UNU World Institute for Development Economics Research (UNU-WIDER) was established by the United Nations University as its first research and training centre and started work in Helsinki, Finland, in 1985. The mandate of the institute is to undertake applied research and policy analysis on structural changes affecting developing and transitional economies, to provide a forum for the advocacy of policies leading to robust, equitable, and environmentally sustainable growth, and to promote capacity strengthening and training in the field of economic and social policy-making. Its work is carried out by staff researchers and visiting scholars in Helsinki and via networks of collaborating scholars and institutions around the world.
United Nations University World Institute for Development Economics Research (UNU-WIDER) Katajanokanlaituri 6B, 00160 Helsinki, Finland www.wider.unu.edu
Inequality in the Developing World
Edited by CARLOS GRADÍN, MURRAY LEIBBRANDT, AND FINN TARP
A study prepared by the United Nations University World Institute for Development Economics Research (UNU-WIDER)
1
Great Clarendon Street, Oxford, OX2 6DP, United Kingdom
Oxford University Press is a department of the University of Oxford. It furthers the University’s objective of excellence in research, scholarship, and education by publishing worldwide. Oxford is a registered trade mark of Oxford University Press in the UK and in certain other countries
World Institute for Development Economics Research of the United Nations University (UNU-WIDER), Katajanokanlaituri 6B, 00160 Helsinki, Finland
The moral rights of the authors have been asserted
First Edition published in 2021
Impression: 1
Some rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, for commercial purposes without the prior permission in writing of Oxford University Press, or as expressly permitted by law, by licence or under terms agreed with the appropriate reprographics rights organization.
This is an open access publication. Except where otherwise noted, this work is distributed under the terms of a Creative Commons Attribution-Non Commercial-Share Alike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO), a copy of which is available at https://creativecommons.org/licenses/by-nc-sa/3.0/igo/.
Enquiries concerning reproduction outside the scope of this licence should be sent to the Rights Department, Oxford University Press, at the address above Published in the United States of America by Oxford University Press 198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2020945720 ISBN 978–0–19–886396–0 DOI: 10.1093/oso/9780198863960.001.0001
Printed and bound by CPI Group (UK) Ltd, Croydon, CR0 4YY
Links to third party websites are provided by Oxford in good faith and for information only. Oxford disclaims any responsibility for the materials contained in any third party website referenced in this work.
Foreword
Inequality is a dominant challenge to development. It influences economic growth and redistribution, and feeds into power asymmetries that can jeopardize democratization and human rights, trigger conflict, and entrench chronic poverty. Inequality is at the core of the United Nations mandate, and is one of the seventeen goals of the 2030 Agenda for Sustainable Goals.
While global inequality fell, by and large, towards the end of the twentieth century, since the start of this century within-country inequality has been steadily rising—underpinning intense public and academic debates to the extent that it has become a prevailing policy concern of many countries and in all multilateral agencies.
There has been strong interest for decades in bringing both developed and developing countries together in the analysis of global inequalities and the forces shaping them. Initially the lack of adequate data for such was a fundamental handicap. To address this, UNU-WIDER established the (freely downloadable) World Income Inequality Database (WIID) which compiles income inequality information from primary sources for developed, developing, and transition countries. In turn, the WIID increased the volume of the voice for analytical research on between-country and within-country inequalities. Hence the launch of UNU-WIDER’s Inequality in the Giants project—as part of a broad international effort designed to shed light on a set of new questions on such inequalities, by generating integrated datasets and applying a consistent methodology to investigate the determinants of inequality dynamics in ten of the world’s largest economies: Brazil, China, France, India, Indonesia, Mexico, Russia, South Africa, the UK, and the US.
Employing advances made in recent decades on the measurement of inequality, as well as in the development of better data to analyse the processes which generate inequality, this book is the result of rigorous scientific work by a large team of international experts, each highly qualified within their respective research niches. I sincerely thank the editors—Carlos Gradín, Murray Leibbrandt, and Finn Tarp— for their sharp analytical and editorial skills, which result in giving us a more layered, more nuanced understanding of inequality in developing countries.
UNU-WIDER gratefully acknowledges the support and financial contributions to its research programme by the governments of Finland, Sweden, and the UK. Without this vital funding our research and policy advisory work would be impossible.
Kunal Sen
Director, UNU-WIDER Helsinki, June 2020
Acknowledgements
This book has roots in discussions in late 2013/early 2014 between Francois Bourguignon, Francisco Ferreira, and Nora Lustig, later adding Murray Leibbrandt, to discuss a project idea put forward by Nora Lustig and referred to as ‘Inequality in the Giants’. The team developed the preliminary set of notes from this meeting into a five-page draft research proposal, originally meant to cover nine of the biggest countries in the world (Brazil, China, France, India, Mexico, Russia, South Africa, the UK, and the US. Subsequently, Indonesia was included, so the target number of case countries ended up being ten.
The intention was to mobilize funding from different sources to implement the project with the team of four (above) as principal investigators. Finn Tarp, then director of UNU-WIDER, was invited to join the team, and a large number of potential contributors to the project were invited to a workshop on the sidelines of the UNU-WIDER conference Mapping the Future of Development Economics, held in Helsinki on 17–19 September 2015.
