Effect of Conditionality Monitoring on Educational Outcomes: Evidence from Brazil’s Conditional Cash Transfer Programme.” IPC-IG Working Paper 144. International Policy Centre for Inclusive Growth, Brasília. https://ipcig.org/pub/eng/WP144_The_effects_of_ conditionality_monitoring_on_educational_outcomes.pdf Pega, F., S.Y. Liu, S. Walter, R. Pabayo, R. Saith, and S.K. Lhachimi. 2017. “Unconditional Cash Transfers for Reducing Poverty and Vulnerabilities: Effect on Use of Health Services and Health Outcomes in Low- and Middle-Income Countries.” Cochrane Database of Systematic Reviews 11: CD011135. Pellerano, Luca, and Valentina Barca. 2017. “Does One Size Fit All? The Conditions for Condi- tionality in Cash Transfers.” In What Works for Africa’s Poorest? Programmes and Policies for the Extreme Poor, edited by David Lawson, Lawrence Ado-Kofie, and David Hulme, chap- ter 13. Rugby, UK: Practical Action Publishing. https://www.developmentbookshelf.com/doi/ pdf/10.3362/9781780448435 Robertson, Laura, Phyllis Mushati, Jeffrey W. Eaton, Lovemore Dumba, Gideon Mavise, Jer- emiah Makoni, Christina Schumacher, Tom Crea, Roeland Monasch, Lorraine Sherr, Geoffrey P. Garnett, Constance Nyamukapa, and Simon Gregson. 2013. “Effects of Uncon- ditional and Conditional Cash Transfers on Child Health and Development in Zimbabwe: A 270 Appendix C: Quantitative Evidence on Conditional and Unconditional Transfers Cluster-Randomised Trial.” Lancet 381 (9874): 1283–92. https://www.thelancet.com/journals/ lancet/article/PIIS0140-6736(12)62168-0/fulltext Schady, Norbert, and Maria Caridad Araujo. 2006. “Cash Transfers, Conditions, School Enrollment, and Child Work: Evidence from a Randomized Experiment in Ecuador.” Policy Research Working Paper 3930. World Bank, Washington, DC. https://papers.ssrn.com/sol3/papers. cfm?abstract_id=917501 Siddiqi, Arjumand, Akshay Rajaram, and Steven P. Miller. 2018. “Do Cash Transfer Programmes Yield Better Health in the First Year of Life? A Systematic Review Linking Low-Income/Mid- dle-Income and High-Income Contexts.” BMJ Journals. Archives of Disease in Childhood 103: 920–26. Taafe, Jessica E., Andrew F. Longosz, and David Wilson. 2016. “The Impact of Cash Transfers on Livelihoods, Education, Health and HIV—What’s the Evidence?” Development Policy Review 35 (5): 601–19. Teixeira, Clarissa, Fábio Veras Soares, Rafael Ribas, Elydia Silva, and Guilherme Hirata. 2011. “Externality and Behavioral Change Effects of a Non-Randomized CCT Program: Het- erogeneous Impact on the Demand for Health and Education.” IPC-IG Working Paper 82. International Policy Centre for Inclusive Growth, Brasília. http://www.ipc-undp.org/pub/ IPCWorkingPaper82.pdf Appendix D: Microsimulations Data and Methodology T he main source of information for the microsimulations in chapter 4 is the Atlas of Social Protection Indicators of Resilience and Equity (ASPIRE) database (http:// datatopics.worldbank.org/aspire/). The database collects both administrative and household survey information on social protection programs for over 120 devel- oping countries. The administrative database collects program-level information on social protection programs including spending, number of beneficiaries, and program design features. We use the database to validate/compare total amounts spent on social protection programs from administrative data, with the total benefit amount of social protection programs captured in the household surveys. Administrative data were available for 9 out of 10 countries (Haiti does not have administrative data, and World Bank staff estimates have been used for the Russian Fed- eration; see table D.1). For those countries where administrative data were available, we only considered programs captured in both the household survey and the administrative data. The year of administrative data used in each country is the same as the year of its household survey—except for Chile, where we used the most recent administrative data for 2015 instead of 2013. Program (or subprogram) information was not always available in the administrative database. In Nepal, the administrative database did not disaggre- gate the old age, single woman, disability, and endangered ethnicities pensions, as well as the child grants, while the household survey shows the information by program. The ASPIRE household survey database collects household-level information on social protection programs, welfare indicators (income, expenditure, or consumption), and household demographic characteristics. We use these databases for the universal basic income (UBI) simulations in each country. The chapter is based on 10 household surveys (table D.1); in each country, the survey was the most recent one in the ASPIRE database. The databases consist of both household surveys and specific social protection surveys, as in Russia. We used the harmonized World Bank welfare indicator for Brazil, Chile, Haiti, Kazakhstan, and Nepal. For Indonesia, Mozambique, and South Africa, the harmonized welfare indicator was not available; for Russia, the survey is a specific house- hold-level social protection survey. For these countries and India, we use the national welfare variable included in the surveys. 271 272 Appendix D: Microsimulations Data and Methodology TABLE D.1  ASPIRE Database Use ASPIRE ASPIRE household survey Income administrative group Country data year Name Year Welfare variable Enquête sur les Conditions Total household Haiti Not available de Vie des Ménages 2012 expenditure (harmonized) après Séisme Low Inquérito Sobre Total household Mozambique 2014 2014/15 Orçamento Familiar expenditure Living Standards Survey— Total household Nepal 2010 2010–11 Third Round consumption (harmonized) National Sample Survey Total household India 2011 2011–12 Lower 68th Round consumption middle Survei Sosial Ekonomi Total household Indonesia 2014 2014 Nasional, Maret expenditure Income and Expenditure Harmonized household Brazil 2015 2015 Survey total income (harmonized) Upper Total household Kazakhstan 2015 Household Budget Survey 2015 middle consumption (harmonized) Income and Expenditure Total household South Africa 2014 2014 Survey expenditure Statistical Survey of Russian Not available Income and Participation 2016 Total household income Federation in Social Programs High Encuesta de Harmonized household Chile 2015 Caracterización Socio- 2013 total income (harmonized) Económica Nacional Table D.2 shows the social assistance programs we selected in each country. Over- all, the total amount spent on each program according to administrative data remains similar to the benefit amount captured in the household survey. Nevertheless, the total amount captured in the household survey often remains slightly lower (as in Brazil, Chile, Kazakhstan, Mozambique, and Nepal). This often relates to the small size of the pro- grams, which are not always accurately captured in the surveys. In South Africa, the total amount from the household survey is greater than that from the administrative database because benefit values were imputed, and some households may not have received all the transfers on a regular basis. Note that these small discrepancies between the survey data and the administrative databases do not significantly affect our analysis as, in each country, the UBI transfer is based on the transfers as captured in each survey; therefore, our findings remain internally consistent. World Development Indicators (http://data.worldbank.org/products/wdi) are used for information on country income groups, gross domestic product (GDP), and purchas- ing power parity (PPP). We use data provided as of July 2017. Country-Specific Issues and Decisions In harmonizing the simulation approach across countries, we encountered several coun- try-specific issues that entailed decisions that had to be made, which we list below. Appendix D: Microsimulations Data and Methodology 273 TABLE D.2  Selected Social Assistance Programs by Country Administrative data Household survey Country Program US$ PPP 2011 % GDP US$ PPP 2011 % GDP Low-income countries Scholarships — — 6,289,190 0.0 Haiti Total 6,289,190 0.0 Basic Social Subsidy Programme (PSSB) 89,553,728 0.3 21,197,833 0.1 Mozambique Total 89,553,728 0.3 21,197,833 0.1 Maternal Incentive Scheme 24,556,124 0.0 2,965,674 0.0 Old-age, single woman, disability, and 325,747,636 0.6 236,056,689 0.4 endangered ethnicities pensions and child grants Nepal Social pension 177,561,870 0.3 Allowance for the widowed 53,837,423 0.1 Disability allowance 4,657,397 0.0 Total 350,303,760 0.7 239,022,364 0.4 Lower-middle-income countries Public Distribution System (PDS): Kerosene 19,853,164,547 0.3 8,749,299,546 0.2 India PDS: Food 50,908,902,586 0.9 35,501,675,585 0.6 Total PDS 70,762,067,133 1.2 44,250,975,077 0.8 Bantuan Langsung Sementara Masyrakat 1,497,015,764 0.1 7,078,179,903 0.3 (BLSM) Beras Untuk Rakyat Miskin (Raskin) 4,386,014,735 0.2 695,717,789 0.0 Indonesia Program Keluarga Harapan (PKH) 1,255,561,609 0.0 594,771,434 0.0 Bantuan Siswa Miskin (BSM) 1,593,597,427 0.1 795,663,662 0.0 Total 8,732,189,535 0.3 9,164,332,788 0.4 Upper-middle-income countries Programa Bolsa Família 13,282,189,862 0.4 9,037,919,181 0.3 Benefício de Prestação Continuada (BPC): 11,331,165,059 0.4 12,173,883,960 0.4 Disabled Brazil Benefício de Prestação Continuada (BPC): 8,954,343,338 0.3 Elderly Total 33,567,698,259 1.1 21,211,803,141 0.7 Targeted social assistance 10,093,813 0.0 4,356,954 0.0 State social allowance (families with 762,502,494 0.2 462,436,917 0.1 children) Kazakhstan State social allowance 1,791,549,011 0.4 1,198,184,919 0.3 Special state allowance 968,989,094 0.2 390,095,847 0.1 Housing assistance 18,053,057 0.0 12,841,221 0.0 Total 3,551,187,470 0.9 2,067,915,859 0.5 Disability grant 3,148,871,388 0.5 3,305,898,388 0.5 Child support grant 7,021,930,184 1.0 8,871,145,324 1.3 Care dependency grant 353,204,767 0.1 361,751,206 0.1 Foster child grant 944,927,895 0.1 806,592,818 0.1 South Africa Old-age grant 7,808,852,658 1.2 9,091,002,484 1.3 Grant in aid 48,573,267 0.0 52,995,699 0.0 War veteran’s grant 1,356,937 0.0 8,362,559 0.0 Social relief 94,464,027 0.0 117,074,213 0.0 Total 19,422,181,124 2.9 22,614,822,692 3.4 (continued) 274 Appendix D: Microsimulations Data and Methodology TABLE D.2  Selected Social Assistance Programs by Country (continued) Administrative data Household survey Country Program US$ PPP 2011 % GDP US$ PPP 2011 % GDP High-income countries Pensión Básica Solidaria de Vejez 1,108,094,928 0.3 1,209,290,676 0.3 Pensión Básica Solidaria de Invalidez 505,536,797 0.1 535,425,800 0.1 Subsidio Familiar (SUF) 305,845,100 0.1 387,402,532 0.1 Aporte Previsional Solidario (APS) 1,295,529,151 0.3 251,730,340 0.1 Asignación Familiar y Maternal 82,936,115 0.0 242,782,310 0.1 Leyes de Reparación de DD.HH- 447,601,702 0.1 194,051,770 0.1 Exonerados Políticos Bonos Protección (SSyOO y Chile Solidario) 85,464,589 0.0 56,046,151 0.0 Chile y Egreso (Chile Solidario) Subsidio Discapacidad Mental 46,288,323 0.0 55,020,172 0.0 Bono Base y Transferencias 125,000,732 0.0 40,420,644 0.0 Condicionadas—SSyOO Bono Invierno 140,825,960 0.0 2,784,600 0.0 Bono Logro Escolar—SSyOO 22,854,945 0.0 670,160 0.0 Subsidio al Consumo de Agua Potable 158,792,300 0.0 123,879,672 0.0 Total 4,324,770,642 1.1 3,099,504,828 0.8 Unconditional allowances and other 42,202,603,824 1.22 33,588,756,864 1.0 social payments for people entitled to social support Child allowances 7,812,588,431 0.23 3,129,270,615 0.1 Poverty-targeted cash transfers 1,612,969,255 0.05 735,618,074 0.0 Benefit for children who lost one parent 2,365,854,706 0.07 2,320,011,461 0.1 Maternal capital 13,848,847,359 0.40 23,639,050,125 0.7 Social pensions including disability, 7,945,456,430 0.23 5,731,926,259 0.2 survivorship Free use of the milk kitchen 224,219,625 0.0 Russian Housing subsidy 5,714,431,336 0.16 6,266,875,399 0.2 Federation Scholarships 3,244,741,369 0.09 2,038,391,633 0.1 Food and transportation allowances 93,633,032 0.0 Food and transportation privileges/ 6,954,628,852 0.20 2,958,096,867 0.1 discounts Transfers for caretakers of people in need 2,463,910,583 0.07 1,051,567,543 0.0 of assistance Other cash transfers for government 611,412,240 0.0 organizations Unemployment benefit 1,276,056,942 0.04 863,246,748 0.0 Total 95,442,089,088 2.75 83,252,076,484 2.4 NOTE: — = not available. Due to data availability, administrative database for Chile corresponds to 2015. In South Africa, the amount used for the budget-neutral UBI is US$20,172,051,843 PPP (or 3 percent of GDP) to correct for households with negative consumption net of transfers. Appendix D: Microsimulations Data and Methodology 275 Haiti. The household survey only captures education scholarships. No administra- tive data were available. Mozambique. The household survey only captures the Basic Social Subsidy Pro- gram, which displays lower coverage in the household survey when compared with administrative data. Also, the program’s variable in the survey is participative instead of monetary. To estimate its benefit amount, we impute the value of the transfers to bene- ficiaries as Mt 344 in 2015 per household per month. Nepal. Disaggregated administrative data were not available for the selected pro- grams in the household survey. India. The largest social assistance program in India is the Public Distribution System (PDS), which provides income support to households through price subsidies for wheat, rice, sugar, and kerosene consumption. The households that possess ration cards are eligible to receive benefits under the PDS. To estimate PDS benefit levels, we imputed the subsidies for rice, wheat, sugar, and kerosene for the three types of ration cardholders defined in the survey (BPL, Antyodaya, Others). The PDS subsidy was esti- mated as follow: PDS subsidy = PDS quantity × (Market price − PDS price) where PDS subsidy= subsidies received by the households; PDS price = (value/quantity) of the above-mentioned PDS goods received by households having ration cards (infor- mation about value and quantity was obtained from the National Sample Survey); and Market price is the market price of PDS goods such as rice, wheat, sugar, and kerosene, which was obtained by state from the Indian government’s database (https://data.gov.in/ resources/variety-wise-daily-market-prices-data-wheat-atta-2001-2012). The market price used for kerosene is Rs 39.83, which corresponds to the PDS retail price of kerosene (Rs 14.83)+ underrecovery of oil marketing companies (Rs 25.00) (prices are from the Indian Petroleum Planning and Analysis Cell). The welfare indicator used is the one used by the government of India for national poverty estimation in the 2011–12 round. Indonesia. The two most recent household survey databases (2015 and 2016) do not include Program Keluarga Harapan (PKH), one of the country’s major social assis- tance programs. Furthermore, one program (Beras Untuk Rakyat Miskin—Raskin) uses a participatory variable, meaning monetary variables for 2015 and 2016 are only available for two programs (Bantuan Siswa Miskin and Bantuan Langsung Sementara Masyrakat). We therefore decided to use the 2014 household survey, which includes all four major social assistance programs, captured by participatory variables. We imputed program benefit values to each of the beneficiaries. Significant differences were found in Raskin spending between administrative records and imputed benefit levels in the household survey (table D.3). The value of actual Raskin transfers is low because of discrepancies between total Raskin rice pro- cured and total purchased, between total benefit promised and total benefit received, and between total number of beneficiaries targeted and actual beneficiaries. Records show that of the Raskin rice procured to deliver promised benefits, only about half of 276 Appendix D: Microsimulations Data and Methodology TABLE D.3  Indonesia Simulated Benefit Levels for Main Social Assistance Programs Benefit level (simulated monthly per household) Program Local currency unit (Rp) US$ PPP 2011 Bantuan Langsung Sementara Masyrakat 200,000 48.29 Beras Untuk Rakyat Miskin (Raskin) 8,000 1.93 Program Keluarga Harapan (PKH) 156,000 37.67 Bantuan Siswa Miskin 61,111 14.76 the procured kilograms (in recent years) are actually purchased by households. Raskin should have made 15 kilograms of rice per month available to poor and near-poor house- holds at a subsidized price of Rp 1,600 (US$0.10) per kilogram. Households purchased (per month on average) 3.5, 4.0, and 4.6 kilograms in 2007, 2010, and 2016, respec- tively. Table D.4 shows the evolution of Indonesia’s social assistance budget between 2014 and 2018 in real terms. Brazil. The largest social assistance programs were included in the household survey: Bolsa Família and Benefício de Prestação Continuada (BPC). Total spending from administrative data remains very similar to total benefits from the household survey. However, the survey does not provide disaggregated data for BPC disabled and elderly program components. We assumed the program included benefits for both. Kazakhstan. The total amount spent on the selected social assistance is 0.9 percent of GDP according to the administrative database, while the household survey only records 0.5 percent of GDP. In the UBI simulations, we used the household survey amount. TABLE D.4  Evolution of Indonesia’s Social Assistance Budget 2014 billion Rp Spending by major programs 2014 2018 Unconditional cash transfer (BLT/BLSM) 6,200 — Subsidized rice/food assistance (Bansos Rastra) 18,165 10,499 Food voucher program (BPNT) — 6,176 Health insurance for the poor (PBI-JKN) 19,900 22,182 Cash transfer for poor and vulnerable students (PIP) 6,600 12,879 Conditional cash transfer (PKH) 5,200 15,222 Child social services (PKSA) 345 62 Disabled social services (JSPACA) 79 23 Elderly social services (ASLUT) 64 43 Cash for work (PKT) — 15,657 Total 56,553 82,744 NOTE: — = not available, because the program did not exist in that year. Because the 2018 budget is expressed in 2014 Rp, the nominal budget would be higher. Appendix D: Microsimulations Data and Methodology 277 South Africa. The total amount of the selected transfers captured by the household survey is 0.5 percent of GDP higher than the amount captured by the administrative data (see above for an explanation). Moreover, some of the benefit levels received by house- holds are larger than total household consumption: for the first quintile, the adequacy of the transfers is more than 100 percent. The welfare net of social transfers was there- fore negative for some households, in which case we replaced the negative values with 0 (and then, for the impact analysis, only attributed to each household the transfer amount needed for households to reach the original survey consumption levels). Accordingly, in South Africa the amount used for the budget-neutral UBI is US$20,172,051,843 PPP (or 3 percent of GDP—lower than that reported in the survey) to correct for households with negative consumption net of transfers. Russia. There is high fragmentation and decentralization of social assistance pro- grams in the country. The household survey we used is the Statistical Survey of Income and Participation in Social Programs, which is the main source for social protection– related data. The household survey groups social assistance programs into 14 categories. Chile. The largest poverty-targeted cash transfer program, the Aporte Familiar Per- manente, is not included in the 2013 national survey, as it began in 2014 and is currently the largest poverty-targeted cash program in terms of spending (0.1 percent of GDP). The data used include the largest cash programs in the country as of 2013. Methodology of the Taxation Simulations To capture the impact on living standards after considering the financing mechanisms, the relevant welfare indicator is per capita consumable income, defined as disposable income (or consumption) minus indirect taxes and plus indirect subsidies.1 To assess whether a scenario is welfare increasing or welfare reducing, we compare the poverty, inequality, and decile-based income averages measured with consumable income (instead of dis- posable income) under each of the transfers-cum-financing scenarios against the same indicators, but measured with consumable income for the baseline scenario. Since in almost all cases the amount spent on subsidies is insufficient to cover the deficit even if subsidies were eliminated in full, we do not report the results of the financing scenario that replaces subsidies. Also, the financing gap for the fourth UBI scenario (where every- one receives an amount equal to the poverty line) turns out to be extremely high, so the “with financing” analysis was not considered. The “with financing” scenarios rely on the fiscal incidence results by decile, avail- able from the Commitment to Equity Data Center (http://commitmentoequity.org), which also has descriptions of the data, methodology, and assumptions by country. It is important to note that the incidence of direct and indirect taxes is not the incidence of statutory rates. Due to tax evasion or informality, which are widespread in many devel- oping countries, a significant number of self-employed and salaried workers may not pay direct taxes; and consumers in rural areas and those who purchase from informal sellers (e.g., street vendors, farmers’ markets) may not pay consumption taxes such as value-added tax or excise taxes. The studies housed in the Commitment to Equity Data 278 Appendix D: Microsimulations Data and Methodology FIGURE D.1  Coverage and Incidence of Baseline Programs a. Haiti b. Mozambique 100 100 Percent Percent 50 50 0 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 c. Nepal d. India 100 100 Percent Percent 50 50 0 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 e. Indonesia f. Brazil 100 100 Percent Percent 50 50 0 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 g. Kazakhstan h. South Africa 100 100 Percent Percent 50 50 0 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 i. Chile j. Russian Federation 100 100 Percent Percent 50 50 0 0 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 D1 D2 D3 D4 D5 D6 D7 D8 D9 D10 Coverage Targeting incidence of benefits NOTE: Coverage indicates the percentage of direct and indirect beneficiaries (i.e., beneficiaries and members of beneficiaries’ households) covered in each decile; Targeting incidence indicates the distribution of benefits across deciles. Center make assumptions about informality and evasion. Typically, individuals who do not report being registered in the social security administration are assumed not to pay personal income and payroll taxes. In the case of consumption taxes, for purchases from informal sellers, it is assumed that no consumption taxes are paid (at least, directly at the Appendix D: Microsimulations Data and Methodology 279 time of purchase, although the price of the good may carry the effect of taxes on inputs). If there is no information on place of purchase, some studies assume that households in rural areas do not pay consumption taxes. Note that the way in which each of the under- lying studies accounts for consumption tax informality is not harmonized; hence, for details we refer to the country studies cited above. To calculate consumable income for the baseline and each of the nine “without and with financing” scenarios for the six countries in question, we applied the respective incidence of indirect taxes and indirect subsidies by disposable income decile from the Commitment to Equity Data Center to the per capita disposable income generated with the household surveys. In using the incidence by decile, we are assuming that there is no variation of incidence within a decile. This, of course, is not empirically fully accurate, but within-decile patterns of consumption are likely to be sufficiently similar for the pur- poses of our analysis. Note also that, given that the impacts change relatively little with respect to the scenarios without financing, we do not expect any bias from the approx- imation to be large. To calculate the financing gap, we first calculate the difference between the cost of transfers under the corresponding UBI scenario and the cost of transfers in the baseline. We call this the gross financing gap; it is likely to be a positive number except for the budget-neutral UBI scenario, where spending on transfers under a UBI is kept the same as in the baseline, so by definition this difference will be zero. Note that the gross financ- ing gap does not fully correspond to the actual financing gap (the amount needed to be raised in additional taxes or reduced in subsidies), because under the UBI scenarios that are not budget neutral, disposable income will be higher (everything else being equal). In turn, higher disposable income generates higher consumption; thus, revenues from indirect taxes and spending on indirect subsidies will increase automatically as a result. The financing gap (the needed additional budgetary resources) is the net effect between the difference obtained in the first step and the latter. The total new automatically induced indirect taxes (and indirect subsidies) are cal- culated multiplying the incidence of taxes (subsidies) by decile from the Commitment to Equity Data Center on the new post-UBI disposable income. Direct taxes, in contrast, are calculated using incidence of direct taxes by decile but multiplied by the baseline disposable income. In essence, we are assuming that cash transfers are not subject to personal income tax, which is generally true in developing countries. To calculate the financing gap of alternative financing schemes, we need to take the difference between these totals and the gross financing gap. The financing gap divided by direct taxes collected in the baseline yields the proportion by which direct taxes paid by each indi- vidual would have to increase in order to eliminate the financing gap. In the case of indirect taxes, because the actual collection is higher than the total in the baseline (as a result of the automatic mechanism described above), this proportion is calculated as the ratio of the financing gap divided by the new post-UBI total indirect taxes. In the sce- nario where the gap is financed by a lump-sum increase in direct taxes for the richest decile, the financing gap just needs to be added to the direct taxes paid by this group in the baseline. 280 Appendix D: Microsimulations Data and Methodology In all our calculations (baseline and simulated scenarios), the prefiscal income considered is market income (earnings, nonlabor income, private transfers) plus contrib- utory pensions. In other words, we assume that contributory pensions are either pure deferred income or are part of the salary package (i.e., public servants are paid lower sal- aries than in the private sector but with more generous pensions to encourage retention) rather than government transfers. Income concepts and formulas can be summarized as follows: •• Consumable Income = Disposable Income − Indirect Taxes + Indirect Subsidies •• UBI Consumable Income = UBI Disposable Income with the Additional Tax − Indirect Taxes Based on UBI Disposable Income with the Additional Tax + UBI Indirect Subsidies with the Additional Tax Note 1. This calculation implicitly assumes the standard inelastic responses in conventional fiscal incidence analysis. In the latter, the burden (benefits) of indirect taxes (subsidies) is fully trans- mitted to consumers in the form of higher (lower) prices in the amount equivalent to the tax (subsidy). Appendix E: Annotated Bibliography Ackerman, Bruce, and Anne Alstott. 2000. The Stakeholder Society. New Haven: Yale University Press. The authors argue that every American citizen has the right to share in the wealth accumulated by preceding generations. The distribution of wealth is currently so skewed that the stakeholding fund could be financed by an annual tax of 2 percent on the property owned by the richest 40 percent of Americans. The authors analyze the initiative from moral, political, economic, legal, and human perspectives. By summoning the political will to initiate stakeholding, they argue, societies could become more democratic, productive, and free. https://yalebooks.yale.edu/book/9780300082609/stakeholder-society Ackerman, Bruce, Anne Alstott, and Philippe Van Parijs. 2006. Redesigning Distribution: Basic Income and Stakeholder Grants as Cornerstones for an Egalitarian Capitalism. London: Verso Books. This book explores two proposals, unconditional basic income and stakeholder grants. In a system of basic income, as elaborated by Philippe Van Parijs, all cit- izens are given a monthly stipend sufficient to provide them with a no-frills but adequate standard of living. This monthly income is universal rather than means tested, and it is uncon- ditional—receiving the basic income does not depend upon performing any labor services or satisfying other conditions. It affirms the idea that as a matter of basic rights, no one should live in poverty in an affluent society. In a system of stakeholder grants, all citizens upon reach- ing the age of early adulthood receive a substantial one-time lump-sum grant sufficiently large so that all young adults would be significant wealth holders (grant should be US$80,000). https://www.versobooks.com/books/93-redesigning-distribution Adams, Martin. 2015. Land: A New Paradigm for a Thriving World. Berkeley: North Atlantic Books. Land has been sought after throughout human history. Even today, people struggle to get onto the property ladder and view real estate as an important way to build wealth. Yet, the act of owning land—and our urge to profit from it—causes economic booms and busts, social and cultural decline, and environmental devastation. The book introduces a radically new eco- nomic model that ensures a fairer system. https://www.northatlanticbooks.com/shop/land/ Akee, Randall K.Q., William E. Copeland, Gordon Keeler, Adrian Angold, and E. Jane Costello. 2010. “Parents’ Incomes and Children’s Outcomes: A Quasi-Experiment Using Trans- fer Payments from Casino Profits.” American Economic Journal: Applied Economics 2: 86–115. This article examines the impact on children’s long-run outcomes from an unconditional cash transfer to Native American tribes (casino dividends). An additional US$4,000 per year for the poorest households increases educational attainment by one year at age 21, and reduces the chances of committing a minor crime by 22 percent for 16- and 17-year-olds. Improved paren- tal quality is a likely mechanism for the change. http://dx.doi.org/10.1257/app.2.1.86 Alik-Lagrange, Arthur, and Martin Ravallion. 2018. “Workfare versus Transfers in Rural India.” World Development 112: 244–58. Using data for India’s National Rural Employ- ment Guarantee Scheme (NREGS), the article shows that the basic income guarantee is more 281 282 Appendix E: Annotated Bibliography welfare enhancing than workfare. Allowing for a welfare loss from casual manual labor implies a more “poor-poor” targeting performance, but this is insufficient to compensate for the direct welfare loss from the work requirement for plausible parameter values. A basic income guar- antee dominates NREGS for a given total outlay on workfare wages. https://doi.org/10.1016/j. worlddev.2018.08.008 Anderson, Christopher, and Jonas Pontussen. 2007. “Workers, Worries and Welfare States: Social Protection and Job Insecurity in 15 OECD Countries.” European Journal of Polit- ical Research 46 (2): 211–35. Based on data on people’s attitudes toward their job as well as levels and kinds of social protection collected in 15 Organisation for Economic Co-operation and Development countries, the article shows that social protection measures reduce employ- ment insecurity. The findings also suggest the need to decompose different components of employment insecurity as well as disaggregate national systems of social protection when examining their impact on job insecurity. https://doi.org/10.1111/j.1475-6765.2007.00692.x Anderson, Jonathan. 2002. “The Alaska Permanent Fund: Politics and Trust.” Public Bud- geting and Finance 22 (2): 57–68. Governments use trust funds to earmark policy decisions and remove resource flows from budget competition. The state of Alaska set aside a portion of its oil revenues through the creation of the Alaska Permanent Fund. A unique feature of the fund is that it pays annual dividends to Alaskan citizens, thus creating a citizen stake in its management. Through this framework, Alaskans have successfully protected a significant stream of revenue (US$21  billion) from political demands. https://doi.org/10.1111/1540- 5850.00073 Andersson, Jan-Otto, and Olli Kangas. 2005. “Universalism in the Age of Workfare. Atti- tudes to Basic Income in Sweden and Finland.” In Normative Foundations of the Welfare State: The Nordic Experience, edited by Nanna Kildal and Stein Kuhnle. London: Routledge. This chapter discusses attitudes to basic income in Sweden and Finland, including presenting the results of nationwide and representative opinion surveys. Gallup of Finland collected data in May 2002 through telephone interviews of 1,000 respondents in age brackets 15–80 years of age. TEMO collected the data for Sweden in June 2002 also through telephone interviews of 1,000 Swedes age 16 years and above. http://hdl.handle. net/1814/22714 Anik, Gris. 2016. System Upgrade v2.016: Solution for a Failing Economy, Wealth Distribu- tion, Declining Democracy, Climate Change, and Robots That Steal Jobs. Amazon Digital Services LLC. This book argues that it is possible to live in a highly automated society, without the need to revert to the Stone Age; that technology and the system, if done prop- erly, can work with nature, not against it; and that humans can have high standards of living without a future economic crisis. https://books.google.co.uk/books/about/System_Upgrade_ V2_016.html?id=1YkmMQAACAAJ&source=kp_book_description&redir_esc=y Arneson, Richard J. 1992. “Is Socialism Dead? A Comment on Market Socialism and Basic Income Capitalism.” Ethics 102 (3): 485–511. This article comments on two others in the same journal issue: Roehmer’s theory of market socialism and Van Parijs’s theory of basic income capitalism, which proposes the level of basic unconditional income grant as a stan- dard of a society’s distributive justice. Arneson’s analysis posits that Roemer and Van Parijs agree quite substantially in the fundamental ethical principles underlying their recommen- dations, even though their institutional recommendations are quite different. https://doi. org/10.1086/293421 Aronowitz, Stanley. 2006. Left Turn: Forging a New Political Future (Radical Imagination). New York: Routledge. Aronowitz diagnoses America’s crisis of democracy and the dangers of the new authoritarianism. He draws on history and political theory and from currents of political change around the globe, from the traditions of the European left to the newest polit- ical trends in Latin America. Demonstrating why Democrats lose when they cling to centrism and compromise their core values, this book shows us what a new left party in America would look like in an era of globalization, terrorism, and a crisis of public confidence in government. Appendix E: Annotated Bibliography 283 https://www.abebooks.com/9781594513114/Left-Turn-Forging-New-Political-1594513112/ plp Atkinson, A.B. 1996. “The Case for a Participation Income.” Political Quarterly 67 (1): 67–70. This article presents the case for a participation income. This scheme would pay a basic income to all children of £12.50 a week, in place of the current child benefit, and a partici- pation income to all those age 18 and over who are in work or self-employment, away from work due to sickness or injury, disabled, unemployed and available for work, caring for depen- dents, or over the minimum retirement age. Its introduction would require adjustments to the taxation system such as abolition of income tax allowances and of the upper earnings limit on national insurance contributions. https://doi.org/10.1111/j.1467-923X.1996.tb01568.x —. 2015. Inequality: What Can Be Done? Cambridge, MA: Harvard University Press. Atkinson recommends ambitious new policies in five areas: technology, employment, social security, the sharing of capital, and taxation. He defends these against the common criticism that intervention will shrink the economy, that globalization makes action impos- sible, and that new policies cannot be afforded. http://www.hup.harvard.edu/catalog. php?isbn=9780674504769 —. 2018. Inequality. Cambridge, MA: Harvard University Press. This book presents an update of Atkinson’s proposal to introduce a participation income at a national level, with the potential to eventually expand to European Union–wide implementation. A variant of the basic income, the benefit would be paid not on the basis of citizenship but of participation— defined broadly as making a social contribution (such as full- or part-time waged employment or self-employment, by education, training, or an active job search, by home care for infant children or frail elderly people, or by regular voluntary work in a recognized association). There would be provisions for those unable to participate on the grounds of illness or disability. Atkinson argues that the participation condition is important for promoting “reciprocity”—a message that is both intrinsically justified and more likely to garner political support. He also maintains that the participation income should complement existing social transfers rather than replace them. http://www.hup.harvard.edu/catalog.php?isbn=9780674979789 Avent, Ryan. 2016. The Wealth of Humans: Work, Power and Status in the Twenty-First Century. New York: St. Martin’s Press. Digital technology is transforming every corner of the economy, fundamentally altering the way things are done, who does them, and what they earn for their efforts. Avent looks to the history of the Industrial Revolution and the work of numerous experts for lessons in reordering society. The future need not be bleak, but as this book argues, we cannot expect to restructure the world without a wrenching rethinking of what an economy should be. https://us.macmillan.com/books/9781250075802 Baldwin, Peter. 1990. The Politics of Social Solidarity: Class Bases of the European Welfare State, 1875–1975. Cambridge, UK: Cambridge University Press. This book examines the social bases of the European welfare state, and the interests developed in or against social policy by various classes of society during the period 1875–1975 in Britain, France, Germany, Denmark, and Sweden. By analyzing the competing concerns of different social actors that lie behind the evolution of social policy, it explains why some nations developed a generous and solidaristic welfare state while others fought long and entrenched battles. https://doi. org/10.1017/CBO9780511586378 Banerjee, Abhijit, Paul Niehaus, and Tavneet Suri. 2019. “Universal Basic Income in the Developing World.” NBER Working Paper 25598. National Bureau of Economic Research, Cambridge, MA. The paper argues that interest in universal basic income reflects both positive results from a number of existing cash transfer programs and dissatisfaction with the perceived limitations of piecemeal, targeted approaches to reducing extreme poverty. This paper discusses what we know (and what we do not) about three questions: what recipients would likely do with the incremental income, whether this would unlock further economic growth, and the potential consequences of giving the money to everyone (as opposed to tar- geting it). https://doi.org/10.3386/w25598 284 Appendix E: Annotated Bibliography Barnes, P. 2014. With Liberty and Dividends for All: How to Save Our Middle Class When Jobs Don’t Pay Enough. San Francisco: Berrett-Koehler Publishers. Barnes argues that because of globalization, automation, and winner-take-all capitalism, there will not be enough high-pay- ing jobs to sustain America’s middle class in the future. Barnes proposes to give every American a share of common wealth—starting with our air and financial infrastructure. These shares would pay dividends of several thousand dollars per year—money that would not be welfare or wealth redistribution but legitimate property income. https://www.penguinrandomhouse.com/ books/575136/with-liberty-and-dividends-for-all-by-peter-barnes/9781626562141/ Barry, Brian. 1996. “Real Freedom and Basic Income.” Journal of Political Philosophy 4 (3): 242–76. Barry provides a fierce critique of the concept of “real freedom” (proposed by Van Parijs) and concludes that a universal basic income cannot be justified on those grounds. https://doi.org/10.1111/j.1467-9760.1996.tb00052.x Bastani, Aaron. 2019. Fully Automated Luxury Communism. London: Verso Books. This book argues that new technologies will liberate us from work, providing the opportunity to build a society beyond both capitalism and scarcity. Automation, rather than undermining an econ- omy built on full employment, is instead the path to a world of liberty, luxury, and happiness for everyone. https://www.versobooks.com/books/2757-fully-automated-luxury-communism Benns, Roderick. 