Chronic Poverty Report 2004-05

Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 1 Black plate (1,1)

The Chronic Poverty Report 2004–05

Chronic Poverty Research Centre Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 2 Black plate (2,1)

Acknowledgements

The Chronic Poverty Report 2004–05 is the outcome of research, analysis, Picture credits engagement, listening and thinking by the Chronic Poverty Research Cen- Front cover photos tre (CPRC) partners listed below. Top left: Uganda. Grandfather and orphaned child. CPRC would like to thank the many people living in chronic poverty who Giacomo Pirozzi/Panos have shared their experience and knowledge with us. This has informed Top right: Nepal. Woman carrying firewood, fodder and child. Steve McCurry/Magnum our understanding of chronic poverty and what should be done about it. Bottom left: Bangladesh. Living on the railway The report was written by Ursula Grant, David Hulme, Karen Moore and tracks. Stuart Franklin/Magnum Andrew Shepherd with advice and contributions from all CPRC partners: Bottom right: Ethiopia. Desperate farmer with Phil Amis, Armando Barrientos, Bob Baulch, Kate Bird, Sarah Bracking, handful of immature wheat. David Dahmen/Panos Chapter One Jonathan Goodhand, Tony German, Mark Gorman, Amanda Heslop, Sam Ethiopia, farmer with wheat: David Dahmen/Panos Hickey, Aasha Kapur Mehta, Uma Kothari, David Lawson, Charles Lwanga Bangladesh – Dacca railway tracks: Stuart Franklin/ Ntale, Rachel Marcus, Dorothy McCormick, Andy McKay, Diana Mitlin, Magnum Photos John Okidi, Judith Randel, Binayak Sen, Amita Shah, Cobus de Swardt, Nepal – carrying firewood and a child: Steve Andries du Toit and Rebecca Yeo. Background papers for The Chronic McCurry/Magnum Photos Poverty Report 2004–05 are listed at the end of the report. Uganda – grandfather and orphaned grandchild: Giacomo Pirozzi/Panos CPRC is very grateful to the following people outside the Chronic Poverty Chapter Two Research Centre who contributed comments and advice: Robert Archer, Namibian farmer: Mark Hakansson/Panos Laure Beaufils, Tony Bebbington, Olivia Bennett, Dirk Bezemer, Donal India – bonded labour rally: Paul Smith/Panos Brown, Arjan de Haan, Frank Ellis, Raghav Gaiha, Dave Gordon, Marty Romanian Roma children: Mark Hakansson/Panos Hanratty, Caroline Harper, Kelly Hawrylyshyn, Margaret Kakande, Nambusi Zambian child: Action on Disability and Development (ADD) Kyegombe, Nathalie LeBrun, Corinne Lennox, Keith Mackiggan, Eduardo Ethiopian older woman: Neil Cooper/Panos Masset, Simon Maxwell, Ruth Meinzen-Dick, Lauchlan Munro, Shailen Chapter Three Nandy, Laure-He´le`ne Piron, Lorraine Wapling, Joanne Wheeler, Howard Indonesian urban life: Martin Adler/Panos White and Shahin Yaqub. Thanks also to IDPM and ODI for administrative Somali villagers: Robert Knoth/Panos and other support and particularly to Roo Griffiths, Karen Morgan-Tallents, Chapter Four Eritrea, horticulture worker: Heldur Netocny/ Panos Jane Northey, Chris Taylor and Catherine Tucker for excellent administra- India, Uttar Pradesh – farmers: Alvaro Leiva/Panos tive, IT and copy editing assistance and to Mark Gordon who spent many Chapter Five person and computer hours on the cartogram. Finally, we would like to Zambia – boys working: ADD thank all the participants at the Chronic Poverty Research Centre confer- India, Bombay: Mark Henley/Panos ence, ‘Staying Poor: Chronic Poverty and Development Policy’, April 2003, Bangladesh, political meeting: ADD whose ideas and analysis have contributed to this report. South Africa, Grandmother and grand children: Paul Weinberg/Panos

The Chronic Poverty Research Centre is an independent research centre, funded with a grant from the UK Department for International Development (DFID). DFID can accept no responsibility for any information provided or views expressed in this report.

The Chronic Poverty Report 2004–05 is published by The Chronic Poverty Research Centre Tel: +44 (0) 161 275 2810 Institute for Development Policy & Management email: [email protected] University of Manchester www.chronicpoverty.org Harold Hankins Building Fax: +44 (0) 161 273 8829 Precinct Centre Oxford Road Manchester M13 9QH UK

The Chronic Poverty Report 2004–05 and/or sections of it can be reproduced without permission but please acknowledge CPRC as the source and provide readers with details of the Chronic Poverty website, www.chronicpoverty.org The full text can be downloaded free of charge from this website. If you would like a hard copy please contact CPRC at the address above or any CPRC partner organisation.

Cover, design and typesetting by discript.com, London WC2N 4BN Printed on recycled paper by Bell & Bain Limited, Glasgow Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 3 Black plate (3,1)

About the Chronic Poverty Research Centre (CPRC)

CPRC is an international partnership of universities, research institutes and NGOs. Director: Professor David Hulme, Institute for Development Policy & CPRC aims Management, University of Manchester, UK . to provide research, analysis and policy guidance Deputy Director: Dr Andrew Shepherd, Overseas Development Institute, London . to stimulate national and international debate . so that people in chronic poverty will have a greater say in the formulation of policy and a greater share in the beneﬁts of progress

Chronic Poverty Research Centre Partners The list of partners with links to websites, people, emails and telephone numbers can be found at www.chronicpoverty.org Action on Disability and Development Gujarat Institute of Development Research Vallis House Gota 57 Vallis Road Ahmedabad 380 060 Frome Gujarat Somerset India BA11 3EG www.gidr.ac.in UK HelpAge International www.add.org.uk PO Box 32832 Bangladesh Institute of Development Studies London N1 9ZN E-17 Agargaon UK Sher-e-Bangla Nagar www.helpage.org Dhaka 1207 Indian Institute of Public Administration Bangladesh www.bids-bd.org Indraprastha Estate Ring Road New Delhi 110002 Childhood Poverty Research and Policy Centre India Save the Children 17 Grove Lane Institute for Development Policy & Management University of Manchester London SE5 8RD UK Harold Hankins Building www.childhoodpoverty.org Precinct Centre Oxford Road Department of Economics and International Development Manchester University of Bath M13 9QH Bath UK BA2 7AY www.idpm.man.ac.uk UK Institute of Development Studies www.bath.ac.uk/econ-dev/ University of Sussex Development Initiatives Brighton Old Westbrook Farm BN1 9RE Evercreech UK Somerset www.ids.ac.uk BD4 6DS UK International Development Department www.devinit.org University of Birmingham Birmingham B15 2TT Development Research and Training UK PO Box 1599 www.idd.bham.ac.uk Kampala Uganda Overseas Development Institute www.drt.or.ug 111 Westminster Bridge Road London SE1 7JD Economic Policy Research Center UK Makerere University www.odi.org.uk PO Box 7841 Kampala University of the Western Cape Uganda Programme for Land and Agrarian Studies School of Government www.eprc.or.ug/ P. Bag X17 Bellville 7535 South Africa www.uwc.ac.za/plaas Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 4 Black plate (4,1)

Welcome to The Chronic Poverty Report 2004–05

This report comes to you from the Chronic Poverty Research Centre, one of a new breed of Development Research Centres supported by the UK Department for International Development. The CPRC is a virtual centre – an international partnership of universities, research institutes and NGOs from Bangladesh, India, Kenya, South Africa, Uganda and the UK. Development research, like all other activities supported by DFID, must contribute to poverty eradication and the achievement of the Millennium Development Goals. Research should not only be aimed at an academic audience, it must reach out to meet the needs of policy-makers and practitioners. This Chronic Poverty Report is designed to do just that. We hope it will stimulate thinking and debate on the ways that policy and practice can change so that the hundreds of millions of people living in chronic poverty are included in development progress. Working in policy-relevant research throws down new challenges. Re- searchers – used to developing and testing hypotheses, gathering data, creating analytical frameworks and producing findings – have to be much more responsive to the policy agenda and timeframes. For NGOs – used to starting with an advocacy position and picking out the evidence to back it up – it means a much more integrated approach to research and policy work. For all involved, this requires a degree of boldness. Short, intuitive mes- sages may understate the complexity or diversity of an issue, but they are what is needed by policy-makers. And the dynamics driving policy are often so tightly bound to specific timeframes that they cannot wait for research findings to be perfected. The Chronic Poverty Report 2004–05 is characteristic of policy-relevant research. Its objective is to open up the debate, not to present completed and closed findings. It is one plank in a raft of outputs including technical working papers, journals, newsletters, conferences, public meetings and, increasingly, direct engagement with policy-makers and practitioners on how to make development work for people in chronic poverty. A second Chronic Poverty Report will be coming out in 2006. We hope you will find The Chronic Poverty Report useful. A visit to the Chronic Poverty Research Centre website, www.chronicpoverty.org, will provide you with detailed background papers, information on CPRC partners and working methods – and if you are interested in chronic poverty and what can be done about it, we would like to hear from you.

David Hulme, Director, CPRC

Andrew Shepherd, Deputy Director, CPRC

Paul Spray, Head of Central Research Department DFID Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 5 Black plate (5,1)

The Chronic Poverty Report 2004–05

Overview

Between 300 and 420 million people are trapped in low-potential rural areas, politically-marginalised regions chronic poverty. They experience deprivation over many and areas that are not well connected to markets, ports years, often over their entire lives, and commonly pass or urban centres. There are also concentrations of poverty on to their children. Many chronically poor peo- chronically poor people in particular slum areas in towns ple die prematurely from health problems that are easily and cities as well as the millions of homeless people preventable. For them poverty is not simply about hav- sleeping in streets, stations, parks and burial grounds. ing a low income: it is about multidimensional depriva- The causes of chronic poverty are complex and usually tion – hunger, undernutrition, dirty drinking water, involve sets of overlaying factors. Sometimes they are the illiteracy, having no access to health services, social isola- same as the causes of poverty, only more intense, wide- tion and exploitation. Such deprivation and suffering ex- spread and lasting. In other cases, there is a qualitative ists in a world that has the knowledge and resources to difference between the causes of transitory and chronic eradicate it. poverty. Rarely is there a single, clear cause. Most This Report’s concern about chronic poverty leads to a chronic poverty is a result of multiple interacting factors focus on poverty dynamics – the changes in well-being or operating at levels from the intra-household to the glo- ill-being that individuals and households experience over bal. This is illustrated by Maymana and Mofizul’s story time (Chapter 1). Understanding such dynamics provides (Chapter 4): their chronic poverty is an outcome of ill- a sounder basis for formulating poverty eradication poli- health, widowhood, a saturated rural labour market, dis- cies than the conventional analysis of national poverty ability, social injustice and poor governance. Some of trends. these factors are maintainers of chronic poverty: they op- The chronically poor are not a distinct group. Many erate so as to keep poor people poor. Others are drivers different people suffer such deprivation (see Chapter 2); of chronic poverty: they push vulnerable non-poor and people who are discriminated against, stigmatised or ‘in- transitory poor people into poverty that they cannot find visible’: socially-marginalised ethnic, religious, indigen- a way out of. ous, nomadic and caste groups; migrants and bonded There are several important maintainers of chronic labourers; refugees and internal displacees; disabled peo- poverty. ple or those with ill-health (especially HIV/AIDS). In 1. No, low or narrowly-based economic growth means many contexts poor women and girls, children and older that there are few opportunities for poor people to people (especially widows) are likely to be trapped in raise their incomes and accumulate assets. poverty. 2. Social exclusion and adverse incorporation interact so While chronically poor people are found in all parts of that people experiencing discrimination and stigma the world (see Chapter 3 for an overview and Chapters 6 are forced to engage in economic activities and social to 10 for specific regions) the largest numbers live in relations that keep them poor – poorly paid, insecure South Asia (135 to 190 million). The highest incidence is work; low and declining assets; minimal access to so- in sub-Saharan Africa, where 30–40% of all present day cial protection and basic services; and dependency on ‘US$1/day’ poor people are trapped in poverty – an esti- a patron. mated 90 to 120 million people. East Asia has significant 3. In disadvantaged geographical and agro-ecological re- numbers of chronically poor people, between 55 to 85 gions poor resources, weak economic integration, so- million, living mainly in China. cial exclusion and political marginality create ‘logjams Within countries there are often distinct geographies of disadvantage’. of chronic poverty, with concentrations in remote and 4. High capability deprivation, especially during Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 6 Black plate (6,1)

vi . The Chronic Poverty Report 2004–05

childhood – poor nutrition, untreated sicknesses, lack discrimination and exploitation that drive and maintain of access to education – can diminish human develop- chronic poverty. ment irreversibly. 5. In weak, failing or failed states economic opportu- Action on chronic poverty needs a framework to: nities are few, lack of health services and social protection means that people can easily fall into desperate Prioritise livelihood security A much greater emphasis poverty, children go uneducated, violence destroys as- is needed on preventing and mitigating the shocks and sets and discourages investment, and poor people have insecurities that create and maintain chronic poverty. few means of asserting their rights. This is not only about providing recovery assistance but 6. Weak and failed international cooperation over the also about giving chronically poor people a secure posi- 1980s and 1990s has deepened poverty through struc- tion from which to seize opportunities and demand their tural adjustment and over-rapid economic liberalisa- rights. Thus, social protection policies are of great tion, allocated aid away from countries with large importance. numbers of chronically poor people and blocked off trade opportunities for poor countries. Ensure chronically poor people can take up oppor- Not all chronically poor people are born into long-term tunities It is crucial both to promote broad-based deprivation. Many slide into chronic poverty after a growth and to redistribute material and human assets, so shock or series of shocks that they cannot recover from. that chronically poor people can take up economic These include ill health and injury, environmental opportunities. shocks, natural disasters, violence, the breakdown of law and order, Obwaavu obumu buba Take empowerment seriously Po- and market and economic collapse. buzaale. Abaana licy must move beyond the cosy These are the drivers of chronic pov- babuyonka ku bazadde rhetoric of participatory approaches, erty. When shocks are severe and/or baabwe, ate nabo decentralisation and theories about repeated, when people have few pri- nebabugabira ku baana. rights. It needs to address the diffi- vate or collective assets to ‘fall back’ – Some poverty passes cult political process of challenging on, and when institutional support from one generation to the layers of discrimination that keep (social protection, public informa- another as if the offspring people trapped in poverty. tion, basic services, conflict preven- sucks it from the mother’s tion and resolution) is ineffective, Recognise obligations to provide breast. such processes are likely to trap peo- resources Chronic poverty cannot Source: group of disabled ple in poverty. be seriously reduced without real Ugandan women The knowledge now available transfers of resources and sustained, about chronic poverty must be used predictable finance. The political in- to mobilise public action and reshape development strat- difference to meeting national and international obliga- egy. While there are many policies that are potentially tions on poverty eradication needs to be challenged and beneficial for the poor and for the chronically poor, ways found to foster social solidarity across households, many people living in chronic poverty are not ‘just like communities and nations. the poor but a little bit further down the poverty spectrum’. Overcoming chronic poverty requires policy- The need for policy change must not mask the fact that it makers to reorder their priorities and set their sights is the chronically poor themselves who are the leading higher than the current consensus on poverty reduction actors in overcoming their poverty. To date, when their policy. existence is recognised, the chronically poor are per- Development strategy needs to move beyond the ceived both in policy and the popular imagination as bounds of its present emphasis on economic growth – dependent and passive. Nothing could be further from hundreds of millions of people are born poor and die the truth. Most people in chronic poverty are striving poor in the midst of increasing wealth. Chronically and working to improve their livelihoods, and the pro- poor people need more than ‘opportunities’ to improve spects for their children, in difficult circumstances that their situation. They need targeted support and protec- they have not chosen. They need real commitment, tion, and political action that confronts exclusion. If matched by actions and resources, to support their ef- policy is to open the door to genuine development for forts to attain their rights and overcome the obstacles chronically poor people, it must address the inequality, that trap them in poverty. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 7 Black plate (7,1)

Contents

PART A: Global chronic poverty in 2004–05 1

Chapter 1 What is chronic poverty and why does it matter? 3 What is different about chronic poverty . . . 3 . . . and why does it matter? 3 A window of opportunity to put chronic poverty on the agenda 4 Deﬁning chronic poverty 4 Poverty dynamics – becoming poor, staying poor, and escaping poverty 5 Disaggregating chronic and severe poverty 6 Multi-dimensional deprivation and low levels of assets 6 Who are the chronically poor? 7 Counting the chronically poor 8 Non-monetary indicators of chronic poverty 8 How many people are chronically poor? 9 Where do chronically poor people live? 9 Chronically poor places 12 Chronic poverty matters 12

Chapter 2 Who is chronically poor? 14 Discrimination reinforcing disadvantage 14 Minorities and indigenous people 15 Discrimination through oppressive labour relations 16 Displacement, powerlessness and chronic poverty 18 Disability and stigmatised illness 19 Women and girls in chronic poverty 21 Chronic poverty, age and the life-cycle 22 Multiple deprivation, discrimination and disadvantage 25

Chapter 3 Where do chronically poor people live? 26 Global dimensions of chronic poverty 26 Spatial poverty traps 26 Ecological characteristics of spatial poverty traps 31 Poor infrastructure 31 Weak institutions and organisations 32 Politics, rights and conﬂict in spatial poverty traps 33 Conclusion 34 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 8 Black plate (8,1)

viii . The Chronic Poverty Report 2004–05

Chapter 4 Why are people chronically poor? 36 Staying poor: the maintainers of poverty 36 Economic growth, growth quality and inequality 36 Social exclusion and adverse incorporation 37 Geography and agro-ecology 39 Cultures of poverty 39 Capability deprivation 40 Weak, ‘failing’ and ‘failed’ states 41 Weak and failed international co-operation 41 Falling into poverty: the drivers of chronic poverty 42 Access to assets 42 The nature of shocks 42 Institutional context 43 Understanding chronic poverty and taking action 46 Conclusion 49

Chapter 5 What should be done about chronic poverty? 50 Security ﬁrst 50 Prioritising livelihood security for chronically poor people 50 Tackling childhood poverty 52 Preventing and interrupting chronic poverty through health services 52 Opportunity is not enough: growth, inequality and redistribution 54 Empowerment: making rights real 55 The national agenda: reaching chronically poor people 56 Delivering basic services 56 Delivering social assistance 56 Using PRSs to prioritise the chronically poor 57 Putting chronic poverty on the international action agenda 58 Using the Millennium Development Goals to address chronic poverty 58 The coherence of MDGs and international policy 58 Financing chronic poverty reduction 58 Persistent poverty requires sustained assistance 60 Financing targeted transfers 60 Obligations on aid must be met 61 Chronic poverty – an agenda for action 61

PART B: Regional perspectives on the experience of chronic poverty 63

Chapter 6 Understanding chronic poverty in sub-Saharan Africa 65 Poverty trends in sub-Saharan Africa 65 How many people are chronically poor in sub-Saharan Africa? 67 Who are the chronically poor in sub-Saharan Africa? 67 Women-headed households and AIDS widowhood 68 People with low and declining access to income, land and other assets 69 Where are the chronically poor in sub-Saharan Africa? 69 Chronic poverty in remote rural sub-Saharan Africa 69 Urban chronic poverty in sub-Saharan Africa 70

Chapter 7 Understanding chronic poverty in South Asia 71 Poverty trends in South Asia 71 How many people are chronically poor in South Asia? 71 Who are the chronically poor in South Asia? 72 The working poor 72 Excluded minorities 72 Poor women, older women, disabled women and widows 73 The hungry, weak and ill 74 Where are the chronically poor in South Asia? 74 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 9 Black plate (9,1)

Contents . ix

Chapter 8 Understanding chronic poverty in Latin America and the Caribbean 79 Poverty trends in Latin America and the Caribbean 79 How many people are chronically poor in Latin America and the Caribbean? 80 Who are the chronically poor in Latin America and the Caribbean? 80 Those with less education 80 Race and ethnicity 80 The working poor 81 Women and women-headed households 81 Where are the chronically poor in Latin America and the Caribbean 81 Rural chronic poverty 81 Chronic poverty in megacities and smaller towns 82

Chapter 9 Understanding chronic poverty in transitional countries 83 Poverty trends in transitional countries 83 How many people are chronically poor in transitional countries? 84 Who are the chronically poor in transitional countries? 84 Ethnic minorities 84 Women 84 Single older people 85 Children 85 Where are the chronically poor in transitional countries? 85

Chapter 10 Understanding chronic poverty in China 87 Poverty trends in China 87 How many people are chronically poor in China? 87 Who are the chronically poor in China? 87 Where are the chronically poor in China? 88

PART C: Measuring global trends on chronic poverty: statistical appendix 89

Chapter 11 Measuring chronic poverty How many chronically poor people are there in the world? 91

APPENDIX A Glossary of terms 131 APPENDIX B References 134 APPENDIX C CPRC working papers and Chronic Poverty Report 2004–05 background papers 139 APPENDIX D The Chronic Poverty Report 2006–07 140 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 10 Black plate (10,1)

x . The Chronic Poverty Report 2004–05

Boxes Box 1.1 Poverty that persists – in their own words 3 Box 1.2 Poverty trends and poverty dynamics in 1990s Uganda 6 Box 1.3 Poverty lines 7 Box 1.4 Child malnutrition: is severe stunting the best proxy for chronic poverty? 8 Box 1.5 Researching chronic poverty 12

Box 2.1 The chronic poverty of the ‘non-existent’ 15 Box 2.2 Wage labour, migration and poverty in rural Bangladesh 17 Box 2.3 ‘We ran to the camps to save our lives, but entered into poverty’ 18 Box 2.4 ‘I am a refugee from the world’ 19 Box 2.5 Chronically poor woman – in a wealthy household 21 Box 2.6 Childhood poverty in Kyrgyzstan – a snapshot of the future 22 Box 2.7 Planning for the future 23 Box 2.8 ‘You are given only once, and if you are unfortunate, that is it’ 24

Box 3.1 Chronic Poverty by region 28 Box 3.2 Risk and vulnerability among Bangladeshi char communities 31 Box 3.3 The Batwa people of the Great Lakes Region 34

Box 4.1 Adversely incorporated, top to bottom 39 Box 4.2 Slums, shocks and chronic poverty 42 Box 4.3 Widowed, abused, and robbed 43 Box 4.4 The 2002 Malawi famine – bad weather, bad policies, increased vulnerability 43 Box 4.5 Violent conﬂict and a collapse into poverty in the Solomon Islands 45 Box 4.6 The cost of a cup of coffee – market collapse and poverty traps 46 Box 4.7 Maymana and Moﬁzul’s Story 47

Box 5.1 Poverty reduction policies that assist the poor and the chronically poor 51 Box 5.2 Lesson learned? HIV/AIDS in India and China 53 Box 5.3 Rojamma and the campaign against domestic violence and drunkenness 55 Box 5.4 Are rich countries and their citizens really committed to reducing poverty? Targetting MDG 8 56

Box 6.1 Understanding chronic poverty in middle income South Africa 68 Box 6.2 Voice from the ﬁeld 69

Box 7.1 ‘My heart feels as if it is being held with forceps’ 74 Box 7.2 Access to non-timber forest products in Orissa 77

Box 8.1 Sebastiana’s story, Ecuador 81 Box 8.2 Indigenous poverty and vulnerability in Bolivia 82

Box 9.1 Women ex-collective farm workers in Kyrgyzstan 85

Box 11.01 Concentrating aid on the poorest? 128 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 11 Black plate (11,1)

Contents . xi

Figures Figure 1.1 The chronically poor, transitory poor and non-poor – a categorisation 5 Figure 1.2 Comparative approaches to geographical concentrations of poverty 10 Figure 1.3 The ‘shape’ of multi-dimensional deprivation for different countries 11

Figure 3.1 The proportion and size of chronically poor populations in world regions 27

Figure 5.1 ODA to least developed countries in real (2001) prices (US$ millions) 60 Figure 5.2 DAC donors – gap between income and aid per capita (at 1988 prices and exchange rates) 61

Figure 6.1 Chronic poverty in sub-Saharan Africa 65

Figure 7.1 Chronic poverty in South Asia 71 Figure 7.2 Bangladeshi districts with highest HPI (2000) 75 Figure 7.3 Pakistani districts with lowest HDI (2003) 75 Figure 7.4.1 Indian states with above average proportion of the population below national poverty line (1993–4) 76 Figure 7.4.2 Indian regions with highest proportions of the population below national poverty and/or severe poverty lines (1993–4) 76 Figure 7.4.3 Indian states with below average HDI (1991) 76 Figure 7.4.4 India’s most deprived districts 76

Figure 8.1 Chronic poverty in Latin America and the Caribbean 79 Figure 8.2 Racial inequality and education in Brazil, 1999 81

Figure 11.1 Aid Concentration Curve for all DAC donors combined (2001) 128 Figure 11.2 Aid Concentration Curves for the Netherlands, Japan and the United States (2001) 128 Figure 11.3 Aid Concentration Curves for the UK, Germany and France (2001) 128 Figure 11.4 Aid Concentration Curves for the EC (2001) 129 Figure 11.5 Aid Concentration Curves for the World Bank (IDA) (2001) 129 Figure 11.6 Aid Concentration Curves for the UN (2001) 129 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P10E-01.3D.3D Folio: 12 Black plate (12,1)

xii . The Chronic Poverty Report 2004–05

Tables Table 1.1 Preliminary estimate of the world’s chronically poor (millions) 9

Table 2.1 Access to basic infrastructure by race in Ceres, South Africa 15 Table 2.2 Poverty head count by social group, Orissa and India, 1999-2000 (%) 16 Table 2.3 HIV-positive people and people living with AIDS, by region 20

Table 3.1 A conceptual framework for understanding spatial poverty traps 30

Table 4.1 Economic growth, chronic poverty and poverty dynamics 38 Table 4.2 The inﬂuence of state, market, society and family on Maymana’s and Moziful’s prospects 48

Table 5.1 The MDGs and the chronically poor – helping each other? 59

Table 6.1 Summary of poverty indicators for Africa 66 Table 6.2 The changing African picture, 1980–2000 66

Table 7.1 Summary of poverty indicators for South Asia 72 Table 7.2 Different approaches to chronic poverty in rural Pakistan 73 Table 7.3 Gender gap in adult literacy in South Asia 74 Table 7.4 Poorest Indian states 77

Table 8.1 Summary of poverty indicators in Latin America and the Caribbean 80 Table 8.2 Income polarisation in Latin America 80

Table 9.1 Summary of poverty indicators in transitional countries 83 Table 9.2 Households characteristics of Roma and non-Roma in Bulgaria, Hungary, and Romania 84

Table 11.1 Panel data on chronic and transitory poverty: selected countries 93 Table 11.1a Chronic Poverty in Nicaragua, 1998–2001 93 Table 11.1b Chronic Poverty in Kwa-Zulu Natal, South Africa 1993–1998 93 Table 11.1c Chronic Poverty in Uganda, 1992–1999 93 Table 11.1d Chronic Poverty in Vietnam, 1993–1998 93 Table 11.1e Chronic Poverty in Egypt, 1997–1999 94 Table 11.1f Chronic Poverty in Rural Bangladesh, 1998–2000 94 Table 11.1g Chronic Poverty in Rural Chile, 1968–1986 94 Table 11.1h Chronic Poverty in Rural China (Sichuan), 1991–1995 94 Table 11.1i Chronic Poverty in Rural India 1970/71 to 1981/82 94 Table 11.1j Chronic Poverty in Urban Ethiopia, 1994–1997 94 Table 11.2 Characteristics of Chronic Poverty in Rural Bangladesh 95 Table 11.3 Characteristics of Chronic Poverty in rural India 96 Table 11.4 Characteristics of Chronic Poverty in KwaZulu-Natal 97 Table 11.5 Characteristics of Chronic Poverty in Uganda 98 Table 11.6 Characteristics of Chronic Poverty in Vietnam 99 Table 11.7 Global Indicators of Chronic Poverty 100 Table 11.8 Global Indicators on Childhood Poverty 107 Table 11.9 Global Indicators on Inequality 114 Table 11.10 Resource Inﬂows 121 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 1 Black plate (1,1)

PART A Global chronic poverty in 2004–05 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 2 Black plate (2,1) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 3 Black plate (3,1)

different types of poverty and expressing the idea of a poverty that persists (Box 1.1). An effective response requires a better understanding of what it means What is chronic poverty to be chronically poor, and better analysis of the characteristics and underlying and why does it social processes that result in sustained 1 matter? and intractable poverty. . . . and why does it matter? Njuma, the gleaner The imperative to confront and eradicate Njuma is a childless widow who is almost 70 years old. She lives in a chronic poverty is a moral one. Interna- remote mountainous area of Uganda. Since her husband died, she has tional obligations to eradicate poverty lived alone, largely dependent on gifts of food from relatives and neigh- cannot be selectively applied, with bours. They do not wish to see her suffer from hunger but they are also chronically poor people excluded on the poor and do not see it as their role to provide beyond her minimum nutri- basis that they are too hard to reach. tional needs. She gets no support from the government or NGOs. If she Addressing chronic poverty is integral gets sick, she just has to wait until she is better, as she has no access to to the Millennium Development Goals health services. and poverty eradication. Persistent im- Despite feeling tired, often low-spirited and having no formal access to poverishment is not only a symptom of land or productive assets, she seeks opportunities to be economically past deprivation, it is also the cause of active. Njuma’s main work is gleaning coffee from neighbours’ bushes future destitution. There is increasing once they have been harvested – it is very hard work to glean from evidence that growth and the prospects poorly-maintained bushes that have already been picked, growing on for long-term poverty reduction are held steep slopes. She earns the equivalent of US$0.02–0.03 for each hour back by inequality and by the low re- she works. turns that the poorest people get on their Economic surveys and the census would, if they recognised her at all, labour. At the most basic level, people class her as poor and not working. The reality is that she is employed in cannot be productive unless their food some of the lowest paid work in the world. intake is enough to ensure that they can Source: Hulme, field notes, March 2000. work. What is different about chronic poverty . . . This report is about people living in health, education and social capital. The distinguishing feature of chronic poverty – people who remain They are the invisible poor, and occupy chronic poverty is extended poor for much or all of their lives, many a blind spot when it comes to the design duration. Such poverty is of whom will pass on their poverty to of development policy and the delivery hard to reverse. their children, and all too often die easily of public services. preventable deaths. The distinguishing feature of chronic People in chronic poverty are those poverty is extended duration. Such pov- who have benefited least from economic erty is hard to reverse. Differentiating Reaching the chronically poor is not growth and development. They, and poverty is not simply an issue for offi- simply a matter of implementing current their children, will make up the majority cials and researchers: people in poor policies more fully. Chronic poverty re- of the 900 million people who will still communities in developing countries also search suggests that millions of people be in poverty in 2015, even if the Millen- have many ways of distinguishing will remain in poverty without policies nium Development Goals are met. Chronic poverty exists in all regions, Box 1.1 Poverty that persists – in their own words and chronically poor people live in many different situations. If and when they Zimbabwe: Nhamo inokandira mazai – Poverty that lays eggs.1 have work, it is insecure, casual and at Uganda: Obwaavu obumu buba buzaale. Abaana babuyonka ku bazadde extremely low rates of pay. Many live in baabwe, ate nabo nebabugabira ku baana. – Some poverty remote rural areas, urban slums or con- passes from one generation to another as if the offspring sucks it flict zones, suffer from chronic ill health from the mother’s breast.2 or impairments. Chronic poverty partic- 3 uarly affects children, older people and Lesotho: Bo-mophela ka thata – Those who live in a hard way. people with disabilities. People in Ghana: A beggar with two bags (someone who has to beg during the chronic poverty face layers of social dis- season of plenty as well as the season of hunger).4 crimination, often based on ethnicity, re- Tamil Nadu, India: Yarukku oruvalai sapadu ellayo avango thane allai. – Those who ligion or language. Chronically poor do not have even a single meal in a day.5 people have little access to productive as- 6 sets and low capabilities in terms of Ethiopia: Wuha anfari. – Those who cook water. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 4 Black plate (4,1)

4 . Chapter 1

addressed later rather than sooner, poverty can become more intractable. Those ‘left behind’ in processes of development often have little choice but to ﬁnd ways of coping that undermine their long-term well-being, and that of society as a whole. When people are so poor that they cannot afford to risk new approaches, evidence suggests this may be at the expense of aggregate growth and long term poverty reduction.8 It is clear that the right to development should not be selectively applied, targeting only those citizens who are the easiest and cheapest to assist. The target of halving absolute poverty must be pursued in a way that includes and beneﬁts people who are chronically poor.

Hundreds of millions of people will remain in poverty without policies that specifically address their situation. The present generation of poverty reduction strategies will not dramatically reduce that specifically address their situation and what is currently on offer from do- with substantial and well targeted assis- nors. The proposal is indicative of a levels of chronic poverty within tance. Understanding the manifestations, growing awareness that the MDGs will an acceptable timeframe. attributes and social dynamics of chronic not be met without adequate funding. poverty is essential in developing such The Millennium Declaration commit- effective public interventions. ted 189 governments to ‘making the Currently, few PRSPs disaggregate right to development a reality for every- poverty adequately, let alone examine A window of opportunity one and to freeing the entire human race poverty dynamics.9 This means that even to put chronic poverty on from want’. Under the declaration, the if the present generation of poverty-re- the agenda right to development is clearly universal. duction strategies are highly effective, This has important implications for the they will not dramatically reduce levels Governments and aid agencies are cur- way in which governments and aid agen- of chronic poverty within an acceptable rently showing an unprecedented interest cies pursue the MDGs. timeframe. in poverty. Many donor agencies have Policy makers seeking to make quick The current focus on poverty, opens a produced policy statements prioritising progress on specific MDGs, may perceive window of opportunity to ensure that poverty reduction. At the national level, trade-offs between efficiency in reaching the political, social and economic conse- Poverty Reduction Strategy Papers headline targets and effectiveness in quences of chronic poverty are better (PRSPs) have become the principal achieving ultimate goals. Certainly some understood by policy-makers. Policies framework within which donors and de- targets could be more easily achieved by that take account more effectively of the veloping countries address poverty. At excluding some of those who are hardest multidimensional nature of poverty, and the international level, the Millennium to reach.7 its impact on present and future genera- Development Goals (MDGs) provide For example, it might be easiest to re- tions, are more likely to target the needs global targets against which govern- duce maternal mortality by three- and rights of chronically poor people ments and aid donors can measure pro- quarters by concentrating resources on and create an enabling environment for gress towards the ultimate goal of cities and well-connected villages while everyone. poverty eradication. The MDGs repre- abandoning remote rural areas, until all sent an unparalleled commitment by of the ‘easy to reach’ have services. A governments from around the world to similar approach to country selectivity Defining chronic poverty create an enabling environment for pov- would see donors focus only on the A number of terms have been used to erty reduction. poor who happen to live in countries identify those who experience poverty The resources necessary to achieve the perceived to have a good record on gov- most intensely – ultra poor, extreme MDGs were defined at the 2002 Financ- ernance and conditionality, while ne- poor, hardcore poor, destitute, poorest ing for Development Summit in Monter- glecting the rights and needs of poor of the poor, and declining poor. Such rey. At that time, rich countries made people elsewhere. distinctions are not new. For example, in commitments to increase aid, but those But such an approach would not only 18th-century France social commenta- promises fall well short of what is overlook the universal right to develop- tors and public officials distinguished the needed. The proposal for an Internation- ment, it would also jeopardise the ulti- pauvre from the indigent and sought pol- al Finance Facility (IFF) is an attempt to mate goal of poverty eradication. If the icies to stop the seasonally poor becom- bridge the gap between what it is needed needs of people in chronic poverty are ing permanently poor.10 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 5 Black plate (5,1)

What is chronic poverty and why does it matter? . 5

There is broad agreement that poverty occurs when someone experiences a fundamental deprivation – a lack of some basic thing or things essential for human well-being. Intuitively, most people think they can recognise poverty - hunger, malnutrition, worn clothing, un- washed bodies, run-down housing (or no home at all), begging, lack of access to clean water, primary schooling or basic health services, and so on. This apparent consensus is, however, illusory – there is no objective way of defining poverty. The US$1/day11 criterion adopted by the MDGs has as many critics as it has supporters. The way that poverty is conceptualised is inherently about value preferences that vary between individuals, organisations and societies. Until the 1990s, poverty was consid- People in chronic poverty are actively working to improve their livelihoods and the prospects for their ered mainly in ‘material’ terms – as low children. income or low levels of material wealth. More recently, vulnerability and multi- . The chronically poor include:12 While the study of chronic poverty is dimensional deprivation, especially of – The always poor, whose poverty particularly interested in the always poor basic capabilities such as health and edu- score in each period is below a de- and usually poor, all of these categories cation, have been emphasised as key as- fined poverty line can be used in a dynamic sense to de- pects of poverty. Indeed, chronic poverty – The usually poor, whose mean pov- scribe poverty transitions. For example, is rarely the result of a single factor. In- erty score over all periods is less a household can be broadly seen as es- stead, combinations of, and interactions than the poverty line, but who are caping chronic poverty (an ‘escapee between, material poverty, extreme ca- not poor in every period. household’) when it moves from being pability deprivation and vulnerability . The transitory poor, who include: usually poor to being only occasionally often characterise the chronically poor. – The fluctuating poor,13 who are poor. Conversely, a household can be While for chronically poor people the poor in some periods but not in viewed as descending into chronic pov- different dimensions of deprivation inter- others, and have with a mean poverty (a ‘descending household’) when its act and overlap, it can be useful for poli- erty score around the poverty line; status shifts from being fluctuating poor cy-makers to keep them separate in – the occasionally poor, who have ex- to being always poor. ‘Fluctuating poor’ order to see how each facet of poverty perienced at least one period in pov- represents people who are frequently but relates to the other, and the consequen- erty; although their mean poverty not continuously in poverty, who are ces for the shape of material poverty. score is above the poverty line. certainly vulnerable, and those whose Thus capability deprivation (e.g. ill- . and the non-poor with poverty scores average position may be above or below health, lack of skills) may both underlie in all periods above the poverty line14 the poverty line. and result from material poverty, but is not the same thing as material poverty. This can enable a more sophisticated ap- Figure 1.1 The chronically poor, transitory poor and non-poor – a proach to policy making – work on ca- categorisation pability enhancement does not have to depend on income poverty reduction, for example. Mean score*

Poverty dynamics – Poverty Line* Poverty Line* becoming poor, staying Mean score* poor, and escaping poverty Poverty is not a static condition. The TimeTime Time Time Time study of poverty dynamics focuses on the Always poor Usually poor Fluctuating poor Occasionally poor Never poor } } } ways in which people’s poverty status changes, or does not change, over time. Chronically poor Transitory poor Non-poor Assessments of changes in poverty over time generally recognise ﬁve main poverty categories, under three main head- *Depending on data availability, poverty could be assessed in terms of household expenditure, income, consumption, a poverty index or scale, nutritional status, or an assessment of assets. ings, described in Figure 1.1: Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 6 Black plate (6,1)

6 . Chapter 1

intuitively plausible that it is much hard- Box 1.2 Poverty trends and poverty dynamics in 1990s Uganda er for someone who is well below a pov- Uganda experienced significant reduction in poverty during the 1990s. The erty line to advance far above it than for aggregate national poverty rate fell by about 20% over the 8 years from 1992 to someone who is closer to it. In Uganda, 1999, with substantial poverty reduction occurring everywhere in the country, there is strong evidence that this is the except the Northern region. However, this aggregate poverty trend tells us nothing case: between 1992 and 1998, the se- about what happened to individual households. A fall of 20% in the national verely poor found it much harder to es- poverty rate does not imply that 20% of households that were permanently poor cape poverty than the poor as a whole. have moved out of poverty, nor that all households have become 20% richer. However, in Kwazulu-Natal, South In fact, the poverty trend masks important poverty dynamics: about 19% of Africa, the picture is different. Between households were poor in both 1992 and 1999 (the chronically poor), and while 1993 and 1998, although most severely almost 30% of households moved out of poverty, another 10% moved in (the poor did improve their situation, the transitory poor). Clearly there are many households that have failed to benefit from large majority remained below the pov- Uganda’s impressive macroeconomic development over this period. erty line. On average, the severely poor This more nuanced understanding of poverty requires the collection of panel had about as much chance of escaping data alongside the standard household surveys. While household surveys collect poverty as the moderately poor. This data from a representative sample of households, these are not necessarily the was attributed to the importance of get- same households in each survey; panel data on the other hand are longitudinal ting a permanent job in the South Afri- datasets that track the same households over time.15 can context. The contrast exists despite Source: Lawson, McKay and Okidi, 2003. the reasonably high levels of economic growth in Uganda over the 1990s, and the economic stagnation in South Africa. Household poverty dynamics depend Concepts of severity and chronicity dis- on many factors – the characteristics aggregate ‘the poor’. ‘Poverty severity’ Multi-dimensional (and changes in characteristics) of the refers to the shortfall below the poverty household itself, trends in the economy, line; it is a static concept, capturing the deprivation and low levels society and physical environment, and fact that the poor are not equally poor to of assets sudden events – both shocks and the same level: some people are slightly People who are chronically poor are windfalls. below the poverty line, while others are likely to be multi-dimensionally de- The indicators that are used to meas- far below it. ‘Poverty chronicity’ on the prived; they experience income and/or ure or assess poverty have an important other hand captures the fact that some of health and/or education deprivations at influence on the degree to which poverty the poor are poor for a short period of the same time. It is the combination of appears to be chronic or transitory. time (the transitory poor) while others capability deprivation, low levels of ma- Some indicators can fluctuate greatly are poor for long periods (the chronically terial assets, and social or political mar- over limited periods of time (income, ex- poor). Poverty chronicity is therefore a ginality that keeps people poor over long penditure, hunger) while others are longitudinal concept, referring to persis- periods. Contextual factors will deter- much more stable (literacy, assets, height tence in poverty.16 mine which of these is particularly im- for age). Studies that focus on income It is commonly assumed that people portant in any particular country or poverty thus report higher levels of who experience the most severe poverty situation. change than those that focus on asset are least likely to escape poverty, and status. that those who have been in poverty for a It is the combination of Poverty dynamics are not the same as long time are most likely to be extremely capability deprivation, low poverty trends. For example, Box 1.2 de- poor. Indeed, a combination of severe levels of material assets, and scribes how, in the 1990s, headcount and chronic poverty unequivocally social or political marginality poverty rates fell by approximately 20% would present the worst form of poverty. that keeps people poor in Uganda – a very positive poverty However, the relationship between pov- over long periods. trend. But while about 30% of poor erty severity and poverty chronicity is households escaped poverty, 10% of pre- complex and only partly understood.17 viously non-poor households became Do high levels of severe poverty occur The international movement ATD poor, and about one-fifth of all house- in the same countries as chronic poverty? Fourth World links chronic poverty very holds remained poor over the decade Limited evidence suggests that there are strongly to severe and multidimensional (the chronically poor). poor countries with high levels of both poverty. La grande pauvrete´ is character- chronic and severe poverty (such as ised by a permanent absence of basic se- Ethiopia, India, Madagascar) and coun- curities, combined with an inability to Disaggregating chronic and tries where there are relatively low levels exercise rights and discharge responsibil- severe poverty of both (such as Chile and the Philip- ities. This poverty makes it impossible The concept of poverty is used to express pines). However, there are also countries for the poor ‘. . . to regain by themselves the idea that, whatever the minimum with high levels of one but not the other their rights and to assume their responsi- level of consumption (or another welfare (such as Peru and Bangladesh). bilities. At that stage the poorest need indicator), there exists ‘the poor’ whose Are the severely poor and the chroni- the support of others in order to emerge consumption is below that minimum. cally poor the same people?Itis from chronic poverty’.18 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 7 Black plate (7,1)

What is chronic poverty and why does it matter? . 7

Chronically poor people age, physical or mental disability. In appear. Many people who are reported Ghana these are called ‘God’s Poor’ as ‘not economically active’ are in fact en- commonly experience several ‘there is no obvious remedy’ for the gaged in domestic labour or in activities forms of disadvantage causes of their poverty.19 with low economic return, such as beg- and discrimination . Those who are economically active ging and gleaning. at the same time. but unable to escape poverty because Commonly, people who are chroni- of the terms of their employment, their cally poor are those who are engaged in lack of access to productive assets; or casual labour, those who live in house- If people possess few material assets social barriers that mean they are dis- holds with high dependency ratios, and (land, housing, equipment), the capabil- criminated against. This is sometimes those with few assets (human and social, ities they have, and the claims they can termed ‘adverse incorporation’.20 as well as physical or financial). Access make on others or society as a whole, This distinction is useful in terms of to assets is critical for exiting poverty. will be very significant in determining helping us to recognise the very different Work in India demonstrated that the whether or not they remain poor. Thus, people who experience chronic poverty main feature distinguishing the chroni- there is a strong argument for under- and the different processes that are oper- cally poor from the poor in general was standing chronic poverty in terms of as- ating to keep them in poverty. However, their limited ability to cope with sets, capabilities and claims rather than it is important to recognise that the dis- shocks.21 Becoming literate, having a monetary poverty alone. This type of tinction is not as clear cut as might house, increasing the area of cultivated analysis avoids the difficulties of calcu- lating income and expenditure, and of fluctuations in monetary poverty from Box 1.3 Poverty lines year to year, which can cloud the underlying picture. It can provide a basis for A fundamental dimension of poverty is the inability to adequately feed oneself and developing better measures than income one’s family and to meet other basic requirements such as clothing, housing and for targeting interventions to the chroni- healthcare. A poverty line represents the level of income or consumption cally poor, an argument furthered below. necessary to meet these minimum requirements. While determining this minimum has an important subjective element, poverty lines are typically anchored to minimum nutritional requirements, plus a modest allowance for non-food needs. Who are the chronically Many countries now report poverty figures relative to national poverty lines defined poor? on such a basis. India’s Planning Commission for example first computed its 28 The chronically poor are not a homo- national poverty line in the 1950s. With appropriate adjustments for inflation and genous group. Chronic poverty clearly other changes in living standard, the same line can be used to follow trends in affects people in many different situa- poverty over time. tions. In specific contexts there are differ- However, national poverty lines are not very suitable for cross-country ing sets of factors associated with comparisons, because the concept of poverty is very often different in different chronic poverty, and the causes of countries. Although many countries anchor their poverty lines in terms of the chronic poverty vary from region to re- minimum amount of food needed, different nutritional norms (normally varying gion, household to household and person between 2100 and 2300 Kcals per person per day) are used. A more fundamental to person. problem arises with regard to essential non-food needs, which are both very difficult The chronically poor are not simply a to determine and will vary among economies. Despite several notable advances in 29 list of vulnerable groups, but people the field in recent years there is still no commonly used method of establishing the who commonly experience several forms non-food component of the poverty line. Therefore, someone defined as poor of disadvantage and discrimination at according to a national line in one country might not be poor according to the poverty the same time (see Box 4.7 for Mayma- line of another. The problems of comparability are accentuated further when OECD na and Moziful’s story). Differing com- countries are considered, as many industrialised countries base their national binations of structural factors (labour poverty lines on a relative norm, such as half median income. and product markets, ethnicity, race, This desire to make cross-country comparisons was a major motivation behind caste, gender, religion, class, disability, the introduction of the US$1/day poverty line by the World Development Report of 30 refugee status, geographic location), life 1990. ‘Dollar a day’ poverty lines attempt to express domestic currency values of cycle factors (widowhood, household income or consumption in a common currency. This conversion requires composition, being young or elderly) purchasing power parity (PPP) values – exchange rates where the same basket of and idiosyncratic factors (natural disas- goods costs the same dollar amount in different countries. The validity of cross- ter, ill health, impairment, robbery) cre- country comparisons depends on the accuracy with which these PPP exchange ate and maintain the poverty of some rates are computed, as well as the comparability and reliability of the income or while giving others the chance to avoid consumption data. Both of these cause serious difficulties. While the currently or escape it. quoted figures in the 2003 World Development Indicators enable comparisons to It is possible to broadly distinguish be made, there are serious questions about the figures in a number of instances two particular groups of people enduring (e.g. Nicaragua, Pakistan, Uganda). chronic poverty: To establish a firmer international baseline, more work is urgently needed on . Those long term poor who are not both the comparability of national poverty lines and on alternative, more robust economically active because of health, methods for estimating PPP exchange rates. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 8 Black plate (8,1)

8 . Chapter 1 land, and increased income from live- lower than for the overall population.25 several countries with large numbers in stock, were all found to help people es- Dependency ratios are commonly very poverty, there are important questions cape poverty.22 high in chronically poor households. about their comparability and accuracy. While many are very small (single parent Given current data availability, a very or orphan-headed households, for exam- approximate estimate, with limited geo- Health shocks are often ple) and/or have too few able-bodied graphic disaggregation, is the most that what cause already poor earners, in the countries for which data can be expected at this stage. households to descend is available, chronically poor households Much work on material poverty is ex- into chronic poverty. have significantly more children than the pressed in monetary terms, measuring average.26 This is not surprising: many income, expenditure or consumption chronically poor households rely on fam- against a ‘poverty line’ (see Box 1.3). ily labour; child mortality rates are high; This especially applies to current meas- Access to land may be critical in some there are limited publicly-provided social ures of chronic poverty, and as such situations, but in others human capital, safety nets or pensions; and there is a these monetary measures are frequently housing, and access to public goods may market for child labour. Large families used in this report. be more important. While rural poverty are a rational choice, but one that can is generally much higher among landless also undermine the possibility that and near-landless people, there is no uni- households, and particularly children, Non-monetary indicators of form association of chronic poverty with can escape poverty. chronic poverty land inequality. In South Asia and South Classifying households by assets rather Africa, the chronically poor are likely to than the more usual expenditure reveals be landless or near-landless, but in Ugan- a bigger gap between rich and poor and da and Vietnam this is not the case.23 Counting the chronically gives a more stable picture over time. Human capital is key in contexts poor For example, an assets index, combin- where access to financial and material The common assumption is that the ‘al- ing household assets and housing qual- assets is highly constrained. Given the ways poor’ are much fewer than the ity, public goods, and land, was a better dependence of chronically poor people ‘sometimes poor’. The 2000/1 World De- predictor of school enrolment in India on their own labour, health is crucial. velopment Report cited studies on than current household expenditures, Ill-health both drives and maintains China, Ethiopia, Russia and Zimbabwe although inter-state differences were chronic poverty (see Chapters Two and to illustrate this.27 The Chronic Poverty also large: ‘On average across India a Four), and health shocks are often what Report 2004–05 presents evidence which rich (top 20% of the asset index) child cause already poor households to de- challenges this assumption. Further re- is 31 percentage points more likely to scend into chronic poverty. search will project this estimate into the be enrolled than a poor child (bottom For many, education may be the crit- future, based on a number of different 40%). This wealth gap varies from a ical pathway out of poverty. Formal edu- scenarios. low of only 5 percentage points in Ker- cation is often found to be strongly Estimating the global numbers of peo- ala, to 38% in Uttar Pradesh, and 43 associated with decreased probability of ple living in chronic povety is fraught percentage points in Bihar.’31 However, chronic poverty, as it improves the qual- with serious difficulties. Figures for glob- the analysis of non-monetary indicators ity of labour as an asset. In some con- al poverty, currently most developed for of poverty needs to proceed with cau- texts, such as Pakistan and China, this is consumption poverty in terms of the tion. They do not always correspond secondary schooling;24 in others, literacy US$1/day poverty line, are already very well with monetary measures – in Viet- alone makes a difference. Data analysis approximate. At the same time, there is nam for instance, the overlap between shows that in each of five countries the limited knowledge about poverty dy- monetary chronic poverty and non- average number of years of schooling for namics within countries. Even though es- monetary chronic deprivation was chronically poor adults was significantly timates of dynamics are available for found to be modest.32 Even an indicator like child stunting is only sometimes associated with monetary poverty, although stunting is Box 1.4 Child malnutrition: is severe stunting the best proxy for expected to be strongly associated with chronic poverty? chronic poverty (see Box 1.4). Data Over 15% of children under five years of age – 91 million children – in the showing high levels of chronic poverty developing world are severely stunted. Stunting is defined in terms of height for are generally from countries with high age; severe stunting is measured at more than 3 standard deviations below the levels of stunting. For example, chil- median international height for age. Over half the severely stunted children under dren in chronically poor households in 34 five live in South Asia. Uganda and Vietnam are more likely There is considerable evidence of serious and significant long term and inter- to be stunted than average children, generational effects of poor nutrition on physical and mental health, mortality and and in Vietnam and urban Uganda chronic poverty. Necessary action includes the promotion of greater food security, this is very markedly the case. There food supplementation, the promotion of later marriage and childbearing, and are nevertheless, still surprisingly high reducing gender bias in child feeding practices. incidences of stunting in the overall Source: ACC/SCN and James Commission 2000 in Harper, Marcus and Moore 2003. population in both countries. Research Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 9 Black plate (9,1)

What is chronic poverty and why does it matter? . 9 on malnutrition33 and poverty in gen- many of the countries with the highest eral has suggested that malnutrition is levels of poverty incidence, including Where do chronically spread across wealth groups, and is Bangladesh, Brazil, China, India, Indo- poor people live? more related to beliefs about appropri- nesia, Pakistan, the Philippines, Russia The CPRC has to date largely focused on ate diet, to feeding and weaning prac- and Vietnam, based on a broadly com- sub-Saharan Africa and South Asia. tices, gender inequality and to parable nutritionally-based national pov- There are good reasons for this focus: maternal deprivation than it is to in- erty line. sub-Saharan Africa arguably has the come. Child stunting may be a partic- Based on figures for these and other highest levels of chronic poverty, while ularly good indicator of chronic countries for which panel data is avail- South Asia almost certainly contains the poverty, but the evidence on this re- able, of the 1.2 billion people that are in majority of the world’s chronically poor. mains to be consolidated. extreme poverty in US$1/day terms, it is This should not, however, obscure the estimated that the global number of the fact that chronic absolute poverty exists Howmanypeopleare chronically poor is between just under in many parts of the world – including 300 million and 420 million. In other East Asia,37 South-East Asia and Latin chronically poor? words, around a quarter to a third of America.38 the people living on less than US$1/day Figure 1.2 illustrates a stark similarity Between are chronically poor. The upper end of in conclusions drawn by three different 300 and 420 million people this range is more plausible, given that data sets: most countries for which the panel data . CPRC data has been mapped to show are chronically poor. are available have had positive rates of both ‘desperately’ and ‘very’ deprived The upper end of this range economic growth for some time, while countries. Also included are those is more plausible. there are a large number of countries, countries for which there is insufficient mainly in Africa (where panel data is data but where other evidence suggests much scarcer), which have not been strongly they would be included in one The lack of panel data, the different pe- growing. or other of these categories. riods between each wave of data collec- The 12 countries with available panel . UNCTAD Least Developed tion, and the different poverty lines data that form the basis of this calcula- Countries.39 utilised mean that, at present, only tion account for 78% of the world’s . UNDP ‘Top Priority’ countries (with broad estimates can be made. An initial US$1/day poor. Adjustments have been entrenched human poverty combined guesstimate gave a range of 389–727 made to data to enable use of a uniform with failing or reversing progress) and million people in 32 developing coun- definition of chronic poverty as ‘still ‘High Priority’ countries (not so des- tries including China and India.35 New being poor after 5 years’,36 and to ex- perately poor but failing or reversing analysis detailed below estimates that trapolate from the parts of a country to progress, or extremely poor with mod- between 300 and 420 million people are the whole. The assumption is that mo- erate progress).40 chronically poor. The new estimate, pre- bility around the US$1/day poverty line In Figure 1.3 the different shapes of sented in Table 1.1, combines US$1/day is the same as around the national pov- chronic poverty in terms of multidimen- poverty figures with the available panel erty lines used in the panel data analy- sionality and intensity among poor coun- data. This is an approximate exercise ses. The other 22% of the world’s tries are explored. It is clear from both given differences between the panel data US$1/day poor are assumed to be Figure 1.2 and 1.3 that, at the global sets and the approaches used for meas- chronically poor at the average rate for level Africa experiences chronic poverty urement. But panel data is available for the 12 countries. on a continental scale. Chapters Six to

Table 1.1 Preliminary estimate of the world’s chronically poor (millions)

Number Average percentage of poor US$1/day Estimated chronic poverty assumed chronically poor poor for Estimated for entire region over a ﬁve year period countries US$1/day where this is poverty for Region Population available entire region Low estimate High estimate Low estimate High estimate

Sub-Saharan 658.7 216.4 303.3 91.0 121.3 30.0% 40.0% Africa East Asia and 1807.8 277.0 312.8 53.7 84.9 17.2% 27.2% Paciﬁc South Asia 1355.1 524.7 535.6 133.9 187.5 25.0% 35.0% Rest of world 1149.6 81.0 88.0 19.8 28.0 22.5% 31.8% All 4971.2 1099.1 1239.7 298.3 421.7

For more detailed information on estimations, methods and data sources see Part C. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 10 Black plate (10,1)

10 . Chapter 1

Figure 1.2 Comparative approaches to geographical concentrations of poverty

Chronic poverty clusters

Desperately deprived

Moderately deprived

Insufficient data, but likely to be in a cluster

Source: CPRC analysis.

Human Development Report 2003 Priority Countries Top priority

High priority

Source: UNDP 2003.

UNCTAD’S 2002 LEAST DEVELOPED COUNTRIES

Source: UNCTAD 2002. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 11 Black plate (11,1)

What is chronic poverty and why does it matter? . 11

Figure 1.3 The ‘shape’ of multi-dimensional deprivation for different Key Severe countries Stunting

Desperately deprived countries Percentage of The mortality the population rate for surviving on less children Burkina Faso Burundi Central Africa Rep. Côte d'Ivoire than US$1/day under five

The probability Illiteracy of dying before among the age of 40 women

Mozambique Ethiopia Mauritania Malawi It is likely that intense or multidimensional poverty will be hard to shed. Figure 1.3 illustrates how the intensity of poverty varies Niger Nigeria Rwanda Tanzania remarkably among poor countries. The more substantial star shaped countries are those where deprivation is greater than those with smaller stars or other shapes. The indicators have been selected to tell us something about likelihood of long Moderately deprived countries duration poverty. Severe child stunting, female illiteracy and very Ghana Bangladesh Cameroon India low life expectancy are all deprivations which are thought likely to be associated with chronically poor people. Countries with high levels of Madagascar Kenya Nepal these ‘bads’ are likely to have high levels of chronic poverty. Multi-dimensionally deprived Lao PDR countries can be divided into two categories: if they have two or three Uganda significant rays their population is Yemen moderately multi-dimensionally deprived; if they have four or five we Senegal can speak about severe multi- Zimbabwe dimensional deprivation. It is these countries where a substantial proportion of poverty is likely to be chronic and where chronic poverty will be hardest to reduce. They Relatively non-deprived countries include: Bangladesh, India, Lao PDR, and Nepal in Asia (though Algeria Bolivia Brazil Colombia Ecuador none of these are in the worst category); Burkina Faso, Burundi, Cameroon, Central African Republic, Guatemala Honduras Cote d’Ivoire, Ethiopia, Gambia, El Salvador Egypt Iran Ghana, Kenya, Lesotho, Madagascar, Malawi, Mauritania, Mozambique, Niger, Nigeria, Rwanda, Senegal, Tanzania, Jordan Mexico Mongolia Morocco Peru Uganda, and Zimbabwe in sub- Saharan Africa. Outside these two regions only Guatemala, Honduras, Bolivia and Yemen might possibly Russian Fed Tunisia Ukraine Venezuela Vietnam fall into this category. Countries without data are likely to be desperately or very deprived. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 12 Black plate (12,1)

12 . Chapter 1

Ten of this report examine the level and by comparing the chronically poor and policy. This is an initial assessment, as characteristics of chronic poverty in dif- the overall population. Even the chroni- knowledge is limited and it takes time to ferent world regions. cally poor in South African and Viet- accumulate (Box 1.5). A fuller analysis is namese towns have better access to planned in The Chronic Poverty Report electricity than the average rural 2006–07. Chronically poor places household. The first five chapters of The Chronic Transcending national borders, chronic Poverty Report 2004–05 examine what poverty is also concentrated in specific is known about chronic poverty, its geographical areas, such as remote rural Chronic poverty matters causes and the policy implications. or poorly-connected areas, areas affected Between 300 million and 420 million Chapters Six to Ten review what is by conflict and so on, where inhabitants people are chronically poor. They live known about chronic poverty in differ- are affected by common vulnerabilities in absolute poverty for extended peri- ent regions of the world, providing read- to natural hazards, pollution, agro-cli- ods, often for all of their lives. People ers who have a particular geographical matic shocks, instability and violence, are born poor, die poor and their pov- focus with information and sources. The for example. erty is transferred to their children. For last part, Measuring Global Trends on Chronic poverty is more prevalent in them, poverty is not simply about Chronic Poverty, is a statistical appendix rural areas than in urban,41 and espe- having a low income: it is about multi- bringing together data about chronic cially so in remote rural regions (which dimensional poverty – hunger, undernu- poverty and encouraging researchers and may include towns and cities). Moun- trition, dirty drinking water, illiteracy, policy-makers to use such data more tainous areas may be particularly likely no access to health services, social isola- extensively. to have concentrations of chronic pov- tion and exploitation. Such deprivation In the next chapter (Chapter Two), erty: examples include China and Viet- and suffering exists in a world that has the social characteristics associated with nam.42 In India, increases in the size of the knowledge and resources to eradi- chronic poverty, especially social exclu- a village over time, and proximity to a cate it. sion and adverse incorporation in the large urban centre connected by good The task of this first Chronic Poverty economy, are reviewed. Chapter Three infrastructure, are also factors that ena- Report is to raise awareness about looks at the spatial patterns of chronic ble exit from chronic poverty.43 Access chronically poor people, summarise poverty – the countries, and types of to public goods in particular varies what is known about their lives, and out- area within countries, where such depri- more by urban or rural location than line what this means for development vation is concentrated. In Chapter Four the causes of chronic poverty are examined: the drivers that push people into Box 1.5 Researching chronic poverty long-term poverty and the maintainers that keep people poor. Chapter Five Deepening understanding of chronic poverty demands an effective combination of examines the implications of this know- quantitative and qualitative methods and interdisciplinary analysis. Academics, ledge for policy and asks ‘What should researchers and policy analysts have written much about this in recent times: be done?’ It highlights the need to pri- unfortunately, the levels of methodological and disciplinary integration that have oritise livelihood security and allocate been achieved remain limited. There are also usually long lags between the more attention and resources to social creation of knowledge and its application to policy. And, when poverty reduction protection policies; the need for growth policies are approved, their implementation is commonly weak or manipulated. to be broad based and equitable; the There are also a number of more specific problems. The longitudinal quantitative need for increased focus on how to em- and qualitative datasets we need to assess the scale of chronic poverty and power poor and chronically poor peo- understand its dynamics are rare – we must strive to create such data and ple; and, the national and international demand that other actors recognise its importance if they genuinely wish to move actions, especially increased aid flows, to more evidence based policy making. There are also opportunities to be more that must underpin such policy imaginative in the ways that we use existing information.44 It is already clear that changes. the processes that underpin chronic poverty are complex and dynamic – this To date, when their existence is recog- makes analysis technically demanding and creates particular difficulties in coming nised, the chronically poor are perceived up with the simple, policy relevant ‘sound bites’ that can influence contemporary both in policy and the popular imagina- decision making processes. tion as dependent and passive. Nothing Particular problems exist in focusing on the chronically poor. In terms of political could be further from the truth. Most organisation and power these are usually the world’s ‘last’. How can one support people in chronic poverty are strategising the empowerment of those who are most likely to see empowerment as so far and working hard to improve their live- away, or as such a false promise, that it can only be a diversion from the pressing lihoods and the prospects of their chil- tasks of survival and nurturing children? dren, in difficult circumstances that they Practically, there are many ethical problems of working closely with the have not chosen. They need real moral chronically poor – how can sensitive researchers and service delivery personnel and political commitment, matched by manage social relations with people who live on the margins of survival? One actions and resources, to support their cannot easily meet with people experiencing hunger or lacking basic medical care efforts to attain their rights, and over- for want of one or two dollars, and then say ‘Thank you for the interview’, and walk come the obstacles that trap them in away. poverty. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P11E-02.3D.3D Folio: 13 Black plate (13,1)

What is chronic poverty and why does it matter? . 13

Notes 1. Admos Chimhowu, personal communication. representative household surveys, in 1992 and 31. Filmer and Pritchett 1998. 2. Lwanga-Ntale 2003, from a group of disabled 1999, and is further supported by annual 32. Baulch 2003. women. monitoring surveys (1993 to 1996), and two 33. Gaiha (2003) used stunting of children <5 years 3. PANOS Mountain Voices http:// national participatory poverty assessments, of age as an indicator of malnutrition. www.mountainvoices.org/. which complement quantitative poverty work 34. Gordon et al. 2003. 4. Oduro and Aryee 2003. by bringing a multidimensional perspective. 35. Hulme and Shepherd 2003:412–413. 5. Solomon 2003. Both of the nationally representative surveys, 36. It is worth noting that within the CPRC, 6. Devereux and Sharp 2003. the Integrated Household Survey (IHS) of 1992 discussions of the time concept focus on life- 7. This argument is further developed in Chapter and Ugandan National Household Survey time and intergenerational periods rather than Five. (UNHS) 1999/2000 adopted two-stage stratified seasonality or longer time frames/histories used 8. Ravallion 2003. random sampling methodologies in the by other researchers. No great claim is made for 9. In Africa, the Malawian and Ghanaian PRSs collection of a 9,886 and 10,696 household the five-year period; it is simply analytically make some distinctions, but there is little observations, respectively. Both provide a rich convenient, being close to that which many sustained analysis, and even more limited source of information on socio-economic, crop studies have worked. Logically, lengthening the follow through into policy prescription and community levels data and form the basis time period would be likely to reduce the (www.odi.org.uk/prspsynthesis/ of a two-wave panel (1992 and 2000) that proportion of a population that is chronically synthesis1.pdf). See also covers 1,398 re-interviewed households. The poor. Comparing the first survey with a survey www.chronicpoverty.org/cpchip.htm. panels only represent small sub-samples of the x years later discounts any movement in 10. Hufton 1974. national surveys, but the poverty incidence between these years, which may be significant. 11. ‘‘The proportion of people below $1 a day is figures based on the panel households are There are complex methodological issues the percentage of the population with average broadly similar to the national level figures. involved, discussed in Part C. consumption expenditures less than $1.08 a day 16. Yaqub 2003:1–2. 37. See McCulloch and Calandrino 2003. measured in 1993 prices converted using 17. Yaqub 2003. 38. Helwege 1995. purchasing power parity (PPP) rates. The $1.08 18. Wodon 2001:30. 39. It should be noted that both the methodology a day standard was chosen to be equal to the 19. Narayan et al. 1999: 28–29. that UNCTAD 2002 use, and the validity of the median of the lowest ten poverty lines among a 20. Patron-client relationships are a common idea of ‘chronically poor countries’ are being set of low-income countries.’’ (UN Statistical example of such circumstances. challenged. Office). 21. Gaiha 1999, 1999. 40. ‘Top Priority’ countries are those failing on at 12. See Box 1.3 for a discussion of poverty lines. 22. Bhide and Mehta 2004. least three goals, or half the goals for which Here ‘poverty scores’ and ‘poverty lines’ refer to 23. See Part C. they have data, with a minimum of three data arbitrary amounts and levels of well-being, 24. McCulloch and Baulch 2000; Jalan and points, or if they only have data for two goals based on income, consumption, nutritional Ravallion 1999, 2000. they are top priority in both. ‘High Priority’ status, assets, human deprivation index, 25. Bangladesh, India, South Africa, Uganda and countries are top or high priority for at least participatory ranking etc. Vietnam. Chronically poor people did not three goals, are top priority for two goals, or 13. Also know as the ‘churning poor’. generally go to secondary school (except in are top or high priority for at least half the 14. The concepts here were derived from Jalan and South Africa, a middle income country), and goals for which they have data, with a Ravallion (2000), but the terms are used with they were much less literate (except in Vietnam, minimum of three data points. If they only have different meanings. For a discussion of how this which has high overall levels of literacy). In data for two goals they are top priority in both categorization can be adapted to incorporate some countries there are also big differences in (HDR 2003). the severity of poverty see Figure 3 in Hulme, primary school attendance between chronically 41. However, it should be noted that some writers Moore and Shepherd (2001:13). However, the poor children and others (rural India, Uganda, claim that urban poverty is significantly reader should note that this categorization Vietnam). See Part C. underreported (see Mitlin 2003). needs further development. For example, a 26. McCulloch and Baulch 2000; Jalan and 42. Baulch 2003; McCulloch and Calandrino 2003. dramatic, short-term downturn or ‘spike’ in the Ravallion 1998, 1999, 2000; Aliber 2001. 43. Bhide and Mehta 2004. welfare of a fluctuating or occasionally poor 27. World Bank 2001, Chapter One. 44. For example, see Baulch et al 2002 who household. 28. Dandekar and Rath 1971, Lipton and Ravallion develop non-monetary indicators of poverty for 15. Panel data are micro-longitudinal datasets that 1995. a dataset that was designed to work on income/ track people over time. The Ugandan panel 29. Ravallion 1994, 1998. consumption poverty. data is drawn from two nationally 30. World Bank 1990. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 14 Black plate (14,1)

For example, the poverty of a low- caste, older widow living in a remote village in India, is sustained through the interaction of discrimination and depri- Who is vation based on caste, gender, age, marital status and geographical location. chronically Discrimination and deprivation often poor? overlap and reinforce each other. Ethnic 2 minorities are, for example, more likely to live in remote rural areas, less likely to speak the national language, more Poverty bites likely to have distinct religious beliefs ‘Chronic poverty is that poverty that is ever present and never ceases. It and customs, less likely to hold official is like the rains of the grasshopper season that beat you consistently and citizenship documents, more likely to ex- for a very long time. You become completely soaked because you have perience economic and political discrimi- no way out. If the whole of your village is poor, then all its residents will nation, and more likely to be the objects be perpetually poor. Some poverty passes from one generation to anoth- of racial abuse and violence than other er, as if the offspring sucks it from the mother’s breast. They in turn pass poor people. These kinds of multiple dis- it on to their children. advantage block access to the opportuni- If you did not inherit land, and you are not a political leader, and you ties and resources necessary to escape did not go to school, and your relations do not feel proud of you, then poverty, and force many into marginal, poverty will bite you very hard – forever and ever – amen. exploitative, unsustainable livelihoods3 Now remember that a disabled person cannot inherit land. A brother’s that permit survival but further under- child may even be preferred in inheritance if he is not disabled. Similarly mine well-being in the longer term. disabled people do not get to leadership positions, and most are not even We often speak of chronically poor educated. Where else can you find this dire poverty?’ households in this report. But some indi- – Group of disabled women in Nkokonjeru Providence Home, viduals in less poor households may also Mukono, Uganda be chronically poor – their poverty is Source: Lwanga-Ntale 2003:8. lasting and hard to escape even when other members of the household enjoy a Discrimination reinforcing . household composition (e.g. young relatively good quality of life. Well-being disadvantage families; households headed by dis- is unequally distributed within house- People who are chronically poor are not abled people, children, older people holds – along lines of gender, age, mari- a homogeneous group – they are trapped and widows). tal status, fertility, wife order, birth in poverty for a range of reasons. In this While not everybody in these groups suf- order, health status and disability – as chapter, we focus on the experiences of fers chronic poverty, those who experi- well as between households. people whose chronic poverty is based ence several forms of disadvantage and Knowledge of the extent of chronic on discrimination and disadvantage due discrimination simultaneously are most poverty among different groups varies to their position within their households, likely to be chronically poor. greatly. The number of refugees and communities, and countries. Several different bases for social marginalisation, discrimination and disadvantage have been identified, by both poverty researchers and the poor themselves:1,2 . ‘ascribed status’ (e.g. ethnicity, race, religion and caste); . oppressive labour relations (e.g. migrant, stigmatised and bonded labourers); . position as an ‘outsider’ (e.g. migrant labourers, refugees and internally displaced people, those without the documents necessary to access citizenship rights); . disability; . stigmatised ill-health, especially HIV/ AIDS; . gender; . age (e.g. children, youth and older Deprivation re-inforcing disadvantage: This man, from a minority ethnic group in Namibia, is also disabled people); and – the victim of a landmine Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 15 Black plate (15,1)

Who is chronically poor? . 15 internally-displaced people around the world, and the problems they face, are Box 2.1 The chronic poverty of the ‘non-existent’ reasonably well-known and monitored. In Bolivia, the discrimination and exclusion of indigenous peoples is both cause By contrast, the numbers of disabled and effect of chronic poverty. Despite constitutional recognition, mainstream Bolivia people in developing countries, and their does not accept the validity of indigenous languages and cultural systems, with needs, problems and means of support, their non-European approaches to names, dates, and inheritance. Combined with remain major omissions in the poverty the high costs associated with remoteness, this means that only a tiny minority of literature. The exclusion of many indigenous people have any documentation, like a birth certificate or identity card. chronically poor people from official sta- Without documents, a person formally does not exist. She or he cannot access tistics both reflects and reinforces their any of the rights available to citizens. The ‘undocumented’ face difficulties in: position of marginality and vulnerability, . enrolling their children in school; often rendering them invisible, even to . registering and selling property; humanitarian organisations and their . accessing communal natural resources; own governments. Indeed, in terms of . getting work; citizenship, many people living in . accessing social benefits; chronic poverty do not even exist . accessing credit; (Box 2.1). . travelling freely; . initiating legal proceedings or defending themselves if taken to court; and . participating in political processes, as voters or candidates. Minorities and indigenous As an example, consider the case of an indigenous Quechua man who does not people have legalised documentation. Without an identity document, he cannot show that Being born into a marginalised group he is the owner of the plot of land that he works, and he cannot access can have serious and enduring repercus- agricultural microcredit programmes, including those aimed at supporting the sions on the extent to which someone is poorest farmers. Unable to meet the needs of his family, he is forced to migrate to able to access resources, build a secure the peri-urban area of Cochabamba, to sell his labour. Without documentation and livelihood, or even imagine a life without marginalised as an indio (a term of abuse or disrespect for those who are or poverty. Indigenous minorities are often appear to be of indigenous origin), he has little chance of getting stable work. He also linguistic and religious minorities, now faces exploitation in the informal sector, as a porter waiting on a housewife and often live in remote regions. Box 2.1 who requires his services. His precarious situation forces his children to beg on describes how indigenous people in Boli- public transport and his wife to beg in the street. via are excluded from mainstream Boli- Source: Leoń et al. 2003:3–5. vian civic, economic and social life, and highlights the chronic poverty that re- disproportionately live in poor, remote 1992–93/1997–98, it only dropped from sults. After migration to an urban area, and mountainous regions. They are less 86% to 75% among minorities.7 the family’s lack of documentaion and than 9% of the total population but In sub-Saharan Africa, research on dif- their status as indigenous people and about 40% of those in absolute poverty.5 ferential rates of child mortality by eth- ‘outsiders’ continue to limit their oppor- Research comparing the living stand- nic group suggests that, in addition to tunities, and undermine their children’s ards of 53 ethnic minority groups in Viet- widespread poverty, many countries are future well-being. nam to that of the Kinh majority also as marked by socio-economic inequality, In China, rural-to-urban migrants shows significant disparities and extreme and severe and chronic poverty, as coun- from ethnic minority groups, speaking discrimination.6 Ethnic minorities make tries in other regions of the world.8 Sur- minority languages, are the most margin- up 29% of the poor but only 14% of the vey data collected in the 1990s in 11 alised in labour markets, access to edu- total population. Further, most people sub-Saharan African countries9 showed cation, and social interactions. Wai di belonging to ethnic minorities remained particularly striking results for Kenya, ren (‘outsiders’) are often blamed for so- trapped in poverty during a period of where the likelihood of mortality for Ki- cial problems, such as crime and otherwise pro-poor growth: while the kuyu infants was 65% lower than for disease.4 Members of the more than 50 poverty headcount among the majority children of other ethnic groups and 74% ethnic minority groups in China fell from 54% to 31% over the period lower for children under five.