The workshop was productive in terms of ideas. Funding for the ‘big’ project did not materialize as originally hoped, however. Instead, UNU-WIDER managed to move forward with a subset of countries—namely Mexico, South Africa, and India—coordinated by respectively Nora Lustig, Murray Leibbrandt, and Peter Lanjouw. Subsequently China and Brazil, coordinated by Shi Li and Marcelo Cortes Neri, joined the project, while Carlos Gradín joined the coordination team. The coordinators and their respective teams of authors produced several studies on each country, which were published in the WIDER Working Paper series. Much of this work also reached academic journals.
We felt, however, that a comprehensive book and coherent statement was called for. We therefore put together a proposal that included both the attempt at analysing in-depth inequalities within our group of five large developing countries that jointly account for more than 40 per cent of the world’s population and the relationship with global inequality issues. Accordingly, we proposed to frame five country syntheses by broader contributions from development economists who are leaders in the field of inequality measurement, trends, impact, and policies. They agreed and each provided an excellent take on a set of key framework-setting issues that contextualize the country case studies and put them in perspective.
We owe a profound debt to the original team of three for their inspiring stimulus. Their early notes and engagement in the work session organized with UNUWIDER in September 2015 has been of great value. We would also like to express
our most sincere gratitude to the chapter authors in this book for their willingness to participate in the research project and for their many insightful contributions. The same goes for the extensive and excellent group of collaborators on the background papers on each country case, synthesized here in this book.
Warm thanks are also due to UNU-WIDER for institutional support and never failing collaboration. We wish in particular to thank Lorraine Telfer-Taivainen, UNU-WIDER Editorial and Publishing Associate, for sound advice, hard work, and making the collaboration with Oxford University Press run smoothly. Adam Swallow, Economics and Finance Commissioning Editor at Oxford University Press, and his colleagues provided expert guidance with the publication process, and we do wish to acknowledge the anonymous referee reports that helped sharpen our focus and change the organization of the book.
Finally, a word of thanks to the donors of UNU-WIDER—Finland, Sweden, and the UK—for their core support to the work programme of UNU-WIDER, without which this book would not have been possible.
Carlos
Gradín, Murray Leibbrandt, and Finn Tarp Helsinki, Cape Town, and Copenhagen
June 2020
2.7.
4.6.
List of Figures
6.4. Primary sector employment in China as a share of total employment (per cent) 1980–2015
6.5. Wage growth of rural–urban migrant workers in China (%)
6.6. The shares of fiscal spending on education, medical insurance, social security, and agriculture in total budgetary expenditures in China
6.7. Growth of exports and imports of China (million US$)
7.1. Disparities in human capital outcomes, by social group
7.2. Impacts of upward mobility on consumption growth (at different percentiles: IV regression 1 (90% confidence bound)
8.1. Gini coefficient, 1989–2014
8.2. Decomposition of differences in the distribution of earnings: 1989–2014
8.3. Relative returns and relative supply of workers by education, college, and high school vs rest
8.4. Fiscal incidence analysis, core income concepts
8.5. Government transfers (% of GDP), 1988–2018
8.6. Fiscal policy and inequality, 1996–2015
8.7. Fiscal
9.1.
9.2.
and poverty, 2008–15
9.4. Distribution of
9.5. The progressivity of each of the three main social cash transfers in South Africa
9.6. Socioeconomic class sizes, 2008–17
10.1.
12.1.
12.2.
of
List of Tables
3.1.
4.1. Descriptive statistics of variables used in the empirical
4.2. Regression analysis: OLS including all contextual
4.3. Regression analysis: country fixed effects selecting some contextual variables
4.4. Regression analysis: country fixed effects selecting some contextual variables—most recent observations (year > 2000)
5.1. Inequality in Brazil by topic, technique, dataset, period of time, and income concept
5.2. Income, equality, and social welfare: contribution to growth ordered by disposable income
6.1. China’s regional and urban/rural income gaps
6.2. Gini coefficients of urban wages and income per capita in China, 1988–2013
7.1.
7.2.
7.3.
7.4.
7.5. Welfare transition dynamics based on synthetic panel data, India 1987/88–2011/12 (%)
8.1. Bound and Johnson decomposition: 1989–94; 1994–2006; 2006–14 (assuming an elasticity of substitution σ = 2 and comparing college and high school-educated workers with rest of workers)
9.1. Real annual growth rates in South Africa since 1990
9.2.