2016. Basic Income: How a Canadian Movement Could Change the World. CreateSpace. Benns spent more than two years with one key purpose—using advocacy journalism to get leaders across Canada talking about the potential of a basic income guarantee. From federal ministers, senators, and members of Parliament, to polit- ical party leaders, and mayors across Canada, Benns used his progressive news site to relentlessly interview as many leaders as possible to help advocate for basic income policy in Canada. The book contains every story and Q&A he personally wrote or conducted on basic income over a two-year period. https://books.google.com/books/about/Basic_Income. html?id=5bJBvgAACAAJ&source=kp_book_description Bergmann, Barbara R. 2004. “A Swedish-Style Welfare State or Basic Income: Which Should Have Priority?” Politics & Society 32 (1): 107–18. According to Bergmann, a gen- erous welfare state is currently incompatible with large universal cash payments because of the problems that accumulate as the rate of taxation is pushed to very high levels. The time to consider introducing universal basic income in each country is after the establish- ment of a well-funded welfare state. It could then be phased in slowly, as productivity rises and the labor needed to produce a given level of gross domestic product falls. https://doi. org/10.1177/0032329203261101 Birnbaum, Simon. 2009. “Basic Income, Sustainability and Post-Productivism.” Basic Income Studies 4 (2): 1–7. This article questions whether the postproductivist argument for basic income can simultaneously satisfy two important requirements: the requirements of justice with respect to citizens and their environments, including “the status of the ecosystem considered independently of citizens’ interests” (environmental justice), and the require- ments of justice with respect to the relations between citizens (social justice). https://doi. org/10.2202/1932-0183.1178 —. 2012. Basic Income Reconsidered: Social Justice, Liberalism and the Demands of Equality. New York: Palgrave Macmillan. This book examines the arguments for and against basic income from the point of view of economic and social justice. Birnbaum argues that a universal basic income is radical in the sense that it demands far-reaching equalization of opportunities. It is, at the same time, liberal by insisting that people must be left free to use their resource shares for a much wider range of purposes and life plans than those typically accessible through existing welfare states. https://link.springer.com/ book/10.1057/9781137015426 Birnbaum, Simon, and Jurgen De Wispelaere. 2016. “Basic Income in the Capitalist Econ- omy: The Mirage of ‘Exit’ from Unemployment.” Basic Income Studies 11 (1): 61–74. A widespread argument in the basic income debate is that the unconditional entitlement to a Appendix E: Annotated Bibliography 285 secure income floor improves workers’ bargaining position vis-à-vis their employers. This arti- cle critically examines the issue and suggests that, under current economic conditions, an exit strategy might end up worsening rather than strengthening the opportunity set and bargain- ing position of the most vulnerable workers. https://doi.org/10.1515/bis-2016-0013 Bregman, Rutger. 2017. Utopia for Realists: And How We Can Get There. London: Blooms- bury. Bregman argues that a universal basic income is what capitalism ought to have been striving for all along, and insists that—in the current context of increasingly flexible jobs, middle-class wage erosion, growing social inequality, and automation-induced unemploy- ment—the time is ripe for the policy’s adoption. He claims that the United States is rich thanks to the institutions, knowledge, and social capital amassed by previous generations, and a basic income would allow all to share in this wealth. He concludes by advocating for small UBI pilots that can challenge the way of thinking and provide the foundation for implementa- tion at scale. https://www.bloomsbury.com/uk/utopia-for-realists-9781408890264/ Browne, James, and Herwig Immervoll. 2017. “Mechanics of Replacing Benefit Systems with a Basic Income: Comparative Results from a Microsimulation Approach.” Jour- nal of Economic Inequality 15 (4): 325–44. Through the use of EUROMOD data, the authors assess the fiscal and distributional effects of a universal basic income in four European states (Finland, France, Italy, and the United Kingdom). A basic income scheme would fix bene- fit coverage gaps that exist in many countries, but would require very substantial tax rises if it were to be set at a meaningful level. As support would not be targeted on those most in need, it would not be a cost-effective way of directly reducing income poverty. https://doi. org/10.1007/s10888-017-9366-6 Bryan, James B., and Sarah Lamarche Castillo. 2012. “Politics, Preservation of Natural Resource Wealth, and the Funding of a Basic Income Guarantee.” In Alaska’s Perma- nent Fund Dividend: Examining Its Suitability as a Model, edited by Karl Widerquist and Michael Howard. New York: Palgrave Macmillan. The authors argue that the political requirements of preserving wealth from a mineral discovery coincide exceptionally well with the funding of the politically popular basic income guarantee. Further, among the options for financing a basic income guarantee, revenues from a resource discovery are among the most palatable with the electorate. Marrying the processes and goals of basic income policies and the goal of natural wealth preservation advances each of these relative to pursuing them sep- arately. https://doi.org/10.1057/9781137015020_5 Brynjolfsson, Eric, and Andrew McAfee. 2014. The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies. New York: W.W. Norton & Company. The authors make the case that we should be optimistic about the future because technologi- cal progress, “the only free lunch that economists believe in,” is accelerating quickly past our intuitions and expectations. But they also argue that we must be mindful of our values and our choices: as technology races ahead, it may leave a lot of people, organizations, and insti- tutions behind. http://secondmachineage.com/ Calnitsky, David. 2016. “‘More Normal than Welfare’: The Mincome Experiment, Stigma, and Community Experience.” Canadian Review of Sociology 53 (1): 26–71. This article examines Mincome’s “saturation” site in Dauphin, Manitoba, where all town residents were eligible for guaranteed annual income payments for three years. Drawing on archived qualita- tive participant accounts, it shows that Mincome participation did not produce social stigma. The social meaning of Mincome was sufficiently powerful that even participants with particu- larly negative attitudes toward government assistance felt able to collect Mincome payments without a sense of contradiction. By obscuring the distinctions between the “deserving” and “undeserving” poor, universalistic income maintenance programs may weaken social stigma- tization and strengthen program sustainability. https://doi.org/10.1111/cars.12091 —. 2017. “Debating Basic Income.” Catalyst 1 (3). This essay surveys the universal basic income (UBI) debate, focusing on the main objections from the left. It evaluates the normative issues at the heart of the proposal and analyzes a range of possible empirical effects, from the 286 Appendix E: Annotated Bibliography impact on wages and labor force participation to gender and collective action. Ultimately, it makes a case for UBI on grounds of freedom and power: insofar as it allows people to escape from Marx’s “double freedom,” the UBI fosters both “exit” and “voice,” and thus has real affinity with the socialist project. https://catalyst-journal.com/vol1/no3/debating-basic-income —. 2018. “The Employer Response to the Guaranteed Annual Income.” Socio-Eco- nomic Review 25. Using an archived survey of local firms that inquires into wage rates, applications, hiring, and work hours, this article examining the impact of Mincome finds that the guaranteed income tightens labor markets and pulls wages up. It examines the mecha- nisms behind the employer subsidy argument and considers the conditions under which a variety of income support policies might increase or decrease wages, and more broadly, foster compromise or conflict in the labor market. https://doi.org/10.1093/ser/mwy009 Calnitsky, David, and Jonathan P. Latner. 2017. “Basic Income in a Small Town: Under- standing the Elusive Effects on Work.” Social Problems 64 (3): 1–25. This article finds that the Mincome experiment reduced labor market participation by 11.3 percentage points, and nearly 30 percent of that fall can be attributed to community context effects. Addition- ally, it shows that withdrawals were driven disproportionately by young and single-headed households. Main reasons for labor market withdrawal included care work, disability and ill- ness, uneven employment opportunities, or educational investment. https://doi.org/10.1093/ socpro/spw040 Caputo, Richard, ed. 2012. Basic Income Guarantee and Politics: International Experi- ences and Perspectives on the Viability of Income Guarantee. New York: Palgrave Macmillan. This collection brings together international and national scholars and advo- cates to provide historical overviews of efforts to pass basic income guarantee legislation in their respective countries and/or across regions of the globe. https://www.palgrave.com/gb/ book/9780230116917 Casassas, David. 2007. “Basic Income and the Republican Ideal: Rethinking Material Independence in Contemporary Societies.” Basic Income Studies 2 (2): 1–7. This article discusses both the motivations and obstacles of a republican case for basic income. It con- siders whether civic duties are disregarded, and then assesses the principle of independence, which seems to favor unconditional ways of granting individuals’ socioeconomic self-suffi- ciency. Casassas goes on to ask whether there could be other preferred ways of granting individuals their material independence, for example, through basic capital rather than a basic income scheme, or through workfare schemes. He concludes that the republican case for basic income requires thoroughly considering and debating both normative goals and the institutional means necessary to achieve them. https://doi.org/10.2202/1932-0183.1081 Chrisp, Joe. 2017. “Basic Income: Beyond Left and Right?” Juncture 23 (4): 266–70. With advo- cates on both left and right, is the vision being blurred by a familiar ideological battleground? Chrisp argues that the apparent agreement results from the wide-ranging use of the basic income label, which describes policies that are quite distinct. Even where the schemes overlap, the pro- posed mechanisms for funding them diverge greatly. Chrisp also assesses that the “right versus left” analysis might not be the best dimension to characterize political division, and a “univer- salist versus particularist” dimension might be more apt. https://doi.org/10.1111/newe.12025 Coady, David, and Delphine Prady. 2018. “Universal Basic Income in Developing Countries: Issues, Options and an Illustration for India.” IMF Working Paper 18/174. International Monetary Fund, Washington, DC. Using India as an illustration, the paper discusses the trade-offs that need to be recognized in adopting a universal basic income (UBI). It shows that replacing the 2011 Public Distribution System (PDS) with a UBI results in losses for many low-income households, although much of this can be reduced by recycling the “out-of-sys- tem” PDS losses and the fiscal savings from excluding the highest-income groups as higher UBI transfers. In contrast, replacing inefficient energy subsidies—raising energy prices to efficient levels to internalize the negative environmental externalities of energy consumption—could deliver unambiguous distributional gains, help address fiscal pressures, and improve energy Appendix E: Annotated Bibliography 287 efficiency with associated environmental and health gains. Implementing such reforms would, of course, require careful communication and implementation to address political barriers to reform. https://www.imf.org/en/Publications/WP/Issues/2018/07/31/Universal-Basic-Income- in-Developing-Countries-Issues-Options-and-Illustration-for-India-46079 Colombino, Ugo. 2015. “Is Unconditional Basic Income a Viable Alternative to Other Social Welfare Measures?” IZA World of Labor, 128. Bonn. Economic theory and empiri- cal evidence suggest that, under certain conditions, unconditional basic income might be an important policy innovation for redistributing the gains from automation and globalization, building a buffer against shocks and systemic risks, and generating positive labor supply incentives among poor people. While an unconditional basic income policy is simple and transparent, with low administration costs, financing it might require higher taxes. Although the evidence on implied efficiency losses is mixed, carefully designed taxes can avoid the risk of wiping out the potential benefits of unconditional basic income through efficiency losses. https://wol.iza.org/articles/is-unconditional-basic-income-viable-alternative-to-other-social- welfare-measures/long Cunliffe, John, and Guido Erreygers, eds. 2004. The Origins of Universal Grants: An Anthology of Historical Writing on Basic Capital and Basic Income. London: Palgrave Macmillan. Should all young adults receive a capital grant? Should all individuals be given a lifetime regular income? Would either form of payment be just or unjust? These questions figure prominently in recent social philosophy and policy discussions on “stakeholding” and basic income. Both types of proposal have a long, but largely unknown, history. This anthol- ogy contains a wide variety of historical contributions, some of which are presented in English for the first time, highlighting striking parallels between past and present debates. https:// www.palgrave.com/gb/book/9781403918963 Davala, Sarath, Renana Jhabvala, Guy Standing, and Soumya Kapoor Mehta. 2015. Basic Income: A Transformative Policy for India. London: Bloomsbury. The book draws on a series of evaluation surveys conducted over the course of the 18 months in which a univer- sal basic income pilot in Madhya Pradesh was in operation, supplemented with detailed case studies of individuals and families. It looks at the impact on health and nutrition, schooling, economic activity, women’s agency, and the welfare of those with disabilities. Above all, the book considers whether a basic income could be transformative, not only in improving indi- vidual and family welfare but in promoting economic growth and development, as well as having an emancipatory effect for people long mired in conditions of poverty and economic insecurity. https://www.bloomsbury.com/uk/basic-income-9781472583116/ Desai, Ray M., and Homi Kharas. 2017. “Is a Growing Middle Class Good for the Poor? Social Policy in a Time of Globalization.” Global Economy and Development Work- ing Paper 105. Brookings Institution, Washington, DC. The authors examine the role of the middle class and social spending on the extremely poor using data covering a range of early and late-industrializing countries between 1870 and the present. They find that poverty reduction occurs alongside a growing middle class, mediated in part through greater spending on health, education, and welfare. Given that antipoverty policies are not likely to be sustained without the tacit support of the middle class, a central implication of the research findings is that developing countries aiming to eradicate extreme poverty will need to focus on raising the volume of social assistance in a package with universal programs designed to cover groups in nonstandard jobs. https://www.brookings.edu/research/is-a-growing-middle-class-good-for- the-poor-social-policy-in-a-time-of-globalization/ Devarajan, Shantayanan. 2018 “How to Use Oil Revenues Efficiently.” Working Paper 1199. Economic Research Forum, Giza. Oil-rich countries systematically misallocate public expenditures relative to non-oil countries—by favoring consumption over capital, and within consumption, inefficient subsidies and public sector wages over targeted transfers. Further- more, for given levels of expenditure, value for money is considerably less in oil-rich countries. This paper argues that the reason for these inefficiencies is that oil revenues go directly to the government without passing through the hands of the citizens, as is the case with tax 288 Appendix E: Annotated Bibliography revenues. To improve public spending efficiency, Devarajan proposes that all oil revenues be distributed directly to citizens, and resources raised through taxation. The author considers possible obstacles to such a reform and shows that they have been overcome by technology, politics, and knowledge exchange. http://erf.org.eg/wp-content/uploads/2018/05/1199_Final. pdf De Wispelaere, Jurgen. 2015. An Income of One’s Own? The Political Analysis of Universal Basic Income. Tampere: UTA Press. This dissertation adopts a public administration per- spective and identifies a range of key implementation challenges that need to be carefully resolved for basic income to become a practical policy proposal. Furthermore, it demonstrates that these administrative challenges are decidedly political in nature, reinforcing the potential for persistent internal disagreement among basic income advocates. https://tampub.uta.fi/ bitstream/handle/10024/98162/978-951-44-9989-0.pdf —. 2016. “The Struggle for Strategy: On the Politics of the Basic Income Proposal.” Politics 36 (2): 131–41. This article reviews two thorny problems that affect the coalition-build- ing efforts of basic income advocates: on one hand cheap political support suggests most basic income support may be of little value to further its implementation. On the other hand, persistent political division argues superficial agreement among committed basic income advocates may mask persistent disagreement on which precise model to adopt. The article discusses the relevance of each of these problems for basic income politics, employing both analytical arguments and brief illustrations taken from debates in various countries. https:// doi.org/10.1111/1467-9256.12102 —. 2016. “Basic Income in Our Time: Improving Political Prospects through Policy Learning?” Journal of Social Policy 45 (4): 617–34. This article reviews a number of policy developments through the lens of the policy transfer literature. In the absence of a fully devel- oped basic income in place, proponents must rely on partially implemented schemes or proposals that differ in crucial respects from the basic income ideal. This paper outlines three sets of empirical cases and analyzes what (if any) lessons we can draw from them regarding the future of basic income schemes. https://doi.org/10.1017/S0047279416000039 De Wispelaere, Jurgen, and Leticia Morales. 2016. “The Stability of Basic Income: A Con- stitutional Solution for a Political Problem?” Journal of Public Policy 36 (4): 521–45. In this article, the authors tackle a political problem any enacting coalition must face: how to ensure the political stability of a basic income over time. They first demonstrate how basic income schemes are particularly vulnerable to processes of policy change as discussed in the recent policy feedback literature. They then analyze whether constitutionalizing basic income in a bill of rights protected by strong judicial review would offer a valuable route for boost- ing basic income’s stability. A careful examination of the decision-making process within judicial review suggests that, caught up in a dilemma between judicial restraint and judicial activism, an enacting coalition would do well not to rely on constitutional mechanisms as the sole avenue for ensuring the political stability of basic income. https://doi.org/10.1017/ S0143814X15000264 De Wispelaere, Jurgen, and José A. Noguera. 2012. “On the Political Feasibility of Univer- sal Basic Income.” In Basic Income Guarantee and Politics: International Experiences and Perspectives on the Viability of Income Guarantees, edited by Richard Caputo. New-York: Palgrave Macmillan. For much of the last two decades, debate around the pro- posal of a universal basic income (UBI) centered on arguing the ethical and economic case for instituting a policy that grants each adult citizen a guaranteed income as a right, without a means test or work requirement. The question of how to bring about such a policy—the ques- tion of political feasibility—has only recently gained traction among UBI advocates. In this chapter, the authors lay out an analytical framework to think about the political feasibility of UBI in a more systematic manner. https://doi.org/10.1057/9781137045300_2 De Wispelaere, Jurgen, and Lindsay Stirton. 2004. “The Many Faces of Universal Basic Income.” Political Quarterly 75 (3): 266–74. This article argues that there is no such thing Appendix E: Annotated Bibliography 289 as a preferred basic income scheme independent of the overall institutional and policy con- text. “Fuzzy” policy design paints a world in which policy reform explicitly acknowledges the many faces of basic income, and uses this feature as its main strength to further the case of social justice across Europe and beyond. https://doi.org/10.1111/j.1467-923X.2004.00611.x —. 