Table 2.1 Access to basic infrastructure by race in Ceres, South Africa

Often gone Never gone Have access Often gone Never gone without without to a ﬂush without fuel without fuel adequate adequate toilet for heating for heating shelter shelter and cooking and cooking

Proportion of African households 32% 50% 62% 27% 36%

Proportion of coloured households 1% 98% 98% 9% 60%

Source: de Swardt, 2003. N=543 households. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 16 Black plate (16,1)

16 . Chapter 2

Racial inequality – and the associated by both tribal (indigenous) and low-caste south Indian states, scheduled castes are spatial inequality – continues to play a peoples, to greater and lesser extents. relatively deprived: survey data from central role in the poverty and well- ‘Caste’ is a complex and dynamic system 1994 showed that per capita incomes being of South Africans, despite the end of social stratification that has evolved among scheduled castes were 24% (in of apartheid. Not only is South Africa from occupational groupings into a Andhra Pradesh) and 41% (in Kerala) one of the most unequal countries in the socio-economic hierarchy into which in- lower than the state averages.13 world (the income inequality ratio be- dividuals are born, and thus is related to Table 2.2 describes the situation in the tween whites and Africans has risen ethnicity as well as class. ‘Tribe’ is an eastern state of Orissa, which has a large from four times in 1994 to six times in ethnic categorisation that also comes at tribal population inhabiting remote rural 2000), but research in the fruit- the bottom of the caste hierarchy. In areas: more than 24% are classified as producing area of Ceres in the Western India, scheduled castes (SCs) are a collec- scheduled tribes, as compared to less Cape demonstrates the extent to which tion of castes formerly known as ‘un- than 9% for India as a whole. 92% of racial hierarchies remain intact (Table touchables’ that have been ‘scheduled’ people in rural southern Orissa belong- 2.1). There are significant disparities in for positive discrimination in education ing to a scheduled tribe are poor – twice income and assets, not only between and employment. Scheduled tribes (STs) the state poverty rate and three and a African and white people, but also be- are identified on the basis of certain cri- half times the national poverty rate. tween African and ‘coloured’ people – a teria including distinctive culture and mixed group of ethnicities historically pre-agricultural modes of production. considered as those who are neither Discrimination through white, nor indigenous, nor Asian. How- When people’s only social safety oppressive labour relations ever, while many ‘coloured’ people have net is other wage labourers, low Members of marginalised social groups better access to basic infrastructure than wages, job insecurity, poor are often also on the losing side of op- Africans, most remain stigmatised and working conditions and gruelling pressive labour relations, which can many experience severe discrimination. work, combine to create a situation range from relatively benign systems of In post-socialist Europe, it is the five of high vulnerabilty to shocks. migration for wage labour, to the most to seven million Roma (often known as extreme forms of forced and bonded la- ‘gypsies’) who are the most vulnerable to bour, and trafficking of children and poverty, over years, lifetimes and genera- Although the harsh oppression associ- women. tions. Discriminated against and often ated with untouchability has been banned isolated, the Roma have far worse living in India for more than half a century, Wage labour and migration conditions and far less education than both groups continue to face discrimina- Households that depend on daily wage non-Roma. In Hungary, ethnicity is the tion. Scheduled tribes suffer highly labour in the agricultural and urban in- most important factor in chronic pov- limited and often declining access to the formal sectors are often chronically erty: one-third of those that are poor for natural resources, especially forests, upon poor, or at high risk of becoming so. four or more years are Roma although which their livelihoods are based. While Low wages, job insecurity, poor working they make up only 4–5% of the popula- some scheduled caste families are margin- conditions and gruelling work, combine tion. 53% of Roma are long-term al small farmers, in rural areas most are to create a situation of high vulnerability poor.10 In Bulgaria, Hungary and Roma- functionally landless and work as agricul- to shocks – when the only social safety nia combined, rates of poverty among tural labourers. In the urban areas, a net is other wage labourers. An illness or Roma range from two to 13 times that large proportion of casual workers in the an accident at work, a daughter’s mar- among non-Roma, depending on the informal sector, including stigmatised riage or bad weather, can initiate a measure used.11 workers such as sweepers, are from the downward spiral from which it is diffi- Throughout South Asia, the links be- scheduled castes. It is estimated that two cult to emerge [see Chapter Four]. tween chronic poverty and ascribed sta- thirds of India’s bonded labourers are Women and children tend to be in an tus are played out through the persistent, from the scheduled castes and scheduled even more precarious position – they severe poverty and exclusion experienced tribes.12 Even in the more ‘progressive’ often receive much lower wages than

Table 2.2 Poverty head count by social group, Orissa and India, 1999-2000 (%)

Rural Urban

Scheduled Scheduled Scheduled Scheduled Tribe Caste All Tribe Caste All TOTAL Coastal Orissa 67 42 32 63 76 42 n/a Southern Orissa 92 89 87 72 85 44 n/a Northern Orissa 62 57 50 54 63 46 n/a All-Orissa 73 52 48 59 72 44 47 All-India 44 35 27 38 39 24 26

Source: de Haan and Dubey, 2003. Figures have been rounded. All = entire population of region Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 17 Black plate (17,1)

Who is chronically poor? . 17 their male counterparts, disproportionate to the amount of work they actually do, Box 2.2 Wage labour, migration and poverty in rural Bangladesh and often only receive payment in kind. Nandu lives with his wife and three daughters in a village in western Bangladesh. As for Nandu (Box 2.2), migration is He has a small piece of land but it is not enough to support the family, so he has often part of a broader set of livelihood to sell his labour to others. He has no specific occupation, but works as a day strategies employed by poor wage la- labourer on whatever is available. He gets work on paddy land during the bourers. Migration presents a set of par- monsoon season, for which he earns about Taka 50–60/day (about US$1). As this adoxes. For some people and in some is not enough to support his family, he looks for other work such as digging soil or forms, migration is an important factor cutting trees. He can earn Taka 40 in a day by doing such work, but it is not in escaping poverty, if not for the mi- common work and he doesn’t get the opportunity to earn extra money very often. grant, then perhaps for his or her chil- Nandu, like many others, is unemployed in the lean period so he is forced to dren. For many, however, moving from migrate to other districts to look for work. He has a lot of problems maintaining his one place to another does not make family, and worries about how he will provide dowries for the marriages of his them better off – for some, migration three daughters. deepens poverty.14 Source: Hossain, Khan and Seeley 2003. Much migration for work undertaken by the poor in South Asia and sub-Sahar- sporadic or low, and when decision- widely used method of forced labour. an Africa is rural-rural, temporary and making powers remain with the absent While bonded labour is practiced around seasonal.15 Chasing scarce, short-term, migrant.18 the world, most bonded labourers are insecure, and low-paid wage labour from found in South Asia, particularly among area to area, migrant labourers often Forced into labour, forced into low status castes and tribes in India, find themselves in a constant battle to poverty Nepal and Pakistan. In these countries, repay debt and maintain household con- There are millions of people around the bondage is most common in agriculture, sumption levels. Migrants such as the world for whom livelihood opportunities but it is also found in sectors such as Quechua family described in Box 2.1 are are far more than ‘limited’: they have ab- mining, gem polishing, brick-kilns, car- often discriminated against within the la- solutely no choice about where they live pets and textiles, and domestic service.19 bour markets, and lack access to health, and work, or how they survive. Often In many areas, social and technological education, housing and other services. working in extremely dangerous condi- change has led to a decline in tied labour tions, for oppressively long hours, little arrangements, but as argued for the case Bonded and trafficked or no remuneration, and away from of India, ‘the skein of patronage may un- labourers are locked into their families and communities, bonded coil only to recoil in the form of debt- and trafficked labourers are locked into bondage and labour attachment’.20 livelihoods that provide no livelihoods that provide no opportunity Bonded labour exists in several forms. opportunity to move out of to move out of poverty. Chronic, intergenerational poverty is poverty. In 1999, the United Nations Working perhaps most clearly exemplified by Group on Contemporary Forms of Slav- bondage that encompasses an entire ery estimated that despite national laws family and is permanent, year-round, In China, living without an official res- and international human rights agree- and inheritable. Labourers are bonded to idence permit means that migrants to ments, there were 20 million bonded la- their employer through various means – urban areas have no legal status, and bourers globally, making it the most often through indebtedness, but also must pay high fees for their children to attend school. According to an official 1997 estimate, enrolment rates were only 12% among the more than 100,000 migrant children aged 6–14 years.16 In Bangladesh, migrants and their families report facing many problems at their destinations, including accommodation difficulties, problems in sending money home, sickness and disease, robbery, and physical harassment.17 Consequently, staying put may be a more successful livelihood strategy. But those left behind are often already chronically poor, with no assets to draw upon, and unable to find work somewhere else. Women, children, older people, and the sick or disabled left behind when others migrate for work can also find themselves at increased risk of chronic poverty when remittances are Children, many of whom work as stonebreakers, at a rally against bonded labour in Maratana, India Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 18 Black plate (18,1)

18 . Chapter 2 through violence, and through feudal- It also does not include people traf- livelihood26 entered conditional share- type systems whereby households are de- ficked within countries. For example, in cropping agreements in which women pendent on landlords for their basic southern Sudan, it has been estimated and children continue to have to work needs. In the most extreme cases, indebt- that between the late 1980s and 2000, in for free.27 On the cottonseed farms of edness may actually grow over time de- the course of the civil war and longer Andhra Pradesh, contracted by both na- spite the complete commitment of all term intra-communal conflict, 14,000 tional and multinational capital, men’s household labour to the employer. ‘Em- women and children were abducted.24 emancipation from bonded labour has ployers’ tend to be less concerned with Often taken to the North, they are fostered the emergence of a young, fe- workers’ well-being and future than with forced to work either for their captors or male, and unfree labour force.28 maintaining a cheap (poor, dependent sold on to other people. and vulnerable) workforce. Marginalised, Displacement, dependent and often illiterate, bonded la- Those who have had to flee powerlessness and chronic bourers have little recourse if an employ- their homes due to poverty er increases an interest rate or charges persecution or violent conflict fees without the worker’s knowledge. The effects of displacement Limiting access to property is a key are often already members of Violent conflict destroys personal assets way to keep people as property: ‘it is a marginalised groups. as well as social and physical infrastruc- cruel irony that in some countries a ture, with long-term effects on the woman may be subjected to a bonded health, education and livelihoods status labour debt but not be able to buy or Even those who escape bondage, traf- of countries, households and individuals inherit the land she could use to produce ficking or enslavement – through, for ex- – particularly of children and older peo- income to cancel it’.21 In Mauritania, ample, improved law enforcement, a ple. Those who have had to flee their where slavery has been abolished several peace treaty, or NGO ‘rescue’ – may not homes due to persecution or violent con- times, the inheritance of servitude is fa- escape poverty. Severe material poverty flict – sometimes many times and over cilitated by the norm that members of combined with a denial of basic human many years – are often already members ‘slave castes’ cannot bequeath property rights – especially a lack of access to of marginalised groups and as such tend to their children.22 basic education and health care – contin- to find themselves in situations of ex- As ‘outsiders’ and ‘illegals’, and often ue to limit opportunity. Enduring asym- treme vulnerability. Like migrant com- completely dependent on their captors metries of power, status, information munities in general, refugees and for survival, people who have been traf- and opportunity mean that former Internally Displaced People tend to suf- ficked are especially vulnerable to long- bonded labourers and other poor casual fer high levels of discrimination and ex- term deprivation. It is estimated that be- workers remain vulnerable to the kinds clusion (see Box 2.3), have few economic tween 700,000 and two million women of oppressive labour relations that main- opportunities, and are vulnerable to and children are trafficked across borders tain poverty (see Chapter Four). forced labour, violence and eviction. Per- each year globally.23 This figure does not A lack of alternatives (e.g. sources of sonal documentation lost or destroyed include those workers who migrated by credit25), can make workers ‘voluntarily during displacement further undermines choice but who are susceptible to forced choose’ tied labour contracts. In Nepal access to state and relief services. Dis- labour on arrival, because they do not for example, after the abolition of placed women, children and older people have proper documentation or because bonded labour in 2001, many ‘freed’ are especially vulnerable to human rights their employers are able to isolate them, bonded labourers who found themselves violations.29 as is often the case with domestic work. forced off the land and out of a In most cases, the displacement caused by conflict exacerbates conflict’s effects, leaving refugees and IDPs Box 2.3 ‘We ran to the camps to save our lives, but entered into chronically poor, both within camps poverty’ and outside them. Limited access to In Uganda, most internally displaced people (IDPs), especially those in camp land and other livelihoods assets along- situations, are very unlikely to escape chronic poverty. side restrictions on work and mobility, . They have no land or other assets, having left everything behind when they fled and poor camp infrastructure seriously their homes. undermine food security. IDP camps in . IDPs and especially refugees are discriminated against and denied access to Somalia, for example, are overcrowded development activities. and have high levels of disease.30 Rates . They are regarded as foreigners or outsiders by the host community. of acute and severe acute malnutrition . They lack social capital, often having lost family and community members in the for under-5s range from four to 20 conflict. times the national average in some . They are often traumatised, affecting their capability to build new livelihoods. Ugandan IDP camps.31 This can have . Many depend on relief aid or begging. negative consequences for the non-dis- In a camp in Kitgum District, northern Uganda, IDPs told researchers: ‘We ran to placed as, in a country such as Eritrea, the camps to save our lives, but entered into poverty’. The host community of this communities that host IDPs have re- camp also reported that they were poorer since the camp was established, ceived little assistance and suffer as a because they had given up land for the IDPs. result of increased demand on their Source: Lwanga-Ntale and McClean 2003. land, services and resources. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 19 Black plate (19,1)

Who is chronically poor? . 19

The problem of IDPs and other poor concern’ are under the age of 18.35 In 22 status in Europe or North America can people can be similar, camps however countries, 16 of which are in sub- often rebuild lives and livelihoods. Con- draw resources away from refugees set- Saharan Africa, at least 55% are under trary to the perceptions of citizens in tled outside camps and local popula- 18. According to Save the Children, rich countries, the main countries of asy- tions. In the Democratic Republic of 60% of IDPs in Sierra Leone are chil- lum, however, are those with high pov- Congo (DRC), insecurity as well as re- dren.36 While only about 7% of ‘people erty levels themselves – Asia and Africa moteness and poor infrastructure has of concern’ are 60 years of age or over, host about three-quarters of all refugees meant that millions of IDPs are effec- in nine countries in East-Central Europe, and two-thirds of ‘people of concern’.38 tively beyond the reach of humanitarian North Africa and Latin America the rate organisations.32 The International Res- is 15% or more. In some refugee camps, cue Committee estimated that the major- the children, youth and/or older people Disability and stigmatised ity of the 2.5 million war-related deaths are the majority. In one Sudanese camp, illness in DRC between August 1998 and April 94% were children aged four or under; By definition, those living in poverty 2001 could be attributed to disease and in another, only 14% were aged between often have limited access to adequate malnutrition, as people fled from vio- 18 and 60. Such ‘communities’ have health care, food, education, shelter, and lence into the bush, or into communities such high dependency ratios that they employment, often enduring hazardous already exhausted and impoverished by are absolutely dependent on external working conditions. All these factors in- war. The FAO estimates that three-quar- provision. crease the risk of illness, injury and ters of the population of DRC was impairment. In Chapter Four, the signifi- undernourished in 1999–2001 – the Disabled people are more cance of ill-health, injury and impair- highest rate of undernourishment in the likely to be poor and to stay ments to the lives and livelihoods of world – up from 62% in 1995–97 and chronically poor people is discussed. 31% in 1990–92.33 Particularly in east- poor. Poor disabled children However, the effects of an impairment ern DRC, chronic conflict and displace- are more likely to die early, cannot be fully understood only by ac- ment, plus a volcanic eruption, has led preventable deaths. counting for money spent on treatment, to a situation of constant and extreme or money lost through days not earning. deprivation. As expressed by people with disabilities In addition to those displaced by in Uganda (Box page 14) and Pakistan The scope of the problem armed conflict, many millions of people (Box 2.4), the impact of stigma and ex- At the beginning of 2003, the United Na- have been forcibly relocated by state and clusion on the lives and livelihoods of tions High Commissioner for Refugees paramilitary forces attempting to control disabled people and their households can identified 20.6 million ‘people of con- insurgency movements or resource-rich intensify these effects, trapping people in cern’ throughout the world – about one territory; by states attempting to recon- poverty. The social aspect of impairment out of every 300 persons.34 This includes figure the ethnic make-up of particular has a name – disability. about 5.8 million IDPs to whom regions, such as in Indonesia and Bangla- Disabled people are more likely to be UNHCR was providing support; the Spe- desh; and even by ‘development’ itself, in poor, and to stay poor. Existing informa- cial U.N. Representative for IDPs esti- the form of, for example, large-scale tion, though extremely limited in its mates that there are in fact as many as dams or exclusion from forests or na- scope and comparability, points to a dis- 25 million IDPs worldwide. Nearly half tional parks. In India, even when there proportionate number of disabled people of the global IDPs have been displaced have been attempts to compensate dis- living in poverty in all countries. In through protracted conflict in just three placed people, it has been argued ‘that urban Uganda in the early 1990s, pov- African countries: Angola, Sudan, and very few of the 21–33 million develop- erty among the 5% of households with a the DRC. In Africa, IDPs outnumber ref- ment-induced IDPs . . . have had their disabled head is estimated to be at least ugees by about four to one, totalling ap- livelihoods fully restored’.37 38% higher than among households proximately 13.5 million people. Those with sufficient education or with a non-disabled head.39 In Sri About 46% of these ‘people of skills who are granted refugee or asylum Lanka, the Department of Social Services

Box 2.4 ‘I am a refugee from the world’ Sixty years ago at the age of five, Muhammad Arif was attacked by the polio virus. Due to a lack of medical facilities, and a lack of money, he did not receive proper treatment and lost both his legs. From that time, Md Arif found himself alone in society – he could not get married and was living with his brother as a burden upon him. The government has not supported him. He applied for zakat* from the government but it was not provided regularly. When fieldworkers met him, he was sitting in the corner of the house weeping. He told them, ‘I have not been able to enjoy the colours of life. I am a refugee from the world’. * An Islamic system of charity, formalised into a safety net programme by the Government of Pakistan. Source: Pakistan Participatory Poverty Assessment Roma children in Hungary, where ethnicity is the – Azad Jammu and Kashmir State Report, 2003. most important factor in chronic poverty. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 20 Black plate (20,1)

20 . Chapter 2

a cause and an effect of poverty, it is not inevitable that impairment, illness and injury lead to stigmatisation, exclusion, discrimination and disability. However, once marginalisation on the basis of an impairment occurs, the likelihood is that a vicious cycle of exclusion, loss of income, and persistent poverty will emerge. Some extreme natural environments including mountains and areas prone to flooding, can present particular challenges for people with physical impairments. However, it is the human-made physical environment that serves to exclude most disabled people in some way. This ‘apartheid by design’,44 including buildings with steps and narrow entran- ces, inaccessible ‘public’ transport, edu- If Kauyne is helped to walk, his impairment may not result in chronic poverty. cation and health facilities, and a scarcity of accessible information all has noted that approximately 8% of the existent or inaccessible specialised health serve to keep disabled people out, population are classified as disabled, services. pushed to the margins and without the 90% of whom are unemployed and de- Physical impairments are not the only information they need to participate pendent on their families.40 In China, ac- way in which disability interacts with equally. cording to 1997 data, of approximately poverty. Cross national surveys have Discrimination against disabled people 60 million disabled people, 17 million shown that common mental disorders is rooted in widely-shared attitudes, val- were absolutely poor; in the wealthy are about twice as frequent among the ues and beliefs. People rationalise the ex- province of Jiangsu disabled people poor as among the rich in Brazil, Chile, clusion and ostracism of disabled people made up over 60% of the poor.41 India, and Zimbabwe.43 and their families in many different When the numbers don’t show that ways. Beliefs that disability is associated disabled people and their households A vicious cycle: impairment, with evil, witchcraft, bad omens or infi- are more likely to be poor, evidence discrimination, disability and delity persist in many parts of the world. suggests that this is because of ‘excess poverty Disabled people also often experience mortality’ (and, in some contexts, ex- Disabled people have a higher likelihood suffocating overprotection and exclusion cess institutionalisation) of disabled of experiencing long-lasting poverty be- from everyday challenges. Low expecta- children in poor households.42 Disabled cause of the environmental, attitudinal tions of disabled people are often held by children in poor households are more and institutional discrimination faced, wider society as well as by themselves. likely to die early preventable deaths from birth or the moment of impairment Institutional discrimination is the than their wealthier counterparts, be- onwards. A person with an impairment process by which disabled people are sys- cause poor households are often unable only becomes disabled when physical tematically marginalised by established to provide the care and treatment re- and social barriers limit her or his oppor- laws, customs or practices, based on quired by a disabled child, particularly tunities. For example, where eyeglasses these discriminatory attitudes. Many aid when there are other children to feed are affordable, available and socially ac- agencies, including NGOs, for instance, who seem more likely to survive and ceptable, short-sightedness is rarely dis- make no attempt to include disabled contribute, and in the context of non- abling. While disability is certainly both people in their work; some implicitly

Table 2.3 HIV-positive people and people living with AIDS, by region

Adult % of HIV- Adults and children Adults and children prevalence positive adults living with HIV/AIDS newly infected with Adult and child deaths rate (%) who are (2003) HIV/AIDS (2003) due to HIV/AIDS (2003) (2003)* women (2002) TOTAL 40 million 5 million 3 million 1.1% 50 (34–46 million) (4.2–5.8 million) (2.5–3.5 million) (0.9–1.3%) Sub-Saharan Africa 25.0–28.2 million 3.0–3.4 million 2.2–2.4 million 7.5–8.5 58 South/South-East Asia 4.6–8.2 million 610 000–1.1million 330 000–590 000 0.4–0.8 37

Child = 0–14 years of age, Adult = 15–49 years of age * Adult prevalence rate = the proportion of adults living with HIV/AIDS in 2003, using 2003 population numbers. Source: UNAIDS/WHO 2002, 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 21 Black plate (21,1)

Who is chronically poor? . 21 exclude them through ‘fitness’ requirements. In many countries, disabled chil- Box 2.5 Chronically poor woman – in a wealthy household dren are not required to go to school and A Tanzanian farmer reported ill-treatment by both her wealthy husband and his there is no special provision for their first wife. Doubts over the paternity of her child make her vulnerable in her needs if they do enrol. Banks often do marriage, and she has no command over household resources. She sells her not accept disabled customers. Employ- labour in order to get extra clothes and food for herself and her child, and depends ers often do not consider the needs of on neighbours to help her with salt and soap. She would like her father to return disabled applicants.45 the bride price so that she can separate from her husband, but he claims that he no longer has the money. She feels she ‘has no language’ to report the situation Illness, stigma and persistent to the hamlet chairman who could intervene, so she just has to stay and tolerate poverty the situation. ‘Those staff members who know Source: Cleaver 2003. about me talk about it. They point at me and say, ‘Look, he is the HIV fel- Women and girls in chronic The cycle of maternal and child mal- low.’ They . . . keep their distance poverty nutrition, morbidity and mortality is one from me and remain aloof. I don’t Throughout their lives, poor women per- of the most significant means through share my tiffin box with them any form a triple role: reproductive work (in- which poverty persists over generations: more. I don’t feel like coming to cluding frequent childbearing and a vicious cycle of low investment in work. I remain absent for 10–15 days responsibility for the care of the house- women and low investment in girls.49 and then lose wages.’ – Tatya, 38- hold); productive work (often irregular, Gender discrimination in access to year-old hospital ward boy, India highly physically burdensome, and for health, nutrition, education and security Source: UNAIDS 2000. extremely low pay); and community exacerbates this process further. work. Gender-based discrimination oc- The transmission of poverty via nutri- Those with illnesses can also face stigma curs within each of these roles. Thus tion can begin in utero. The child of an and discrimination normally associated some women and girls are chronically inadequately nourished mother is likely with physical and mental impairments. poor even though they reside in non- to grow less rapidly than that of an ad- Stigmatised illnesses include tuberculosis, poor households, as Box 2.5 illustrates. equately nourished mother. Around 30 depression, cancer and sexually transmit- For poor women exposed to domestic, million children are born each year in ted diseases. While there have been large community or state-sponsored violence – developing countries with impaired improvements in how people with these psychological and emotional as well as growth due to poor nutrition during foe- conditions are perceived, stigma remains physical and sexual – escape from pov- tal life.50 Babies born with a low birth and is very strong in many parts of the erty is especially difficult. Women and weight are much more likely to die, and world. In the new millennium, it is atti- girls are most at risk of persistent pov- to be stunted and underweight in early tudes towards HIV and AIDS in particu- erty in contexts where gender-based dis- life, increasing their chances of ill-health lar that will stigmatise and exclude crimination is chronic, severe, and and death in childhood and beyond.51 millions. In 2003, between 34 and 46 overlapping with other forms of margin- Although evidence of the extent to million people were living with HIV/ alisation such as age, marital status or which poor foetal growth is related to AIDS, about two-thirds of whom lived in ethnicity.47 future disease is contested, it is clear that sub-Saharan Africa (Table 2.3). The neg- girls who grow up stunted or anaemic ative relationships between chronic pov- The cycle of maternal and are more likely to be underdeveloped for erty and the health and economic effects child malnutrition, morbidity childbirth, and face higher risks of ma- of chronic ill-health are discussed in de- and mortality is one of the ternal and child mortality, and of low tail in Chapter Four. most significant means birthweight and stunting among their In addition to the huge direct and indi- through which poverty own children.52 This is often com- rect costs of full-blown AIDS to liveli- persists over generations. pounded by an earlier start to childbear- hoods and economies, however, there is ing among poorer women than their also widespread social and economic dis- better off counterparts – an estimated crimination against people who have The effects of gender 17% of babies in the least developed AIDS or are HIV-positive. People with discrimination in the household countries are born to women aged 15 to AIDS, and often their households as Parental investment in children is 19 years, compared to 8% in the more well, continue to be excluded from strongly affected by localised norms of developed regions.53 Their babies are at work, access to services, and family and entitlement. In many parts of South Asia, greater risk of having a low birthweight community life. This exclusion is based China, the Middle East and North Afri- and being less healthy, leading to a cycle on misperceptions about the source and ca, girls in poor households are less of harmful long-term effects. effects of the illness, and amplified by ex- likely than boys to receive adequate gen- isting social inequalities, especially those eral care, education, nutrition and health Death and the chronically poor: of gender, sexuality, race and class. care.48 The impact of women’s educa- 100 million missing girls and The impoverishment of widows by tion and their control over resources on women asset-grabbing in-laws has worsened due children’s (and often especially girls’) Premature and preventable death due to to HIV/AIDS, as women are blamed for welfare can be much more significant poverty-related factors is the most funda- their husbands’ illness.46 than that of men. mental evidence of chronic poverty. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 22 Black plate (22,1)

22 . Chapter 2

From this perspective, females are dis- number of casualties of major epidemics based on a combination of strong son proportionately among the chronically such as the 1918–20 global influenza epi- preference and the ‘one child policy’, poor. While estimates vary, it is indisput- demic or the currently ongoing AIDS which allows rural households to have able that, based upon gender discrimina- pandemic’.55 a second child if the first is a boy.56 tion from conception to grave, millions Globally, the number of missing While in most countries gender bias has of girls and women are simply missing women increased in absolute terms over been reduced through improved female from the world. They die preventable the past decade, although it fell as a education and employment opportuni- deaths through selective abortion, infan- share of the number of women alive. Re- ties, in China (and also increasingly in ticide, overwork, ill-health, and neglect. gionally, trends vary: India) an increasing recourse to sex- Over a decade ago, Amartya Sen . Sub-Saharan Africa has a relatively selective abortions has worsened it.57 brought the issue of extreme gender bias good sex ratio, but it is deteriorating, in mortality to the attention of the and it is estimated that 5.5 million world. In the late 1980s and early 1990s, sub-Saharan African women are Chronic poverty, age and Sen estimated that over 100 million missing. the life-cycle women were ‘missing’ from the global . There have been sizable improvements The experience of chronic poverty is not population. Global trends in gender- in the sex ratios of Pakistan, Bangla- only associated with structural factors; biased mortality have recently been re- desh, and most countries of the Mid- individual and household life-cycles viewed.54 Depending on the assumptions dle East and North Africa. have a significant influence. Poverty ex- made regarding the ‘expected’ ratio of . India has seen only small overall im- perienced at a certain point in an indi- women to men in the world, estimates of provements in the sex ratio, although vidual’s life does not only affect that the total number of missing women vary there are significant differences among individual at that particular time, but from 60 to 113 million. The preferred and within states. Worsening survival can have consequences over the entire estimate is 101.3 million missing rates among infants and young girls course of that person’s life, as well as women. 80 million of these are Indian has all but cancelled out clear im- over the lives of other members of the and Chinese women – a staggering 6.7% provements in survival among older household. of the expected female population of women. China, and 7.9% of the expected female . In China the situation is deteriorating. Childhood population of India. Due to discrimina- In 1999, between 100 000 and 160 At the turn of the 21st century, an esti- tion against girls and women, from be- 000 orphans and abandoned children mated 600 million children are growing fore birth throughout the life-cycle, more were included in official figures. 90% up in absolute income poverty.58 In women are missing ‘than the combined were girls; most of the remaining 10% terms of broader indicators of child number of casualties of all famines in the were boys with a disability. Second rights, more than half the children in de- 20th century, [or . . . ] the death toll of and subsequent daughters are at high veloping countries – more than one bil- both World Wars combined, [or . . . ] the risk of abandonment or abortion, lion children – suffer from severe

Box 2.6 Childhood poverty in Kyrgyzstan – a snapshot of the future Over half of Kyrgyzstani children and youth live in households – Stunting was identified in 29% of rural children and 16% with incomes too low to satisfy their minimum needs, and of urban children. between one-fifth and one-quarter of rural children live in – Anaemia was identified in 28% of rural children and 19% households with incomes too low to buy enough food. A of urban children. greater proportion of children live in poverty than people of – 87% of adolescents living in southern Kyrgyzstan, and any other age group. 52% of adolescents in the north had symptoms of iodine As both the official and unofficial costs of accessing key deficiency, associated with mental impairment. social services rose in the early 1990s, households have been . In 2001: making less use of these services, including education and – 65% of rural infants below 1 year old and 57% of urban health care. As schooling costs bite, children drop out to save infants were living in poverty. costs and generate an income, particularly in families with – 60% of rural children below 10 years old and 54% of many children. For many poor families, children’s clothes, and urban children were living in poverty. particularly shoes, are major and sometimes prohibitive – 21% of rural children and 12% of urban children were expenses. The very cold winters render adequate clothing and living in severe poverty (about 60% of the national footwear a more pressing concern than in other parts of the poverty line). world. Some families attempt to cope by having children share Recent data analysis has shown that approximately 40% of shoes and attend school in shifts; notebooks and pens are young children (birth to four years) in Kyrgyzstan have spent also shared. Others make use of clothing donated as part of four years in income/expenditure poverty. Given the long-term humanitarian aid, although this can be considered shameful implications of persistent poverty in the early years, this is an and stigmatising, and leads to some children dropping out. extremely serious situation in terms of the future well-being of The depletion of human capital now is undermining the future the country. prospects of both individual children and society. Sources: Yarkova et al. 2004; Eversmann, 1999; Howell, 1996; . In the late 1990s: CC/WB, 1999; Falkingham and Ibragimova, forthcoming. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 23 Black plate (23,1)

Who is chronically poor? . 23 deprivation of at least one basic human need, and over one third (674 million) suffer from absolute poverty (i.e. two or more severe deprivations).59 Intuitively, children who have had a ‘good’ start in life should be at much less risk of being poor as adults, and of initiating another cycle of poverty with their own children; indeed there is much evidence to support this.60 An effective attack on chronic poverty clearly must be grounded in an attack on childhood poverty and the ways poverty is transmitted over lives and generations. ‘Childhood poverty’ and ‘chronic poverty’ both emphasise the particular time in the lifecycle when it occurs, and its duration. One rationale for distinguishing chronic and transitory poverty is that long periods in poverty may well have Older people, like this woman in Tigre, Ethiopia, have to rely on hard physical labour to survive. more damaging long-term effects than short periods.61 Similarly, the urgency of Older people addressing childhood poverty derives ‘I am an old woman and used to get which few if any can be expected to es- partly from the vulnerability of young support from my sons and daughters cape; even a relatively short spell of old- people to the impacts of poverty, both in but they have all died of AIDS leaving age poverty, if it leads to a premature childhood and throughout their lives. In me with six orphans to look after. I and preventable death, must be consid- particular, the long-term effects of poor have found it very difﬁcult to pay ered as chronic poverty. health and nutrition in the womb and in school fees, feed them, clothe and pay early life, compounded by limited educa- their medical bills. This has been wors- 355 million of the world’s tional opportunities, create signiﬁcant ened by my inability to carry on farm older people live in obstacles to escaping poverty. activities due to old age. I just pray for Many chronically poor children, are the government to offer some support developing countries where members of so-called ‘vulnerable groups’ for my grandchildren’. – Poor older formal social security is – orphans, street children, working chil- woman, southern Uganda minimal or non-existent dren, child sex workers, child heads of Source: Lwanga-Ntale and households. Many of these children, McClean 2003. and many more children not included in Conditions of poverty in older age are these categories, live within chronically 355 million of the world’s older people associated with an absence of income se- poor households. Children living in live in developing countries63 where for- curity; inadequate family or social sup- poor households without any form of mal social security is minimal or non- port; and poor health combined with social protection or access to basic so- existent. Research suggests that chronic inadequate health care. Discrimination, cial services are at high risk of staying poverty is disproportionately experi- neglect and abuse of older people occurs poor (see Box 2.6). Discrimination and enced by older people. Poverty in old age throughout the world, contributing to in- deprivation based on ethnicity or caste is both a cause and effect of intergenera- secure livelihoods as well as to feelings can also condemn marginalised children tional poverty. It is a condition from of worthlessness and powerlessness. to a lifetime in poverty or an early death. Box 2.7 Planning for the future Certain children are discriminated against within households, and can be Surya is a 50-year-old Indian male migrant worker who works at a mining site in chronically poor within households that hazardous conditions. He has developed ailments that make him feel, at 50, an old are not chronically poor. In some coun- person in every respect. With no savings, lack of job security in later life, poor tries, particularly in South Asia, girls are health and no economic support from his children, old age is a phase of life he particularly vulnerable; in others, birth does not look forward to. . . . He copes with his frail health alone, and only in an order matters. Many child domestic emergency visits the government hospital. While he does receive free medical workers live in chronic poverty even consultation, he cannot afford to take days off work, nor purchase the medicine within wealthy households. In some prescribed for his aliments. He views beggary as an alternative means of living if cases, fostered and adopted children face he is jobless, and thinks he will have to resort to this within couple of years. a similar situation, although this depends ‘In my old age with frail health who will give me a job? For the present job I had on the reasons for fostering, the resour- to coerce my employer, he agreed after I offered to work at half the salary he pays ces available, and the bonds formed be- to other employees, at least now I have some income to meet my needs.’ tween children and carers.62 Source: Shankardass 2002. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 24 Black plate (24,1)

24 . Chapter 2

Older people are particularly vulnerable old’ people are still collecting water for the poorest older people in their ef- to the ravages of war, unable to flee and and that chronically poor people in forts to achieve security. The differential not considered priority for relief efforts. this age group are more than twice as impacts of age on women and men is Such a situation is not inevitable. In likely to be collecting wood and water only beginning to be understood. Factors South Africa, significant and universal than the not-poor. This underscores indicating high risk of chronic poverty non-contributory pensions not only sup- the economic role that continues to be for older women include their greater port older women and men, but also played by older people into late old longevity and likelihood of widowhood, play an important part in maintaining age, and shows that chronic poverty inequitable inheritance laws, and low ac- families. At the same time, poor, older implies that the transition from process to education and health services. South Africans face a set of challenges ducer to consumer is not an option for Risk factors for men in some commun- that hinder the possibility of livelihood many older people.’65 ities include abrupt loss of status and improvement, including illiteracy, an low levels of support from children. ever-increasing burden of childcare, and years of apartheid policies that under- Widespread institutional and Household composition mined their capacity to save for a secure social exclusion on the basis of Poverty based on structural discrimina- old age. age, gender and disability tion can be exacerbated and entrenched Of those South Africans above the age represent formidable barriers for if it occurs at certain points in the life- of 50, the chronically poor are more the poorest older people in their cycle of a household, as well as that of than twice as likely to be illiterate; two- efforts to achieve security. an individual. Household size and de- thirds of chronically poor ‘frail old’ peo- pendency ratios, as well as headship, are ple (84+) are illiterate compared to important factors in determining the ex- about 40% of the non-poor.64 While On the other hand, around the world perience of poverty and chronic poverty. those who are chronically poor but not those older people who can still earn a And, as discussed above, changes in old are as likely to report illness as the living suffer early retrenchment due to household composition – through mar- same non-poor age group, this situation their age. Financial support to start small riage, divorce, abandonment, birth, changes in the older age groups, where business is limited because lending agen- death and migration – are also differenti- chronically poor older persons are more cies have an age limit for giving credit; ated along lines of gender, age and likely to report illness than the not poor. even some donor agencies refuse to fund health status. And even many of the very old chronic- work with older people.66 So earning Poor households tend to be signifi- ally poor continue to work, even if it is from labour, often the most critical asset cantly larger than others, even allowing not recognised as such. ‘Retirement’ is of poor people, is reduced in old age, for economies of scale within house- illusory for poor older people: pushing older people such as Surya holds, and to have higher dependency ra- ‘Not only are chronically poor people (Box 2.7) into marginal, low-paying oc- tios.67, 68 Where there is high child more likely to collect wood and water cupations that further undermine their mortality, limited publicly-provided so- than the not-poor, they are also more health. cial safety nets, and a market for child likely to continue to perform this Widespread institutional and social labour, it is a rational decision for poor work as they age. Indeed, it is alarm- exclusion on the basis of age, gender and households to have many children, but ing to note that almost 20% of ‘frail disability represent formidable barriers one that can undermine the chances that

Box 2.8 ‘You are given only once, and if you are unfortunate, that is it’ Grace was widowed shortly after she and her husband granddaughter. She feels that she has no-one else to go to escaped the violence that took place in Uganda in the mid to for help in the village, as there are no clan leaders or late 1980s. They escaped to another district with a little members of her tribe in the camp, and although her three money, but lost almost all their accumulated assets. With the surviving daughters and her son are all in the camp they sound of bullets coming closer they had to make stark rarely give her any food or other support. When she is ill it choices between saving a cooking pot or a child. is difficult for her to go to the clinic, as ‘you have to go with Soon after Grace’s husband spent the money they had to your brother’, meaning that you have to take a bribe for the buy some land, he was murdered by the land’s original doctor. She does not have anyone who will give her the owners. Grace was driven away, and settled in an internally- money. displaced people’s camp. She has now been there for over Nevertheless she is not entirely without a support network. ten years, but, twice a refugee, Grace has been able to re- A young man lent her a small patch of land during the last accumulate very little. She lives in a simple one-roomed agricultural season, on which her children helped her to thatch hut, which is her only asset. She owns no land and cultivate sweet potatoes. An old man built her a granary next ‘even the hoes I had have been stolen.’ to her house, where she planned to store the potatoes. Grace has limited support from others. Although she had Unfortunately pests destroyed the crop, leaving her no better 13 children, only five lived beyond early childhood. Of the off than she had been before. She does not expect to be surviving children, the youngest daughter died some time offered land again ‘you are given only once, and if you are ago of AIDS leaving three of her own children. Two of these unfortunate, that is it.’ children died and Grace is now bringing up the third, a Source: Bird and Shinyekwa 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P12E-02.3D.3D Folio: 25 Black plate (25,1)

Who is chronically poor? . 25 the household, and particularly the chil- mothers’ determination to give their chil- Multiple deprivation, dren, will be able to move out of dren a better future, and, in many cases, poverty.69 support from other family members. discrimination and While households headed by children Similarly, in polygynous households, disadvantage or the highly marginalised, including dis- where women are essentially responsible While there is no single, dominant char- abled people, are clearly among the for providing for their children as in acteristic of the chronically poor, such as poorest and likely to stay that way, evi- West Africa, there is no clear evidence being older or being a woman, there is a dence on the relationship between fe- that children are necessarily disadvan- set of identiﬁable elements of socio- male-household headship and chronic taged. However, where provision of economic structure and life-cycle stage poverty is far more mixed.70 This is in resources is principally a male responsi- that underpins long-term poverty. Pro- part because of the range of household bility, and in cases where women are cesses of exclusion and exploitation structures that ‘female-headed’ can truly unsupported or stretched, unequal based on ethnicity, religion, caste, class, imply. Where social inequalities and dis- distribution of resources can result in displacement, old and new forms of slav- crimination reduce single mothers’ access women and children living in poverty.71 ery, disability, ill-health, gender, age and to resources, they and their children may Widow- and divorced woman-headed household headship, keep many millions be worse off, but this is sometimes medi- households are disproportionately among in poverty by limiting access to assets, ated by a range of factors, including the poorest (see Box 2.8). services and positive social relationships.

Notes 1. While participatory poverty assessments suggest 15. de Haan and Rogaly 2002. same time, it must be noted that some women that such categories often relate closely to those 16. Wu 2001. use the possibility of witchcraft to their used by poor people, they may not always be 17. Hossain, Khan and Seeley 2003. advantage. In the context of changing gendered part of the mental constructs that individual 18. Kothari 2003. agrarian relations in Kenya, witchcraft ‘remains poor people use to understand their world. The 19. Daru and Churchill 2003. a powerful weapon through which women can social advantages enjoyed by most poverty 20. Harriss-White 2003:24. level intra-household disparities and, more analysts – class, wealth, education, race, gender 21. ILO 2001:74. broadly, challenge the legitimacy of social – make it possible to introduce analytical 22. Anti-Slavery International 2002. practice’ (Dolan 2002:678). In other cases, frameworks that enhance our ability to 23. US Centre for the Study of Intelligence, in Kaye women – particularly those from tribal understand and compare experiences across 2001. communities – have particular marketable skills time and space. 24. Kaye 2001. as soothsayers and healers (Lalita 2003) – in 2. It is also important to recognise that these 25. Genicot 2000. some cases one of the only businesses that ‘categories’ of chronically poor people are only 26. Herzfeld 2002. expands in a declining economy. ‘groups’ inasmuch as they share common 27. Daru and Churchill 2003. 48. In sub-Saharan Africa and the Americas, there experiences of poverty based on common 28. Venkateshwarlu and daCorta 2001. is less clear evidence of intrahousehold gender characteristics. An individual or household that 29. HAI 2000; SCF 2003. discrimination, and in some cases it is boys who an outsider would place within a speciﬁc 30. Global IDP Project 2002. are in the weaker position. . category may not identify themselves as such. 31. UN SCN 2003. 49. Harper, Marcus and Moore 2003. Social identities are multiple and overlapping, 32. Global IDP Project 2002. 50. James Commission 2000. and there is often little sense of social solidarity 33. FAO 2003. 51. ACC/SCN 2000; James Commission 2000; between those who are similar in some ways 34. UNHCR 2003a. Kielman and McCord 1988 in Tudawe 2001b. but not others. A teenage girl with learning 35. UNHCR 2003b. 52. ACC/SCN 2000. disabilities whose father is a bonded labourer 36. Global IDP Project 2002. 53. UN Population Division 2004. from a minority ethnic group, and the 37. Pettersson in UNHCR 2002. 54. Klasen and Wink 2003. matriarch of a family inﬂuential in politics and 38. UNHCR 2002. 55. Klasen and Wink 2003:264. business experiencing impaired mobility based 39. Hoogeveen 2003. This estimate is based on 56. Young 2000c in LeBrun 2003. on age, are unlikely to feel a bond based on recent innovative work combining 1991 57. Klasen and Wink 2003. shared gender or disability status. Population and Housing Census data and 1992 58. UNICEF 2000. 3. See Begum and Sen 2003. Integrated Household Survey data. 59. Gordon et al 2003. 4. LeBrun 2003. 40. Tudawe 2001a. 60. Harper, Marcus and Moore 2003. 5. World Bank 2001. 41. World Bank 2001. 61. Chase-Lansdale & Brooks-Gunn, 1995. 6. Baulch et al. 2002. 42. Erb and Harriss-White 2001; Masset and White 62. Harper, Marcus and Moore 2003. 7. Strikingly, even if minority households had the 2003. 63. HAI 2001. same levels of human and physical capital as 43. Patel et al. 1999, cited in WHO 2001. 64. May 2003. majority households, only about one-third of 44. Imrie 1996 in Yeo and Moore 2003. 65. May 2003:22-3. the expenditure gap would be closed (Baulch et 45. Yeo and Moore 2003. 66. HAI 2001. al, 2002): social assets are crucial to well-being. 46. Human Rights Watch 2003. 67. Children, older people, the ill and disabled are 8. Brockerhoff and Hewett 2000. 47. Human rights abuses against those accused of generally considered as dependents, although 9. Central African Republic, Coˆ te d’Ivoire, Ghana, being ‘witches’ are common in many cultures, each often contributes both directly and Kenya, Mali, Namibia, Niger, Rwanda, and women – especially older women – are indirectly to household income. Senegal, Uganda, and Zambia. much more susceptible than men to charges of 68. de Haan and Lipton 1998. 10. Braithwaite et al. 2002. witchcraft. ‘Witches’ are often the scapegoat if 69. Kabeer 2000. 11. Revenga et al. 2002. something goes wrong in a community. 70. Chant 2000, Quisumbing et al 2001, Appleton 12. Sankaran 2000. Subsequent exclusion of, and even violence 1996. 13. NCAER/Oxford 2001. towards, an accused woman can undermine her 71. Harper, Marcus and Moore 2003, Masset and 14. Kothari 2003. livelihood, her well-being, and her life. At the White 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 26 Black plate (26,1)

classic cases being the Kalahandi- Koraput region of southern Orissa, the southern and western areas of Madhya Pradesh and adjacent regions in Where do Maharashtra. In Vietnam, chronic poverty is found chronically poor disproportionately in the Northern Up- people live? lands region,3 and in the Philippines it is 3 strongest in the provinces experiencing protracted conflict within Mindanao.4 There are areas within the poorest parts A place of multiple deprivations of south-west China where household Bitare village is located in a steep, hilly region of south-west Uganda, consumption is falling while identical near the conflict-prone DRC border. It is also close to the Bwindi ‘impene- households in better off areas enjoy rising trable’ forest, a National Park that cuts the community off from relatives consumption, indicating how neighbour- and employment opportunities in neighbouring districts. hood endowments of physical and human It is 28 km – nearly 2 hours in a 4-wheel drive vehicle – from Kisoro capital influence the productivity of a town, where there are hospitals and other services. A connecting road for household’s own capital.5 In Nicaragua, the village was built in 2001, but the last kilometre is still a narrow foot- chronically, extreme poor ‘agricultural path. The sub-county headquarters (16 km away) has a health unit but households’ are disproportionately repre- no secondary school. There are primary schools, but the quality is very sented in the population of the Central poor, particularly as there are no teachers’ houses to attract good quality Region, even more so than the moder- teachers to the area. Electricity does not extend beyond the periphery of ately poor.6 Kisoro town. All of these are examples of spatial The high concentration of extreme poverty in Bitare (54 out of 121 poverty traps – geographical areas which households are very poor) constrains local economic growth, by inhibiting remain disadvantaged, and whose people demand for goods and services, and producing little other than a few remain multi-dimensionally deprived and forest products and eggs to sell. poor over long periods of time. The remoteness and the rugged terrain drive and maintain chronic poverty. The terrain has encouraged soil erosion and reduced agricultural productivity; made access and to schools, health services,1 markets and Global dimensions of information difficult; provided camouflage for rebel activity; and increased chronic poverty construction costs. Remoteness has also reduced labour opportunities. Figure 3.1 illustrates two main features Shocks – weather, crop failure,2 animal diseases, landslides, rebel skir- of the international distribution of mishes and the absence of adequate public responses to this remote chronic poverty. First, the countries with area – have also contributed to keeping people poor, and pushing others the highest levels of chronic poverty (des- into poverty. ignated by the darkest shading) accord- Source: Ssewaye, 2003 ing to CPRC’s cluster analysis [see Chapter One] are found solely in sub- Chronically poor people live in all re- regions have more financial resources Saharan Africa. Second, the countries gions, but are concentrated in certain available to overcome poverty, specific with the greatest numbers of chronically places. This is true globally and within groups and regions have benefited little poor people (designated by country size) countries. The problem is deepest in sub- from growth. In transitional countries, are found in South and East Asia.7 Box Saharan Africa where a high proportion rising levels of poverty and chronic pov- 3.1 summarises the different regional ex- of people are poor and remain poor over erty reflect the insecurities of newly periences of chronic poverty; further de- long periods of time. In South Asia, emerging market economies, combined tails can be found in Part B. where economic growth rates have been with a loss of state protection for some – improving, poverty rates are lower but notably single older people. large numbers of people remain in Within countries, chronic poverty is Spatial poverty traps chronic poverty, reflecting the popula- usually unevenly distributed. Most na- Chronic poverty tends to be concen- tion size. In some areas of South Asia tional household survey data shows a trated in regional poverty traps rather such as Sri Lanka’s ‘wet zone’, chronic significant regional dimension to the in- than evenly spread across a country. poverty is becoming a localised problem, cidence of poverty, with greater propor- Looking at multidimensional deprivation but in India and Bangladesh chronic tions of poor households in remote, less- across regions reveals increases in pov- poverty remains nationally significant, favoured or conflict-affected areas. In erty and destitution in certain areas even increasingly concentrated within persis- Uganda, persistent income poverty is in countries where overall income pov- tently poor states and districts. concentrated in the east and north. In erty is otherwise declining. The famine- In other regions, such as Latin Ameri- India, chronic poverty is concentrated in prone region of Wollo in Ethiopia is an ca, and East and South-East Asia, the centre and east of the country, but extreme example of this, where hundreds chronic poverty has declined in recent within that, extreme poverty is generally of thousands of people need food aid decades. Although countries in these concentrated in remote regions, the every year.8 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 27 Black plate (27,1)

Where do chronically poor people live? . 27 China Philippines Indonesia Source: refer to technical annex on page 35. Vietnam KH Myanmar RU Bangladesh NP India MG SO KE MZ TZ Pakistan UG MW BI RW IQ Afghanistan Ethiopia ZW ZM Sudan ZA CF TD CM DRC AO EU CG Nigeria NE ML BJ TG BF GH CI LR GN SN SL Code Country NameAOBF Angola FasoBI Burkina BJ Code Burundi Country NameCF Benin Code CountryName CGRepublic African Central Code HT CountryName Congo the of CI Republic EU HaitiCM IQ Europe d'Ivoire Cote ECCO Cameroon Iraq Ecuador Colombia GH ML MZ Ghana Mali MG GN Code Mozambique Country Name KE Madagascar Guinea TD NE Kenya RW KH MW Niger Chad Rwanda Malawi Cambodia SL LR MX Leone Sierra Mexico Liberia NP PE TG Nepal SN Peru Togo Senegal SO RU Somalia Russia TZ VE ZA UG Tanzania of Republic United Venezuela Africa Uganda South US United States ZM Zambia ZW Zimbabwe Brazil VE PE CO EC HT MX US Desperately Poor Very Poor Relatively Non-Poor Data Insufficient OECD Figure 3.1 The proportion and size of chronically poor populations in world regions Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 28 Black plate (28,1)

28 . Chapter 3

Regions likely to have concentrations research, approximately 1.8 billion peo- of the multiple deprivations associated of chronic poverty can be characterised ple, most of them poor, live in less-fav- with spatial poverty traps: as: oured or low potential areas – deﬁned . Poor agro-ecology: soil quality, slope, . Remote: areas that are far from the here as areas ‘less favoured by nature or rainfall quality and distribution, tem- centres of economic and political ac- by people’. This is about two-thirds of perature, vulnerability to natural tivity. ‘Far’ is calculated in terms of the rural population of developing coun- hazards. not only distance, but also time taken tries.11 About 40% of the rural poor live . Poor infrastructure: poor road, rail, to get there. These areas are deﬁned in in highlands and drylands,12 many of river connections, leading to high terms of distance multiplied by time: which are also low potential and transport costs. ‘frictional distance’.9 remote.13 . Weak institutions/organisations: espe- . Low potential: areas that have low While terms such as ‘remote’ and ‘low cially weak market institutions, lead- agricultural or natural resources, often potential’ indicate the characteristics of a ing to high transactions costs. crudely equated with drylands and spatial poverty trap, ‘marginal’, ‘less fav- . Political isolation: especially associ- highlands. oured’ and ‘weakly integrated’ begin to ated with weak political parties and . Less favoured: politically disadvan- suggest explanations for why such an networks, weak claims on local and taged areas.10 area – or rather a substantial proportion central government services. . Weakly integrated: areas that are not of its population – remains poor over Often these characteristics overlay and well-connected, both physically and in long periods. reinforce each other. The case of Bitare terms of communication and markets. Although the nature and incidence of village in Uganda provides an example Large numbers of very poor people live chronic poverty is different in different of how different forms of deprivation in such regions. According to recent areas, there are common characteristics combine to create a poverty trap

Box 3.1 Chronic Poverty by region and western India, and in Bangladesh. Most human Chronic Poverty in Sub-Saharan Africa development indicators have also improved over the past two Sub-Saharan Africa is both the poorest and most chronically decades, although in Afghanistan years of war have poor part of the world. Between 28% and 38% of the absolute obstructed almost all potential progress. However, South Asia poor population in sub-Saharan Africa is estimated to be has very high population levels and still has the largest chronically poor, totalling between 90 and 120 million people. number of chronically poor people in the world – an estimated There are particularly high levels of absolute poverty in West 135 to 190 million people – including 110–160 million Indians, and Central Africa, where one in every 5 or 6 people is both 9–13 million Bangladeshis, 10–15 million Pakistanis, perhaps chronically and severely poor. 5 million Afghans, and 2–3 million Nepalese. Because of such There are 22 sub-Saharan countries for which there is both high numbers of people living in persistent poverty, even small a US$1/day figure and sufficient data to undertake the cluster areas of severe deprivation in this region can affect a very analysis detailed in Chapter One. Of these 22 countries, large number of individuals. approximately 310 million people live in 12 desperately Chronic poverty in the region is most pronounced in areas deprived countries, about 150 million of whom live on less that have significant minority populations [see Chapter Two], than US$1/day; perhaps 45–60 million of these people are that are economically stagnant, where agrarian class chronically poor. Almost 110 million people live in the 10 structures and gender relations are exploitative, and where moderately deprived countries for which we have data, close governance is weak. At a regional level, most indicators show to 40 million of whom are absolutely poor, and 10–15 million a swathe of poverty cutting across eastern and southern of whom are chronically poor. Several countries with Pakistan, central India, western Nepal, and northern and insufficient data – including Angola, Guinea, Liberia, Mali, south-eastern Bangladesh. Over 70% of India’s poor for Somalia, Sudan and Zambia – make up the ranks of those example live in six states (out of 28). There are pockets of countries with high proportions and numbers of absolute and improvement, lower levels of poverty and even relative chronically poor. prosperity in this region, however there are also large areas Chronic poverty in sub-Saharan Africa is most pronounced where deprivation is the norm. in both urban and rural areas of remote, less favoured, and Most poor South Asians still live in rural areas, and it is weakly integrated regions. There are many of these, likely that the proportion of chronic poor is greater in rural particularly in areas affected by violent conflict, suffering areas, given the greater economic opportunities in towns and economic stagnation or decline, and where HIV/AIDS and cities. However, in India the proportions of severely poor other diseases are endemic. The majority of chronically poor people in rural and urban areas are similar at about 15%. Africans live in countries with large numbers of poor people, a Chronic poverty also tends to follow the ‘contours of history of low economic growth, and problems of governance. conflict’.14 Notwithstanding the current peace processes, the There is a strong possibility that they will remain poor. longterm poverty found in northern and eastern Sri Lanka and mid-west Nepal is likely to be relatively intractable. The Chronic poverty in South Asia relationship between violent insurgency and the isolation of The headcount ratio for the chronically poor has been regions is further examined below. declining in many parts of this region – particularly in southern Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 29 Black plate (29,1)

Where do chronically poor people live? . 29

(Introduction to this chapter). Spatial poverty traps do not only occur in rural areas. In many countries there are urban and peri-urban poverty traps – Kibera in Nairobi, East Delhi, Soweto outside of Johannesburg are examples. Both rural and urban poverty traps are underpinned by a combination of location-specific factors and by the relationships they have (flows of people, labour, finance, resources) with other areas. A conceptual framework for ana- lysing spatial poverty traps is outlined in Table 3.1. This identifies the social, economic, political and environmental factors that promote uneven development and lead to some areas having populations with high incidences of chronic poverty. Marginal urban environments, like this railway track at Senen, Indonesia, are often home to chronically poor people.