9.3.
components in per capita terms (real 2014 prices, rand)
9.4. Dynamic decompositions including household composition and re-rankings, 1993–2008
9.5. Gini coefficients at different thresholds
9.6. Labour market summary statistics, 2000, 2011, and 2014
10.1. Descriptive statistics: Afrobarometer
10.2. Descriptive statistics: Asianbarometer
10.3. Descriptive statistics: Latinobarometer 242
10.4. Descriptive statistics: Eurobarometer 242
10.5. Economic conditions and inequality: OLS results in the Afrobarometer 244
10.6. Economic conditions and inequality: OLS results in the Asianbarometer 244
10.7. Economic conditions and inequality: OLS results in the Latinobarometer 245
10.8. Economic conditions and inequality: OLS results in the Eurobarometer 245
10.A1. List of items per dataset 249
10.A2. Number of observations per country per wave: Afrobarometer 250
10.A3. Number of observations per country per wave: Asianbarometer 250
10.A4. Number of observations per country per wave: Latinobarometer 251
10.A5. Number of observations per country per wave: Eurobarometer 252
10.A6. Economic conditions and inequality: OLS results in the Afrobarometer—all controls 253
10.A7. Economic conditions and inequality: OLS results in the Asianbarometer—all controls 254
10.A8. Economic conditions and inequality: OLS results in the Latinobarometer—all controls 255
10.A9. Economic conditions and inequality: OLS results in the Eurobarometer—all controls 256
12.1. Manufacturing share of GDP (%) 285
12.2. Deindustrialization in sub-Saharan Africa 296
List of Abbreviations
BRICS Brazil, Russia, India, China, South Africa group
CCT conditional cash transfer
CES constant elasticity of substitution
CFPS China Family Planning Survey
CHIP China Household Income Project
DHI disposable household income
DISE District Information System for Education
ENIGH National Survey on Households’ Income and Expenditures
FDI foreign direct investment
GATT General Agreement on Tariffs and Trade
GCIP Global Consumption and Income Project
GDIM Global Database of Intergenerational Mobility
GDP gross domestic product
GGC Great Gatsby Curve
GIC growth incidence curve
GNI gross national income
HDI Human Development Index
IBGE Brazilian Institute of Geography and Statistics
IHDS India Human Development Surveys
IMF International Monetary Fund
IPR intellectual property rights
LIS Luxembourg Income Study
LIT learning, industrial, and technology
LSMS World Bank Living Standard Measurements Surveys (LSMS)
MDGs Millennium Development Goals
MENA Middle East and North Africa
MLD Mean Log Deviation
MRP mixed recall period
NBS National Bureau of Statistics
NIDS National Income Dynamics Study
NSS National Sample Survey Organization
OECD Organisation for Economic Co-Operation and Development
PIT personal income tax
PME Pesquisa Mensal de Emprego
PNAD Pesquisa Nacional de Amostras a Domicílio
PNADC Pesquisa Nacional de Amostras a Domicilio Contínua
PPP purchasing power parity
PSID Panel Study of Income Dynamics
QE quantitative easing
QLFS Quarterly Labour Force Survey
R&D research and development
RAIS Registro Anual de Informações Sociais
RIF re-centred influence function
SARS South African Revenue Services
SCs Scheduled Castes
SDGs Sustainable Development Goals
SIA scale invariance axiom
SNA System of National Accounts
SOEs state-owned enterprises
SSA sub-Saharan Africa
SWIID Standardized World Income Inequality Database
TSTSLS two-sample two-stage least squares
US United States
WID World Inequality Database
WIID UNU-WIDER World Income Inequality Database
WIPO World International Properties Rights Organization
WTO World Trade Organization
Notes on Contributors
Raymundo Campos-Vazquez obtained his PhD in economics from the University of Califormia, Berkeley in 2009. Since then, he has worked as a professor in El Colegio de Mexico in Mexico City. He has publications in the American Economic Review, Oxford Development Studies, and American Economic Journal: Economic Policy, among others. Currently he is on academic leave at the Banco de Mexico.
Daniele Checchi is a professor of economics at the University of Milan. He also serves as a Secretary General of the LIS Cross-National Data Center in Luxembourg. He has published in journals such as Economic Policy, Journal of Public Economics, European Sociological Review, and others, and his research interests range from inequalities in educational attainments to educational policies, from wage inequalities to the role of labour market institutions in shaping inequalities of opportunities.
Andrew E. Clark is CNRS research director and a professor of economics at the Paris School of Economics. He pioneered the introduction of subjective well-being in economics in the early 1990s, and its use for the analysis of contextual effects. He has more than 30,000 citations on Google Scholar, and nine papers cited more than 1000 times each. He was among the ‘Highly Cited Researchers 2018’—the list that recognizes world-class researchers selected for their performance, demonstrated by production of multiple highly cited papers that rank in the top 1 per cent by citations for field and year in the Web of Science.
Andrej Cupak works as a data expert and research associate at the LIS Cross-National Data Center in Luxembourg. He is also affiliated with the National Bank of Slovakia. His research interests include household consumption and financial behaviour, human capital, income and wealth inequality, survey data, and applied microeconomics more generally.
Conchita D’Ambrosio is a professor of economics at the Department of Behavioural and Cognitive Sciences, University of Luxembourg. Her research interests revolve around the analysis and measurement of individual well-being, both theoretically and empirically. In this respect, she has proposed a number of different axiomatically characterized well-being indices, including measures of social exclusion and of poverty over time, that disentangle the effects of chronic and persistent poverty. She has also worked on the empirical analysis of these indices using data from a variety of different countries. She has published widely in the fields of social-index numbers and income distribution.