2007. “The Public Administration Case against Participation Income.” Social Ser- vice Review 81 (3): 523–49. This article suggests what the authors call the “trilemma” of participation income: only at significant costs to administrators and welfare clients does the scheme retain its apparent ability to satisfy the requirements of both activation and universal approaches to welfare. Consequently, the main apparent strength of participation income—its capacity to unite different factions within the basic income debate—is shown to be illusory precisely because competing factions strongly prefer different resolutions of the trilemma. This has far-reaching implications for the political strategy of basic income advocates as well as the wider debate on universal welfare reform. https://doi.org/10.1086/520939 —. 2011. “The Administrative Efficiency of Basic Income.” Policy and Politics 39 (1): 115–32. This article exposes several misconceptions, generalizations, or assumptions that per- meate discussion of the administrative properties of basic income. Each of these obscures a significant constraint on the possibility of administrative savings, or else inflates the likely size of such efficiencies where they do exist. The analysis also reveals a number of important political choices faced by policy makers and advocates intent on implementing an administra- tively efficient basic income policy. https://doi.org/10.1332/030557311X546352 —. 2012. “A Disarmingly Simple Idea? Practical Bottlenecks in Implementing a Uni- versal Basic Income.” International Social Security Review 65 (2): 103–21. The authors identify and examine three important practical bottlenecks that may prevent a basic income scheme from attaining universal reach: (1) maintaining a populationwide cadaster of eligi- ble claimants ensuring full takeup, (2) instituting robust modalities of payment that reach all intended beneficiaries, and (3) designing an effective oversight mechanism in a policy context that actively opposes client monitoring. http://dx.doi.org/10.1111/j.1468-246X.2012.01430.x —. 2012. “The Politics of Unconditional Basic Income: Bringing Bureaucracy Back In.” Political Studies 61 (4): 915–32. The authors identify three essential tasks any income maintenance policy must accomplish: defining criteria of eligibility, determining who meets such criteria, and disbursing payments to those found to be eligible. Building on the work of Christopher Hood, they contrast two alternative ways in which the design of a universal basic income (UBI) might apply the principle of “using bureaucracy sparingly” to the per- formance of each of these three tasks. Relating these alternative designs to the politics of basic income, they show a correspondence between contrasting senses of using bureaucracy sparingly and “redistributive” and “aggregative” UBI models. https://doi.org/10.1111/j.1467- 9248.2012.01004.x Dowding, Keith, Jurgen De Wispelaere, and Stuart White, eds. 2004. The Ethics of Stake- holding. London: Palgrave Macmillan. Stakeholder policies have been hailed by academics and policy makers as one of the most promising tools for combating poverty, unemployment, and inequality in modern welfare states. This collection by leading academics offers a com- prehensive overview of stakeholding and critically explores the ethical foundations of the stakeholder society. https://www.palgrave.com/gb/book/9781403905802 Downes, Amy, and Stewart Lansley, eds. 2018. It’s Basic Income: The Global Debate. Bris- tol: Policy Press. This book gathers a number of insightful articles on the idea of a basic income and related issues, bringing in international contributors from a variety of economic and social approaches. It offers assessments of trials that have been conducted, obstacles to implementation, and possible avenues for political and economic action. https://policy. bristoluniversitypress.co.uk/its-basic-income Dyer, Gwynne. 2018. Growing Pains: the Future of Democracy (and Work). Brunswick, Vic- toria, Australia: Scribe Publications. Dyer examines the rise of automation and calls for 290 Appendix E: Annotated Bibliography finding new ways to provide income support to affected populations. https://scribepublications. com.au/books-authors/books/growing-pains Eisenstein, Charles. 2011. Sacred Economics: Money, Gift, and Society in the Age of Tran- sition. North Atlantic Books. The book explores avant-garde concepts of negative interest currencies, local currencies, resource-based economics, gift economies, and the restoration of the commons. https://www.northatlanticbooks.com/shop/sacred-economics/ Ensor, Will, Anderson Fraley, Matt Jensen, and Amy Xu. 2017. “A Budget-Neutral Universal Basic Income.” AEI Economics Working Paper 2017-03. American Enterprise Insti- tute, Washington, DC. The authors simulate a three-part policy reform in the United States: (1)  repeal most welfare and transfer programs, including social security and Medicare; (2) repeal most base-narrowing features of the individual income tax system; and (3) replace those programs with a UBI on a budget-neutral basis. They use Tax Calculator and Tax Data to estimate the consequences of the tax provisions, and use the new Current Population Survey Transfer Augmentation Model to estimate the consequences of repealing welfare and transfer programs. https://www.aei.org/publication/a-budget-neutral-universal-basic-income/ Eyal, Nir. 2010. “Near-Universal Basic Income.” Basic Income Studies 5 (1): 1–26. Under what Eyal calls “near-universal basic income” (NUBI), everyone receives basic income except the rich. NUBI is therefore only near universal and requires means testing. Eyal’s thesis is that if standard considerations that are often advanced to support UBI against social relief are suc- cessful, then these combined considerations probably lend NUBI even greater support. The considerations examined focus on adequacy, cost cuts, equality, freedom, the social bases of self-respect, and political resilience. https://doi.org/10.2202/1932-0183.1149 Felber, Christian. 2015. Change Everything: Creating an Economy for the Common Good. London: Zed Books. Felber lays out a vision for the Economy for the Common Good, a rap- idly growing international movement, reorienting our relationship to work, money, and the purpose of both. https://www.zedbooks.net/shop/book/change-everything/ Ferguson, James. 2015. Give a Man a Fish: Reflections on the New Politics of Distribution. Durham: Duke University Press. Interested in an emerging “politics of distribution,” Fergu- son shows how new demands for direct income payments (including universal basic income) require us to reexamine the relation between production and distribution, and to ask new questions about markets, livelihoods, labor, and the future of progressive politics. https://doi. org/10.1215/9780822375524 Fitzpatrick, Tony. 1999. Freedom and Security: An Introduction to the Basic Income Debate. New York: Palgrave Macmillan. This book provides an introduction to the basic income debate, examining a range of arguments for and against, and charts the future direction for the welfare state. https://www.palgrave.com/gb/book/9780312223137 Forget, Evelyn. 2011. “The Town with No Poverty: The Health Effects of a Canadian Guar- anteed Annual Income Field Experiment.” Canadian Public Policy 37 (3): 283–305. This article documents the historical context of Mincome, a Canadian guaranteed annual income field experiment (1974–79). Specifically, it uses routinely collected health administration data and a quasi-experimental design to document an 8.5 percent reduction in the hospitalization rate for participants relative to controls, particularly for accidents and injuries and mental health. The author also found that participant contacts with physicians declined, especially for mental health, and that more students continued into grade 12. She found no increase in fertility, family dissolution rates, or improved birth outcomes. https://doi.org/10.3138/ cpp.37.3.283 —. 2018. Basic Income for Canadians: The Key to a Healthier, Happier, More Secure Life for All. Toronto: James Lorimer & Company Ltd. This book reports on research from the original test in Manitoba in the 1970s. The evidence shows that basic income improves family and community health and well-being, financial resilience, and access to education and training. http://www.lorimer.ca/adults/Book/3070/Basic-Income-for-Canadians.html Appendix E: Annotated Bibliography 291 Francese, Maura, and Delphine Prady. 2018. “Universal Basic Income: Debate and Impact Assessment.” Working Paper WP/18/273. Fiscal Affairs Department, International Monetary Fund, Washington, DC. This paper discusses the definition and modeling of a universal basic income (UBI). After presenting the arguments in favor and against, an analyt- ical approach for its assessment is proposed. The adoption of a UBI is discussed with regard to the policy objectives (shaped by social preferences) it is designed to achieve. Key design dimensions to be considered include coverage, generosity of the program, overall progressiv- ity of the policy, and financing. https://www.imf.org/en/Publications/WP/Issues/2018/12/10/ Universal-Basic-Income-Debate-and-Impact-Assessment-46441?cid=em-COM-123-38103 Frankel, Sid, and James P. Mulvale. 2014. “Support and Inclusion for All Manitobans: Steps toward a Basic Income Scheme.” Manitoba Law Journal 37 (2): 425–64. The authors outline an incremental approach to developing a provincial basic income, and some of the political and practical challenges that will need to be taken into account. https://ssrn.com/ abstract=2822768 Frayne, David. 2016. The Refusal of Work: Rethinking Post-Work Theory and Practice. London: Zed Books. Frayne delves into the reasons that people disconnect from work, the strategies they develop for coping, and what they do with their free time. https://www.press. uchicago.edu/ucp/books/book/distributed/R/bo22554138.html Fritz, Verena. 2014. “Dealing with a Resource Shock: Political Economy Analysis and Its Impacts in Mongolia.” In Problem-Driven Political Economy Analysis: The World Bank’s Experience, edited by Verena Fritz, Brian Levy, and Rachel Ort. Washington, DC: World Bank. As part of the tale on the political economy of managing resource windfalls, Fritz offers insights into Mongolia’s short-lived universal basic income (UBI) program. The chronicle of the events, as, for example, nicely captured in table 2.1, provide information on one of the most underexplored, and yet large-scale, experiences with UBI worldwide. Impor- tantly, the chapter examines the UBI conception and implementation within the broader political economy dynamics that shaped the country over the turbulent 2008–12 period. https://elibrary.worldbank.org/doi/pdf/10.1596/978-1-4648-0121-1_ch2?download=true Funiciello, Theresa. 1994. Tyranny of Kindness: Dismantling the Welfare System to End Pov- erty in America. New York: Atlantic Monthly Press. Funiciello’s own welfare story, as well as those of the many others she has come in contact with, forms the backdrop of this book. Reviewing the history of assistance for the poor, she calls for bypassing the bureaucracy to give monetary assistance directly to the poor. https://groveatlantic.com/book/tyranny-of-kindness/ Furman, Jason, and Robert Seamans. 2018. “AI and the Economy.” NBER Working Paper 24689. National Bureau of Economic Research, Cambridge, MA. The authors review cur- rent and potential policies around AI that may help boost productivity growth while mitigating any labor market downsides. These include evaluating the pros and cons of an AI-specific reg- ulator, expanded antitrust enforcement, and alternative strategies for dealing with the labor market impacts of artificial intelligence, including universal basic income and guaranteed employment. https://www.nber.org/papers/w24689 Gamble, Andrew. 2016. Can the Welfare State Survive? Cambridge, UK: John Wiley & Sons. In this book, Gamble explains why Western societies need generous and inclusive wel- fare states for all their citizens. Welfare states can survive, he argues, but only if there is the political will to reform them and to fund them. https://www.wiley.com/en-gb/Can+the+Wel fare+State+Survive%3F-p-9780745698748 Ghatak, Maitreesh, and François Maniquet. 2019. “Universal Basic Income: Some The- oretical Aspects.” Annual Review of Economics 11: 895–928. The authors review the possible theoretical justifications of introducing a universal basic income (UBI) and examine the determinants of its feasibility and scope. They begin by contrasting the unconditionality of UBI with the many conditions that typically accompany welfare policies. Possible justi- fications for an unconditional UBI range, in the authors’ view, from philosophical reasons 292 Appendix E: Annotated Bibliography to practical challenges in targeting and inefficiencies inherent in public agencies. They also explore the conditions that determine the feasibility and size of a UBI. Normative and practical considerations make UBI easier to defend as a tool of poverty alleviation in poor economies than a tool to achieve social justice in rich ones. http://personal.lse.ac.uk/ghatak/ UBI_theory.pdf Giugale, Marcelo, and Nga Thi Viet Nguyen. 2014. “Money to the People: Estimates of the Potential Scale of Direct Dividend Payments in Africa.” CGD Policy Paper 043. Center for Global Development, Washington, DC. This paper focuses on the distribution of resource rents as cash transfers to citizens, so-called direct dividend payments. It expands on recent related literature by calculating such transfers, whether universal or targeted, for every African country for which data are available, and compares them to measures of poverty depth under both national and global definitions. It then extends the analysis to a different kind of resource flow enjoyed by most African countries—foreign aid. The authors find that direct dividend payments can account for a large proportion of the income Africa’s poor need to live above the poverty line. https://www.cgdev.org/sites/default/files/money-people- estimates-direct-dividend-payments-africa.pdf Goldsmith, Oliver Scott. 2010. “The Alaska Permanent Fund Dividend: A Case Study in Implementation of a Basic Income Guarantee.” Working paper. Institute of Social and Economic Research, University of Alaska Anchorage. The Alaska Permanent Fund Dividend program has attracted considerable interest because it is a unique example of a basic income guarantee. This paper describes the structure of the dividend program, its economic effects, and some of its unintended consequences. It concludes with a number of observations about how the dividend might be structured differently. http://hdl.handle. net/11122/4170 Gourevitch, Alex, and Lucas Stanczyk. 2018. “The Basic Income Illusion.” Catalyst 1 (4): 1–22. The authors argue that the familiar claim regarding the emancipatory potential of basic income for workers is misconceived. They suggest that a universal basic income (UBI) high enough to be genuinely liberating would require enormous financial commitments. Con- sequently, a prerequisite for UBI would be an organized working class powerful enough to mobilize significant funding. https://catalyst-journal.com/vol1/no4/the-basic-income-illusion Government of India. 2017. “Universal Basic Income: A Conversation With and Within the Mahatma.” In Economic Survey 2016–17, 172–211. New Delhi. This chapter lays out the conceptual pros and cons of universal basic income (UBI), and then outlines a “qua- si-UBI” proposal covering 75 percent of the Indian population. The wealthiest 25  percent of Indians would be excluded for reasons of political and fiscal feasibility. At a predicted cost of 4.9  percent of gross domestic product, the survey acknowledges that the program would be unaffordable for the government unless some existing social welfare programs were replaced. Gradual, phased implementation is advised, starting with specific subpopulations such as women, vulnerable individuals, or urban households. https://www.indiabudget.gov. in/budget2017-2018/es2016-17/echapter.pdf Graeber, David. 2018. Bullshit Jobs: A Theory. New York: Simon and Schuster. Graeber explores how a number of current jobs may not generate value (and have meaning) for soci- eties, economies, and individuals. Instead, the jobs are often meant to perpetuate power relations, including through the concept of “buying people’s time.” By being detached from work, a universal basic income, he argues, would help reframe those power relations. https:// www.simonandschuster.com/books/Bullshit-Jobs/David-Graeber/9781501143311 Greene, Leonard M. 1998. The National Tax Rebate: A New America with Less Govern- ment. Washington, DC: Regnery Publishing. Greene argues that a national tax rebate would replace the costly myriad of targeted, inefficient government assistance programs with a simple, universal cash grant of US$1,000 a month for every American family of four. The rebate would give half of the U.S. budget back to the American people. https://www.regnery. com/books/the-national-tax-rebate/ Appendix E: Annotated Bibliography 293 Groot, Loek. 2004. Basic Income, Unemployment and Compensatory Justice. Dordrecht: Kluwer. This book argues that implementing a substantial basic income is the best policy response to deal with job insecurity, social exclusion, and poverty and to improve labor market flexibility and boost low wage employment and part-time work. https://www.springer. com/gp/book/9781402026140 —. 2006. “Reasons for Launching a Basic Income Experiment.” Basic Income Stud- ies 1: 1–7. This paper argues that a basic income experiment would provide the benchmark or baseline against which the results of all workfare-oriented experiments can be evaluated. https://ssrn.com/abstract=2185877 Groot, Loek, Ruud Muffels, and Timo Verlaat. 2019. “Welfare States’ Social Investment Strategies and the Emergence of Dutch Experiments on a Minimum Income Guar- antee.” Social Policy and Society 18 (2): 277–87. This article addresses cultural values and principles, such as personal autonomy (capacitating people by providing opportunities and therewith “free choice”) and trust (activating people by putting trust in their self-management capacities) which in day-to-day policy practice means more tailor-made, demand-oriented integrated mediation and coaching while rewarding people instead of penalizing them. https://doi.org/10.1017/S1474746418000283 Groot, Loek, and Robert van der Veen. 2003. Basic Income on the Agenda: Policy Objec- tives and Political Chances. Amsterdam University Press. This book explores the potential of an unconditional basic income, without means test or work requirement, com- pared to policies of subsidized employment. It also assesses the political chances of basic income in various European countries. https://www.aup.nl/en/book/9789048505029/basic- income-on-the-agenda Haagh, L. 2019. “The Developmental Social Contract and Basic Income in Denmark.” Social Policy and Society 18 (2): 301–17. This article discusses why steps toward basic income “from within” the state are institutionally plausible in Denmark, yet this transition is contested in Danish society. The author argues that implementation since the 1990s of the flexicurity regime—labor flexibility with social transfers and training—has stretched the tradi- tion for broadly inclusive reforms. Rising inequality and precarity in Denmark make a case for basic income within the public sector, but the source of this inequality—neo-liberalism—also puts it at risk. https://doi.org/10.1017/S1474746418000301 —. 2019. The Case for Universal Basic Income. Cambridge, UK: Polity Press. In this book, Haagh argues that universal basic income is essential to advancing freedom and democracy in the 21st century. Far from being a silver bullet that will transform or replace capitalism, a universal basic income can be implemented in a range of different contexts across the globe. http://politybooks.com/bookdetail/?isbn=9781509522958&subject_id=4 Haarmann, Claudia, Dirk Haarmann, Herbert Jauch, Hilma Shindondola-Mote, Nicoli Nattrass, Ingrid van Niekerk, and Michael Samson. 2009. “Making the Difference! The BIG in Namibia; Basic Income Grant Pilot Project Assessment Report April 2009.” Namibia NGO Forum. This report, one in a series of publications on the basic income grant in Namibia, reflects the results of the pilot project in Otjivero-Omitara. The report provides an overview of findings and possible lessons for national implementation of a basic income grant in Namibia. http://www.bignam.org/Publications/BIG_Assessment_report_08b.pdf Halmetoja, Antti, Jurgen De Wispelaere, and Johanna Perkiö. 