Chronic poverty in Latin America and the poverty alongside rapid economic and political liberalisation. It Caribbean is estimated that between 10% and 20% of the extreme poor Within the developing world, Latin America and the Caribbean population in transitional countries is now chronically poor: have some of the best average human development between two and five million people. This number is rising indicators, alongside impressive economic growth. However, quickly, due to childhood poverty [See Chapter Two], long term there is significant variety within the region, with some unemployment without social protection, and older people with countries, such as Haiti and Bolivia, faring badly. Extreme pensions that do not meet even their basic needs. poverty in the Latin America and the Caribbean region is Spatially, chronic poverty is most evident in remote rural relatively low compared to other developing regions but the areas, and ‘one-company towns’ where many households proportion of the poor who are chronically poor remains were previously reliant on employment within a single sector. relatively high. It is estimated that between 30% and 40% of Former employees of these now unprofitable enterprises have the extreme poor population in Latin America and the fallen prey to market forces and redundancy. Chronic poverty Caribbean is chronically poor: between 16 and 22 million is also most evident where there are large minority people. populations, notably the Roma [see Chapter Two]. Persistent poverty in Latin America and the Caribbean is largely a distributive problem; inequality levels are among the Chronic poverty in China highest in the world and undermine the potentially positive Much of the global poverty reduction of the last 30 years is impacts of growth on the poor, as well as hindering growth down to progress in China. However, much depends on the itself. Race and ethnicity are important dimensions of the statistics and measures that are used. According to the most region’s geographic concentrations of persistent poverty, and recent estimates, the number of poor with expenditures below evidence suggests that access (or rather the lack of access) US$1 in 1999 was 235 million, more than twice that under the to social services has a particularly powerful role in income poverty line.17 It is also clear that the benefits of determining and shaping these patterns. recent, rapid economic growth have been highly unequal, both Urban poverty is particularly significant in Latin America and socially and spatially. the Caribbean – 64% of the poor and 75% of the total Available evidence indicates that chronic poverty is highest population of Latin America live in urban areas.15 The in rural areas and that it is particularly concentrated in north- probability of being poor or extremely poor is still much higher west, west and south-west areas, away from the dynamic in the rural areas – 37% if you live in a town or city, but 63% if coastal region. These areas are remote from growth centres, you live in a rural area.16 Several primarily rural regions stand are agro-ecologically poor,18 and have high ethnic minority out as persistently poor, such as the pan-Andean region, populations. Any industry was largely state-owned and likely including parts of Bolivia, Peru, Ecuador and Colombia. to have been closed during the economic restructuring. Rising economic and social inequality means that Chronic Poverty in transitional countries substantial numbers of chronically poor people also live in Although relatively low, chronic poverty is growing fastest in the rural areas that are seen as prospering.19 Significant numbers ‘transitional’ states of Central Asia, the Balkans, East-Central of ‘new poor’ have moved to cities and are unable to meet Europe and the former Soviet Union. Over the last 10 to 15 their minimum needs because of low paid casual work, and years, the region has seen rising inequality and long-term no access to state provided services. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 30 Black plate (30,1)

30 . Chapter 3 Politically marginal, unstable, liable to political fragmentation and conflict. Poor representation in political assemblies. Lack of protection against abuseofficials, by lack of institutions ablesafeguard to and further citizen rights,safety no net. Political characteristics not usually considered but natural disadvantage may affect societal perceptions of people from such areas leadingstigma, to discrimination and inequality. Illegal land holding increases vulnerability. Excluded. Relatively small (often fragmented) constituencies. Political access more constrained because less competitive. Voices rarely heard, especially if alsoor ethnic religious minority. Political isolation Adversely incorporated into markets through exploitative or uncompetitive economic relationships: markets are fragmented and function weakly. Limited market access, low population density, ‘residual’ populations left behind, old, very young, disabled, ill, discriminated. Poor economic and social infrastructure, ‘over-population’, low human and financial capital. Out- migration or commuting with positive and negative consequences depending on migrants’ endowments. Includes poor areas within growth centres. Low economic diversity and lackgrowth. of Dependence on agriculture or natural resources, which arereturn low and lowest wage sectors. Little wage labour available: out- migration or commuting ‘solutions’, but usually into low skill/returninsecure and occupations. Few accumulation or expansion possibilities due to low demand.opportunities Few to augment skills, save, get credit. High risk forSocial investments. capital may be high,excludes but the often poor or notsecuring useful access for to other resources. markets) migrate; limited information on opportunities and rights. illegal residents and enterprises. Low public investment in social protection and basic services leading tocash low circulation. Risk of fallinglabour out market of due to injury or death. Multiple costs to meet basic(shelter, needs water, transport, health, education) in settlements that are often unsafe and insecure. Lowcirculation cash as a result ofproductivity. low Dependence on remittances, public subsidy. High infrastructure costs lead toquality poor or absent provision. Poor road, rail, river connections leadhigh to transport costs. No clear patterns. Poor opportunities to commute or No clear patterns. Lack of services for informal and High ecosystem diversity, fragile or degraded land resources, climatic variation. Bio-physical constraints limited rains, poor soils, steep slopes. Vulnerable to hazards, displacement. Geographically isolated, may have low or high population densitiesdifferent with implications for resource exploitation. Geographical obstacles, such as slopes, ravines and marshes, contribute to isolation. Can include high and lowagrarian potential environments, poorly serviced and connected peri-urban and urban areas. Can include high and lowenvironments potential and pockets. Lower levels of infrastructure and services, stigmatised, ‘hardship posting’. Private sector avoids investment; savings invested outside the area. Poor locations for built orenvironment: productive hillsides, roadsides, canalsides, riversides, dumps. Limited possibilities for technical change in natural resource based production systems. Definition Ecological characteristics Poor infrastructure Weak institutions (including Can include high and lowenvironments. potential Costs of centrally supplied infrastructure and services are higher. Generally lower potential for non- farm activity, though remoteness offers some protection from competition. Poor urban residential areas remote from workplaces, with weak connections. Weakly integrated regions (poorly linked and economically disadvantaged) Less favoured areas (politically disadvantaged) Low potential or marginal areas (ecologically disadvantaged) Spatial poverty trap description Remote regions and areas (frictional distance and locational disadvantage) Table 3.1 A conceptual framework for understanding spatial poverty traps Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 31 Black plate (31,1)

Where do chronically poor people live? . 31 Ecological characteristics of spatial poverty traps Box 3.2 Risk and vulnerability among Bangladeshi char The ecological characteristics of a loca- communities tion can directly increase the vulnerabil- The riverine and coastal areas of Bangladesh, known as chars, are home to the ity of those living there. Over half of the poorest and most vulnerable communities in the country. 5% of the total population world’s rural poor live in areas of low live in these char regions, over 80% of whom are in extreme poverty. agricultural potential.20 Half a billion The chars are exposed to a range of environmental shocks and stresses people in developing countries live in (particularly erosion and flooding) associated with the instability of the local arid regions with no access to irrigation environment. Villages and agricultural lands are swept away during floods and the systems; another 400 million are on rivers frequently change their course. This instability means that only temporary lands with soil unsuitable for agriculture; settlements and infrastructure are built. Large scale flood/erosion protection 200 million live in mountainous and infrastructure has not worked in these regions. hilly regions; and more than 130 million As a result of their isolation and peripatetic nature, communities are excluded are in fragile forest ecosystems.21 These from mainland services and infrastructure and neglected by local government regions are also highly vulnerable to cli- officials. Ill-health is common due to lack of access to clean water and sanitation. mate fluctuations, pests, diseases, and The lack of social services perpetuates high morbidity and poor educational status, man-made and natural disasters, which particularly for women and children. Poor members of communities frequently lose make food supplies precarious and control of their re-emerging land to wealthier and more powerful neighbours (who unstable.22 employ thugs). They have little access to justice. Heavy dependence on daily In Bangladesh, panel data for 1987–8 labour, often through migration for part or all of the year, shows the constraints and 2000 indicate that 15% of house- which char dwellers face in obtaining income through farm or non-farm enterprises, holds that had descended into poverty and interacting effectively with markets. had experienced a shock related to a nat- Source: DFID, 2002. ural disaster, suggesting that poor geographic capital at the most local level Urban location can be a crucial asset Tribal Areas in the west, and large areas played a role. Poverty rates are highest for the poor, but as urban populations of Balochistan, North West Frontier in extremely low-lying areas that are fre- grow and land becomes increasingly Province and Sindh – areas which are quently flooded, including chars (river- scarce, large and increasing numbers of also dominated by oppressive tribal and/ islands that seasonally disappear – Box people live and work in high risk, low or feudal agrarian and gender relations. 3.2), and in tribal areas where social and potential or marginal urban environ- In India, populations living in forested geographical disadvantage overlap.23 ments, such as roadsides, rail embank- areas, such as south-western Madhya Rural development policies have ments, steep slopes and rubbish dumps. Pradesh, are disproportionately margin- tended to neglect low potential areas in Alongside poor sanitation and hygiene, alised tribal groups.29 favour of investment in high potential these environments present risks of dis- zones,24 where returns to investment are ease, accidents and infections, a substan- perceived to be higher.25 In Africa, these tial part of the experience of urban Poor infrastructure neglected areas include arid and semi- chronic poverty. Insufficient physical infrastructure re- arid regions, deep forests, and mountain- stricts local access to markets, and main- ous areas. In India, less-favoured areas The chronically poor are tains spatial, political, and social include the drylands (characterised by more likely to be excluded marginality. This includes roads, mar- frequent crop failure and sporadic op- kets, electricity, water and sanitation, ir- from opportunities to portunities for employment) and forested rigation and telecommunications. A areas (especially in hilly regions with a migrate, reproducing regional comparative study within China indi- predominance of tribal populations, and inequality and exclusion cates a highly significant positive rela- with limited access to natural resources, tionship between higher road densities information and markets).26 In China, and consumption growth in an area.30 living in a village in a mountainous area Evidence suggests that chronic poverty Remote rural areas experience deficits has a sizable and significant negative ef- is most harsh where ecological and social in all forms of physical infrastructure. In fect on consumption growth. deprivation overlap. In Latin America, it many less-favoured rural areas, sparse In semi-arid areas of rural Andhra Pra- is clear that living on a hillside or road- population densities increase the cost of desh and Maharashtra, one study found side is detrimental to lives and liveli- service provision, and the extension of that over one-fifth of the population was hoods in the urban favelas. However, it physical infrastructure, compared to poor in all nine years between 1975–76 is the discrimination faced by these resi- more densely populated, often urban, and 1983–84, while 60% were poor in dents, combined with the fear of rising areas, where political pressure for invest- at least five of the nine years.27 Further gun and drug crime, that renders a very ment may be stronger. It is also more analysis of this dataset suggested that real sense of physical and psychological difficult and costly to monitor service even relatively affluent households are vulnerability and isolation among the provision and quality in these areas, and highly vulnerable to long spells of pov- poor, despite any improvements in phys- governments’ response to citizens’ rights erty when severe crop shocks occur.28 ical infrastructure [see Chapter Eight]. In and demands can be weak.31 The importance of ecological condi- Pakistan, the mountainous environment This lack of physical infrastructure tions is also evident in urban settings. is harsh in the Federally-Administered undermines the maintenance and Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 32 Black plate (32,1)

32 . Chapter 3

Nowhere to sell, no-one to hire Imperfections in goods and factor markets, including land, labour, rental, finance and insurance markets, and information flows are common in rural areas.35 These undermine attempts to promote agricultural and other development, which is further aggravated by a lack of support services. For this reason, movements in and out of poverty can be strongly regionally-differentiated. In Uganda, for example, upward mobility has been associated with the more rapidly growing coffee-producing areas in the 1990s – followed by sharp downward mobility when prices collapsed (see Box 4.6). In Zimbabwe, the most dynamic of three semi-arid districts Villagers are forced to move by flooding in Somalia – cholera is endemic in the wet season so some was found to be near the city of Bula- members of the family have to be carried. wayo, benefiting from inward investment and employment opportunities promotion of human capital. Distance characterised by problems associated for migrants.36 and time increase the opportunity cost of with the inadequate provision of serv- The link between high levels of re- accessing education and health care, and ices, infrastructure and transport, and moteness, low levels of public and pri- market imperfections resulting from typified by illegal squatting and informal vate investment and high incidence of transport deficiencies prevent access to sub-division of agricultural land. The chronic poverty is clear.37 In remote re- medication, while poor water and sanita- trend of urbanisation in Latin America gions, market imperfections limit agri- tion increase health risks.32 Preventable from the 1950s, for example, has led to cultural productivity and households’ deaths become more likely, and human a growing number of mega-cities ringed ability to accumulate capital and other capital is depleted by ill-health and with illegal land occupations (favelas assets. In turn, the capacity of poor impairment. and villas), and large numbers of urban households to hire farm labour when Remote rural areas tend to be by- poor. Unreliable and costly transport household labour is insufficient (in terms passed by flows of people, ideas, services links to the city centre mean that, despite of numbers, strength or knowledge)38 is and goods. This increases the isolation of a lack of opportunities in the peripheral constrained, and their ability to rent oc- certain areas and people and deepens the areas, many people work locally with casional-use equipment, such as oxen for ‘frictional distance’ for migration.33 Mi- very low remuneration. Inner city areas, ploughing, is undermined. This encour- gration is often an expression of spatial in contrast, tend to be characterised by ages inefficient overstocking of tools and patterns of opportunity, and people tend overcrowding, high levels of competition equipment of production,39 thereby in- to move in expectation of improved live- for work and resources, and commodifi- creasing vulnerability to risks.40 lihood chances, to areas that offer im- cation of land and services. The potential for livelihood diversifi- proved connectivity and services, better cation tends to be significantly higher in markets and higher potential.34 How- The link between high levels non-remote areas, as is the case in semi- 41 ever, the poorest are more likely to be of remoteness, low levels of arid Zimbabwe. While the chronically excluded from opportunities to migrate, poor are least likely to benefit directly public and private investment partly reflecting transportation con- from such economic diversification, in- straints, and so reproducing regional in- and high incidence of chronic creased levels of local economic activity equality and exclusion. poverty is clear. raise the likelihood of increased demand In urban settings, service connections for casual labour and may make it more are often illegal and therefore may be in- likely that non-poor relatives and neigh- secure and often disrupted by the author- Weak institutions and bours can provide support to the ities. Households that do not have any organisations chronically poor and those likely to fall access at all are more likely to fall into The problems experienced by concentra- into long-term poverty. and remain in poverty. In many countries, tions of very poor people can be similar Increasing agricultural wage rates can rising urban population trends have been to those of extremely sparsely populated be a crucial factor in poverty reduction; accompanied by declines in the quality areas. Economically, market opportuni- in much of South Asia, this is probably and availability of services, although the ties can be severely limited and fragile: the best single explanation for the slow experience is different in different types there are few generators of employment, but steady reduction in the depth of con- of urban settlement. Fire is a particular less demand for local produce, and few sumption poverty. However, getting hazard that drives many slum-dwellers opportunities to save and acquire the as- work does not always translate into exit- into chronic poverty [see Chapter Four]. sets and human capital that can be the ing poverty. In agrarian economies with Peripheral urban areas tend to be springboard for economic growth. large casual labour markets, the number Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 33 Black plate (33,1)

Where do chronically poor people live? . 33 of days of work obtained in a given pe- advantage of alternative economic op- Politics, rights and riod can be as important as the wage portunities within the changing labour level itself. Discrimination and inequal- market due to their lack of skills, conflict in spatial poverty ity within the labour market is a charac- contacts, capital or freedom of traps teristic of chronic poverty, even in movement.44 Politically, concentrations of very poor growth centres. Gender divisions within people are often characterised by a less labour markets restrict employment op- Concentrations of very poor organised civil society, less responsive portunities for women, though the de- people are often characterised government, and even a less visible mand on women to work is strong NGO presence.47 Pockets of chronic by less organised civil society, within poor and chronically poor poverty therefore exist where socio-polit- households. less responsive government ical exclusion – often on the basis of lan- Labour market position is crucial and less visible NGO presence. guage, identity, or gender – shapes the within more urbanised regions charac- prospects of a significant proportion of terised by high wage dependencies the population. coupled with low and falling real Although there may be more opportu- Political marginality extends to specif- wages, as well as high unemployment, nity for employment in the inner city, ic groups within regions. Areas where casualisation and under-employment. and poverty may be more responsive to prospects have been undermined by un- These in turn are related to patterns of growth there, economic conditions vary even development, widening inequality growth or economic stagnation. The widely, living costs are likely to be high- and lack of access to essential services, question is whether all of the urban er and there are fewer possibilities for and where people have common ethnic, poor can respond to opportunities. In subsistence farming. As population den- religious, or linguistic characteristics, Madagascar, urban poverty has re- sities increase in towns, greater pressure can produce populations who share a mained at high levels, despite rapid is felt by residents to secure sources of strong sense of grievance associated with growth in average household incomes in income, access to which becomes more their identity. In Mindanao in the Philip- the capital city (by 50% between 1995 and more competitive. pines, for example, feelings of political and 2000). 65% of the urban popula- In inner city areas, if the informal sec- disenfranchisement have risen through tion remained poor between 1997 and tor is unable to absorb an ever-expand- years of government bias against the re- 1999. Few had access to electricity and ing urban labour force, trading on the gion in policy and public expenditure.48 running water. Most chronically poor black market and other illegal activities Where marginality provides the con- households lived in poor neighbour- can provide viable alternative liveli- tent for social and political movements, hoods, and their jobs were largely low hoods.45 Although many aspects of these are likely to have a potentially ex- quality and in the informal, low wage urban life are illegal for the poor, in- plosive ethnic, regional or linguistic ele- sector.42 cluding their rights to homes, settlement. Areas experiencing long-term Chronic urban poverty is high among ments and workplaces, this may not be conflict, resulting in widespread damage those who are unable to participate or a major issue when things like illegal to the resource base and people’s capa- are excluded from participating fully in connections to services are so common bilities, will almost certainly experience the labour market. These people can in- that they are generally ignored by the persistent poverty.49 Remote areas may clude sick or older people, children and authorities. In other cases, though, it be the site for protracted insecurity, people with disabilities, as well as other can increase vulnerability. The stigma linked not only to regional politics and ‘dependent’ groups and areas incorpo- associated with living in areas associ- the proximity of national (or internal) rated into markets on unfavourable or ated with high criminality reduces resi- boundaries, but also to the waning pres- exploitative terms. dents’ chances of accessing employment, ence of national security bodies. Under-employment is a significant markets and other city-wide institutions. Many conflicts are fought in border re- problem of the Ethiopian urban Prejudicial attitudes may play a role in gions, which have historical legacies of chronically poor. Over one-quarter of creating criminals: residents may be ar- marginality, limited voice and persistent chronically poor household-heads work rested on suspicion and criminalised be- poverty. Civil wars largely occur in as casual labourers or in ‘women’s busi- cause they cannot defend themselves; countries that have low human and eco- ness activities’, compared to only about and expectations within an area may nomic development. Borderlands and 8% among the never poor.43 These themselves play a role in determining other regions typified by weak state pres- occupations are insecure and give low life choices. ence may provide fertile ground for returns, so it is not surprising that While this section has focused on mobilisation of militant groups and or- the chronically poor are disproportion- weak markets, other institutional failures ganised criminality. The powerlessness ately represented. maintain spatial poverty traps. Public of the poorest makes them unlikely driv- Increased occupation of rural land ad- services are usually further away from ers of conflict, but most likely to feel its versely draws poor rural residents – in- low income settlements and disadvan- devastating effects in full. cluding tenant farmers, sharecroppers, taged rural areas, and are of poor qual- Conflict creates refugees and internally those who rely on common property re- ity. Evidence from Bangladesh,46 where displaced people – now in their millions sources – into the urban transition. there are several high-performing NGOs, in developing countries, especially Africa Without the necessary political connec- indicates that NGOs systematically – pushed into countries and places with tions to protect their interests, they are avoid operating in the areas where few resources to welcome assetless in- left in a weak position, unable to take chronic poverty is most common. comers. Internally displaced people Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 34 Black plate (34,1)

34 . Chapter 3

disaggregate by ethnicity. Regional ag- Box 3.3 The Batwa people of the Great Lakes Region gregations of numbers, therefore, obscure In the Great Lakes region little is known or reported about one of the poorest and the condition of particular indigenous most vulnerable communities, the Batwa Pygmies. The Batwa are traditionally forest peoples, and tend to underestimate num- hunter-gatherers. They have been steadily dispossessed and evicted from their bers. Where the figures are small, they traditional forest lands, as forests have been claimed by other interests, such as may go unnoticed. Where they are small game reserves in south-western Uganda. In Rwanda, legislation in the 1970s but noticed, they can be ignored by outlawed hunting as a way of life and by the 1990s those Batwa still practising policy-makers as politically insignificant, clandestine hunting and gathering were forced to the edges of their ancestral forests as with the case of the Batwa people to make way for national parks and military training areas. In the Kahuzi-Biega (Box 3.3). However, in some settings, in- National Park of the Democratic Republic of Congo, Batwa people risk violent digenous groups become drawn into con- repercussions and imprisonment for any attempt to gain access to the forest. flict when the land, resources and The Batwa now live largely with limited resources, no access to social services, traditional practices are threatened by and often depend on begging as a form of livelihood. No government in the region other interests.52 In the DRC, for exam- has formal policies concerning Batwa people. Frequently any support provided is ple, the Batwa are reputed to take retri- dependent on the Batwa renouncing their traditional values and way of life for a bution into their own hands when sedentary agricultural lifestyle. In Rwanda, the post-1994 government does not seriously wronged, as they claim they are acknowledge the Batwa as a group that is marginalised and discriminated against, ignored by the authorities.53 and this has led to them being ignored in government programmes to provide social services such as primary healthcare, education, shelter and clean water. Census data on Batwa populations is rarely complete or accurate in any of the Conclusion countries in this region. Commonly these people are not acknowledged by the Different regions of the world have very state in the same way as other citizens. They rarely have citizenship documents – different incidences of chronic poverty birth certificates, identity cards, health cards. In Uganda, representation of the and, in particular, sub-Saharan Africa Batwa in Local Councils and other political leadership roles is not encouraged. and parts of South Asia are foci for con- Batwa are generally represented by people from other ethnic groups and those cern. There are also very different re- that are elected are elected because of their degree of assimilation and not gional dynamics for chronic poverty. On because of their capacity to fight for the rights of Batwa people. the coast of China, chronic poverty (and Sources: Baker, 2001; Lewis, 2000; Mugarura and Ndemeye, 2000. poverty in general) have declined rapidly over the 1990s. By contrast, in Central (IDPs) are a significant group of the Groups with no citizenship entitlements Asia and the CIS, growing numbers of chronically poor [see Chapter Two]. are particularly vulnerable, including il- people have fallen into chronic poverty Away from their original residence they legal immigrants and refugees who are and large numbers of children are being are often alienated and isolated, as the likely to have to accept the lowest paid born into it. support groups that existed in their own employment without recourse to avail- Within countries, chronic poverty is villages are no longer available to them. able legislation.51 Weak networks and almost always unevenly distributed, with They may be crowded into camps or links into patronage combined with the certain areas – rural and urban spatial areas where they are not welcome, and social stigma of being an outsider often poverty traps – having high levels. Ad- onto land that does not belong to them. make access to the labour market harder. verse ecological conditions, poor infra- In Sri Lanka, for example, the govern- In urban India, outsiders can find it near structure, weak markets and other ment estimates that the total number of impossible to get access to informal sec- institutions and political isolation char- IDPs (including those in welfare centres) tor resources such as market/vending acterise such areas. Commonly, such is 650,000, equal to roughly one-third of pitches on streets, which are often firmly areas also have high levels of the socio- the total population currently living in controlled by one ethnic group. economic characteristics associated with conflict-affected areas. In one affected People in spatial poverty traps are persistent poverty that were identified in area alone, it is estimated that as many often invisible to policy-makers, partly as Chapter Two – the spatial and socio- as 80% of the current population are a reflection of difficulties in counting and economic dimensions of chronic poverty displaced, having arrived as part of a collecting accurate statistics within re- are clearly two sides of the same coin in major influx of people from the adjoin- mote and difficult terrain, or in reaching many ways. As the next chapter argues, ing conflict areas in 1995.50 marginalised groups. In the case of indig- chronic poverty often seems to occur be- The effect of coming from a poor re- enous people, this is exacerbated by the cause several causal processes are in op- gion and suffering social stigma is acute. tendency for regional data to fail to eration at one and the same time. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P13E-01.3D.3D Folio: 35 Black plate (35,1)

Where do chronically poor people live? . 35

Technical note to Figure 3.1 The proportion and size of chronically poor populations in world regions The cartogram was designed by Mark Gordon for the Chronic the Population Division of the Department of Economic and Poverty Research Centre (CPRC). ERSI’s ArcView 3.3 with Social Affairs of the United Nations Secretariat.4 In cases Andy Adenda’s ‘Cartogram’ ArcView Script was used, based where the UN 2003 mid-year population was not available,5 on Charles B. Jackel’s script.1 the ESRI-provided 1994 estimated population of the country The initial shape files were drawn from Environmental prepared by National Center for Geographic Information and Systems Research Institute, Inc. (ESRI) ‘World Countries Analysis was employed. 2002’ shapefile from the ESRI Data & Maps Series published The chronic poverty cluster ranks were provided by the in 2002. The shapefile was projected in the WGS-84 CPRC. The estimated number of absolute poor was Projection and then modified. First, Antarctica and 35 small calculated by multiplying the 2003 mid-year population by the island nations with both low populations and rates of absolute most recent World Bank estimates of rate of absolute poverty, poverty were removed.2 Second, the 25 countries and islands or, in its absence, CPRC estimates. of Western Europe were merged into one polygon.3 Third, the 70 iterations were run to produce the presented cartogram. entire shapefile was generalized Douglas-Peucker-algorithm The average square of the percentage change in area by the from 165,797 vertices to 83,562 vertices using the Generalize final iteration was 0.025%. Tool in the ‘Point & Polyline Tools V1.2’ by Soeren Alsleben. The cartogram layout was prepared in ESRI’s ArcMap 8.3 Population data represent the 2003 mid-year estimates from and then exported as both a JPEG and Adobe .PDF file.

Notes 1. See ‘Using ArcView to Create Contiguous and Noncontiguous Area Islands, Samoa, South Georgia and the South Sandwich, Spratly Islands, Cartograms,’ Cartography and Geographic Information Systems, vol. 24, Svalbard, Tokelau, Tonga, Wallis and Futuna. no. 2, 1997, pp. 101–109. 3. Austria, Belgium, Denmark, Finland, France, Germany, Gibraltar, Greece, 2. American Samoa, Antarctica, Baker Island, Bouvet Island, British Indian Guernsey, Ireland, Italy, Jersey, Liechtenstein, Luxembourg, Man, Isle of, Ocean Territory, Cape Verde, Cook Islands, Falkland Islands (Islas Monaco, Netherlands, Norway, Portugal, San Marino, Spain, Sweden, Malvinas), Faroe Islands, Fiji, French Polynesia, French Southern & Switzerland, United Kingdom, Vatican City. Antarctic Lands, Glorioso Islands, Greenland, Heard Island & McDonald 4. http://www.un.org/esa/population/publications/wpp2002/POP-R2002- Islands, Howland Island, Jan Mayen, Jarvis Island, Johnston Atoll, Juan DATA_Web.xls De Nova Island, Kiribati, Midway Islands, Niue, Paracel Islands, Pitcairn 5. Cocos Island, Christmas Island, Norfolk Island, Taiwan, Wake Island, Mayotte Island.

Notes 1. All but the poorest households belong to 14. Goodhand 2001. 30. Jalan and Ravallion 1997. ‘ambulance clubs’ and pay a monthly insurance 15. CEPAL 2001 in Wheeler 2003. 31. Farrington and Gerard 2002 in Bolt 2003. contribution which will pay for strong young 16. Wheeler 2003. 32. World Bank 2000 in Bolt 2003. men to carry them to the health centre in a 17. World Bank 2003a. 33. Kothari 2002. stretcher should they fall sick. The poorest are 18. For example, McCulloch and Caladrino (2003) 34. Black et al. 2003. excluded because they can not afford the cash found that chronic poverty was significantly 35. Holden et al. 2001 in Bolt 2003. payments. more likely to occur in upland areas in Sichuan 36. Bird and Shepherd 2003. 2. Passionfruit was recently introduced but has than lowland areas. 37. Shinyekwa et al. 2003; Bird and Shepherd succumbed to disease. Coffee wilt reduces 19. For example, in Shanxi Province, 3.8 million 2003. coffee yields. poor people live in the 34 designated poor 38. Van de Walle 2000 in Bolt 2003. 3. Baulch and Masset 2003. counties but another 1.5 million are recorded in 39. Holden et al 2001 in Bolt 2003. 4. Reyes 2002a/b. the province’s non-poor counties (Cook and 40. World Bank 2003b. 5. Jalan and Ravallion 1997. White, 1997). Also see Riskin, 1994. 41. Bird and Shepherd 2003. 6. Davis and Stampini 2002. 20. Ashley and Maxwell 2001. 42. Herrera and Roubaud 2003. 7. In the cartogram we have used absolute poverty 21. World Bank 2003b. 43. Kedir and McKay 2003. (US $1/day poverty) as a proxy for chronic 22. FAO 1996a in Bolt 2003. 44. Satterthwaite and Tacoli 2002. poverty. For those countries for which there is 23. Sen 2003; Sen and Ali 2003. 45. Some urban poor choose criminal activities no international data, rates have been 24. Notably Green Revolution investments. See because of their high returns. estimated. See technical annex. Bolt, 2003: 2. 46. Thornton 2001. 8. Devereux and Sharp 2003. 25. Recent research based on India and China (Fan 47. Bebbington 2003. 9. This conventional definition is currently being and Hazell 2000) has suggested that 48. Malapit, Clement and Yunzal 2003. taken forward by the work within the CPRC. In investments (especially in agricultural research 49. Bird et al 2002. Uganda, work is proceeding on an ‘index of and development, roads and education) in low 50. Korf and Silva, 2003; ‘. . . 302,586 individuals isolation’ which measures remoteness from potential and less-favoured areas make a (78,828 families) returned to their homes within services, and access to information as well as stronger contribution to reducing poverty than Sri Lanka from January 2002 up to and physical remoteness. It will be interesting to see similar investments in high potential areas. This including April 2003.’ (Durable Solutions, the extent to which this can account for the is not generalisable to Africa, however; 2003). geographical distribution of Uganda’s poverty. investment in high-productivity areas has yet to 51. In practice, even those groups that do have 10. Note that others define ‘less favoured’ as areas reach the point of diminishing returns. citizenship may not have recourse to available with low agricultural potential. 26. Mehta and Shah 2003. legislation. 11. Pender and Hazell 2000. 27. Gaiha and Deolalikar 1993. 52. Bourne 2003. 12. Ruben et al. 2003. 28. Gaiha and Imai 2003. 53. Lewis 2000. 13. Alexandratos 1995. 29. Shah and Sah 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 36 Black plate (36,1)

the wealthiest countries of the world.3 The paucity of data means that there is little precise evidence about the impact of economic change on the chronically Why are people poor. By the nature of the deprivations they experience, chronically poor people chronically (particularly older and disabled people), poor? are likely to find it most difficult to par- 4 ticipate in growth as well as being the most difficult group to benefit through macro level changes and growth.4 There Chronic poverty is not just a snapshot of those who are poor now. It is are many examples of where growth has essential to understand the social, economic and political processes that failed to deliver improvements in key made people poor and keep them poor. At the present time, our know- human development indicators closely ledge about the causes of chronic poverty is limited. The longitudinal correlated with chronic poverty, and data and analyses that are required to understand poverty that persists others where human development indi- are rare. It is clear that processes vary greatly from context to context. In cators have improved without economic some cases, the causes of chronic poverty may be the same as the growth. Growth therefore may not deliv- causes of poverty in general, only more intense, widespread, or longer er short term improvements in these lasting. In other cases there is a clear qualitative difference between the human development indicators. Ulti- causes of transitory and chronic poverty and these may need quite differ- mately though there remains a strong ent policies to reduce them. correlation between the levels of GNI and human development indicators. What is critically important is a coun- The lives of people who are chronically try’s choice of how to allocate the in- poor follow many different trajectories. Staying poor: the creased resources arising from growth.5 At one extreme is an individual born maintainers of poverty Proponents of growth-mediated pov- into multidimensional poverty, experi- Barriers to accumulating or accessing as- erty reduction argue that achieving encing a lifetime of deprivation and sets and pursuing opportunities are the growth is the only policy issue of signifi- transferring that poverty to her/his chil- key maintainers of chronic poverty. They cance. However, the widely cited Dollar dren. At the other extreme is someone hold individuals, households, commun- and Kraay argument6 that, on average, born into a well-off family who suffers a ities and states back from making the the incomes of the bottom 20% im- shock that drives them into income pov- economic and social investments needed proved one for one with those of the erty for many years, but who eventually to move out of poverty. Access to assets population as a whole as a result of eco- escapes poverty and has a family with and opportunities is constrained by low nomic growth needs to be carefully as- prospects. Among the different trajecto- rates of economic growth, inequality, so- sessed.7 It does not allow for variation ries are those – all too many – that end cial exclusion and adverse incorporation, around the average (which is known to with an easily preventable death. geography and agro-ecology, violent be significant), it uses a relative concept Understanding chronic poverty is diffi- conflict, state failure, ineffective interna- of poverty, the dataset used has been cult because of multiple overlapping and tional co-operation and, perhaps, ‘cul- criticised,8 it does not consider poverty interacting factors. The available evi- tures of poverty’. depth, and researchers using a different dence indicates that chronic poverty econometric approach with the same rarely has a simple ‘single cause’. Rather, Chronically poor people data9 have produced contradictory 10 different forces, from the household to are likely to be the findings. the global level, interact to trap people hardest group for macro Growth figures describe an average for in poverty. a whole nation or state. The distribution- In this chapter, the causes of chronic level changes, such as al pattern varies widely from case to poverty are discussed in terms of main- growth, to benefit. case11 and so does the impact on chronic tainers and drivers. Maintainers make poverty. Key to this impact is the initial poverty persistent and trap people in pov- level of inequality and the nature and erty. Drivers cause individuals and house- Economic growth, growth pace of the growth experience. It is holds to fall or slide into types of poverty quality and inequality broadly recognised that higher inequality – severe, recurrent, multidimensional – Global poverty has dramatically reduced will mean: that are hard to escape. Maintainers and in absolute terms since the early 19th . higher levels of income poverty (and drivers cannot always be precisely distin- century,1 closely associated with in- probably non-income poverty) for a guished from each other, but this device creased rates of economic growth and given average GNI per capita; helps explain how the socio-economic rising levels of GNI per capita.2 But, de- . growth will be less effective at reduc- environment (economic growth, social spite these impressive declines, 300 to ing poverty; and, exclusion, bad governance) and concrete 420 million people remain trapped in . generally, growth rates will be lower. experiences (a sickness, a drought, a chronic poverty. These estimates do not These same arguments hold for the beating) make and keep people poor. take account of chronic poverty within chronically poor.12 Economic growth in Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 37 Black plate (37,1)

Why are people chronically poor? . 37 countries with high initial levels of in- leaving this high dependency group most almost twice as many people who were equality and increasing inequality will be vulnerable to economic shock. Even chronically poor as were transitorily relatively ineffective at assisting the where high levels of private and/or pub- poor (measured at US$2/day). These chronically poor. Public service expan- lic transfers exist, the effectiveness of limited results suggest that economic sion is one product of growth but coun- both formal and informal schemes growth has a wider impact in cities and tries differ on the level and distribution varies. that there may be more poor people in of service provisioning, and so the The available evidence indicates that rural areas with structural obstacles to growth impact on the chronically poor participation in economic growth. also varies. This has considerable implications for policy and points to the need If exclusion is the problem for ‘growth plus other policies’ strategies then it could be assumed Social exclusion and to help the chronic poor. that inclusion is the answer. adverse incorporation In considering the relationship be- This is not necessarily right. Exclusion from social, political and eco- tween growth or economic change and nomic institutions is part of a vicious chronic poverty, it is important to recog- cycle in which exclusion leads to lower nise the diversity of the chronically poor. capabilities, which in turn reduces the Typically the majority are engaged in no growth, low growth and narrowly prospects for escaping poverty and peo- some form of work or productive activ- based growth increase the probability of ple’s ability to assert their rights. Com- ities, though almost by definition on ad- people being trapped in poverty for long monly, such exclusion operates through verse terms, disadvantaged by differing periods or all of their lives, and transfer- varying forms of active discrimination factors such lack of assets, insecurity of ring that poverty to the next generation. directed against certain identifiable employment opportunities, ill-health or So, promoting economic growth is im- groups (as a result of ethnicity, race, reli- malnutrition or lack of social networks. portant for reducing chronic poverty, gion, caste, culture, migration), which is Others may not be able to work at all as and the absence of growth is a severe often reinforced by discrimination on the a consequence of old age or disability constraint to reaching the chronically basis of personal characteristics (age, (although many in this group may still poor, but growth is not enough nor is it gender, impairment). At other times, ex- be engaged in very marginal activities – ‘almost enough’. clusion is relatively passive, based on the see Box 4.1).13 Table 4.1 provides support for this, ignorance or preferences of more power- For those that are able to work, the based on a limited number of cases. Eco- ful and better off groups, simply not sectoral composition of growth really nomic growth had very clear benefits for knowing about or leaving out certain in- matters, particularly as to whether it in- the urban poor in Vietnam and Uganda, dividuals or groups of the chronically cludes broad-based agricultural growth and its absence was clearly bad news for poor. and is in sectors where the employment- the urban poor in Egypt. However, the If exclusion is the problem then it elasticity for unskilled labour is high. picture was not quite as predicted for could be assumed that inclusion is the The existence, extent and type of barriers rural India and Vietnam and urban answer. This is not necessarily right. to markets are also important, including Ethiopia, where average consumption Many of the poorest people are included processes of social exclusion and adverse growth did not translate into a greater in economic activity, but on extremely incorporation (see below), and high lev- likelihood of escaping from poverty. unfavourable terms. In many parts of the els of inequality. Where the economic Rural Egypt’s negative growth produced world, migrant labourers and their context is narrowly based, such as in oil or mineral extraction for example, and social exclusion and/or elite domination is high, the chronically poor will experience few benefits from growth and their position may worsen. The chronically poor are not able to increase their access to productive assets, or increase the productivity of those assets, or achieve better incomes in the labour market. For the ‘non-working’ group the situation is quite different. The benefits of economic growth for this group derive from a mix of private transfers from relatives and friends, and publicly-funded social protection, such as resources for nutrition support, education and health services, as well as public transfers like pensions and employment schemes. The degree to which additional resources are channelled into such schemes is often re- This woman, employed on a large horticultural farm in Eritrea, is an important part of the economy – but sponsive to broader economic contexts, on very disadvantageous terms. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 38 Black plate (38,1)

38 . Chapter 4 National poverty headcount: ﬁnal year (%) National poverty headcount: ﬁrst year (%) Descending into poverty (%) Of whom . . . Escaping from poverty (%) %in transitory poverty % in chronic poverty 6.3 33.3 36.7 24.0 12.7 ...... 5.78.7 10.2 30.1 6.5 24.1 19.4 17.3 6.0 27.8 2.1 10.3 23.8 8.5 4.26.9 18.9 28.7 40.1 32.1 29.7 27.4 10.1 4.7 55.7 56.1 35.0 33.5 4.06.1 20.5 33.9 41.8 35.1 30.7 29.7 11.1 5.4 59.7 63.6 39.1 39.2 –8.9 25.2 27.1 17.9 9.2 34.4 42.9 –3.3 42.6 22.6 ...... –8.1 14.2 18.0 ...... –5.9 19.0 20.4 6.3 14.1 ...... Average change in standard of living measure, annualised (%) Regression based on calorie requirements (2200 kcal/day per capita Calorie based; norms between 2360 and 2499 per capita depending on location Planning Commission poverty line – calorie based Calorie based 2100kcal/person/day plus non-food allowance Calorie based; norms between 2360 and 2499 per capita depending on location Calorie based plus non-food component (3000kcal/day per adult) Calorie based 2100kcal/person/day plus non-food allowance Calorie based; norms between 2360 and 2499 per capita depending on location Calorie based plus non-food component Calorie based 2100kcal/person/day plus non-food allowance Calorie based plus non-food component Standard of living measure Poverty line Expenditure per adult equivalent Consumption expenditure per capita Expenditure (per capita) Consumption expenditure per capita Expenditure per adult equivalent Expenditure (per capita) Consumption expenditure per capita Expenditure per adult equivalent Expenditure (per capita) Expenditure per adult equivalent Country/period Urban Ethiopia, 1994–97 Rural Egypt 1997–99 Urban Vietnam 1993–98 Urban Egypt 1997–99 Urban Uganda 1992–99 National Vietnam 1993–98 National Egypt 1997–99 National Uganda 1992–99 Rural Vietnam 1993–98 Rural India 1968–70 Rural Uganda 1992–99 Table 4.1 Economic growth, chronic poverty and poverty dynamics ‘. . .’ means not available Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 39 Black plate (39,1)

Why are people chronically poor? . 39 families, without local social networks, are excluded from public services and Box 4.1 Adversely incorporated, top to bottom the institutions of governance, but are Fruit picked at Ceres in South Africa regularly appears on the supermarket shelves ‘adversely incorporated’ into the labour in developed countries. Workers there depend on the seasonal labour generated market, that is, forced to take work at by the global fruit market. Although race (as well as gender) remains a central low rates, in bad conditions and on pre- component of the South African political-economy of poverty, the dynamics that carious terms. This is sometimes the re- keep fruit workers poor are not just about exclusion, but adverse incorporation into sult of the way local labour markets are the global labour market. structured and relate to other institu- The fruit workers and the globalised fruit industry are interdependent but, for the tions. In South Asia, households that workers, this is on unfavourable terms. They are trapped in a situation which face social exclusion because of religion, forces them to rely on poorly paid, scarce and insecure jobs. ethnicity or caste, are significantly more As a result of the casualisation of labour, more than half the farms that used to vulnerable to labour market exploitation provide on-site housing have ceased to do so. For landless workers, relying on and debt bondage than other economi- seasonal and temporary demand for labour, dependence on a paternalistic farmer cally poor families.14 for employment has been replaced by dependence on a contractor who acts as a Interrelated processes of social exclu- labour broker. sion combine with adverse incorporation Due to low incomes and weak access to economic and natural resources for to deepen workers’ dependency. An ex- basic household food production, people often have to buy food on credit, in small ample is bonded labourers’ dependence amounts and at higher prices. Networks of mutual aid and support enable basic on manipulative employers who may survival, but provide only limited protection against shocks. Limited social safety construct intricate systems of advances, nets and few alternative livelihood options have given rise to a burgeoning payments in kind, and hidden interest underground economy. Rising criminality, violence and exploitative relationships rates, which maintain the disadvantaged effectively increase the vulnerability of poor fruit workers. status of the workers. The capability pov- Chronic poverty is maintained among fruit workers in Ceres by both erty of the chronically poor, such as illit- ‘downstream’ factors – the ways in which products are processed, shipped, eracy and innumeracy, along with wider marketed and consumed, and ‘upstream’ factors – the relationships that are institutional indifference (the absence of involved on the input side of agricultural production. People in chronic poverty are an effective legal system to assert the linked to the fruit consumers of Europe via the economics and politics of the export rights of the poorest, the direct and indi- chain. rect influence of employers or their rela- Source: du Toit 2003. tives in locally elected bodies and systems), prevents labourers from taking low potential areas that compound any need for sustained political energy, both action to challenge their position.15 geographical disadvantage. These log- in the wider polity, and in the region it- Global commodity chains, such as the ex- jams are made up of social and political self, where political leaders may benefit port fruit industry in South Africa, can exclusion owing to ethnic status, thin from the status quo. The combination of produce similar conditions (see Box 4.1). and interlocked markets, poor gover- political demand from the region and the The structures of social exclusion – nance, and high levels of exposure to willingness to redistribute or target re- discrimination, stigma, and invisibility – asset depleting risks. These different sources from the centre is critical. often create the basis for processes of ad- forms of disadvantage reinforce each The danger is that integration, while verse incorporation – declining assets, other where both social exclusion and good for opportunity (especially for la- low wages, no job security, restricted ac- adverse incorporation dominate the lives bourers), leads to loss of regional or cess to social protection, and dependency of the poor. local character and may have costs. Bet- on a patron. The ways in which risk and It is the concentration of a combina- ter placed areas may out-compete more vulnerability shape social relations is tion of factors that keeps places poor, remote areas in production terms once key. Chronically poor people often man- often bolstered by a history of purpose- integration is enhanced. Supplementary age vulnerability by developing patron- ful neglect which deters investors, both policies such as social protection, market client ties that produce desirable, imme- public and private. At the extreme, pla- development, improved governance and diate outcomes (access to food, access to ces treated as labour reserves or closed cultural and environmental conservation health services and so on) by trading-off areas can becomes sites where criminal- measures may be needed to mitigate the their longer-term needs and rights (ability ity and illegal economies prevail, and costs of integration. to accumulate assets, rights to change em- where state institutions are ineffective. ployer or vote freely, for example). ‘Stay- This is summed up in the notion of ‘geo- ing secure [often means] staying poor.’16 graphic capital’.18 Cultures of poverty It takes public and private investment Whether the ways in which children are to raise the geographic capital of an born and raised in poverty influences Geography and area, and the prospects for growth in a them to become poor adults and pass agro-ecology spatial poverty trap depend on govern- poverty on to their own children is a Geography and agro-ecology can be ments and community organisations controversial issue and often avoided. maintainers of poverty. There are often overcoming the negative perceptions and Yet, in terms of the persistence of pov- ‘logjams of disadvantage’17 in less fav- marginalisation of the area among in- erty, it is important to ask whether there oured, weakly integrated, remote and/or vestors.19 Underlying the ‘problem’ is the is something about the ways in which Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 40 Black plate (40,1)

40 . Chapter 4 people cope with poverty (psychologi- Lack of investment in school, and miss out on critical health cally as well as economically and so- care.24 Women face increased demands cially) that makes poverty more difficult primary health care, or to fulfil the triple roles of child and fam- to escape. primary education and other ily care, income generation, and partici- Elites have long used ‘cultures of pov- life skills results in permanent pation in the wider community. This erty’ to blame the poor for their poverty life-long deprivation. reduces their time for child care which – ‘they are poor because they are drunk may be delegated to others – in some 20 or lazy or ignorant’. Such assertions have adapted to and coped with poverty cases, to children, who may be unable to are often based on little understanding over years and generations. Coping and care for their younger siblings 25 about the poor. Aspirations for children, survival strategies passed on from one adequately. attitudes to risk, rules of behaviour, generation to the next can help survival Poor nutrition has serious, long term emotional resilience, are all important in bad or deteriorating conditions, some- and intergenerational effects on physical for coping with and escaping from pov- times keeping the poor from destitution and mental health, mortality and chronic erty. But the beliefs and norms of society or death. However, they may also con- poverty. Even a relatively short period of as a whole or some poor people in par- tain elements which help to maintain deprivation in childhood can harm child ticular, may limit the possibilities for es- poverty, through reproduction of the so- nutrition, health, education and aspira- cape. These are real problems, which cial and economic structures that ob- tions, with serious repercussions for the limit the agency of the poor and play a struct escape – a form of adverse long-term well-being of both the child 26 role in the persistence of poverty; they incorporation. This means that effective and her/his own children. The negative need to be understood and overcome just policies to reduce chronic poverty have effects of stunting are almost impossible as much as the mechanisms of social ex- to deal with the difficult issues of values, to reverse. Lack of investment in primary clusion or adverse incorporation. attitudes, aspirations and confidence. health care, or primary education and Culture affects what is transferred other life skills results in permanent life- from one generation to the next. Norms long deprivation.27 of entitlement affect intergenerational Capability deprivation Acquiring capabilities also has lifelong transfers by determining who has access People trapped in persistent poverty tend and intergenerational benefits. There is to and control over resources, and who to experience multiple capability depri- strong association between parents’ and is dependent on others. The aspirations vations, such as poor education, illiter- children’s education levels.28 Knowledge of parents may have substantial effects acy, bad health, inadequate nutrition, and skills, and in many cases a formal on children’s up-take of education. lack of human rights and civil rights. qualification, can facilitate upward eco- Norms about diet influence maternal nu- These constrain opportunities and nomic and social mobility, offering a trition and the health of babies in the choices.23 They have mutually reinforc- means to getting a better-paying, safer womb, as well as after birth. Norms also ing impacts among themselves and job; accessing credit; keeping accounts; influence perceptions of ‘strategic’ trans- across generations. extending social networks to include fers, for example of resources to younger Most chronically poor people have in- more influential contacts; and garnering generations in order to ensure support in secure and inadequate livelihoods. increased respect. Women’s education in one’s old age.21 Households suffer hunger and enormous particular is also strongly associated The ‘culture of poverty’ theory22 links stress, sell off key assets, and get into with improved child health and nutrition poverty to the ways in which the poor debt. Children may have to drop out of and children’s own educational success.29 Capability deprivation is intimately tied to public policy, sometimes reflecting discrimination within policy processes against particular children, households, groups and areas. The chronically poor may have less access to public services, either because they live in remote areas, or regions where government has not prioritised development, or because of discrimination among public servants, and language, or other cultural barriers. Exclusion of minority, indigenous children from school, for example, is often a result of discriminatory treatment either by teachers or pupils, a curriculum that perpetuates negative representations of minorities and indigenous peoples, lack of funding to the regions where they are located, and the lack of These farmers in northern India rise at dawn to work. Like many chronically poor people, their (or low-quality) education provision in opportunities and choices are very limited. minority and indigenous languages. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 41 Black plate (41,1)

Why are people chronically poor? . 41

Similarly, the chronically poor may face HIV/AIDS, governance, social integra- and 2000), being marginal to the world direct and indirect discrimination in tion and disintegration and conflict. economy and, with few exceptions, to health facilities. Older and disabled peo- While poor performance is attributed, by the geo-political interests of the world’s ple, for instance, report negative atti- most international commentators, to dis- big powers. They have suffered from a tudes from health staff which dissuade abling internal conditions for invest- relatively unregulated international arms them from seeking treatment.30 The par- ment, aid, and peace in the countries trade and have been at the receiving end ticipation of chronically poor people in concerned, there are also international of declining terms of trade for primary programme design and evaluation is explanations and causes to be addressed, commodities. rarely invited so public services are not and opportunities to grasp. There are many complex debates attuned to their cultures and lifestyles. about if, when, and how the internation- State failure means that al community should help failed and social protection and social weak states. The conviction that aid is Weak, ‘failing’ and ‘failed’ most effective when spent in good policy services, such as education states countries, and that all aid resources A large number of chronically poor peo- and health, do not operate, so should be directed to those countries has ple live in countries or states where pov- human capital is weakened. led donors to ignore the imperative to erty is as bad in depth, absolute address all poverty and to avoid the chal- numbers, or proportion as it was 30 Exactly which states are ‘weak’ or lenge of how to reduce suffering and years ago. Typically, state capacity is ‘failing’ depends on the criteria used. poverty in failed, fragile and weak states. low or non-existent and violent conflict The World Bank’s classification ‘Low In- The international community needs to disrupts economic and social activity. come Countries Under Stress’ (LICUS) make much faster and stronger progress Panel data on these areas is rare, so it is identifies 27 countries; UNDP’s ‘fragile on improving terms of trade, particularly difficult to give precise estimates of states’ cover 46 countries. ‘Desperately for primary commodities, and on the chronic poverty, however, the processes deprived’ and ‘moderately deprived’ arms trade to enable weak and failed that trap people in poverty are evident. countries – classified by CPRC, using states to improve. Violence destroys assets and discourages human development indicators – are gen- both domestic and foreign investment erally ‘poorly-performing’ on one or (except for illegal goods and enclave ex- other of the above analyses. By contrast, Weak and failed tractive activities) so that growth is low the ‘not deprived’ countries are hardly international co-operation or negative and is not pro-poor. State ever included as poor performers, except A key explanatory factor for the failure failure means that social protection and some of the Central Asian transitional to reduce poverty in many countries over social services, such as education and countries, where economic and human the 1980s and 1990s, and for the in- health, do not operate, so human capital development has gone into reverse. What crease in poverty in the former Soviet is weakened. And the breakdown of law does this mean? Countries whose popu- Union and Central Asia, is the ineffec- and order raises the likelihood of in- lations are ‘desperately’ and ‘moderately’ tiveness of the international system. creased inequality as violence, or the deprived are typically seen international- Structural adjustment, prescriptions for a threat of violence, determines access to ly to: rapid move to the market,32 reduced vol- assets and incomes. . have weak rule of law; umes of aid, the tying of aid, the alloca- The Human Development Report . have been prone to conflict; tion of aid away from the poorest 2003 identified more than 50 countries . be subject to low levels of civil and countries,33 the continued agricultural that grew poorer over the 1990s, symp- political rights; protectionism of the OECD countries, toms of failed economic growth and the . have governments which are weakly the unregulated trade in arms, reckless HIV/AIDS pandemic. Human develop- responsive to their citizens; lending to known corrupt despots, and ment indicators got worse in 21 coun- . in some cases, avoid engaging much in the failure to cancel debt, all helped to tries. 59 ‘top’ and ‘high’ priority the economic and governance reforms maintain poverty levels. They also con- countries failed to make progress to- promoted by the international com- tribute to maintaining levels of chronic wards the Millennium Development munity since 1980, or, where they poverty. Goals.31 38 of these countries are in sub- have, the reforms have not born fruit. The international system filed against Saharan Africa. This omits the states, re- However, it is also possible to see these social protection for poor countries dur- gions and provinces of large countries countries as affected by particularly diffi- ing the 1980s and 1990s as being unaf- which may also be stagnating or even cult structural economic conditions – fordable and not likely to contribute to experiencing reversals, while other re- poorly connected, small economies, with development. That assumption needs ur- gions progress. Including the poorly per- weak human development positions to gent review. forming states of India and the remote start with. They are unable to move into Aspects of the international system re- regions of China would increase the more beneficial international trading main hostile to the interests of chronic- number of people affected by persistent, lines – the manufactured exports and ally poor countries: trade in primary territorial poor-performance by scores of service economy which more rapidly commodities; the unregulated arms millions. growing countries have entered. They trade; and the volume of export earnings Why these states ‘fail’ is a multi-fac- have not benefited from reliable aid poor countries still have to pay in debt eted issue involving strategies for eco- flows (LDCs have seen a decline per cap- service, despite significantly lower debt nomic growth, human development, ita from US$29 to $19 between 1994 stocks in particular, are examples. The Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 42 Black plate (42,1)

42 . Chapter 4 failure of OECD countries to meet their . the nature of the broader institutional but discrimination based on gender obligations on aid volume and alloca- context, including systems of social and marital status strips away any tions to basic social services is a failure protection, basic services, and conflict assets that could be used to bounce of international co-operation, and one prevention and resolution, as well as back. which denies essential resources to public information. . Fire: Box 4.2 shows how household chronically poor people. The structures that maintain people in and neighbourhood poverty, and ex- poverty play the largest role in hindering clusion from basic services exacerbate Falling into poverty: recovery from shocks, as they affect both the effects of a shock. assets and the institutional context. A . In Box 4.7, the story of Maymana and the drivers of chronic poor household living in a remote area Mofizul, a combination of shocks at poverty and suffering discrimination is unlikely the micro-level and structures at the Not all chronically poor people are born to have access to the resources or public community level combine to maintain into long-term deprivation – many slide services required to bounce back. The in- and deepen their poverty. into chronic poverty after a shock, or teraction between drivers (shocks) and series of shocks. What drives people into maintainers is illustrated in the boxes chronic poverty is not very different to throughout this section: Access to assets what drives people into poverty in gener- . Market collapse: In Box 4.6, a shock Limited access to assets, both private al. The types of shocks are the same – ill- to households and entire economies is and collective, is what drives already health and injury, environmental shocks, traced back to a lack of global eco- poor people into deeper and more in- violence, the breakdown of law and nomic governance. tractable poverty after a shock. Those order, market and economic collapse. . Civil war: Box 4.5 examines the mac- with few material, financial, natural or But while some households recover rela- roeconomic effects of the eruption of social assets are vulnerable to relatively tively quickly, others are pushed into a conflict. minor shocks, which nonetheless cause downward spiral of asset depletion and . Famine: Box 4.4 details how bad catastrophic declines into persistent pov- increased vulnerability, culminating in weather (a shock), bad policy (a fail- erty. The coping strategies undertaken severe and chronic poverty. The factors ure of national and international gov- by a poor household without ‘reserves’ that combine to determine this ability to ernance), and reduced resilience are those that, while possibly ensuring bounce back from a shock (resilience) combined to cause hundreds, perhaps short-term survival, can undermine well- can be divided into three main thousands, of preventable deaths and being in the medium to long-term. Any categories: have trapped many ‘survivors’ in in- crucial productive assets are liquidated, . the level of private and collective as- tractable poverty. and households often resort to reducing sets to which people have access; . Property grabbing: Box 4.3 describes a consumption in ways that have poten- . the nature of the shock(s), in terms of situation where not only is there no tially irreversible welfare effects – eating depth, breadth and sequencing; and safety net to fall back on after a shock, a smaller amount of less nutritious food, avoiding essential medical expenditures, withdrawing children from school. Even when a poor household does Box 4.2 Slums, shocks and chronic poverty possess assets, these assets are themselves Many people in low-income settlements in Delhi, Dhaka and Cape Town report often highly susceptible to shocks. Hous- that their descent into chronic poverty was initiated by a house fire. They give a ing can be destroyed; cash and jewellery range of similar reasons why a small fire can become a shock from which recovery can be stolen; land and livestock can suc- is impossible: cumb to erosion and disease. Financial . Houses built with poor quality, flammable materials and crowded together favour assets are susceptible to economic shocks the spread of fire. and devaluation. Social capital is vulner- . Narrow streets and political marginality mean that the fire brigade cannot or will able when the shock affects entire com- not come to a slum fire. munities or countries. . Limited water supplies mean slum dwellers can do little to stop fires from spreading. . Injuries to (or death of) household members leads to increased health bills and The nature of shocks reduced income. Discussion of the relationship between . Damaged or lost workplace or equipment (often also located in the slum) reduce shocks and poverty is not new.34 But income. what types of shocks result in irreversible . Distributions of relief from public funds are insignificant, inegalitarian and decline? Intuitively, the shocks from intermittent. which it is very difficult to recover are . A lack of access to savings and insurance services means households have those that are severe, sudden, unpredict- limited or no ‘crisis funds’ to fall back on. able, sequential and affecting many . The destruction of identity documents and pension books means the household people. loses access to public entitlements (e.g. subsidised food, free health care, The severity of a shock can be meas- pensions). ured by the extent to which it causes the . Informal, intra-community safety nets are undermined as neighbouring loss of productive assets and/or increased households are in a similar position and cannot offer support. resource demands. The devastating effect Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 43 Black plate (43,1)

Why are people chronically poor? . 43 of losing a home is compounded by losing tools and working premises with Box 4.3 Widowed, abused, and robbed which it would have been possible to re- Both in law and in practice, women are denied equal property rights, particularly in build a livelihood. It is compounded sub-Saharan Africa, South Asia and the Middle East. An unequal or absent means through injury that both limits the ca- of inheriting, managing and trading property, especially land, is a major maintainer pacity to work and requires resources to of poverty among women and their dependents. The effect of this discrimination is be spent on care and treatment; death multiplied when it emerges suddenly and brutally after a husband’s death, by means lost income and funeral expenses. which time many households are already reduced to extreme poverty. Throughout southern and eastern Africa, a high proportion of widows and their children are left The shocks of nature destitute when their dead husband’s relatives snatch their property – theft in the Sudden, unpredictable shocks can have guise of tradition, fostered by unequal and unenforced laws. huge consequences even for those with Sources: Human Rights Watch 2003; the financial and social resources to plan UN Integrated Regional Information Networks 2002. ahead. People without such resources find themselves made destitute from one rural areas, due to the density of popula- assets, reduced capabilities – a lack of day to the next, perhaps permanently. tion and infrastructure. Effects can be felt animal transport means children can’t Sudden shocks are often natural disasters far beyond the city when the economy is get to school or the sick to health care; and environmental, agro-ecological or interrupted, particularly if the national concentrated grazing undermines pasture agro-climatic shocks. distribution of goods is affected. sustainability. It is increasingly recognised that the in- Sequential shocks, like poor weather teraction between a natural hazard and a year after year, are particularly difficult vulnerable human population is what to recover from, especially if they follow Institutional context causes a livelihoods shock (see Box 4.4). in close succession or trigger off a series The extent to which the broader institu- Severe and unpredictable shocks – floods; of further shocks. In Mongolia, for ex- tional environment enables people to re- droughts and El Nin˜ o episodes; land- ample, different combinations of sum- sist and recover from the negative effects slides; volcanic eruptions; earthquakes; mer drought followed by severe winters of shocks is a crucial determinant of extreme temperature; fires; windstorms; – known locally as dzud – can have dev- whether poverty will be transitory and infestations; and disease outbreaks – es- astating effects on herding families, as isolated, or chronic and endemic. Three pecially when successive, can all under- entire flocks are lost. This sets into mo- common types of shock – illness, vio- mine otherwise secure livelihoods and tion a downward spiral of reduced lence and market collapse – demonstrate lead to impoverishment, diminished capabilities and preventable death. In much of the developing world, ex- Box 4.4 The 2002 Malawi famine – bad weather, bad policies, treme climatic events are regular but increased vulnerability largely unpredictable. These events have The food crisis in Malawi in early 2002 resulted in several hundred – perhaps been increasing in severity and fre- several thousand – hunger-related deaths. Starving Malawians resorted to eating quency, probably due to global climate unripe and unconventional foods, including flour ‘fortified’ by maize cobs and change, and will continue to do so.35 As sawdust, much of which made them ill. Malnutrition was high, not only among climate zones shift over the coming deca- young children, older people and the ill, but also among working adults. An des, those with the fewest resources will estimated 30% of the population required emergency aid. be the most vulnerable to new patterns The famine can be explained in two ways. The ‘technical’ view is that an of extreme conditions. The chronically environmental shock (bad weather), limited information, and import bottlenecks poor are almost certain to be among resulted in famine. The ‘political view’ attributes blame to different actors, those least able to adapt rapidly to cli- depending on who one talks to: the IMF for recommending the sale of strategic mate change, unless unprecedented levels food stocks; Malawian politicians for selling off the entire food reserve, and making of effective support are provided. money on the side; complacent government and donor officials; and profiteering It is more difficult to recover from traders inflating prices. The truth lies somewhere in a combination of the technical shocks that operate at an aggregate level, and political views. affecting entire communities, countries In addition to these immediate causes of famine, there are a number of and regions. When the total pool of re- underlying vulnerability factors that left poor Malawians unable to cope with a sources available is decimated, people in production shock that was actually less severe than the drought of 1991–2: the same, sinking boat can’t help each . declining soil fertility and neglect of smallholder agriculture, particularly in remote other enough for recovery. Even rela- areas; tively affluent households can be highly . deepening poverty that decimated asset buffers (foodstocks, savings); vulnerable to long spells of poverty when, . weakened informal systems of social protection in poor communities; for example, severe crop shocks occur.36 . the demographic and economic consequences of HIV/AIDS. The location of a shock can also exac- Better weather in mid-2003 has been followed by a period of low rainfall. Recent erbate its effects – urban areas prone to reports describe a country struggling through the lean season, on the brink of natural disasters (floods in Bangladesh; another food crisis. volcanic eruptions in Goma, DRC; earth- Sources: Devereux 2002; with additions from Bird, Booth and Pratt 2003; quakes in Iran) can suffer more than Cammack et al. 2003; WFP 2004. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 44 Black plate (44,1)

44 . Chapter 4 the extent to which a lack of health serv- also least likely to be able to afford the rising expenses and liquidating assets can ices, social protection and conflict pre- many direct and indirect costs of treat- reduce the household to a state of vention are implicit in the generation ment. People with impairments and chronic poverty by the time the ‘bread- and maintenance of chronic poverty. some socially stigmatised illnesses face winner’ dies, often prematurely. the added challenge of disabling discrim- The likelihood that illness and injury When there is no health care or ination, further driving both income and will deteriorate into a longer-term im- social protection multi-dimensional poverty. In the past, pairment, or death, increases when a Social protection – including cash grants diseases such as tuberculosis and cancer household is unable to pay for the neces- and income support, free access to basic typified such ill-health spirals but today sary treatment, or cannot endure the services, food and training, pensions, in- HIV/AIDS is increasingly the associated temporary absence from work due to ill- surance schemes, and emergency relief – diagnosis. ness. Illness episodes experienced by the can help prevent people from sliding into Evidence from nine developing coun- poor tend to last longer than those suf- chronic poverty, and enable poor people tries37 shows that the poor spend a sig- fered by the rich.42 Tuberculosis is par- to escape. It can shore up current con- nificantly higher proportion of their ticularly problematic in this way. A sumption so that potentially irreversible income on health care than the non- private doctor running a clinic in Uttar welfare effects do not occur. It can pre- poor.38 The proportion spent by the Pradesh, India explained that most TB vent the erosion of savings and other as- poorest ranged from about one-and-a- patients who are poor do not come to sets, and help the poor to avoid half to ten times that spent by the rich- him until the disease is already at an ad- becoming trapped in debt. It provides est: in a large city in northern Thailand, vanced stage. He estimates 90% of poor the security that permits poor and very the poorest quintile spent 21% of house- people discontinue treatment for TB poor households to invest in economic hold income on health, while the richest once started, and often start and stop activities and human capital. By encour- spent only 2%.39 In South India the di- treatment several times because it is ex- aging a greater level of household-level rect cost of treatment and equipment for pensive and because they feel better. If risk-taking, it promotes economic diver- disabled people averages three months’ treatment has been aborted several times sification which can create more robust income. This does not include the indi- and the TB becomes resistant, the cost local economies, and reduce the degree rect costs to families and carers, or the becomes exorbitant.43 Further, as TB is to which whole communities are ex- opportunity costs of income foregone.40 highly contagious, other household posed to covariant risk. members are likely to become infected, Without adequate systems of social Ill-health and poverty are thus multiplying the costs. protection, particularly health care, the mutually reinforcing: the poor effects of any shock are intensified. are more vulnerable and less When violence isn’t prevented, Health shocks are highly significant resilient to illness and injury, and conflict isn’t resolved events in the lives and livelihoods of and the sick and injured are Violence, conflict, and the weakening of poor people. Ill-health and poverty are more likely to become poor. the rule of law drive individuals, house- mutually reinforcing: the poor are more holds and entire populations into pov- vulnerable and less resilient to illness erty, and act as a maintainer. At the and injury, and the sick and injured are When a household member requires micro level, case study research (and in- more likely to become poor. And, evi- care, this means the loss of income or vestigative journalism) provides evidence dence suggests, they are more likely to labour of an additional household mem- of the ways in which violence can drive stay poor. Ill-health – particularly of a ber. Household capacity to employ assets children onto the streets;44 force women household’s main income earner – is per- such as land and livestock effectively is into sex work; destroy the livelihood, so- haps the most common driver of chronic constrained. To try to maintain con- cial status and self-esteem of older poverty at the individual and household sumption levels, households compensate women who are punished for being level. In the context of overstretched, in- by selling off natural and physical assets, witches; and lead to girls who are accessible and, in many cases, non-exis- drawing down any financial savings, tak- ‘spoiled’ by rape being turned out of tent health care systems, the direct and ing on debt, pulling children out of their homes and made destitute. Much, indirect costs of treatment and care can school to work, and mobilising social if not most, violence is hidden away, but in themselves initiate rapid asset deple- support. The additional costs of medical there is a growing awareness of the tion, leading to chronic poverty. The expenses, especially in the case of many ways in which violence, or the story of Maymana and Mofizul (Box chronic illness, add to the household’s threat of violence, can force people into 4.7) illustrates the downward spiral that problems. chronic poverty and close off mecha- can follow a health problem. The critical issue here is whether the nisms by which they might escape. When the illness is chronic or termi- household falls below the threshold from Less dramatic breakdowns of the rule nal, the strain on a household’s financial which, assuming the individual recovers of law can also drive households into and human capital is very great, increas- sufficiently to return to work, a liveli- poverty. In Bangladesh, the way in ing the chances of it falling into a spiral hood can still be generated and the which both policemen and magistrates of debt and ill-being. Poverty doubly af- household can start to re-accumulate. can be ‘bought off’ means that wealthier fects the health of the poor. Not only are Some forms of employment, such as rick- people can impoverish poorer people by the poor most likely to suffer food inse- shaw pedalling, are inherently ‘unsus- mounting false legal cases against them curity and exposure to hazards in domes- tainable livelihoods’, mining the health and having them arrested and jailed.45 tic and work environments, but they are of the worker.41 A spiral of lost income, Such cases can be manufactured to scare Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 45 Black plate (45,1)

Why are people chronically poor? . 45 poor people off their land or settle dis- brickbreaking, begging. For children and in India in 2000–01, no longer able to putes in favour of the wealthy. The debts families dependent on garment wages, keep their businesses going in the face of incurred by taking on a court case can this began a spiral of asset depletion; the cheap imports following liberalisation.47 trap entire families in poverty. subsequent Memorandum of Under- ‘Last year, we had a spate of suicides At the macro level, the outbreak of a standing has brought some children back by power loom weavers in Sircilla, violent conflict impoverishes people di- into garment work with a provision for Andhra Pradesh. One example is the rectly, through the destruction of assets, part-time free education; the extent to case of Konda Kishtiah, 32 years old, impairment and death, and indirectly, by which this has interrupted a downward with dependent old parents, two chil- disrupting the functioning of markets and spiral for individual children or their dren. His wife died of tuberculosis governments. This induces rapid econom- families needs to be assessed. and despite being skilled, the shock of ic decline, impacting on the livelihoods of The consequences of international eco- a combination of factors such as mar- poor people and causing a breakdown in nomic change on the lives of the poor ket based fluctuations, globalisation, public services, so that ill-health and re- are dramatic – and all too often invisible subsidy withdrawal, withdrawal of duced ability to cope with shocks makes to decision makers. At other times an marketing support by the State, lack chronic poverty more likely. economic shock may operate indirectly, of alternative income earning oppor- The situation in the Solomon Islands is as in the suicides of powerloom weavers tunities, mounting debts, starvation illustrative (Box 4.5). A country in which there was evidence of progress in Box 4.5 Violent conflict and a collapse into poverty in the Solomon poverty reduction suddenly and rapidly Islands moved into reverse. Income poverty The Solomon Islands have struggled with inter-ethnic tension since independence soared and maternal and child mortality in 1978, although violent conflict only broke out in late 1998. About 20,000 people rates rose. Recent attempts to stop fur- were forced to abandon their homes. In June 2000 there was a coup, and violence ther violence seek to prevent conflict as a continued particularly in remote rural areas. driver becoming conflict as a maintainer The effects on the economy and the well-being of the people have been of chronic poverty. marked. Modest growth in national income has turned into severe decline. Net When markets collapse . . . inflows of foreign direct investment declined from US$33.8 million in 1997 to US$1.4 million in 2000; by 2001, there was a net outflow of US$5.1 million. Debt Economic shocks are powerful drivers. grew by almost 50% between 2000 and 2001. While aid is about 23% of gross The collapse of coffee prices (Box 4.6) national income, the debt burden is about 58%. has been catastrophic in countries such as Burundi, Ethiopia and Uganda. Small scale producers and casual labourers de- 15 pendent on the crop experience dramatic drops in income, take on extra debt and 10 GDP (% change) risk spiralling into chronic poverty. A second shock, ill-health or ‘unavoidable’ social expenses, such as funeral costs for 5 a family member, can drive the family to a point at which it cannot recover its ear- 0 lier position and rebuild its assets. The collapse of the Asian financial markets in the late 1990s generated large -5 amounts of transitory poverty, for exam- GDP per head ple South Korea’s poverty head count -10 leapt from around 3% to 25%, and may Total GDP no well have generated persistent poverty data among particular groups of people. In -15 1994 1995 1996 1997 1998 1999 2000 2001 poorer countries, mitigating measures, such as unemployment benefits and so- Source: Asian Development Bank, Asian Development Outlook. cial assistance, are rarely in place to deal with these shocks. The Solomons are far behind on the Millennium Development Goals. Particularly While many of these shocks have com- in remote areas, declining funding of essential social services has adversely plex chains of causality, occasionally affected health and education indicators, now the lowest in the region. Maternal they can be attributed to specific policy mortality rates are particularly high. New vulnerable groups are emerging such as measures.46 For example, the garment those displaced by the conflict, the unemployed, especially women, and youth. industry in Bangladesh was affected by More than two years later, Australia and a few other Pacific neighbours offered the US ‘Harkin Bill’ against child labour, to send in a small peacekeeping force. The Solomon Islands parliament approved which inadvertently led to thousands of the force on July 17, 2003. By the end of 2003, a measure of stability had been children, mostly girls, going into jobs restored. This may be the chance to prevent conflict as a driver from becoming that are less safe, lower paying, with conflict as a maintainer of chronic poverty. more limited prospects – sex work, Sources: ADB 2003; BBC 2003; World Bank 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 46 Black plate (46,1)

46 . Chapter 4

Box 4.6 The cost of a cup of coffee – market collapse and poverty traps Over the past five years, massive coffee overproduction has support to particular coffee-producing countries, and of the led to a collapse in prices, contributing to chronic poverty at pressures for hard currency that encourage indebted countries both the micro and macro level. to increase exports, the absence of governance has There are some 20–25 million small farmers dependent on undermined the livelihoods of millions of people. the global coffee market, living in about 50 countries of Latin America, Asia-Pacific and Africa. In Ethiopia, more than Selected Share of world Dependence on 700,000 smallholders are involved in coffee production, and countries production (%)1 coffee exports (%)2 the livelihoods of a further 15 million rely in part on the coffee Americas 65 economy. For these small farmers, the collapse in coffee Brazil 33 prices – sometimes to a fraction of the production cost – has meant the collapse of livelihoods, increased food insecurity, Colombia 12 17 asset depletion, resource degradation, and a decimated ability Guatamala 4 21 to cope with shocks. For many children, it has meant leaving Honduras 3 25 school, as coffee production was the only way their families Mexico 3 could earn cash for fees. Diversification and entry into niche markets are risky, long- El Salvador 2 26 term strategies. With few livelihood alternatives, the most Nicaragua 1 30 vulnerable farmers stay in the coffee market. Others, along Others 6 with under- and unemployed coffee plantation labourers, migrate in search of a new livelihood. Asia-Paciﬁc 22 At the macro level, the countries hit hardest by the price Vietnam 13 collapse are mostly Least Developed Countries, many Indonesia 5 desperately poor. Over the last five years, Uganda lost about India 3 half the amount it has received as debt relief under HIPC. Others 2 Ethiopia lost half its annual export earnings over a two year period. Of the 36 countries facing food emergencies identified Africa 13 by the FAO in June 2003, four are Central American countries Uganda 4 55 experiencing food insecurity because of the international Coˆte d’Ivoire 3 coffee crisis, and six are African countries heavily dependent on coffee exports. The economies of some of the poorest Ethiopia 2 67 countries in the world – including much of Eastern Africa and Kenya <1 14 Central America – are highly dependent on exports of coffee Tanzania <1 16 from small farms (see table). Burundi <1 80 While a range of factors has contributed to the coffee crisis, Rwanda <1 43 the main problem is a lack of governance since the International Coffee Agreement broke down in 1989. The Others 2 global coffee market is extraordinarily unequal: an average 1June 2002–May 2003; www.ico.org farmer has about 15 bags of coffee to sell each year, while 21998; FAO/ILO in Oxfam 2001. each of the biggest coffee roasters buys about 15 million. In Sources: Charveriat 2001; Lines and Tickell 2003; the context of commodity market speculation, of IFI production Barrett et al. 2002; FAO/GIEWS 2003.

and no social safety nets, created a sit- Understanding chronic level, ineffective policies meant that the uation where he and many other poverty and taking action lack of broad based growth, especially in power loom weavers found the situa- Maintainers and drivers of chronic pov- agriculture and the rural economy, re- tion so hopeless that they killed them- erty overlap and reinforce each other. duced opportunities to work, earn and selves and their families.’ This can be seen clearly by unpicking the accumulate. Poor governance meant that Although the prediction of speciﬁc mar- story of Maymana and Moﬁzul, two inheritance laws were ignored and social ket and economic collapses is problem- chronically poor people in Bangladesh, protection was not provided. Civil soci- atic, there is a widespread consensus that whose experience is described in greater ety, in the form of NGOs, bypassed this globalisation means that such shocks are detail in Box 4.7. household. likely to be more frequent, more rapid At the micro-level, their poverty is ex- Internationally, the protection that and on a larger scale than in the past.48 plained by the ill-health of Maymana’s wealthy nations provide to their farmers But mechanisms for international gover- husband and by not being able to access (US$320 billion per annum) and textile nance to address social impacts of glob- affordable and reliable health services. industries reduces growth in Bangladesh alisation, not least on chronic poverty, Moving up a layer, social exclusion at and the capacity of its government to in- have yet to be developed. community level permitted Maymana’s crease public revenues. OECD resistance father-in-law to seize her land and fur- to proposals from Bangladesh at Cancu´ n ther deplete her assets. At the national to open up the global market in unskilled Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 47 Black plate (47,1)

Why are people chronically poor? . 47

Box 4.7 Maymana and Moﬁzul’s Story Maymana and Mofizul recounted their life story during a a tiny and irregular income and a sickly child. Although number of long conversations over a period of 12 months Mofizul was only 12 and often sick, he sometimes found from October 1999. Maymana and Mofizul were selected as casual employment at a local timber mill. At a daily rate of 10 one of 42 poor households interviewed as part of a study on taka (US$0.20), however, his income did not make a financial behaviours and preferences of the poor. For a year, significant difference. Bangladeshi researchers visited them every fortnight and The situation declined further when Maymana’s father-in-law collected information about their financial and economic took control of the household’s agricultural plot. Maymana activities, as well as gathering their life history and exploring began to borrow, glean and beg for food. Fortunately her their perceptions as to how and why they had declined into married daughters, wider family, neighbours and the mosque poverty, their strategies for addressing the difficulties they committee provided some support so she and Mofizul face, and the persistent nature of their poverty. survived. At the end of 1999, despite threats and warnings Maymana and Mofizul live in a village about 30 kilometres against doing so, Maymana took her father-in-law to the outside the city of Mymensingh. It is relatively favoured in village court (shalish) for the return of Hafeez’s land to her Bangladeshi terms. The area is flat, fertile and densely and her son. Under Bangladeshi law she has rights to the populated, and it does not experience severe flooding. land, but the shalish ruled against her. Following the spirit of Agricultural productivity has been rising with the spread of compromise that often guides the shalish, her father-in-law ‘green revolution’ and fish farming technologies. The village is told the court that he would pay for any medical expenses near a main road, so the economy is fairly diversified and arising from Mofizul having his back treated. He has only services are relatively accessible. In addition, a high density of partly honoured this promise. NGOs operates in the area, including the mega-NGOs BRAC and Proshika, as well as several smaller ones. The widely Enduring poverty acclaimed Grameen Bank also has a large presence in the area. In October 1999, Maymana and Mofizul occupied a one room house with mud walls and an old iron roof. They also had a Sliding into poverty small kitchen hut with mud walls and plastic sheeting on the In the early 1990s, the household was made up of Maymana, roof. These huts and 0.06 acres of homestead land were their her husband Hafeez, two daughters and a son. Hafeez owned only assets. They had no furniture, equipment or livestock, not three rickshaws that he hired out on a daily basis and an acre even chickens, and only a few cooking utensils. Their huts of paddy land, thus the household had a reasonably secure stood at the back of a number of better-constructed buildings income and an asset base to fall back on in hard times. In belonging to an uncle. Maymana’s words, life was balo (all right/OK). However, as The household of two met its livelihood needs from a the two daughters approached their teens there was the variety of sources including casual work, gleaning, borrowing, expense of dowry to consider, and the son Mofizul had a begging and receiving charity, but it was unable to acquire or growth on his back and was often unwell. accumulate any significant financial or physical capital. Their At this time Hafeez developed a cough and complained of a human capital remained low, with poor health and no new painful throat. After purchasing ineffective medicines from a skills acquired, and, having angered Maymana’s father-in-law ‘pharmacist’ in the bazaar, he visited the nearby government- and taken loans of grain and cash that she was unable to run health centre, where staff asked for bribes and were repay, their social networks were weakening. generally uninterested in his illness. Next he went to a private Maymana sought paid work in others’ homes, but found it ‘doctor’ in a nearby town, who recommended costly increasingly difficult as she was ageing, nearly deaf and often medicines. When these also failed to work, he referred unwell. Maymana did not know her age but was probably in Hafeez to a colleague in Mymensingh. In order to meet these her late 40s. She had only two years schooling and was medical bills, a rickshaw was sold. As Hafeez’s condition illiterate. Whenever possible Maymana gleaned rice. worsened and further X-rays and tests were required, a Mofizul was 13 with no education, as is the norm for second rickshaw had to be sold. The weekly income children with an impairment in Bangladesh, and being plummeted, the family reduced its consumption levels, and old ‘disabled’ was part of his social identity. But despite his youth, clothes and utensils were no longer replaced. disability, ill-health and lack of education Mofizul was As Hafeez got sicker the elder daughter, concerned that her determined to find employment. As he got older his daily rate family would not be able to provide for her dowry, acquired a for casual labour rose to 30 taka/day – half the adult male kid, fattened it, sold it and repeated this cycle. In this way she rate. However, employment was irregular. Once the police saved for her own dowry. Her younger sister adopted the shut down the mill for a month, claiming that it was sawing same strategy. Much to Maymana’s relief, male family logs taken from a protected area. members arranged marriages for the girls, with some Sometimes the household received gifts or charity. During involvement from Hafeez. Eid the mosque committee gave them 150 taka (four days’ By now Hafeez was confined to the house and there were pay for Mofizul), a sari and meat. Maymana also borrowed no more rickshaws to hire out. The household was dependent and begged for food and money. Distinguishing between these on rice produced from its small plot of land and on Maymana strategies is not easy: during the year Maymana arranged finding occasional work as domestic help. When Hafeez died several loans from family and neighbours that she was unable in 1998, Maymana was in despair. She was now a widow with to repay – loans gradually converted into ‘gifts’. By October Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 48 Black plate (48,1)

48 . Chapter 4

2000 she had borrowed 500 taka from one daughter and 36 kg Explaining poverty of rice from the other daughter, a son-in-law, and a neighbour. Maymana identified three main factors to explain her poverty. At It was unclear how this could be paid back. the heart of the explanation was the prolonged illness and In October 1999, Maymana held a Vulnerable Groups eventual death of her husband, which led to a dramatic decline Development (VGD) card entitling her to 30 kg of wheat each in income, a rise in expenditure and the sale of productive month. This is World Food Programme grain provided to female- assets. Second, the seizure of her husband’s land by her father- headed households identified by local government as being in-law undermined the household’s ability to produce at least vulnerable to hunger. However, Maymana only ever received some food each year. Finally, there was the composition of her 7.5 kg and subsequently had to return the card to the councillor. household – two daughters needing dowries and a disabled The reasons for this were complicated, but were related to the son. These explanations, however, were more broadly councillor belonging to a different political party than the uncle in understood and explained by her as Allah’r ichcha – God’s will. whose compound she lives. The uncle suspected that the Maymana and Mofizul are not poor because of any lack of councillor had ulterior motives and was trying to get wider family action on their part. Their agency may be severely members to change their political allegiances. constrained by a host of structural factors but they are Despite these difficulties, Maymana reported that 2000 had constantly seeking out ways of improving their position – they been much better than the previous year, as her son’s may be down but they refuse to be out. earnings meant that they could more often substitute borrowing for begging. Table 4.2 The inﬂuence of state, market, society and family on Maymana’s and Moziful’s prospects

Sectors What support has this sector provided What constraints has this sector placed on their welfare and in for them? what ways has it failed them?