Hai-Anh H. Dang is an economist in the Analytics and Tools Unit, Development Data Group, World Bank. His main research is on international development, poverty, inequality, human development topics, and methodology to construct synthetic (pseudo) panel data from cross-sections. He has published in various development journals, including Economic Development and Cultural Change, Journal of Development Economics, Journal of Development Studies, World Bank Economic Review, World Development, and has chapters
in books published by leading academic publishers. He also serves as a co-editor of Review of Development Economics, and on the editorial boards of other journals.
James Davies is a professor of economics in the Department of Economics, University of Western Ontario in London, Canada. His field is public economics and his research has ranged over many topics, from tax policy to inequality measurement. In recent years he has focused on policy to mitigate the economic impacts of climate change and natural disasters, and on the global distribution of personal wealth.
Carlos Gradín is a research fellow at the United Nations University World Institute for Development Economics Research (UNU-WIDER) in Helsinki, and a professor of applied economics at the University of Vigo (on leave of absence). His main research interest is the study of poverty, inequality, and discrimination in both developed and developing countries, especially inequalities between population groups (i.e. by gender, race, or ethnicity). His research deals with enhancing the empirical evidence as well as methodological tools for the measurement and understanding of those issues. His research has been widely published in international journals.
Pippa Green is a South African journalist and writer. She has served as deputy editor for a number of South African newspapers, as head of Radio News at the South African public broadcaster, and as a visiting professor of journalism at Princeton University. She is South Africa’s Press Ombudsman and before this worked in SALDRU, editing and writing on employment, inequality, and inclusive growth in South Africa.
Peter Lanjouw is a professor at the School of Business and Economics, VU University Amsterdam. His research focuses on the measurement of poverty and inequality as well as the analysis of rural development, notably via the study of a village economy in rural India and the broader process of rural non-farm diversification. He has co-authored several books and has also published in such leading economics journals as Econometrica and the Economic Journal, as well as numerous field journals. He is currently editor of the World Bank Research Observer.
Murray Leibbrandt holds the National Research Foundation Chair in Poverty and Inequality Research at the University of Cape Town and is a senior research fellow of UNU-WIDER. He is the director of the Southern Africa Labour and Development Research Unit and the African Centre of Excellence for Inequality Research. He has published widely in development economics using survey data and especially panel data to analyse South Africa’s poverty, inequality, and labour-market dynamics.
Shi Li is a professor of economics at Zhejiang University, director of the China Institute for Income Distribution at Beijing Normal University, non-resident senior research fellow at UNU-WIDER, and a research fellow at IZA. His research focuses on income and wealth distribution, poverty, labour markets, and rural migration and the labour market in China. He has published in journals such as Review of Income and Wealth, Oxford Bulletin of Economics and Statistics, Economic Development and Cultural Change, and Journal of Development Economics. His publications include several edited volumes, including Rising Inequality in China (with H. Sato and T. Sicular, Cambridge University Press 2013).
Nora Lustig is Samuel Z. Stone Professor of Latin American Economics and the founding director of the Commitment to Equity Institute (CEQ) at Tulane University. She is also a
non-resident senior fellow at the Brookings Institution, the Center for Global Development, and the Inter-American Dialogue. Her research on economic development, inequality, and social policies has been published across more than sixty articles, close to ninety chapters, and twenty-five books and edited volumes. Professor Lustig is a founding member and President Emeritus of the Latin American and Caribbean Economic Association (LACEA) and serves on the editorial board of the Journal of Economic Inequality. She received her doctorate in economics from the University of California, Berkeley.
Teresa Munzi is the director of operations at the LIS Cross-National Data Center in Luxembourg, where she is responsible for managing and overseeing all operations in the LIS office. Her research interests include the comparative study of welfare systems and their impact on poverty, inequality, and family wellbeing; and gender differences in employment and earnings.
Ambar Narayan is a lead economist in the Poverty & Equity Global Practice of the World Bank. He leads and advises teams conducting policy analysis, evaluations, and research in development from a microeconomic perspective, and has published research papers and policy briefs on issues that reflect the eclectic mix of topics on which he has worked over the years. He has been a lead author for large World Bank studies, including a recent global report on intergenerational mobility, and regional or country reports on inequality of opportunity, poverty, and the impact of economic shocks. He has a PhD in economics from Brown University in the United States.
Marcelo Neri is director of FGV Social at the Getulio Vargas Foundation (FGV). He holds a PhD in economics, Princeton University. Previously he was the secretary-general of the Council of Economic and Social Development (CDES), president of the Institute for Applied Economic Research (Ipea), and Minister of Strategic Affairs in Brazil. He has evaluated policies in more than a dozen countries and also designed and implemented policies at three government levels in Brazil. Neri’s research focuses on social policies, microeconometrics, and well-being. He teaches graduate and undergraduate courses at EPGE/FGV, writes regularly in scientific journals, and has published ten books.
Vimal Ranchhod is a professor at the School of Economics and the Deputy Director of the Southern Africa Labour and Development Research Unit. He is a labour economist who has published widely on labour markets, education, and poverty and inequality.
Martin Ravallion currently holds the inaugural Edmond D. Villani Chair of Economics at Georgetown University. Prior to joining Georgetown in 2013 he was director of the World Bank’s research department, the Development Research Group. Martin’s main research interests over the past thirty years have concerned poverty and policies for fighting it. He has published extensively on this topic and advised numerous governments and international agencies.