2019. “A Policy Comet in Moominland? Basic Income in the Finnish Welfare State.” Social Policy and Society 18 (2): 319–30. Finland is widely considered a front-runner in the European basic income debate, primarily because of the decision to design and conduct the first national basic income exper- iment (2017–18). The Finnish basic income experiment builds on several decades of public and policy debate around the merits and problems of basic income, with the framing of basic income over time changing to fit the shift of the Nordic welfare state to embrace the activa- tion paradigm. The authors find several discrete, relatively small and unintended institutional 294 Appendix E: Annotated Bibliography developments that have arguably aligned the design of Finnish unemployment security closer to a partial basic income scheme. https://doi.org/10.1017/S1474746418000258 Häni, Daniel, and Philip Kovce. 2016. Voting for Freedom: The 2016 Swiss Referendum on Basic Income: A Milestone in the Advancement of Democracy. CreateSpace. Switzer- land is the first country ever to have voted on the implementation of an unconditional basic income (UBI). While the UBI was rejected at the ballot (only 23 percent of the population voted in favor of it), Häni and Kovce discuss why the Swiss referendum on basic income is a milestone in the advancement of democracy. https://books.google.com/books/about/Voting_ for_Freedom.html?id=spm_jwEACAAJ&source=kp_book_description Hanna, Rema, and Benjamin A. Olken. 2018. “Universal Basic Incomes vs. Targeted Trans- fers: Anti-Poverty Programs in Developing Countries.” NBER Working Paper 24939. National Bureau of Economic Research, Cambridge, MA. This paper examines the potential trade-offs between targeting transfers toward low-income households versus provid- ing a universal basic income. It starts by discussing how the fact that most households in poor countries do not pay income taxes changes how we conceptually think about universal basic income. It then analyzes data from two countries, Indonesia and Peru, to document the trade- offs involved. The results suggest that, despite the imperfections in targeting using proxy means tests, targeted transfers may result in substantially higher welfare gains than universal programs. However, targeted transfers do create an implied tax on consumption in the region where benefits are phased out. The authors discuss how alternative targeting approaches, such as community targeting and self-targeting, can be used to further improve targeting in some situations. https://www.nber.org/papers/w24939 Harris, Tom, David Phillips, Ross Warwick, Maya Goldman, Jon Jellema, Karolina Goraus, and Gabriela Inchauste. 2018. “Redistribution via VAT and Cash Transfers: An Assess- ment in Four Low and Middle Income Countries.” IFS Working Paper W18/11. Institute for Fiscal Studies, London. The authors use microsimulation methods to estimate the impact of preferential value-added tax (VAT) rates on revenues, poverty, inequality, and across the consumption distribution in four countries (Ethiopia, Ghana, Senegal, and Zambia). Despite being completely untargeted, a universal basic income (UBI) funded by the revenue gains from a broader VAT base would create large net gains for poor households and reduce inequal- ity and most measures of extreme poverty in each of the countries studied—even if only 75 percent of the additional VAT revenue was disbursed as UBI payments. https://www.ifs. org.uk/uploads/WP201811.pdf Harvey, Philip. 2005. “The Right to Work and Basic Income Guarantees: Competing or Complementary Goals.” Rutgers Journal of Law & Urban Policy 2 (1): 8–59. This arti- cle argues that basic income advocates have been too ready to reject the right to work, too willing to embrace the assumption that it cannot be secured by reasonable means, and too quick to conclude that a basic income guarantee would provide an adequate substitute for it. The author does not, however, reject the basic income idea. Rather than viewing it as a substitute for securing the right to work, he argues that it should be seen as a founda- tion for policies designed to secure what he refers to as the right to income support. https:// rutgerspolicyjournal.org/sites/jlpp/files/vol2issue1PhilipHarvery.pdf Healy, Seán, and Brigid Reynolds. 2012. “Ireland and the Prospects for Basic Income Reform.” In Basic Income Worldwide: Horizons of Reform, edited by Matthew Murray and Carole Pateman. London: Palgrave Macmillan. Ireland has produced a large number of reports, beginning in 1977, that include or recommend basic income, and in 2002 a gov- ernment Green Paper on the subject was issued. This chapter discusses the reports and their reception in some detail. The chapter closes with an argument for refundable tax credits as a major move toward a basic income, and reflects on the impact of Ireland’s major economic crisis on the debate about basic income. https://doi.org/10.1057/9781137265227_8 Henry, Malcolm. 2013. Our Money: How to Shrink Government, Boost Business, Eliminate Poverty and Make the Economy Work Properly for Everyone. QEI Projects. This book Appendix E: Annotated Bibliography 295 explains what money is, how we could use it to make the economy work better, and how to ensure that everyone has enough cash to meet their basic needs. http://www.qei.co.uk/our- money.html Hiilamo, Heikki, and Kathrin Komp. 2018. “The Case for a Participation Income: Acknowl- edging and Valuing the Diversity of Social Participation.” Political Quarterly 89 (2): 1–6. This article investigates how Atkinson’s idea of a participation income draws attention to the various ways in which people contribute to society. Current discussions on social partici- pation income is reflected in Danish, Finnish, German, and Dutch social policies. For instance, Denmark adopted a narrow focus on paid work; Finland seeks tailor-made solutions for the long-term unemployed; Germany reformed many policy areas to achieve better activation; and the Netherlands acknowledges a wide range of social participation forms. These country differences highlight that governments can adapt the idea of a participation income to achieve the focus they desire. https://doi.org/10.1111/1467-923X.12511 Hoynes, Hilary W., and Jesse Rothstein. 2019. “Universal Basic Income in the US and Advanced Countries.” NBER Working Paper 25538. National Bureau of Economic Research, Cambridge, MA. This paper discusses the potential role of universal basic income (UBI) within the existing constellation of programs in the United States. A UBI would direct much larger shares of transfers to childless, non-elderly, nondisabled households than existing programs, and much more to middle-income rather than poor households. A UBI large enough to increase transfers to low-income families would be enormously expen- sive. The authors also review the labor supply literature for evidence on the likely impacts of a UBI. They argue that the ongoing UBI pilot studies will do little to resolve the major outstanding questions. https://gspp.berkeley.edu/assets/uploads/research/pdf/Hoynes- Rothstein-UBI-081518.pdf Hughes, Chris. 2018. Fair Shot: Rethinking Inequality and How We Earn. New York: St. Mar- tin’s Press. This book tells the story of how the author, a cofounder of Facebook, came to embrace a guaranteed income program (not a UBI). He argues for cash transfer of US$500 a month to every adult who lives in a household that makes less than US$50,000 and who works in the formal economy, does caregiving at home, or is enrolled in school. This would be financed through a tax on the top 1 percent of earners, through (1) adjusting the tax code to eliminate the special tax rate on capital gains and dividends for those who make more than US$250,000, (2) capping deductions at 28 percent for the wealthiest Americans and closing tax loopholes, and (3) raising the tax rates on income above US$250,000 back to 50 percent (the historical average for much of the 20th century). https://www.bloomsbury.com/in/fair- shot-9781408899779/ International Monetary Fund (IMF). 2017. Fiscal Monitor: Tackling Inequality. Washington, DC: IMF. This issue of the Fiscal Monitor includes a section offering an insightful discussion on universal basic income (UBI), including in terms of historical ways of defining it, a policy analysis of choices and trade-offs, and a set of static microsimulations. Based on the analysis, the section devises a typology of countries based on coverage, incidence, and adequacy of UBI (pegged at a share of median income) relative to targeted transfers. https://www.imf.org/en/ Publications/FM/Issues/2017/10/05/fiscal-monitor-october-2017 Ipsos MORI. 2017. “Half of UK Adults Would Support Universal Basic Income in Principle.” Polling commissioned by the Institute for Policy Research, University of Bath. Nearly half of all adults age 18–75 in the United Kingdom (49  percent) would support the U.K. government’s introduction of a universal basic income (UBI) to cover basic needs. About a quarter (26 percent) would oppose it. However, public support drops down to 22 percent (and opposition would rise to 47 percent) if introducing a UBI meant increases in taxes and cuts in current welfare benefits. https://www.ipsos.com/ipsos-mori/en-uk/half-uk-adults-would- support-universal-basic-income-principle Jameson, Robert. 2016. The Case for a Basic Income. IMOS.org.uk. This book takes a pas- sionate view in favor of universal basic income (UBI) and exposes the supposed flaws of 296 Appendix E: Annotated Bibliography current criticism to the idea. The analysis focuses on potential benefits of UBI, how it could be afforded, and how it might best be introduced. http://imos.org.uk/ Jones, Damon, and Ioana Marinescu. 2018. “The Labor Market Impacts of Universal and Permanent Cash Transfers: Evidence from the Alaska Permanent Fund.” NBER Working Paper 24312. National Bureau of Economic Research, Cambridge, MA. Since 1982, all Alaskan residents have been entitled to a yearly cash dividend (a universal basic income, UBI) from the Alaska Permanent Fund. Using data from the Current Population Survey and a synthetic control method, the study shows that the dividend had no effect on employment and increased part-time work by 1.8 percentage points (or by 17 percent). Although theory and prior empirical research suggest that individual cash transfers can decrease household labor supply, the authors interpret the results as evidence that general equilibrium effects of widespread and permanent transfers tend to offset negative impacts, at least on the extensive margin. They also present suggestive evidence that tradable sec- tors experience employment reductions, while nontradable sectors do not. Overall, Alaska’s UBI does not significantly decrease aggregate employment. https://www.nber.org/papers/ w24312.pdf Jordan, Bill. 1988. “The Prospects for Basic Income.” Social Policy & Administration 22 (2): 115–23. This article provides an analysis of the prospects for the adoption of a universal basic income (UBI) policy in the United Kingdom in the late 1980s. In 1985–86, the chances for a policy seemed encouraging, with the formation of a pro-basic income movement attracting wide-ranging academics and politicians. However, by the time of publication, the idea’s political chances had become slim, with the British government looking at America, rather than Europe, for welfare policy reform. https://doi.org/10.1111/j.1467-9515.1988. tb00296.x —. 2011. “The Perils of Basic Income: Ambiguous Opportunities for the Implementa- tion of a Utopian Proposal.” Policy & Politics 39 (1): 101–14. Using the example of the tax benefit reforms scheduled to be introduced in the United Kingdom, the article argues that there are risks of undesirable consequences (in terms of justice and equal citizenship) from the adoption of partial variants of a universal basic income (UBI) scheme, but that they rep- resent necessary first steps toward a full UBI. https://doi.org/10.1332/030557311X546343 —. 2012. “The Low Road to Basic Income? Tax-Benefit Integration in the UK.” Jour- nal of Social Policy 41 (1): 1–17. This article discusses proposals for a partial integration of the United Kingdom’s tax and benefits systems. Such proposals are set out by the Centre for Social Justice in 2009 and endorsed by the new Secretary of State for Work and Pensions. The author argues that these proposals have some features of a universal basic income (UBI) scheme, and asks whether they would dismiss UBI or rather introduce it from the back door. https://doi.org/10.1017/S0047279411000353 Kalleberg, Arne. 2018. Precarious Lives: Job Insecurity and Well-Being in Rich Democ- racies. Cambridge, UK: Polity Press. This cross-national analysis shows how active labor market policies and generous social welfare systems can help protect workers and give employers latitude as they seek to adapt to the rise of national and global competition and rapid, sweeping technological changes. Such policies form elements of a new social contract that offers the potential for addressing many of the major challenges resulting from the rise of precarious work. https://www.wiley.com/en-us/Precarious+Lives%3A+Job+Insecurity+ and+Well+Being+in+Rich+Democracies-p-9781509506491 Kalleberg, Arne L., and Steven P. Vallas. 2018. “Probing Precarious Work: Theory, Research, and Politics.” Research in the Sociology of Work 31: 1–30. This article presents evidence on the proliferation of precarious work, both in the United States and other advanced nations, and discusses disparities in its spread and how this affects different groups in the labor force. It next explores the literature on how workers have responded to precarious work, both individually and collectively, and offers a framework to consider policy responses. http:// arnekalleberg.web.unc.edu/files/2018/01/Precarious-Work-CH-1.pdf Appendix E: Annotated Bibliography 297 Kangas, Olli, Signe Jauhiainen, Miska Simanainen, and Minna Ylikännö, eds. 2019. The Basic Income Experiment 2017–2018 in Finland. Preliminary Results. Helsinki: Ministry of Social Affairs and Health. This evaluation sets out the results from the two-year basic income pilot in Finland. Its extensive empirical results show that the scheme did not affect employment (a result that could be interpreted both as an encouraging and negative effect), while improving participants’ select qualitative dimensions of well-being. http://julkaisut. valtioneuvosto.fi/bitstream/handle/10024/161361/Report_The%20Basic%20Income%20 Experiment%2020172018%20in%20Finland.pdf?sequence=1&isAllowed=y Kennedy, Joseph V. 2008. Ending Poverty: Changing Behavior, Guaranteeing Income, and Transforming Government. New York: Rowman & Littlefield Publishers. This book pro- poses an aggressive, conservative, governmental reform plan that is institutionalized through an income contract between the individual and the government. Based on three main principles— government policy is more effective when it channels market forces than when it overrides them, individuals should have more control over the resources government spends on their behalf, and access to a decent life free of poverty is a conditional right in American society—the book sets out reform guidelines for education, health care, housing, retirement savings, and taxa- tion. https://books.google.com/books/about/Ending_Poverty.html?id=O90GOat_eDcC&hl=en Khosla, Saksham. 2018. India’s Universal Basic Income: Bedeviled by the Details. New Delhi: Carnegie India. This report discusses proposals for universal basic income (UBI) in India, with a particular focus on the UBI proposal outlined in the Ministry of Finance’s 2016– 17 Economic Survey. The report argues that the Economic Survey deserves praise for bringing substantial rigor to the debate, thrusting a UBI into the national spotlight, and prudently con- cluding that the time has not yet come for implementation. However, should future Indian policy makers wish to implement a UBI, the survey’s central design features offer a weak foundation. If enacted without deeper analysis, debate, or sufficient evidence demonstrating improvement in development outcomes, the Economic Survey’s blueprint for an Indian UBI will produce underwhelming results. https://bit.ly/2OwVkmD King, Martin Luther. 1967. Where Do We Go from Here: Chaos or Community? New York: Beacon Press. In 1967, King isolated himself from the demands of the civil rights movement, rented a house in Jamaica with no telephone, and labored over his final manuscript. In this work, he lays out his thoughts, plans, and dreams for America’s future, including the need for better jobs, higher wages, decent housing, and quality education. “I’m now convinced that the simplest approach will prove to be the most effective—the solution to poverty is to abolish it directly by a now widely discussed measure: the guaranteed income,” Dr. King writes. http:// www.beacon.org/Where-Do-We-Go-from-Here-P1376.aspx Korpi, Walter, and Joakim Palme. 1998. “The Paradox of Redistribution and Strategies of Equality: Welfare State Institutions, Inequality, and Poverty in the Western Coun- tries.” American Sociological Review 63 (5): 661–87. The article argues that traditional arguments in favor of targeting and flat-rate benefits neglect three policy-relevant consider- ations. First, the size of redistributive budgets is not fixed but reflects the structure of welfare state institutions. Second, a trade-off exists between the degree of low-income targeting and the size of redistributive budgets. And third, outcomes of market-based distribution are often more unequal than those of earnings-related social insurance programs. The authors’ conclu- sions are that the more benefits are targeted to the poor, as opposed to a broader approach encompassing wider populations, the less likely societies are to reduce poverty and inequality. https://www.jstor.org/stable/2657333 Lavinas, Lena. 2013. “Brazil: The Lost Road to Citizen’s Income.” In Citizen’s Income and Welfare Regimes in Latin America: From Cash Transfers to Rights, edited by Rubén Lo Vuolo. New York: Palgrave Macmillan. This essay reflects on the impasses in the imple- mentation of citizen’s income in Brazil. While the country represents an exception in adopting citizen’s income by law, its experience raises questions on the attainability of moving from conditional cash transfer programs to the unconditional, universal citizen’s income. https:// www.palgrave.com/us/book/9780230338210 298 Appendix E: Annotated Bibliography Lee, Soomi. 2018. “Attitudes toward Universal Basic Income and Welfare State in Europe: A Research Note.” Basic Income Studies 13 (1): 101–9. This research note examines the relationship between public attitudes toward universal basic income (UBI) and country-level socioeconomic conditions in 21 European countries. Despite abundant theoretical and empir- ical research on UBI, a comparative analysis of public appetite for it has been unavailable due to data limitations. This research note takes advantage of the 2016 European Social Survey to explore the connection between public support for UBI and levels of social protection and economic insecurity. http://dx.doi.org/10.2139/ssrn.3209161 LeGrand, Julian. 2006. “Implementing Stakeholder Grants: The British Case.” In Rede- signing Redistribution: Basic Income and Stakeholder Grants as Cornerstones for an Egalitarian Capitalism, edited by Erik Olin Wright. New York: Verso. Britain is a lead- ing country in the process of implementing stakeholder grants. The government of Tony Blair committed itself to introducing a Child Trust Fund. This is a universal grant, topped up by a means-tested addition, given to every baby at birth, invested in a fund, and available for use when the child reaches 18 years of age. Together with earlier contributions to the British debate, this note explores theoretical and practical issues around emerging experi- ences with this particular kind of grants. https://www.ssc.wisc.edu/~wright/Redesigning%20 Distribution%20v1.pdf Liebermann, Sascha. 2012. “Germany: Basic Income in the German Debate.” In Basic Income Worldwide: Horizons of Reform, edited by Matthew Murray and Carole Pateman, 173–99. London: Palgrave Macmillan. The chapter begins by tracing a brief his- tory of the current basic income debate in Germany. It then turns to its precursors in the 1980s and provides some reasons why a basic income might have been off the agenda in the 1990s following German Reunification. It then presents the main arguments and objections to basic income, followed by a discussion on the prospects for basic income in Germany. https:// doi.org/10.1057/9781137265227_9 Livingston, James. 2016. No More Work: Why Full Employment is a Bad Idea. Chapel Hill: University of North Carolina Press. Livingston explains how and why Americans still cling to work as a solution rather than a problem—why it is that both liberals and conservatives announce that “full employment” is their goal when job creation is no longer a feasible solu- tion for moral or economic hurdles. The book encourages us to reimagine ways of finding meaning, character, and sustenance beyond our workaday world—and shows us that we can afford to leave that world behind. https://muse.jhu.edu/book/48045 Lord, Clive, and Miriam Kennet. 