State . Vulnerable Group Development card . Card withdrawn

. Basic health services . Poor quality, and has failed to regulate the quality of private health service providers

. Primary education . Only taken up by Maymana for two years

. Law and order . Failed to uphold Maymana’s rights to land inheritance

Market . Labour market . Provides Moﬁzul with poorly paid, casual work. Maymana unable to get work

. Product market . Used by Maymana’s daughters to sell goats for dowries

. Insurance . No health or life insurance available to manage Hafeez’s decline

. Health Services . Provided services to Hafeez that did little for his health but dramatically depleted household assets

Society . Charity . Neighbours give food when Maymana begs and permit her to glean from their land

. Mosque Committee . Provides gifts at Eid

. Informal loans . Neighbours provide loans of money and grain that may turn into gifts

. Village court . Cheated Maymana out of her land rights and greatly reduced her asset base

. NGOs . Do not provide support to Maymana – not a suitable client

Family . Father-in-law . Seized her land, greatly reduced her asset base, does not buy health care for Moﬁzul

. Daughters and sons-in-law . Provide loans of food and money that may not be repaid

. Uncle . Provides physical security (as the household is part of the uncle’s compound), food loans and gifts. Blocked Maymana from using her VGD card and discourages her from begging

. Maymana’s father (mother is dead) . Unable to provide support as he is old, sick and poor. Maymana wishes she could help him

Source: Hulme 2002 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P14E-01.3D.3D Folio: 49 Black plate (49,1)

Why are people chronically poor? . 49 labour also works against Maymana and that is pro-poor in terms of both the community-based savings and insurance Mofizul. The USA’s abrogation of the labour market and public revenues; schemes can foster escape from chronic Kyoto Treaty on global warming makes . persuading NGOs to extend their serv- poverty, when everyone involved is it likely that opportunities will be further ices to the chronically poor; chronically, and often severely, poor. constrained in the future as land is lost . identification of strategies that will re- Building social relationships upwards and ‘natural’ hazards become more duce social exclusion; may help people to survive but, without frequent. . lobbying aid agencies to focus on as- significant improvements in social solid- Historically, the destruction of Ben- sisting the chronically poor by financ- arity, this will have little chance of inter- gal’s textile industry by the British, and ing innovative programmes (e.g. NGO rupting the processes that drive and the creation of an agrarian social struc- ultra-poor programmes, projects in maintain chronic poverty. Governments ture based on clientelism and the parti- less favoured areas and non-contribu- and the international system clearly have tioning of India (and the subsequent war tory pensions); duties to fulfil. of Liberation) have left a legacy of im- . contributing to efforts to improve The length of time that different policy poverishment in Bangladesh. governance. interventions take before they bear fruit While many factors may explain differs radically. The removal of socio- chronic poverty, only some can be cultural barriers to citizenship, services changed. Action needs to focus on issues Conclusion and improved governance are vital for that may be influenced. Internationally, The drivers and maintainers of chronic chronic poverty reduction, but difficult efforts to reform global governance, in- poverty operate at every level. Processes to challenge in the short-term, especially ternational trade, and environments, that affect the individual and household, with limited resources. Other macro pol- need vigorous support if the chronically as well as communities, countries and icy changes, which are crucial for the poor are not to be further marginalised the international system, need to be tar- medium to long-term, are unlikely to and impoverished by globalisation. The geted if the downward spirals that result deal with the chronic poverty that indi- international community must be kept to in chronic poverty are to be interrupted. vidual and households are experiencing its commitments on resource flows and At the individual and household level now. What is needed immediately is in- on the right to development. In Bangla- this means managing vulnerability. Peo- creased investment in health and educa- desh, there is a significant agenda for ple in chronic poverty try to cope with tion matched with asset transfers and those who seek to reduce chronic vulnerability and keep afloat through social insurance, so that chronically poor poverty: livelihood diversification and building people can move towards being able to . improved health services and social social relationships, both horizontally always access services and make choices protection for the chronically poor; and vertically. But there are obvious lim- about their own futures, without com- . the promotion of broad based growth its to the extent that things like promising short or long-term well-being.

Notes 1. Worryingly, global relative poverty may have unpublished working paper that had not been 28. Castan˜ eda and Aldaz-Carroll 1999. greatly risen over the 19th and 20th centuries, independently reviewed by a scientific journal – 29. MHHDC 2000; Ray 1999; Watkins 2001. because of increased inequality. reveals more about the political economy of 30. HelpAge International, State of the World’s 2. Maddison 2001. why people stay poor than it does about the Older People 2002: 11. 3. See Corcoran 1995 for a discussion for the USA technical evidence. An ideological drive to 31. Top priority countries fail on at least 3 goals, or and/or take a walk around any US or UK city promote neo-liberal ideas allied to a half of those for which they have 3 data points, centre late at night to see scores of homeless technocratic belief that ‘average’ policies should or, if data is only available on 2 and they are people sleeping on the street. be pursued by all countries. top priority in both. ‘High Priority’ countries 4. McKay 2004. 8. Atkinson and Brandolini 2001. are top/high for at least 3 goals, top for 2, or 5. Countries were grouped according to their 9. Timmer 1997. top/high for at least half the goals for which average rate of economic growth (with those 10. For a review of these concerns see McKay they have data (3 data points) or, if they only whose rate was less than 2% identified as 2004. have data for 2 goals and they are top priority ‘collapses’) over the 1980s and 1990s. Three 11. McKay 2004. in both. HDR 2003 page 347. ways of measuring performance in the 12. McKay 2004. 32. Stiglitz 2002. reduction of infant mortality rates (absolute 13. See introduction to Chapter One for a discussion. 33. Baulch 2003. numbers, relative to all developing countries, 14. Daru and Churchill 2003. 34. See Chambers 1983 on ‘triggers’; Sinha, Lipton and conditional on initial levels) were 15. Daru and Churchill 2003. et al. 1999 on ‘damaging fluctuations’; and a correlated with the country groups for each 16. Wood, 2003: 456. wealth of work on sustainable livelihoods on decade. See Anderson and Morrissey 2003 for a 17. de Haan and Lipton, 1998: 13. ‘vulnerability context’. detailed discussion of the performance 18. Jalan and Ravallion 1997. 35. CRED (Centre for Research on the measures. For those countries with adequate 19. Jalan and Ravallion 1997. Epidemiology of Disasters) http://www.cred.be data, there are almost no significant differences 20. At the extreme end of this, there are those in Skoufias 2003. between the average values of IMR reduction (Banfield 1968) that believe that much if not 36. Cammack et al. 2003; Gaiha and Imai 2003. performance of the four growth quartiles, with most poverty is based upon the ‘innate’ 37. Rural Nepal; Bangladesh; rural tribal Madhya two exceptions. First, the quartile of countries characteristics of the poor, sometimes called the Pradesh, India; urban northern Thailand; with the lowest economic growth rates in the ‘underclass’. This approach has highly racist Vietnam; Kuba District, Azerbaijan; Egypt; The 1990s had a statistically significant lower and classist overtones. Within this view, any Gambia; rural Sierra Leone. average value of the conditional measure of attempt to eradicate or alleviate poverty among 38. Fabricant, Kamara and Mills 1999. performance. Second, countries that the ‘underclass’ is doomed to fail. 39. Pannarunothai and Mills 1997. experienced an economic collapse performed 21. Kabeer 2000. 40. Harriss-White 2001. significantly worse on all three measures in the 22. Originally put forward by Oscar Lewis in the 41. Begum and Sen 2003; see also Kabir 2003. 1980s and significantly worse on the late 1950s. 42. Afsaw 2003. conditionality measure in the 1990s. 23. Based on Sen’s capabilities and freedoms 43. Ruthven and Kumar 2003. 6. Dollar and Kraay 2001. approaches; see for example Sen 1999. 44. Conticini 2003. 7. Indeed, it can be argued that the Dollar and 24. Marshall 2004. 45. Landell-Mills 2003. Kraay work – reported as a virtual law of 25. Harper, Marcus and Moore 2003. 46. See Oxfam 2002. poverty reduction by influential media such as 26. Harper, Marcus and Moore 2003.. 47. Mehta and Shah 2002. the Economist at a time when it was an 27. Marshall 2004. 48. Stiglitz 2002; Gray 2002. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 50 Black plate (50,1)

There is now an unprecedented window of opportunity for pushing forward with the eradication of poverty: . the MDGs provide an agreed frame- What should be work for action; . the ‘Monterrey consensus’ and the done about proposal for an International Finance chronic poverty? Facility (IFF) indicate a new serious- 5 ness in the provision of resources; . we are a decade on from the World Summit on Social Development in Co- There are many policies that are potentially beneﬁcial for the poor AND penhagen when heads of state signed for the chronically poor. But people living in chronic poverty are not ‘just up to the global commitment to erad- like the poor but a little bit further down the poverty spectrum’. Overcom- icate poverty; and ing chronic poverty requires policy makers to reorder their priorities and . globally, people are recognising the in- set their sights higher than the current consensus on poverty reduction terconnectedness of the world and the policy. responsibilities that this creates. Development strategy needs to move beyond the bounds of its present Analysis of chronic poverty points to the emphasis on economic growth – hundreds of millions of people are born need to reduce vulnerabilities by invest- poor and die poor in the midst of increasing wealth. Chronically poor ing in human capacity, supporting the people need more than ‘opportunities’ to improve their situation. They acquisition of assets and strengthening need targeted support and protection, and political action that confronts people’s capacity to assert their rights. In exclusion. If policy is to open the door to genuine development for the short-term this is challenging and chronically poor people, it must address the inequality, discrimination and costly, but it is essential to achieve and exploitation that drive and maintain chronic poverty. sustain MDG gains.

Action on chronic poverty needs a framework to: Security first Prioritise livelihood security: A much greater emphasis is needed on preventing and mitigating the shocks and insecurities that create and Prioritising livelihood maintain chronic poverty. This is not only about providing recovery assis- security for chronically tance but also about giving chronically poor people a secure position poor people from which to seize opportunities and demand their rights. Thus, social The current development consensus, ex- protection policies are of great importance. pressed in the World Development Re- port 2000–01,1 argues that because of Ensure chronically poor people can take up opportunities: It is cru- their interconnectedness there is ‘no hier- cial both to promote broad-based growth and to redistribute material and archy’ between opportunity, empower- human assets, so that chronically poor people are in a position to take up ment and (livelihood) security. And then, opportunities and can better cope with shocks. apparently by chance, opportunity always comes first and security always Take empowerment seriously: Policy must move beyond the cosy rhet- comes last. A related invisible hand oric of participatory approaches, decentralisation and theoretical ap- seems to guide the drafting of PRSPs: proaches on rights. It needs to address the difficult political process of economic policies and growth rates take challenging the layers of discrimination that keep people trapped in pov- priority. This covert prioritisation of eco- erty nomic opportunity may or may not be appropriate for the poor. For the Recognise obligations to provide resources: Chronic poverty cannot chronically poor, however, one thing is be seriously reduced without real transfers of resources and sustained, clear – such an emphasis is incorrect. predictable finance. The political indifference to meeting national and Livelihood security must come first. In- international obligations on poverty eradication needs to be challenged, security dominates their lives, constrain- and ways found to foster social solidarity across households, commun- ing their ability to take advantage of ities and nations. opportunities or risk pushing for empowerment. Chronic insecurity means The need for policy change must not mask the fact that it is the chronic- that the long-term poor engage in eco- ally poor themselves who are the leading actors in overcoming their own nomic activities that destroy their chronic poverty. As Maymana and Mofizul reveal (Box 4.7), poor people human capital,2 and social relationships are not passively waiting for assistance: they are actively working to that block off opportunities for asset maintain and improve their circumstances. Most of the action to tackle accumulation.3 chronic poverty is at individual, household and community levels. The Different poverty reduction strategies benchmark of effective policy is whether it enables people to get a better are appropriate for different mixes of return on the efforts that they are already making. chronic and transitory poverty. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 51 Black plate (51,1)

What should be done about chronic poverty? . 51

. In a country where poverty is more transitory than chronic, where ‘the Box 5.1 Poverty reduction policies that assist the poor and the poor’ at any particular time have a chronically poor high probability of improving their po- . Pro-poor, broad-based economic growth sition, policies should focus predomi- . Peace-building and conflict prevention nantly on social safety nets that help . HIV/AIDS prevention (especially in India, China and the CIS) and greater access people to avoid descending into to retroviral treatment (in Africa) chronic poverty, rapidly return to a . Slowing down global warming non-poor status and reduce vulnerabil- . Strengthening national and international governance ity. This includes limited term unem- . Making trade fair (especially removing the obscene agricultural protectionism of ployment allowances, social grants, rich countries) workfare, micro-credit and new skills . Effectively managing national indebtedness (through debt relief and fiscal acquisition programmes. prudence) . In a country where a significant pro- . Improving the effectiveness of basic service delivery in the public and non-profit portion of the poor are chronically sectors poor, then policies to redistribute as- . Making markets work for all sets, direct investment toward basic physical infrastructure, reduce social opportunity and empowerment, allied to redistribution, reaching poor households exclusion (from employment, markets the agency of chronically poor people, and the poorest children. The gross im- and public institutions) and provide means that social protection policies and pact of pension incomes is estimated to long-term social security will be neces- expenditures are often directly produc- reduce poverty by 12.5%.11 sary if poverty is to be significantly tive. They permit people to spend more Given the large number of people who reduced. days labouring (because they are not ill). are chronically poor, a major re-orienta- Quite different national development They are often used for income generation of international thinking about so- strategies, roles for the state and forms tion and children’s education – the vari- cial protection policy is called for, and levels of international support are ous uses and impacts of old age pensions focusing on protecting breadwinners’ needed in the two cases. in South Africa are good examples6 – and carers’ incomes, and working to in- and they create the space for chronically The benchmark of effective poor people to choose between economic policy is whether it enables and social relationships (rather than hav- Social protection policies ing to take on ‘last resort’ options that people to get a better return mortgage their future opportunities). and expenditures are often on the efforts that they are Examples of the positive effects of directly productive. already making. social protection on growth and asset creation include the Maharashtra Em- ployment Guarantee Scheme and India’s Prioritising livelihood security entails Midday Meal Programme.7 Indeed, the crease their (and their children’s) assets, developing cost-effective social protec- most innovative schemes for helping the whether financial, physical or human. tion systems for the poor and chronically chronically poor escape poverty, such as This is not to deny the continued impor- poor, rather than the residual concept of BRAC’s Income Generation for Vulner- tance of preventing descent into poverty, ‘modular social safety nets’4 that has able Group Development and Ultra-poor but to add to the social protection agen- shaped contemporary policy. As the da a significant new and urgent dimen- chronically poor are predominantly both Prioritising livelihood sion – helping the chronically poor avoid dependent on their physical labour to security entails developing catastrophic situations leading to destitu- make a living, and exposed to high inci- cost-effective social tion, family breakdown and early death, dences of health problems, providing and providing social assistance to people them with the means to maintain their protection systems for the who cannot work enough but who are health is of over-riding importance as poor and chronically poor. nevertheless contributors to household both a goal in itself and a means to im- welfare through caring and other roles. proving their lives. For those chronically There is now significant evidence that poor who are not economically active – programmes,8 and Bonded Labourer social protection – in combination with sometimes, but not always,5 the old and Schemes in Nepal,9 conceptualise social other policies and interventions – can en- infirm, the severely disabled, the chronic- protection and income generation as able persistently poor people to escape ally ill, the stigmatised, those with full- integral, not as a choice between con- poverty. It can shore up consumption so time caring duties – then short-term assis- sumption/welfare or investment/growth potentially irreversible welfare effects tance is inadequate and mechanisms for expenditures. Transfers enhance effi- (reducing nutrition, avoiding essential longer term support must be established. ciency and growth when they reduce risk medical expenditures or withdrawing Recognising the need to prioritise the and excessive inequality that prevent children from school) do not occur. It security of the chronically poor is not a people participating in and contributing can prevent the erosion of savings and recipe for creating welfare dependency. to growth.10 The South African pension other assets, and help poor people avoid The interconnectedness of security, has been shown to be an effective tool of becoming trapped in debt.12 It also Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 52 Black plate (52,1)

52 . Chapter 5

There is now significant transmitted over lives and generations. Policies to interrupt the evidence that social protection Even a relatively short period of depriva- relationship between ill-health – in combination with other tion in childhood can harm child nutri- and poverty are essential to policies and interventions – can tion, health, education and aspirations, prevent the chronically poor enable persistently poor people with dramatic and irreversible conse- from becoming destitute and to quences for the long-term well-being of to escape poverty. both the child and her/his own permit exit from poverty. children.15 provides the security that permits very Sectoral policies should focus on the poor households to invest in economic Even a relatively short period most crucial aspects of child well-being: activities and human capital. By permit- of deprivation in childhood can . First, action to foster child health and ting a greater level of household risk- have irreversible consequences nutrition includes the promotion of taking, it facilitates economic diversifica- for the long-term well-being of greater food security; food supplemen- tion, making local economies more ro- both the child and her/his own tation; the promotion of later marriage bust and reducing the degree to which children. and childbearing, helping to prevent communities are exposed to covariant the intergenerational transmission of risk. There is emerging evidence that so- poor nutritional and health status; and cial protection measures can also achieve Access to basic services and household combating gender and other biases in a more gender neutral distribution of assets are crucial to children’s survival, child-feeding practices. benefits than other development protection and development. Needs are . Second, enhancing and equalising op- initiatives.13 well understood: adults without suffi- portunities for both child and adult There is an urgent need to synthesise cient time and assets cannot adequately education requires substantial financial experience in this field and see whether it nurture their children; children in remote investment; a wider environment that can form a base for innovative larger- areas often cannot attend school; states prioritises and enables this investment; scale programmes – by governments and/ without adequate resources cannot fi- an enabling social context, involving or the profit and non-profit sectors. nance education. Investments that allow public action (to promote girls’ educa- Knowledge about how to design strat- chronically poor people to take up option for example); and sustained ef- egies that permit chronically poor people portunities for development are key to forts to create skilled employment to move from partial dependence on wel- the interruption of child and intergenera- opportunities for youth. fare through to independent livelihoods tional poverty. At a minimum, this re- . Third, the negative effects of work in should soon be sufficient to develop ef- quires the development of adequate childhood can be countered through fective programmes during the coming adult labour markets; financing of state enhancing school quality and accessi- decade.14 provision of public services and social bility, particularly for girls; developing protection; and programmes that sup- adult education; more effective regula- port asset generation and retention. It tion of working conditions; and a Tackling childhood poverty can also require campaigns and legal ac- wide range of poverty reducing meas- Any attack on chronic poverty must in- tion to prevent discrimination against ures that dthe need for children to corporate an attack on present childhood particular children, households and work. Blanket policies regarding child poverty, and on the ways poverty is groups. work must be treated with caution. . Finally, policies that acknowledge the crucial role of good adult-provided care and nurture in child development.

Preventing and interrupting chronic poverty through health services Chronically poor people rely on their personal labour power and hence their health for survival: they have few other assets to utilise. In study after study, ill- health comes out as a major driver and maintainer of long-term poverty.16 Maintaining the health of breadwinners and carers in chronically poor households is absolutely critical, combined with social protection against the effects on the ability to labour of ill-health and other crises. Mounting evidence suggests This boy, like many others, is working to help his family survive, but missing out on the education that that the ill-health, morbidity, disability might give him better prospects. or premature mortality of adults, Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 53 Black plate (53,1)

What should be done about chronic poverty? . 53 particularly household breadwinners, through avoiding protracted illness local manufacture of generic medi- matter more to the economic viability of and death, and allow poor carers to cines, and increasing aid that permits households than was previously recog- maintain their own health while look- anti-retroviral drugs to be provided nised.17 Sick people are more likely to ing after others. free or at low cost, are means of mov- become poor, while poor people are . Linking health services to social pro- ing this agenda forward. more vulnerable to accidents, disease tection, so that where there are impov- . Raising awareness of the consequences and disability.18 Ill-health shocks can erishing effects these can be mitigated of the HIV/AIDS epidemic, and how it trap already resource-poor households through accessing an appropriate pro- might be minimised, in ‘late starter’ and individuals in poverty. tective scheme. This can be done uni- countries such as China, India and the versally, through health insurance, or former Soviet Union. If countries such There is a need to focus in a targeted way if there is a range of as these do not pursue effective and policy on the role that protective public policy instruments to high-profile campaigns to lower trans- redistribution can play in hand (the issue is to structure these to mission rates of HIV/AIDS then one of include the chronically poor). the greatest drivers of chronic poverty extending the opportunity of Special efforts are needed to tackle the will reshape poverty dynamics in these the poorest. ways in which HIV/AIDS is driving countries (see Box 5.2). households into long-lasting, deep pov- . Focusing greater resources on the de- erty. This involves: velopment of a vaccine against HIV/ Policies to interrupt the relationship . Making low-cost anti-retroviral treat- AIDS as a global public good. As most between ill-health and poverty are essen- ments available so that breadwinners, AIDS deaths have been in Africa, the tial to prevent the chronically poor from parents and caregivers can continue in resources put into finding a vaccine or becoming destitute and to permit exit their roles for as long as possible. Per- cure have been much lower than from poverty. There are two main as- suading pharmaceutical companies to would have been the case had these pects to this: preventing ill-health lower prices, facilitating the import or deaths been in high-income countries. through better environmental and occupational health, greater health awareness, and better infant, child and Box 5.2 Lesson learned? HIV/AIDS in India and China maternal nutrition and care; and, preventing the impoverishing effects of ill- The significance of HIV/AIDS for households and states in much of sub-Saharan health through better and more accessi- Africa, in terms of its importance in driving and maintaining chronic, severe and ble and affordable curative care, drug multidimensional poverty and deprivation, is overwhelming. While the pandemic is availability and insurance. This is espe- nowhere near under control, there are some positive trends. In Uganda, for cially important for breadwinners and example, sustained and large-scale prevention efforts focused on young people carers as their ill-health can set off cycles fostering significant changes in sexual behaviour, and political commitment to of asset depletion that have irreversible reducing stigma and providing treatment, have combined to lead to a marked consequences for household well-being. decline in infection rates over the past decade. Outside Africa, in Brazil, HIV/AIDS While most countries and the interna- death rates have been cut by half and hospitalisation by 80% by making generic tional community already focus signifi- anti-retroviral drugs free to all who need them. cant attention on the former (but In other parts of the world – notably the populous countries of India, China and arguably insufficient attention on occu- the Russian Federation, already battling huge poverty problems – HIV/AIDS rates pational health, nutrition or reproductive are increasing rapidly, and not only among high risk groups. Low overall health), there is little focus on combating prevalence rates mask huge regional differences. It is crucial that Asian and the impoverishing effects of ill-health. European states learn from the successes and failures experienced in Africa, and This demands a focus on curative serv- that the international community – including drugs companies – facilitate this ices. In particular: process. . Making public curative health services China presently has about one million people with HIV/AIDS. UNAIDS expect more accessible and affordable to the the incidence of HIV to soar in the context of ever-widening socio-economic chronically poor through reducing the disparities and massive amounts of migration. Thus prevalence rates are estimated direct costs (fees, medication) and to rise tenfold by the end of the decade. In 2001, China launched a five-year AIDS transaction costs (travel, time, food action plan, signalling a growing recognition of the huge task at hand. expenses) of treatment. This would After South Africa, India has the most people living with HIV/AIDS of any also foster a reduction in the number country – an estimated 3.97 million as of the end of 2001, and rising. If HIV/AIDS of chronically poor people delaying is not brought under control, it is likely to undermine progress made in reducing seeking treatment, or withdrawing poverty, particularly in the southern states. In July 2003, a National Parliamentary early from treatment programmes. Convention on HIV/AIDS was convened, in which over 1000 political leaders from . Strengthening curative health services mayors to ministers took part. The Executive Director of UNAIDS described the (public and private), so that they can event as ‘historic’: cope with the major killer and impov- ‘Never before, in any nation of the world, has there been such a large and erishing diseases (such as malaria and committed gathering of the leaders from every level of decision-making, TB). This would permit breadwinners dedicated to the common cause of fighting AIDS.’ to get back to work more quickly Sources: UNAIDS 2002; Joint UNAIDS/Parliamentary Forum on AIDS 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 54 Black plate (54,1)

54 . Chapter 5

Opportunity is not they depend on, and prioritising the rapid poverty and chronic poverty reduc- enough: growth, type of trade that increases employ- tion achievements have derived in part ment, improves working conditions and from policies involving the redistribution inequality and wage levels. This may require evolving of land, and public expenditures on redistribution a different balance between public regu- health and education that have been Enabling chronically poor people to en- lation and the private sector’s (espe- skewed towards the poorest and thus are gage in economic activity, and to reduce cially the informal sector’s) freedom to redistributive.23 The challenge is to gen- their unemployment, underemployment determine working conditions. This erate the political commitment to imple- and low-productivity work, is a vital presents significant policy dilemmas in ment land and other asset reforms, since step on the road out of chronic poverty. cases where such measures might under- such policies can dramatically improve Economic growth can directly in- mine fragile growth prospects, but the economic opportunities for chronic- crease incomes and assets, and expand should be easier where growth is more ally poor people. the revenues available to government to robust. Avoiding undermining the labour In a significant number of countries, provide services and promote progres- productivity of the chronically poor is a growth is likely to remain low in the sive social change.19 However, the qual- major challenge in today’s globalised short to medium-term (at least) because ity of economic growth is as important economy. of structural factors and severe gover- for the chronically poor as is the rate of nance problems. In addition to the growth: growth must be broad-based so Confronting deeply engrained humanitarian arguments, there are two that increased demand for their labour, attitudes, social structures and reasons why the international commun- goods and services occurs and so that economic interests that deny ity should not abandon such countries. they are able to improve their productiv- chronically poor people their First, they require forms of support that ity. And, given the evidence that the rights presents difficult challenges protect poor populations (many of poorest 20% of the population benefit but should not deter action. whom are chronically poor) from pre- less from economic growth than do ventable deaths and capability-depleting others,20 it must be supported by other experiences (malnutrition, ill-health, im- actions, especially those that increase There is substantial evidence that pov- pairment), and to maintain their capaci- livelihood security. Middle-income coun- erty reduction is achieved most rapidly, ties to seize opportunities when (and if) tries such as Brazil and South Africa and is most likely to reach the poorest, growth returns.24 Second, effective inter- have managed to grow their economies when income and asset inequality are at national transfers require relationships over the years, but millions remain trap- modest levels. It is therefore crucial to of trust, and when regime change occurs ped in poverty. encourage growth that does not rapidly these are most likely to be rapidly estab- Patterns of growth need to be fostered increase inequality.21 There is a need to lished if there are pre-existing relation- that make use of the labour of the poor- focus policy on the role that redistribu- ships to build on. est, expand demand for the services and tion can play in extending the opportu- The distinct geographies of chronic goods they produce, and enable them to nity of the poorest. The massive declines poverty in most countries point to the increase their labour productivity. In in chronic poverty in South Korea22 and need for growth policies to have a strong many economies, this means focusing Taiwan were based on economic growth regional dimension. Much depends on on and investing in agricultural and cas- following effective land reforms that dra- the nature of the factors that discourage ual labourers and the labour markets matically reduced inequality. Malaysia’s growth and underpin high levels of chronic poverty in a region. Where these are based on a lack of connectivity then major investments in infrastructure – through foreign aid or public funds lev- eraging in private investment – can provide the lead. Malaysia’s land settlement schemes and industrial plans led to opportunity being relatively broadly distributed across peninsular Malaysia and a rapid assault on chronic poverty across rural areas.25 In Bangladesh, the aid-financed Jamuna Bridge has transformed the patterns of opportunity available to the poor and chronically poor on the west side of the river.26 In other contexts, migration may be the mechanism by which people seize opportunities, but this can have both beneficial and adverse impacts on the chronically poor.27 Ac- cessible services that facilitate the transfer and use of remittance monies can Growth and prosperity exist side by side with chronic poverty help.28 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 55 Black plate (55,1)

What should be done about chronic poverty? . 55

Empowerment: Box 5.3 Rojamma and the campaign against domestic violence and making rights real drunkenness Everyone has the right to a standard When Rojamma, from Andhra Pradesh, attended a literacy class, she read a story of living adequate for the health and which described a life very like her own. A poor woman was struggling to make well-being of himself and of his fam- ends meet. Whatever her husband earned, he spent on liquor, and then, drunk ily, including food, clothing, housing and violent, he attacked her because she had no food to give him. Unable to stand and medical care . . . Everyone is enti- the continuing violence, the woman went from house to house in her village, to tled to a social and international order find other women who had the same story to tell. They got together and decided in which the rights and freedoms set to picket liquor shops and stop liquor being sold. Their husbands then would have forth in this Declaration can be fully no liquor to drink, and the money they earned would be saved. realised. Rojamma was inspired by the story, collected her friends together, and began to – Articles 25 and 28 of the Universal picket liquor shops. The campaign spread. In village after village, women refused Declaration of Human Rights.29 to allow their husbands to squander money on liquor. And, they succeeded. The sale of liquor was banned by the government in Andhra Pradesh, savings went up, Issues of livelihood security and econom- violence went down, and lives began to improve. ic growth fit relatively easily into debates Source: Butalia 1998. about policy. Empowerment is an issue that has quite different dimensions. Ex- attainment of other goals. A rights-based and evaluation, to engender changes in isting patterns of power relations are the approach means that poverty is not just social relations. The work of CPRC re- outcome of deeply embedded historical regarded as morally unacceptable, but searchers32 indicates that, at best, such processes and social structures. Interven- that clear duties are identified that differ- approaches achieve mixed results in ing in such social and political relation- ent institutions must fulfil in order to terms of empowering the poorest. While ships is about engagement with long- combat poverty.31 This is not purely an they create a political space for previ- term social change, not simply about argument for social justice (an important ously ‘unheard’ voices to engage with ‘policy decisions’. Unfortunately, where goal in its own right) but for the eco- decision-making, they can often re- empowerment is most needed – for ex- nomic logic of securing rights. A rights- inforce pre-existing forms of social ample where repressive regimes operate based approach is not anti-growth; on stratification.33 or agrarian relations are quasi-feudal – it the contrary, many, if not most, rights At present the CPRC partnership is exis also most difficult to foster. are pro-growth – when people are prop- ploring the degree to which concepts Many of the debates around empower- erly fed, healthy and educated they can such as ‘political space’,34 ‘social solidar- ment in recent years have focused on contribute more to their local and na- ity’35 and ‘social energy’36 might aid the rights-based approaches to development tional economies and raise global levels understanding of how to promote rights and poverty reduction. Empowerment is of demand. and mobilise poor and non-poor people particularly important for the chronic- But this still leaves the problem of to support such efforts. The rhetoric of ally poor as they are the most likely to how to promote empowerment. Much support for the poor and chronically be denied their rights and the most prone faith has been put in the capacity of par- poor is easy to generate – but the impor- to discrimination. The big question ticipatory approaches – to gather infor- tant issue is moving beyond that into ac- around the promotion of empowerment mation, plan implementing programmes tion. MDG Goal 8 provides a classic and rights is not so much about ‘what?’ as ‘how?’. Confronting deeply-engrained attitudes, social structures and economic interests that deny chronically poor people their rights, especially when they are embedded in state institutions, private businesses and local communities, can appear both hopeless and very dangerous for the chronically poor.30 They may be ‘punished’ or ‘taught a lesson’ by those whom they challenge when they seek their rights. Occasionally, though, there are advances, as illustrated by Rojamma and her associates who successfully con- fronted domestic violence (Box 5.3). The more such events happen and succeed, and the more they are publicised and lauded, the more likely it is that they will take place again in the future. Empowerment and the achievement of rights are direct goals for development, Holding powerful institutions to account: Bangladeshi disabled people challenge the processes that keep but they also contribute to the them poor. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 56 Black plate (56,1)

56 . Chapter 5 example of this dilemma (Box 5.4). The The national agenda: Demand is stimulated both by the citizens of rich countries are content to reaching chronically poor knowledge that the service exists and on let their leaders make big promises about how to use it, and by knowing that using poverty reduction, but they lack the so- people it is a right. However, changing demand cial solidarity with the poor in the poor- for services may involve complex eco- est countries, and are not prepared to Delivering basic services nomic and cultural calculations and take on vested interests in their own Improving the security and enabling power relations: examples would be the countries to demand that their leaders chronically poor people to take up op- demand for family spacing/planning honour such promises. portunities requires access to basic serv- services, and for education, both contin- ices (education, health, water and gent on logical ‘cost-benefit analyses’ sanitation, social assistance). To achieve and subject to the influence of powerful A rights-based approach is this three interlinked issues must be ad- individuals in households and commun- not anti-growth; on the dressed. First, access barriers need to be ities who determine that boys should be contrary, many, if not most, reduced. Second, the quality of service prioritised over girls, or that women rights are pro-growth. outputs needs to be improved so they are should ideally bear a certain number of capable of assisting people out of long- children, or that contraception is term poverty. Finally, attitudes and per- immoral. ceptions of the value of services, and Neither access nor demand will im- Discussions about empowerment and therefore the demand for services, need prove outcomes if quality is not rights are often naively critical of elites. to be fostered among the chronically addressed. The degree of quality en- Analytical frameworks and practical ac- poor. hancement required to make access tion must look at the real world settings Some access barriers are relatively sim- worthwhile and contribute to stimulat- in which pro-poor elites can play key ple to bring down – formal charges on ing demand is substantial in many situa- roles in promoting the rights of disad- basic services can be abolished, although tions. Encouragingly, there is much work vantaged groups. How can energetic this is not the case for informal charges. in this area, through sectoral pro- leaderships and mass action be mobilised Others may be more difficult, due to re- grammes. The quality of private provi- both at the ‘top’ of the social spectrum mote location, unclear land or residential sion, on which the chronically poor and at the ‘grassroots’, in order to force rights, disability, or discrimination. The often depend, is a particular concern. the rights of chronically poor people lack of information can constitute a for- Quality relates also to the level of service onto national and international agendas? midable barrier, especially for people provided. If this is too basic, demand for How can the perceptions of elites and who do not have access to many media, the service may be limited: for example, middle classes be shaped, through the and who are isolated, illiterate or don’t if there are no referrals from primary to media, education and civil society, so speak a national language. Information secondary health services, the incentive that those trapped in poverty are not in- provision – through media accessible to to attend a clinic is greatly reduced. visible, ignored or believed to be non- the chronically poor (e.g. local radio) is deserving? These difficult questions need a critical means of both expanding access to be prioritised by poverty analysts. and raising demand. Delivering social assistance If it is accepted that the poor have a right to social protection, as agreed in the Universal Declaration of Human Box 5.4 Are rich countries and their citizens really committed to Rights, how would this best be accom- reducing poverty? Targetting MDG 8 plished? The key issues in delivering so- The bold Millennium Declaration from 189 countries that ‘‘We are committed to cial assistance are whether and what making the Right to Development a reality for everyone and to freeing the entire form of targeting will benefit the human race from want’’ is not yet matched by bold actions on the part of the chronically poor; whether the value of developed world. transfers can be sufficient to have devel- Goal 8 of the Millennium Development Goals declares the establishment of a opment as well as relief outcomes (so global partnership for development. It promises the Least Developed Countries that the chronically poor may accumu- (where the greatest concentrations of chronically poor people live) tariff and quota- late assets); and how technology and in- free access for their exports; an enhanced programme of debt relief for HIPC and stitutional innovation can make cancellation of official bilateral debt; and more aid for countries committed to targeting and delivery easier. poverty reduction. Where administration is weak, target- What has followed this declaration? Trade talks at Cancun have failed as OECD ing is difficult. In such situations, com- countries refused to open up their highly-protected agricultural markets; debt mon in the poorest countries, the reduction has proceeded at a snail’s pace; and rich countries have made promises politics of transferring resources to the of more ODA at Monterrey but have not committed anything like the necessary poor is much easier if the non-poor ben- resources. efit too. Universal provision is thus pref- MDG Goals 1 to 7 – all of which are the primary responsibility of developing erable, wherever resources permit. countries – have agreed targets that are regularly monitored. There is less Where targeting is imperative, self- emphasis on quantifiable targets, however, for MDG Goal 8, which is about what targeting or targeting by readily- rich countries do. observable indicators is preferable to Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 57 Black plate (57,1)

What should be done about chronic poverty? . 57 more complex administrative approaches scheme within six years of start-up, de- involving means-testing. Self-targeting spite the physical and communication schemes, where resources being trans- obstacles in this mountainous state, pro- ferred (such as wages or food in a public vides evidence of the feasibility of effec- works scheme) are kept at low levels, are tive social assistance programmes for the only attractive to people with no other chronically poor. The value of the pen- options. Such programmes are more sion, at 2.5 days’ agricultural wages per suited to preventing destitution than re- month, is low, but this could be incre- ducing chronic poverty, although well mentally increased over time to permit it administered public works programmes, not merely to provide security but to fa- like the Maharashtra Employment Guar- cilitate opportunities for the younger antee Scheme, can contribute to both.37 members of the household.44 Social assistance is not about Evidence is mounting that ‘doles’. Pensioners support even very small transfers can grandchildren’s schooling; have positive impacts on public works schemes provide welfare and investment. the savings for small scale business start-ups.

Alvera, like many millions of older people, is using Studies of ‘free’ school lunches, subsi- her pension to support and educate her Both the coverage and value of social dised basic foods and pensions in India grandchildren assistance programmes are affected by also point to the protective and promo- how social protection is perceived by the tional roles of such interventions.38 were almost 1.2 million ‘destitute’ pen- non-poor. There is a need to point out to Spatial or social targeting (for instance sioners (defined as those with no other national and international policy-mak- targeting poor areas, minority ethnic sources of support) receiving Rs 200 ers, the middle classes and the general groups, older people or children) may (US$5) per month. Without such assis- public, how transfers are often used not have developmental benefits for the area tance the state’s growing population of only for current consumption but also or group and, depending on the target chronically poor older people would for saving, investment and further redis- group, may result in targeting on chronic have been larger and even more tribution within the household. Social as- poverty.39 Demographic targeting can be deprived.42 sistance is not about ‘doles’. Pensioners effective in contexts where the number New forms of delivery can help reduce support grandchildren’s schooling; pub- of children in a household is strongly the often considerable costs of transfer- lic works schemes provide the savings linked to chronic poverty, but there is a ring resources. For example, direct trans- for small scale business start-ups.45 risk of creating perverse incentives. fers through post offices, banks, mobile Community-based identification of the vans, and taking advantage of computer- poor and poorest has become popular, ised information systems and electronic Using PRSs to prioritise the but may fall victim to elite capture, espe- identity cards, may offer significant ad- chronically poor cially in unequal communities.40 vantages over other possibilities for get- A major advantage of national Poverty To make policy prescription even ting resources to poor people, and can Reduction Strategies is that they put the more complex, context has been found inject substantial resources into weakly issue of poverty at the heart of dialogue to be a key determinant of the success of integrated areas43 and marginal groups, between donors and developing coun- targeting in reaching the poor or very provided they are adequately registered. tries. They also offer the potential for poor.41 This makes generalisation diffi- Linking service delivery with social pro- improved analysis and coordination of cult but suggests that it may be best to tection transfers through ‘conditional efforts to target poorer groups. build on local experience rather than to transfers’ can make a difference to de- But from the perspective of chronic adopt ‘best practice’ solutions that may mand: the Food for Education pro- poverty, evidence to date suggests that be inappropriate in new contexts. gramme in Bangladesh and the Progresa PRSPs could play a more effective role. In many countries, the coverage of so- scheme in Mexico have both resulted in Few PRSs disaggregate poverty ad- cial assistance schemes is constrained by parents keeping more children out of the equately; few specifically recognise the public expenditure ceilings. If these are labour force for longer, thus reducing needs of the chronically poor, though reduced or removed, as schemes mature the likelihood of their being poor in the many discuss vulnerability; most exag- and demonstrate their effectiveness, then future. gerate the extent to which economic the chance of reaching the chronically The value of a transfer is a critical as- growth will deliver poverty reduction; poor can be greatly improved. For exam- pect of scheme design as it shapes cost and few emphasise strongly protecting ple, in the social protection trend-setting and potential impact – but evidence is the assets and rights of the poor. Interna- state of Tamil Nadu in southern India, mounting that even very small transfers tional structural factors that contribute budget ceilings for pensions and other al- can have positive impacts on welfare and to chronic poverty, (such as casualisation lowances were dropped in the early investment. The 80% national coverage and insecurity for workers in the global 1990s. As a result, by March 2003 there achieved by Nepal’s old age pension fruit market), are not currently part of Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 58 Black plate (58,1)

58 . Chapter 5 most PRSP processes. (See Chapter 4). Using the Millennium MDGs and their stance on trade. It is In principle, the ‘joined up thinking’ difficult to see how developed countries that is supposed to inform PRSPs, could Development Goals to can justify entering into trade meetings enable government and donor agencies address chronic poverty knowing ‘that the better they succeed, to take better account of the multidimen- The implications of the eight MDGs, the more people will die of poverty’.48 sional nature of chronic poverty. and their many targets and indicators, for the chronically poor are varied (see ‘Joined up thinking’ in PRSPs, Table 5.1). Only Goal 2 sets a universal The Millennium Declaration could enable government and target – primary education for all chil- on eliminating poverty should donor agencies to take better dren. This goal cannot be reached un- inform all government actions account of the less children in chronic poverty are multidimensional nature of included. – not only those on aid and chronic poverty. The other targets for 2015 are not development co-operation. measured by universal achievement. This means that policymakers can decide to What PRSs have achieved to date is exclude those who are hardest to reach, Financing chronic poverty the establishment of processes that may from efforts to achieve the MDGs. Most reduction ultimately create disaggregated strategies significantly, the Goal 1 target of halving to be developed that take account of the US$1/day poverty by 2015, may be Resources from governments and particular problems facing chronically achieved, by concentrating on those Official Development Assistance poor people. But sustained commitment nearest the poverty line.46 The resources to finance poverty reduc- and attention from governments and do- But as Table 5.1 illustrates, the case tion mainly come from developing coun- nors will be required if PRSPs are to re- for insisting that people in chronic pov- try governments, aid donors and, sult in effective measures to tackle erty must be fully incorporated in efforts importantly but often overlooked, from multidimensional and persistent poverty. to achieve the MDGs rests not only on poor people themselves. Government re- rights. The sustained reduction and even- sources in many developing countries are Putting chronic poverty tual eradication of absolute poverty will under immense pressure. During the be achieved more effectively if current 1990s, many countries in Africa experi- on the international policy and action is informed by chronic enced a worsening of both economic and action agenda poverty analysis. poverty indicators, as communities and Inevitably, the whole framing of devel- whole nations suffered a range of shocks, opment policy is in the hands of people including conflict, deteriorating terms of who have never experienced malnutri- The coherence of MDGs trade and HIV/AIDS.49 tion, disabling exclusion, and the humil- and international policy The only international resource that is iating lack of capabilities like illiteracy. The MDGs cannot be viewed in isolation meant to be focused on poverty reduc- For the global elite (north and south), from other international policies and ne- tion is international aid – Official Devel- perhaps the biggest challenge is to under- gotiations, some of which result in con- opment Assistance (ODA). But over the stand the real obstacles facing people ditions in which chronic poverty is 1990s, global aid experienced a sharp re- who have never had these advantages. sustained and may even increase. There versal, falling 20% in real terms by There is a kind of intellectual recidi- are still too many stark cases when 2001. Between 1990 and 2001, ODA aid vism in policy making, which means that ‘northern’ interests are crudely asserted. fell from 0.33% to 0.22% as a share of however well intentioned, policy keeps For example, shortly after agreeing to donor GNI – compared to the UN target returning to the idea of broad based the complete eradication of hunger by of 0.7%. In 2002, ODA rose by 7% in growth as a complete solution. But the 2015 at the World Food Summit in real terms from its 2001 total, up mod- reality is that many people are never 1996, the US government published an estly to 0.23% of donor country GNI.50 going to be able to grasp the opportuni- ‘interpretive statement’ that ‘the attain- But just one donor out of 22 DAC do- ties that the global market is supposed to ment of any ‘right to adequate food’ or nors was giving more in real terms in offer. Facing up to this reality, and prop- ‘fundamental right to be free from hun- 2002 than they had a decade earlier. Just erly taking into account the rights of ger’ is a goal or aspiration to be realised to restore aid to its per capita levels in 300–400 million people in chronic pov- progressively that does not give rise to 1990 would require US$23 billion in ad- erty, requires more than the grudging any international obligations’.47 As a ditional funding – a 45% increase. provision of patchy safety nets and poli- consequence, the MDGs seek only to In 2002, the year of the Financing for cies governed by the illusion that oppor- halve global hunger by 2015 – only 900 Development (FfD) Summit in Monter- tunity will be enough. A much more million people will need to suffer hunger rey, Mexico, global aid stood at almost creative and dynamic policy approach to longer than agreed in 1996! US$58.3 billion. Estimates prepared for chronic poverty is not only essential on The Millennium Declaration on elimi- the FfD Summit by the Zedillo Panel rights grounds, but it also recognises nating poverty should inform all govern- suggested that to achieve the MDGs, an emerging evidence that measures to in- ment actions – not only those on aid and additional US$50 billion/year in ODA crease security for the poorest can con- development co-operation. At present was needed.51 This was almost certainly tribute to aggregate growth and long there is a major inconsistency between an underestimate, as it did not include term poverty reduction. developed countries’ commitment to the costs for clean water and sanitation Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 59 Black plate (59,1)

What should be done about chronic poverty? . 59

Table 5.1 The MDGs and the chronically poor – helping each other?

Millennium Signiﬁcance of this goal for the chronically poor Signiﬁcance of including the chronically poor Development Goals for the achievement of the MDG

Goal 1 . Target 1 (halve the proportion of those living on less than US$1/ . Progress on reducing child malnutrition is very Eradicate extreme day) could lead to a focus on the less poor at the expense of the slow. Addressing chronic poverty could speed poverty and hunger chronically poor. this up considerably, since chronically poor . Although the introduction of indicators 2 (poverty gap) and 3 (share households are significantly more likely to of poorest quintile) improved the ability of this goal to focus contain under and malnourished children. attention on the chronically poor, the lack of a specified target . Progress on increasing the share of the poorest means that these indicators exercise relatively little influence on in national income would be enhanced by policy-makers. specific attempts to transfer income to the . Target 2 (halve the proportion of people who suffer from hunger, in chronically poor. terms of prevalence of underweight children under five and proportion of population below minimum level of dietary energy consumption) is useful in guiding policy towards the needs of the chronically poor, and those likely to become chronically poor.

Goal 2 . Universal Primary Education must, by deﬁnition, include chronically . This goal cannot be achieved without the Achieve Universal poor children. chronically poor. Primary Education . In the shorter term, some countries may target more resources to . In low literacy countries (where many those more likely to enrol, in order to improve rates as much as chronically poor people live) the ‘increase possible, acknowledging the impossibility of achieving 100% literacy among youth’ indicator may encourage enrolment by 2015. a focus on the ‘easy to reach’ poor.

Goal 3 . Potentially mixed impacts. Essential in the long term, but could . Interrupting chronic poverty requires girls to go Promote gender work against the interests of the chronically poor in the short and to school and stay in school for longer, among equality and empower medium term, if non-poor women (and their households) gain other things, which fosters the achievement of women privileged access to education, employment and parliament at the this goal. expense of poorer men (and their households).

Goal 4 . Reducing the number of preventable deaths is crucial to the . Very slow progress, especially in sub-Saharan Reduce child mortality interruption of chronic poverty. Africa, where there is the biggest incidence of . However, as the target is articulated in terms of improvements in chronic poverty. Addressing the national averages, there is a real danger that the focus will initially multidimensional and intergenerational nature of be on the non-poor and less poor. Redirection of resources away chronic poverty would signiﬁcantly improve the from the chronically poor may even worsen their situation. chances of meeting the targets.

Goal 5 . Improved maternal health is crucial to the interruption of the . Current rates of progress on the maternal Improve maternal intergenerational transmission of poverty. mortality indicator is one-third that required. health . However, as the target is articulated in terms of improvements in Addressing chronic poverty requires a national averages, there is a real danger that the focus will initially comprehensive and multi-sectoral approach to be on the non-poor and less poor. Redirection of resources away maternal health, and would contribute from the chronically poor may even worsen their situation. signiﬁcantly to the decline in maternal mortality.

Goal 6 . Indicators 18 and 20 (reduced HIV prevalence in young, pregnant . Reducing chronic poverty is likely to improve Combat HIV/AIDS, women and reduced numbers of AIDS orphans) are central to the capacity of poor people to access, afford malaria and other reducing chronic poverty, particularly in Africa. In the coming and complete treatment, helping to interrupt diseases decade this indicator will have increasing signiﬁcance in many other transmission paths of contagious diseases. parts of the world (e.g. India and China). . Indicators 21–24 (combating malaria and TB) are important: in many countries the chronically poor have their physical strength, health and livelihoods undermined, and die preventable deaths . However, as the target is articulated in terms of improvements in national averages, there is a real danger that the focus will initially be on the non-poor and less poor. Redirection of resources away from the chronically poor may even worsen their situation.

Goal 7 . Could have beneﬁcial effects for the chronically poor (e.g. . On target to achieve indicator 25. Ensure environmental indicators 29 and 30 on improved access to safe water and sustainability sanitation) and directly negative effects (e.g. indicators 25 and 26 would limit the opportunity for the chronically poor to modify land use in ways they believe will improve their livelihoods). The lack of progress by industrialised nations on reducing energy use (indicator 27) and CO2 emissions (indicator 28) undermines this goal.

Goal 8 . A wide range of indicators for ofﬁcial development assistance, . A focus on chronic poverty can help to achieve Develop a global market access, debt reduction, employment and access to this goal. Public support for aid in donor partnership for medicines and ICT. Indicators include: proportion of aid to basic countries is strong based on humanitarian development social services, and to Least Developed Countries, both of which principles and a wish to see resources spent on should beneﬁt the chronically poor. the poorest people. Serious efforts to create public support and solidarity around combating chronic poverty could help to mobilise public commitment to the MDGs.

Sources: CPRC analysis; Christianson, Conway and Lui 2003; Gaiha 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 60 Black plate (60,1)

60 . Chapter 5

(estimated by the World Bank at be- impact of aid on chronic poverty. US$70–100 billion a year is tween US$5 and US$20 billion/year), or Aid to basic social services, like health for reducing child and maternal mortal- and education, receives a small fraction needed to finance the MDGs. ity (estimated at US$20–25 billion/year); of total bilateral commitments. 2002 This could be achieved if and education may have been underesti- was a record year for spending on basic donors fulfil commitments to mated by more than US$6 billion/year. health and basic education. But total reach the 0.7% target. These revisions suggest that between commitments from all donors amounted US$70 and US$100 billion/year is to only US$1 billion for basic education needed, on top of current levels of ODA. and US$1.3 billion for basic health – less incidence of poverty is important. This This increase could be achieved if donors than 5% of bilateral commitments. would mean giving far more aid to South were to fulfil commitments on reaching Overall spending on health also reached Asia and to larger high-poverty coun- the 0.7% target.52 However, two years its highest level in 2002 (US$2.4 billion) tries, particularly in sub-Saharan Africa on from FfD, a major financing gap puts but education has yet to match its 1995 where chronic poverty seems the most the 2015 goals for poverty reduction in figure of US$6 billion. Spending in 2002 intractable. See Box 11.01, ‘Aid Concen- jeopardy. The modest US$16 billion/year was US$4.4 billion (Figure 5.1).53 tration Curves’, for an analysis of aid in increases promised by donors for In 1990, donors held their first Least flows to countries with the largest num- 2003 falls far short of what is required. Developed Countries conference. This bers and highest concentration of people One pragmatic response to this short- set a target for donors of 0.15% GNI in living on less than US$1/day. fall, is the proposed International Finance aid to this group of particularly deprived Facility. This proposal, backed by the UK countries, home to 700 million people, a and France, would substantially increase disproportionate number of whom are in Persistent poverty requires aid spending in the years to 2015 by using chronic poverty. In 1992, DAC donors sustained assistance additional aid pledged at FfD to back the were allocating 0.05% of their total GNI The intractable nature of chronic poverty issue of bonds. Resources generated by to Least Developed Countries (LDCs). clearly requires flexible, predictable and the sale of bonds would be immediately Ten years later, aid to LDCs was just long-term financial support. The record available for spending on poverty reduc- 0.04% of GNI, despite being at its high- on this is not good. Aid has been domi- tion. Children whose growth may be est level of the decade in real terms.54 nated by a project based approach with stunted by malnutrition and who run the Despite the rhetoric, poverty is not the relatively short time horizons and which risk of growing up illiterate, and grand- only consideration governing aid alloca- frequently bypass national aid coordina- parents struggling to care for AIDS or- tions. Regional preferences, political pri- tion mechanisms. In Ethiopia, for in- phans, are among the people in chronic orities, commerce and history all have stance, recurrent cost budget finance is poverty whose need for assistance cannot their effect on the allocation of ODA. needed to deliver the public expenditure be put off – and who would stand to Aid generally has a positive effect on set out in the Poverty Reduction Strat- benefit most from the IFF. economic growth.55 But the special sig- egy. This would need to be sustained for nificance of aid is that, unlike other 20 years, and the reliability of donor Aid allocation, as well as volume, flows which can contribute to growth, commitments will be critical. Recently is important aid can be targeted to those who most initiatives such as the DAC Task Force It is not only volumes of aid that matter. need assistance – people and countries on Donor Practices have signalled a The sectors and countries in which aid is whose poverty is persistent and severe. growing awareness of the need for pro- spent have a major effect on the potential Therefore, aligning aid with the gress on aid harmonisation, genuine developing country leadership and medium to long-term financial frameworks. Figure 5.1 ODA to least developed countries in real (2001) prices Without reliable aid commitments, (US$ millions) 10000 governments may be reluctant to make long-term commitments to recurrent costs which cannot yet be financed from 8000 revenues. The proposed International Fi- nance Facility would require donors to make legally binding commitments over the long term. This predictable finance 6000 could enable developing country governments to plan interventions in areas such as social services, on the basis of assured 4000 availability of recurrent funding needs.

2000 Financing targeted transfers Generally, it is the poor who ﬁnance poverty reduction, both through their 0 own efforts, and through gifts, loans and 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 remittances from family and neighbours. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 61 Black plate (61,1)

What should be done about chronic poverty? . 61

Putting more resources directly in the . Donors may be well placed, especially international transfers are essential. hands of the poor can therefore be crit- in the context of PRS processes, to en- Political commitment is the essential ical for sustained poverty reduction. sure that the interests of chronically factor in achieving these changes. There Social assistance is one means of poor groups are taken properly into are some grounds for optimism: the pub- reaching very disadvantaged people di- account by governments who face lic and politicians in donor countries reg- rectly. But are social assistance transfers, pressure from domestic interest groups ularly re-state their support for economic such as pensions, affordable? Evidence is with more political weight. justice and aid, particularly for the poor- clear that they can be, even for an LDC est. Proposals such as the ILO’s Global Generally, it is the poor who finance like Nepal. Nepal’s universal pension for Trust to finance global social assistance poverty reduction, both through their men and women over 75, which reaches are based on concept of international sol- own efforts, and help from family between 83% and 86% of those eligible, idarity in the effort to achieve basic social and neighbours. Putting more costs only 0.4–0.6% of public expendi- resources in the hands of the poor protection. The international community ture. Namibia’s old age cash transfer could be critical to sustained is currently discussing different proposals programme costs just under 2% of GDP, poverty reduction. to secure reliable finance for poverty re- and South Africa’s pensions cost around duction and global public goods includ- 1.4% of GNP.56 ing the International Finance Facility This experience of existing social . Helping developing country govern- and forms of global taxation.57 transfer programmes means that blanket ments to finance social assistance pro- But optimism needs to be tempered by assumptions on the unaffordability of grammes could fit well with increasing the knowledge that in 13 countries, aid transfers should be replaced by detailed efforts by donors to coordinate their is less than half way to the target of thinking on issues, such as the balance aid within a framework of developing 0.7% of GNI established in 1970. GNP between universal and targeted schemes. country ownership, using budget sup- growth in donor countries has risen from Attention also needs to be focused on port where appropriate. US$11,000 per person in 1960 to how existing social assistance can be ex- . There is no doubt that public support US$28,000 in 2002 (in real terms). Aid tended and most effectively managed. for aid in OECD donor countries is at has increased by just US$6 per person Many aid donors have been very reluc- its strongest around humanitarian and over the past 40 years.58 tant to promote, or help foot the bill for, basic needs expenditure. social assistance. Donors prefer projects which produce visible short-term results. Chronic poverty – They are wary of becoming involved in Obligations on aid must be an agenda for action government-managed schemes. Many met Between 300 and 420 million people are donors also fear ongoing commitment to Changes in aid modalities will not be trapped in poverty. They experience what some have perceived as welfarist enough to address chronic poverty – a multiple forms of deprivation over many schemes. But there are sound reasons for step change in the levels of financing for years, often over their entire lives, often donors revisiting this issue: basic services and predictable funding passing poverty on to their children, and . There is increasing evidence that im- flows is indispensable. If chronic poverty often dying premature, preventable proved social protection can be both is to be eradicated, it is necessary to look deaths. The causes of their poverty are redistributive, a productive invest- beyond 2015 and to recognise that, complex and involve sets of overlaying ment, and therefore a very sound use while some chronic poverty can be re- factors. These causes commonly include of aid money. duced through policy changes, increased social exclusion, adverse incorporation

Figure 5.2 DAC donors – gap between income and aid per capita (at 1988 prices and exchange rates) 250 1961=100 US$27,518

ODA per capita GNP per capita 200

150

US$11,303

100

US$61 US$65

50 1961 1963 1965 1967 1969 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P15E-02.3D.3D Folio: 62 Black plate (62,1)

62 . Chapter 5 in the economy and living in parts of the . tackle child poverty by ensuring adult will mean the chronically poor are not world that are poorly governed and/or livelihood security and investing in invisible, ignored, pushed aside or ex- weakly connected to national and inter- nutrition, health and education ploited? This issue should be a major national economies. programmes; concern for policy-makers and will be a This first Chronic Poverty Report has . improve health care systems, including key task for the second Chronic Poverty set out to raise awareness of the extent curative services, so that households Report. and experience of long term poverty. It are not caught in health shock-induced At the national level, this report high- has also drawn some initial conclusions spirals of asset depletion; and, lights the need for developing more ef- about policy and action. Most impor- . reduce the disastrous effects of HIV/ fective service delivery for chronically tantly, although many of the components AIDS through preventive programmes poor people and reveals that this is possi- of the contemporary ‘orthodoxy’ on pov- and by rapidly increasing the avail- ble through, for example, non-contribu- erty reduction can also benefit chronic- ability of affordable anti-retroviral tory pensions in Brazil, Nepal and South ally poor people (Table 5.1) it is clear treatments. Africa, food-for-work and employment that to tackle chronic poverty it is neces- Increasing economic opportunity is also guarantee schemes in many African and sary to go beyond this orthodoxy. important and, for the chronically poor, Asian countries, and initiatives linking Increasing opportunity, reducing vul- the quality of growth – broad-based and access to assets and education. nerability, and empowerment are all nec- reaching rural areas – is probably as im- At the international level, the MDGs essary for chronic poverty reduction, but portant as the rate of growth. Growth provide a partial – but only a partial – livelihood security must be prioritised. that is associated with rapidly increasing framework for tackling chronic poverty. The chronic insecurity of those trapped inequality is unlikely to bring much ben- The big issues remain – rich country in poverty constrains their ability to pur- efit to those trapped in poverty. Asset commitment to economic justice through sue opportunities and seek empower- transfers, such as land reform, and indi- aid, trade and debt reform – as do de- ment so much that social protection rect redistribution, such as progressive tailed questions about whether some policies must play a major role in strat- public revenue and expenditure systems MDGs encourage policymakers to avoid egy. There are many elements to such that skew educational resources towards the poorest. At the global level, greater social protection policies; CPRC research the poorest, are important elements of transfers of resources from North to highlights the need to: chronic poverty reduction strategies. South, and the political commitment to . put security first and use existing evi- While this report concurs with many co-operate on tackling conflict and dis- dence to develop larger scale, cost ef- on the importance of empowerment, it crimination are urgently needed. fective social protection programmes finds that the big question is not so much Chronic poverty is a global challenge. that enable chronically poor people to ‘what?’ but ‘how?’. How can we foster It demands a response that requires ac- make choices that do not undermine the social solidarity – at household, com- tion at all levels. People in chronic pov- their current or future well-being; munity, national and global levels – that erty cannot wait for change.