John Scott is a professor-researcher at the Economics Department at the Centro de Investigación y Docencia Económicas (CIDE) in Mexico City, and academic researcher at the Consejo Nacional de Evaluación de la Política de Desarrollo Social (CONEVAL), a public institution responsible for poverty measurement and the evaluation of social programmes in Mexico. He has a BA in philosophy from NYU and an M.Phil in economics
from the University of Oxford. His principal research areas include redistributive policies, social and fiscal incidence analysis, poverty and inequality analysis, evaluation of social policy, rural development policies, agricultural and energy subsidies, and health and social security.
Anthony Shorrocks is an honorary professor at Manchester University and director of Global Economic Perspectives Ltd, having previously served as director of UNU-WIDER from 2000 to 2009. He has published extensively on topics concerned with income and wealth distribution, inequality, poverty, and mobility. His recent research on the level and distribution of global household wealth has been published annually in the Credit Suisse Global Wealth Report and Databook.
Terry Sicular is a professor of economics at the University of Western Ontario. She has conducted research on China’s economy since the 1980s and has published works on topics ranging from China’s economic transition to China’s rural households and rural economy, and education in China. Her current research focuses on topics related to income distribution and poverty in China. She has edited and contributed to several books, including Rising Inequality in China (with S. Li and H. Sato, Cambridge University Press 2013). She has served as a consultant to international organizations such as the World Bank, Asian Development Bank, and UNU-WIDER.
Joseph E. Stiglitz is a professor at Columbia University. He was awarded the Nobel Prize in Economics in 2001 for his work on asymmetric information. He is the co-chair of the High-Level Expert Group on the Measurement of Economic Performance and Social Progress at the OECD. He served as chairman of the Council of Economic Advisors under President Clinton and as chief economist of the World Bank.
Finn Tarp is a professor of development economics at the University of Copenhagen, and was director of UNU-WIDER from 2009 to 2018. His field experience covers more than two decades of in-country work in thirty-five countries across Africa and in the developing world more generally. Professor Tarp’s research focus is on issues of development strategy and foreign aid, with an interest in poverty, income distribution and growth, micro- and macroeconomic policy and modelling, agricultural sector policy and planning, and household and enterprise development. He has published widely and is a member of a large number of international committees and advisory bodies.
Roy van der Weide is a senior economist in the Poverty and Inequality Research team within the Development Research Group of the World Bank. He leads the poverty and inequality mapping research within the department. His other research is concerned with the empirics of inequality of opportunity and poverty reduction, and axiomatic approaches to income measurement. His work has been published in a range of academic journals, including the American Economic Review, the Journal of Econometrics, the Journal of Applied Econometrics, and the World Bank Economic Review. He holds a PhD from the University of Amsterdam.
1
Setting the Scene
Carlos Gradín, Murray Leibbrandt, and Finn Tarp
1 Introduction
Inequality has emerged as a key, perhaps the key, development challenge of the past decade. It holds implications for economic growth and redistribution, and some of the most influential social science of the first half of the past decade showed that it also translates into power asymmetries that can endanger democratization and human rights, create conflict, and embed social exclusion and chronic poverty (Wilkinson and Pickett 2010; Stiglitz 2012; Deaton 2014; Piketty 2014). As such, inequality is at the core of the United Nations’ mandate, and ‘reducing inequality within and among countries’ is one of the seventeen goals of the 2030 Agenda for Sustainable Goals (SDGs) approved by the UN General Assembly in September 2015. Positioning inequality (SDG 10) as a stand-alone goal confers high visibility upon it. Moreover, the way in which the associated, more specific targets have been formulated links the inequality goal with many of the other SDGs, meaning that ‘reducing inequality’ is one of the most heavily inter-linked themes of the 2030 SDG agenda.
These concerns also underpin intense public and academic debates, to the extent that inequality has become a dominant policy concern within many countries, in all multilateral agencies, and even for the World Economic Forum (World Bank 2016; Ostry et al. 2019; World Economic Forum 2020). It would be very worrying if the possibility of investigating inequality in the real world in greater depth than was previously the case had not accompanied this intense public concern with inequality. Fortunately, there have been a number of advances in recent decades regarding measurement of inequality, as well as in the development of better data to analyse the processes generating inequality and changes in inequality.
The empirical literature began with a focus on analysing inequality in the industrialized world in detail. The scarcity of in-depth analyses focused on developing countries was associated with a lack of adequate data and research capacity and was not helped by a stubborn and widespread misconception that inequality was not a relevant and pressing issue in poorer countries. This idea, which had its roots in the work of the classical economists of the eighteenth century, generally
Carlos Gradín, Murray Leibbrandt,
Finn
held that reducing inequality could conflict with higher priorities such as boosting growth or reducing poverty, by reducing the incentives of the most productive people. Within development economics, a more nuanced version of this view held that there was no need to push for a more equal society because this would be an automatic outcome that would eventually result from higher development, in line with the inverted-U hypothesis proposed by Simon Kuznets. A more textured understanding of inequality in developing countries was facilitated by gradual improvements in the available data. The intense research focus on inequality in both developed and developing countries over the past decade has replaced these ideas with a growing consensus that all highly unequal societies are dysfunctional in many ways, including compromised growth and poverty reduction.