2011. Citizen’s Income and Green Economics. Oxford, UK: Green Economics Institute. This book attempts to weave several different strands whose importance is only apparent when considered together. In particular, it aims to provide a basis for a new approach to ecological problems, including with a universal basic income at its core. https://www.greeneconomicsinstitute.com/product-page/a-citizen-s-income-and- green-economics Lo Vuolo, Rubén, ed. 2013. Citizen’s Income and Welfare Regimes in Latin America: From Cash Transfers to Rights. New York: Palgrave Macmillan. Social protection systems in Latin America developed in a context of widespread informal and precarious work. As such, social insurance institutions could only provide limited coverage. A citizen’s income policy, the book argues, is not only a cash transfer to alleviate poverty or food security, but a basic right to improve democracy and empower people living in unequal societies. https://www. palgrave.com/us/book/9780230338210 Lowrey, Annie. 2018. Give People Money: How a Universal Basic Income Would End Poverty, Revolutionize Work, and Remake the World. New York: Crown. In this book, Lowrey exam- ines the universal basic income (UBI) movement from many angles. She explores the potential of such a sweeping policy and the challenges the movement faces, among them contradictory aims, uncomfortable costs, and, most powerfully, the entrenched belief that no one should get something for nothing. In the end, she shows how a UBI has the potential to solve some of our Appendix E: Annotated Bibliography 299 most intractable economic problems, while offering a new vision of citizenship and a firmer foundation for our society in this age of turbulence. https://www.penguinrandomhouse.com/ books/551618/give-people-money-by-annie-lowrey/9781524758769/ Marinescu, Iona. 2018. “No Strings Attached: The Behavioral Effects of U.S. Unconditional Cash Transfer Programs.” NBER Working Paper 24337. National Bureau of Economic Research, Cambridge, MA. This review covers the evidence on universal basic income (UBI) and discusses its potential impact in the United States. In particular, it reviews rel- evant initiatives, including small- and larger-scale pilots and programs. Available evidence finds no statistically significant effect of an unconditional cash transfer on labor supply. Stud- ies show that an unconditional cash transfer can improve health and educational outcomes, and decrease crime, drug, and alcohol use, especially among the most disadvantaged youths. https://doi.org/10.3386/w24337 Martin, Danielle. 2017. Better Now: Six Big Ideas to Improve Health Care for All Canadians. Toronto: Penguin Random House. Martin uses real patient stories to illustrate what works in the Canadian health care system, and what does not. Most importantly, she proposes six ideas to make health care achievable and affordable, including creating a guaranteed income program. https://www.penguinrandomhouse.com/books/541501/better-now-by-dr-danielle- martin/9780735232617/ Martinelli, Luke. 2017. “Assessing the Case for a Universal Basic Income in the UK.” IPR Policy Brief. Institute for Policy Research, University of Bath, Bath. This report reviews core theoretical and empirical literature on universal basic income, assesses its feasibil- ity as a realistic proposal in the United Kingdom, and considers various options regarding policy design and implementation strategies. The paper also includes original results from microsimulations. https://www.bath.ac.uk/publications/assessing-the-case-for-a-universal- basic-income-in-the-uk/attachments/basic_income_policy_brief.pdf Martinelli, Luke, and Nick Pearce. 2019. “Basic Income in the UK: Assessing Prospects for Reform in an Age of Austerity.” Social Policy and Society 18 (2): 265–75. While the United Kingdom has large shares of precarious and low-paid workers, significant barriers remain to the emergence of a large and coherent constituency of support for universal basic income (UBI). However, options falling short of a full UBI—but retaining some of its core features— appear relatively feasible. https://doi.org/10.1017/S147474641800026X Marx, Axel, and Hans Peeters. 2008. “An Unconditional Basic Income and Labor Supply: Results from a Pilot Study of Lottery Winners.” Journal of Socio-Economics 37: 1636–59. This article sets out results from a pilot survey of specific types of lottery winners to investi- gate the consequences of introducing a universal basic income (UBI). In particular, analysis on Belgian Win for Life lottery winners points to no extreme consequences of introducing a UBI, with very few changes with regard to exiting work, diminishing working time, or becoming self-employed. https://doi.org/10.1016/j.socec.2007.08.007 Mays, Jenni, Greg Marston, and John Tomlinson, eds. 2016. Basic Income in Australia and New Zealand: Perspectives from the Neoliberal Frontier. New York: Palgrave Mac- millan. This book asks whether a universal basic income (UBI) offers a viable solution to the income support in Australia and New Zealand. The contributors’ essays and case studies explore the historical basis on which a UBI program might stand in these two countries, the ideological nuances and complexities of implementing such a policy, and ideas for future regional and global applications. https://www.palgrave.com/gb/book/9781137535313 McGuirk, Eoin, Anand Rajaram, and Marcelo Giugale. 2016. “The Political Economy of Direct Dividend Transfers in Resource-Rich Countries: A Theoretical Consideration.” Policy Research Working Paper 7575. World Bank, Washington, DC. The acceleration of natural resource discoveries across many parts of the developing world has highlighted the urgent need for solutions to the mismanagement of windfalls that has blighted many countries over the past half-century. One proposal involves distributing annually a share of 300 Appendix E: Annotated Bibliography resource rents to citizens in the form of direct dividend transfers. This paper models the deci- sion of political leaders to allocate resource revenues between cash transfers, public goods, power-preserving activities, and personal consumption. The analysis finds first that conducive political conditions, including competitive elections, limited patronage networks, and a high degree of budgetary accountability, increase the share of resource revenues to be spent on cit- izens’ welfare. The paper then shows that high poverty and inefficient public institutions will each strengthen the political incentive to provide direct dividend transfers relative to public goods. This combination of conditions is rare, which may explain why relatively few coun- tries have implemented or plan to implement direct dividend transfers. http://documents. worldbank.org/curated/en/552901468186276244/pdf/WPS7575.pdf McKay, Ailsa. 2001. “Rethinking Work and Income Maintenance Policy: Promoting Gender Equality through a Citizens’ Basic Income.” Feminist Economics 7 (1): 97–118. Imple- mentation of universal basic income (UBI) would bring into question the institutionalized relationship between work and welfare, a basic premise of modern welfare states. Yet social policy reforms should take account of all gender inequalities and not just those relating to the traditional labor market. This article argues that UBI can be a tool for promoting gender-neu- tral social citizenship rights, but that any future “marriage” of justice and efficiency must first “divorce” work from income. https://doi.org/10.1080/13545700010022721 —. 2007. “Why a Citizen’s Basic Income? A Question of Gender Equality or Gender Bias.” Work, Employment and Society 21 (2): 337–48. This article highlights the gen- der-blind nature of universal basic income (UBI) debates, and presents a case for a UBI based on a more inclusive notion of citizenship. https://doi.org/10.1177/0950017007076643 McKay, Ailsa, and Jo Venevery. 2000. “Gender, Family, and Income Maintenance: A Feminist Case for Citizens Basic Income.” Social Politics 7 (2): 266–84. The authors con- ceptualize welfare provision as a partnership between state, labor market, and family. The latter element is the focus of the article: drawing on feminist critiques of the male breadwin- ner family and the need to develop welfare policy suitable for a “post-familial” society, this article argues that a universal basic income has the potential to provide a basis for a truly uni- versal citizenship. https://doi.org/10.1093/sp/7.2.266 Mea, ‘A’ohe. 2018. Where We Go from Here: Chaos to Community. A Modest Proposal for the Livable Income Security Act of 2019. Creative Commons License 4.0. This book is the result of intensively studying Dr King’s intentions and strategies, and offers a road map for putting in practice his vision, including actively exploring universal basic income as a central topic. https://books.google.co.uk/books/about/Where_We_Go_from_Here_Chaos_to_ Community.html?id=AzOevQEACAAJ&source=kp_book_description&redir_esc=y Miller, Annie. 2017. A Basic Income Handbook. Edinburgh: Luath Press Ltd. This book argues that the current social security system in the United Kingdom is complex, unjust, and unfit for purpose. In this regard, a universal basic income redefines the relationship between the state, society, and the individual. https://www.luath.co.uk/politics-and-current-issues/a-basic- income-handbook Moffitt, Robert. 2003. “The Negative Income Tax and the Evolution of U.S. Welfare Policy.” NBER Working Paper 9751. National Bureau of Economic Research, Cambridge, MA. This essay reviews Friedman’s proposal for a negative income tax and outlines what he saw to be its chief advantages. It then considers two challenges, namely its purported work incen- tives and the possible superiority of work requirements. The latter in particular have gained ground in actual U.S. welfare policy over the last 30 years, a development counter to the neg- ative income tax. On the other hand, the earned income tax credit has produced a negative income tax–like program on a vast scale. https://www.nber.org/papers/w9751.pdf Munger, Michael C. 2018. Tomorrow 3.0: Transaction Costs and the Sharing Economy. Cambridge, UK: Cambridge University Press. With the growing popularity of apps such as Uber and Airbnb, there has been a keen interest in the rise of the sharing economy. Appendix E: Annotated Bibliography 301 Munger shows how these technologies are reducing transaction costs, rather than reducing the costs of products themselves. He predicts that smartphones will be used to commod- ify excess capacity, and a universal basic income will be required as a consequence of this new transaction costs revolution. https://www.cambridge.org/gb/academic/subjects/ management/entrepreneurship-and-innovation/tomorrow-30-transaction-costs-and-sharing- economy?format=PB Murray, Charles. 2016. In Our Hands: A Plan to Replace the Welfare State. Washington, DC: American Enterprise Institute Press. Murray proposes a plan for the United States to replace all targeted transfer programs with a single, universal basic income (UBI) of US$13,000 per year for every (non-incarcerated) citizen age 21 and older. Some details of the proposal include the following: US$3,000 must be devoted to catastrophic health insurance; for those with a personal income above US$30,000, a surtax would be levied that reimburses the grant incrementally up to a maximum of half its total (i.e., people making US$60,000 or more would get a UBI of US$6,500); and programs to be replaced include social security, Medicare, Medicaid, welfare programs, social service programs, agricultural subsidies, and corporate welfare. Murray argues that this UBI would be cheaper than the current welfare system, and by 2020 could cost close to US$1 trillion less per year. The book states that the greatness of the American project was that it set out to let everyone live life as each person saw fit, and he sees UBI as the best means to reach this goal. A UBI policy, he argues, is politically inevi- table, based on two historical forces: (1) America’s continually rising gross domestic product will make poverty and deprivation less and less acceptable, and (2) government’s inherent limitations in trying to manage complex human needs. http://www.aei.org/publication/in-our- hands-a-plan-to-replace-the-welfare-state/ Murray, Matthew, and Carole Pateman, eds. 2012. Basic Income Worldwide: Horizons of Reform. London: Palgrave Macmillan. This book discusses existing examples of universal basic income in both high- and low-income countries, and considers its prospects in other places around the world. https://www.palgrave.com/gb/book/9780230285422 Murray, Michael L. 1997. …and Economic Justice for All: Welfare Reform for the 21st Cen- tury. New York: Routledge. The author presents an argument for a system of social insurance that replaces welfare with a guaranteed adequate income. The book reviews public assis- tance programs and evaluates other plans that have been proposed. https://www.routledge. com/and-Economic-Justice-for-All-Welfare-Reform-for-the-21st-Century-1st/Murray/p/ book/9781563249891 Nell, Guinevere Liberty, ed. 2013. Basic Income and the Free Market: Austrian Economics and the Potential for Efficient Redistribution. London: Palgrave Macmillan. This com- pilation of essays discusses whether universal basic income could offer an alternative to both laissez-faire and existing welfare systems in high-income countries. https://www.palgrave. com/gb/book/9781137263582 Noguera, José A. 2019. “The Political Debate on Basic Income and Welfare Reform in Spain.” Social Policy and Society 18 (2): 289–99. Two factors have boosted the recent debate on universal basic income (UBI) in Spain: on the one hand, the combination of welfare budget cuts and growing poverty rates has spread claims for a radical reform of Spanish wel- fare policies; on the other hand, the emergence of Podemos as a new key actor in the Spanish political arena has generated a vivid discussion on UBI proposals. Noguera argues that the radical rejection of means testing by basic income defenders fails to grasp the different types and degrees of conditions an income guarantee system may establish. The main lesson from the Spanish experience is that UBI supporters should be ready to compromise and accept gen- erous means-tested guaranteed income programs as steppingstones. https://doi.org/10.1017/ S1474746418000271 Noguera, José A., and Jurgen De Wispelaere. 2006. “A Plea for the Use of Laboratory Experiments in Basic Income Research.” Basic Income Studies 1 (2): 1–8. In recent years, 302 Appendix E: Annotated Bibliography interest in conducting a European universal basic income (UBI) experiment, along the lines of the famous negative income tax experiments in the United States and Canada in the 1970s and 1980s, has grown considerably. In this short commentary, the authors caution against embarking on this route. They offer two reasons against the use of field experiments in UBI research: its vulnerability to political manipulation and its incapacity to offer real answers to the pressing questions that need to be addressed before a basic income scheme can be implemented. Instead, they argue that laboratory experiments carefully mimicking selec- tive social situations might offer crucial insight in the operation of UBI schemes. https://doi. org/10.2202/1932-0183.1044 O’Brien, Paul. 2017. Universal Basic Income: The Irish Context. Stroud, UK: The History Press. O’Brien explains how a universal basic income might work in an Irish context—and how these arguments are forming the basis of what might redefine the Irish social protection system. https://www.thehistorypress.co.uk/publication/universal-basic-income/9781845883676/ Offe, Claus. 2001. “Pathways from Here.” In What’s Wrong with a Free Lunch?, edited by Philippe Van Parijs. Boston: Beacon Press. Offe considers the opposition to universal basic income (UBI) and obstacles that have prevented it from being fully implemented in the Euro- pean Union. He discusses the concerns that UBI can provoke among employers, employees, citizens, and corporate actors. Offe calls for UBI proponents to take these fears seriously, and advocates the principles of gradualism and reversibility to provide a vehicle for people to change their preferences. Yet there is, in his view, one dimension in which gradualism is not feasible: single-country implementation in the European Union. According to Offe, such uni- lateralism is likely to trigger migration effects that are bound to undermine the political and economic viability of any even less-than-complete solution. http://www.beacon.org/Whats- Wrong-With-A-Free-Lunch-P132.aspx —. 2008. “Basic Income and the Labor Contract.” Basic Income Studies 3 (1): 1–27. The article argues that universal basic income (UBI) would help address unemployment, pov- erty, and disempowerment. Next, it provides a rebuttal to the main arguments raised by UBI opponents, namely that the “idle should not be rewarded,” the “prosperous do not need it,” and “there are so many things waiting to be done in the world.” https://doi.org/10.1515/auk- 2009-0103 Ortiz, Isabel, Christina Behrendt, Andrés Acuña-Ulate, and Quynh Anh Nguyen. 2018. “Universal Basic Income Proposals in Light of ILO Standards: Key Issues and Global Costing.” ESS Working Paper 62. International Labour Office, Geneva. This paper shows that the International Labour Organization’s (ILO) Social Protection Floors Recommendation (No. 202) includes principles that are highly relevant to guide the universal basic income (UBI) debate. These include (1) adequacy and predictability of UBI benefits to ensure income security, set at least at the national poverty line; (2) social inclusion, including of persons in the informal economy; (3) social dialogue and consultation with stakeholders; (4) enactment of national laws regulating UBI entitlements, including indexation of benefits; (5) coherence with other social, economic, and employment policies; and (6) sustainable and equitable financing. Based on criteria such as replacement of programs, level of benefits, and source of funding, the paper shows that some models of UBI can be in accordance with ILO standards, while others are not. https://www.ilo.org/wcmsp5/groups/public/---ed_protect/---soc_sec/ documents/publication/wcms_648602.pdf Osterkamp, Rigmar. 2013. “The Basic Income Grant Pilot Project in Namibia: A Critical Assessment.” Basic Income Studies 8 (1): 71–91. Since January 2008, a monthly basic income grant of N$100 (about US$15) has been paid to around 1,000 citizens of the Namibian village of Otjivero on a pilot basis. (In January 2010, it was reduced to N$80.) The design and implementation of the pilot, it is argued, are not in line with standards of empirical socioeco- nomic research; also, the project did not address political issues, which are key for scale-up, adequately. The article concludes by drawing recommendations for possible future UBI pilot projects. https://doi.org/10.1515/bis-2012-0007 Appendix E: Annotated Bibliography 303 Parolin, Zachary, and Linus Siöland. 2019. “Support for a Universal Basic Income: A Demand-Capacity Paradox?” Working Paper 19.01. Herman Deleeck Centre for Social Policy, University of Antwerp, Antwerp. This paper investigates the determinants of support for a universal basic income (UBI). Using data from the 2016 European Social Survey (ESS8), the authors find that the relative size of the welfare states has a large effect on UBI support (higher levels of support for UBI can be found in countries with less developed welfare states). This presents a demand-capacity paradox: countries that are presumably best equipped to implement a UBI also see the least support for the policy. Also, where welfare state spending is low, welfare state chauvinism and left ideology have little effect on UBI sup- port. https://doi.org/10.31219/osf.io/fvh92 Pateman, Carole. 