Notes 1. WDR2000/1:41. avoided as they reduce opportunity and create 35. Esping-Anderson 1999. 2. Begum and Sen 2004. transitory and chronic poverty (Stiglitz 2001). 36. Uphoff 1992 Gal Oya book. 3. Wood 2003. 20. Timmer 1997. 37. Gaiha and Imai 2003. 4. WDR 2000/1. 21. McKay 2004. 38. Guhan 1993. 5. See Chapter Two. 22. Park and Kim 1998. 39. Ravallion 2003: 17–18. 6. Lund 2002. 23. Bruton et al. 1992. 40. Ravallion 2003:22 Bangladesh example. 7. Dreze 2003 in EPW. 24. For example, aid agencies supported NGOs 41. Coady, Grosh and Hoddinott 2003. 8. Matin and Hulme 2003. throughout the civil war/Taliban era in 42. Government of Tamil Nadu 2003. 9. Daru and Churchill 2003. Afghanistan. When ‘peace’ arrived in 2002, the 43. Devereux 2002; Farrington et al. 2003. 10. Ravallion 2003. staff of these NGOs provided the basis for the 44. Rajan 2003. 11. Case and Deaton 2003. re-establishment of a civil service – few other 45. Devereux 2002. 12. This is particularly true of cash transfers, which Afghans had relevant training and experience in 46. Paradoxically, this target might be more rapidly can be used for whatever purpose the recipient civil management, record keeping, civilian achieved if chronically poor people die, thus decides. logistics, accounting, electronic reducing the proportion of people below a 13. Although micro-ﬁnance programmes provide communications etc. dollar a day. examples in which women are most usually the 25. Bruton et al 1992. 47. Pogge (forthcoming: 10). development receipient. 26. Sen and Ali 2003. 48. Pogge (forthcoming: 20). 14. See later for a discussion of how the chronically 27. Kothari 2002. 49. Table 12, Human Development Report 2003, poor can best be targeted for social assistance. 28. China is developing plans to relocate long-term UNDP. 15. Harper, Marcus and Moore 2003. poor people from spatial poverty traps in rural 50. OECD DAC Statistics, Table 1. 16. The likelihood that an episode of ill-health will areas to more dynamic regions, but few other 51. Zedillo Commission 2001. lead to a household or individual falling into countries have considered such ‘closures’ of 52. If ODA had reached 0.7% of GNI in 2002 it chronic poverty as opposed to a period of regions. The implications for the regions would have been US$174 billion. transitory poverty depends largely on the expected to absorb such populations are 53. OECD DAC Statistics Table 5. nature, severity and duration of the episode, uncertain. 54. OECD DAC Statistics Table 2a. and its resultant effect on income, employment, 29. UNHCHR 1948/2004. 55. Tarp 2000; Morrissey 2001. household assets, level of indebtedness, 30. While the assassination of Chico Mendes, the 56. South Africa Department of Social redistribution of household income and assets, leader of the rubber tappers’ union in Brazil, Development 2002: 20. and whether the episode leads to lasting was widely reported the deaths and beatings of 57. As the costs of borrowing are much less than morbidity, disability, incapacitation or death. thousands of other frontline activists for the the rates of return from aid investments (and 17. Corbett 1989. poor goes unmentioned in the media. because people in developing countries prefer to 18. WHO (n.d.): 4 and Goudge and Govender 31. Piron 2003. have funds now rather than later (the Social 2000. 32. Most obviously Uma Kothari in Cooke and Time Preference Rate) this should result in 19. Conversely, policies that lead to economic Kothari 2001. increased poverty reduction. collapse, such as the over rapid liberalisation of 33. Mosse 2001. 58. The Reality of Aid 2004, forthcoming. the former Soviet Union’s economy, must be 34. Hickey 2003; Esping Anderson. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P26E-01.3D.3D Folio: 63 Black plate (63,1)

PART B Regional perspectives on the experience of chronic poverty Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P26E-01.3D.3D Folio: 64 Black plate (64,1) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P26E-01.3D.3D Folio: 65 Black plate (65,1)

– Female literacy is high in Lesotho and Zimbabwe – Each of these six countries has at least three indicators that show ex- Understanding treme deprivation. Over the past two decades, many indica- chronic poverty in tors have worsened, for sub-Saharan sub-Saharan Africa Africa as a whole as well as for particu- 6 lar sub-regions and countries (see Table 6.2, and PART C for country-level detail). Due to HIV/AIDS and conflict, Sub-Saharan Africa is the poorest part of the world but also the region with overall life expectancy has reduced, espe- the highest share of its population living in chronic poverty. The region as a cially in southern Africa and among whole has not experienced economic growth over the last two decades, women. On average, household con- and the opportunities available to the poor have been highly constrained. A sumption has stagnated, and, in West high proportion of poverty is likely to be chronic. Best estimates are that Africa, declined. between 30% and 40% of the absolute poor population in sub-Saharan Nevertheless, some positive signs are Africa is chronically poor – between 90 and 120 million people. It is a apparent: gloomy picture, the human development results of which are summarised . Generally reduced illiteracy rates, es- in Table 6.1. Chronic poverty in the region is most pronounced in areas that pecially for women are remote, affected by protracted violent conflict, suffering economic stag- . Reduced infant mortality rates (except nation or decline, and where HIV/AIDS and other diseases are endemic. where HIV/AIDS has led to an increase) Poverty trends in sub-Saharan Africa . Over the past two decades, a few High rates of infant mortality and stunt- – Of the 23 ‘moderately deprived’ countries including Cape Verde, Equa- ing, and low life expectancy, (exacer- countries, 15 are sub-Saharan Afri- torial Guinea, Guinea, Mauritius, and bated by the HIV/AIDS pandemic) occur can, and nine are East or Southern possibly Sudan,1 have registered signif- across sub-Saharan Africa, despite differ- African icant improvements in human develop- ences in income poverty levels. . Few sub-Saharan countries have low ment as well as sustained positive levels of deprivation on any one indi- growth The most extreme, persistent, cator. Exceptions include the middle . A number of other countries, many of multi-dimensional poverty is income countries of Southern Africa, which are recovering from conflict or which nonetheless have extremely low economic collapse, have begun to 100% a sub-Saharan African life expectancy, perhaps reflecting the grow in the 1990s, supported by in- – and particularly a Central HIV/AIDS epidemic flows of aid and West African – problem . Among low income countries, Coˆ te . Benin, Burkina Faso, Ethiopia, Ghana, d’Ivoire has been doing relatively well Mozambique and Uganda have all in- in terms of proportion of people living creased consumption per capita during . All of the 16 countries flagged up as on less than US$1/day, and, along the 1990s, the last two countries dra- ‘desperately deprived’ are in sub- with the Gambia, Senegal and Togo, matically so, and most appear to be Saharan Africa, (see Chapter 1) it has had relatively low rates of sustaining that growth into this – 12 are in Central or West Africa stunting decade.

Figure 6.1 Chronic poverty in sub-Saharan Africa

About one-quarter of One in every 6 or 7 people About one in every the world’s chronically poor in sub-Saharan Africa three poor people in live in sub-Saharan Africa is chronically poor sub-Saharan Africa is chronically poor Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P26E-01.3D.3D Folio: 66 Black plate (66,1)

66 . Chapter 6 Adult male, 2000 illiteracy rate, Source: See Part C. Source: See Part C. capita (%) Average annual consumption per change in household Adult female, 2000 illiteracy rate, (% points) Life illiteracy rate Change in male male, 2000 expectancy, though further analysis is required, the data is broadly Life expectancy, female, 2000 (% points) illiteracy rate Change in female a 2000 s.d., 1992– Stunting <- 2 2000 Infant Change in male life expectancy (# years) mortality rate, 2001 Under-5 (per 1,000 live births), mortality rate a expectancy (# years) Change in female life te d’Ivoire, Equatorial Guinea, Gabon, Gambia, The, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Mauritania, Niger, Nigeria, Sao Tome, Senegal, ˆ Average 1989–99 shortfall of poor below US$1/day (%), a 3 – 53 44 22 68 64 53 30 (% points) mortality rate US$1/ Change in infant on less than people living day,1989–99 Percentage of West AfricaCentral AfricaSouthern AfricaEast AfricaSub-Saharan AfricaNorth AfricaSub-Saharan Africa + North Africa –18 –5 –21 –6 –15 –17 –56 1 –5 0 –8 –3 –3 11 2 –2 –6 2 –1 –1 10 –29 –28 –15 –25 –25 –25 –24 –25 –23 –13 –20 –21 –18 –19 –3.9 –0.8 –0.3 –1.5 –2.3 –0.3 1.2 Sub-Saharan Africa + North Africa 34 – 156 99 38 53 52 48 30 West AfricaCentral AfricaSouthern AfricaEast AfricaAll sub-Saharan AfricaNorth Africa 58 50 43 29 30 45 36 36 27 27 185 196 174 165 155 111 121 109 111 100 37 44 40 36 44 51 50 50 47 49 50 47 48 45 47 52 49 46 29 47 35 28 30 19 31 Sierra Leone, Togo Central Africa = Burundi,Southern Central Africa African = Rep., Angola, Dem.East Botswana, Rep. Africa Lesotho, Congo, = Malawi, Rep. Comoros, Mozambique,North Congo, Djibouti, Namibia, Africa Rwanda Eritrea, South = Ethiopia, Africa, Algeria, Kenya, Swaziland, Egypt, Madagascar, Zambia, Libya, Mauritius, Zimbabwe Morocco, Somalia, Sudan, Tanzania, Tunisia Uganda Table 6.1 Summary of poverty indicators for Africa* Figures have been rounded. Table 6.2 The changing African picture, 1980–2000* Figures have been rounded. West Africa = Benin, Burkina Faso, Cameroon, Cape Verde, Chad, Co * These tables contain a reconﬁguration of the data provided in PART C based on a more detailed regional disaggregation (African Development Bank). Al compatible with that in PART C, but highlights the extreme regional disparities discussed in the text. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P26E-01.3D.3D Folio: 67 Black plate (67,1)

Understanding chronic poverty in sub-Saharan Africa . 67

However, even these ‘turnaround’ coun- in human development despite little or extreme poverty in a US$1/day sense. In tries are often only growing in limited no growth in consumption. The Gambia, sub-Saharan Africa, South Africa domi- regions and sectors and progress is frag- Madagascar and Niger, for example, nates, with other middle income coun- ile. The dramatic reduction in poverty in have seen large improvements in infant tries (Botswana, Namibia and post-conflict Uganda has enabled a large mortality rates, life expectancy and liter- Swaziland) linked economically and po- number of people who were poor – in- acy, despite high proportions of the pop- litically to it. The major source of infor- cluding some of the severely poor – in ulation living on less than US$1/day and mation on chronic poverty to date has 1992 to escape poverty by 1996 or declining consumption levels. The Gam- been the KwaZulu-Natal Income Dy- 1999. Uganda’s infant mortality rate has bia now has one of the lowest rates of namics Survey. While it cannot be taken seen an above-average decline, despite infant mortality and stunting on the con- as representative of middle income the HIV/AIDS epidemic. However, re- tinent, and life expectancy in Madagas- southern Africa, primarily because of its cently the infant mortality rate has stag- car is significantly above both East and geographical limitations, it is clearly in- nated. The reasons for this are both sub-Saharan African averages. dicative of the extent to which there direct (disease – malaria, diarrhoea, can be large populations living in HIV/AIDS, malnutrition, acute respira- How many people are chronic, severe and multi-dimensional tory infection (ARI) and maternal condi- poverty in relatively wealthy countries tions) and proximate (linked to maternal chronically poor in (Box 6.1). and household characteristics, utilisation sub-Saharan Africa? Above the Sahara, panel data is avail- of health services and socio-economic There is a limited number of recent, able for Egypt – where only about 3% factors).2 As a result of population high quality, representative and compa- live on less than US$1/day but over half growth, the absolute number of prevent- rable panel surveys from which the ex- the population, or close to 35 million able infant deaths and illiterate adults in tent of chronic poverty in sub-Saharan people, lives on less than $2/day. This Uganda may actually be as great as it Africa can be determined. Based on data suggests that the middle income was in 1980. panel data from Ethiopia, Coˆ te d’Ivoire, countries of North Africa may also have Progress on poverty has excluded Uganda, urban Madagascar and rural significant numbers of people living in about 20% of Ugandans – the chronic- Zimbabwe, best estimates are that be- chronic poverty. Egyptian panel data ally poor – largely made up of large tween 30% and 40% of the absolute suggests that about one-fifth of all house- rural households: poor food crop farm- poor population in sub-Saharan Africa is holds are chronically poor, and that, of ers unable to benefit from periodic price chronically poor – between 90 and 120 the households that were poor at one increases for crops like cotton and cof- million people. Nearly one-third live in time, 48% were always poor, so that the fee; people unable to diversify their live- Nigeria, and over half live in West Afri- overall contribution of chronic poverty lihoods; severely poor people with ca in total. The rates of both absolute to total poverty is high.5 limited or no education; and those in the and chronic poverty are high in Western poorest eastern regions and conflict-af- and Central Africa where one in every 5 Who are the chronically fected north. Indeed, while Uganda is or 6 is both absolutely and chronically commonly referred to as post-conflict, its poor. poor in sub-Saharan northern region is still affected by There are 22 sub-Saharan countries Africa? chronic conflict. The household survey for which there is both a US$1/day figure Chronically poor households tend to be for 1999 indicated that 15% of house- and sufficient data to undertake the clus- larger and with high dependency ratios holds reported being affected by civil ter analysis detailed in Chapter One. Of (Uganda,6 South Africa’s KwaZulu- strife, compared to 7% in 1992, and these 22 countries, approximately 310 Natal,7 urban Madagascar8). In civil strife was seen as being a greater million people live in 12 desperately de- KwaZulu-Natal, it is particularly cause of economic damage than theft or prived countries, about 150 million of women-headed, older households, as personal violence. whom live on less than US$1/day; per- well as larger, rural households who are Unsurprisingly, escape is most diffi- haps 45–60 million of these people are chronically poor. Such households usu- cult for the poorest; and in Uganda chronically poor. Almost 110 million ally have low overall years of schooling, urban households emerged from poverty people live in the 10 moderately de- low levels of productive assets such as at a much higher rate than rural.3 Ugan- prived countries for which we have data, chickens, livestock (Uganda) and arable dan panel data suggests that movement close to 40 million of whom are abso- land (South Africa), and get low returns out of poverty favoured those house- lutely poor, and 10–15 million of whom from their land (Rwanda).9 holds close to the poverty line: 68% of are chronically poor. Several countries The chronically poor include those households within 5% of the poverty with insufficient data – including Ango- who have missed out on economic op- line in 1992 had moved out of poverty la, Guinea, Liberia, Mali, Somalia, portunity and have been unable to diver- by 1996, compared to only 31% of Sudan and Zambia – make up the ranks sify or secure their livelihoods. Disabled those with expenditures less than 50% of those countries with high proportions people, older people, ethnic minorities below the poverty line initially.4 The ex- and numbers of absolute and chronically are frequently marginalised or excluded tent to which sustained growth can fa- poor. from participating in what opportunities cilitate an escape from poverty – even in Middle income countries with high exist in, for example, education, health, the longer term – for those left behind levels of inequality can still have signifi- employment, politics and community is debatable. cant populations of chronically poor support. They are often unable to access Several states have seen improvements people, even if this is not necessarily limited livelihood options to help them Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P26E-01.3D.3D Folio: 68 Black plate (68,1)

68 . Chapter 6

Box 6.1 Understanding chronic poverty in middle income South Africa South Africa is a middle income country, with per capita GDP sector unemployment from 24% to 46%. Semi-skilled black of $3,985 in 2000, comparable to countries in parts of East- workers became increasingly detached from the formal Central Europe and Latin America. However, about half of all economy, rendering their households susceptible to chronic South Africa’s 45 million people presently live in poverty, and poverty. Since the end of apartheid, a further one million about one quarter of all households are trapped in chronic jobs were lost in the formal sector, and official poverty. Whilst many poor people manage to escape poverty unemployment of 16% in 1995 has since doubled. for short periods (a couple of months or even a few years), Moreover, many employed people do not earn enough to they fail to do so in a more permanent way. These escape poverty. households may not be chronically poor at present, but they 3. Rural poverty: Nearly three-quarters of all rural people are are very vulnerable to becoming chronically poor in the future. poor, most of them chronically so. Poor rural households do Five of the central aspects underlying chronic poverty in not have the assets (land, finances, tools) to progress as South Africa are: agriculturalists, and land-based livelihood strategies fail to 1. Historic asset depletion: From the mid-17th century provide enough to accumulate, inhibiting incremental onwards, black workers were selectively incorporated into a escape from chronic poverty over time. severely repressive labour system, limiting their ability to 4. High inequalities: South Africa has both very high income accumulate and use their assets and skills. Today the inequality (with the richest fifth of households receiving over effects are still evident, as the majority of poor black 70% of income, and the poorest two fifths less than 4 households continue to lack direct access to basic percent), and highly racialised economic geographies due to economic resources and household food production assets. apartheid spatial planning. It is unlikely that they will be able to escape inter- 5. HIV/AIDS epidemic: HIV/AIDS incidence in South Africa is generational poverty through their own efforts. amongst the highest in the world. HIV/AIDS related deaths 2. Post-1970s economic restructuring. South Africa’s are now a major cause of chronic poverty. The causal economic restructuring is rooted in the economic decline relationship between AIDS and chronic poverty is complex between 1974 and 1994. During this time, the number of and controversial, but there is some evidence that chronic potential African workers increased from 6.9 to 9.3 million, poverty makes households more susceptible to HIV/AIDS but the number employed in the formal sector actually exposure and thus infection. declined from 5.2 to 5.0 million, expanding African formal Source: de Swardt, 2003.

diversify and improve their position. access to key resources such as land and Women-headed households In areas where the economy is largely basic services are severely limited. and AIDS widowhood agricultural, many, if not most, older Older people (and their households), Often the product of AIDS deaths or people missed out on formal education especially when they are responsible for conflict, but also of estrangement, aban- in their youth, making it difficult for the care of vulnerable children orphaned donment and family breakdown, them to participate in less physically de- as a result of HIV/AIDS and conflict, are women-headed households are not in manding, off-farm activities.10 The poor also particularly vulnerable to deepening general more likely to be poor compared and especially the poorest often depend and ongoing poverty. to male-headed households. There is only on their own labour. This capacity The situation is made worse for older much variety in the picture across the naturally declines with age, increasing people and others who are excluded continent, as throughout the world. the vulnerability of older people, particu- from decision making and for those who Male-headed polygamous households in larly if they do not enjoy support from are physically unable to access services, Nigeria’s rural areas, for example, are kin. They may have to resort to begging such as food and health facilities, and disproportionately poor, affecting a large and charity to survive. are disadvantaged by the way services number of women.11 It is plausible, how- The chronically poor include large are provided and structured. This leaves ever, that women-headed households numbers of people displaced or disabled people highly vulnerable to ill-health and who are poor are more likely to be by persistent conflict, including refugees disease, and dependent on fragile sup- chronically poor, although this does not and internally displaced people – of port systems. Loss of individuality and necessarily mean that women-headed whom there are more in sub-Saharan control, as well as the humiliation of de- households represent a disproportionate Africa than anywhere else in the world; pendency, often define their lives. segment of the absolute poor. abducted children, orphans, widows, There are more HIV/AIDS sufferers in In semi-arid Zimbabwe, women- people with impairments and illnesses. sub-Saharan Africa than anywhere else headed households reported less recovery People affected by conflict spend long pe- in the world, and their households have from the 1991/2 drought, and were more riods on the run from violence, with or to bear the isolation associated with this likely to not have recovered at all by without basic assets, which may have as well as almost all the burden of care 1998. They found entry into the critical been left behind in the hurry to escape. for this and other chronic illnesses (see non-farm occupations more difficult This limits their ability to find food and Box 6.2). Many households have than did male-headed households.12 shelter. Life within refugee camps may witnessed several deaths from AIDS, and Women-headed households were also not be better – most are overcrowded most deaths are in the adult breadwinner more disadvantaged in urban Madagas- with very poor living conditions, and and/or homemaker and carer category. car.13 This may be the case in many Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P26E-01.3D.3D Folio: 69 Black plate (69,1)

Understanding chronic poverty in sub-Saharan Africa . 69 urban settings where education, health and nutritional status – assets that are Box6.2 Voicefromthefield often less common in women-headed Illness and marginalization among the working poor in South Africa households – make a large difference to ‘They appreciated that I came to visit them because no one visits their house. the work and income available. What touched me is that four members of the family were sick. One had TB, one Widows commonly return to their pa- had HIV, one woman had a breast problem, and her baby had diarrhoea. They rent’s village and beg for a small parcel have never been treated because they could not pay (US$0.3 – 30 cents) for of land to cultivate. Some will seek a transport. The baby is fed on thin porridge diluted with water and breastfed on the protector, often marrying again soon healthy breast. I had to help them with money. There is no food at all, they after their widowhood. This option is depend on anyone who comes in and gives food . . . The family is tired of them, not available to AIDS widows however, they do not want to assist them anymore, because they are always hungry. There 14 who are often shunned. AIDS widows are no blankets except one, which is shared between the baby with diarrhoea and make-up between 24% and 60% of all the person with HIV. The roof is leaking during the rainy season and it gets widows, and are commonly stigmatised, flooded sometimes. This house belonged to a family member who died of TB. No- accused of being witches and bringing one in this family has an ID, and they have no money to apply for IDs, or even to 15 the disease into the family. come to town.’ Source: Mount Frere fieldworker, translated from Xhosa, in de Swardt (2002). People with low and declining access to income, and child-headed households.17 Chronic poverty in remote land and other assets Those that are unable to access bor- rural sub-Saharan Africa The poorest occupational groups are rowing or rental systems either due to Chronic rural poverty in Africa is char- often non-cash crop farmers, whose agri- weak family connections, diminishing acterised by topography and remote- cultural productivity is low, assets mini- common lands (in some cases related to ness. In remote areas such as semi-arid, mal, and access to credit and output wildlife conservation projects), or inabil- coastal, deep forests, borders and moun- markets weak. This is widely true in ity to pay rent are more vulnerable to tainous areas, poor people are likely to West Africa. Cash crops can give good long-term poverty. Land has classically experience poverty along several dimen- results from time to time, the benefits of provided subsistence when the market or sions, including low incomes, low rates which can be capitalised and used to ex- state has been hostile, and provides a of literacy, and political marginality. pand enterprise, and provide access to basis for accumulation when conditions They are likely to have underdeveloped state loans, extension and research often are propitious. Loss of land has occurred and highly imperfect markets, limited denied to others. However, tenants, both on an individual basis, as for the non-farm livelihood opportunities, poor sharecroppers and migrant labourers can AIDS widows and abandoned women social and physical infrastructure, and remain poor for long periods. Poor discussed above, and on a collective weak social and political institutions. households who are able to diversify out basis, as a result of conflict. Quality basic services tend not to reach of farm activities into off- and non-farm deep into rural areas. Towns in these activities are in many situations the Where are the chronically regions are more likely to have stag- lucky few. Many are only able to enter nant economies than urban centres low cost and low return activities, often poor in sub-Saharan elsewhere. dependent on shrinking and heavily used Africa? Extreme poverty in the Gambia is common property resources. These are The African chronically poor tend to live highly concentrated in the rural areas, not escape routes from poverty. in countries with large numbers of poor with about 35% of the rural community The chronically poor are likely to have people, and a history of low economic living in absolute poverty, compared to experienced a series of shocks of differ- growth, such that the prospects of re- 15% in urban areas. Low agricultural ent sorts – health, drought, conflict, fam- maining poor are strong. Poverty is more productivity and an inability to accumu- ily breakdown – from which their low prevalent in rural than urban areas late capital and other assets are key, asset position hardly allows them to re- across the whole of sub-Saharan Africa, with livelihoods being constrained by a cover, or from which recovery is at best despite regional differences in income lack of tools and equipment, inadequate slow and erratic. and levels of urbanisation. storage facilities, difficult to access mar- Landlessness is particularly important, There are pronounced regional differ- kets and credit. as land has been the continent’s prime ences in the incidence of poverty through- In Uganda, while 61% of the urban safety net. There is an expanding group out Africa, and chronic poverty tends to households that were poor in 1992 were for whom this is no longer the case. follow similar distribution patterns able to move out of poverty by 1996, Whether the ‘negotiability’ of access to across remote, less-favoured, weakly-in- only 39% of rural households moved land offers sufficient potential for social tegrated regions. There are many regional out of poverty over the same period, mobility for the disadvantaged is ques- poverty traps in sub-Saharan Africa, leaving a large proportion of the rural tionable.16 Land adjudication processes where people are generally highly de- poor in a state of chronic poverty. With- in Kenya, for example, have ignored the pendent on food crops and livestock in rural areas, this was strongly region- claims of some, notably young men with- farming (or agro-pastoralism), and these ally differentiated, with mobility much out clear inheritance rights such as sons are often inhabited by minority ethnic higher in the more rapidly growing cof- of widows or divorcees, and women- groups, both in rural and urban areas. fee producing regions than elsewhere. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P26E-01.3D.3D Folio: 70 Black plate (70,1)

70 . Chapter 6

Urban chronic poverty in areas. Today, however, urban as well as stability, good weather and improved sub-Saharan Africa rural Coˆ te d’Ivoire is in a very different macro-economic management, urban In countries experiencing sustained situation after a succession of shocks, household welfare declined during this growth, the chronically poor are mainly combined with the current political and period. That one-quarter of urban located in regions, sectors or sub-sec- economic collapse which has propelled households remained poor reflects the tors which are not contributing to and many into chronic poverty. long-term impacts of the serious agro- benefiting from that growth. There is Not all urban centres are responsive to ecological and conflict-related shocks some evidence that urban Africa – at growth. Madagascar followed neo-liber- that affected Ethiopia before 1994 (and least in the large cities – seems to real prescriptions during the late 1980s have done again since 1997). Chronic spond to growth with significant and and 1990s, but economic growth did not poverty in urban Ethiopia is associated fairly rapid urban poverty reduction. materialise until 1997. Poverty remained with low asset ownership, low levels of However, some groups of urban high, slightly worse in rural areas, and education, as well as age and ethnicity chronically poor people may not benefit far worse than it was in the 1970s. (the Gurage being more likely to be from growth. In smaller towns, particu- Although there was rapid growth in chronically poor than the Tigre), and, larly in more remote and less econom- average household incomes in the capital to a certain degree, with women- ically dynamic regions, it may be that city (50% between 1995 and 2000 com- headedness. the urban poor are more likely to re- pared to an overall GDP increase of only Under- and unemployment are signifi- main absolutely poor. 2.3%), 65% of the urban population cant problems of the urban Ethiopian In urban Coˆ te d’Ivoire poverty unam- were poor in both 1997 and 1999. Few chronically poor. Over one-quarter of biguously rose between 1985 and 1995 had access to electricity and running chronically poor household heads work at both the $1/day and $2/day level but water. Most chronically poor households as casual labourers or in ‘women’s busi- was accompanied by rising poverty se- lived in poor neighbourhoods, and their ness activities’, compared to 8% among verity and worsening inequality. High jobs were largely low quality and in the the never poor. These occupations are rates of urban poverty began to decline informal, low wage sector. insecure and give low returns, so it is not (although not to below 1985 rates) after Between 1994 and 1997, 26% of surprising that the chronically poor are economic recovery began in 1995, such urban Ethiopians remained poor, com- disproportionately represented. Simi- that by 1998 poverty incidence ($1/day) pared to 24% in transitory poverty and larly, in urban Uganda the main liveli- fell to 1.2% in Abidjan, 10% for other 51% non-poor.18 Despite economic re- hood activity of 47% of the ‘always urban centres, and 5.9% for all urban covery associated with post-conflict poor’ is in agricultural subsistence.

Notes 1. Data is questionable for Sudan. 2. Government of Uganda 2002. 3. Deininger and Okidi 2003. 4. Lwanga-Ntale and McClean 2003. 5. Haddad and Ahmed 2003. 6. Lawson, McKay and Okidi 2003. 7. Aliber 2003. 8. Herrera and Roubaud (2003). 9. Muller (2000). 10. Najjumba-Mulindwa, I 2003. 11. Oduro (2003). 12. Bird and Shepherd 2003: 603–4. 13. Herrera and Roubaud (2003). 14. Bird and Shinyekwa 2003. 15. UNDP 2002:49–50. 16. Woodhouse (2003:6). 17. Hunt (1994). 18. Kadir & McKay 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P27E-01.3D.3D Folio: 71 Black plate (71,1)

A third survey, collected only in semi- arid rural Andhra Pradesh and Mahara- shtra, found that over one-fifth of the population was poor in all nine years be- Understanding tween 1975/6 and 1983/4, while 60% were poor in at least five of nine years.5 chronic poverty in Further analysis of this dataset suggested South Asia that even relatively affluent households 7 are highly vulnerable to long spells of poverty when severe crop shocks occur.6 The Indian National Sample Survey South Asia has the largest number of chronically poor people in the world – reported that the number of poor people an estimated 135 to 190 million people. Chronic poverty in the region is increased by 13 million between 1987– most pronounced in areas that have significant minority populations,1 that 88 and 1993–94, while data from 1999– are economically stagnant, where agrarian class structures and gender 2000 shows a very large reduction in the relations are exploitative, and where governance is weak. second half of the decade. This finding is intensely disputed, however, due to Poverty trends in South years of war have obstructed almost changes in the way the national figures Asia all potential progress. have been calculated, and as such it re- . 44% of the population of India lives mains difficult to estimate the absolute below the international US$1/day pov- How many people are numbers of chronically poor people today. Due to the very nature of chronic erty line. chronically poor in South . In Nepal, Pakistan and Bangladesh the poverty, however, it is unlikely that the figures are also relatively high (at Asia? proportion of people in chronic poverty 38%, 31% and 29% respectively). The number of recent, high quality, rep- has declined at anything like the rates of . In Bhutan and Afghanistan, where resentative and comparable panel sur- poverty in general. data is unavailable, the proportion of veys available to determine the extent of For instance, village-level research in people living on US$1/day is likely chronic poverty is very limited. Best esti- Rajasthan, where headcount poverty has comparable and much higher, mates suggest that about one-third of the unambiguously declined, suggested that respectively. poor population in South Asia is about 18% of the total population was . Internationally, South Asia has the chronically poor – between 135 and 190 poor both 25 years ago and in 2002. worst indicators of stunting and fe- million people, of whom 110–160 mil- This figure ranged from 8% to 31% male illiteracy, and very poor rates of lion are Indians. Bangladesh and Paki- across districts, and was highest among child mortality and female illiteracy. stan account for the majority of the scheduled tribes, more than two-thirds . The headcount ratio for the chronic- remainder. of whom had stayed in poverty over the ally poor has been declining in many A survey of rural Bangladesh suggests past 25 years.7 parts of the region – particularly in that close to one-third of the rural popu- For Pakistan, a significant amount of southern and western India, and in lation was poor in both 1987/8 and analysis has been undertaken using one Bangladesh. 2000.2 In India, two national sample sur- particular dataset.8 Different approaches . Most human development indicators veys suggest that in the late 1960s3 and to defining chronic poverty and the pov- also have improved over the past two between 1970 and 19814, almost half erty line have led to a wide range of esti- decades, although in Afghanistan the rural poor were chronically poor. mates of chronic poverty. The best all-

Figure 7.1 Chronic poverty in South Asia

Over one-third of the world’s About one in every chronically poor live in India, About one in every 3 or 4 poor people and almost half live in 8 people in South Asia in South Asia is South Asia as a whole is chronically poor chronically poor

India

South Asia excluding India Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P27E-01.3D.3D Folio: 72 Black plate (72,1)

72 . Chapter 7

Table 7.1 Summary of poverty indicators for South Asia

Average depth of poverty (the number of Under-ﬁve Percentage percentage mortality Infant Proportion of people points by which rate mortality of children Life Adult Adult living on the poor fall (per 1,000 rate (per under 5 expectancy, Life illiteracy illiteracy less than below the live births) 1,000 live who are female, expectancy, rate, female, rate, male, US$1/daya poverty line)a 2001 births) 2000 stunteda 2000 male, 2000 2000 2000

Afghanistan – – 257 165 52.0b –––– Bangladesh 36.0 22.5 77 54 44.8 59.5 59.4 70.1 47.7 Bhutan – – 95 77 40.0b 63.3 60.8 – – India 44.2 27.1 93 69 45.5b 63.8 62.8 54.6 31.6 Maldives – – 77 59 26.9 65.8 67.3 3.2 3.4 Nepal 37.7 25.7 91 72 54.1b 58.3 58.8 76.0 40.4 Pakistan 31.0 20.0 109 85 – 59.9 60.2 72.1 42.5 Sri Lanka 6.6 15.2 19 17 17.0 75.3 69.5 11.0 5.6 Regional average 40.7 26.1 98.1 72.4 45.5 63.0 62.2 57.3 33.9 a. Data refer to the most recent year available b. Data differ from the standard deﬁnition Source: See Part C.

Pakistan estimate of rural chronic pov- particularly vulnerable to chronic pov- cases this can result in people becoming erty, based on mean income over five erty in the region. Many chronically more vulnerable to exploitative employ- years, is 26% – this represents about poor live in persistently poor Indian ment (see Box 7.1). Much migration for 50% of households classified as poor in states and/or less favoured or remote work undertaken by the poor in South the first year of the survey, and about areas. Asia is this rural-rural, temporary and 6% of households classified as non-poor seasonal movement,11 although mi- in the first year. Table 7.2 presents a grants are also often among the urban summary of these different approaches The working poor chronically poor. This is not to say, and estimates, and includes another sur- Contrary to the common perception that however, that all migrants are chronic- vey that is more recent, but also contains the chronically poor are ‘unproductive’ – ally poor. For some, migration has fewer households, fewer waves and is unable or unwilling to work – the work- proved to be an effective means of escap- confined to a single province. ing poor actually constitute a significant ing poverty. There are no panel data from which to proportion of the chronically poor. The determine the numbers of chronically largest group of chronically poor people poor in Sri Lanka. It is clear, however, in rural India are casual agricultural la- Excluded minorities that although per capita GDP passed the bourers; cultivators, the second largest Excluded minorities, including ‘tribals’, US$800 hurdle in 1999, poverty persists. group. Most of the chronically poor are people of ‘low’ caste and religious mi- The proportion of the population living either landless or near-landless, and norities, find it more difficult to marshal on less than US$1/day, and the nutrition- highly dependent on wages.10 the necessary social, political and eco- ally ‘ultra poor’,9 both seem stable at Agricultural wages have been rising nomic resources to progress, and are just above 5% of the population. slowly in much of the sub-continent, and much more likely to experience long- The extent to which the 40% of Sri this is probably the best single explana- term and absolute poverty. As touched Lankans who survive on between US$1 tion for the slow but steady reduction in upon in Chapter Two, both Scheduled and US$2/day are likely to be chronically the depth of consumption poverty. How- Castes (SCs) and Scheduled Tribes (STs) poor is an empirical question, and fur- ever, getting work does not always trans- are stigmatised groups, within which ther research is needed to understand the late into exiting poverty. In agrarian many suffer extreme discrimination poverty dynamics of the ultra poor, poor economies with large casual labour mar- although the harsh oppression associated and non-poor in Sri Lanka. kets, the number of days of work ob- with untouchability has been banned.12 tained in a given period, is almost as In rural India, for example, a SC or ST important as the wage level. household was more likely to be poor in Who are the chronically Migration is often part of a broader both 1970–71 and 1981–82 than other poor in South Asia? set of livelihood strategies employed by caste households. Scheduled Caste The chronic poor in South Asia are dis- poor wage labourers. Chasing scarce, women have one of the lowest levels of proportionately made up of excluded mi- short-term, insecure, and low-paid wage literacy of all groups in India – in the norities, including tribal peoples; people labour from area to area, migrant la- 1991 Census more than 80% rural SC belonging to perceived low status castes; bourers often find themselves in a con- women were found to be illiterate. STs and casual and migrant labourers. stant battle to repay debt and maintain have literacy rates of just 40%, com- Women and girls also tend to be household consumption levels. In some pared to 54% national average, with Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P27E-01.3D.3D Folio: 73 Black plate (73,1)

Understanding chronic poverty in South Asia . 73

Table 7.2 Different approaches to chronic poverty in rural Pakistan

Sample Timeframe Source Poverty line Deﬁnition of Proportion chronic poverty chronically poor

Poorest quintile 6% 727 households from 1986/7–1988/9 Adams and Jane (income) Poorest quintile in all 3 IFPRI rural survey (12 waves) (1995) Poorest quintile years 10% (expenditure)

Poor at least 4 out of 5 7% 686 households from 1986/7–1990/1 Baulch and McCulloch periods IFPRI rural survey (5 annual waves) (1998) Poor in all 5 periods 3%

About 50% of households classified as Mean income over five Baulch and McCulloch 2100 Kcal/day – Rs poor in the first year "" 2000 (approximates years below poverty (1999) About 6% of households poorest quintile); line classified as non-poor in welfare measure real the first year income per adult equivalent Poor in all periods 5% Baulch and McCulloch "" Mean income over five (2000) years below poverty 26% line

Poorest quintile in both " " CPRC calculations 10.3% 1986 and 1991

39.7% (northern irrigated plains 34.3%, barani Mean expenditure level 1986/7–1990/1 plains 25.9%, dry " World Bank (2002) Rs. 2850 is below the poverty (2 annual waves) mountains 46.7%, line southern irrigated plains 46.4%)

Rs 7,140 (WB 1995 299 households from 1996–1999 Kurosaki (2002) adjusted for rural CPI) 63.2% rural NWFP survey 2 waves (expenditure)

Poor in both periods 43.7% – 58.3% (depending on: observed Ofﬁcial national poverty Kurosaki (2003) or ﬁtted consumption line (expenditure) values, poverty line or 90% poverty line)

Source: CPRC analysis; Yaqub 2000 only a quarter of ST women being liter- Poor women, older women, underweight children.18 Property and in- ate.13 This varies greatly from state to disabled women and heritance laws are highly gender discrim- state, with female literacy ranging from widows inatory across the South Asian region, about 88% to just 9% in 1991.14 Poor women feature prominently as a and ignorance and misapplication of While per capita incomes are lowest group of the chronically poor in South these laws often mean that women do among SCs followed by STs, tribal status Asia. They are generally less educated not even enjoy the minimal protection is more significant than caste status in (see Table 7.3), triply burdened16, less that they can afford.19 In much of north- determining poverty persistence.15 STs in well connected and informed, and often ern India and Pakistan, for example, India are often located in isolated areas unable to ensure that they benefit from strong patriarchal traditions of owner- where opportunities to diversify income husbands’ income.17 Gender divisions ship and inheritance continue to domi- earning strategies is low. within labour markets restrict the em- nate despite legal provisions to protect The chronic poverty dimension of trib- ployment opportunities for women, women’s ownership rights. In Nepal, real status is most pronounced in the con- though the demand on women to work cent constitutional changes that ensure text of social movements and conflict. is strong within poor and chronically equal property rights for women present Indigenous peoples of south-eastern Ban- poor households. a significant and positive opportunity for gladesh, for example, have only recently The position of women is particularly poor women and their children to avoid emerged from years of struggle against vulnerable to continued poverty when slipping further into deep, inescapable Bengali in-migration cum colonisation. they reach old age and/or are widowed poverty. Agitation for separate states in parts of and/or become disabled. In India, wid- Since women usually move to their India has taken root partly in response ows represent 6.5% of the total female husband’s village on marriage, they do to rising resentment within deprived re- population – 30 million in absolute not have strong support systems if they gions and tribes. terms, perhaps three times the number of are widowed. Although not always the Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P27E-01.3D.3D Folio: 74 Black plate (74,1)

74 . Chapter 7

tend to spend a large proportion of their improvement between 1986 to 2001 – Table 7.3 Gender gap in adult earnings on food, often without meeting the absolute numbers of stunted and literacy in South Asia minimum energy and nutrient require- wasted Pakistani children have grown.22 Country Difference in ments. Families facing chronic food inse- Breadwinner illness is a major cause of percentage points curity are caught in a hunger trap. The the financial deterioration for poor between female inadequacy and uncertainty of their food households – almost one-fifth of all dete- 23 and male literacy supply make it difficult for them to take rioration in Bangladesh, for example. rates (2000) advantage of any development opportu- The costs are direct (medical fees and nities that might emerge. treatments) and indirect (lost wages or Bangladesh 22.5 Despite India’s position as a net food production, care, withdrawal of children exporter, 268 million people are still from school, asset depletion and long- India 23.0 considered food insecure in India. Al- term indebtedness). Chronic diseases Maldives –0.2 most half the women aged between 15 such as TB have particularly devastating and 49, and three-quarters of children, results.24 Severe or prolonged illness or Nepal 35.6 are anaemic. Of the 204 million people accidents are more likely in very poor Pakistan 29.6 that are currently undernourished in households. Clean water, and good Sri Lanka 5.4 India, there is a significant subset of household and community sanitation, those that are unable to access two are increasingly recognised as factors in Regional Average 23.4 meals a day throughout the whole year. determining not only the health of chil- What is particularly worrying about dren but also of adults.25 Source: See Part C. low food intake is the compounding ef- The despair caused by the combina- fect it has on individual and household tion of long term hunger, ill-health and case, many widows do not receive eco- ill-health, debt and inability to work (or poverty, responsibility for older people nomic support from family or wider study), as well as rising anxiety and and other dependants, lack of employ- community unless they are taken in by stress. Low energy leaves people, notably ment opportunities or any hope in the adult sons.20 That said, relatives may children, particularly susceptible to dis- future for children, further debilitates the provide the only access to charity on ease. It is estimated that India has 20% chronically poor. Multiple deprivations which widows can depend as they get of the global child population but ac- and starvation are reported to have cul- older and more frail. However, where counts for 40% of the world’s malnour- minated in suicides by skilled power- families are poor themselves, this charity ished children.21 loom weavers in India.26 Such reports can be limited. In rural Pakistan, children by the age highlight the hopelessness and despair of five have a 62% probability of being often experienced by the desperate, fac- stunted, a 45% chance of being under- ing the prospect of chronic poverty. The hungry, weak and ill weight and a 12% probability of being Although hypertension and heart dis- Hunger and ill-health are both contribu- wasted, representing high levels of ease are commonly considered problems tors to and results of chronic poverty. chronic malnutrition. Stunting is worst of the middle class, they also are signifi- Malnutrition is not specially associated in the south-western province Balochi- cant problems for the long-term poor with poverty, but it may be with chronic stan, with a 75% probability. Further, (Box 7.1). Studies warn about heart dis- poverty. Those below the poverty line there seems to have been no ease and diabetes reaching epidemic proportions in India.27 The choices chronically poor people are forced to Box 7.1 ‘My heart feels as if it is being held with forceps’ make in order to survive can be highly detrimental to their health. Some of Poverty and hypertension in an Indian slum these decisions may have high physical and psychological costs, such as heart at- After her husband’s death, Amina Khatun* had to think of a way to support herself tacks and high blood pressure. and her two sons. Illiterate, and being from a Muslim community where women normally don’t work outside home, she had few marketable skills and limited livelihood options. She only managed to keep her house after a Dubai-based cousin Where are the chronically invested in rebuilding it after a fire. In return, Amina takes care of her cousin’s sister poor in South Asia? who has epilepsy, and the woman’s two children who have learning difficulties. Chronic poverty in South Asia has both Talking about the stress she feels and her inability to work she says, ‘Inside, my macro and micro-level features. At a re- heart feels as if it is being held with forceps. I feel a tightness inside my head. The gional level, most indicators show a sight in one eye is almost gone. I can’t see properly.’ She suffers from constant swathe of poverty cutting across eastern burning in her stomach, and often complains of a heaviness in her chest. Each and southern Pakistan, central India, time they met, Amina wept as she spoke to the researchers, especially when western Nepal, and northern and south- mentioning how she suffers when she has to accept help from relatives. She told eastern Bangladesh. Within this general them that she has felt suicidal several times, and once tried to commit suicide by ‘poverty tract’, however, there are pock- jumping into the river Krishna. ets of improvement, lower levels of pov- (*Name has been changed). erty and even relative prosperity – Source: Lalita 2003. sometimes urban areas, sometimes areas Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P27E-01.3D.3D Folio: 75 Black plate (75,1)

Understanding chronic poverty in South Asia . 75

density in the world,28 even small pock- Figure 7.2 Bangladeshi districts ets of severe distress can affect a very Figure 7.3 Pakistani districts with highest HPI large number of people. Panel data for with lowest HDI (2003) 33 (2000) 1987–88 and 2000 indicate that 15% of Up to 30 households that had descended into pov- 30.1–35 erty had experienced a shock related to a Above 35 natural disaster, suggesting that poor HDI34 <0.400 geographic capital at the most local level No data played a role. Poverty rates are highest in extremely low-lying areas that are frequently flooded, including chars (river-islands that seasonally disappear; see Box 3.2), and in tribal areas where social and geographical disadvantage overlap.29 In India, there is significant but incom- plete overlap of areas with the highest poverty rates and those with the lowest human development indicators, and of poor regions, states and districts (see Fig- ure 7.4). At the regional level, the mar- Source: Sen and Ali 2003. Source: UNDP 2003c. ginality of central and eastern India is explained largely by adverse agrarian re- dependent on remittances or strong lations, and poverty has persisted in Out of 379 districts in fifteen states, the NGO programmes. Similarly there are these regions despite a good endowment same 52 to 60 districts are consistently pockets of deprivation in otherwise well- of natural resources and a relatively identified as the most deprived, despite off regions – areas, both rural and urban, strong focus of Indian development plan- computing nine different indices with less-favoured by nature and/or man. ning on ‘backward areas’. State, district different combinations of indicators and Most poor South Asians still live in and rural indicators broadly follow this methodologies (see Map 4 in Figure 7.4). rural areas, and it is likely that the pro- general regional sketch, with one or two 80% of the districts identified are located portion of chronic poor is greater in exceptions. Urban indicators show a in one of the five states with high persis- rural areas, given the greater opportuni- markedly different trend. tence of poverty.35 ties in towns and cities. However, in Over 70% of India’s poor reside in six 20% of the most deprived districts ac- India the proportion of severely poor states: Uttar Pradesh, Bihar, Madhya cording to the multidimensional indices people in rural and urban areas is similar Pradesh, Maharashtra, West Bengal and (including one of the seven districts suf- at about 15%, indicating that urban Orissa.30 In four of these states – Bihar, fering extreme deprivation) are in Rajas- chronic poverty may be greater than Orissa, Madhya Pradesh and Uttar Pra- than. This north-western state is supposed. desh, plus Assam, persistently high levels something of an anomaly in the pattern. In Bangladesh, spatial inequalities in of poverty in excess of 30% have oc- Poverty rates are significantly below the human development are considerable, curred for several decades.31 As most all-India average, and have been declin- with the central and south-western re- central Indian states are the size of large ing much faster than average in the late gions doing relatively well (see Fig- countries – Uttar Pradesh would have 1990s. Rajasthan does not show up at ure 7.2). However, modest reductions in the world’s sixth largest population if it all on the National Sample Survey list of spatial inequalities have occurred, during were a country – numbers of people suf- regions (clusters of districts) with the the late 1990s in particular. The north- fering persistent poverty and deprivation highest rates of poverty and severe pov- west and southeast are beginning to are huge. erty (see Figure 7.4, Map 2). At the same catch up, based upon two main factors: In Assam, both income poverty and time, the state’s HDI is significantly a better-integrated national market, and human development performance de- below the all-India average, although in decreased conflict. The construction of clined strongly in the 1990s, from al- the late 1990s some improvement in this the Jamuna bridge – representing a mas- ready low levels. In the mid-1990s, 46% index has also been noted, in part due to sive public investment – helped to inte- of rural households in the lowest expen- enormous progress on education indica- grate long-neglected northern and diture class could not access two meals tors. Yet it contains one-fifth of the most western districts with the rest of the per day throughout the year, compared deprived districts in India. country, while the peace process in the to an all India average of 15%.32 Comparing Figure 7.4 Maps 2 and 4, Chittagong Hill Tracts removed some At the micro-level, severe deprivation it is clear that even within the core five obstacles to improvement in that region. is remarkably concentrated in India. Dis- persistently poor states, overlap is There are also pockets of poverty in trict-level multidimensional indices have sketchy, and that there are several re- areas much smaller than districts, due been developed combining indicators of gions that the National Sample Survey to variations in agro-ecological vulner- literacy and enrolment, infant mortality identifies as poorest that do not contain ability, or the presence of minority rate, agricultural productivity, and infra- any of the most deprived districts. As populations. And, as Bangladesh is structural development – low levels of has been found in Vietnam,36 there is characterised by the highest population which can reflect persistent deprivation. not the expected near-universal or exact Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P27E-01.3D.3D Folio: 76 Black plate (76,1)

76 . Chapter 7 correspondence between changing levels . Second, forested regions, especially in degree to which Indian states have miti- of income poverty and other dimensions hilly regions with predominance of gated the effects of drought. On the face of deprivation. The reasons for this are tribal populations, with limited access of it, drought-related chronic poverty is likely to relate to differing patterns of to natural resources, information and most likely in arid areas in poorly gov- economic growth and socio-economic markets.37 erned states. However, many dryland inequality. These areas are not only persistently in- populations have been able to develop Many remote rural areas in India are come poor, but are generally much less coping strategies to facilitate their resil- largely populated by scheduled tribes, well-endowed with human capabilities. ience to drought, including groundwater who face extreme marginalisation and Tribal populations living in forested areas development, economic diversification discrimination. In general, two types of affected by consecutive years of drought, with infrastructural development, area are viewed as less-favoured on the such as south-western Madhya Pradesh, drought relief safety nets, and migration. basis of agro-ecological and socio- face extreme deprivation.38 Geography The latter is especially significant. For- economic conditions. These areas also is only part of the reason why access to est-based regions have few of these pos- exist in less poor states. resources may be limited. See Box 7.2 sibilities. Migration is more likely to be . First, large tracts of dryland char- for a discussion of the effects of some from distress, since regions of economic acterised by frequent crop failure government lease oriented policies on growth are often further away, and mar- and sporadic opportunities for traditional access to resources in Orrisa. kets function less well so that invest- employment. There is significant variation in the ments at home have less effect.39

Figure 7.4, Map 1 Indian states Figure 7.4, Map 2 Indian regions with highest proportions of the with above population below national poverty and/or severe average poverty lines (1993–4) proportion of the population below national poverty line (1993–4)

Rural only Bihar (central, northern, southern) Uttar Pradesh (central) West Bengal (Himalayan) Urban only Karnataka (inland northern) Madhya Pradesh (central) Maharashtra (inland eastern, inland northern) Rural and urban Madhya Pradesh (southwestern) Maharashtra (inland central) Orissa (southern) Source: Derived from Mehta et al. 2001 Uttar Pradesh (southern) Tables 2 and 4. Source: Derived from Mehta et al. 2001 Tables 2 and 4.

Figure 7.4, Map 3 Indian states Figure 7.4, Map 4 India’s most deprived districts with below average HDI (1991) Extremely deprived districts on multidimensional indices (54)

Most deprived districts on all nine multidimensional indices (7)

Multidimensional indices developed using different combinations of: – education (female literacy, total literacy, 11–13 year old in school), – health (IMR) – income (agricultural productivity) – infrastructural development (roads, proportion cultivated land under irrigation, electricity, toilet facilities, post/telegraph) Source: Derived from Mehta et al. 2001 Tables 2 and 4. Source: Derived from Mehta et al. 2001 Tables 2 and 4. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P27E-01.3D.3D Folio: 77 Black plate (77,1)

Understanding chronic poverty in South Asia . 77

Table 7.4 Poorest Indian states

States with the highest number of people in poverty (1999–2000) UP, Bihar, MP, MA, WB, Orissa . 72% of India’s poor and 56% of the population live in these six states. . 48% of India’s poor and 36% of the population live in UP, Bihar and MP

States with above average proportions of people in poverty . 1993–1994 Bihar, Orissa, MP, Assam, UP, MA . 1999–2000 Orissa, Bihar, MP, Assam, UP, WB

States with above average proportions of the rural population in poverty (1993–4) Bihar, Orissa, Assam, UP, WB, MP, MA

States with above average proportions of the rural population in severe poverty Bihar, Orissa, UP, MP, MA (three-quarters poverty line) (1993–4)

States with above average proportions of the urban population in poverty (1993–4) MP, Orissa, KA, TN, AP, UP, MA, Bihar

States with above average proportions of the urban population in severe poverty MP, Orissa, KA, MA, TN, UP, AP (three-quarters poverty line) (1993–4)

States with below average HDI (1991) Bihar, UP, MP, Orissa, RA, Assam, AP

States with above average HPI (1991) Bihar, UP, Assam, Orissa, RA, MP, AP

States with above average rural hunger (1993–4) Orissa, WB, Kerala, Assam, Bihar

States with above average urban hunger (1993–4) Kerala, Orissa, WB, Assam, Bihar, TN, AP

AP (Andhra Pradesh); KA (Karnataka); MA (Maharashtra); MP (Madhya Pradesh); RA (Rajasthan); TN (Tamil Nadu); UP (Uttar Pradesh); WB (West Bengal).

Urban poverty and hunger, particu- relations – the Federally-Administered ‘contours of conflict’.40 The absolute larly urban hunger, do not conform to Tribal Areas in the west, and large areas poverty found in north-eastern Sri Lanka the broad notion that persistent and ab- of Balochistan, North West Frontier and mid-west Nepal is likely to be rela- solute poverty is concentrated in central Province and Sindh. Third, inner city and tively intractable, even within the current and north-eastern India. The southern urban periphery slums, particularly in context of peace processes. Violent insur- states of Karnataka, Andhra Pradesh Karachi and in the Afghan refugee camps gency has increased the isolation of re- and Tamil Nadu have above average around Peshawar, some of them long-es- gions with low levels of ‘geographic rates of urban poverty and urban hun- tablished. The extent to which the capital’. In Sri Lanka, outside of conflict ger, while Kerala – India’s showcase changed political and security context in zones – for which there is very limited state in terms of high levels of human Afghanistan will foster escape from data, poverty is concentrated in arid, development – has the highest and third chronic poverty in that country, much unirrigated rural areas. Rates of poverty highest urban and rural hunger rates in less among the hundreds of thousands of and severe poverty are almost twice as India. refugees in Pakistan, remains to be seen. high in rural and estate (plantation) Andhra Pradesh suffers a low and de- Chronic poverty tends to follow the areas as in urban areas.41 clining HDI in contrast to its low levels of income poverty. This may suggest that growth and public investment have been Box 7.2 Access to non-timber forest products in Orissa less than pro-poor, with particularly adverse effects on the urban population. In India, rural poverty is generally considered to be related to a lack of access to On the other hand, Karnataka, and in cultivatable land or its low productivity. Approximately 100 million people living in particular Kerala and Tamil Nadu have and around forests in India derive their livelihood support from the collection and strong HDIs and governance is relatively marketing of non-timber forest products (NTFPs), making the issue of rights and pro-poor. Urban poverty is clearly a spe- access to, and income from NTFPs vital to the sustenance and livelihood of forest cific and complex problem. dwellers. In Pakistan, available evidence sug- Some government lease-oriented policies have given private companies, gests that chronic poverty exists in sev- monopoly access to some NTFPs including kendu, bamboo and sal seed. eral areas, and is harshest where Attempts to remedy the situation, by enabling gram panchayats (local government) ecological and social deprivation overlap to regulate the purchase, procurement and trade of NTFPs, in order to provide (see Figure 7.3). First are the harsh envi- primary gatherers with a fair price, have been largely impotent. Though three years ronments – the mountainous Northern have passed since the gram panchayats were accorded control, the market Areas, and arid parts of Balochistan and situation has not improved. Most traders are unregistered, and Panchayats make Sindh in the west and south. Second, no efforts to enforce the prices that are fixed by the District Magistrates. This has areas dominated by oppressive tribal been partly responsible for reducing traditional access to resources. and/or feudal agrarian and gender Source: Saxena 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P27E-01.3D.3D Folio: 78 Black plate (78,1)

78 . Chapter 7

Notes 1. In this context, the term minority is used to 11. de Haan and Rogaly 2002: 14. 31. Mehta and Shah 2003. distinguish groups that experience 12. In much the same way that purdah transcends 32. Mehta and Shah 2001. discrimination and particular forms of Islam and influences the lives of Hindu women 33. HPI = Human Poverty Index = composite index exclusion and not only those which constitute a in northern India in particular, the strictures of representing: deprivation in longevity – small proportion of national population. In caste operate outside of Hinduism and of India, probability of dying before age 40; deprivation India, for example, this broadly refers to and perceptions of low caste continue to foster in knowledge – adult illiteracy, children aged 6– scheduled caste and scheduled tribe persistent poverty throughout the region. 10 not in school; and deprivation in economic populations. 13. Kumar 2003. provisioning – share of population without 2. Sen 2003. 14. Mehta and Shah 2003. access to health services (children not 3. Gaiha 1989. 15. Bhide 2003. immunised, deliveries not attended by trained 4. Bhide and Mehta 2003. 16. With responsibilities concerning household worker), safe tubewell water, electricity; 5. Gaiha and Deolalikar 1993. productive activities, household reproduction children under 5 malnourished. 0.00 = no 6. Gaiha and Imai 2003. activities and community and social human poverty. 7. Krishna 2003. maintenance obligations. 34. HDI = Human Development Index = composite 8. The IFPRI (International Food Policy Research 17. MHHDC 2000. index representing income, life expectancy and Institute) Pakistan Panel Survey was 18. Dreze and Sen 2002: 263, in Amis 2003. adult literacy, gross combined enrolment. administered in 14 waves over five years from 19. MHHDC 2000. 1.00 = complete human development. 1986–1991, to approximately 800 rural 20. Dreze and Sen 2002: 265 in Amis 2003. 35. Aasha Kapur Mehta, Multidimensional Poverty households. Analysis undertaken on poverty 21. Measham and Chatterjee 1999. in India: District Level Estimates, from Mehta, dynamics has used data on 686 households over 22. UNDP 2003. Ghosh, Chatterjee and Menon (edited) Chronic five years or 727 over three. The surveys were 23. Sen 2003. Poverty in India, CPRC-IIPA, New Delhi, 2003. conducted in three less-developed districts of 24. Kamolratankul et al. 2000 in Pryer et al. 2003. 36. Baulch and Masset 2003. Punjab, Sindh and NWFP, and one relatively 25. Mehta, Panigrahi, and Sivramkrishna 2003. 37. Mehta and Shah 2003. well-developed and irrigated Punjab district. 26. Kala and Mehta 2002. 38. Shah and Sah 2003. 9. Nanayakkara 1994, in Tudawe 2002. The ultra 27. WHO 2003. 39. Mehta and Shah 2003. poor are households who spend more than 80% 28. Excluding city states and small islands. 40. Goodhand 2001. of their total expenditure on food, but achieve 29. Sen 2003; Sen and Ali 2003. 41. Tudawe 2001a. less than 80% of their food energy requirement. 30. Including the new states of Uttaranchal, 10. Gaiha 1989, in Bhide and Mehta 2003. Jharkhand and Chhatisgarh. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P28E-02.3D.3D Folio: 79 Black plate (79,1)

2001 these absolute gains had been largely erased, and the picture worsened in 2002 due to negative growth. Persistent poverty in Latin America is, Understanding chronic to quite a large extent, a distributive problem (see Table 8.2). Where growth poverty in Latin America has been achieved, its potential positive and the Caribbean impact on the poor has been undermined 8 by inequality – which also holds back growth itself. Regional levels of inequality are among the highest in the world While rates of absolute poverty in Latin America and the Caribbean (LAC and did not decline significantly in any region) are relatively low compared to other developing regions, the propor- Latin American country during the tion of the poor that are chronically poor is relatively high. Despite wide- 1990s; four countries showed significant spread economic growth in the 1990s, the picture has not improved, and at increases.4 present rates the region will not meet the MDG of halving absolute poverty There are two main processes behind by 2015. Persistent poverty is generally attributed to the high levels of the persistence of inequality. First, neo- inequality in the region, much of which is associated with race and ethnicity, liberal reforms have contributed to wid- which give people few chances to escape poverty. ening economic inequalities as the benefits of growth have been very unequally Poverty trends in Latin expenditure has barely risen – on average spread. Second, underlying socio-eco- less than half a percent – despite economic structures are deeply rooted in America and the nomic recovery and positive growth in Latin America’s colonial past. Inequality Caribbean the 1990s. In some countries, such as persists on the basis of the concentration Although aggregate rates of absolute Peru, poverty rates rose and poverty of land in the hands of the very few; ex- poverty in the LAC region are relatively tractive economic industry (rubber, min- low, national rates range from negligible Persistent poverty in Latin erals, oil and natural gas); commodity- numbers in countries like the Dominican America is largely a distributive based exports (sugar, bananas and cof- Republic, up to a fifth of the population problem. Inequality undermines fee); and the history of slavery and in- in Ecuador, El Salvador and Paraguay, the potentially positive impacts dentured labour that irreversibly and almost a quarter in Honduras. In of growth on the poor, as well transformed societies. Racial and ethnic Bolivia and Haiti these figures are likely as hindering growth itself. stratification form the bases for exclu- to be much higher. Although Latin sion and adverse incorporation. Further, America and the Caribbean has some of many countries of the region have long the best human development indicators, gaps widened alongside substantial eco- been characterised by powerful, central- on average, there is significant diversity nomic growth. It is estimated that the ised and often clientelistic states, as well between countries (see Table 8.1). Haiti absolute number of poor1 people rose by as several decades of military dictator- plus most of Central America and Boli- nearly 11 million between 1990 and ship, and not yet been transformed by via fare the worst. 1999 – 7.6 million in the last two years.2 recent democratic reforms. These politi- Although human development indica- The numbers of indigent3 people de- cal structures have further exacerbated tors have improved over the past two clined by about 4 million over the same the concentration of power and accentu- decades, aggregate per capita household period, but estimates suggest that by ated prevailing inequalities.

Figure 8.1 Chronic poverty in Latin America and the Caribbean 5% of the world's One in every two or three chronically poor live poor people in the in the Latin America / LAC region is Caribbean region chronically poor

About one in every 25 people in the LAC region is chronically poor Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P28E-02.3D.3D Folio: 80 Black plate (80,1)

80 . Chapter 8

Table 8.1 Summary of poverty indicators in Latin Table 8.2 Income polarisation in Latin America America and the Caribbean (US$, yearly PPP-adjusted GDP per capita)

LAC Time Poorest 1% Richest 1% Income Ratio Percentage of people living on less than US$1/day, 11.5 1989–1999a 1970 112 40,711 363 The number of percentage points by which the poor 34.7 a fall below the poverty line, 1989–1999 1975 170 46,556 274 Under-five mortality rate (per 1,000 live births), 2001 36.5 Infant mortality rate (per 1,000 live births), 2000 31.7 1980 184 43,685 237 Proporton of under five children stunted, 1992–2000a 17 Proporton of under five children severely stunted, 5.4 a 1985 193 54,929 285 1992–2000 Life expectancy at birth, 2000 69.8 Adult illiteracy rate, female, 2000 13.9 1990 180 64,948 361 Adult illiteracy rate, male, 2000 11.8 a: Data refer to the most recent year available during the period 1995 159 66,363 417 specified. Source: Part C. Source: Londonõ and Sze´kely, 1997, in Wheeler (2003).

How many people are better educated members tend to be bet- educated, and among girls in rural areas chronically poor in Latin ter off and experience lower levels of in- and boys in cities. Few countries in the come variability. region have seen drop-out rates decrease America and the In a study of Monterrey, Mexico,8 significantly over the past decade, and Caribbean? both educational and occupational mobi- rural-urban gaps remain wide, fostering Best estimates are that between 30% and lity have increased since 1968. However, the intergenerational persistence of pov- 40% of the absolute poor population in while, overall levels of education have erty and inequality. the LAC region is chronically poor: be- improved, the ‘social minimum’ for edu- tween 16 and 22 million people. cation has also increased across all social Poverty is a way of being in which individuals become unable to classes, and educational opportunities re- exercise their rights. The cultural main stratified. Opportunities for effec- Who are the chronically deprivation imposed by the absolute tive upward mobility are still scarce. absence of rights, suppresses human poor in Latin America Further, average real incomes for lower and the Caribbean? dignity, and leads to material white-collar workers have decreased deprivation and political exclusion.6 Chronically poor households in the LAC since the 1960s, so occupational mobility region tend to be of indigenous or Afri- from manual jobs into lower white-col- can descent, with high dependency ra- lar jobs does not automatically translate Race and ethnicity tios, and low levels of education. Very into better economic status. In Brazil, (as Race and ethnicity are strongly linked to little data is available on the relationship with South Africa), a strong legacy of prevailing structures of inequality in the between chronic poverty and disability social inequality means that even declin- LAC region, and likely to be major pre- in the LAC region.5 ing (but still significant – see Figure 8.4) dictors of chronic poverty. Often it is not educational inequality is not yet translat- only a lack of access to basic services ing into reduced income inequality.9 and political power that makes indige- Those with less education This is particularly true for marginal- nous people poor, but an overarching ex- Based on household survey data from 18 ised groups. For indigenous peoples in clusion from citizenship11 that denies LAC countries, education and occupa- Latin America, discrimination in labour them such access, as well as less tangible tion were found to be the two most im- markets and limited opportunities for aspects of well-being such as dignity and portant factors in determining poverty.7 quality education mean that education is cultural integrity. Whilst the case for occupation appears less strongly correlated with income than Regions within Latin America with to be stronger, as it is related to incomes, for other socio-economic groups.10 large indigenous populations, such as the the relationship between education and While there are few children who have Pan-Andean region, the Central Ameri- poverty is highly contextual. less than one year of education, there are can lowlands and south-eastern Mexico, The average difference between the extremely high non-completion rates and areas with large populations of Afri- number of years in school of the poorest across the region. Dropping out – or can descent such as north-eastern Brazil, and richest quintiles is 6 years, and the being pushed out by poverty and under- tend to be the poorest. In Brazil 45% of study suggests that on average, 29% of financed schools – is highest in rural the poorest decile is black, and 85% of poverty would be eliminated if there was areas, among those living in the poorest the richest 1% is white.12 This has no such disparity. Households with households, with mothers who are less hardly changed over the past century, Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P28E-02.3D.3D Folio: 81 Black plate (81,1)

Understanding chronic poverty in Latin America and the Caribbean . 81 underscoring the resilience of these Figure 8.2 Racial inequality and education in Brazil, 1999 socio-economic structures to change. In 97% Mexico, Oaxaca is both the poorest state and that with the highest proportion of Black/mixed 87% indigenous people.13 The only time that White 75% the problems of indigenous regions seem 6.6 to receive any serious consideration, ei- years ther from national authorities or key in- 57% ternational institutions, is when they are the sites of armed rebellion, as with the 47% 4.4 years Zapatista uprising in south-east Mexico.14 Indigenous migrants to cities and 26% other non-indigenous areas are often 20% among the poorest, but they may have higher incomes and more opportunities 8% to access basic services than in their indigenous areas. Illiteracy rate <4 years study <8 years study <11 years study Mean years study . In Panama, 70% of the indigenous population is extremely poor com- (>= 15 years old) (>= 25 years old) Source: PNAD, 1999, in Henriques, 2001. pared to 13% of the non-indigenous population.15 . Virtually all monolingual indigenous minimal if any reduction of their non- perhaps most importantly, potentially speakers are extremely poor. waged household work, and rarely is different expenditure patterns. The ex- . While 87% of indigenous people living enough to reduce poverty. Nor has it tent to which headship is linked to per- within indigenous areas are extremely contributed to sufficiently improved sta- sistent poverty needs to take into poor, 25% of those living outside tus for women in the household; domes- account a range of factors. these areas are below the extreme pov- tic violence against women is rising erty line. significantly. Where are the chronically The evidence that households headed by women – of which the proportion is poor in Latin America The working poor large and in many cases growing in the and the Caribbean Several million new jobs generated by LAC region – are disproportionately Urban poverty is particularly significant growth in the 1990s have been largely in among the extreme poor is mixed. In in the LAC region – 64% of the poor the informal sector, low-waged, insecure some countries with relatively low over- and 75% of the total population of Latin and unprotected. The number of ‘work- all poverty levels, female-headed house- America live in urban areas.18 However, ing poor’ has thus increased. A 2000 holds tend to be disproportionately the probability of being poor or ex- study suggested that three-quarters of represented among the extreme poor. In tremely poor is still much higher in the the employed population in Latin Amer- Costa Rica, for example, they make up rural areas. It is 37% if you live in a ica does not generate enough income 28% of total households but 56% of the town or city, but 63% if you live in a from their jobs to surpass the poverty extreme poor. In poorer countries like rural area.19 line,16 fostering a dependence of the Honduras and Nicaragua, households working poor on a portfolio of activities headed by women are only marginally alongside a ‘main’ job. In urban areas, among the poorest.17 Figures such as Rural chronic poverty the highest rates of poverty were found these do not take into account the prob- Several primarily rural regions stand out to occur among those employed domesti- lematic definitions of ‘female-headed- as persistently poor, such as the pan- cally, as manual workers, or informally ness’, nor the poverty gap between male- Andean region including portions of in commerce or services. and female-headed households, nor, Bolivia, Peru, Ecuador and Colombia.

Women and women-headed Box 8.1 Sebastiana’s story, Ecuador households ‘Misery did not separate us, because my mother refused to hand us over to the Although women are increasingly incor- landowner, who offered so many advantages in return. Our only protection was our porated into the economy, they tend to work, first that of my grandparents in haciendas in the Puno region, then the work enter at the lowest levels, where low of my parents, which took us from Llave to Santa Rosa, from Taquile to the wages, limited security and gender-based Tiquina Strait, from Yunguyo to Moya de Ayaviri, and from there to Cusco. It was discrimination all undermine the pov- there that my father worked on building the Huatanay River canal, helped build the erty-reducing potential of paid work. regional hospital, worked in the railway station, and also as a bricklayer and The contribution of poor women’s journeyman. My mother sold food on the street, sold everything she could, was a wages to household income is increas- labourer like me, my brothers, and now like my male children.’ ingly signiﬁcant, but this has led to Source: Ochoa (2001: 56) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P28E-02.3D.3D Folio: 82 Black plate (82,1)

82 . Chapter 8

Megacities are increasingly divided Box 8.2 Indigenous poverty and vulnerability in Bolivia into favelas for the poor and gated com- Bolivia has one of the largest indigenous populations in Latin America. 63% of the munities for the rich. There are also new total population is unable to satisfy their basic education, health and housing favelas accommodating informal sector needs. The incidence of poverty is particularly high in rural areas: in 1997, 77.3% construction workers and other service of the rural population was considered poor and 58.2% severely poor, while 50.7% providers who work within the gated of the urban population was poor and 21.6% severely poor. The economy is highly communities. Residential location se- dependent on mining – tin and other mineral mines have replaced the famous verely limits the possibilities of upward silver mines of the colonial period. Natural gas and mineral exports account for the mobility: the prejudices of the elite and large majority of Bolivian exports – and the price for these goods is determined in middle classes against favela dwellers ex- the global commodity market, making vulnerability to external shocks in these clude the latter from better jobs, and the markets one of Bolivia’s biggest problems in addressing poverty.24 Bolivia middle classes have opted out of public epitomises the characteristics common to other geographic concentrations of services and into private health and edu- poverty in Latin America. cation, leaving wide quality differences Source: Wheeler (2003) between service provision. Although the Rio de Janeiro favelas are racially mixed, there is evidence that Race and ethnicity explain much of the reduced. Also, day labouring for low the white favela population has been far geographic concentration of persistent wages is often the only available work, more successful than the black popula- poverty: in this Pan-Andean region, the sometimes bonded by debt. tion in moving out to neighbouring resi- northeast of Brazil, where 60% of Bra- People in these regions are excluded dential areas over the last three decades. zil’s poor are located20, and southwest from both quality public services and Discrimination based on skin colour is Mexico, where poverty is increasing de- their broader citizenship rights. They widely perceived as the greatest obstacle spite a reduction in the national pov- also tend to be adversely incorporated faced by the urban poor.23 erty level.21 A similar pattern is found into very limiting economic relation- Many in the favelas are second or in the Caribbean lowlands of Central ships. World markets have provided vol- third generation migrants from rural America. atile and even declining terms of trade areas, and the favelas continue to grow Access (or rather the lack of access) to for traditional products from these re- with new illegal land occupations in re- social services also has a powerful role. gions, and alternatives are hard to come sponse to employment opportunities. In rural Peru, consumption growth is as- by. The favelas are afﬂicted by rising rates of sociated with local provision of medical crime, violence, murder and the associ- care, the level of education, and the Chronic poverty in ated stigma and immobility, as well as prevalence of diarrhoeal diseases, more precarious environmental conditions. than it is with other geographic differ- megacities and smaller Hill- or roadside location, for example, ences (such as altitude, road network towns is detrimental to lives and livelihoods in density, and the percentage of paved Latin America is not only highly urban- the favelas. The discrimination faced by roads).22 ised, but it also contains two of the residents, combined with fear of rising In persistently poor rural areas, health world’s largest cities and several other gun and drug crime, renders a very real and education services tend to be weak. megacities of around ten million. The pe- sense of physical and psychological vul- Inadequate rural land reforms mean that ripheries and marginal areas of the cities nerability and isolation among the poor, huge historical disparities between land comprise ‘illegal’ settlements of poor despite any improvements in physical owners and labourers have not been communities. infrastructure.