This interest in inequality has also crossed the traditional national borders. There has been much analysis of cross-country comparisons of inequalities, in general. This was complemented by a focus on specific issues such as the middle class or the effects of the great recession (see Gornick and Jäntti 2013; Jenkins et al. 2013; Nolan 2018, among others). The gradual emergence of large databases, such as the Luxembourg Income Study (LIS) discussed in detail in Chapter 4 of this book, facilitated this research, providing researchers with easy access to harmonized microdata for many countries. Initially focused on developed countries, LIS has more recently been successful in incorporating several key middle-income countries.
Increasing interest in analysis of global inequalities and the role of forces like globalization in shaping them has brought developed and developing countries together. Two recent books by Bourguignon (2016) and Milanovic (2016) are prominent examples. Initially, the lack of adequate data for such work was a clear handicap. As reviewed in Ferreira et al. (2015), in response to this there have been many efforts to provide compilations of data from different sources, such as the UNU-WIDER World Income Inequality Database (WIID), which collates information from many primary sources such as LIS, the World Bank (PovCalnet) and other international organizations, Eurostat and National Statistical Authorities, and research studies. Another good example of this new age of inequality analysis with a global perspective is the 2018 World Inequality Report by the World Inequality Lab at the Paris School of Economics (Facundo et al. 2018). This initiative pursues an ambitious research agenda that includes producing new distributional data combining survey, administrative data, and national accounts, with special attention given to the top of the distribution.
Global inequalities are the result of combining between-country and withincountry inequalities. While according to, for example, the widely used Gini measure, global relative income inequality has been falling steadily for more than three decades, this trend reflected convergence in GDP per capita across nations. Relative inequality within countries remained roughly constant in the 1990s and
has been rising since 2000. This increase in average within-country relative inequality arose from a very heterogeneous picture among countries and regions, and the same goes for the associated and very sizeable increases in absolute inequality (see Niño-Zarazúa et al. 2017).
Inequalities between countries mostly reflect how successful developing regions have been in accelerating economic growth as compared to richer countries. Asian countries have been quite successful in using their particular exportoriented model and development strategies, as argued in Nayyar (2019). While African countries have experienced robust growth and poverty reduction since 2000 (see Addison et al. 2017), they have struggled to escape from a complex set of initial conditions and have seen more difficulties in finding a pathway to development. However, in these and all other contexts, inequalities within countries are much more heterogeneous. The same forces that can reduce inequality between countries—like globalization—might contribute to higher inequality within both developed and developing countries. Other forces, such as technological progress, have contributed to increased market inequalities everywhere and to increasing profit-shares. Some countries have been more successful in using active policies to curb the market processes that push for higher inequality or, at least, to compensate them using taxes and benefits. In this respect, developing countries—traditionally small states with weak policy capability and generally regressive welfare regimes—face clear disadvantages.
This book aims to contribute to the literature and public debate on inequality by bringing together an analysis of global inequality and a new and comprehensive view of the trends in inequality in five of the world’s largest developing countries—Brazil, China, India, Mexico, and South Africa—jointly accounting for more than 40 per cent of the world’s population. While this is not a majority population share and omits other important developing country inequality contexts—such as Indonesia, Pakistan, Nigeria, Bangladesh, and Russia—each of these cases are important inequality contexts. Understanding inequalities in these emerging economies is of great value in coming to grips with contemporary inequalities worldwide. This remains difficult due to both data and analytical challenges.
For this reason, UNU-WIDER engaged with a group of highly qualified international researchers to develop the five country case studies within a common framework under its Inequality in the Giants project.1 These country case leaders, together with their respective country teams, produced a series of in-depth studies for each of these countries. The objective of each country project was to assess
1 The outputs of the project can be accessed at www.wider.unu.edu/project/inequality-giants. The main findings of the project were presented in a special panel during the UNU-WIDER Think Development—Think WIDER development conference held in Helsinki, September 2018. Videos are available on the conference website www.wider.unu.edu/parallel-session/inequality-developing-giants.
the level and trend of income inequality and to understand the processes driving this quantitative picture. Use was made of the best available data and techniques. Most countries have developed a system of household surveys providing rich information to measure inequality as well as to investigate its drivers. The limitations of these household surveys, in terms of underestimation of certain sources of income or consumption and the misrepresentation of certain population groups, are well known. For that reason, other complementary sources such as administrative data, public listings, or big data were used in order to correct for lack of response when this was an issue in underestimation of the top income sources and shares. In some cases, researchers produced new data sets for the research.
Using these rich data sources, different decomposition techniques and the use of regression analyses facilitated identification of the main inequality drivers. Each of these country cases summarize the main findings about long-run trends in income inequality and its driving factors. They contain detailed analyses of three issues: (i) the role of earnings inequality and its determinants; (ii) the role of top incomes (when administrative records or other sources can be combined with household surveys); and (iii) the distributive impact of public policies. Yet, we tailored each country study to its specific needs and authors discuss specific issues that are important to that country, such as spatial inequalities or intergenerational mobility.