2004. “Democratizing Citizenship: Some Advantages of a Basic Income.” Politics & Society 32 (1): 89–105. This article argues that if the focus or primary objective of universal basic income (UBI) is democratization, including women’s freedom, then a UBI is preferable to stakeholding approaches. https://doi.org/10.1177/0032329203261100 Peeters, Hans, and Axel Marx. 2006. “Lottery Games as a Tool for Empirical Basic Income Research.” Basic Income Studies 1 (2): 1–6. The authors survey the behavior of Win for Life (W4L) lottery winners to help explore universal basic income’s (UBI) labor market effects. They argue that lifelong receipt of the W4L grant produces more accurate evidence of behav- ioral changes than a short-term basic income experiment. They also note that W4L lottery games exist in several countries, hence results from W4L research in multiple contexts can offer valuable insights into the interaction of a basic income with local institutions. https://doi. org/10.2202/1932-0183.1040 Pereira, Richard, ed. 2017. Financing Basic Income: Addressing the Cost Objection. London: Palgrave Macmillan. This book argues that universal basic income (UBI) at an adequate level is affordable. The contributors approach the topic from the perspectives of three different countries—Australia, Canada, and Switzerland—to overcome objections that a universal program to keep all citizens above the poverty line would be too expensive to imple- ment. They assess the complex array of revenue sources that can make UBI feasible, from reducing welfare program redundancies to publicly owned assets. https://www.palgrave.com/ gb/book/9783319542676 Ravallion, Martin. 2018. “Guaranteed Employment or Guaranteed Income?” CGD Work- ing Paper 482. Center for Global Development, Washington, DC. This paper critically reviews the arguments for and against both employment guarantees and universal basic income (UBI) programs, with special reference to India. Evidence on India’s National Rural Employment Guarantee Act shows that work is often rationed by local leaders in poor areas, and the poverty impact is small when all the costs are considered. Yet it cannot be pre- sumed that switching to a UBI will reduce poverty more than workfare or finely targeted transfers. Nonetheless, more incentive-neutral, universal, and/or state-contingent transfer schemes merit consideration in settings in which existing public spending is skewed against poor people and/or there is scope for raising taxes on the rich. http://dx.doi.org/10.2139/ ssrn.3208865 Raventós, Daniel. 2007. Basic Income: The Material Conditions of Freedom. London: Pluto Press. This book is a guide to universal basic income—what it is, how to manage it, and how it can benefit the majority in different spheres of their lives. https://www.jstor.org/stable/j. ctt18fsbw8 Raventós, Daniel, and Julie Wark. 2018. Against Charity. Chico, CA: AK Press/Counter- punch. Raventós and Wark argue for a universal basic income above the poverty line, and paid for by progressive taxation, to both eradicate poverty and empower recipients. https:// www.akpress.org/against-charity.html Reich, Robert B. 2015. Saving Capitalism: For the Many, Not the Few. New York: Vintage. The book asserts that the critical choice ahead in the United States is one between a market 304 Appendix E: Annotated Bibliography organized for broad-based prosperity and one designed to deliver ever more gains to the rich- est in society. He presents the path that, he argues, needs to be taken to restore America’s fundamental promise of opportunity and advancement, with a universal basic income emerg- ing as an important element for realizing such vision. https://www.penguinrandomhouse. com/books/227780/saving-capitalism-by-robert-b-reich/9780345806222/ Reza Farzin, Mohammad, Dominique M. Guillaume, and Roman Zytek. 2011. “Iran—The Chronicles of the Subsidy Reform.” Working Paper 11/167. International Monetary Fund, Washington, DC. On December 18, 2010, the Islamic Republic of Iran increased domestic energy and agricultural prices by up to 20 times, making it the first major oil-export- ing country to reduce substantially implicit energy subsidies. This paper reviews the economic and technical issues involved in the planning and early implementation of the reform, includ- ing the distribution of compensatory transfers in the form of a universal basic income to all Iranians. The paper delves into the critical role played by public communication campaigns and concludes by laying out main challenges for the second phase of the reform. https://ssrn. com/abstract=1888910 Rogers, Brishen, Philippe Van Parijs, Dorian Warren, Tommie Shelby, and Diane Coyle. 2017. Work, Inequality, Basic Income: Boston Review Forum 2, Cambridge, MA. Some of the volume’s contributors explore universal basic income’s (UBI’s) merits, not only as an antipov- erty tool, but as a path toward justice and equality. Others, more skeptical, highlight the perils of a UBI designed without attention to workers’ power and quality of work. Together, they offer a nuanced debate about what it will take to tackle inequality. https://store.bostonreview. net/backissues/work-inequality-basic-income Ruckert, Arne, Chau Hunh, and Ronald Labonté. 2017. “Reducing Health Inequities: Is Uni- versal Basic Income the Way Forward?” Journal of Public Health 40 (1): 3–7. The authors argue that a universal basic income (UBI) has the potential to sustainably reduce health inequi- ties. The article first focuses on what is known about UBI’s impact on a range of specific health pathways, including birth outcomes, educational attainment, housing, and mental health. Given the existence of various UBI models, it then discusses some of the advantages and drawbacks of different approaches to UBI from a health equity perspective. It finally addresses some of the main criticisms that have been raised about UBI in relation to work disincentives, gendered impacts, and economic costs. https://doi.org/10.1093/pubmed/fdx006 Salehi-Isfahani, Djavad. 2014. “Iran’s Subsidy Reform: From Promise to Disappointment.” Policy Perspective 13. Economic Research Forum, Giza, Egypt. This paper explores the reasons why the Islamic Republic of Iran’s subsidy reform program was downsized (although still reaching most of the population) after its initial successful implementation. The author argues that the government set the level of cash transfers well above new revenues from the price increases, and printed money to pay for the deficit. The resulting inflation eroded public support for the program and caused the Parliament to freeze further price adjustments. Moreover, about a year and a half after the program began, international sanctions targeting Iranian oil exports and the country’s access to global trade tightened considerably, caused oil exports to fall by half, and disrupted industrial production. The national currency collapsed, and prices spiraled out of control. The resulting economic crisis further eroded public support. The main lessons from the Iranian experience with energy price reform are that cash transfers are an important part of the reform package—they can reduce poverty and inequality while increasing the reform’s general acceptance—and they must be self-financing and not depend on the general budget. http://erf.org.eg/publications/irans-subsidy-reform-from-promise-to- disappointment/ Salehi-Isfahani, Djavad, and Mohammad H. Mostafavi-Dehzooei. 2018. “Cash Transfers and Labor Supply: Evidence from a Large-Scale Program in Iran.” Journal of Devel- opment Economics 135: 349–67. This article examines the impact of a universal basic income program on labor supply in the Islamic Republic of Iran. In 2011, the country started monthly deposits of cash into individual accounts covering more than 70 million people and Appendix E: Annotated Bibliography 305 amounting to 28 percent of the median per capita household income. The article uses panel data to study the causal effect of the transfers on labor supply using exogenous variation in the time households started receiving transfers and in the intensity of treatment (defined as the share of net transfers from the program in total per capita household expenditures). The authors find no evidence that cash transfers reduced labor supply, in terms of hours worked or labor force participation. To the contrary, they find positive effects on the labor supply of women and self-employed men. https://doi.org/10.1016/j.jdeveco.2018.08.005 Schutz, Robert R. 1996. The $30,000 Solution: A Guaranteed Annual Income for Every American. McKinleyville, CA: Fithian Press. Schutz puts forward an ambitious plan for a minimum income of US$30,000 per year for every adult in the United States. The primary financing tool in his proposal is to tax all unearned income (including all rent, profits, returns on capital) at 100 percent. In this book, he outlines the many social, economic, and ecological problems that he believes such a policy would solve. https://books.google.co.uk/books/about/ The_30_000_Solution.html?id=ys_EAAAACAAJ&redir_esc=y Shafarman, Steven. 2017. Basic Income Imperative: For Peace, Justice, Liberty, and Per- sonal Dignity. Bookbaby. This book makes the case for a universal basic income in the United States, arguing that it will be a “monthly reminder” about citizenship, equality, unity, and cooperation. https://www.booktopia.com.au/basic-income-imperative-steven-shafarman/ prod9781543902044.html Sheahen, Allan. 2012. Basic Income Guarantee: Your Right to Economic Security. London: Palgrave Macmillan. The debate on universal basic income (UBI) in the United States is fraught with questions: why should the county adopt a UBI? Can the United States afford it? Why do current welfare programs not work? Why not guarantee everyone a job? Would anyone work if his or her income were guaranteed? Has a basic income guarantee ever been tested? The book explores these questions and provides support to the notion that a UBI would likely be a superior option than alternative policy measures in the United States. https:// www.palgrave.com/gb/book/9781137005700 Simpson, Wayne, Greg Mason, and Ryan Godwin. 2017. “The Manitoba Income Exper- iment: Lessons Learned 40 Years Later.” Canadian Public Policy 43 (1): 85–104. The Manitoba Basic Annual Income Experiment (Mincome), conducted some 40 years ago, was an ambitious social experiment designed to assess a range of behavioral responses to a neg- ative income tax, a specific form of guaranteed annual income. This article reviews that experiment, clarifying what exactly Mincome did and did not learn about how individuals and households responded to the income guarantees. It discusses the relevance of Mincome to modern-day income experiments and describes how researchers may access these valu- able data. https://doi.org/10.3138/cpp.2016-082 Smith, Justin R. 2017. Basic Income Policy and National Debts: The Basic Blueprint for a Global Planned Economy. This book lays out a universal basic income plan for citizens contingent on employment, with exceptions only for the young or disabled. It also discusses the thorny issue of debt and provides a path out of burdensome lending. https://books.google.com/books/ about/Basic_Income_Policy_and_National_Debts.html?id=ORHOtAEACAAJ&source=kp_ book_description Sommer, Maximilian. 2016. A Feasible Basic Income Scheme for Germany: Effects on Labor Supply, Poverty, and Income Inequality. Heidelberg: Springer. This book analyzes the consequences that would arise if Germany’s means-tested unemployment benefits were replaced with a negative income tax (calibrated to be both financially feasible and compatible with current constitutional legislation). Using data from the German Socio-Economic Panel, Sommer examines the impact of the reform on households’ labor supply as well as on both poverty and inequality measures. By applying reasonable values for both the benefits and the implied marginal tax rate imposed on earned incomes, efficiency gains can be reconciled with generally accepted value statements. Furthermore, as the proposal is geared for families, child poverty could be reduced considerably. https://www.springer.com/gb/book/9783319240626 306 Appendix E: Annotated Bibliography Standing, Guy. 2002. Beyond the New Paternalism: Basic Security as Equality. London/ New York: Verso. This book argues that the era of market regulation has ended in an era of fiscal regulation: new social and economic insecurities have spread around the world, boosted by globalization and flexible labor markets, and compounded by privatization and increased selectivity of social policy. This global insecurity has spawned growing and vastly underes- timated inequalities. To overcome these seemingly endemic insecurities and inequalities, Standing argues for a complex egalitarianism, in which universal basic income is recognized as a right for all. Work (including voluntary, community, and care work), and not labor, should be the basis of a good society, and policies should be judged by their capacity to promote occupational security. https://books.google.com/books/about/Beyond_the_New_Paternalism. html?id=0zSjF0Vrk5QC —. 2008. “How Cash Transfers Promote the Case for Basic Income.” Basic Income Studies 3 (1): 1–30. This article reviews recent evidence on various types of schemes imple- mented in developing countries, including several pilot cash transfer schemes, assessing them through the lens of social justice. It concludes that experiences with cash transfers are strengthening the case for a universal basic income. https://doi.org/10.2202/1932-0183.1106 —. 2011. The Precariat: The New Dangerous Class. London: Bloomsbury. Standing offers an account of an emerging class of people facing insecurity, moving in and out of precarious work that gives little meaning to their lives. He warns that the growth of the pre- cariat is producing instabilities in society. Its internal divisions have led to the villainization of migrants and other vulnerable groups, and some are susceptible to the dangers of political extremism. Standing argues for a new politics that puts the fears and aspirations of the precar- iat at the heart of a progressive strategy of redistribution and income security. https://www. bloomsbury.com/uk/the-precariat-9781849664561/ —. 2011. “Responding to the Crisis: Economic Stabilisation Grants.” Policy & Politics 39 (1): 9–25. This article reviews some of the mainstream policies proposed to tackle the eco- nomic crisis of 2008–09 and its aftermath, and goes on to advocate a policy of economic stabilization grants. It argues that such grants, which would be paid to every citizen at a rate that could be varied according to the severity of the crisis, would be more effective in boosting aggregate demand and more efficient in terms of resource allocation. Unlike the alternatives, economic stabilization grants would also directly address two key issues deriving from the process of globalization, namely the growth of systemic uncertainty and rising inequality. https://doi.org/10.1332/030557311X546389 —. 2013. “Why a Basic Income Is Necessary for a Right to Work.” Basic Income Stud- ies 7 (2): 19–40. This article makes the proposition that a right to work can only exist if an individual has a prior right to a universal basic income. It criticizes the perspective that maxi- mizing the number of jobs is a meaningful way of advancing the right to work, since activity in subordinated labor is scarcely consistent with a freedom-enhancing right to work. In recall- ing the historical right to practice an occupation, it rejects the notion of a “job guarantee,” as neither feasible nor desirable in a free society or as part of a progressive vision of a good society. https://doi.org/10.1515/bis-2013-0007 —. 2016. The Corruption of Capitalism: Why Rentiers Thrive and Work Does Not Pay. London: Biteback Publishing. This book seeks to explain how global capitalism is rigged in favor of “rentiers capitalism” to the detriment of societies, especially the precariat, and how it perpetuates itself via ownership of assets and access to political power. https://www. bitebackpublishing.com/books/the-corruption-of-capitalism —. 2017. Basic Income: A Guide for the Open-Minded. New Haven: Yale University Press. In this book, Standing covers in detail the effects of universal basic income (UBI) on the economy, poverty, work, and labor; dissects the standard arguments against it; explains what we can learn from pilots across the world; and illustrates why, In his view, a UBI has now become such an urgent necessity. https://books.google.co.uk/books/about/Basic_Income. html?id=GCvGAQAACAAJ&redir_esc=y Appendix E: Annotated Bibliography 307 Steensland, Brian. 2006. “Cultural Categories and the American Welfare State: The Case of Guaranteed Income Policy.” American Journal of Sociology 111 (5): 1273–326. There is considerable evidence that cultural categories of worth are central to the ideological foun- dation of the American welfare state. However, existing perspectives on U.S. welfare policy development grant little explanatory power to the role of culture. For this reason, they cannot adequately explain the dynamics of an important, but frequently overlooked, episode in American welfare state history: the rise and fall of guaranteed annual income proposals in the 1960s and 1970s. The author outlines three mechanisms—schematic, discursive, and insti- tutional—through which culture can influence policy outcomes. He then argues that cultural categories of worthiness affected welfare policy development through their constitutive con- tribution to cultural schemas, their deployment by actors as resources in expert deliberation and public discourse, and their institutionalization in social programs that reinforced the sym- bolic and programmatic boundaries between categories of the poor. https://www.jstor.org/ stable/10.1086/499508 —. 2007. The Failed Welfare Revolution: America’s Struggle over Guaranteed Income Policy. Princeton: Princeton University Press. During the 1960s and 1970s, policy makers in three presidential administrations tried to replace the nation’s existing welfare system with a program to guarantee Americans basic economic security. Surprisingly from today’s van- tage point, guaranteed income plans received broad bipartisan support in the 1960s. One proposal, President Nixon’s Family Assistance Plan, nearly passed into law in the 1970s, and President Carter advanced a similar bill a few years later. The failure of these propos- als sowed the seeds of conservative welfare reform strategies under President Reagan and beyond. Steensland’s account, based on extensive original research in presidential archives, draws on mainstream social science perspectives that emphasize the influence of power- ful stakeholder groups and policy-making institutions. But Steensland also shows that some of the most potent obstacles to guaranteed income plans were cultural. Most centrally, by challenging Americans’ long-standing distinction between the “deserving” and “undeserving” poor, the plans threatened the nation’s cultural, political, and economic status quo. https:// press.princeton.edu/titles/8548.html Stern, Andy. 2016. Raising the Floor: How a Universal Basic Income Can Renew Our Economy and Rebuild the American Dream. New York: Public Affairs. With major tech- nological advances eliminating more middle-class jobs, new systems of universal support are required. Stern proposes instituting a universal basic income (UBI) of US$1,000 per month for all adults between the ages of 18 and 64—as well as for all seniors who do not receive at least US$1,000 per month in social security payments. This will cost around US$1.75–US$2.5 tril- lion per year in government spending, which would require ending many of the current 126 welfare programs, adjusting the pension system, creating a new and more cost-effective non-employer-based health care system, some redistribution of government spending and taxation expenditures, and increased revenue from new sources. He closes with organizing ideas for getting a UBI off the ground, including the goals of getting Basic Income Party can- didates to run nationwide in the 2020 congressional primary elections, and of petitions to get UBI included as an item on the ballot in states where citizens’ initiative processes exist. https://www.publicaffairsbooks.com/titles/andy-stern/raising-the-floor/9781610396264/ Tabatabai, Hamid. 2011. “The Basic Income Road to Reforming Iran’s Price Subsidies.” Basic Income Studies 6 (1). The Islamic Republic of Iran has become one of the first coun- tries in the world to provide a de facto universal basic income to all its citizens. This article reviews the replacement of fuel and food subsidies with direct cash transfers to the popu- lation, and shows how this emerged almost as a byproduct of an attempt to transform an inefficient and unfair system of price subsidies. https://doi.org/10.2202/1932-0183.1172 Tcherneva, Pavlina R. 2013. “The Job Guarantee: Delivering the Benefits That Basic Income Only Promises—A Response to Guy Standing.” Basic Income Studies 7 (2): 66–87. The article offers three critiques of universal basic income (UBI). First, there is a fundamental ten- sion between the way income in a monetary production economy is generated, the manner 308 Appendix E: Annotated Bibliography in which a UBI wishes to redistribute it, and the subsequent negative impact of this redistri- bution on the process of income generation itself. A UBI, she argues, is “dependent for its existence on the very system it wishes to undermine.” Second, the macroeconomic effects of UBI on contemporary economies are destabilizing. The job guarantee, by contrast, stabilizes both the macroeconomy and the currency while helping transform the nature of work itself. https://doi.org/10.1515/bis-2013-0010 Torry, Malcolm. 