Notes 1. ECLAC deﬁnition – income less than twice the 8. Solis 2002, in Wheeler 2003 15. Vakis and Lindert 2000 cost of basic food basket. 9. Lam 1999 16. CEPAL 2000, in Wheeler 2003 2. ECLAC 2002 Social Panorama of Latin 10. Castan˜ eda and Aldaz-Carroll 1999 17. CEPAL 2002 I in Wheeler 2003 America. 11. Citizenship requires a state which recognises the 18. CEPAL 2001 in Wheeler 2003 3. ECLAC deﬁnition – income less than the cost of formal equality of citizens, regardless of their 19. Wheeler 2003 basic food basket. origin and poverty levels. A notion of political 20. Wheeler 2003 4. Argentina, Bolivia, El Salvador and Nicaragua community may be needed to allocate resources 21. Wheeler 2003 (Sze´kely 2001) so that substantial equality can be achieved. 22. De Vreyer and Mesple-Somps 2003 5. Yeo and Moore 2003, in Wheeler 2003 12. IPEA 2002, in Wheeler 2003 23. Perlman 2003 6. Dagnino 2003: 5 13. Wheeler 2003 24. Andersen and Nina (2000) in Wheeler (2003) 7. Attanasio and Sze´kely 1999, in Wheeler 2003 14. Gonzales-Parra and Perez-Bustillo 2001 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P29E-01.3D.3DBlack Folio: plate 83 (83,1)

majority of those living in the region, as more and more people become reliant on social transfers as well as informal and in-kind incomes. Alongside this wide- Understanding spread decline, however, is rising inequality: only an elite minority are ex- chronic poverty in periencing significant economic improve- transitional countries ments. Human development in this 9 region is summarised in Table 9.1. It is not easy to compare poverty rates across the pre- and post-transition peri- Over the last 10 to 15 years Central Asia, the Balkans, East and Central ods as officially no poverty or unemploy- Europe and the former Soviet Union have seen persistent income and ment existed during socialism and there asset poverty, catastrophic declines in capabilities at the household level, remains today limited data on poverty and soaring levels of preventable deaths in tens of millions of previously dynamics. Those data that are available secure households.1 The loss of social protection has hit hard. Rapid are highly unreliable. Indications point economic and political liberalisation in the early 1990s contributed to rising to a sharp increase in poverty as a result inequality and the creation of long-term poverty. Maldistribution and ex- of regime change; however, poverty inci- tremely unequal development, exacerbated by domestic politics and elite dence was probably already consider- capture of resources,2 is entrenching chronic poverty in the region.3 able, particularly in Central Asia and Prospects within the region differ. In the poorer Central Asian econo- other predominantly rural parts of the mies, chronic poverty is a problem of growth and economic opportunity as former Soviet Union, where international well as distribution. Countries like Kyrgyzstan and Tajikistan may be estimates put the figures at around 30% headed for deep chronic poverty and exclusion from, or extremely adverse of the population.4 For most countries it incorporation into, the world economy. In Central Asia, natural resource is unknown whether many households constraints, such as water shortages, are also major contributors to pov- are poor for a limited duration, or fewer erty. In parts of East-Central Europe, ethnicity – specifically being a Roma households poor for longer periods of – remains the most significant factor in chronic poverty, and, in several time.5 parts of the region (especially but not only in Tajikistan and Armenia/Azer- Many transitional countries experi- baijan) inter-ethnic/community conflict and violence is also significant. enced pre-transition recessions. This was particularly severe in Poland between processes are involved, particularly 1989 and 1992. Although not officially Poverty trends in changes in the level, distribution, and recognised, some poverty was extreme transitional countries structure of incomes, and profound and income mobility was low, indicating Although there are marked differences in changes in economic systems, which significant long-term poverty. Indeed, changes in average incomes between have had both immediate and longer- poverty6 rose during 1989–1992 from Central and Eastern Europe, the Balkan term effects. These include unemploy- 15 to 17%, and ‘repeated poverty’ (pov- Countries, and the former Soviet Union, ment – caused by the loss of state enter- erty for more than a year) increased at real disposable household incomes have prise subsidies and economic disruption twice the overall poverty rate, although generally declined, due to increasing in- at the beginning of transition, and by a transitory poverty decreased. House- equality and inflation. reduction in state-provided social trans- holds experiencing poverty over two The nature of poverty has changed fers and services – as well as high infla- consecutive years increased from 45% of during the economic and political transition. Indeed, transition has been a all poor in 1988–89 to 72% in 1991–92 tion from socialism. Several interrelated process of increasing hardship for the but then declined during the growth

Table 9.1 Summary of poverty indicators in transitional countries

Percentage of Average Under-5 people living shortfall of mortality rate on less than poor below (per 1,000 Infant Stunting <-2 Life Life Adult Adult US$1/day US$1/day (%), live births), mortality rate, s.d., 1992– expectancy, expectancy, illiteracy rate, illiteracy rate, Region* 1989–1999 1989–99 2001 2000 2000 female, 2000 male, 2000 female, 2000 male, 2000

Balkans 3 24 21 20 10 74 67 4 2 Former Soviet Union and 5212720147362<1<1 Central and Eastern Europe Central Asia 12 26 72 54 25 71 64 1 <1

Transitional countries 62141321673631<1

* Balkans = Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Macedonia, Romania, Slovenia, Yugoslavia; Former Soviet Union and Central and Eastern Europe = Armenia, Azerbaijan, Belarus, Czech Republic, Estonia, Georgia, Hungary, Latvia, Lithuania, Moldova, Poland, Russian Federation, Slovakia, Ukraine Central Asia = Kazakhstan, Kyrgyz Republic, Tajikistan, Turkmenistan, Uzbekistan Figures are rounded. The regional averages differ from those in Part C because Turkey has been excluded from the analysis. Source: See Part C. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P29E-01.3D.3DBlack Folio: plate 84 (84,1)

84 . Chapter 9 years 1993–1996, to 60–62%.7 Balkans, and in Central Asia, the remote rather well economically – for example, In contrast, poverty and inequality are rural poor in particular. the ethnic Hungarians living in northern low in Hungary compared to other tran- and central Romania are economically 13 sitional economies. Although rates are Who are the chronically better-off than ethnic Romanians – dis- lower, poverty seems more entrenched, crimination by majority populations is concentrated among the poorly edu- poor in transitional common. The Roma are the largest ex- cated, those living in remote rural areas, countries? cluded and vulnerable group in Central and those weakly positioned within the The same correlates of poverty fre- and Eastern Europe. They have poor ac- labour market, as well as the Roma. quently arise in transitional countries, cess to services, and few productive 7.5% of the Hungarian population was for example household composition and household assets. The World Bank con- long-term poor (a poverty line of 50% of labour market status. Historically, asset siders their situation ‘the biggest chal- mean equivalent income) between 1992 status has not been measured and this lenge to poverty alleviation in Central and 1997. This figure is based on house- may slant discussion of poverty problems and Eastern Europe’.14 hold data, and therefore omits the home- in transition away from asset ownership Roma households are found among less and the institutionalised. Although and access and towards income.10 Where the poorest in different studies over time, these two groups are relatively small in poverty is causally related to a perma- and in all countries in the region. Histor- size – roughly 20,000–30,000 people are nent household feature (such as ethnic- ically a stateless people, the Roma today homeless within a total population of ity), chronic poverty can perhaps be make up an estimated seven to nine mil- about 10 million – they are very likely to inferred without time series data. lion people throughout the region, have large proportions of people living In Poland, the households with the although they are a minority presence in in chronic poverty.8 highest risk of falling into chronic each country. Ethnicity is the most im- For Russia, using either a consumption poverty are those with unemployed portant contributor to chronic poverty in or income poverty line, about 50% of members – 58% more likely than those Hungary, where one-third of the chronic- the population was living below the pov- with an employed member, for each ad- ally poor are Roma, accounting for only erty line by mid-1992. Between 1992 ditional year of unemployment). Those 4–5% of the population, and 53% of the and 1996, both overall poverty and tran- households having mainly wage income Roma are long-term poor. There are, of sitory poverty decreased considerably, are more likely to live in chronic poverty course, large variations in income and but chronic poverty moved very little, re- compared to those mainly relying on so- wealth within the Roma community, but maining at about 10%.9 cial transfer income (300% more likely as a group they tend to experience signifi- in urban areas, 55% in rural areas).11 cant social exclusion by majority popula- How many people are Three-quarters of the chronically poor in tions and often live in exclusively Roma Hungary are unemployed. They often settlements, which are poorly serviced chronically poor in live in low-growth areas and in places with limited access to electricity, gas and transitional countries? where poverty reduction policies and ac- running water, a lack of sanitary facilities Best estimates are that between 10% and cess to information are less effective.12 and sewerage, and poor quality housing. 20% of the absolute poor population in Chronic poverty in Russia is statistically transitional countries is chronically poor: significantly related to location, human Women between 2 and 5 million people. Chronic capital and asset wealth. Households The economic position of women in poverty in Russia and Uzbekistan ac- with higher dependency ratios are more transitional countries has deteriorated. counts for a significant proportion of likely to be poor. Under the socialist system, they experi- this number. A large majority of the enced greater equality in pay, had high chronically poor in Central and Eastern Ethnic minorities labour market participation, as well as Europe are members of the community Ethnic minorities in transitional coun- state-provided child care and health of an estimated 7–9 million Roma living tries are generally more prone to persis- care. Women are increasingly triply- throughout East Central Europe and the tent poverty. While some minorities fare burdened,15 particularly when husbands

Table 9.2 Households characteristics of Roma and non-Roma in Bulgaria, Hungary, and Romania

% of households in poverty

Bulgaria Hungary Romania

Expenditure based poverty lines Roma others Roma others Roma others

50 % of median, per equivalent adult 36.1 3.8 24.5 4.5 39.5 10.9

50 % of median, per capita 37.2 3.4 26.3 3.6 43.1 11.1

PPP $ 2.15 per capita per day 41.4 4.1 6.6 0.5 37.6 7.3

PPP $ 4.30 per capita per day 80.1 36.8 40.3 6.9 68.8 29.5

Note: In the source there are ‘PPP $ 2.15 per capita’ and ‘PPP $ 4.30 per capita’ poverty lines. We assume this is expenditures per day. Source: Revenga et al. 2002: 13 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P29E-01.3D.3DBlack Folio: plate 85 (85,1)

Understanding chronic poverty in transitional countries . 85 migrate for work. They are vulnerable to around 10–15% of school-age children function poorly, and rural areas in the layoffs and experience pay disparities. are not attending school, while 25% of 1990s had low monetisation, often rely- The collapse of state-provided support children miss 20 or more school days per ing on barter. This meant that people and price hikes associated with privatisa- year. Government spot-checks have re- had very limited cash resources. To some tion mean that health and child care is vealed that up to 20% of primary and extent this has resulted in the re-emer- now often out of their reach and must be secondary school pupils do not attend gence of feudal social relationships, often provided within the household. school regularly.19 In some cases, this is based on debt and huge interest rates Particularly in Central Asia, poverty associated with discrimination against where people enter arrangements like may also be sharpening gender inequal- minority groups. The Turks and Roma share-cropping on highly disadvanta- ities, with intergenerational implications children in Bulgaria, for example, experi- geous terms.22 With limited alternative as families resort to practices to keep enced a sharp fall in education levels be- income sources, people sometimes resort costs down. Rural parents often have to tween 1995–2000, with secondary to extreme measures such as drug traf- decide whom among their children to school attendance falling to 15% com- ficking, international prostitution and il- send to school, and in such situations pared to 50% for other households.20 legal migration. Whilst these measures boys usually receive precedence, while Since the fall of the Soviet Union in may aid the escape from poverty, they girls are married off as a means of pay- 1991, some countries have seen a sharp often present a serious risk to life and ing the fees. Reports from Uzbekistan increase in child abandonment due to health.23 suggest a growing incidence of marrying economic hardship. Dinara, a 35-year- Households living in remote rural re- girls off at a young age (before the end old resident of Osh, abandoned her four- gions are often older, less well-educated of compulsory education) to cut down day old daughter at a maternity home. ‘I and less mobile than the rest of the pop- on the number of mouths to feed, again cannot feed the elder five children, my ulation, limiting the prospect of escape curtailing girls’ education. These mar- husband is jobless and I get only US $25 from poverty. riages are illegal and so cannot be regis- a month [salary]. Maybe those who will tered. This makes the woman vulnerable adopt her will be able to give her a prop- Single sector/enterprise if the couple subsequently divorce. er upbringing and education’.21 Infants settlements Often, a young wife is left with the chil- with weak health and congenital impair- These settlements emerged during the so- dren and has no access to social security ments are particularly affected, and cialist era in an effort to encourage re- as the marriage did not officially take rarely adopted. Thousands of migrants gional specialisation and large-scale place.16 going to Russia, Kazakhstan and other production. The ‘one-company town’ (or countries, are leaving their children be- village) is an extreme case, where many Single older people hind without proper care. There are no former employees of now unprofitable Households with older people tend to be accurate statistics on the numbers of enterprises such as arms factories or col- less poor, as pension transfers help to children within this group but orpha- lective farms tend to be concentrated keep households above the poverty line. nages report many problems, due to lack (Box 9.3). Many local households often However, the decline in the real value of of resources. rely totally on employment in that one the pension often leaves those without sector or enterprise, leaving them vulner- wider family or household support with Wherearethechronically able to market fluctuations. This is partic- a significant reduction in their only ularly the case in countries where smaller source of income, leaving them at greater poor in transitional scale or diversified production was ab- risk of perpetual poverty. In Hungary, countries? sent, and where limited private sector among single elderly people, women are Markets in remote rural areas often employment opportunities emerged. 19 times more likely to be chronically poor than men. Box 9.1 Women ex-collective farm workers in Kyrgyzstan Children The disintegration or privatisation of the collective farms has resulted in the Despite large country differences, house- dismantling of entire community infrastructures. The collective farms system had holds with larger numbers of children provided comprehensive social assistance: in-kind food donations, education of tend to be more at risk of remaining in children, and distribution of benefits. Prior to the privatisation of collective farms, poverty. 18% of all Moldovans were people appeared to be able to weather the economic crisis better. poor in 1997, but 42% of those living in A 38-old Kyrgyz women from a rural area compared the past and present families with 3 or more children were systems: poor.17 In Hungary too, 21% of house- ‘Right now, we don’t know whom to turn to for help. Things were better when holds with three or more children are the collective farms were working. At least there was a director, and we had chronically poor. In Poland, households someone to complain to. Now it is as if the government didn’t exist. The only with children under 15 are more likely one you can rely on is yourself.’ to be chronically poor (27% for each The economic networks that managed transactions have been severed, and the additional child), and similarly, in Azer- workers themselves do not have the same bartering power. Unemployed rural baijan the ‘very poor’ were found in women seem least informed about the economic changes underway in the country 1997 to have twice the number of chil- and especially about the government services still available to them, including dren as non-poor households.18 employment services and unemployment benefits. In Kyrgyzstan, at any one time, Source: Dudwick et al. 2002:38 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P29E-01.3D.3DBlack Folio: plate 86 (86,1)

86 . Chapter 9

Notes 1. This section draws directly from an unpublished 6. Based on a poverty line of lowest-quintile 15. Triple burden – economic livelihood work, CPR background paper authored by Dirk consumption expenditures per equivalent adult. household care work, and community level Bezemer, ODI, 2003. 7. Okrasa 1999 work. 2. Abazov 1999 8. Braithwaite et al. 2001 16. IRIN News.org, 2003a. 3. UNDP 2002; World Bank 2003a 9. Commander et al. 1999 17. Banjeri 1999: xv 4. Falkingham 1998 10. Thanks to Rachel Marcus at SCF–UK for this 18. O’Keefe and Holson 1997: 5 5. This chapter recognises the extreme diversity comment. 19. UNICEF 1999: 48 within the region, but draws heavily from three 11. Okrasa 1999 20. Ringold 2002 countries, Hungary, Poland and Russia (where 12. Braithwaite et al. 1999 21. IRIN News.org, 2003b there is relatively strong data) to deduce 13. Bezemer and Davies 2003 22. Kuehnast and Dudwick 2002 implications for transitional countries as a 14. Revenga et al. 2002 23. IOM 2001; PMC 2000 whole. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P210E-01.3D.3DBlack Folio: plate 87 (87,1)

estimates are that between 40 and 65 million people live in persistent poverty, or between about one-fifth and one- quarter of the country’s absolute poor. Understanding Chronic poverty may be on the rise as the ‘new poor’ join the ranks of those chronic poverty left behind by growth. in China Available panel data covering five con- 10 secutive years between 1991 and 1995 in rural Sichuan,6 shows that only around 6% of households were consumption For the Chinese peasantry of the 19th and early 20th century, lifelong poor in all five years, compared to 44% poverty, hunger and vulnerability were the norm. In the late 20th century, who were poor in at least one year. But a unique combination of revolution, land reform and central planning, there is still a high degree of persistence, followed by the gradual opening up of the economy, has dramatically once a household has fallen below the reduced the proportion of people trapped in poverty. Indeed, much of the poverty line.7 achievement in global poverty reduction over the last 30 years is attribut- On the basis of panel data from four able to progress in China. southern provinces8, almost 60% of However, the benefits of recent rapid economic growth have been rural poverty in the three poorest provin- highly unequal, both between regions and across social groups. Growth, ces, but less than 20% in the better-off and its benefits, have been concentrated in urban and coastal areas province, is chronic.9 while neo-liberal reforms, including the introduction of user fees in health Other indicators suggest that chronic and education, the closure of state-owned enterprises, and the with- poverty remains a significant problem in drawal of the ‘iron rice bowl’ (guaranteed access to food)1 have reduced China. For example, on the ‘preventable the incomes of many poorer households while increasing their costs. The death’ criterion of chronic poverty, historical disadvantages that minority groups have faced continue and China has tens of millions of ‘missing may well have deepened in recent times. women’. Huge numbers of children – about one-quarter according to a 1993 Poverty trends in China daily expenditure down just below.3 survey – are malnourished.10 Since the beginning of the economic re- At the same time as per capita incomes forms in 1978, China has experienced have risen, income inequality has wors- rapid economic growth: official figures ened markedly,4 and this is increasingly Who are the chronically suggest 9% on average between 1978 and matched by inequalities in social indica- poor in China? 2000. This has been accompanied by a tors. The improvements in national The chronically poor in China tend to dramatic reduction in absolute rural pov- human development indicators over the have several characteristics that combine erty, at least until the mid- to late-1980s. past three decades have been impressive, to trap them in poverty. Some of these Chinese statistics, which use an official in- but not for everyone. The gaps in health can be seen as ‘historically’ predisposed come poverty line of about US$0.67/day, outcomes between regions and social to chronic poverty while others are the indicate a massive decline in the number groups are enormous; for example, while ‘new poor’, impoverished by recent pro- of rural Chinese in absolute poverty the national infant mortality rate has cesses, many of whom seem likely to from 250 million in 1978 to 34 million dropped from 48 per 1000 live births in form tomorrow’s chronically poor. in 1999. World Bank (2003b) estimates, 1975, to 32 in 2000, IMR among the US$1/day poverty line, report a slightly poorest quartile of the rural population Ethnic minorities: China’s non-Han more modest decline in poverty measured remains 3.5 times higher than that populations have faced discrimination based on household incomes from 260 to among city dwellers.5 While the majority for centuries, and despite the profound 97 million over the same period. of the country’s population may have ex- socio-economic changes in the country However, much depends on the meas- perienced improvements in their well- over the last 50 years, this discrimination ure that is used. According to the most being, a large minority are still waiting to appears to be a constant thread in pat- recent estimates, the number of rural see the benefits of growth ‘trickle down’. terns of impoverishment. There are more poor with daily expenditures below than 50 ethnic groups, comprising only US$1 in 1999 was 235 million, more How many people are 9% of the total population, but 40% of than twice that under the income poverty the absolute poor in the country.11 Their line.2 Indeed, using the expenditure crite- chronically poor in difficulties are exacerbated by the remote rion, poverty in China actually increased China? and rugged areas in which a large major- in the late 1990s, from 214 million In estimating numbers in chronic poverty ity of minority peoples live. (1996) to 235 million (1999). The differ- in China, we take the US$1/day poverty ence in these figures seems to be due to estimates based on household expendi- Older, sick, or disabled people, and the large number of households clustered tures, to be consistent with other coun- households with high dependency around the US$1/day poverty line: mini- tries. There are no national-level panel ratios: In rural areas, the poorest house- mal saving by those whose daily income data for China, and limited reliable data holds are those who support older, sick is just above the poverty line pushes their of any kind on urban poverty. Best and, increasingly, disabled family Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P210E-01.3D.3DBlack Folio: plate 88 (88,1)

88 . Chapter 10 members. Poverty is particularly persis- They have the ability and willingness to from the dynamic coastal region. Often tent in households where one or two eco- work, but economic growth is insuffi- such areas are remote from growth nomically active people have to support cient and there are no jobs available. centres, of low agro-ecological poten- both young children and ageing parents. The government introduced a ‘minimum tial,20 and have large ethnic minority In urban areas, chronic poverty has income guarantee’ (MIG) for urban poor populations. Any industry the area once been identified with the ‘three ‘‘Nos’’ . . . in 1996. The number of people under its had was largely state-owned and likely no ability to work, no savings and no coverage has increased rapidly.16 to have been closed during the economic friends or relatives to depend on’.12 People resettled for major infrastruc- restructuring. It is likely that chronic While the old age pension system limits tural projects often face similar problems poverty is greatest in the counties offi- the extent of chronic poverty in old age when their livelihood is lost and new op- cially classed as ‘poor’; the government and reduces the pressure on households portunities are not made available to has increased the number of these from with older members, the introduction of them.17 There is also emerging evidence 258 in 1986 to 592 in 2003. However, user fees for health care has increased that a proportion of rural to urban mi- evidence is emerging of the large num- the likelihood of households with older, grants are becoming part of this ‘new bers of people trapped in poverty in sick or disabled members becoming trap- poor’ group. These wai di ren (‘outsiders’) ‘non-poor’ counties. The growing eco- ped in poverty. ‘. . . are more likely to take up odd, infor- nomic and social inequality of rural mal jobs and become the bottom segment China means that substantial numbers of Orphaned and abandoned children, within the migrant population’.18 In chronically poor people live in areas that and street children: Official Chinese 1997 an official estimate for Beijing was have been prospering.21 estimates in 1999 suggested there were that 88% of the school age children of about 100,000 abandoned children, pri- wai di ren were not enrolled at school. Chronic poverty in urban areas: An marily girls, but also disabled boys.13 estimated 5% of the urban population There are also 150,000 street children, (or 14 million people) experience US$1/ most of whom are not included in the Wherearethechronically day income poverty22 – confirming the ‘orphan’ statistics. These figures are poor in China? broad consensus that poverty in China likely to grow as HIV/AIDS rates contin- Aggregate figures conceal the enormous remains mainly a rural problem. How- ue to rise. Chinese children living outside variations in poverty and well-being that ever, urban poverty is probably underes- families are often severely and variously now characterise China. The growing timated because 100 million of China’s deprived, with extremely limited access provincial inequalities in economic urban population are still recorded as to education and poor nutritional status. growth rates and per capita incomes are living in their home villages. Contracting social expenditures have matched by increasing differences in While poverty is lower in urban areas, begun to turn around, and are increasing social indicators. In terms of life expect- and much more likely to be transitory, both in absolute terms and relative to ancy and income, Shanghai is compara- there is evidence that the ‘old’ urban total government expenditure and to ble to Portugal, and Beijing to Costa long term poor are now accompanied GDP alongside transfer expenditures.14 Rica, while the poorest regions are at the and even outnumbered by significant level of Tajikistan, Vietnam and Bolivia, numbers of ‘new poor’23 – those who The ‘New Poor’. Over the 1990s, many and approaching the low income coun- have moved to cities but are unable to have become poor because of economic tries of South Asia.19 meet their minimum needs due to the restructuring. As in the transitional low paid casual work into which they economies, the collapse of state-owned Chronic poverty in rural areas: The are forced, and the very limited access to industries, particularly in one industry available evidence indicates that chronic state-provided services. In areas away towns and cities, has meant that millions poverty is highest in rural areas and that from the coastal zone, factory closures of people have lost their livelihoods and it is particularly concentrated in north- have created deep concentrations of per- their social security at the same time.15 west, west and south-west areas, away sistent poverty.

Notes 1. Cook 2003. 11. World Bank 2001. urban standards, but it is not clear whether 2. World Bank 2003. 12. Hussain 2003. their living standards improve or worsen by 3. Hussain 2003. 13. Lebrun 2003. moving into urban areas (Shaoguang Wang, 4. The national Gini coefﬁcient rose from 28.8 in 14. Shaoguang Wang, The Chinese University of The Chinese University of Hong Kong, Dept. of 1981, to 32.1 in 1990 and 41.6 in 1999. Hong Kong, Dept. of Government and Public Government and Public Administration, pers. 5. World Bank 2003b, in LeBrun 2003. Administration (pers. comm.). comm.). 6. China’s most populous province, with over 100 15. Despite China’s recent economic growth, 9 19. UNDP 2003. million people. million of the 26 million workers who lost jobs 20. McCulloch and Calandrino 2003 found that 7. McCulloch and Calandrino 2003: 624, 620. in state owned enterprises between 1998 and chronic poverty was signiﬁcantly more likely to 8. Yunnan, Guangxi, Guizhou, and the better-off 2002 had not found new employment (World occur in upland areas in Sichuan than lowland Guangdong. Bank 2003). areas. 9. Jalan and Ravallion 1998; they use a 16. Shaoguang Wang, The Chinese University of 21. For example, in Shanxi Province 3.8 million components approach, and a consumption- Hong Kong, Dept. of Government and Public poor people live in the 34 designated poor based poverty line for this estimate. Administration (pers. comm.). counties. Another 1.5 million are recorded in 10. Lebrun 2003. This is an extremely complicated 17. The most obvious example is the Three Gorges the province’s non-poor counties (Cook and issue. For a detailed discussion, see Yong Cai Project, which is disrupting the livelihoods of White 1997). See also Riskin 1994. and William Lavely, ‘China’s Missing Girls: more than a million people who have been 22. Using an expenditure poverty line of US$1/day Numerical Estimates and Effects on Population moved to upland areas that are agro- raises this to 12% or 37 million people. Growth,’ The China Review, Vol. 3, No. 2 ecologically fragile. Hussain 2003. 2003, pp. 13–30. [www.chineseupress.com]. 18. Wu 2001. Urban migrants may be poor by 23. Hussain 2003. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 89 Black plate (89,1)

PART C Measuring global trends on chronic poverty: statistical appendix Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 90 Black plate (90,1) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 91 Black plate (91,1)

Measuring chronic 11 poverty

How many chronically poor people are there in the world? There are a number of serious difficulties WDI 2003 suffer from severe implausi- on the incidence of poverty in other to face in estimating the likely global bility, and US$1/day estimates some- countries perceived to be similar). On numbers in chronic poverty. Current fig- times change sharply when they are this basis 1.18bn of the 4.97bn living in ures for global poverty, currently most revised, raising questions about reliabil- these 134 countries fall below the US$1/ developed for consumption poverty rela- ity. Weaknesses in US$1/day estimates day poverty line. tive to the US$1/day poverty line, are al- commonly reflect difficulties in working As noted above, much less information ready very approximate. And at the out PPP exchange rates, a problem likely is available on poverty dynamics. Panel same time there is limited knowledge to be most severe in transitional or re- data sets are available for eight of the about poverty dynamics within countries pressed economies. Such exchange rate ten countries with the highest numbers (movements into and out of poverty). problems are common to most global of absolutely poor people (representing And even though estimates of dynamics poverty estimates. 77% of poverty among these countries), are available for several of the countries US$1/day poverty figures have been and for several others (see Table 8 in the with the largest numbers in poverty, taken from World Development Indica- Statistical Appendix). Unfortunately, the there are important questions about their tors (WDI) 2003 in general. Four indi- panel data sets that are available for comparability and accuracy. Given cur- vidual cases, reported in Table 11.7 these countries are not always nationally rent data availability, a very approxi- below, though were considered suffi- representative (sometimes having been mate estimate, with little geographic ciently implausible not to use. The table conducted in one region, or only in rural disaggregation, is the most that can be explains the basis for this judgement and or urban areas for instance). In addition expected at this stage. the action taken in these cases. In a few the time between different waves of The US$1/day (or absolute) poverty es- cases WDI 2002 figures were used when these panel data sets differs between timates, pioneered by the World Bank, a figure was not available in 2003. These countries (from one year in Indonesia to seem to provide the best basis for such an are the same figures as reported in Table ten in India). The time spanned by a estimate of global chronic poverty. Com- 11.7. US$1/day poverty estimates are panel is important for comparability parability across countries is very impor- available for 80 or the 134 countries in across countries because it might be con- tant for this exercise, and the US$1/day this table, with most of those for which sidered that the probability that a cur- estimates offer greater comparability figures are unavailable being small in rently poor individual will still be poor than poverty estimates based on national population terms. These estimates are one year later is higher than the proba- poverty lines, at least when the latter available for 18 of the 20 most populous bility that the same individual will still cover countries at significantly different countries and 26 of the 30 most popu- be poor ten years later. There is no scien- stages of development (Ravallion, Datt lous. Available poverty figures were used tific process though of making an adjust- and Van de Walle, 1991). The US$1/day to estimate total numbers in US$1/day ment for this difference. figures though do suffer from a number poverty in each of the regions identified The available panel data were used to of problems. They have been subject to a in Table 1, generally assuming that the compute the proportion of the poor in number of criticisms (Deaton, 2001; missing countries collectively had the these data sets that are chronically poor, Reddy and Pogge, 2003), although at same proportions in poverty as the re- that is poor at two points in time as present alternatives that have been dis- gional average. However, in four large close to five years apart as possible. Ad- cussed or developed, such as by UN- countries (Democratic Republic of justments then need to be made to these CTAD, in the Least Developed Congo, Myanmar, North Korea and proportions where the time interval dif- Countries Report 2002, seem to suffer Sudan) there was a strong prima facie fers from five years and /or when the sur- from much more serious methodological case for higher levels of poverty than the veys are not nationally representative. problems (Ravallion, 2003). In addition corresponding regional averages, so Unfortunately this involves various sub- a few individual US$1/day estimates in higher numbers were substituted (based jective judgements, drawing on Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 92 Black plate (92,1)

92 . Chapter 11 information such as the differences be- in the range of between 300m and lower numbers for South Asia and, to a tween poverty persistence in urban and 420m. In other words, around one quar- lesser extent, Africa. This different geo- rural areas in other countries, or infor- ter to one third of the total number of graphic distribution applies both to over- mation from instances of multi-year pan- US$1/day poor, are chronically poor. all poverty and specifically to chronic els where it is possible to compare For the reasons mentioned above, we be- poverty. This is likely to reflect the fact poverty persistence over short durations lieve that the global extent of chronically that the national poverty lines are likely with longer durations. poverty is probably nearer the upper end to be higher in the former groups of To reflect the extremely imprecise na- of this range. Unfortunately, the limited countries and lower in the latter group. ture of these estimates, a range was esti- availability of panel data means that it is Typically estimates of non-food needs in- mated for the proportion likely to be not possible to draw conclusions about cluded in national poverty lines tend to chronically poor. The lower end of the the geographic pattern of chronic pov- be somewhat higher in less poor coun- range corresponded to what seemed to erty at anything other than a highly ag- tries, this being a key element of the be the lowest plausible proportion that gregated level. Our estimates suggest comparability problem discussed in would be persistently poor over a five that 29% of the world’s chronic poor Chapter One (Box 1.3). Another point of year period. However, the upper limit are in Africa (compared to around 24% difference though that it is not necessary should not necessarily be considered as a of the global US$1/day poor), with the to use PPP exchange rates for these na- maximum because the effects of meas- corresponding figures for South Asia tional poverty line computations. urement error in panel data often make being 48% and 45%. Another 20% of On the basis of these, admittedly very it appear that there is more volatility in the world’s chronic poor are found in imprecise estimates, it can concluded consumption levels than is actually the East Asia (including China), although that there are at least 270 million case. Thus our estimates of chronic pov- this region accounts for nearly one quar- chronically poor people in the world. erty (as for instance in Table 11.1 in the ter of global poverty. The remaining The actual numbers of chronically poor Statistical Appendix) are liable to be regions account for only a small propor- people are likely to be closer to our underestimated. Unfortunately, there is tion of both absolute poverty and upper estimate around 420 million. insufficient information to be able to chronic poverty. judge the extent of such measurement Finally, an analogous calculation using References error and so to make a correction for national poverty estimates (most of Deaton, A., 2001, ‘Counting the world’s poor: problems and possible solutions’, World Bank this. It should also be noted that many which are based on nutritionally based Research Observer, 16(2): 125–147 countries in conflict or recovering from poverty lines) produced a very similar Ravallion, M., 2003, ‘Pick your number in the great globalisation debate’, Mimeo, Development conflict do not have poverty information range of estimates of the extent of Economics Research Group, World Bank, available, although they are likely to chronic poverty (270 million to 410 mil- Washington DC have high levels of poverty and chronic lion) but with a somewhat different geo- Ravallion, M., Datt, G. and Van de Walle, D., 1991, ‘Quantifying absolute poverty in the developing poverty than on average. As noted graphic distribution. Using national world’, Review of Income and Wealth, 37(4): above, we have only been able to allow poverty lines rather than the US$1/day 345–262 Reddy, S. and Pogge, T, 2003, ‘How not to count for this very approximately here for a poverty line produces somewhat higher the poor’, Mimeo, Columbia University, New few large countries. estimates of poverty in Latin America York UNCTAD (2002 The Least Developed Countries As discussed in Chapter One, the glob- and the Caribbean; Europe and Central Report 2002: Escaping the Poverty Trap, al number of chronic poor is likely to be Asia; and North Africa, but somewhat Geneva Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 93 Black plate (93,1)

Measuring chronic poverty . 93

Table 11.1 Panel data on chronic and transitory poverty: selected countries Panel data on chronic and transitory poverty These figures report on chronic and transitory poverty, based on non-poor in both periods and the percentage of people who panel data where the same households are surveyed over two were poor in both periods – people in chronic poverty. time periods. This reveals the extent to which households remain trapped in chronic poverty while other households move Key to tables: into or out of poverty. (The sample ﬁgures are taken from 11.1a – Rural chronic Panel data sets are relatively rare but this table compiles poverty in Nicaragua) information from a range of available panel data sets to draw out patterns of chronic poverty. The levels of poverty are not People who moved out of poverty People in necessarily comparable between countries, first because they (poor in 1998, chronic poverty are based on national poverty lines and second, because non-poor in 2001) methodology and time periods are different. However the shares of chronic and transitory poverty can be broadly compared across countries. 17.0% 42.0% All are based on monetary measures of poverty (income or consumption) and all relate to two wave panels. For panel data sets comprising three or more waves figures are still reported in relation to two of these waves only. In each case households 30.0% 11.0% are classified according to their poverty status (poor or non- poor) in the first and second waves of the panel. These figures show what happened to people over two time People who were People who moved into periods: the percentage of people who moved out of poverty, not poor in either poverty (non-poor in period 1998, poor in 2001) the percentage that became poor, the percentage remaining

Table 11.1a Chronic Poverty in Nicaragua, Table 11.1b Chronic Poverty in Kwa-Zulu Natal, 1998–2001 South Africa 1993–1998

RURAL URBAN RURAL URBAN

17.0% 42.0% 10.0% 14.0% 8.4% 24.3% 5.8% 8.3%

30.0% 11.0% 69.0% 7.0% 32.2% 35.2% 75.2% 10.7%

Table 11.1c Chronic Poverty in Uganda, 1992–1999 Table 11.1d Chronic Poverty in Vietnam, 1993–1998

RURAL URBAN RURAL URBAN

30.7% 20.5% 24.1% 10.2% 29.7% 33.9% 17.3% 6.5%

37.7% 11.1% 59.6% 6.0% 31.1% 5.4% 74.1% 2.1% Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 94 Black plate (94,1)

94 . Chapter 11

Table 11.1e Chronic Poverty in Egypt, 1997–1999 Table 11.1f Chronic Poverty in Rural Bangladesh, 1998–2000

6.3% 19.0% 25.8% 31.4%

60.6% 14.1% 25.1% 17.7%

Table 11.1g Chronic Poverty in Rural Chile, 1968– Table 11.1h Chronic Poverty in Rural China 1986 (Sichuan), 1991–1995

23.3% 54.1% 15.2% 9.6%

14.4% 8.2% 67.9% 7.3%

Table 11.1i Chronic Poverty in Rural India 1970/71 Table 11.1j Chronic Poverty in Urban Ethiopia, to 1981/82 1994–1997

22.8% 25.3% 9.2% 25.0%

38.5% 13.3% 47.8% 17.9% Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 95 Black plate (95,1)

Measuring chronic poverty . 95

Table 11.2 Characteristics of Chronic Poverty in Rural Bangladesh These tables are based on panel data sets available to CPRC members or, in the India case, on published data. In each case they reveal key characteristics of households who are always poor. While there has been an attempt to collect similar information for all countries, in practice the information, as well as precise definitions, vary from case to case reflecting the precise surveys carried out in different countries. The tables report average characteristics for households classified as chronically poor (poor in both periods in a two wave panel) by comparison with the average for the entire sample. This allows us to see to what extent the characteristics of the chronic poor differ from those of the average household. The tables have been highlighted to show characteristics where the difference between people who are chronically poor and the overall population is greatest. Rural Always Poor Overall People (million) 29.6 94.3 Average household size 5.46 5.19 Percentage of children under 5 who are wasted na 11.9% Percentage of children under 5 who are underweight na 52.8% Percentage of children under 5 who are stunted na 50.7% Average number of children aged 0–14 in h’hold 4.24 3.45 Average number of people aged 15–59 years in h’hold 5.06 5.90 Average number of people aged 60+ years in h’hold 0.71 0.65 % of h’holds with no members aged between 15 and 59 years 1.9% 0.9% % of children who are engaged in Child Labour 15.8% 11.8% % of households headed by women 14.5% 8.66% % of households headed by widows na na % of households with children under 16 who have been orphaned na na % of households with any member disabled na na % of h’holds with at least one member who is long term ill (15 days or more out of every 30) 24.8% 17.8%

Patterns of Expenditure and Income % of expenditure spent on food na 54.0% % of expenditure spent on housing na 5.2% % of expenditure spent on medical care na 2.6%

% of income from agricultural subsistence activities 44.0% 24.3% % of income from agricultural wage labour 15.4% 10.3% % of income from non-agricultural non wage 18.2% 33.6% % of income from non-agricultural wage labour 18.2% 20.9% % of income from remittances 4.3% 10.8% All sources 100.0% 100.0%

Occupation of the household head % Agricultural Subsistence 46.7% 44.9% % Agricultural wage labour 19.0% 8.9% % Non Agricultural wage labour 12.4% 13.5% % Non-agricultural self-employed 10.5% 22.5% % Unemployed/Not working/Retired/Disabled/Other 11.4% 10.2% All 100.0% 100.0%

Use of Public Services % of those ill or injured not seeking health care na 22.7% % of primary school aged children not attending primary school 28.0% 25.0% % of secondary school aged children not attending secondary school 55.0% 40.9%

Household Public Goods % of households without clean water na 3.8% % of households without access to toilet na 79.4% % of households with no electricity na 81.3%

Physical Assets % of households not owning dwelling 7.6% 4.9% % of households not owning radio or tv na na % of households not owning bicycle na na % of landless households 39.0% 28.6% % of households ‘near’ landless 30.4% 19.1% % of households with no livestock na na

Human Capital % of adults illiterate 52.1% 34.9% % of adults who have not completed primary school 69.9% 48.2% % of adults who have not completed secondary school 98.4% 90.6% Average number of years schooling for individuals aged 15+ yrs 5.9 10.4

Note: (1) Average household size and % of households female-headed have been estimated by using the sample ratios of ‘always poor’ to ‘overall’ (based on panel data) and applied to national averages (based on HIES). ‘Agricultural subsistence activities’ include rice, non-rice crop and non-crop agriculture. Non-agricultural non-wage income includes ‘other income’ such as informal and formal transfers and rental income from housing. Main occupation is given by household head and is estimated from panel data for rural areas, while the matched urban data are from HIES. % child labour represents proportion of earners who are children. Average number of years of schooling for individuals (15+ yrs) is given for earners only. Landless is deﬁned as having no agricultural land other than homestead; near landless is deﬁned as having agricultural land up to 0.49 acre. (2) Rural data for ‘always poor’ and ‘overall’ are estimated from primary panel survey data except for ‘Expenditure’, ‘Use of public services’, ‘Household public goods’, ‘Child anthropometry’ for which HIES and CNS data of BBS have been used. Source: Population Census 2001, Household Income and Expenditure Survey 2000, Child Nutrition Survey 2000 of BBS; IRRI-IFPRI 21 Village panel data for 1987 & 2000. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 96 Black plate (96,1)

96 . Chapter 11

Table 11.3 Characteristics of Chronic Poverty in rural India Rural India 1968–1970 panel Rural Always Poor Overall People % of people living in households with six or more persons 55.0% 49.3%

Occupation of the household head % Cultivators 26.9% 44.8% % Casual agricultural labourers 56.3% 32.3% % Casual non-agricultural labourers 4.4% 5.3% % Permanent wage earners 2.3% 6.2% % Artisans 8.5% 8.6% % Dependent on transfer income 1.7% 2.9% All 100.0% 100.0%

Physical Assets % of households cropping less than 1 hectare 71.5% 57.2%

Human Capital % of adults illiterate 56.2% 47.8% % of adults with primary school education or below 22.8% 21.5%

Note: characteristics refer to initial tear values (1968). Source: NCAER Panel 1968–70, as reported by Gaiha (1989)

Rural India 1970/71 – 1981/82 panel Rural Always Poor Overall People Average household size 6.7 6.7 Average number of children 4.0 3.8 % that are scheduled caste or scheduled tribes 32.7% 18.4%

Sources of income % of income from agricultural subsistence activities 45.5% 63.4% % of income from agricultural wage labour 36.1% 18.4% % of income from non-agricultural non wage 5.8% 3.3% % of income from non-agricultural wage labour 12.6% 15.0% All 100.0% 100.0%

Physical Assets Average land size (hectares) 1.7 3.7 % of landless households 44.0% 32.7% % of ‘near’ landless 19.6% 12.6%

Note: characteristics refer to initial year values (1970/71). Source: NCAER Panel 1970/71–81/82, as computed by Bhide and Mehta (2003) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 97 Black plate (97,1)

Measuring chronic poverty . 97

Table 11.4 Characteristics of Chronic Poverty in KwaZulu-Natal South Africa (KwaZulu-Natal) Rural Urban Total Always Overall Always Overall Always Overall Poor Poor Poor People

Average household size 7.8 6.8 8.6 5.6 8.0 6.5

Average number of children aged 0–14 in h’hold 3.9 3.3 3.7 2.0 3.8 2.8 Average number of people aged 15–59 years in h’hold 3.7 3.2 4.5 3.3 3.8 3.2 Average number of people aged 60+ years in h’hold 0.4 0.5 0.4 0.3 0.4 0.4 % of h’holds with no members aged between 15 and 59 years 2.4% 2.4% 0.0% 1.4% 2.1% 2.0% % Child labour na na na na na na % of households headed by women (de facto) 16.3% 23.3% 0.0% 2.8% 13.7% 16.0% % of households headed by women (de jure) 36.3% 32.6% 63.3% 29.7% 40.5% 31.6% % of households headed by widows 25.3% 26.2% 46.7% 20.9% 28.6% 24.3% % of households with children under 16 who have been orphaned 18.50% 14.20% 23.30% 10.50% 19.30% 12.90% % of households with any member disabled na na na na na na % of h’holds with at least one member who is long term ill (15 days or more out of every 30) na na na na na na

Patterns of Expenditure and income % of expenditure spent on food 59.7% 58.4% 60.0% 41.0% 59.7% 52.3% % of expenditure spent on housing 7.3% 8.5% 9.9% 16.6% 7.7% 11.4% % of expenditure spend on medical care 1.2% 1.3% 0.7% 1.1% 1.1% 1.2%

% of income from wages 31.7% 29.8% 45.5% 56.8% 33.9% 39.2% % of income from agriculture 6.8% 7.8% 0.7% 0.2% 5.9% 5.2% % of income from agricultural wages 13.6% 5.4% 0.0% 0.0% 11.4% 3.5% % of income from non-farm self employment 7.2% 6.0% 4.0% 5.9% 6.7% 6.0% % of income from remittances 19.4% 22.5% 8.7% 5.5% 17.8% 16.6% % of income from other sources OAP 19.7% 20.1% 24.3% 15.0% 20.4% 18.3% % of income from casual work 7.8% 4.4% 6.8% 1.9% 7.7% 3.5% % income from remaining sources 7.2% 9.2% 10.1% 13.0% 7.7% 10.5%

Occupation of the household head % Not economically active 23.7% 28.8% 9.6% 9.6% 21.2% 22.9% % Regular employment 13.7% 14.0% 21.3% 33.9% 15.0% 20.1% % Casual employment 3.8% 3.2% 2.9% 2.1% 3.7% 2.9% % Self-employed 2.7% 3.8% 2.2% 5.2% 2.7% 4.2% % Housewife/husband 13.4% 13.3% 5.9% 13.5% 12.1% 13.4% % Unemployed 30.5% 23.5% 47.1% 25.0% 33.4% 24.0% % Retired 7.6% 10.4% 8.1% 7.7% 7.7% 9.6% % Other 2.4% 1.4% 1.5% 1.1% 2.3% 1.3% % Disabled 2.0% 1.5% 1.5% 1.9% 1.9% 1.6% All 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Use of Public Services % of those ill or injured not seeking health care % of primary school aged children not attending primary school 11.9% 10.1% 14.3% 3.6% 12.3% 8.2% % of secondary school aged children not attending secondary school 22.6% 21.4% 40.0% 17.0% 25.0% 20.0%

Household Public Goods % of households without clean water 59.3% 50.5% 0.0% 0.0% 50.0% 28.9% % of households without access to toilet 33.3% 22.2% 0.0% 0.6% 28.1% 14.5% % of households with no electricity 79.6 75.4% 60.0% 17.4% 76.6% 55.0%

Physical Assets % of households not owning dwelling 17.9% 7.9% 36.7% 28.7% 20.8% 15.2% % of households not owning radio or tv 30.9% 19.0% 20.0% 5.5% 29.2% 14.3% % of households not owning bicycle 90.7% 87.6 93.3% 85.1% 91.1% 86.7% % households not owning gas or electric stove 90.7% 78.3 66.7% 21.5% 87.0% 58.3% % of landless households 54.9% 42.4% 96.7% 98.9% 61.5% 62.3% % of households with no livestock 66.0% 65.6% 96.7% 99.4% 70.8% 77.5%

Human Capital % of adults illiterate 44.5% 35.9% 25.5% 13.1% 41.2% 28.9% % of adults who have not completed secondary school 96.4% 89.9% 93.0% 72.1% 95.8% 84.4% Average number of years schooling for individuals aged 16+ yrs 4.0 5.0 5.0 7.0 4.0 5.3

Note: The ﬁgures in this table refer to the ﬁrst year of the panel, 1993 Source: Calculations based on KwaZulu Natal Income Dynamics Survey 1993/98 Panel Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 98 Black plate (98,1)

98 . Chapter 11

Table 11.5 Characteristics of Chronic Poverty in Uganda

Rural Urban National Always Overall Always Overall Always Overall Poor Poor Poor People Average household size 6.4 5.7 6.6 5.9 6.5 5.8 Percentage of children under 5 who are wasted 4.4% 3.9% 4.7% 2.2% 4.5% 3.7% Percentage of children under 5 who are underweight 26.6% 21.8% 3.3% 20.5% 27.2% 21.6% Percentage of children under 5 who are stunted 45.3% 39.8% 57.1% 34.6% 46.4% 39.1% Average number of children aged 0-14 years in h’hold 3.42 2.92 3.82 3.06 3.45 2.94 Average number of people aged 15-59 years in h’hold 2.81 2.56 2.71 2.73 2.80 2.59 Average number of people aged 60+ years in h’hold 0.21 0.26 0.12 0.15 0.21 0.24 % of h’holds with no members aged between 15 and 59 years 1.6% 2.3% 0.0% 0.6% 1.4% 2.1% % of households headed by women 17.2% 18.2% 41.2% 34.3% 19.1% 20.7% % of households headed by widows 6.2% 7.8% 11.8% 12.7% 6.7% 8.5% % of households with children under 16 who have been orphaned 4.1% 5.3% 11.8% 7.2% 4.8% 5.6% % of h’holds with any member unable to work due to disability 5.7% 3.3% 0.0% 4.2% 5.3% 3.4% % of h’holds with at least one member who is long term ill (15 days or more out of every 30) 10.4% 13.2% 11.8% 7.2% 10.5% 12.3%

Patterns of Expenditure and income % of expenditure spent on food 73.5% 74.8% 72.9% 71.3% 73.4% 72.4% % of expenditure spent on housing 3.2% 2.9% 3.7% 5.3% 3.2% 3.2% % of expenditure spend on medical care 0.3% 0.6% 0.1% 0.6% 0.3% 0.6%

% of income from agricultural subsistence activities 52.1% 51.1% 32.0% 15.0% 50.5% 45.7% % of income from agricultural wage labour 1.0% 1.3% 3.9% 1.8% 1.2% 1.4% % of income from non-agricultural non wage 2.6% 3.8% 8.4% 23.9% 3.0% 6.8% % of income from non-agricultural wage labour 5.7% 8.8% 21.4% 25.9% 7.0% 11.3% % of income from remittances 12.3% 11.6% 4.6% 11.2% 11.7% 11.5% % other income 26.3% 23.5% 29.7% 22.1% 26.6% 23.3% All sources 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Occupation of the household head % Agricultural Subsistence 79.2% 73.1% 47.1% 22.9% 76.6% 65.5% % Agricultural wage labour 2.1% 2.1% 0.0% 2.4% 1.9% 2.2% % Non Agricultural wage labour 9.4% 13.8% 29.4% 38.6% 11.0% 17.5% % Non-agricultural self-employed 4.2% 6.2% 11.8% 28.3% 4.8% 9.5% % Unemployed/Not working/Retired/Disabled/Other 5.2% 4.8% 11.8% 7.8% 5.7% 5.3% All 100.0% 100.0% 100.0% 100.0% 100.0% 100.0%

Use of Public Services % of those ill or injured not seeking health care 55.0% 33.9% 50.0% 8.3% 54.5% 31.6% % of primary school aged children not attending primary school 51.0% 38.9% 41.7% 23.6% 50.1% 36.3% % of secondary school aged children not attending secondary school 96.2% 88.4% 81.9% 73.1% 95.4% 85.8%

Household Public Goods % of households without clean water 40.1% 39.0% 23.5% 24.7% 38.8% 36.8% % of households without access to toilet 22.9% 15.5% 11.8% 9.0% 22.0% 14.5% % of households with no electricity 99.5% 99.3% 100.0% 73.5% 99.5% 95.4%

Physical Assets % of households not owning dwelling 2.6% 4.2% 5.9% 38.0% 2.9% 9.2% % of households not owning radio or tv 67.7% 54.3% 52.9% 28.3% 66.5% 50.4% % of households not owning bicycle 54.7% 49.6% 58.8% 63.3% 55.0% 51.7% % of landless households 15.6% 13.6% 29.4% 44.6% 16.7% 18.2% % of households ‘near’ landless 17.7% 16.2% 29.4% 45.2% 18.7% 20.6% % of households with no livestock 24.0% 28.4% 29.4% 52.4% 24.4% 32.0%

Human Capital % of adults illiterate 50.4% 40.1% 42.0% 21.0% 49.7% 37.2% % of adults who have not completed primary school 79.8% 75.2% 69.3% 47.0% 79.0% 71.0% % of adults who have not completed secondary school 97.1% 96.0% 91.0% 81.7% 96.6% 93.8% Average number of years schooling for individuals aged 16+ yrs 3.2 3.8 3.7 5.9 3.3 4.1

Note: Figures in this table are based on data from the ﬁrst wave of the panel (1992 ). Source: Based on IHS/UNHS 1992/99 panel data set Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 99 Black plate (99,1)

Measuring chronic poverty . 99

Table 11.6 Characteristics of Chronic Poverty in Vietnam Vietnam, 1993–98 Rural Urban National Always Overall Always Overall Always Overall Poor Poor Poor People Average household size 5.4 5.0 6.1 5.0 5.4 5.0 % of households with wasted children 2.4% 1.7% 3.8% 1.6% 2.4% 1.7% % of households with stunted children 33.4% 22.1% 40.0% 10.5% 33.7% 19.9% % of households with malnourished children 25.3% 17.0% 25.0% 8.7% 25.2% 15.4% Average number of children aged 0-14 in h’hold 2.47 1.99 2.69 1.49 2.48 1.90 Average number of people aged 15–59 years in h’hold 2.58 2.67 3.04 3.09 2.60 2.75 Average number of people aged 60+ years in h’hold 0.39 0.42 0.44 0.48 0.39 0.44 % of h’holds with no members aged between 15 and 59 years 3.1% 3.9% 0.0% 2.8% 2.9% 3.7% % Child labour 16.3% 15.5% 28.6% 24.7% 16.9% 16.9% % of households headed by women 18.5% 22.0% 32.7% 43.2% 19.1% 26.0% % of households headed by widows 8.2% 10.7% 2.5% 10.0% 8.1% 10.6% % of households with ethnic minority head 29.5% 14.9% 28.6% 24.7% 16.0% 16.9% % of households with children under 16 who have been orphaned 21.2% 21.6% 7.5% 15.0% 20.7% 20.8% % of households with any member disabled 1.4% 1.8% 0.0% 2.9% 1.4% 1.9% % of h’holds with at least one member who is long term ill (15 days or more out of last 30) 15.7% 18.2% 20.0% 18.0% 15.9% 18.2%

Patterns of Expenditure and income % of expenditure spent on food 70.5% 64.1% 62.4% 47.8% 70.2% 61.1% % of expenditure spent on housing 3.4% 3.7% 8.3% 8.6% 3.6% 4.2% % of expenditure spent on medical care 7.1% 8.1% 4.8% 6.8% 7.1% 8.0%

% of income from agricultural subsistence activities 71.1% 53.3% 14.6% 26.0% 69.2% 50.3% % of income from wage labour 17.0% 20.3% 43.1% 18.8% 17.9% 20.1% % of income from non-agricultural enterprises 2.1% 12.5% 22.6% 33.1% 2.7% 14.8% % of income from gifts & remittances 4.5% 8.0% 5.2% 8.1% 4.5% 8.0% % other income 5.3% 5.8% 14.4% 13.9% 5.6% 6.7%

Occupation of the household head % Agricultural Subsistence 77.3% 71.4% 30.0% 22.8% 75.7% 66.0% % Agricultural wage labour 2.8% 4.4% 2.5% 2.5% 2.8% 4.2% % Non Agricultural wage labour 7.6% 9.4% 40.0% 37.2% 8.7% 12.5% % Non-agricultural self-employed 10.1% 11.7% 27.5% 41.3% 10.7% 15.0% % Unemployed/Not working/Retired/Disabled/Other 0.6% 0.6% 1.9% 1.6% 0.6% 0.8%

Use of Public Services % of those ill or injured not seeking health care 56.6% 58.2% 69.8% 61.0% 57.1% 57.3% % of primary school aged children not attending primary school 16.4% 9.1% 11.4% 3.0% 16.2% 8.2% % of secondary school aged children not attending secondary school 56.7% 47.3% 61.1% 47.3% 56.9% 47.3%

Household Public Goods % of households without clean water 31.8% 36.1% 10.0% 10.2% 31.1% 33.2% % of households without access to toilet 48.5% 52.5% 20.0% 21.1% 47.6% 49.0% % of households with no electricity 63.0% 61.1% 12.5% 9.8% 61.3% 55.4%

Physical Assets % of households not owning dwelling 3.0% 3.2% 15.0% 17.5% 3.4% 4.8% % of households not owning radio or tv 88.0% 84.6% 55.0% 54.5% 86.9% 81.3% % of households not owning bicycle 37.6% 38.7% 17.5% 17.3% 36.7% 36.3% % of landless households 5.7% 7.8% n/a % of households ‘near’ landless 24.7% 28.1% n/a % of households with no livestock 2.5% 2.2% n/a

Human Capital % of adults illiterate 15.4% 16.1% 3.8% 6.0% 15.0% 14.9% % of adults who have not completed primary school 21.4% 23.9% 11.5% 10.4% 21.1% 22.3% % of adults who have not completed secondary school 49.8% 52.0% 34.4% 27.3% 49.2% 49.2% Average number of years schooling for individuals aged 16+ yrs 5.1 5.9 5.1 8.1 5.1 6.3

Note: There are only 40 always poor households in urban areas Note: Figures in this table A27are based on data from the ﬁrst panel wave (VLSS, 1992–93) Source: Bob Baulch based on 1992–93 to 1997–98 VLSS panel with 4302 households Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 100 Black plate (100,1)

100 . Chapter 11

Table 11.7 Global Indicators of Chronic Poverty Tables 11.7, 11.8 and 11.9: Global Indicators These tables report indicators that are available for the vast majority of countries from international sources. At this level direct estimates of chronic poverty are not available; the data reported are of series potentially linked to chronic poverty and which are available on a comparable basis across all countries.

Table 11.7 Global Indicators of Chronic Poverty Chronic poverty estimates are not available for many countries, but the incidence of absolute poverty (US$1/day) and measures of the average depth of poverty (an indicator of the extent to which many of the poor lie a long way below the poverty line) may give some indication of likely patterns of chronic poverty. The distinguishing characteristic of chronic poverty is persistence. Child stunting (height for age more than 2 standard deviations below the reference level for that age) is generally taken as an indicator of long term or persistent malnutrition. Illiteracy also represents persistent deprivation. Life expectancy and infant and child mortality are expected to be strong correlates of chronic poverty. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 101 Black plate (101,1)

Measuring chronic poverty . 101 prices) 2000 (1995 capita US$ Real GDP per 2000 rate for men, Adult illiteracy rate for women, 2000 Adult illiteracy Life birth, 2000 expectancy at refer stunting data Year to which Proportion of who are stunted, most recent year children under ﬁve 2000 live births), rate (per 1,000 Infant mortality Under-ﬁve births), 2001 mortality rate (per 1,000 live refer poverty data Year to which number of recent year) poverty (The fall below the Average depth of percentage points by which the poor poverty line) (most – – 260 172 – 45.3 – – 506 Percentage of people living on (most recent year) less than US$1/day te d’Ivoire 12.3 19.5 1995 175 102 21.9 c 1998–99 47.8 61.4 45.5 743 ˆ Sub-Saharan Africa Countries East and Southern Africa Angola BotswanaBurundiComorosCongo, Dem. Rep.EritreaEthiopiaKenyaLesothoMadagascarMalawiMauritiusMozambique 23.5 –Namibia 58.4 bRwanda –SomaliaSouth AfricaSudan – 31.3 32.8SwazilandTanzania 49.1 23.0 43.1 bUganda 42.6 –ZambiaZimbabwe 37.9 41.7 – –West 1993 Africa Benin 25.6 34.9Burkina – Faso 37.3 35.7 26.1 1998Cameroon 47.1 11.5Cape – Verde b 110Central African 31.7 35.5 Rep.Chad – 1995 –Congo, – 19.9 Rep. 190 40.1Co 1999 1997 21.6 1993 36.7 15.7 63.7 74 36.0 205 b 1997–98 1996 172 – 114 61.2 79 136 122 66.6 132 1983–85 1993 24.1 – 33.4 – – 183 128 111 23.1 – 1993 – 197 117 51.3 26.7 183 56.8 61 – 86 77 92 41.7 67 – 117 1993 57.2 45.2 73 19 35.3 126 71 51.2 2000 – 1990–91 100 1998 42.3 – 48.6 37.2 2000 44.0 56 165 1994 225 49.0 38.4 1993 – 123 17 35.9 55 40.1 1995 1996 149 202 – c 107 42.7 2000 c 40.6 124 2000 104 133 197 28.4 2000 2000 1996 180 20.2 2000 51.4 73 155 1995 25.4 9.6 101 1997 112 43.9 66 59.6 c 59.8 2000 81 105 52.7 43.8 50.7 23.3 158 45.7 115 25.5 49.8 40.0 1992 52.0 26.5 38 95 39.3 69.1 59.0 1994 – 43.8 1995 51.3 40.2 108 40.3 3951 24.0 38.3 36.8 6.4 98 200 26.9 53.5 38.9 55.5 44.7 1999 1999 71.3 c 52.8 30 34.6 141 52.1 71.4 1999 36.8 39.8 1999 26.4 11.1 81 27.5 25.5 118 32.7 18.8 1998–99 1995 – 25.0 51.1 39.9 116 – 15.4 2000 16.2 18.7 436 c 42.9 26.3 246 328 1998 41.3 551 169 46.6 155 18.8 17.2 191 44.0 33.5 28.3 44.4 14.0 1996 c 44.4 12.2 242 – 15.3 56.0 50.0 28.5 1994 2408 85.9 43.2 1998–99 16.1 3985 4429 21.4 53.8 65.1 2000 53.7 7.2 30.5 14.8 – 69.4 66.1 51.3 190 22.5 19.2 76.4 40.3 621 45.7 30.5 392 17.6 34.3 252 25.6 – 348 1476 339 47.9 319 66.0 675 15.5 12.5 414 48.4 1519 841 218 Equatorial GuineaGabonGambia, TheGhanaGuineaGuinea-BissauLiberiaMaliMauritania –NigerNigeriaSao TomeSenegal 59.3Sierra Leone –Togo – 44.8 – – – 48.6 28.6 72.8 – 38.6 61.4 70.2 – – 57.0 26.3 – 1998 31.8 – 51.4 1999 – 55.2 49.7 126 153 69.3 – 26.6 1995 100 1994 1995 103 92 1997 90 – 211 1989 183 1995 169 231 58 265 235 183 18.7 60 – 132 316 120 138 112 142 25.9 159 74 157 110 180 – 28.0 2000 44.0 80 26.1 141 39.8 1998 58 45.5 – 46.3 33.9 c 19.0 2000 1996 51.0 80 1999 56.8 26.0 70.6 2000 1999 44.8 2000 51.5 25.6 2000 47.5 37.1 21.7 52.7 56.0 45.2 51.7 c 1996 76.7 51.4 38.9 69.9 53.4 7.5 19.7 – – – 370 91.6 1998 44.3 45.6 – 65.6 49.2 1599 – 413 72.4 – 62.3 76.2 27.6 51.8 – 210 496 51.1 – 52.7 – 203 254 29.9 603 57.5 4378 288 609 – 147 27.6 – 341 327 Table 11.7 Global Indicators of Chronic Poverty Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 102 Black plate (102,1)

102 . Chapter 11 prices) 2000 (1995 capita US$ Real GDP per 2000 rate for men, Adult illiteracy rate for women, 2000 Adult illiteracy Life birth, 2000 expectancy at refer stunting data Year to which Proportion of who are stunted, most recent year children under five Infant births), 2000 mortality rate (per 1,000 live Under-five births), 2001 mortality rate (per 1,000 live refer poverty data Year to which number of recent year) poverty (The fall below the Average depth of percentage points by which the poor poverty line) (most – – 138 95 46.0 2000 56.3 42.9 20.2 297 Percentage of people living on (most recent year) less than US$1/day Asia Countries East Asia and Pacific Cambodia ChinaFijiIndonesiaKiribatiKorea, Dem. Rep.Lao PDRMalaysiaMarshall IslandsMicronesia, Fed. Sts.Mongolia 18.8MyanmarPapua 7.0 New – Guinea –Philippines –SamoaSolomon 26.3 Islands – – <2 22.3ThailandTonga 13.9Vanuatu – – – 13.9Vietnam –South Asia 24.0 – 1998Afghanistan – – 14.6 2000 –Bangladesh –Bhutan –India 22.3 1997–98 39 – <2Maldives 45Nepal – 18.5 1997Pakistan – 100 –Sri 17.7 Lanka – 32 55 1995 21 – 35 – 69 36.0 – 8 2000 90 24 66 16.7 23 76 – – 18 18.6 – c 44.2 52 – 38 94 22.5 8 40.7 – 20 2000 55 37.7 31.0 109 59.5 62 1998 6.6 2.7 1998 c – 24 – 30 79 27.1 – 28 2000 – 25 78 – 2000 – 24.6 70.6 25.7 1998 20.0 38 15.2 1993 21 29.9 20 – 1997 42 25 77 53.6 23.7 21 37.2 74.9 66.3 1995 2000 257 1996 30 69.1 1995–96 – 17 66.8 93 35 16.0 54 8.3 1996 18.0 – – 62.9 2000 95 165 – 109 91 36.4 19 9.2 69.8 77 – – 35.9 824 69.3 – 69 44.8 8.2 – 56.1 1.2 – 77 52.0 85 72 – 16.6 56.7 5.1 17 450 59 2000 c – – 994 4.9 45.5 1999–2000 19.5 2395 0.9 40 c – 43.2 68.4 8.6 54.1 70.3 1997 – 17.0 68.3 59.4 26.9 69.4 c c 4.5 – – 1998–99 11.1 428 4797 29.4 – 6.1 561 – 1998 8.6 1167 70.1 1999 21.0 – 68.2 1735 2000 63.3 1602 – 1995 927 2.9 58.6 – 47.7 62.0 18.8 4.5 – – 54.6 72.4 – 66.6 2805 60.1 1440 373 76.0 356 643 – 31.6 11.0 – – 3.2 – 72.1 40.4 459 5.6 – 1177 1768 – 3.4 42.5 241 860 1933 532 516 Table 11.7 Global Indicators of Chronic Poverty (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 103 Black plate (103,1)

Measuring chronic poverty . 103 prices) 2000 (1995 capita US$ Real GDP per 2000 rate for men, Adult illiteracy rate for women, 2000 Adult illiteracy Life birth, 2000 expectancy at refer stunting data Year to which Proportion of who are stunted, most recent year children under ﬁve Infant births), 2000 mortality rate (per 1,000 live Under-ﬁve births), 2001 mortality rate (per 1,000 live refer poverty data Year to which number of recent year) poverty (The fall below the Average depth of percentage points by which the poor poverty line) (most – – 30 27 31.7 2000 73.3 23.0 7.9 899 Percentage of people living on (most recent year) less than US$1/day Europe and Central Asia Countries Albania ArmeniaAzerbaijanBelarusBosnia and HerzegovinaBulgariaGeorgiaKazakhstanKyrgyz RepublicLatvia –Lithuania 12.8 3.7Macedonia, FYRMoldova <2RomaniaRussian Federation 4.7 <2Tajikistan 25.8 – – 1.5Turkey –TurkmenistanUkraine – – <2 <2Uzbekistan 29.8 1998Yugoslavia, Fed. Rep. 6.1 20.0 – 22.0 2001 – 2.1 2000 35 10.3 2001 – – – 12.1 105 1996 <2 18 19.7 – 26.4 20 28.6 19.1 2.9 25 16 25.2 74 76 1998 2000 15 21.5 2000 2001 17 29 – 13.6 61 2000 – 15 42.4 20.7 19.6 60 1998 21 9 9.7 21 1998 26 32 24 – 53 21 2000 – 2000 1998 1999 72 17 9.7 2000 17 18 99 22 27 2000 11.7 24.8 19 43 72.9 68 20 c 54 71.6 – 12.7 – 52 19 1999 – 6.9 9.6 2.4 1999 38 7.8 1997 c 51 17 – 69.0 – 64.9 17 – 71.1 1995 – 1999 73.3 0.7 1996 67.8 16.0 31.3 1991 15.4 0.6 – – c 2.1 66.7 5.1 976 73.1 66.7 – – – 70.7 72.3 70.0 1998 506 0.3 1996 2000 – 1.0 0.6 1526 – 1.7 2000 – 0.2 67.6 – 2760 0.5 2.7 69.8 66.3 69.0 1503 1512 68.5 0.3 0.5 – 499 0.2 885 1.2 23.5 0.3 1.0 – 1.2 2455 0.5 636 2597 2530 2056 1460 0.4 6.5 – 0.4 – 0.3 3134 386 485 1377 – 896 1240 Table 11.7 Global Indicators of Chronic Poverty (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 104 Black plate (104,1)