The country case studies are summarized in Part III of this book. It is imperative to contextualize this type of fine-grained analysis of within-country inequality within the global context. Therefore, for Part II of this volume, we invited a number of outstanding researchers to provide their take on a set of key frameworksetting issues to precede and put them in perspective. Accordingly, Part II focuses on global inequalities in income and wealth, the role of globalization, and challenges in analysing cross-country comparisons. The aim is to provide an overview of the state of the art in our knowledge of inequalities between and within countries, of the main practical and conceptual challenges in undertaking these analyses, and of the necessary steps to move forward in this crucial field of social and economic inquiry. With this as the global context, the five country synthesis studies form Part III. In Part IV, this volume then returns to analysing economic inequality in a broader context. It contains three chapters, which profile, respectively, subjective wellbeing across the world, socioeconomic mobility in China and the United States against the context of seventy-three other countries, and the need for a new growth and development model in developing economies. Collectively, these three substantive parts contain much detail on global and local inequalities and, in Part V, we synthesize key findings and outline policy implications.
Before proceeding, in what follows we provide a synopsis of the eleven individual chapters that make up Parts II–IV of this book to help set the scene and give some up-front guidance on what to expect.
2 Global Inequality and Inequality within Countries
Part II starts with Chapter 2, in which Martin Ravallion clarifies the underlying assumptions behind the conflicting narratives on what has been happening to global income inequality in the past decades. After reviewing the main issues involved, with regard to assessing global trends, he identifies a set of elements that put into perspective the fact that many prominent international scholars produce solid evidence in support of a conventional view according to which global inequality has been declining since the 1990s, while at the same time other prominent scholars produce solid evidence for the opposite view that it has been rising. These elements include people’s main reference group—their country or the world—and what type of income growth keeps inequality constant, which depends on the relative versus absolute view. Crucially, and unlike the measurement and analysis of poverty, a focus on inequality requires the inclusion of the full distribution of wellbeing. This is true whether it is explicitly recognized by the analyst or not, and Ravallion shows that opposing empirical pictures of the trends in global inequality can be understood by being explicit about the ethical weight given to the poor or the aversion to inequality from the top of the distribution that is being assumed in each of these analyses. Conclusions drawn about what has happened depend critically on the exact combination of assumptions made. Understanding the drivers of inequality, even income inequality, requires an analysis of how many dimensions of inequality intersect to produce a given inequality outcome. In Chapter 3, James Davies and Anthony Shorrocks emphasize this important point by analysing the joint distribution of wealth and income and their decomposition into between- and within-country components. This is a particularly insightful approach because both of these metrics feature prominently in the discussions of contemporary inequality, with most contention being about the relationship between them. The authors make this contribution drawing on the Credit Suisse Global Wealth Report and Credit Suisse Global Wealth Databook which they have produced (Davies et al. 2018a, 2018b). The chapter contrasts the trends in these wealth data, which have featured prominently in international inequality deliberations, with income data from the WIID. For both income and wealth, and for all the inequality indices considered, the degree of inequality attributable to differences in mean income and wealth across countries accounts for much, if not most, of the level of global inequality. As regards changing inequality over time, changes in mean income, mean wealth, and population size have induced a strong downward element in the trend in global inequality, regardless of the inequality index selected.
In Chapter 4, the Luxembourg Income Study (LIS) team (Daniele Checchi, Andrej Cupak, and Teresa Munzi) investigates the development patterns in economic inequality for several low- and middle-income countries, including the five country cases analysed in depth in Part II. They, along with Russia, now form
part of the LIS harmonized data, which has so far largely focused on developed economies. Thus, these data are a valuable source of information on inequality within the BRICS and the chapter splices a discussion of Russian inequality into the book. The authors describe the process by which each country data set is prepared for inclusion in the harmonized LIS data. This is invaluable in making clear the many challenges in moving from within-country data and analysis to wellgrounded cross-country analyses of inequality for developing countries, and then to analysis to inform strategies to overcome inequality. Without explicit description of what was done to the data in moving from country data sets to a comparable cross-country data set, and even to merged global data sets, one does not have a clear sense of the extent to which the observed inequality levels and trends are driven by differences in the underlying data rather than in the forces driving inequality.
3 Inequality in Five Developing Giants
Part III begins with Chapter 5, in which Marcelo Neri puts forward his synthesis of how, from the dawn of the new millennium, Brazil—one of the world’s most prominent high-inequality countries—experienced an unprecedented decline in income inequality that lasted until 2014, with a reversion after that date. Lower earnings inequality was the main driver behind this downward trend. Neri discusses a number of different possible channels that might have helped to reduce earnings inequality. These include heterogeneity among firms and the role of a higher minimum wage. The expansion of education, alongside falling returns, stands out among the driving factors helping to mitigate the particularly high inter-generational inertia. The chapter also addresses the complementary role of tax redistribution and conditional cash transfers in underpinning the reduction in educational and health inequalities among the Brazilian populace.