2016. The Feasibility of Citizen’s Income. London: Palgrave Macmillan. This book assesses the feasibility of a universal basic income (UBI) with a primary, but not exclusive, focus on the United Kingdom. It opens by defining a UBI, namely an unconditional income paid automatically to every legal resident as a right of citizenship, with the amount varying with age only and not in relation to any other conditions. The rest of the book assesses the feasibility of UBI in financial, psychological, administrative, behavioral, political, and policy process terms. Torry ultimately assesses that a citizen’s income is in principle feasible, highlighting two conclu- sions in particular. First, he argues that policy process feasibility is the most crucial determinant. Second, he asserts that it would be feasible to implement the scheme one step at a time: either via a variety of incremental policy changes (such as the individualization of benefits, the imple- mentation of a participation income, or the reduction of means-tested benefit withdrawal rates), or one age group at a time. https://www.palgrave.com/in/book/9781137530776 van der Veen, Robert. 1997. “Real Freedom and Basic Income: Comment on Brian Barry.” Journal of Political Philosophy 5 (3): 274–86. Van der Veen provides a rebuttal to Barry’s critique of Van Parjis’s case for universal basic income (UBI) on the basis of “real freedom.” Among other critiques, he maintains that Barry’s argument rests on assuming identical choice sets when comparing the situation of an individual under UBI or current welfare benefits. https://doi.org/10.1111/1467-9760.00035 van Oorschot, Wim. 2000. “Who Should Get What, and Why? On Deservingness Criteria and the Conditionality of Solidarity among the Public.” Policy & Politics 28 (1): 33–48. To understand the form of and societal acceptance for welfare, the article argues that it is important to know which deservingness criteria are acknowledged by the public and their rel- ative importance; whether people differ in the degree to which their solidarity with others is conditional, which groups in society tend to be most or least conditional in their views, and factors that might explain differences in people’s views. Based on results from a public opin- ion survey carried out in the Netherlands in 1995, deservingness criteria and differences in conditionality can be linked to three different sets of explanatory variables: socioeconomic and demographic characteristics, opinions on and perceptions of social security and the wel- fare state, and basic values and attitudes. https://doi.org/10.1332/0305573002500811 —. 2006. “Making the Difference in Social Europe: Deservingness Perceptions among Citizens of European Welfare States.” Journal of European Social Policy 16 (1): 23–42. This article aims at contributing to a theoretical and empirical understanding of the popular cultural context of welfare rationing. It examines European public perceptions of the relative deservingness of four needy groups (elderly people, sick and disabled people, unemployed people, and immigrants). Hypotheses, deduced from a literature review, are tested against data from the 1999/2000 European Values Study survey. It is found that Europeans share a common and fundamental deservingness culture: across countries and social categories, there is a consistent pattern that elderly people are seen as most deserving, closely followed by sick and disabled people; unemployed people are seen as less deserving still, and immi- grants as least deserving of all. Conditionality is greater in poorer countries, in states with lower unemployment, and in countries where people have less trust in fellow citizens and state institutions. At the national level, there is no relationship with welfare regime type or spending. https://doi.org/10.1177/0958928706059829 Van Parijs, Philippe. 1991. “Why Surfers Should Be Fed: The Liberal Case for an Uncondi- tional Basic Income.” Philosophy & Public Affairs 20 (1): 101–31. Van Parijs argues that a liberal theory of justice—one that is truly committed to an equal concern for all and to Appendix E: Annotated Bibliography 309 nondiscrimination among conceptions of the good life—does justify, under appropriate factual conditions, a substantial universal basic income (UBI). He challenges the recurring objection that a UBI would be unfair: if liberal justice consists in “maximinning” the real freedom to pursue the realization of one’s life, those who take an unfair share of society’s resources are not those who opt for such a low-production, low-consumption lifestyle but those who, thanks to the attractive job they were given, appropriate a huge employment rent. https://www.jstor. org/stable/2265291 —, ed. 1992. Arguing for Basic Income: Ethical Foundations for a Radical Reform. London: Verso. The central objection to universal basic income (UBI) is simple: there is a widespread feeling that it would be unfair because hard workers would be exploited by loafers. By describing the type of society in which UBI would be legitimate, this volume’s con- tributions question and clarify some of the central principles of modern political philosophy. https://www.versobooks.com/books/2631-arguing-for-basic-income —. 1996. “Basic Income and the Two Dilemmas of the Welfare State.” Political Quar- terly 67 (1): 63–66. The article makes a case for universal basic income (UBI) as a potential solution to what Van Parijs calls “the two dilemmas of the welfare state.” The first quandary is the tension between fighting exclusion and exploitation: the more a country does to improve the situation of the poorest workers, the scarcer jobs become, and the more people there are who are deprived of work. The second dilemma is the tension between economic and political capacity: in the European Union, a generous national welfare state would struggle to sustain itself economically given the level of mobility, but a generous transnational welfare policy faces immense political resistance. The article then discusses how UBI would help solve both dilemmas. https://doi.org/10.1111/j.1467-923X.1996.tb01567.x —. 2001. What’s Wrong with a Free Lunch? Boston: Beacon Press. Van Parijs argues that a universal basic income (UBI) would reduce unemployment, improve women’s lives, and pre- vent the environmental damage caused by overproduction and fast growth. At the heart of his proposal is the intention to secure real freedom for all. He acknowledges that an idle surfer might not deserve a UBI, but that the surfer’s good luck would be no different than the good fortune enjoyed by those who benefit from the current distribution of resources. http://www. beacon.org/Whats-Wrong-With-A-Free-Lunch-P132.aspx —. 2013. “The Universal Basic Income: Why Utopian Thinking Matters.” Politics & Society 41 (2): 171–82. The article summarizes the main arguments that support the utopian, radical proposal of universal basic income (UBI). It mentions a number of contexts in which it is being taken seriously, and sketches out ways in which sociological insights and research are crucially relevant to the UBI debate. https://doi.org/10.1177/0032329213483106 —. 2018. Basic Income and the Left: A European Debate. Social Europe Edition. Is a universal basic income compatible with social justice and individual self-worth? Does it open up the space for an end to demeaning labor and a resurgence of voluntary work and cul- tural life? Is it affordable? This collection of short essays, all previously published in Social Europe, allows both proponents and opponents to make their case. https://www.socialeurope. eu/book/basic-income-and-the-left-a-european-debate Van Parijs, Philippe, and Yannick Vanderborght. 2017. Basic Income: A Radical Proposal for a Free Society and a Sane Economy. Cambridge, MA: Harvard University Press. This book presents a comprehensive defense of universal basic income, advocating it as our most realistic hope for addressing economic insecurity and social exclusion in the 21st cen- tury. The authors combine philosophy, politics, and economics as they compare the idea of a basic income with alternative policy options against poverty and unemployment. They trace its history and lay out how such an apparently implausible idea might be viable financially and achievable politically. http://www.hup.harvard.edu/catalog.php?isbn=9780674052284 Vanderborght, Yannick. 2005. “The Basic Income Guarantee in Europe: The Belgian and Dutch Back Door Strategies.” In The Ethics and Economics of the Basic Income 310 Appendix E: Annotated Bibliography Guarantee, edited by Karl Widerquist, Michael Lewis, and Steven Pressman. New York: Routledge. This chapter scrutinizes the political chances of universal basic income (UBI) in Belgium and the Netherlands, and possible incremental steps in that direction. It provides a brief account of more than 25 years of UBI debate in both countries and unbundles reasons for its failure to gain support (e.g., moral objection to provide income without work). Refund- able tax credits or a participation income constitute two possible steps that may prove to be more promising than a full UBI. https://www.taylorfrancis.com/books/e/9781351890540/cha pters/10.4324%2F9781315239934-16 Vanderborght, Yannick, and Toru Yamamori, eds. 2014. Basic Income in Japan: Prospects for a Radical Idea in a Transforming Welfare State. New York: Palgrave Macmillan. A seminal English volume entirely devoted to the prospects for a universal basic income in Japan, it includes a dozen contributors and new empirical data to inform the debate in the country. https://www.palgrave.com/gp/book/9781137356574 Walker, Mark. 2016. Free Money for All: A Basic Income Guarantee Solution for the 21st Century. New York: Palgrave Macmillan. The book makes a case for a universal basic income of US$10,000 per adult U.S. citizen. It argues that UBI would increase “gross national happiness” and “gross national freedom,” while helping to mitigate some of the worst consequences of rising technological unemployment. https://www.palgrave.com/us/ book/9781137471321 Wehner, Burkhard. 2018. Universal Basic Income and the Reshaping of Democracy: Towards a Citizens’ Stipend in a New Political Order. Heidelberg: Springer. This book discusses the political logic of a universal basic income (UBI) and its controversial political and scientific implications. The author studies the institutions, rules, and decision-making processes of conventional democracies to reveal an institutional framework in which a UBI for all citizens could eventually become politically viable. The work addresses a broad range of topics, such as nationwide experiments with a UBI, voters’ support for the idea, and the effects of a UBI on business cycles and demographic policies. https://www.springer.com/la/ book/9783030058272 White, Stuart. 2012. “Basic Capital: A Policy Whose Time Has Come…and Gone?” The Good Society 21 (1): 61–73. This article examines the idea of a universal capital grant or “basic capital,” and how it affected public policy in the United Kingdom in the form of the Child Trust Fund. Some supporters of the fund (such as the author) viewed it as a potentially important first step toward the creation of something like a “property-owning democracy.” However, the government abolished the Child Trust Fund in 2010 and the article ponders future prospects for basic capital. https://doi.org/10.1353/gso.2012.0011 Widerquist, Karl. 2005. “A Failure to Communicate: What (If Anything) Can We Learn from the Negative Income Tax Experiments?” Journal of Socio-Economics 34 (1): 49–81. The U.S. and Canadian governments conducted five negative income tax experiments between 1968 and 1980. The labor market findings of these experiments were an advance for understanding the effects of a basic income guarantee, but their conclusiveness is often overstated. A review of nonacademic articles on the experiments reveals poor understand- ing of the results. One often overlooked cause of this misinterpretation was the failure of researchers to make clear that the experiments could not estimate the demand response and therefore could not estimate the market response to the program. https://doi.org/10.1016/j. socec.2004.09.050 —. 2013. Independence, Propertylessness, and Basic Income: A Theory of Freedom as the Power to Say No. New York: Palgrave Macmillan. This book argues that philoso- phers have focused too much on scalar freedom and proposes a theory of status freedom as effective control self-ownership: the power to have or refuse active cooperation with other willing people—or simply freedom as the power to say no. https://www.palgrave.com/us/ book/9781137274724 Appendix E: Annotated Bibliography 311 —. 2017. “The Cost of Basic Income: Back-of-the-Envelope Calculations.” Basic Income Studies 12 (2). This article shows how the cost of universal basic income (UBI) is often misunderstood and greatly exaggerated. It then presents simple, “back-of-the-envelope” estimates of the net cost of a UBI set at about the official poverty line in the United States: US$12,000 per adult and US$6,000 per child with a 50 percent marginal tax rate. The result- ing net cost of this UBI is US$539  billion per year, or about one-sixth its often-mentioned gross cost of about US$3.415 trillion. The net cost of this UBI scheme is less than 25 percent of the cost of current U.S. entitlement spending, less than 15 percent of overall federal spend- ing, and about 2.95 percent of gross domestic product. The average net beneficiary is a family of about two people making about US$27,000 per year in market income, and its net UBI ben- efit would be nearly US$9,000. https://doi.org/10.1515/bis-2017-0016 —. 2018. A Critical Analysis of Basic Income Experiments for Researchers, Policymak- ers, and Citizens. New York: Palgrave Macmillan. Although experiments can produce much relevant data about universal basic income (UBI), they are crucially limited in their abil- ity to enlighten on big questions that bear on the discussion of whether to implement a UBI as a national or regional policy. Results of UBI experiments are particularly vulnerable to mis- understanding, sensationalism, and spin. This book examines the difficulties of conducting a UBI experiment and reporting the results in ways that successfully improve public under- standing of the probable effects of a national UBI. The book makes recommendations on how researchers, reporters, citizens, and policy makers can avoid these problems and get the most out of UBI experiments. https://www.palgrave.com/gp/book/9783030038489 Widerquist, Karl, and Michael Howard, eds. 2012. Alaska’s Permanent Fund Dividend: Examining Its Suitability as a Model. New York: Palgrave Macmillan. Contributors to this volume discuss Alaska’s Permanent Fund Dividend as a model both for resource policy and for social policy. The book explores whether other states, nations, or regions would bene- fit from an Alaskan-style dividend, while also laying out ways to alter and improve the model. https://www.palgrave.com/gb/book/9780230112070 Widerquist, Karl, Michael Anthony Lewis, and Steven Pressman. 2005. The Ethics and Eco- nomics of the Basic Income Guarantee. New York: Routledge. This book explores the political, sociological, economic, and philosophical issues of the basic income guarantee. Trac- ing the history of the idea, from its origins in the late 18th century through its political vogue in the 1970s, when the Family Assistance Plan narrowly missed passage in the U.S. Congress, it also examines the philosophical debate over the issue. https://www.routledge.com/The-Ethics- and-Economics-of-the-Basic-Income-Guarantee/Widerquist-Lewis/p/book/9780754641889 Widerquist, Karl, José A. Noguera, Yannick Vanderborght, and Jurgen De Wispelaere, eds. 2013. Basic Income: An Anthology of Contemporary Research. Oxford, UK: Wiley-Blackwell. This volume is the first major compilation on universal basic income (UBI) combining six decades of literature. It includes extracts from influential empirical research and theoretical arguments on virtually all aspects of UBI. https://www.wiley.com/en-gb/Basic +Income%3A+An+Anthology+of+Contemporary+Research-p-9781405158107 Wiederspen, Jessica, Elizabeth Rhodes, and H. Luke Shaefer. 2015. “Expanding the Dis- course on Antipoverty Policy: Reconsidering a Negative Income Tax.” Journal of Poverty 19: 218–38. This article proposes the replacement of the current means-tested safety net in the United States with a negative income tax. It highlights gaps in service provision that leave millions in poverty, explains how a negative income tax could help fill those gaps, and compares current expenditures on major means-tested programs to estimated costs of a negative income tax. Finally, it addresses the financial and political concerns that are likely to arise if a negative income tax proposal gains traction among policy makers. https://doi.org/1 0.1080/10875549.2014.991889 Yang, Andrew. 2018. The War on Normal People: The Truth about America’s Disappear- ing Jobs and Why Universal Basic Income Is Our Future. New York: Hachette Books. Yang, who at the time of writing this appendix is a Democratic presidential nominee for the 312 Appendix E: Annotated Bibliography 2020 U.S. election, proposes a universal basic income (UBI) on the grounds of technological change and automation-induced unemployment. His UBI proposal would be funded by the introduction of a value-added tax. https://www.hachettebooks.com/titles/andrew-yang/the- war-on-normal-people/9780316414241/ Yeung, Ying, and Stephen Howes. 2015. “Resources-to-Cash: A Cautionary Tale from Mongolia.” Development Policy Centre Discussion Paper 42, Crawford School of Public Policy, Australian National University, Canberra. The paper is one of the few mate- rials documenting Mongolia’s experience with universal basic income. While the scheme has showed mixed results, it only lasted two years due to a range of factors, including its financing linked to the boom-and-bust of commodity prices (in this case copper), which also affected its political and public support. https://im4dc.org/wp-content/uploads/2015/09/Combined- Yeung.pdf Zelleke, Almaz. 2018. “Work, Leisure, and Care: A Gender Perspective on the Partici- pation Income.” Political Quarterly 89 (2): 273–79. In this article, Zelleke considers the participation requirement of Atkinson’s “participation income” from a gender perspective. In particular, he discusses the notions of dependence, independence, reciprocity, and free riding that underlie normative arguments for conditional over unconditional benefits. Employing such a perspective, the article argues, would show the superiority of unconditional benefits in achieving more efficient and effective income support and reducing inequality—all of which lie at the heart of Atkinson’s proposal. https://doi.org/10.1111/1467-923X.12518 Zwolinski, Matt. 2012. “Classical Liberalism and the Basic Income.” Basic Income Studies 6 (2): 1–14. This article provides a brief overview of the relationship between libertarian political theory and universal basic income (UBI). It distinguishes between different forms of libertari- anism and argues that at least one form, classical liberalism, is compatible with and provides some grounds of support for UBI. A classical liberal UBI, however, is likely to be much smaller than the sort of UBI advocated by those on the political left. Moreover, there are both empir- ical and moral reasons for doubting that the classical liberal case for UBI will ultimately be successful. https://doi.org/10.1515/1932-0183.1221 U niversal basic income (UBI) is emerging as one of the most hotly debated issues in development and social protection policy. But what are the features of UBI? What is it meant to achieve? How do we know, and what don’t we know, about its performance? What does it take to imple- ment it in practice? Drawing from global evidence, literature, and survey data, this volume provides a framework to elucidate issues and trade-offs in UBI with a view to help inform choices around its appropriateness and feasibility in different con- texts. Specifically, the book examines how UBI differs from or complements other social assistance programs in terms of objectives, coverage, incidence, adequacy, incentives, effects on poverty and inequality, financing, political economy, and implementation. It also reviews past and current country experiences, surveys the full range of existing policy proposals, provides original results from micro–tax ben- efit simulations, and sets out a range of considerations around the analytics and practice of UBI. ISBN 978-1-4648-1458-7 SKU 211458