104 . Chapter 11 prices) 2000 (1995 capita US$ Real GDP per 2000 rate for men, Adult illiteracy rate for women, 2000 Adult illiteracy Life birth, 2000 expectancy at refer stunting data Year to which Proportion of who are stunted, most recent year children under ﬁve Infant births), 2000 mortality rate (per 1,000 live Under-ﬁve births), 2001 mortality rate (per 1,000 live refer poverty data Year to which number of recent year) poverty (The fall below the Average depth of percentage points by which the poor poverty line) (most <2 – 1998 42 36 15.4 1998 68.9 30.7 16.8 1649 Percentage of people living on (most recent year) less than US$1/day Middle East and North Africa Countries Middle East Iran, Islamic Rep. IraqJordanLebanonOmanSaudi ArabiaSyrian Arab RepublicWest Bank and GazaYemen, Rep.North Africa AlgeriaDjibouti <2 –Egypt, Arab – – Rep.Libyan Arab Jamahiriya – –Morocco –Tunisia 15.7 – – – – – <2 – – 3.1 – – 28.7 1997 <2 <2 12.9 – – 1998 33 – 133 32 28 – 2000 107 28 24 1995 – 28 13 105 28 24 41 85 1999 24 22 49 12 7.8 19 1995 22.1 143 12.2 20.8 37 51.7 44 19.9 7.2 50 22.9 17 27 1997 2000 102 24.9 1996 1995 41 1997 18.0 1996 1996 70.4 15.1 22 1995 – 25.7 73.1 71.2 60.6 22.6 1997 71.6 16.1 71.1 – 2000 12.3 1995 54.1 19.7 39.5 1996 74.8 67.3 33.1 38.4 4.9 1992 69.6 – 70.7 34.4 2000 11.7 7.9 32.5 43.0 56.2 16.9 1616 19.9 42.9 67.6 31.8 2891 839 – 70.2 – 314 6729 45.6 33.4 23.8 63.9 – 9.2 1365 39.4 1226 24.4 1606 38.2 18.6 – 783 1370 2470 Table 11.7 Global Indicators of Chronic Poverty (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 105 Black plate (105,1)

Measuring chronic poverty . 105 prices) 2000 (1995 capita US$ Real GDP per 2000 rate for men, Adult illiteracy rate for women, 2000 Adult illiteracy Life birth, 2000 expectancy at data refer all stunting Year to which Proportion of who are stunted, most recent year children under ﬁve Infant births), 2000 mortality rate (per 1,000 live Under-ﬁve births), 2001 mortality rate (per 1,000 live refer poverty data Year to which number of recent year) poverty (The fall below the Average depth of percentage points by which the poor poverty line) (most – – 40 34 74.1 6.8 6.7 3141 Percentage of people living on (most recent year) less than US$1/day Latin America and the Caribbean Countries Belize BoliviaBrazilColombiaCubaDominicaDominican RepublicEcuadorEl SalvadorGrenadaGuatemala 14.4Guyana 14.0 <2Haiti 9.9HondurasJamaica – –Mexico 37.5Nicaragua 20.2 21.4 57.9Panama 32.3 –Paraguay – 16.0PeruSaint 1999 – Lucia – –St Vincent 28.7 1998 36.9 1998 23.8Suriname – 1998 <2Venezuela 28.7 77 – 8.0 – 23 1995 36 1997 7.6 – 47 19.5 48.7 2000 62 – – – 15.5 25 30 32 26.2 39 – 15 42 9 – 1998 15.0 38.2 50.3 58 25.6 13.5 25 c 2000 10.5 34 34.8 1998 14 25 – 6.1 7 38 44 1998 – 1998 46.0 1998 72 – 27.1 20 23.3 2000 29 123 1996 21 1996 – 32 c 4.6 25 46.4 2000 62.6 30 55 1998 c 43 17 71.5 68.1 25 81 1999 1998 39 – 38.5 67.4 20.7 1998–99 20 26 2000 c 10.1 22 19 37 8.3 14.6 3.4 70.4 70.2 25 c 17.7 31.9 40 65.1 16.4 32 8.0 1996 76.4 14.4 10.9 20 17 14.9 1997 – 8.3 24.9 10.0 23.9 21 1998–99 16.4 38.8 25.8 1994–95 952 1999 27 66.0 3.4 4624 2290 1997 63.2 13.6 18.4 1998 2062 – 6.7 – – 73.1 52.8 23.9 1998 75.3 – 25.5 1996 3.2 – 1752 1425 74.5 70.3 1.9 1558 10.5 52.2 1999 – 68.8 – 9.3 25.3 69.1 – 8.7 7.8 1.1 48.0 6.6 3371 33.2 73.3 711 17.1 – 14.7 73.4 941 3819 367 7.5 5.6 1785 33.7 – 7.9 3832 70.6 5.3 3279 1700 – 466 6.9 2368 – – – 3300 – – 3968 2771 994 Table 11.7 Global Indicators of Chronic Poverty (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 106 Black plate (106,1)

106 . Chapter 11 prices) 2000 (1995 capita US$ Real GDP per 2000 rate for men, Adult illiteracy rate for women, 2000 ta are from World Development Indicators, 2002. Adult illiteracy Life expectancy: Human Development Report 2002. Life birth, 2000 expectancy at Adult illiteracy rates: World Bank World Development Indicators 2002 Real GDP per capita: World Bank World Development Indicators 2002 data refer all stunting Under five mortality rates: UNICEF ChildInfo, http://childinfo.org/cmr/revis/db2.htm Year to which Proportion of who are stunted, most recent year children under five Child stunting data: UNICEF Global Database on Child Malnutrition, http://childinfo.org/eddb/malnutrition/index.htm Infant births), 2000 mortality rate (per 1,000 live Under-five births), 2001 mortality rate (per 1,000 live refer poverty data Year to which number of recent year) poverty (The fall below the Average depth of percentage points by which the poor poverty line) (most 6.5 22.5 – 40.6 33.0 16.0 – 68.3 4.6 1.7 2370.5 11.5 34.7 – 36.5 31.7 17.0 – 69.8 13.9 11.8 3437.1 Percentage of people living on (most recent year) less than US$1/day Sources: Poverty incidence and depth: World Development Indicators, 2003, Table 2.7 except for Ethiopia, Pakistan, South Africa and Uganda where da Sub-Saharan Africa East and Southern AfricaWest AfricaAsia East Asia and PacificSouth AsiaEurope 30.4 and Central Asia Middle East and North Africa Middle EastNorth AfricaLatin America 16.8 and 58.3 the Caribbean 27.0 41.4 20.6 45.1 15.7 – 3.1 25.8 163.7 – – 28.7 12.9 104.6 – 184.0 44.2 – 42.3 110.6 98.1 – 35.0 66.0 37.0 72.4 21.0 42.6 – 53.7 45.5 39.5 48.0 – – 23.6 22.1 43.0 50.7 – 69.4 27.2 – 52.1 62.6 21.2 – 796.6 68.7 34.4 57.3 8.1 68.1 39.2 376.8 33.9 955.2 52.7 19.7 458.8 30.2 2206.0 1432.1 Regional averages Countries Table 11.7 Global Indicators of Chronic Poverty (continued) Notes a. Data refer tob. the Data most refer recent to yearc. a available Data different during refer time the to periodPoverty period a to percentage specified different that and age specified depth groupUnder averages (often five are 0–35 and calculated months) infant usingStunting mortality the averages averages total are are population calculated calculatedIlliteracy figures using using averages for the the are 1998 under number calculated 5<2 of using population refers births the in to in total 2000 percentages 2000 population of in less 2000 than 2% Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 107 Black plate (107,1)

Measuring chronic poverty . 107 2000 children aged Percentage of 5–14 who work, are working Child Labour: 2000 10–14 in the labour force, children aged Percentage of The percentage of children who Under-five e of income for many chronically poor births), 2001 mortality rate (per 1,000 live spectively below the reference weight short to medium term nutritional status, (0–5 years). 2000 Mortality rates for infants (0–1 year old) and children live births), rate (per 1000 Infant mortality five data malnutrition Year of under Notes (%) Severe wasting under five who are Under five wasting: (%) Wasting The percentage of children underweight for their height (most recent year 1995–2000) (%) Severely underweight the percentage of (%) Under five malnutrition: underweight for their age children under five who are (most recent year 1995–2000) Underweight (%) boys rates, 1996–98a. (%) girls primary school age. (Note 1) a percentage of all children of The number of children of any age attending primary school as Gross primary school enrolment Sub-Saharan Africa Countries East and Southern Africa AngolaBotswanaBurundiComorosCongo, Dem. Rep.EritreaEthiopiaKenyaLesotho 105.3 82.7Madagascar 43.7Malawi 45.8 69.7Mauritius 105.6Mozambique 98.7Namibia 48.4 48.1Rwanda 56.2 48.1 82.0Somalia 92.1 12.5 106.2South 92.1 AfricaSudan 34.4 58.3Swaziland 45.1 78.6 124.8 107.8 25.4 60.0Tanzania 92.1 97.0 94.0Uganda 2.4 127.4Zambia 10.2 43.7 137.8 113.9Zimbabwe 13.3 47.1 107.7 82.8 8.5 124.5 22.7 6.6 16.0 33.1 5.0 124.8 17.0 114.0 114.9 9.6 25.4 51.5 16.0 16.4 7.5 129.4 26.1 – 11.5 64.6 12.6 6.5 11.1 4.0 146.2 26.2 1.1 120.7 16.4 29.0 111.3 83.7 59.9 10.7 3.5 5.9 2.2 0.5 65.1 9.1 3.7 25.8 162.1 13.7 6.3 5.0 5.7 114.8 89.0 3.1 2000 7.1 1.4 5.5 15.0 29.4 1995 7.9 c 2000 25.5 6.9 4.6 2000 1.4 3.0 13.0 8.6 25.0 6.7 74 1995 1.2 3.5 128 2000 6.5 114 1.5 17.2 6.7 61 2000 – 1.5 1.5 2000 c 1996 1.3 110 73 172 117 5.4 2000 – 205 1995 3.5 5.3 190 – 1997 86 6.4 79 77 92 1992 14.4 4.0 2000 111 260 28.6 117 0.6 172 17 48.5 126 0.9 1999 37.6 1.6 136 122 c 132 56 – – 38.4 100 26.1 41.1 – 32 183 1999 19 197 133 1995 37 34.1 39.2 20.7 1999 – 67 – 183 – 31.5 1999 104 32.4 2.0 81 55 227 19 36 25 73 17.4 101 41.3 – 66 112 – 165 – 31.3 124 71 123 37 – 149 107 202 36.9 36 43.8 0.0 27.0 12.2 27.4 15.6 – – – – 12 21 – mortality rates and available estimatesgiven of a child child’s labour. height; Wasting andfamilies. and underweight severe and wasting severe indicate underweight weight indicate for analogous height concepts more for than a child’s 2 or weight 3 for standard age. deviations Child re labour is a major sourc Table 11.8 GlobalThe Indicators indicators on in Childhood this Poverty table are not directly indicators of chronic poverty but are key indicators of child development in terms of school attendance, Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 108 Black plate (108,1)

108 . Chapter 11 2000 children aged Percentage of 5–14 who work, are working Child Labour: 2000 10–14 in the labour force, children aged Percentage of The percentage of children who Under-five births), 2001 mortality rate (per 1,000 live (0–5 years). 2000 Mortality rates for infants (0–1 year old) and children live births), rate (per 1000 Infant mortality five data malnutrition Year of under Notes (%) Severe wasting under five who are Under five wasting: (%) Wasting The percentage of children underweight for their height (most recent year 1995–2000) (%) Severely underweight the percentage of (%) Under five malnutrition: underweight for their age children under five who are (most recent year 1995–2000) Underweight (%) boys rates, 1996–98a. (%) girls primary school age. (Note 1) a percentage of all children of The number of children of any age attending primary school as Gross primary school enrolment te d’Ivoire 66.4 89.4 21.4 4.0 10.3 0.9 c 1998–99 102 175 18.6 49 ˆ Sub-Saharan Africa (continued) Countries West Africa Equatorial GuineaGabonGambia, TheGhanaGuineaGuinea-Bissau 115.2LiberiaMaliMauritania 146.3Niger 74.6 153.5NigeriaSao Tome 49.2 71.1Senegal 45.5 154.6 88.1Sierra LeoneTogo 70.3 85.4 78.9 80.8 43.6 71.9 17.0 24.2 95.3 74.4 – 23.1 24.9 85.6 42.2 62.6 62.9 23.2 3.5 37.7 89.4 107.2 – 23.0 5.2 5.2 – 61.4 76.4 5.1 8.6 39.6 141.1 27.3 – 10.1 9.2 9.5 27.2 18.4 16.0 9.1 1.2 14.3 25.1 10.7 2.3 7.2 1.4 8.7 4.1 5.0 – 2.1 14.1 12.4 6.7 2000 – 3.2 9.8 8.3 2000 4.8 1998 3.2 1999 12.3 4.9 92 103 132 1.9 – – 1996 58 112 c 2.1 2000 1999 126 60 153 c 120 211 2000 100 2000 1996 169 159 1998 110 33.8 32.0 90 157 36.7 183 180 12.0 142 31.1 80 58 27 265 80 183 – 14.1 65 235 22.1 – 316 – 231 138 43.6 74 23.9 141 – 15.4 – 13.9 51.1 27.3 – 70 – 26.8 – 72 40 – 66 20 BeninBurkina FasoCameroonCape VerdeCentral African Rep.ChadCongo, Rep.Co 34.5 66.4 45.7 81.5 143.2 50.0 102.0 69.0 55.5 99.0 145.6 49.3 34.3 29.2 24.3 59.3 21.0 13.5 85.1 11.8 7.4 13.9 6.0 4.2 1.8 27.6 13.2 14.3 3.0 8.9 4.5 5.6 9.8 2.5 2.7 3.9 2.1 11.7 0.8 1.0 c b 1998–99 1996 0.9 2.9 2000 1994 1998 c 105 98 1998–99 115 2000 30 95 197 81 158 118 180 155 38 43.5 26.5 108 – 200 23.0 13.6 – 25.4 – 36.6 58 64 – – 66 Table 11.8 Global Indicators on Childhood Poverty (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 109 Black plate (109,1)

Measuring chronic poverty . 109 2000 children aged Percentage of 5–14 who work, are working Child Labour: 2000 10–14 in the labour force, children aged Percentage of The percentage of children who Under-five births), 2001 mortality rate (per 1,000 live (0–5 years). 2000 Mortality rates for infants (0–1 year old) and children live births), rate (per 1000 Infant mortality five data malnutrition Year of under Notes (%) Severe wasting under five who are Under five wasting: (%) Wasting The percentage of children underweight for their height (most recent year 1995–2000) (%) Severely underweight the percentage of (%) Under five malnutrition: underweight for their age children under five who are (most recent year 1995–2000) Underweight (%) boys rates, 1996–98a. (%) girls primary school age. (Note 1) a percentage of all children of The number of children of any age attending primary school as Gross primary school enrolment Asia Countries East Asia and Pacific CambodiaChinaFijiIndonesiaKiribatiKorea, Dem. Rep.Lao PDRMalaysia 110.5Marshall IslandsMicronesia, Fed. Sts. 108.7Mongolia 110.4 127.8Myanmar – 110.8Papua New Guinea –Philippines 106.0 101.7Samoa 114.9 – – 45.9 113.3Solomon 98.6 Islands –ThailandTonga 9.6 119.7 78.4 – 26.4 95.6Vanuatu 113.8 13.4 98.7Vietnam 7.9 – –South 118.0 Asia 60.0Afghanistan 92.1 40.0 91.2 91.9Bangladesh 8.1 115.1 102.4 15.3 18.3Bhutan 0.8 114.1India 92.2 – 101.5Maldives 12.9 101.0 12.7 110.8Nepal 36.0 – – 2.6 4.0 107.1Pakistan 1.2 8.2 28.2Sri 95.4 Lanka 34.2 119.9 18.7 15.4 116.0 2.8 8.7 113.1 – – – – 19.3 2000 0.5 18.6 68.5 124.8 – 126.6 93.1 c – 3.0 – – 5.5 33.1 9.7 99.6 62.2 22.6 110.3 – 1998 – 48.0 95 129.9 47.8 1999 1993 107.0 5.6 – 2000 1.2 127.8 5.8 1998 1.5 109.1 – – – 111.7 18.7 – 43.2 13.1 32 – 138 47.0 35 18 – 5.9 1999 47.1 90 38.2 5.6 23 33.0 2000 3.1 2000 – 10.1 10.0 18.0 23.7 25.0 39 – – 45 21 8 12.0 12.8 1996 – 100 4.8 62 2.6 55 16.8 78 1.0 15.5 52 – – 7.9 20 7.8 55 0.0 6.7 – 1993 30 25.4 15.0 8 c 79 0.5 0.0 3.4 2000 109 76 2.8 1999–2000 69 – – 1997 – c 0.5 24 32 66 25 21 – – 2.3 38 54 21 – 22.9 94 30 1.4 1999 1998–99 1995 – 165 – – 1998 – 35 28 5.4 24 17.2 69 – 77 1995 2000 – 25 77 59 38 17 – 257 – – 72 17 12.2 24.2 – 27.7 42 85 17 93 – 5.2 95 77 20 24.2 – – 91 – – 12.1 109 51.0 – 19 – 3.8 – – 42.1 14 15.4 – – 2.0 – – – – – Table 11.8 Global Indicators on Childhood Poverty (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 110 Black plate (110,1)

110 . Chapter 11 2000 children aged Percentage of 5–14 who work, are working Child Labour: 2000 10–14 in the labour force, children aged Percentage of The percentage of children who Under-five births), 2001 mortality rate (per 1,000 live (0–5 years). 2000 Mortality rates for infants (0–1 year old) and children live births), rate (per 1000 Infant mortality five data malnutrition Year of under Notes (%) Severe wasting under five who are Under five wasting: (%) Wasting The percentage of children underweight for their height (most recent year 1995–2000) (%) Severely underweight the percentage of (%) Under five malnutrition: underweight for their age children under five who are (most recent year 1995–2000) Underweight (%) boys rates, 1996–98a. (%) girls primary school age. (Note 1) a percentage of all children of The number of children of any age attending primary school as Gross primary school enrolment Europe and Central Asia Countries AlbaniaArmeniaAzerbaijanBelarusBosnia and HerzegovinaBulgariaGeorgiaKazakhstanKyrgyz 110.4 73.9 RepublicLatvia 90.0 103.4LithuaniaMacedonia, FYR 99.0 108.9Moldova 73.5 103.4Romania 84.9 99.4 102.9Russian Federation 97.2 95.1 103.3Tajikistan 14.3Turkey 4.1 16.8 101.9 102.4Turkmenistan 2.5 104.9 99.5 99.5Ukraine 96.8 95.5Uzbekistan 116.2 4.3Yugoslavia, 96.8 Fed. 103.8 Rep. 0.6 101.8 4.3 11.0 106.0 0.1 101.5 4.2 3.1 117.2 93.8 109.3 11.1 97.9 103.9 6.0 94.8 6.3 69.9 7.9 1.7 2.0 81.3 79.3 0.4 0.2 96.5 3.0 108.8 3.6 104.1 3.2 5.7 68.8 0.7 1.9 – 1.9 3.4 0.3 82.6 81.5 1.8 2.3 0.5 – 8.3 0.6 2000 1.9 3.6 2000 0.7 18.8 2000 3.0 2000 0.2 0.5 3.9 – c – 1.4 2.5 27 0.4 0.5 2.5 15 5.0 1997 74 0.5 25 1.6 1999 1999 1.9 0.3 – – 3.7 – 30 1999 11.6 6.4 53 b 18 105 35 60 24 1995 0.4 17 0.7 1991 0.3 22 2.8 1.3 1996 15 0.0 0.0 61 0.0 c 18 76 29 1998 19 32 20 2000 1996 18 13 26 27 17 2000 0.0 17 16 – 0.0 0.0 21 38 0.0 17 21 51 0.0 – 32 17 0.0 21 9 54 30 52 – 0.0 – 43 0.0 19 – 68 – 0.0 0.0 20 0.0 – 72 99 – 7.8 0.0 37 0.0 – 0.0 – 0.0 0.0 – – 23 – 25 – Table 11.8 Global Indicators on Childhood Poverty (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 111 Black plate (111,1)

Measuring chronic poverty . 111 2000 children aged Percentage of 5–14 who work, are working Child Labour: 2000 10–14 in the labour force, children aged Percentage of The percentage of children who Under-five births), 2001 mortality rate (per 1,000 live (0–5 years). 2000 Mortality rates for infants (0–1 year old) and children live births), rate (per 1000 Infant mortality five data malnutrition Year of under Notes (%) Severe wasting under five who are Under five wasting: (%) Wasting The percentage of children underweight for their height (most recent year 1995–2000) (%) Severely underweight the percentage of (%) Under five malnutrition: underweight for their age children under five who are (most recent year 1995–2000) Underweight (%) boys rates, 1996–98a. (%) girls primary school age. (Note 1) a percentage of all children of The number of children of any age attending primary school as Gross primary school enrolment Middle East and North Africa Countries Middle East Iran, Islamic Rep.IraqJordanLebanonOmanSaudi ArabiaSyrian 95.0 Arab RepublicWest Bank and GazaYemen, Rep.North Africa 101.7Algeria 80.1 69.4 107.8Djibouti 98.8Egypt, 70.2 Arab Rep. – 72.5Libyan Arab Jamahiriya 10.9Morocco 95.6 112.7 68.4 108.6Tunisia 55.1 72.7 76.7 152.3 – 1.5 95.9 15.9 103.6 12.9 3.0 100.2 5.1 32.0 14.3 23.6 154.3 104.3 113.8 4.9 4.4 86.9 46.1 3.7 0.3 0.5 115.7 45.6 2.8 3.9 4.7 11.7 106.8 6.0 0.9 14.5 122.5 8.7 2.9 1.9 18.2 – 10.7 13.0 0.6 2.8 9.0 1.2 12.9 4.0 1998 2.5 – 2.7 0.2 5.9 2.2 1.6 2.8 6.1 1.8 2.8 2.6 0.6 12.9 36 – 1995 1997 1996 1996 0.4 1995 2000 1.8 2.3 0.6 2.2 1997 2.8 42 24 28 1996 28 24 105 12 1995 0.4 1997 2000 0.5 85 1996 2.6 22 28 33 32 28 133 17 13 1992 37 50 107 102 2000 – 2.3 0.0 0.0 24 0.0 2.2 0.0 19 41 18.7 41 49 22 143 – – – – – – – 0.0 – 9.2 44 0.0 – 27 17 – – 0.6 – – 0.0 – – Table 11.8 Global Indicators on Childhood Poverty (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 112 Black plate (112,1)

112 . Chapter 11 2000 children aged Percentage of 5–14 who work, are working Child Labour: 2000 10–14 in the labour force, children aged Percentage of The percentage of children who Under-five births), 2001 mortality rate (per 1,000 live (0–5 years). 2000 Mortality rates for infants (0–1 year old) and children live births), rate (per 1000 Infant mortality five data malnutrition Year of under Notes (%) Severe wasting under five who are Under five wasting: (%) Wasting The percentage of children underweight for their height (most recent year 1995–2000) (%) Severely underweight the percentage of (%) Under five malnutrition: underweight for their age children under five who are (most recent year 1995–2000) Underweight (%) boys rates, 1996–98a. (%) girls primary school age. (Note 1) a percentage of all children of The number of children of any age attending primary school as Gross primary school enrolment Latin America and the Caribbean Countries BelizeBoliviaBrazilColombiaCubaDominicaDominican RepublicEcuadorEl Salvador 111.5Grenada 103.6Guatemala 111.9 152.0 129.6GuyanaHaiti 114.6 107.8Honduras 99.2 –Jamaica 112.0 156.2 136.3 112.9Mexico 109.5Nicaragua 6.2 99.8Panama 9.5 – 96.0 –Paraguay 6.7 5.7 113.2 4.6 112.8 101.0PeruSaint Lucia 1.3 109.8 152.7St 107.5 4.1 Vincent 1.7Suriname – 98.0 14.8 0.8 103.0 0.6 11.8 1.0Venezuela 112.6 102.9 107.0 150.5 – 102.7 24.2 0.4 113.6 1.8 97.4 11.8 1.9 114.3 0.8 2.3 0.8 100.5 1.5 125.4 24.5 – 27.5 108.1 117.2 4.7 – – 2.0 0.5 – 3.9 92.5 – 12.2 7.5 0.1 0.4 1.1 0.1 127.4 c b 4.0 8.1 6.8 – – 5.0 2.5 0.4 – 1998 1992 90.1 – 1.9 1.2 0.1 11.5 7.8 2000 1996 2000 1.5 7.8 – 0.9 2000 4.7 62 34 2.2 2.0 3.6 – 1.1 1998 25 32 0.1 42 1.5 1999 1.1 c 1998–99 c 1.0 0.7 7 0.5 77 0.6 40 – 1.1 34 1997 23 36 1996 1994–95 47 25 44 – – 3.1 11.4 – 1.9 1998–99 0.3 1998 9 – 14 55 14.4 1999 81 6.0 13.2 39 32 – 30 58 0.6 26 1997 25 – 1998 37 0.0 1996 21 12 – 13.7 17 – 123 15 72 38 14.2 4.3 20 1999 26 29 – 40 – 43 22.8 25 – 0.0 20 7.1 – – 20 25 4.9 30 12.0 – – 39 – 0.0 – – 17 21 2.5 – 22 5.8 – 27 1.8 – – 19 – 0.0 25 – 32 – – – – 0.4 – – – Table 11.8 Global Indicators on Childhood Poverty (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 113 Black plate (113,1)

Measuring chronic poverty . 113 2000 children aged Percentage of 5–14 who work, are working Child Labour: 2000 10–14 in the labour force, children aged Percentage of The percentage of children who Under-five births), 2001 mortality rate (per 1,000 live (0–5 years). 2000 Mortality rates for infants (0–1 year old) and children live births), rate (per 1000 Infant mortality egional average birth rates. Infant mortality rate: UNICEF ChildInfo, http://childinfo.org/cmr/revis/db2.htm five data malnutrition Year of under Sources: Gross primary enrolment rates: World Bank World Development Indicators, 2002 Child labour, children 10–14 (% of age group): World Bank World Development Indicators 2002 Notes (%) Severe wasting Child labour, proportion of children 5–14 years of age that work: UNICEF, http://childinfo.org/eddb/work/images/rank.gif under five who are Under five wasting: (%) Wasting The percentage of children underweight for their height (most recent year 1995–2000) Child underweight and wasting rates: UNICEF Global Database on Child Malnutrition, http://childinfo.org/eddb/malnutrition/index.htm (%) Severely underweight the percentage of (%) Under five malnutrition: underweight for their age children under five who are (most recent year 1995–2000) Underweight (%) boys rates, 1996–98a. 98.5 101.5 7.3 1.5 4.4 1.2 33.0 40.6 1.5 23.5 (%) 125.6 128.0 8.3 1.3 2.1 0.5 31.7 36.5 9.0 19.4 girls primary school age. (Note 1) a percentage of all children of The number of children of any age attending primary school as Gross primary school enrolment Regional averages Countries Middle East and North Africa Middle EastNorth AfricaLatin America and the Caribbean 83.6 98.6 96.2 109.4 18.2 9.4 4.8 2.1 8.1 4.4 1.7 1.2 53.7 39.5 66.0 42.6 4.0 4.3 17.0 – Sub-Saharan Africa East and Southern AfricaWest AfricaAsia East Asia and PacificSouth Asia 76.5Europe and Central Asia 108.7 87.2 66.0 108.1 91.5 32.5 83.1 108.9 17.0 9.8 26.6 45.8 7.6 8.5 8.9 16.5 4.0 11.2 1.9 14.8 2.2 3.4 2.5 104.6 35.0 110.6 163.7 72.4 30.9 184.0 44.2 98.1 28.7 25.9 8.3 14.9 65.4 18.1 14.0 Table 11.8 Global Indicators on Childhood Poverty (continued) Primary school enrolment averagesNutritional are averages calculated are using calculated theInfant using population and the aged under-five under 6–11 mortality 5 in averages population 2000 are in calculated 2000 using the number of births in 2000 Notes: 1. Gross primary enrolment2. can Number be of more births than3. figures 100% Due for if to mortality children data rates outside unavailability, in the number Grenada primary of and school births Kiribati age data are range for calculated are Dominica based attending and on school. Marshall data Islands from are WDI imputed 2002. using the total population for that country and the r a. Data refer tob. the Data most refer recent to yearb. a available Data year during refer or the to periodc. period an other Data specified earlier than refer year that to specified a different age group (often 0–35 months) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 114 Black plate (114,1)

114 . Chapter 11 a school 1996–1998 the number of girls every boy attending enrolment F/M ratio, attending school for Gross primary school le reports the gender differentials 2000 tials are reported as the average vels of chronic poverty. The Gini Adult literacy F/M literate women for every literate man, ratio: the number of than men than women (positive numbers) (negative numbers) expect to live longer The number of years or men to live longer by which women can refers to Year income distribution data 1 inequality Gini index of by lowest 40% Income share held – – 2.70 – 0.84 by lowest 20% Income share held BotswanaBurundiComorosCongo, Dem. Rep.EritreaEthiopiaKenyaLesothoMadagascarMalawi –Mauritius –Mozambique 5.1 –NamibiaRwandaSomalia b,c –South Africa 7.1Sudan 5.6Swaziland 6.4 2.8 15.4Tanzania b,cUganda – b,c –Zambia 6.5 b,c b,c,f b,cZimbabwe 18.0 – 14.9 b,c 9.7 17.1 9.3 42.5 – 2.9 b,c – b,c 17.3 b,c,f b,c b,c,f – 2.7 – b,c 40 – 6.8 44.9 22.9 7.1 1998 56 b,c 38.1 d,e 3.3 – 4.7 b,c b,c 8.4 b,c b,c,f b,c 39.6 b,c b,c,f 8.5 b,c 1995 b,c 17.8 1986–87 1997 28.9 1.70 1999 – 18.2 b,c – d,e 10.9 12.7 b,c 1996–97 – b,c,f 59.3 b,c -0.10 60.9 1.40 b,c -0.20 1983–85 b,c – 0.72 38.2 1.50 2.50 2.30 2.80 37.4 – d,f 52.6 1.80 50.1 b,c 1.07 2.70 0.66 1993–94 1.29 b,c 1.50 1994 0.86 0.81 0.69 0.81 b,c 1993 b,c 0.77 1996 0.48 1998 -0.40 0.66 1.00 7.70 3.70 0.61 0.82 1.09 1.00 0.90 1.40 0.98 2.10 0.10 0.85 1.30 0.73 – 0.62 -0.90 0.83 0.93 0.98 0.99 -0.70 0.97 2.80 0.79 0.98 0.73 0.84 0.91 – 1.00 0.96 0.91 0.67 0.94 f 0.99 1.02 0.90 0.94 0.52 0.97 0.86 f Sub-Saharan Africa Countries East and Southern Africa Angola coefficient and the shares ofnumber the of bottom years 20% by or whichin 40% women literacy are can and reported expect primary for to school income live enrolment. or longer consumption than (depending men on (a which negative is figure available). signifying Life that expectancy men differen live longer on average). The remainder of the tab Table 11.9 GlobalThe Indicators motivation on behind Inequality this table is that – depending on the average income level – higher levels of inequality are likely to be associated with higher le Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 115 Black plate (115,1)

Measuring chronic poverty . 115 a school 1996–1998 the number of girls every boy attending enrolment F/M ratio, attending school for Gross primary school 2000 Adult literacy F/M literate women for every literate man, ratio: the number of than men than women (positive numbers) (negative numbers) expect to live longer The number of years or men to live longer by which women can refers to Year income distribution data 1 inequality Gini index of by lowest 40% Income share held – – 3.40 0.45 0.65 by lowest 20% Income share held te d’Ivoire 7.1 b,c 18.3 b,c 36.7 b,c 1995 0.60 0.71 0.74 ˆ Equatorial GuineaGabonGambia, TheGhanaGuineaGuinea-BissauLiberiaMali –MauritaniaNigerNigeriaSao 4.0 Tome –SenegalSierra Leone 5.6 2.1Togo b,c 6.4 – b,c b,c,f 11.6 b,c 6.4 4.6 15.6 8.6 2.6 4.4 b,c 16.8 b,c – b,c 6.4 1.1 b,c b,c,f b,c b,c 50.2 – b,c 17.6 12.6 – b,c b,c,f 56.2 40.7 9.7 12.6 b,c 40.3 b,c – b,c 16.7 3.1 b,c,f b,c 1998 b,c b,c b,c 37.3 50.5 1991 b,c b,c,f 1999 – 1994 50.5 50.6 b,c b,c 62.9 41.3 2.80 1995 b,c b,c 1994 – 2.80 2.60 b,c,f 3.20 b,c 1996–97 1.00 1995 1989 0.67 1995 – 2.40 3.20 0.43 2.00 0.78 0.80 – 0.40 0.60 0.85 2.60 3.70 – – 0.59 0.70 0.58 0.90 0.79 0.77 0.35 0.63 – – 0.58 0.94 0.54 0.99 0.70 0.83 0.64 2.40 – 0.82 0.69 0.74 0.59 – 0.76 Sub-Saharan Africa (continued) Countries West Africa Benin Burkina FasoCameroonCape VerdeCentral African Rep.ChadCongo, Rep.Co 4.6 2.0 4.6 – b,c b,c b,c – – 11.8 12.9 6.9 b,c b,c b,c 55.1 47.7 61.3 b,c b,c b,c 1998 – 1996 1993 – – 2.00 1.50 3.30 0.41 0.84 0.58 5.80 0.69 4.20 2.40 0.82 0.66 0.78 0.85 0.66 0.98 0.94 0.58 Table 11.9 Global Indicators on Inequality (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 116 Black plate (116,1)

116 . Chapter 11 a school 1996–1998 the number of girls every boy attending enrolment F/M ratio, attending school for Gross primary school 2000 Adult literacy F/M literate women for every literate man, ratio: the number of than men than women (positive numbers) (negative numbers) expect to live longer The number of years or men to live longer by which women can refers to Year income distribution data 1 inequality Gini index of by lowest 40% Income share held 6.9 b,c 17.6 b,c 40.4 b,c 1997 4.70 0.72 0.86 by lowest 20% Income share held ChinaFijiIndonesiaKiribatiKorea, Dem. Rep.Lao PDRMalaysiaMarshall IslandsMicronesia, Fed. Sts.MongoliaMyanmar 5.9Papua New Guinea 9.0Philippines – –Samoa – d,eSolomon Islands b,cThailand 7.6 – –Tonga 4.4 16.1VanuatuVietnam 21.5 b,c 4.5 7.3South d,e Asia – –Afghanistan d,e –Bangladesh b,c – 19.0 5.4 b,cBhutan b,c 12.5 –India – – 40.3 –Maldives 31.7 b,c b,c 12.4Nepal 6.4 19.5 d,ePakistan d,e –Sri – Lanka b,c – 37 – 14.2 – b,c 8.0 b,c b,c 49.2 1998 – – 1999 – 8.7 – – b,c b,c 50.9 – b,c 16.2 33.2 d,e – b,c 1997 46.2 1997 8.1 – b,c – 19.4 b,c 4.30 b,c – – 3.90 7.6 9.5 20.7 1996 1995 – 8.0 b,c b,c b,c 41.4 – b,c – b,c 2.50 10.30 0.83 1997 b,c b,c – 19.7 0.89 36.1 b,c 7.40 – 3.50 19.1 – 22.4 1.90 – 33.6 4.00 – 1998 19.8 b,c b,c 0.52 0.91 – – – 1.03 b,c 4.00 b,c b,c 0.96 1998 – b,c 37.8 0.96 0.80 1.00 1995–96 – 36.7 – 31.2 5.90 4.80 34.4 – b,c 0.85 1.00 1.00 – – b,c 2.50 b,c 4.70 1997 0.98 – b,c 6.60 0.86 1.04 0.10 1995–96 1996–97 0.97 0.91 – 1995 1.03 – 3.10 – – – 0.96 0.97 1.00 0.57 -0.50 – -0.30 0.97 0.99 5.80 – – 2.50 0.95 0.91 0.66 1.01 -1.50 0.40 0.96 0.48 – 0.94 – 0.96 – 1.00 0.87 0.78 0.57 0.50 0.99 f 0.85 0.98 Asia Countries East Asia and Paciﬁc Cambodia Table 11.9 Global Indicators on Inequality (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 117 Black plate (117,1)

Measuring chronic poverty . 117 a school 1996–1998 the number of girls every boy attending enrolment F/M ratio, attending school for Gross primary school 2000 Adult literacy F/M literate women for every literate man, ratio: the number of than men than women (positive numbers) (negative numbers) expect to live longer The number of years or men to live longer by which women can refers to Year income distribution data 1 inequality Gini index of by lowest 40% Income share held – – – 5.90 0.84 1.01 f by lowest 20% Income share held Europe and Central Asia Countries Albania ArmeniaAzerbaijanBelarusBosnia and HerzegovinaBulgariaGeorgiaKazakhstanKyrgyz RepublicLatviaLithuania –Macedonia, 5.5 FYR 6.9MoldovaRomania 11.4Russian Federation b,c d,e 10.1TajikistanTurkey b,c 6.1 6.7 7.6Turkmenistan 14.9 18.4 d,eUkraine –Uzbekistan – 26.6 d,e b,c b,c 7.6 7.8Yugoslavia, Fed. Rep. b,c d,e 24.0 4.4 5.6 b,c 17.5 18.2 19.3 d,e 8.0 b,c 44.4 36 d,e b,c 8.0 d,e 21.7 d,e b,c b,c 6.1 20.5 – b,c 20.4 5.8 b,c – – 26.4 d,e 13.0 b,c 8.8 15.8 4.0 b,c 37.1 35.4 34.6 b,c d,e 1996 21.1 b,c b,c 1995 d,e b,c 1998 20.9 b,c d,e b,c d,e b,c b,c 16.3 32.4 b,c 32.4 16.0 1997 48.7 b,c 1996 1996 1999 40.6 22.1 13.5 – – 6.00 b,c d,e 31.1 b,c 7.00 b,c 11.60 b,c 34.7 d,e b,c 1998 b,c 1996 40.8 b,c 7.70 41.5 1998 – 11.20 0.98 8.20 1997 7.90 b,c – 29 44.7 1998 b,c 1.00 b,c – 1998 11.10 10.40 0.99 1998 b,c b,c 1994 – – 12.40 – 1.06 – 7.50 1998 6.80 0.96 1999 1.00 1.00 1.00 4.30 5.80 0.97 f 6.70 1.00 5.10 0.99 1.00 1.00 0.98 1.00 0.98 10.80 5.80 0.94 – 0.99 0.98 – 0.82 0.99 0.99 – 0.98 1.00 0.99 f 0.97 0.98 0.91 1.00 – 0.98 0.97 1.02 f Table 11.9 Global Indicators on Inequality (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 118 Black plate (118,1)

118 . Chapter 11 a school 1996–1998 the number of girls every boy attending enrolment F/M ratio, attending school for Gross primary school 2000 Adult literacy F/M literate women for every literate man, ratio: the number of than men than women (positive numbers) (negative numbers) expect to live longer The number of years or men to live longer by which women can refers to Year income distribution data 1 inequality Gini index of by lowest 40% Income share held – – – 1.80 0.83 0.93 by lowest 20% Income share held IraqJordanLebanonOmanSaudi ArabiaSyrian Arab RepublicWest Bank and GazaYemen, Rep.North Africa AlgeriaDjiboutiEgypt, Arab – Rep. 7.6Libyan – Arab – JamahiriyaMorocco – – –Tunisia b,c 7.4 19.0 – – b,c 9.8 7.0 – – – b,c – – – 19.6 b,c b,c 6.5 36.4 5.7 b,c 23.0 18.6 b,c – – b,c b,c – – – 33.4 b,c – b,c – – 17.1 1997 15.6 b,c 28.9 35.3 b,c b,c 1998 – b,c b,c 2.70 39.5 – 41.7 1995 1995 b,c 2.20 b,c 0.88 – 1998–99 3.10 2.40 2.50 1995 – 2.90 3.20 3.00 0.37 1.02 3.70 0.70 0.87 0.68 0.81 0.77 4.00 0.66 2.50 0.75 – 2.50 0.55 0.58 0.84 0.96 0.91 0.97 0.75 0.74 0.95 0.92 0.91 – 0.72 0.81 0.99 0.94 0.70 f Middle East and North Africa Countries Middle East Iran, Islamic Rep. Table 11.9 Global Indicators on Inequality (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 119 Black plate (119,1)

Measuring chronic poverty . 119 a school 1996–1998 the number of girls every boy attending enrolment F/M ratio, attending school for Gross primary school 2000 Adult literacy F/M literate women for every literate man, ratio: the number of than men than women (positive numbers) (negative numbers) expect to live longer The number of years or men to live longer by which women can refers to Year income distribution data 1 inequality Gini index of by lowest 40% Income share held – – – 2.80 1.00 0.97 by lowest 20% Income share held Latin America and the Caribbean Countries Belize BoliviaBrazilColombiaCubaDominicaDominican RepublicEcuadorEl SalvadorGrenadaGuatemalaGuyana 4.0Haiti 2.2 3.0Honduras 5.1Jamaica b,c – –Mexico d,e d,eNicaragua d,ePanama 5.4 3.3 13.2ParaguayPeru – 7.6 9.6 3.8 13.7Saint Lucia b,c d,e b,cSt Vincent 6.3Suriname d,e d,e d,e – – d,e 2.2Venezuela – 14.8 10.6 44.7 6.7 b,c 3.5 60.7 57.1 10.6 2.3 47.4 d,e b,c – d,e b,c 3.6 b,c 17.0 1.9 d,e d,e d,e d,e b,c d,e 8.6 43.7 5.2 1999 52.2 4.4 b,c 17.4 – – b,c – – d,e 1998 1996 1998 10.8 – 55.8 8.2 d,e 3.0 b,c d,e d,e d,e 11.7 b,c 40.2 7.9 – d,e d,e 1995 b,c 3.50 1998 56.3 d,e 15.1 13.5 b,c 37.9 b,c 7.90 6.60 5.20 d,e 1998 53.1 60.3 – – d,e 11.2 d,e 1993 d,e – 48.5 b,c 0.86 57.7 5.20 d,e 1998 6.00 b,c 1.00 1.00 d,e 42.6 1.00 2000 46.2 b,c 5.80 1998 d,e 1998 3.90 – 49.5 8.40 d,e 1997 d,e 0.96 0.96 1998 0.93 – 5.70 0.97 1.00 – 0.95 1995 4.00 1996 d,e 0.80 – 1.00 6.00 4.70 – 0.99 1998 4.60 1.00 0.97 1.00 4.60 6.00 1.09 5.30 0.89 5.00 – 0.96 0.99 1.01 0.98 0.99 – 5.80 0.98 1.03 0.92 1.01 – 0.90 0.99 f 1.02 – – 5.20 0.99 0.95 0.97 1.01 f 0.98 – – – 1.03 – – Table 11.9 Global Indicators on Inequality (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 120 Black plate (120,1)

120 . Chapter 11 a school 1996–1998 the number of girls every boy attending enrolment F/M ratio, attending school for Gross primary school 2000 Adult literacy F/M literate women for every literate man, ratio: the number of ealth than men than women ional average for that age group. (positive numbers) (negative numbers) expect to live longer The number of years or men to live longer by which women can Income shares and Gini index: World Bank World Development Indicators, 2002 Life Expectancy differential: computed using data from Human Development Report 2002 Literacy ratio: computed using data from in World Bank World Development Indicators 2002 refers to Year income Enrolment rate ratio: computed using data from World Bank World Development Indicators, 2002 distribution data Sources: Total population, female & male population ages 15 or over: World Bank World Development Indicators 1 inequality Gini index of by lowest 40% Income share held 6.13.0 16.6 9.8 40.6 55.5 – –x 9.34 6.53 0.97 0.98 0.97 0.98 by lowest 20% Income share held Middle East and North Africa Middle EastNorth AfricaLatin America and the Caribbean 7.4 8.1 19.5 20.2 34.1 33.6 – – 2.09 3.22 0.76 0.68 0.87 0.90 Regional averages Countries Sub-Saharan Africa East and Southern AfricaWest AfricaAsia East Asia and PacificSouth AsiaEurope and Central Asia 5.9 6.4 4.7 15.5 8.3 16.8 13.1 44.0 20.1 39.5 48.5 36.6 – – – 1.92 – 4.37 1.28 0.78 0.82 0.86 0.73 0.88 0.65 1.01 0.79 0.84 Income inequality averages areIlliteracy calculated averages using are the calculated totalEnrolment using population averages the figures are male for calculated and 1998 using female the population population aged agedPopulation 15 6–11 aged and in 15–24 over 200 and in 6–11: 2000 United Nations PopulationNotes: Division, http://esa.un.org/unpp/sources.html 2. Due to dataa. unavailability, Data the refer 6-11 to ageb. the group Data most for refers recent primary to yearc. enrolment expenditure available Ranked in shares during by Yugoslavia by the per is percentilesd. period capita imputed of Data specified expenditure using population refers the toe. total income Ranked population shares by for by per that percentiles country of and population the reg Table 11.9 Global Indicators on Inequality (continued) 1. The Gini index measures income inequality. A score of 0 would mean that all incomes were equal and a score of 100 would mean that one person held all the w Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 121 Black plate (121,1)

Measuring chronic poverty . 121 Average 1990–2000 growth rate Average 1980–1990 growth rate ative numbers for FDI mean that more s. These are all expressed as a proportion of Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Average 1981–1990 Sub-Saharan AfricaCountries East and Southern Africa Aid as proportion of GNI Net FDI inﬂows as proportion of GNI Openness (Exports plus imports/GNI) Growth of real GDP per capita AngolaBotswanaBurundiComorosCongo, Dem. Rep.EritreaEthiopiaKenyaLesotho 2.3 7.6Madagascar 16.1 5.9Malawi 31.4Mauritius 11.3Mozambique 2.1Namibia 19.0 – 3.6 8.6 16.6RwandaSomalia 8.7 6.5 14.4 9.7South 0.6 Africa 13.8Sudan 20.0 15.2 . 9.2 . 18.5 .Swaziland 19.8 3.6Tanzania 8.5 1.7 12.8 7.9 3.6Uganda 0.2 25.3 1.6 10.9Zambia –0.2 26.6 0.7 10.6 44.6Zimbabwe 1.0 51.7 11.4 5.0 – 8.5 3.6 0.1 0.2 – 5.0 0.0 26.8 0.0 30.5 7.2 6.3 0.4 24.9 0.5 23.7 – 19.2 0.4 0.2 1.7 0.3 0.6 8.0 1.7 4.4 1.2 18.1 14.4 0.0 4.0 4.5 . 3.5 0.0 . 3.6 0.1 . 17.7 0.7 – 0.2 15.3 105.6 0.7 . . 0.4 . 15.6 34.5 27.8 – 0.8 0.9 1.5 46.2 5.8 61.7 2.3 0.9 6.1 3.1 11.6 1.2 13.1 1.1 –0.4 – – 2.1 – 13.3 32.7 28.5 25.4 60.4 0.2 2.7 – – – 0.0 4.7 2.5 3.7 0.1 139.6 6.1 52.6 164.4 34.6 – 0.0 32.6 – 0.9 33.2 2.2 57.5 53.4 0.8 –0.1 32.5 1.2 135.4 5.2 – – 115.0 1.5 . . . 64.6 –0.8 49.5 2.1 1.6 . . 46.1 0.8 . 6.4 3.6 65.1 0.3 28.9 116.8 51.5 –3.0 128.5 1.5 3.4 109.0 101.8 –2.3 2.1 62.1 –2.4 59.4 –3.8 58.6 3.6 50.0 64.5 2.4 –2.4 33.1 6.9 153.9 130.6 54.5 2.2 1.1 – – – – –2.2 0.6 27.4 45.6 – –1.0 166.2 32.4 69.4 –1.4 5.1 1.5 1.9 43.2 –1.2 –0.9 – – 49.5 – 32.1 55.1 146.3 –1.0 72.1 1.5 – 3.2 4.1 73.0 37.9 33.0 –2.1 35.9 –1.2 3.1 –1.9 76.4 61.0 –1.4 1.9 –0.3 – –2.0 0.1 0.7 – 5.2 0.1 –1.9 3.3 –0.5 investment is leaving the countryThis than table is reports being on received. measuresgross of national inflows income to (GNI) countries,The and specifically reported table aid for also flows, the reports foreign 1980s per and direct capita investment, 1990s. growth and rates openness for measured the as 1980s total and trade the flow 1990s. Table 11.10 ResourceThis Inﬂows table allows assessment of the extent to which chronically poor countries are less open to trade or receive less FDI or aid inflows than others. Neg Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 122 Black plate (122,1)

122 . Chapter 11 capita Average 1990–2000 growth rate Average 1980–1990 growth rate Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Average 1981–1990 te d’Ivoire 3.1 9.0 4.1 0.5 1.6 1.1 69.3 74.0 85.3 –1.7 –0.5 ˆ BeninBurkina FasoCameroonCape VerdeCentral African Rep.ChadCongo, Rep.Co 13.3 10.0 15.0 2.7 35.7 18.8 13.1 13.0 5.3 5.7 14.2 25.8 15.5 11.1 8.0 11.6 14.2 17.2 4.6 0.1 0.0 0.5 1.5 9.4 0.2 0.9 0.6 1.1 0.2 1.2 1.1 2.1 0.2 0.5 1.4 0.5 0.2 1.0 1.8 0.4 40.9 49.4 50.6 0.4 50.7 1.1 – 40.5 46.1 38.8 102.5 44.3 41.6 71.9 40.4 44.3 116.1 29.1 57.6 46.4 85.0 1.7 0.7 120.4 –1.0 –0.4 48.6 – 2.5 1.8 –0.4 –0.7 –1.1 1.1 3.1 –2.7 –0.5 Equatorial GuineaGabonGambia, TheGhanaGuineaGuinea-Bissau 39.4LiberiaMaliMauritania 34.2 21.2Niger 2.3Nigeria 50.1Sao 6.9 Tome 11.8Senegal 16.9 4.3 2.1Sierra Leone 10.2Togo 50.2 10.1 25.1 10.4 20.9 11.8 1.3 0.3 14.6 – 39.6 0.4 42.0 12.1 23.9 17.6 5.2 0.7 14.5 7.0 31.8 1.9 16.0 110.3 0.5 0.8 23.3 0.2 13.7 . . . 15.9 0.6 12.1 2.8 21.7 8.9 –2.2 11.7 80.4 0.6 0.5 10.2 0.8 1.5 1.3 0.1 0.8 9.9 29.6 3.3 – 3.0 0.3 0.0 0.0 2.0 5.8 0.5 2.1 – 1.4 111.3 –0.8 2.1 0.3 159.2 95.0 0.4 2.2 52.0 28.0 4.1 0.7 0.5 118.2 3.3 0.1 – . 1.4 . 153.2 . 87.1 0.8 21.5 55.3 120.4 70.3 108.9 2.6 1.3 49.9 0.2 – 2.4 46.8 – 71.8 47.4 78.6 118.8 90.0 95.6 –0.1 44.1 2.5 57.5 37.4 67.9 57.3 –0.8 17.0 – 38.8 112.3 97.9 0.2 –0.0 0.9 81.9 96.6 65.4 45.4 67.2 –0.4 – 115.6 38.6 – –0.9 0.1 1.8 93.4 72.3 –0.3 50.7 70.1 –2.3 – 1.2 – –1.1 1.1 85.4 1.3 –3.4 0.5 –1.5 –0.7 –0.2 –1.8 – –6.3 0.7 –1.4 Sub-Saharan Africa (continued) Aid as proportion of GNI Net FDI inﬂows as proportion of GNI Openness (Exports plus imports/GNI) Growth of real GDP per Countries West Africa Table 11.10 Resource Inﬂows (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 123 Black plate (123,1)

Measuring chronic poverty . 123 Average 1990–2000 growth rate Average 1980–1990 growth rate Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Aid as proportion of GNI Net FDI inflows as proportion of GNI Openness (Exports plus imports/GNI) Growth of real GDP per capita Average 1981–1990 Asia Countries East Asia and Pacific CambodiaChinaFijiIndonesiaKiribatiKorea, Dem. Rep.Lao PDRMalaysia 2.4Marshall IslandsMicronesia, Fed. 0.4 Sts.Mongolia 1.3 – 12.2Myanmar 3.4 33.3Papua New GuineaPhilippines 0.4 8.7Samoa – – 1.1 12.6 0.7 2.6Solomon – Islands 23.9ThailandTonga 0.2 12.0 17.2 – 32.6Vanuatu 43.2 1.2 0.0 0.1 – 2.0 21.8Vietnam . . .South 1.8 Asia 22.4 16.8 0.6 8.2Afghanistan 39.6 56.5 24.0 22.7 0.4Bangladesh 4.1 0.1 2.5 – –Bhutan – 1.1India 1.5 16.0 0.2 4.3 7.5 23.4Maldives 22.8 20.7 – – 28.6 0.7 3.9Nepal 3.3 2.6 1.9Pakistan – . . 0.6 . 24.0 – 0.7 0.7Sri Lanka 4.0 17.3 3.6 4.2 11.6 6.5 – 19.5 -3.0 – – – 5.2 0.0 4.3 12.3 – 0.5 2.2 . 0.7 . – . 12.1 23.1 . 4.2 22.1 . 3.5 20.5 . 0.7 0.4 3.8 47.2 2.1 62.2 98.4 1.9 10.0 . . 20.0 . . 5.4 . . 2.9 130.5 1.2 4.2 – 1.9 41.4 8.5 0.1 14.4 – – 3.4 . 5.8 0.6 . 59.5 1.6 . 2.5 120.1 115.6 3.1 87.0 71.7 10.7 9.3 – 0.0 1.8 – 2.6 49.1 3.6 2.7 4.5 1.2 3.7 93.7 12.6 69.2 . – . 131.6 0.3 186.5 . 0.0 – – 0.0 93.2 – 0.0 6.9 7.2 – 1.2 129.0 2.8 51.9 – 7.6 1.7 1.3 229.8 1.4 13.3 – 9.4 139.9 4.4 0.2 0.0 – 0.2 – – – 0.0 0.0 2.2 – 4.1 0.4 – 56.8 86.9 – 9.0 108.5 -1.2 92.5 147.4 0.7 3.1 2.4 – 2.5 0.2 0.4 – 0.6 – . . . 0.0 0.2 106.5 – 91.9 0.9 – – 0.1 – 1.3 4.4 2.3 24.8 – 0.5 -1.3 114.7 – 65.5 126.0 0.1 -0.7 – 3.7 0.5 – 83.0 – 1.1 3.2 – – 27.1 15.3 -0.1 -1.3 -0.5 1.9 79.1 31.7 6.0 – 0.7 34.7 66.2 – -2.0 – 33.2 24.3 1.0 1.9 89.5 52.4 – 3.4 38.0 – 78.2 189.4 30.5 2.1 -0.4 5.3 55.8 2.7 34.6 90.2 – 3.6 3.0 5.6 3.2 2.5 3.5 2.7 3.6 – 4.6 2.5 1.4 3.9 Table 11.10 Resource Inflows (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 124 Black plate (124,1)

124 . Chapter 11 Average 1990–2000 growth rate Average 1980–1990 growth rate Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Aid as proportion of GNI Net FDI inﬂows as proportion of GNI Openness (Exports plus imports/GNI) Growth of real GDP per capita Average 1981–1990 AlbaniaArmeniaAzerbaijanBelarusBosnia and HerzegovinaBulgariaGeorgiaKazakhstan 0.5Kyrgyz Republic – –Latvia –Lithuania –Macedonia, FYR 16.9Moldova 29.2 0.1Romania 8.2 2.9 – –Russian – FederationTajikistan 0.5 8.3 16.0Turkey 11.2 1.9 –Turkmenistan – 2.8 –Ukraine 10.3 6.1 0.5Uzbekistan 0.1 0.0 0.0 – –Yugoslavia, Fed. Rep. – 2.7 0.6 – 3.4 17.6 1.1 1.2 5.6 1.1 – – 2.5 0.5 – 0.0 3.4 0.5 0.0 0.9 3.9 – 7.1 12.1 – 1.3 0.9 – – 0.3 – 6.9 3.8 0.7 0.6 0.0 0.7 9.0 0.2 2.8 1.2 7.3 – 2.5 6.9 – – 2.3 0.9 2.6 4.4 15.2 37.2 1.1 0.0 0.3 – 0.7 – 0.2 8.3 0.0 – 13.4 – 1.3 –0.2 4.7 1.8 2.7 4.3 56.5 6.9 – 2.5 0.7 – – 79.9 0.2 – 1.8 2.7 88.3 – 4.9 – – 5.7 – 59.3 3.3 – – – 1.1 1.1 120.1 100.7 0.5 10.0 2.2 2.8 85.0 74.1 0.0 – –0.8 84.0 79.1 – – 0.9 83.1 137.2 122.5 – 2.4 – 1.0 0.5 47.2 – . . . 0.0 98.7 87.4 0.7 106.4 103.0 84.1 1.9 2.6 106.2 – 1.3 59.3 31.1 58.8 121.7 – –4.5 107.5 100.1 – – 96.7 –1.1 –6.6 –1.3 – 70.7 –1.0 – 126.9 44.3 – 73.9 – 2.8 –13.0 – –5.1 80.4 1.5 55.8 –0.4 –2.7 68.5 165.4 1.8 –3.5 116.4 –1.6 118.4 81.9 –3.7 –3.9 –1.5 82.8 –9.7 2.8 – – – 1.9 –10.6 –6.0 –7.6 – –2.3 Countries Europe and Central Asia Table 11.10 Resource Inﬂows (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 125 Black plate (125,1)

Measuring chronic poverty . 125 Average 1990–2000 growth rate Average 1980–1990 growth rate Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Aid as proportion of GNI Net FDI inﬂows as proportion of GNI Openness (Exports plus imports/GNI) Growth of real GDP per capita Average 1981–1990 Middle East and North Africa Countries Middle East Iran, Islamic Rep.IraqJordanLebanonOmanSaudi Arabia 0.0Syrian Arab RepublicWest Bank and GazaYemen, Rep.North Africa 0.2 13.6 0.1Algeria 5.5Djibouti 4.7 0.0Egypt, – Arab 1.1 Rep.Libyan Arab Jamahiriya 0.1 8.4 –Morocco 1.8 8.4 2.3Tunisia 0.0 12.2 0.5 –0.1 0.0 6.6 6.0 0.3 . 1.1 . . 5.2 1.0 – 0.0 12.5 . . . 0.0 – 0.9 3.7 4.7 0.7 – 0.1 3.6 0.3 16.3 2.8 – – 1.5 0.0 1.8 2.3 1.3 0.3 . –2.7 . 0.7 . – 0.7 12.6 1.3 – – 0.8 22.2 6.7 –0.0 1.3 2.6 – 2.6 1.8 0.7 0.0 1.2 . . . 42.4 . . . 119.2 ...... 0.0 0.4 1.2 –2.4 – 43.9 – 0.6 1.6 – 55.5 121.9 86.6 – – 0.0 0.7 1.3 – 67.1 70.0 0.0 2.4 . . 111.0 . –0.7 – 75.3 – – 48.5 55.5 72.6 0.0 65.8 50.8 – 3.9 –1.8 – 2.5 75.3 – 52.7 49.6 –1.1 54.3 – 84.6 91.9 80.8 0.6 – –4.8 – 64.3 38.9 59.5 2.7 4.7 88.9 5.3 107.4 –0.5 – –0.2 68.6 2.9 – – 91.6 – 1.5 –0.3 1.6 2.4 1.1 –3.8 0.4 – 3.1 Table 11.10 Resource Inﬂows (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 126 Black plate (126,1)

126 . Chapter 11 Average 1990–2000 growth rate Average 1980–1990 growth rate Average 1991–2000 2000 Average 1981–1990 Source: All statistics computed using data from World Bank World Development Indicators, 2002 Average 1991–2000 2000 Average 1981–1990 Average 1991–2000 2000 Aid as proportion of GNI Net FDI inﬂows as proportion of GNI Openness (Exports plus imports/GNI) Growth of real GDP per capita Average 1981–1990 BelizeBoliviaBrazilColombiaCubaDominicaDominican RepublicEcuadorEl Salvador 8.3Grenada 8.0Guatemala 0.1 0.2Guyana 2.1Haiti 4.2 16.1 –Honduras 9.5Jamaica 0.0 0.2 1.2 0.7Mexico 8.5 1.9Nicaragua 8.6 12.7 – 5.9Panama 1.8Paraguay 0.1 0.2 14.8 1.4 0.3 3.5Peru 1.8 6.5Saint Lucia 4.3 7.9 0.5 7.8 1.6St Vincent . . . 7.0 0.6 1.4Suriname 26.5 1.2 1.1 1.4Venezuela 10.7 0.1 3.0 3.4 4.4 12.0 9.7 6.1 1.0 1.4 – 1.5 1.8 16.4 2.1 2.3 0.6 3.2 41.5 0.3 0.1 5.1 2.2 9.0 1.6 4.1 5.1 7.8 8.9 8.4 0.8 1.3 1.3 0.1 – 5.6 0.3 5.5 2.9 26.6 3.1 0.1 4.8 –0.0 1.4 122.5 5.1 3.9 1.1 8.9 47.4 0.4 0.8 6.7 0.2 1.1 13.4 13.0 17.2 1.1 28.6 0.4 59.0 5.2 0.0 112.4 . 0.1 . . 1.0 1.4 113.7 1.6 0.8 48.4 9.0 0.2 2.2 2.0 0.5 1.2 18.8 36.3 4.4 9.4 0.4 4.0 110.7 66.9 50.3 5.6 – 49.2 116.8 0.1 2.3 10.4 114.2 42.6 0.1 0.3 4.8 34.7 3.1 6.1 128.4 23.0 42.3 – 1.7 6.2 69.1 58.7 2.2 114.9 10.6 57.3 0.2 2.3 114.6 – 8.6 –1.9 3.1 43.4 58.3 226.7 43.5 15.5 –0.4 6.1 1.3 102.1 73.2 1.1 0.4 54.2 2.1 70.3 – 136.2 30.9 2.9 33.9 6.9 1.3 207.8 33.4 88.1 1.3 47.9 8.4 104.9 73.3 1.3 0.8 –0.5 51.1 96.6 –1.5 4.1 51.7 5.1 150.0 3.7 . – . . –3.0 39.9 98.7 33.4 –1.6 149.1 99.1 75.6 –0.3 121.3 72.6 2.4 64.6 135.0 3.1 45.8 81.8 –2.3 –0.7 122.0 30.5 4.5 1.4 – 72.1 1.2 –4.0 55.6 121.2 –0.3 50.2 32.4 128.9 –2.7 33.8 0.4 –0.7 0.1 –0.3 0.4 5.5 1.8 46.4 34.7 5.1 –2.9 2.7 –0.7 1.1 –1.7 –1.7 2.5 2.2 –0.1 1.8 Countries Latin America and the Caribbean Table 11.10 Resource Inﬂows (continued) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 127 Black plate (127,1)

Measuring chronic poverty . 127 Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 128 Black plate (128,1)

128 . Chapter 11

Figure 11.1 Aid Concentration Curve for all DAC donors combined (2001) 100 Middle income countries The steep rise in the with relatively small populations curve shows that some small and Russia receive disproportionately large very poor countries, mostly in sub-Saharan shares of aid. 80 Africa, receive relatively greater amounts of aid per capita – but this may still be below the level needed to achieve the China Egypt MDGs. 60 The length of the Chinese Indonesia segment is shorter than that for India, India Pakistan because a smaller share of people are estimated to live in absolute 40 Nigeria poverty in China. 37% of the world’s absolute poor

Cumulative share of aid (%) Cumulative live in India, but receive less than their 20 Bangladesh proportionate share of world aid. Tanzania Large aid ﬂows go to poor Mozambique South Asian countries such as Bangladesh and Pakistan. 0 0 20406080100 Cumulative share of world population living under $1 per day

Box 11.01 Concentrating aid Figure 11.2 Aid Concentration Curves for the Netherlands, Japan on the poorest? and the United States (2001) Aid concentration curves show 100 whether donors’ Official Development US Assistance is targeted towards or Japan away from the poorest countries. 80 DAC all members The diagonal line shows what aid allocation would look like if it was Netherlands 60 allocated in direct proportion to the share of the world’s poor living on

less than US$1/day in each country. 40 If most of a donor’s aid goes to the

poorest countries, then its share of aid (%) Cumulative concentration curve will be above the 20 diagonal line and vice versa. Figure 11.1 shows the concentration curve for all DAC 0 0 20406080100 donors combined. Cumulative share of population living under $1 per day (%)

The distribution of development assistance by the Figure 11.3 Aid Concentration Curves for the UK, Germany and main bilateral aid donors France (2001) Figures 11.2 and 11.3 show aid 100 France concentration curves for the ‘big six’ bilateral donors – the United States, Germany 80 Japan, Germany, the UK, France and DAC all members the Netherlands – together with, for comparative purposes, the aid UK 60 concentration curve for all members of the DAC. The aid concentration curves of the 40 Netherlands and the UK are broadly progressive, while those of Japan and Cumulative share of aid (%) Cumulative the United States are fairly regressive. 20 The aid concentration curves of France and Germany lie in between, being fairly progressive at the low end 0 0 20406080100 of the global poverty distribution, but Cumulative share of population living under $1 per day (%) then becoming highly regressive at the Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 129 Black plate (129,1)

Measuring chronic poverty . 129

top end of the distribution. The main reason why the Netherlands and UK’s Figure 11.4 Aid Concentration Curves for the EC (2001) bilateral aid programme are progressive is that they give large 100 Population Brazil amounts of aid (relative to their South Africa numbers of poor people) to a number Population living under $1 per day Turkey 80 of poor African countries, such as Tunisia Ghana, Mozambique and Tanzania. Romania Both countries also give significant 60 (although still relative small) volumes China of aid to the populous countries of Pakistan India South Asia (Bangladesh, Pakistan 40 Bangladesh and most importantly India) where

around 45% of the world’s $1/day share of aid (%) Cumulative poor live. 20 Nigeria In contrast, France and the United Mozambique States give large amounts of aid 0 (both absolutely and relatively) to 0 20406080100 middle-income developing countries Cumulative share of population living under $1 per day (%) such as Egypt, Morocco and Russia (US) and Peru and Thailand (Japan) and less to the poorest (mostly sub- Figure 11.5 Aid Concentration Curves for the World Bank (IDA) Saharan African) countries and South (2001) Asia. France and Germany resemble 100

the Netherlands and the UK in giving China relatively large amounts of aid to poor 80 African countries but also resemble the Japan and the US, in giving Pakistan India Vietnam considerable amounts to relatively 60 prosperous middle income countries Nigeria (notably, Egypt, Jordan, Morocco and Bangladesh Russia). 40 Population Uganda

The distribution of aid from share of aid (%) Cumulative Population living under $1 per day multilateral agencies 20 Malawi The three main providers of Ethiopia concessional multilateral development 0 assistance – the European Union, 0 20406080100 World Bank and the UN System – Cumulative share of population living under $1 per day (%) distribute their aid in quite different ways. The aid provided by the World Bank through IDA appears relatively Figure 11.6 Aid Concentration Curves for the UN (2001) well targeted towards the poorest countries. In contrast, the European 100 Commission spends large amounts of its aid on relatively well-off middle Brazil 80 income countries (such as Brazil, Jordan South Africa, Turkey, Tunisia, and a number of countries in Eastern 60 India China Europe and the former Soviet bloc). Pakistan Indonesia The third largest multilateral donor, Nigeria the UN System, has a marginally 40 progressive distribution of aid. Bangladesh Population living under $1 per day Source: CPRC Working Paper No 35, Cumulative share of aid (%) Cumulative Tanzania Population Baulch 2003 20