In Chapter 6, Shi Li, Terry Sicular, and Finn Tarp present China as a country with a classic development trajectory characterized by structural change, increased market integration, and labour absorption, moving along the upward side of the Kuznets inverted-U curve. This process has generated strong economic growth and poverty reduction. However, an incomplete transition from a planned to a market economy has generated increased differentiation of incomes, while creating continued opportunities for rent-seeking, corruption, and hidden income. More recent countervailing factors include expanding government efforts to moderate inequality, especially through social and welfare programmes. Whether the country has already started moving along the downward section of the Kuznets curve will depend on the direction of future reforms.
Hai-Anh Dang and Peter Lanjouw offer in Chapter 7 a comprehensive analysis of inequality trends in India over the past three decades. As with Brazil and
China, the backdrop was one of strong growth and poverty reduction. In the Indian context, this was accompanied by rising inequality in all dimensions, but to differing degrees depending on the dimension considered and the measurement method employed. In order to reconcile these growth, poverty, and inequality levels and changes, the authors interrogate inequality trends across and within the national, state, and microscopic village levels. They show that local-level inequality accounts for the bulk of overall inequality in India. They explore these inequality dynamics further, using synthetic panel data constructed at the household level, to examine intra-generational income mobility over time. This mobility has risen over time, implying that lifetime inequality will be lower than cross-sectional inequality. However, while poverty has fallen, most of the poor who have escaped poverty continue to face high risks of falling back into poverty. Particularly concerning is the fact that intragenerational educational mobility is low and not improving. Moreover, those who remain poor are increasingly chronically poor, and may be particularly difficult to reach via the introduction or expansion of safety nets.
In Chapter 8, Raymundo Campos-Vazquez, Nora Lustig, and John Scott investigate the ‘rise–decline–rise again’ pattern in income inequality in Mexico. The growth and poverty context for these inequality changes is itself much less positive and more variable in Mexico than in Brazil, China, and India. As in the case of Brazil, the evolution of labour income inequality is key to understanding this inequality pattern, with the skill premium being the key determinant of earnings inequality. The authors highlight two key implications. The first is the need to continue the expansion of access to higher levels of (quality) education and the increase in minimum wages to their 1980 levels. The authors also show that direct cash transfers, including the famous Progresa programme, largely targeted the poor in Mexico, but also that their overall redistributive impact remains limited because of their small coverage. Furthermore, the redistributive effect of the entire fiscal system has declined significantly since 2010, as transfers have become less progressive and net indirect taxes have increased.
Murray Leibbrandt, Pippa Green, and Vimal Ranchhod show in Chapter 9 that, in a context of sluggish growth but some reduction in poverty rates, inequality has remained exceptionally high in post-apartheid South Africa. This stubbornly high level of aggregate inequality masks a number of important changes in the texture of this inequality. Again, the dynamics of the labour market are key to understanding this. Education policy has effected a reduction in inequality in years of schooling driven by increases in average education levels. Under the conditions prevailing in the mid-1990s this would have reduced inequality. But over the intervening years the patterns of growth and employment were such that earnings inequality increased as the result of strongly increased returns to tertiary education and to experience, and strongly decreasing returns to levels of education lower than complete secondary. Using tax data, the authors also show that
those at the top end of the income distribution have experienced much higher rates of real income growth than the remainder of the population. This is linked to the rising labour market inequality and to strong returns to capital income sources that are only found at the top end of the income distribution. While government grants played an important role in reducing inequality (and poverty) by supporting households at the bottom of the income distribution, these effects have become weaker more recently. The direct tax system is progressive overall, and social benefits are well targeted, but some of the tax benefits are indeed regressive.
4 Inequality in a Broader Context
Part IV of the book moves the analysis from the detail in our country cases back to a broader context of global inequality. Thus far, the chapters in the book have interrogated the measurement and analysis of money-metric inequalities at global and national levels. Full interrogation of these inequalities has required an understanding of their intersection with inequalities in education, gender, race, ethnicity, space, and many other dimensions. These are complicated interactions and they raise the question of how perceptions of wellbeing are correlated with subjective and measured inequalities. In Chapter 10, Andrew E. Clark and Conchita D’Ambrosio empirically explore exactly this question. They explore whether individuals’ subjective evaluation of their present and future living conditions is associated with their own current levels of functioning, their subjective assessment of their position relative to other members of the society, and the level of measured economic inequality in their country. Using data from a series of Barometer surveys conducted in 76 countries in Africa, Latin America, Asia, and Europe, combined with money-metric measures contained in the WIID data, the authors are able to look in particular at how these relationships differ across the four world regions. Their results confirm that, across the world, those whose lives are more objectively deprived have lower subjective evaluations of current living standards and, to a lesser extent, lower expectations regarding future living conditions. While they also find evidence about the implications of relative comparisons, these are more heterogenous across world regions. In the developed countries of Europe, those who see themselves as better off relative to others in their country have higher levels of subjective wellbeing. In developing countries this is not a strong relationship. Indeed, Hirschman’s famous tunnel effect seems to prevail, especially in Latin America and Africa, in that seeing oneself as worse off relative to others is positively correlated with subjective wellbeing in the present and the future. Finally, the higher the aggregate level of inequality in the country of residence, the higher the expectations of future living conditions globally, with more mixed results on present living conditions also depending on the region.