Ethiopia

0 0 20406080100 Cumulative share of population living under $1 per day (%) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P311E-02.3D.3D Folio: 130 Black plate (130,1) Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 131 Black plate (131,1)

income or expenditure (DFID 2001:174– APPENDIX 186). Disability The outcome of the interaction between a person with an impairment and the environmental and attitudinal barriers that Glossary s/he faces (WHO’s International Classifica- tion of Functioning). of Discrimination Refers to the institutional, environmental and attitudinal factors that terms work to exclude certain people from activ- A ities, organisations and institutions. Displaced person: (see Internally Displaced People) Dollar-a-day ($US1/day) An absolute poverty line introduced by the World Bank in 1990 to estimate global poverty. The dollar amount is revised over time to keep pace with inflation and now stands at $1.08 in Absolute poverty A person living in absolute Common property resources Assets owned by 1996 prices. This is converted into local cur- poverty is not able to satisfy his or her mini- a group or society whose use is not re- rencies using purchasing power parity (PPP) mum requirements for food, clothing or stricted to a single individual. Communal exchange rates (DFID 2001:174–186). shelter. The dollar a day poverty line is ac- management is needed to make sustainable Economic growth An increase in a country’s cepted internationally as an absolute pov- use of common property (DFID 2001:174– total output. It may be measured by the an- erty line. (See relative poverty) (DFID 186). nual rate of increase in a country’s Gross 2001:174–186). Coping strategy How a household responds National Product (GNP) or Gross Domestic Adverse incorporation Where people are in- when faced with an unexpected event such Product (GDP) as adjusted for price cluded in social, political and economic in- as illness, drought or unemployment. Typi- changes. The increase in GNP, at constant stitutions and processes, but on extremely cal responses include taking children out of prices per head of population, indicates unfavourable terms. school, drawing on support from the ex- changes in the average living standards in Assets framework A framework which can be tended family or other households, or re- that country but says nothing about the dis- used to identify the poor and vulnerable. ducing expenditure on food and other tribution of the levels for different social The framework takes into account factors items. In addition, some households may groups around that average (DFID that cause vulnerability to poverty and con- migrate (DFID 2001:174–186). 2001:174–186). siders these in relation to poor people’s as- Covariate shock An unexpected event that af- Empowerment The process whereby people sets. It is part of the sustainable livelihoods fects all the members of a group. An exam- gain more power over the factors governing framework (DFID 2001:174–186). ple is a drought, which typically reduces the their social and economic progress. This Bonded labour People become bonded labour- agricultural output of all the households in may be achieved through: increasing the in- ers by taking or being tricked into taking a a village. (See idiosyncratic shock) (DFID comes and assets of the poor; interventions loan for as little as the cost of medicine for 2001:174–186). that aim to enhance confidence and self-re- a sick child. To repay the debt, they are Decentralisation The process of transferring spect; by developing collective organisation forced to work long hours, seven days a control over, and administration of, services and decision-making and by reforming po- week, 365 days a year. They receive basic from national to local level (DFID litical institutions to make them more inclu- food and shelter as ‘payment’ for their 2001:174–186). sive. Empowerment is one aim of setting up work, but may never pay off the loan, Dependency ratio The ratio of economically- participatory processes (DFID 2001:174– which can be passed down through several active household members to those who are 186). generations (Anti-Slavery International). economically dependant. Equivalence scales A ratio applied to the pov- Capabilities A term developed by Amartya Deprivation A lack of welfare, often under- erty line to reflect the fact that children cost Sen that refers to the means which enable stood in terms of material goods and re- less than adults do. Two households with people to function. The term is distinguishes sources but equally applicable to the same resources can have very different intrinsic and instrumental capabilities (in- psychological factors, relative to the local financial circumstances, depending upon the come, education, health, human rights, civil community or the wider society or nation size and composition of the household rights etc). Sen’s conceptualisation of pov- to which an individual, family or group be- (DFID 2001:174–186). erty as capability deprivation focuses on the longs (Gordon and Spicker, 1999:37). Exclusion The economic, political and cultur- failure of some basic capabilities to func- Depth A measure of the average distance of al processes that lead to the isolation of tion, for example, being adequately nour- poor individuals or households below the some groups in society, including ethnic mi- ished, leading a long and healthy life, being poverty line. The depth of poverty is also norities or the long-term unemployed. Dif- literate. (Gordon and Spicker, 1999: 22) known as the poverty gap (DFID ferent interpretations of this concept range Capability deprivation Poverty defined in rela- 2001:174–186). from notions of discrimination to under- tion to the failure to achieve basic capabil- Destitution Refers to the total, or virtually standing the social consequences of poverty. ities such as being adequately nourished, complete, absence of resources. Although Forced labour People illegally recruited by leading a healthy life or taking part in the indicative of extreme poverty it is not neces- governments, political parties or private in- life of the community. The emphasis on ca- sarily equivalent; a person may become des- dividuals, and forced to work – usually pabilities shifts focus away from money- titute immediately through fire or natural under threat of violence or other penalties based measures such as income or expendi- disaster, while someone in chronic or ex- (Anti-Slavery International). ture onto the kind of life the individual can treme poverty may have experienced long- Food energy method A method for deriving live (DFID 2001:174–186). term malnutrition and disadvantage (Gor- absolute poverty lines. It sets the poverty Chronic poverty Poverty experienced by indi- don and Spicker 1999:38). line by identifying the level of income or viduals and households for extended peri- Dimensions of poverty The individual and so- expenditure needed to obtain sufficient food ods of time or throughout their entire lives. cial characteristics of poverty such as lack to provide enough calories to meet the min- Also called ‘persistent poverty’. Chronic of access to health and education, power- imum energy requirement of an individual poverty must be distinguished from transi- lessness or lack of dignity. Such aspects of (DFID 2001:174–186). tory poverty or being non-poor. [For a full deprivation experienced by the individual Food insecurity A situation that exists when definition see Chapter One] or group are not captured by measures of people lack secure access to sufficient Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 132 Black plate (132,1)

132 . Appendix A

amounts of safe and nutritious food for nor- groups of equal size. The bottom 10 per there are few goods and services; and many mal growth and development and an active, cent is the first or bottom decile. If divided people cannot meet their basic needs healthy life. It may be caused by the un- into five groups the groups are called quin- (World Bank). availability of food, insufficient purchasing tiles (DFID 2001:180). Low birth weight Newborn infants who power or the inappropriate distribution or Income (or consumption) poverty Poverty de- weigh less than 2.5 kg at birth (FAO). inadequate use of food at the household fined with respect to a money-based poverty Marginalised people Those who are physically level. Food insecurity may be chronic, sea- line for income or expenditure. The distinc- or socially remote (see also exclusion). They sonal or transitory (FAO). tion is made between this and other con- are by-passed by most economic, political Geographic capital A combination of social, cepts that emphasise the many dimensions and social activity and likely to have very cultural, political, environmental and eco- of poverty (DFID 2001:180). precarious livelihoods (DFID 2001:181). nomic factors that are specific to a geo- Inclusive policies Policies which acknowledge Market failure A situation in which markets graphic area. that socially excluded, poor or vulnerable do not function properly. A common cause Gini coefficient An aggregate numerical meas- people are not a homogeneous group and of market failure is imperfect information. ure of income inequality ranging from 0 have a right to be included in poverty allevi- For instance, the difficulty of determining (perfect equality) to 1 (perfect inequality) ation and development work. which potential borrowers are creditworthy (FAO). Income distribution The allocation of national is given as a reason for badly functioning Global public goods Items that benefit every- income between persons or households; an rural credit markets and a rationale for the one: for example, international research, indicator of economic and social inequality high interest rates charged by money lend- environmental agreements or measures for where some people have more than others. ers (DFID 2001:181). conflict management and resolution (DFID (See gini co-efficient) (Gordon and Spicker Money-metric A reference to poverty from a 2001:179). 1999:71). strictly income perspective (DFID Gross Domestic Product (GDP) The total Income inequality See Income distribution 2001:181). value of all goods and services produced do- Income poverty Income is a key concept in al- Multidimensionality Multi-dimensional ap- mestically by a nation during a year. It is most all definitions and studies of poverty. proaches capture a fuller range of depriva- similar to Gross National Product (GNP), Classically, income has been defined as the tions that constitute poverty, and may give which is the value of output produced by a sum of consumption and change in net ‘voice’ to the poor and include non-mone- country’s labour and capital regardless of worth (wealth) in a period (Gordon and tary dimensions. whether it is in the country or not (DFID Spicker 1999:77). Internationally, the in- National poverty lines Poverty lines drawn by 2001:179). come poverty line is set at a dollar a day. national governments or national statistical Headcount index (see Poverty headcount) Indicator A numerical measure of quality of offices to measure poverty. It is not possible Household life-cycle The sequence of events life in a country. Indicators are used to illus- to make comparisons between countries (birth, death, marriage, moving together or trate progress of a country in meeting a using national poverty lines as each is calcu- away from other household members) range of economic, social, and environmen- lated on the basis of criteria specific to that which characterise the formation, growth tal goals. Since indicators represent data country (DFID 2001:182). and disappearance of a household. The that have been collected by a variety of Oblasts Administrative and territorial divi- household’s likelihood of being in poverty agencies using different collection methods, sions in some republics of the former Soviet is related to its position in the household there may be inconsistencies among them Union. life cycle (DFID 2001:179). (World Bank). Panel survey Quantitative longitudinal study. Human capital Factors such as knowledge, Indigence A person who is indigent is in need (See Longitudinal study). skills and health, which increase the pro- and lacks the means for subsistence. The Poverty correlates The characteristics that are ductivity of the individual (DFID United Nations Economic Commission for closely associated with being poor such as 2001:174–186). Latin America has referred to an indigence living in a rural area or having a large num- Human Development Index (HDI) An index line, which at half the value of the poverty ber of children. These can be used to target introduced by UNDP in 1990, which com- line is supposed to cover only basic nutri- public expenditure in the absence of de- bines the three measures of life expectancy, tional requirements (cited in Gordon and tailed information relating to every house- educational attainment (itself a composite Spicker 1999:81). (See extreme poverty, hold (DFID 2001:183). of literacy and school enrolment) and GDP destitution) Poverty dynamics Changes in individual or per head. The index theoretically ranges Infant Mortality Rate Statistical summary rate household poverty status over time. from 0 for the least developed to 7 for the based on the number of infant deaths occur- Poverty gap See Poverty depth. most (DFID 2001:179). ring during the same period of time, usually Poverty headcount Refers to the proportion of Human Poverty Index (HPI) A composite a calendar year, usually given in relation to individuals, households or families that falls index introduced by UNDP in 1997, which 1 000 live births occurring among the pop- under the poverty line. Divides the number focuses on those who do not achieve mini- ulation during the same year (UNECE). of people identified as poor by the total mum standards of health, education and liv- Internally displaced people IDPs are people number of people in the community. The ing conditions. This index contrasts with who are displaced but remain within the headcount ratio ranges from zero (nobody that of the HDI, which measures average border of their country of origin. Usually is poor) to one (everybody is poor) (Gordon achievements (DFID 2001:179). applied to people fleeing their homes be- and Spicker 1999:73). Kilocalorie (kcal) A unit of measurement of cause of an armed conflict, civil disturbance Poverty incidence See Incidence. energy: 1 kcal = 1 000 calories. In the Inter- or natural disaster (CRED). Poverty line Represents the level of income or national System of Units (ISU), the universal Intra-household allocation The way resources consumption necessary to meet a set of min- unit of energy is the joule (J). 1 kcal = are distributed between members of the imum requirements to feed oneself and 4.184 kilojoules (kJ) (FAO). household on the basis of their age, gender one’s family adequately and/or to meet Idiosyncratic shock An unexpected event that and role of the household member (DFID other basic requirements such as clothing, affects one household or individual. An ex- 2001:180). housing and healthcare. Those with in- ample of an idiosyncratic shock would be Longitudinal study A study that observes the comes or expenditure equal to or above the the death of the breadwinner (DFID same group of individuals, households or line are not poor. While what the minimum 2001:179). villages over time, also known as panel should be has an important subjective ele- Impairment: An individual’s condition – phys- study (DFID 2001:181). ment, poverty lines are typically anchored ical, sensory, intellectual or behavioural. Low-income country A country having an to minimum nutritional requirements plus a Incidence The percentage of people living Gross National Income (GNI) per capita modest allowance for non-food needs. (see below the poverty line (DFID 2001:179). equivalent to $755 or less in 1999. There Chapter One). Income deciles and quintiles The population is are currently about 64 low-income coun- Poverty severity A static concept, capturing ranked by income and divided into ten tries where the standard of living is lower, the fact that the poor are not equally poor Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 133 Black plate (133,1)

. 133

to the same level. It is the average value of experience. This concept contrasts with that remain multi-dimensionally deprived and the square of depth of poverty for each in- of uncertainty, in which the likelihood is poor over long periods of time dividual. Poorest people contribute rela- unknown. An individual or household may Stunting Low height for age, reflecting a sus- tively more to the index. Also called Foster assess that the likelihood of a bad event, tained past episode or episodes of undernu- Greer Thorbeke (or P2) (ADB). such as drought, occurring is high enough trition (FAO). Poverty Reduction Strategy Papers (PRSP) A to alter their livelihood strategy (DFID Targeting The process by which expenditure national strategy for poverty reduction. All 2001:185). is directed to specific groups of the popula- countries that are eligible for World Bank Scheduled Castes In India, a collection of tion defined as poor or disadvantaged, in concessional lending or for debt relief under castes formerly known as ‘untouchables’ order to increase the efficiency of the use of the Heavily Indebted Poor Countries (dalits) that have been ‘scheduled’ for posi- resources (DFID 2001:186). (HIPC) Initiative are producing PRSs. The tive discrimination in education and Trafficking Involves the transport and/or PRS is intended to be the basis for all donor employment. trade of humans, usually women or chil- support, including the IMF and World Scheduled Tribes In India, identified on the dren, for economic gain using force or de- Bank (DFID 2001:184). basis of certain criteria including distinctive ception. (Anti-Slavery International). Poverty spell The period of time spent in pov- culture and pre-agricultural modes of Transitional Countries Those countries whose erty. In longitudinal studies, often the dura- production. economies used to be centrally planned by tion of poverty and the causes of the Selectivity The allocation of development as- the government but are now changing – or transition out of poverty are analysed sistance prioritising those with good anti- ‘transitioning’ – to base their economies on (DFID 2001:184). poverty policies (DFID 2001:185). the market (World Bank). Poverty trends How aggregate poverty levels Severe poverty Persons who fall below a Transitory poverty Short term poverty. Pov- change over time Public Goods A good that lower poverty line. For example, in 1993 erty experienced as the result of a tempo- is provided for users collectively, see Global the World Bank defined an upper poverty rary fall in income or expenditure although Public Good. line of US$ 1 income per day and extreme over a longer period the household resour- Purchasing Power Parity (PPP): A method of poverty as persons living on less than US$ ces are on average sufficient to keep the measuring the relative purchasing power of 0.75 income per day (both in 1985 prices). household above the poverty line (DFID different countries’ currencies over the same These measures are converted into local 2001:186). types of goods and services. Because goods currencies using purchasing power parity Undernourishment Food intake that is contin- and services may cost more in one country (PPP) exchange rates. Other definitions of uously insufficient to meet dietary energy than in another, PPP allows us to make this concept have identified minimum sub- requirements (FAO). more accurate comparisons of standards of sistence requirements, the denial of basic Ultra-poverty This is another term for ex- living across countries. PPP estimates use human rights or the experience of exclusion treme poverty. It is sometimes specifically price comparisons of comparable items but (DFID 2001:174–186). used to refer to those who spend more than since not all items can be matched exactly Sex ratio The relative proportion of males and 80 per cent of their income on food but across countries and time, the estimates are females in a given population, usually ex- obtain less than 80 per cent of their food not always ‘robust’ (World Bank). pressed as the number of males per 100 energy needs. The low food intake of this Refugees Those who are forced to cross inter- females. particular group will affect their productiv- national borders because of conflict or po- Slavery Traditional or ‘chattel’ slavery in- ity and ability to get out of poverty (DFID litical instability. volves the buying and selling of people. 2001:186). Relative poverty Poverty defined in relation to They are often abducted from their homes, Vulnerability Relates to risk. People are vul- the social norms and standard of living in a inherited or given as gifts (Anti-Slavery nerable to poverty when they are more at particular society. It can therefore include International). risk than others (Gordon and Spicker 1999: the individual’s ability to take part in activ- Social exclusion See exclusion. 141–2). While income poverty may be re- ities that society values even if they are not Social protection Policies and programmes duced by borrowing; debt may make the necessary for survival. Relative poverty can which aim to prevent and mitigate the poor more vulnerable (DFID 2001:186). also refer to the nature of the overall distri- shocks that create and maintain chronic Wasting Low weight for height, generally the bution of resources (DFID 2001:184). poverty, and provide recovery assistance by result of weight loss associated with a re- Rights-based approach An approach based on protecting incomes and building the assets cent period of starvation or disease (FAO). understanding of the links between develop- of the poor. Examples include pensions, Waves the occasions on which a survey is con- ment and civil, political, economic, social and food for education programmes. ducted to make up panel data e.g. a three and cultural rights (DFID 2001:185). Spatial poverty trap Geographical areas which wave panel dataset has conducted compara- Risk Understanding of the likelihood of events remain disadvantaged, and whose people ble surveys at three different times on the occurring, for example, on the basis of past same individuals or households.

Sources DFID (2001) ‘Poverty: Bridging the Gap – Guidance notes’ The Centre for Research on the Epidemiology of Disasters DFID Issues paper: Department for International (CRED): Development, London. http://www.cred.be/emdat/intro.htm

Gordon, D. and P. Spicker (eds.), (1999), ‘The International FAO: Glossary on Poverty’, ZED books: London/New York. http://www.fao.org/docrep/003/y1500e/y1500e06.htm#P0_2

Anti-Slavery International: UNECE Gender statistics glossary: http://www.antislavery.org/homepage/antislavery/modern.htm) http://www.unece.org/stats/gender/web/glossary.htm

Chronic Poverty Research Centre: World Bank: http://www.chronicpoverty.org; http://www.worldbank.org/html/schools/glossary.htm Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 134 Black plate (134,1)

APPENDIX B References

Abazov, R., (1999). ‘Policy of economic transition in Kyrgyzstan’, Central Asian CPRC. http://www.chronicpoverty.org/pdfs/Aid%20for%20the%20Poorest- Survey, Vol. 18, No. 2, 197–223. %20WP35.pdf ACC/SCN (United Nations Administrative Committee on Coordination’s sub Baulch, B. and E. Masset (2003) ‘Do Monetary and Non-Monetary Indicators Committee on Nutrition) & IFPRI (International Food Policy Research Insti- Tell the Same Story About Chronic Poverty? A Study of Vietnam in the 1990s’, tute) (2000). 4th Report On The World Nutrition Situation: Nutrition World Development, Vol. 31, No. 3, pp. 441–454 Throughout The Life Cycle, Geneva: ACC/SCN, IFPRI. Baulch, B., Chuyen, T. T. K., Haughton, D., and Haughton, J. (2002). ‘Ethnic ADB (Asian Development Bank) (2003). Millennium Development Goals in the Minority Development in Vietnam: A Socioeconomic Perspective’. World Bank Pacific: Relevance and Progress. Manila, Philippines: ADB. http://www.ad- Research Working Paper 2836. Washington DC: World Bank. http://econ.- b.org/documents/books/MDG_Pacific/mdg.pdf worldbank.org/files/15032_wps2836.pdf ADB (2002). Priorities of the Poor in Papua New Guinea, Manila, Philippines: BBC (British Broadcasting Coorporation) (2003) Q&A: Crisis in the Solomons Asian Development Bank http://www.adb.org/Documents/Reports/Priorities_- http://news.bbc.co.uk/1/hi/world/asia-pacific/3023106.stm Poor/PNG/priorities_poor_png.pdf Bebbington, A. (2003) ‘NGO Geographies, Livelihood Geographies and the ‘non- Adelman, L., Middleton, S., and Ashworth, K. (2003). Britain’s Poorest Children: viable’ Poor in the Andes’, Conference Paper presented at ‘Staying Poor: Severe and persistent poverty and social exclusion. London: Save the Children. Chronic Poverty and Development Policy’, Manchester, 7–9 April 2003. http:// http://www.savethechildren.org.uk/temp/scuk/cache/cmsattach/ idpm.man.ac.uk/cprc/Conference/conferencepapers.htm 949_poorestchildren.pdf Begum, S. and Sen, B. (forthcoming). ‘Unsustainable Livelihoods, Health Shocks Afsaw, A. (2003) How Poverty affects Health Status and Health Care Demand and Urban Chronic Poverty: Rickshaw-pullers as a Case Study’. Forthcoming Behaviour of Households – The Case of Rural Ethiopia. Conference Paper Chronic Poverty Research Centre Working Paper. Manchester: Institute of De- presented at ‘Staying Poor: Chronic Poverty and Development Policy’, Man- velopment Policy and Management/CPRC. chester, April 2003. http://idpm.man.ac.uk/cprc/Conference/conferencepapers/ Bezemer, D and D. Davis, (2003). Rural Livelihoods in Romania. Mimeo, Impe- Abay%20Asfaw%2007.03.03.pdf rial College and the National Resources Institute, University of Greenwich. Alexandratos, N. (ed.) (1995). World Agriculture Towards 2010, An FAO Study. Bhide, S. and A. K. Mehta, (forthcoming) ‘Chronic Poverty in India: Issues and Chichester, UK: John Wiley and Sons, and Rome: FAO. Findings from Panel Data’, Journal of Human Development, Volume 5 Num- Aliber, M. (2003) ‘Chronic Poverty in South Africa : Incidence, Causes and Poli- ber 2, July 2004 cies’ World Development, Vol 31, No 3, pp. 473–490 Bird, K. & A. Shepherd, 2003, ‘Livelihoods and Chronic Poverty in Semi-Arid Aliber, M. (2001) ‘Study of the incidence and nature of chronic poverty and Zimbabwe’, World Development Vol31, No 3, pp: 591–610 development policy in South Africa: an overview’, CPRC Working Paper 3. Bird, K. and Shinyekwa, I. (2003) ‘Multiple shocks and downward mobility:learn- Manchester: IDPM, University of Manchester. ing from life histories of rural Ugandans’. Chronic Poverty Research Centre Amis, P (2003) ‘Chronic Poverty in India: Lessons from recent research, Confer- Working Paper 36. Manchester: Institute of Development Policy and Manage- ence Paper presented at ‘Staying Poor: Chronic Poverty and Development Poli- ment/CPRC. http://www.chronicpoverty.org/pdfs/life%20histories%20WP36. cy’, Manchester, April 2003. http://idpm.man.ac.uk/cprc/Conference/ pdf conferencepapers.htm Bird, K., Booth, D. and Pratt, N. (2003) The Contribution of Politics, Policy Fail- Anderson, E. and Morrissey, O. (2003). A statistical approach to identifying poor ures, and Bad Governance to the Food Security Crisis in Southern Africa. performers. Background paper for ODI Study for DFID on Poorly Performing Forum for Food Security in Southern Africa. www.odi.org.uk/food-security- Countries. forum Anti-Slavery International (2002) Forced labour in Mauritania. Submission to the Bird, K, D. Hulme, K. Moore and A. Shepherd, (2002) ‘Chronic Poverty and United Nations Commission on Human Rights, Sub-Commission on the Pro- Remote Rural Areas’, Chronic Poverty Research Centre Working Paper 13. motion and Protection of Human Rights, Working Group on Contemporary Manchester: Institute of Development Policy and Management/CPRC. Forms of Slavery (27th Session, Geneva, 27–31 May 2002). http://www.antisla- Black, R., Sabates-Wheeler, R., Skeldon, R, Waddington, C., & A. Winters very.org/archive/submission/submission2002-mauritania.htm (2003) ‘Mapping Study of Migration Issues’ Sussex Centre for Migration Re- Appleton, S. (1996) ‘Women-headed households and household welfare: An em- search working paper, Sussex University: Brighton pirical deconstruction for Uganda’, World Development, Vol. 24, No. 12, pp. Bolt, V. J. (2003) ‘Lessons from Low Potential Areas: the Causes of and Possible 1811–1827. Responses to Chronic Poverty in Remote Rural Areas’ DRAFT Chronic Pov- Ashley, C. and Maxwell, S. (2001) ‘Rethinking Rural Development’, Develop- erty Research Centre Working. Manchester: Institute of Development Policy ment Policy Review 19 (4): 395–425. and Management/CPRC. Atkinson, A. B. and Brandolini, A. (2001). ‘Promise and Pitfalls in the Use of Bourguignon, F. (1999). Absolute poverty, relative deprivation and social exclu- ‘Secondary’ Data-Sets: Income Inequality in OECD Countries as a Case Study’. sion. Villa Borsig Workshop Series, Policy Dialogue on Inclusion, Justice, and Journal of Economic Literature, Vol. 39, No. 3, pp. 771–799. Poverty Reduction: Berlin. http://www.dse.de/ef/poverty/bourgign.htm Ayres, R. and Torrijos S. M. (2003). ‘Education, poverty and sustainable live- Bourne R. (2003) Invisible lives – Undercounted, underrepresented and under- lihoods in Tamil Nadu: inequalities, opportunities and constraints’. Review of neath: the socioeconomic plight of Indigenous peoples in the Commonwealth, Political Economy, Vol. 15, No. 2, pp. 211–229. Commonwealth Policy Studies Unit, University of London, DRAFT REPORT Baker, W. G. (2001) ‘Uganda: The Marginalization of Minorities’, Minority p.19 Rights International Report, MRG: London, UK. http://www.minorityright- Bradbury, B., Jenkins, S. P., and Micklewright, J. (2000). ‘Child poverty dynamics s.org/admin/Download/Pdf/UgandaReport.pdf in seven nations.’ Working Paper 78. Florence: UNICEF Innocenti Research Banfield, E. C. (1968) The Unheavenly City: The Nature And Future Of Our Centre. http://www.unicef-icdc.org/publications/pdf/iwp78.pdf Urban Crisis. Boston: Little, Brown. Braithwaite, J (2001) Hungary: Long-Term Poverty, Social Protection, and the Banjeri, A. (1999) ‘Moldova: Poverty Assessment’. World Bank Country Study Labor Market. Volume I: Main Report. Report no. 20645-HU. Poverty Reduc- 19926. Washington, D.C.: World Bank tion and Economic Management Unit, Europe and Central Asia Region. Wash- Barrett, C. B., Blume, L. E., McPeak, J. G., Minten, B., Murithi, F., Okumu, B. ington D.C: World Bank N., Pell, A., Place, F., Randrianarisoa, J. C., and Rasambainarivo, J. (2002). Brockerhoff, M. and Hewett, P. (2000). ‘Inequality of child mortality among ‘Poverty traps and resource degradation’. BASIS Brief: Broadening Markets ethnic groups in sub-Saharan Africa’, Bulletin Of The World Health Organiza- and Strengthening Input Market Systems, No 6, Madison, WI: BASIS. http:// tion, Vol.78, No.1, 30–41. www.basis.wisc.edu/live/basbrief06.pdf Bruton, H, J. et al (1992) The Political Economy of Poverty, Equity, and Growth: Baulch, B. (2003) Aid for the Poorest? The distribution and maldistribution of Sri Lanka and Malaysia. New York, OUP international development assistance. Chronic Poverty Research Centre Work- Burnside, C. and Dollar, D.(2000) ‘Aid, Policies, and Growth’. American Eco- ing Paper 35. Manchester: Institute of Development Policy and Management/ nomic Review, Vol. 90, no. 4, pp. 847–68. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 135 Black plate (135,1)

. 135

Butalia, U. (1998) ‘The women’s movement in India: action and reflection.’ Third DFID, (2002) Chars Programme, Prepared by the Department for International World Resurgence No. 94, first printed in Communique (Nos. 42–43, July-Aug Development, for the Government of Bangladesh. http://www.dfid.gov.uk/ 1997). http://www.twnside.org.sg/title/india1-cn.htm Pubs/files/chars_livelihoods_prog.pdf Cammack, D., Chulu, O., Khaila, S. and Ng’ong’ola, D. (2003) Malawi Food Dolan, C (2002) ‘Gender and witchcraft in agrarian transition: the case of Kenyan Security Issues Paper. Forum for Food Security in Southern Africa. www.o- horticulture’, Development and Change, Vol. 33, No. 4, pp. 659–681. di.org.uk/food-security-forum Dollar, D. and Kraay, A. (2001) Growth is good for the poor. Mimeo. World CastaEˆ eda, T., and E. Aldaz-Carroll 1999 The Intergenerational Transmission of Bank: Washington DC. http://www.worldbank.org/research/growth/pdfiles/ Poverty: Some Causes and Policy Implications. Unpublished background paper growthgoodforpoor.pdf for Breaking the Poverty Cycle: Investing in Early Childhood seminar at Inter- Durable Solutions, (2003). Progress Report No. 11, The Bulletin of the United American Development Bank Annual Meeting March 14, 1999, Paris. Avail- Nations Inter-Agency IDP Working Group, 3 June 2003. http://www.peaceins- able: http://www.iadb.org/sds/doc/1258eng.pdf rilanka.org/insidepages/RRR/UNHCR/UNHCR1 1.asp Chant, S. (2003) ‘Female Household Headship and the Feminisation of Poverty: Du Toit A. (2003) Hunger in the Valley of Fruitfulness: Globalization, Poverty Facts, Fictions, and Forward Strategies.’ New Working Paper Series Issue 9. Dynamics and ‘Social Exclusion’ in the Ceres, South Africa. Conference Paper London: London School of Economics Gender Institute presented at ‘Staying Poor: Chronic Poverty and Development Policy’, Man- Chambers, R. (1983) Rural Development: Putting the Last First. London, chester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/Conference/conferencepa- Longman. pers/Du%20Toit%20Andries%20%20- Charveriat, C. (2001) Bitter coffee: how the poor are paying for the slump in %20Chronic%20Poverty%20in%20Ceres,%20South%20Africa.PDF coffee prices. Oxford: Oxfam. http://www.oxfam.org.uk/what_we_do/issues/ ECLAC, (2002) Statistical Yearbook for Latin America and the Caribbean,pp trade/downloads/bitter_coffee.pdf 64–5, http://www.eclac.cl/publicaciones/Estadisticas/0/LCG2190PB/p1_1.pdf Chase-Lansdale, P.L., & Brooks-Gunn, J. (1995). ‘Introduction’. In Chase-Lans- Elwan, A. (1999) ‘Poverty and disability: a review of the literature.’ Background dale, P.L., & Brooks-Gunn, J. (Eds), Escape from poverty. What makes a dif- Paper for the World Development Report 2000/2001. Washington, D.C.: ference for children? Cambridge: Cambridge University Press. World Bank. Cleaver, F. (2003) ‘The Inequality of Social Capital: Agency, Association and the Erb, S. & Harriss-White, B. (2001). The economic impact and developmental Reproduction of Chronic Poverty’ Conference Paper presented at ‘Staying implications of disability and incapacity in adulthood – a village study from S. Poor: Chronic Poverty and Development Policy’, Manchester, 7–9 April 2003. India. Paper to the workshop ‘Welfare, demography and development’, Sep- http://idpm.man.ac.uk/cprc/Conference/conferencepapers/ tember 11–12, 2001, Downing College, Cambridge. franciscleaverpaper.pdf Esping-Andersen, G. (1999). Social Foundations of Postindustrial Economies. Ox- Conticini, A. (2003) ‘We are the kings’: the children of Dhaka’s streets. PhD in ford, Oxford University Press. progress, IDPM: Manchester Eversmann, E. (1999) School attendance in the Kyrgyz Republic – Report for Cook, S. (2003) ‘From iron rice bowl to safety net.’ China Development Brief, UNICEF Bishkek. Bishkek, UNICEF. vol II, n.4, dec 1999-march 2000. Fabricant, S.J., Kamara, C.W., and Mills, A. (1999) Why the poor pay more: Cook, S and White, G., 1997. China: a poverty profile’. Briefing paper prepared Household curative expenditures in rural Sierra Leone. International Journal of for DfID China Health Planning and Management 14:179–199. Corcoran, M. (1995). Rags to rags: poverty and mobility in the United States. Falkingham, J., 2003, Inequality and Poverty in the CIS-7, 1989–2002, paper Annual Review of Sociology, Vol. 21, pp. 237–267. presented at CIS-7 Conference, Lucerne Counterpart Consortium/ World Bank, (1999). Kyrgyz Republic: Consultations Falkingham, J. 1998, ‘Poverty in Central Asia’ in UNDP, 1998, Central Asia with the Poor, WB/ CC, Bishkek. 2010, Prospects for Human Development, UNDP, Bratislava. Crook, R. (no date) Civil war in Cote d’Ivoire; behind the headlines, unpublished Falkingham, J. and Ibragimova (forthcoming) Trends in childhood poverty in paper, Sussex: Institute of Development Studies. http://www.ids.ac.uk/ids/gov- Kyrgyzstan. CHIP Report, CHIP, London. ern/pdfs/cdicivilwarlongv.pdf Fan, S. and Hazell, P. (2000) Promoting Sustainable Development in Less-Fa- Dadekar, V. and Rath, N. (1971), Poverty in India, Pune: Indian School of Polit- vored Areas – Returns To Public Investment: Evidence From India And China. ical Economy Washington, D.C.: IFPRI, 2000 (2020 Vision for Food, Agriculture and the Dagnino, E (2003). ‘Citizenship in Latin America: An Introduction’, Latin Amer- Environment Focus 4, Brief 5). http://www.ifpri.org/2020/focus/focus04/ ican Perspectives. Vol. 30, No. 2, pp: 3–18. focus04.pdf Dalgaard, C. J., Hansen, H. and Tarp, F. (2002) ‘On the empirics of foreign aid FAO (2003) The State of Food Insecurity in the World. Rome: Food and Agricul- and growth.’ CREDIT Research Paper No.02/08, University of Nottingham. ture Organisation. http://www.nottingham.ac.uk/economics/credit/research/papers/CP.02.08.pdf FAO/GIEWS (Food and Agriculture Association/Global Information and Early Daru, P. and Churchill, C. (2003) The Prevention of Debt Bondage with Micro- Warning System. (2003). Foodcrops and shortages, no. 2, June 2003. http:// finance and Related Services: Preliminary Lessons. Conference Paper presented www.fao.org/docrep/005/J0269e/J0269e00.htm at ‘Staying Poor: Chronic Poverty and Development Policy’, Manchester, April Farrington, J., N.C. Saxena, T. Barton and R. Nayak (2003) Post offices, pensions 2003. http://idpm.man.ac.uk/cprc/Conference/conferencepapers/Daru Churchill and computers: new opportunities for combining growth and social protection 02 04.03.03.pdf in weakly integrated rural areas London: Overseas Development Institute, Nat- Davis and Stampini, (2002). ‘Pathways towards prosperity in rural Nicaragua: or ural Resource Perspectives 87 why some households drop in and out of poverty, and some policy suggestions Filmer and Pritchett (1998) ‘Estimating wealth effects without income or expendi- to keep them out’, Agriculture and Economic Development Analysis Division ture data – or tears: Educational enrolment in India,’ World Bank Policy Re- Working paper, FAO: Rome. search Working Paper No. 1994. Washington DC: Development Economics Deaton, A., (2001). ‘Counting the world’s poor: problems and possible solutions’, Research Group (DECRG), World Bank. World Bank Research Observer, Vol. 16, No. 2, pp: 125–147. Foster, M. and A. Keith (2003) The Case for Increased Aid. London: Department De Haan, A. and Dubey, A. (2003) Extreme deprivation in remote areas in India: for International Development, Draft final report. social exclusion as explanatory concept. Conference Paper presented at ‘Staying Gaiha, R. (2003a) Are Millennium Goals of Poverty Reduction Useful? Oxford Poor: Chronic Poverty and Development Policy’, Manchester, 7–9 April 2003. Development Studies, Vol. 31, No. 1, pp. 59–84. http://idpm.man.ac.uk/cprc/Conference/conferencepapers/Dehaan%20Ar- Gaiha, R. (2003b) Anthropometric Failure and Persistence of Poverty in Rural jan%20-dubey%2007.03.03.pdf India Conference Paper presented at ‘Staying Poor: Chronic Poverty a nd De- De Haan, A. and Rogaly, B. (2002) ‘Introduction: Migrant Workers and Their velopment Policy’, Manchester, April 2003. Role in Rural Change’. Journal of Development Studies, Vol. 38, No. 5, pp. 1– Gaiha, R. (1992). On the Chronically poor in rural India. Journal of International 14. Development, Vol. 43, No. 3, pp. 273–289. De Haan and Lipton (1998) ‘Poverty in Emerging Asia: Progress, Setbacks, and Gaiha, R. (1989), ‘Are the chronically poor also the poorest in rural India?’ Log-jams.’ Asian Development Review, vol. 16, no. 2, pp. 135–176. http:// Development and change, Vol.20, No.2, pp.295–322 www.adb.org/Documents/Periodicals/ADR/ADR_Vol16_2.pdf Gaiha, R. and Imai, K. (2003) Vulnerability, Shocks and Persistence of Poverty – Deininger, K. and Okidi. J. (2003) ‘Growth and poverty reduction in Uganda, Estimates for Semi-Arid Rural South India. Conference Paper presented at 1992–2000’ Development Policy Review, Volume 21 No 4, July, pp 481–509 ‘Staying Poor: Chronic Poverty a nd Development Policy’, Manchester, April Devereux, S. (2002) State of disaster – causes, consequences & policy lessons 2003. http://idpm.man.ac.uk/cprc/Conference/conferencepapers/ from Malawi. Report commissioned by ActionAid Malawi. http://www.actio- Gaiha%20Raghav_Imai_REVISED2.pdf naid.org/newsandmedia/the_malawi_famine_of_2002.pdf Gaiha R., and A.B. Deolalikar (1993) Persistent expected and innate poverty – Devereux, S. (2002) ‘Can social safety nets reduce chronic poverty?’ Development estimates for semi- arid rural South India, 1975–1984, Cambridge Journal of Policy Review, Vol. 20, No. 5, pp. 657–675. Economics Viol 17, No 4. pp 409–421 Devereux, S. (2001). ‘Social Protection for the poor?’ IDS Working Paper 142, Genicot, G. (2002) ‘Bonded labor and serfdom: a paradox of voluntary choice’, Sussex: Institute of Development Studies Journal of Development Economics, Vol. 67, No. 1, pp.101–127. Devereux and Sharp (2003) Is Poverty Really Falling in Rural Ethiopia? Confer- Global IDP Project & Norwegian Refugee Council. (2002) Internally-Displaced ence Paper presented at ‘Staying Poor: Chronic Poverty and Development Poli- People: A Global Survey (2nd ed.) Earthscan Publications Ltd. cy’, Manchester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/Conference/ Gomanee, K., Girma, S. and Morrissey, O. (2002) ‘Aid and Growth in sub-Sahar- conferencepapers.htm an Africa: Accounting for Transmission Mechanisms.’ CREDIT Research De Vreyer and Mesple-Somps (2002) Consumption growth, and spatial poverty Paper 02/05. http://www.nottingham.ac.uk/economics/credit/research/papers/ traps: a theoretical and econometric analysis of the effect of social services and CP.02.05.pdf community infrastructure on living standards in Peru, Document de travail Gonzales-Parra, C. (2001) ‘Indigenous peoples and Mega-Projects’ in van Genugt- DIAL no DT/2002/15, 40 p. en, W. and Perez-Bustillo, C. (eds) The Poverty of Rights London: Zed Books Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 136 Black plate (136,1)

136 . Appendix B

Goodhand, J. (2003) Enduring Disorder and Persistent Poverty: A Review of the ILO (International Labour Organisation) (2001) Stopping Forced Labour. http:// Linkages Between War and Chronic Poverty. World development, vol. 31, no. www.ilo.org/public/english/standards/decl/publ/reports/report2.htm 3, pp. 629–646. IOM, (International Organization for Migration) (2001), Deceived Migrants Goodhand, J., (2001). ‘Violent Conflict, Poverty and Chronic Poverty’, Chronic from Tajikistan. A Study of Trafficking in Women and Children, IOM, Poverty Research Centre Working Paper 6. Manchester: Institute of Develop- Dushanbe; ment Policy and Management/CPRC. IRIN News.org, 2003a, KYRGYZSTAN: Early marriages on the rise, say experts, Gordon, D., S. Nandy, C. Pantazis, S. Pemberton, and P. Townsend, 2003. ‘Child 1 December 2003, http://www.irinnews.org/report.asp?ReportID=38172&- Poverty in the Developing World’, Studies in Poverty, Inequality and Social SelectRegion=Central_Asia&SelectCountry=KYRGYZSTAN Exclusion, Bristol: Policy Press. IRIN News.org, 2003b, KYRGYZSTAN: Focus on abandoned children in south, Goudge, J. and Govender, V. (2000) A Review of Experience Concerning House- 27 November 2003, http://www.irinnews.org/report.asp?ReportID=38094&- hold Ability to Cope with the Resource Demands of Ill Health and Health SelectRegion=Central_Asia&SelectCountry=KYRGYZSTAN Care Utilisation. EquiNet Policy Series No3. EQUINET, Centre for Health Jalan, J. and Ravallion, M. (1998) Transient poverty in postreform rural China. Policy, Wits University and Health Economics Unit, University of Cape Town. Journal of Comparative Studies, vol 26, 338–357. Government of Tamil Nadu (2003) Social Welfare and Nutritious Meal Pro- Jalan J and M. Ravallion (2002) ‘Geographic Poverty Traps? A Micro Model Of gramme Department, Policy Note, 2003–4. http://www.tn.gov.in/policynotes/ Consumption Growth In Rural China’. Journal of Applied Econometrics, Vol. sw2003–04-main.htm 17, No. 4, pp. 329–346 Government of Uganda (2002) ‘Infant Mortality in Uganda 1995–2000 Why the James Commission. (2000). Ending malnutrition by 2020: an agenda for change non-improvement?’ Discussion Paper 6, Ministry of Finance, Planning and Eco- in the millennium. http://www.unsystem.org/scn/Publications/UN_Report.PDF nomic Development: Kampala, Uganda. Jenkins, S. P., and Rigg, J. (2001). The dynamics of poverty in Britain. Colchester: Gray, J. (2002) False Dawn: The Delusions of Global Capitalism. London/New Institute for Social and Economic Research (ISER), University of Essex; Depart- York, Granta Books. ment for Work and Pensions. Guhan, S. (1993) ‘Social Security for the Poor in the Unorganised Sector: a Fea- Joint UNAIDS/Parliamentary Forum on AIDS (2003) Press Release www.unaid- sible Blueprint for India.’ In Parikh, K. S. and Sudrashan, R. (eds.) Human s.org/whatsnew/press/ eng/IndiaParlForum250703_en.html Development and Structural Adjustment, London, Macmillan. Kabeer, N. (2000) ‘Inter-generational contracts, demographic transitions and the Haddad, L. and Ahmed, A. (2003). ‘Chronic and Transitory Poverty: Evidence ‘quantity-quality’ trade-off: parents, children and investing in the future’, Jour- from Egypt, 1997–99’. World Development, Vol. 31, No. 1, pp: 71–85. nal of International Development, Vol. 12, No. 4, pp. 463–482. HAI (HelpAge International) (2001) Training manual on Ageing in Africa. Lon- Kala, S. and A. K. Mehta (2002) ‘Hunger and Starvation’, The Alternative Eco- don: HelpAge International. nomic Survey 2001–2002, Rainbow Publishers. HAI (2000) Older people in disasters and humanitarian crises: Guidelines for best Kaye, M. (2001) Forced labour in the 21st century. Anti-Slavery International/ practice. London: HelpAge International. http://www.helpage.org/images/pdfs/ International Confederation of Free Trade Unions. bpg.pdf Kedir, A and McKay, A. (2003) Chronic Poverty in Urban Ethiopia: Panel Data Hansen, H. and Tarp, F. (2001). ‘Aid and Growth Regressions.’ Journal of Devel- Evidence. Conference paper presented at CPRC Conference, Staying Poor: opment Economics, Vol. 64, No. 2, pp. 547–570. Chronic Poverty and Development Policy, IDPM, University of Manchester, 7– Harper, C., Marcus, R. and Moore, K. (2003) ‘Enduring poverty and the condi- 9 April, 2003 http://idpm.man.ac.uk/cprc/Conference/conferencepapers/ tions of childhood: lifecourse and intergenerational poverty transmissions’, kedir&mckay.pdf World Development, Vol. 31, No. 3, pp. 535–554. Killick, T. (2004) ‘Politics, evidence and the new aid agenda’ Development Policy Harriss-White, B. (2003) India working: essays on society and economy. Cam- Review, Volume 22, No 1 bridge: Cambridge University Press. Klasen, S. and Wink, C. (2003) ‘‘Missing Women’: Revisiting the Debate’, Femi- Harriss-White, B. (2003). The Political Economy of Disability and Development, nist Economics, Vol. 9, No. 2–3, pp. 263–299. with special reference to rural India. Conference Paper presented at ‘Staying Korf B., and K. T. Silva (2003). Poverty, Ethnicity and Conflict in Sri-Lanka, Poor: Chronic Poverty and Development Policy’, Manchester, April 2003. Conference Paper presented at ‘Staying Poor: Chronic Poverty and Develop- http://idpm.man.ac.uk/cprc/Conference/conferencepapers/HarrissWhite.pdf ment Policy’, Manchester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/Confer- Helwege, A. (1995). ‘Poverty in Latin America: back to the abyss’. Journal of ence/conferencepapers.htm Interamerican Studies and World Affairs, Vol 37, pp. 99–123. Kothari, U. (2003) ‘Introduction (to ‘Staying Put, Staying Poor’ Policy Arena)’, Henriques, R. 2001. Desigualdade Racial no Brasil: Evolu±o das Condi±es de Journal of International Development, Vol. 15, No. 5, pp. 607–609. Vida na D cada de 90. Rio de Janeiro: Instituto de Pesquisas de Estatisticas Kothari, U. (2002). ‘Migration and Chronic Poverty.’ Chronic Poverty Research Aplicadas (IPEA) Texto para Disscuss±o No 807. http://www.ipea.gov.br/pub/ Centre Working Paper 16. Manchester: Institute of Development Policy and td/2001/td_0807.pdf Management/CPRC. http://www.chronicpoverty.org/pdfs/MigrationJun-02.pdf Herrera, J and Roubaud, F 2003. Urban Poverty Dynamics in Peru and Madagas- Krishna, A (2003) Escaping Poverty and Becoming Poor: Who Gains? Who car 1997–1999: A Panel Data Analysis. Conference Paper presented at ‘Staying Loses and Why? Accounting for Stability and Change in 35 North Indian Poor: Chronic Poverty and Development Policy’, Manchester, 7–9 April 2003. Villages, Conference Paper presented at ‘Staying Poor: Chronic Poverty and http://idpm.man.ac.uk/cprc/Conference/conferencepapers/Herrera%20Jav- Development Policy’, Manchester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/ ier%20-Roubaud-ENGLISH%20version27.02.03.pdf Conference/conferencepapers.htm Herzfeld, B. (2002) ‘Slavery and gender: women’s double exploitation’, Gender Kuehnast , K. and N. Dudwick (2002) ‘Better 100 Friends than 100 Roubles? and Development, Vol. 10, No. 1, pp. 50–55. Social Networks in the Kyrgz Republic’ World Bank Economic Forum,No2. Hickey, S. (2003) ‘The Politics of Staying Poor in Uganda’, Chronic Poverty pp. 51–88 Research Centre Working Paper 37. Manchester: Institute of Development Pol- Kumar, A. (2003) ‘Political Sociology of Poverty: A Note on State, Caste-Class icy and Management/CPRC. and Poverty in India’, DRAFT CPRC working paper, Manchester: Institute of Hoogeveen, H. (2003) Census-based welfare estimates for small populations: pov- Development Policy and Management/CPRC. erty and disability in Uganda. Presented at Human Development Learning Lalita, K. (2003) Urban Chronic Poverty in Vijayawada: Insights from houshold Week 2003, World Bank. http://info.worldbank.org/etools/bspan/ profiles over time. Conference Paper presented at ‘Staying Poor: Chronic Pov- PresentationView.asp?PID=953&EID=494 erty and Development Policy’, Manchester, 7–9 April 2003. http://idpm.man.a- Hossain, M. I., Khan, I. A., and Seeley, J. (2003) Surviving on their Feet: Charting c.uk/cprc/Conference/conferencepapers/lalitakay.pdf the Mobile Livelihoods of the Poor in Rural Bangladesh. Conference Paper Lam, D. (1999). Generating Extreme Inequality: Schooling, Earnings, and Inter- presented at ‘Staying Poor: Chronic Poverty and Development Policy’, Man- generational Transmission of Human Capital in South Africa and Brazil. Ann chester, April 2003. http://idpm.man.ac.uk/cprc/Conference/conferencepapers/ Arbor, Population Studies Center At The Institute For Social Research Univer- Seeley.pdf sity Of Michigan, Report No. 99–439. http://www.psc.isr.umich.edu/pubs/pa- Howell, J. (1996) ‘Coping with the transition: household coping strategies in pers/rr99–439.pdf Kyrgyzstan.’ Working Paper 15. London, Save the Children. Lawson, D., A. McKay and J. Okidi, (2003), ‘Poverty Persistence and Transitions Hufton, O H. 1974. The Poor of Eighteenth Century France, 1750–1789. Ox- in Uganda: A Combined Qualitative and Quantitative Analysis’, Chronic Pov- ford: Clarendon Press. erty Research Centre Working Paper 38. Manchester: Institute of Development Hulme, D. and A. Shepherd (2003). ‘Conceptualizing Chronic Poverty’, World Policy and Management/CPRC. http://www.chronicpoverty.org/cpreports2.htm Development, Volume 31, No 3, pp: 403–424. LeAˆ n, R., Goulden, J., Rea, C., Salinas, H., Medrano, L., and Schollaert, J. Hulme, D. (2002) ‘Thinking ‘small’ and the understanding of poverty: Maymana (2003) Social Exclusion, Rights and Poverty in Bolivia. Conference Paper pre- and Mofizul’s story’ Chronic Poverty Research Centre Working Paper 22. sented at ‘Staying Poor: Chronic Poverty and Development Policy’, Manchester, Manchester: Institute of Development Policy and Management/CPRC. 7–9 April 2003. http://idpm.man.ac.uk/cprc/Conference/conferencepapers/ Hulme, D., K. Moore and A. Shepherd (2001) ‘Chronic Poverty: Meanings and Goulden%20Jay%20SocialExclusionandChronicPoverty.pdf Analytical Frameworks’. CPRC working paper 2. Manchester: IDPM, Univer- Lewis, J., (2000). The Batwa Pigmies of the Great Lakes Region, Minority Rights sity of Manchester. Group International Report, MRG: London, UK. http://www.minorityright- Human Rights Watch (2003) ‘Double Standards: Women’s Property Rights Vio- s.org/Profiles/profile.asp?ID=4 lations in Kenya.’ KENYA Vol. 15, No. 5 (A) – March 2003. http:// Lewis, O. (1959). Five families: Mexican case studies in the culture of poverty. www.hrw.org/reports/2003/kenya0303/kenya0303.pdf New York: Basic Books. Hunt, D. (1994) Rural Livelihood Systems and Farm Non-Farm Linkages in Lines, T. and Tickell, S. (2003). ‘Walk the talk: a call to action to restore coffee Lower Embu, Eastern Kenya: 1972/4 to 1993/4, ESCOR Project: End of proj- farmers’ livelihoods. Oxford: Oxfam Briefing Paper 44. http://www.oxfam.or- ect report to ODA, London. g.uk/what_we_do/issues/trade/downloads/bp44_coffeefarmers.pdf Hussain, A. (2003) Urban poverty in China: measurement, patterns and policies. ILO paper. Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 137 Black plate (137,1)

. 137

Lipton, M., and Ravallion, M., (1995), ‘Poverty and policy’, in Behrman, J, and Oduro, A. D. and I. Aryee (2003) ‘Investigating Chronic Poverty in West Africa’, Srinivasan, T.N., (eds) Handbook of Development Economics, Volume 3: Am- Chronic Poverty Research Centre Working Paper 28. Manchester: Institute of sterdam: North-Holland Development Policy and Management/CPRC. Lund, F. (2002) ‘Crowding in’ care, security, and macro-enterprise formation – OECD DAC Statistics: http://www.oecd.org/department/ revisiting the role of the state in poverty reduction, and in development. Journal 0,2688,en_2649_34447_1_1_1_1_1,00.html of International Development, Vol. 14, No. 6, pp. 681–694. O’Keefe, P. and J. Holson, (1997) Azerbaijan: Poverty Assessment. Report no. Lwanga-Ntale, C. (2003). Chronic Poverty and Disability in Uganda. Conference 15602-AZ. Poverty Reduction and Economic Management Unit, Europe and Paper presented at ‘Staying Poor: Chronic Poverty and Development Policy’, Central Asia Region. Washington DC: World Bank. Manchester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/Conference/conferen- Okrasa, W. (1999) ‘Who Avoids and Who Escapes from Poverty during the Tran- cepapers/lwangaNtaleUganda%20disability%20paper%2006.03.03.pdf sition? Evidence from Polish Panel Data, 1993–96’. World Bank Policy Re- Lwanga-Ntale, C. and McClean, K. (2003) The Face of Chronic Poverty in Ugan- search Working Paper no. 2218. World Bank Development Research Group: da As Seen by the Poor Themselves – Constraints and Opportunities. Confer- Poverty and Human Resources. Washington, D.C.: World Bank ence Paper presented at ‘Staying Poor: Chronic Poverty and Development Okidi, J. and A. McKay, 2003. ‘Poverty Dynamics in Uganda, 1992–2000’, Policy’, Manchester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/Conference/ Chronic Poverty Research Centre Working Paper 35. Manchester: Institute of conferencepapers/lwangantalemclean.pdf Development Policy and Management/CPRC. Macrae, J., Shepherd, A., Morrissey, O. et al. (2004) ‘Development, Performance OPM/ODI (Oxford Policy Management and Overseas Development Institute) and Sovereignty: challenging the ‘poor performance’ paradigm in international (2002) General Budget Support Evaluability Study Phase 1 Final Synthesis Re- development’, London: Overseas Development Institute, Research Report. port, Report to UK Department for International Development (DFID) Maddison, A. (2001). The world economy: a millennial perspective. Paris: Devel- Oxfam (2002) Rigged Rules and Double Standards: trade globalisation and the opment Centre of the OECD. fight against poverty. Oxford, Oxfam. http://www.maketradefair.com/assets/ Malapit, H., Clement T., and J. Yunzal, (2003) Does Violent Conflict Make english/report_english.pdf Chronic Poverty More Likely? The Mindanao Experience, Conference Paper Pakistan Participatory Poverty Assessment – Azad Jammu and Kashmir State Re- presented at ‘Staying Poor: Chronic Poverty and Development Policy’, Man- port, 2003. chester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/Conference/ Park, C. and M Kim (1998) Current Poverty issues and Counter Policies in conferencepapers.htm Korea. Seoul and New York, UNDP Marcus, R. and Wilkinson, J., (2002), Whose Poverty Matters, Vulnerability and Pender, J. and Hazell, P. 2000. Promoting sustainable development in less-fa- Social Protection in PRSPs, CHIP Working Paper 1, London: CHIP, Save the vored areas – Overview. Washington, D.C.: IFPRI, 2000 (2020 Vision for Children – UK. Food, Agriculture and the Environment Focus 4, Brief 1). http://www.ifpri.org/ Marshall, J. (2004) Children and Poverty: Some Questions Answered. London, 2020/focus/focus04/focus04.pdf Save the Children – UK, CHIP Briefing 1. http://www.childhoodpoverty.org/ Perlman, J. E (2003) ‘Marginality: From Myth to Reality in the Favelas of Rio De index.php/action=documentfeed/doctype=pdf/id=46/ Janeiro 1969 -2002’, CPRC Conference Paper, Manchester: CPRC Matin, I. and Hulme, D. (2003). ‘Programs for the Poorest: Learning from the Perez-Bustillo, C. (2001) ‘Human Rights, Poverty and Indigenous People’s Strug- IGVGD Program in Bangladesh.’ World Development, Vol. 31, No. 3, pp. gles in the Americas: New Directions and Case Studies from Colombia, Guate- 647–665. mala, Mexico and Nicaragua, in van Genugten, W. and Perez-Bustillo, C. (eds) May, J. (2003) ‘Chronic Poverty and Older People in South Africa’. Chronic The Poverty of Rights London: Zed Books Poverty Research Centre Working Paper 25. Manchester: Institute of Develop- PMC (Professional Manager Consulting), 2000, Research on Trafficking in Mi- ment Policy and Management/CPRC. http://www.chronicpoverty.org/pdfs/ grants, Kyrgyz Republic, 1999, Bishkek: International Organisation on JMay.pdf Migration. McCulloch, N., and Calandrino, M. (2003). ‘Vulnerability and Chronic Poverty Pryer, J, Rogers, S and Rahman, A (2003) ‘Work Disabling Illness and Coping in Rural Sichuan.’ World Development, Vol. 31, No. 3, pp. 611–628. Strategies in Dhaka Slums, Bangladesh’, CPRC Conference Paper, Manchester: McKay, A. (2004) How Can the Poorest Benefit from Growth? Issues Paper CPRC Prepared for Reaching the Very Poorest Team, DFID: London. Quisumbing, A. R., Haddad, L. and PeEˆ a, C. (2001) ‘Are women overrepresented Mehta, A. K., Panigrahi, R. and S. Sivramkrishna, (2003). Operationalising Mul- among the poor? An analysis of poverty in 10 developing countries’, Journal of tidimensional Concepts of Chronic Poverty: An Explanatory Spatial Analysis, Development Economics, Vol. 66, No. 1, pp. 225–269. Conference Paper presented at ‘Staying Poor: Chronic Poverty and Develop- Rajan, S. I. (2003) Oldage Allowance Programme in Nepal. Conference Paper ment Policy’, Manchester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/Confer- presented at ‘Staying Poor: Chronic Poverty and Development Policy’, Man- ence/conferencepapers.htm chester, April 2003. http://idpm.man.ac.uk/cprc/Conference/conferencepapers/ Mehta, A.K and Shah, A (2003) ‘Chronic Poverty in India: Incidence, Causes and rajan27.03.pdf Policies,’ World Development, Vol. 31, No. 3, pp 491–513 Ravallion, M. (2003a) Targeted Transfers in Poor Countries: Revisiting the MHHDC Mahbub ul Haq Human Development Center, (2000) Human develop- Trade-Offs and Policy Options. Chronic Poverty Research Centre Working ment in South Asia 2000- the Gender Question. Oxford: MHHDC/Oxford Paper 26. Manchester: Institute of Development Policy and Management/ University Press CPRC. http://www.chronicpoverty.org/pdfs/martinravallion.pdf Minority Rights Group (no date) ‘War: The Impact on Minority and Indigenous Ravallion, M., (2003b), ‘Pick your number in the great globalisation debate’, Children’, http://www.minorityrights.org/Profiles/profiile.asp?ID=7 Mimeo, Development Economics Research Group, World Bank: Washington Mitlin, D. (2003). ‘The Economic and Social Processes Influencing the Level and DC. Nature of Chronic Poverty in Urban Areas’. Chronic Poverty Research Centre Ravallion, M, (1998), ‘Poverty lines in theory and practice’, LSMS Working Working Paper 29. Manchester: Institute of Development Policy and Manage- Paper 133, World Bank, Washington DC ment/CPRC. http://www.chronicpoverty.org/pdfs/mitlin.pdf Ravallion, M., (1994), Poverty Comparisons, Fundamentals of Pure and Applied Morrissey, O. (2001), ‘Does Aid Increase Growth?’ Progress in Development Economics vol 56, Chur: Switzerland Harwood Academic Press. Studies, Vol. 1, No. 1, 37–50. Ravallion, M., and Wodon, Q.T. (1999). Poor Areas or Only Poor People? Jour- Mosse, D. (2001) ‘People’s Knowledge, Participation and Patronage: Operations nal of Regional Science, Vol. 39, pp. 689–711. and Representations in Rural Development’, in B. Cooke and U. Kothari (eds) Ravallion, M., Datt, G., and Van de Walle, D., (1991), ‘Quantifying absolute Participation: The New Tyranny? pp. 16–35. London: Zed Books. poverty in the developing world’, Review of Income and Wealth, 31(4): 345– Mugarura, B. and Ndemeye, A. (2000) ‘Batwa Land Rights in Rwanda’ Minority 362 Rights Group International micro study, MRG: London, UK. http://www.mi- Ray, R. (1999) How child labour and child schooling interact with adult labour. norityrights.org/admin/Download/Pdf/Batwa%20Report.pdf Working Paper no 2179. Washington DC: World Bank. Muller, Christophe (2000), ‘Transient – Seasonal and Chronic Poverty of Peas- Reddy, S. and Pogge, T., (2003), ‘How not to count the poor’, Mimeo, Columbia ants: Evidence from Rwanda’, University of Nottingham, Discussion Papers in University: New York. Economics No 00/9, April. Revenga, A., Ringold, D. and Tracy, W. (2002) Poverty and Ethnicity – A Cross- Najjumba-Mulindwa, I. (2003) Chronic Poverty Among the Elderly in Uganda: Country Study of Roma Poverty in Central Europe. Washington, DC: World Perceptions, Experiences and Policy Issues, Conference Paper presented at Bank ‘Staying Poor: Chronic Poverty and Development Policy’, Manchester, 7–9 Reyes, C. (2002a), ‘Poverty Profile of the Philippines,’ MIMAP Research Paper, April 2003. http://idpm.man.ac.uk/cprc/Conference/conferencepapers.htm Policy and Development Foundation, Inc. National Council of Applied Economic Research 2001. South Indian Human Reyes, C. (2002b). ‘The Poverty Fight: Have We Made An Impact?’ Perspective Development Report. New Delhi/Oxford: NCAER/OUP. Paper Symposium Series, Philippines Institute for Development Studies, Narayan, D., with R. Patel, K. Schafft, A. Rademacher, and S. Koch-Schulte Philippines. (2000). Voices of the Poor: Can Anyone Hear Us? New York, N.Y: Published Ringold, D. (2002). Bulgaria Poverty Assessment Report no 24516-BUL, Human for the World Bank, Oxford University Press. Development Department, Europe and Central Asia Region. Washington DC: Nord, M. (1997). Overcoming persistent poverty – and sinking into it: income World Bank. trends in persistent-poverty and other high-poverty rural counties, 1989–94. Riskin, C, (1994). ‘Chinese rural poverty: marginalized or dispersed?’ American Rural Development Perspectives, Vol. 12, No. 3, pp. 2–10. http://www.ers.us- Economic Review, 84. da.gov/publications/rdp/rdp697/rdp697b.pdf Rodgers, J. L., and Rodgers, J. R. (1993). Chronic Poverty In The United States. Ochoa, M.A.U. (2001) ‘Poverty and Human Rights in the Light of the Philosophy Journal of Human Resources, Vol. 28, No. 1, pp. 25–54. and Contributions of Father Joseph Wresinski’ in van Genugten, W. and Perez- Ruben, R., A. Kuyvenhoven and P. Hazell (2003). ‘Investing in Poor People in Bustillo, C. (eds) The Poverty of Rights London: Zed Books Less-favoured Areas: Institutions, Technologies and Policies for Poverty Allevi- ODI (2003) Forum for Food Security in Southern Africa. www.odi.org.uk/food- ation and Sustainable Resource Use’ Conference Paper presented at ‘Staying security-forum Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 138 Black plate (138,1)

138 . Appendix B

Poor: Chronic Poverty and Development Policy’, Manchester, 7–9 April 2003. UNAIDS/WHO. (2002). Report on the global HIV/AIDS epidemic: July 2002. http://idpm.man.ac.uk/cprc/Conference/conferencepapers.htm http://www.unaids.org Ruthven, O. and Kumar, S. (2003) Making & Breaking Poverty in Koraon, Utter UNCTAD (2002) The Least Developed Countries Report 2002 – Escaping the Pradesh, India. Conference Paper presented at ‘Staying Poor: Chronic Poverty Poverty Trap, United Nations Conference on Trade and Development: Geneva. and Development Policy’, Manchester, April 2003. http://idpm.man.ac.uk/cprc/ UNDP (2003a) ‘Human Development Report: Millennium Development Goals: A Conference/conferencepapers/RuthvenOrlanda18rd%20March.pdf compact Among Nations to End Human Poverty’, New York/Oxford: Oxford Sankaran, S. R. (2000). ‘Welfare of Scheduled Castes and Scheduled Tribes in University Press. Independent India – An Overview of State Policies and Programmes’, Journal UNDP (2003b) Human development trends (presentation). http://hdr.undp.org of Rural Development, Vol.19, No.4. UNDP (2003c) ‘Pakistan Human Development Report 2003: Poverty, Growth Satterthwaite, David and Cecilia Tacoli (2002) .Seeking an understanding of pov- and Governance’ UNDP Pakistan erty that recognizes rural-urban differences and rural-urban linkages’. In Rako- UNDP, (2002b), The macroeconomics of poverty: case study of the Kyrgyz Re- di with Lloyd-Jones (eds), Urban Livelihoods: A people-centred approach to public. Report of a Mission to the Kyrgyz Republic of the United Nations reducing poverty. Earthscan Publications Ltd., London pp 52–70. Development Programme, UNDP, Bishkek; Saxena, N, C. (2003) ‘Livelihood Diversification and Non-Timber Forest Prod- UNHCR (2003a) Refugees by numbers 2003. http://www.unhcr.ch ucts in Orissa: Wider Lessons on the Scope for Policy Change’, ODI Working UNHCR (2003b) 2002 UNHCR Population Statistics (Provisional). http:// Paper 223, Overseas Development Institute: London www.unhcr.ch Sen, A. (1999) Development As Freedom. Oxford: Oxford University Press. UNHCHR (United Nations Office of the High Commissioner for Human Rights) Sen, B (2003) ‘Drivers of Escape and Descent: Changing Household Fortunes in (1948,2004) The Universal Declaration of Human Rights. Geneva: UNHCHR. Rural Bangladesh’, World Development, Volume 31: Number 3, pp 513–534 http://www.unhchr.ch/udhr/index.htm Sen, B. and Z. Ali (2003) ‘Spatial Inequality in Social Progress in Bangladesh’, UNICEF (2000) UNICEF: ending poverty begins with children. Press release. Conference Paper presented at ‘Staying Poor: Chronic Poverty and Develop- http://www.unicef.org/newsline/00pr54.htm ment Policy’, Manchester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/Confer- UNICEF, (1999), School Attendance in the Kyrgyz republic – Report for UNICEF ence/conferencepapers.htm Bishkek, GoKR and UNICEF, Bishkek SCF (2003) State of the World’s Mothers: Protecting women and children in war UNPD (Population Division of the Department of Economic and Social Affairs of and conflict. Westport CT, SCF. http://www.savethechildren.org/publications/ the United Nations Secretariat), World Population Prospects: The 2002 Revi- SOWMPDFfulldocument2.pdf sion and World Urbanization Prospects: The 2001 Revision, http://esa.un.org/ Shankardass, M. K. (2002) ‘Voices of the older poor in India’ in Situation and unpp, 16 January 2004; 2:08:50 PM. voices: the older poor and excluded in South Africa and India. Population and UN SCN (2003) Report on the nutrition situation of refugees and displaced pop- Development Strategies Number 2, UNFPA, 113–144. ulations. RNIS 41. http://www.unsystem.org/scn/Publications/RNIS/rnis41.pdf Shah, A. K and A. Sah, (2003). Poverty Among Tribals in Madhya Pradesh: Has Vakis, R. and Lindert, K. (2000). Poverty in Indigenous populations in Panama: Anything Changed Over Time? Conference Paper presented at ‘Staying Poor: a study using LSMS data. Annex 6 of World Bank Panama Poverty Assessment Chronic Poverty and Development Policy’, Manchester, 7–9 April 2003. http:// 2000. http://www.worldbank.org idpm.man.ac.uk/cprc/Conference/conferencepapers.htm Venkateshwarlu, D. and daCorta, L. (2001) ‘Transformations in the age and Sharma, A.N., Karan, A.K., Sarkar, S. (2002) Dynamics of Employment, Poverty gender of unfree workers on hybrid cotton seed farms in Andhra Pradesh’, and Human Development in Rural Bihar. New Delhi: Institute for Human Journal of Peasant Studies, Vol. 28, No. 3, pp.1–36. Development. Watkins, K. (2001) The Oxfam education report. Oxford: Oxfam. Shinyekwa, I., Bird, K. and J. Okidi (2003) ‘Spatial Differentiation in Access to WFP (World Food Programme) (2004) Emergency Report No. 7 of 2004. http:// Goods and Services in Uganda’, DRAFT Chronic Poverty Research Centre www.reliefweb.int Working Paper. Manchester: Institute of Development Policy and Manage- WFP (2001) ‘Food Security Atlas of Rural India’, M.S Swaminathan Research ment/CPRC. Foundation, Chennai Singh, J. (1989). Banks, gods, and government – institutional and informal credit World Health Organisation (WHO) (2003) ‘The World Health Report 2003 – structure in a remote and tribal Indian district (Kinnaur, Himachal Pradesh) Shaping the Future,’ Geneva: WHO. 1960–1985. Stuttgart: Steiner-Verlag Wiesbaden. WHO (2001) ‘The World Health Report 2001, Mental Health: New Understand- Sinha, S.; Lipton, M. et al. (1999) Damaging Fluctuations, Risk and Poverty: A ing, New Hope’, Geneva: WHO. Review. Background Paper for the World Development Report 2000/2001 Wodon, Q. (2001) ‘Attacking extreme poverty: learning from the experience of Poverty Research Unit at Sussex, University of Sussex. http://www.worldban- the International Movement ATD Fourth World’, Report Number World Bank k.org/poverty/wdrpoverty/background/sinhaliptn.pdf technical paper; no. 502 Skoufias, E. (2003). ‘Economic crises and natural disasters: coping strategies and Wood, G. (2003) Staying Secure, Staying Poor: The ‘Faustian Bargain’. World policy implications’ World Development 31/7:1087–1102. Development, Vol 31, No. 3, pp: 455–471. South Africa Department of Social Development (2002) National Report on the Woodhouse, (2003) ‘Perceptions of Poverty Among Local Elites in Uganda’, Con- Status of Older Persons 1994–2002. Report to the Second Assembly on Ageing, ference Paper presented at ‘Staying Poor: Chronic Poverty and Development Madrid, Spain. Policy’, Manchester, 7–9 April 2003. http://idpm.man.ac.uk/cprc/Conference/ Stewart, F. and M. Wang, (2003), Do PRSPs Empower Poor Countries and Dis- conferencepapers.htm empower the World Bank or Is it the Other Way Round?, QEH Working World Bank, (2003a) World Development Report 2002/03: Sustainable Develop- Paper Series 108, http://www.eurodad.org/uploadstore/cms/docs/ ment in a Dynamic World: Transforming Institutions, Growth and the Quality WBevalMay03.pdf of Life. World Bank: Oxford University Press. Stiglitz, J. (2002) Globalization and Its Discontents. New York, Allen Lane. World Bank (2003b) China country assistance strategy, 2003–2005. The World Sz kely, M. 2001. The 1990s in Latin America: another decade of persistent Bank Group. http://www.worldbank.org.cn/English/Content/cas03.pdf inequality, but with somewhat lower poverty. IADB Research Department World Bank, (2003c), Kyrgyz Republic. Enhancing Pro-poor Growth, Report Working Paper #454. NO. 24638-KG, PREM, ECA Region, World Bank, Washington, DC Tarp, F. (ed.) with Hjertholm, P. (ed.) (2000) Foreign Aid and Development: World Bank (2003d) Solomon Islands Data Profile. http://devdata.worldbank.org/ Lessons Learnt and Directions for the Future. London: Routledge Studies in external Development Economics. World Bank (2001). China: overcoming rural poverty. Washington: World Bank. Third World Resurgence No. 94, June 1998 http://www.twnside.org.sg/title/ http://www.worldbank.org.cn/English/content/poverty.pdf india1-cn.htm World Bank (2000) World Development Report 2000–2001: Attacking Poverty. Timmer, C.P. 1997. How well do the poor connect to the growth process? Cam- Oxford University Press: New York. http://www.worldbank.org/poverty/ bridge, MA: Harvard Institute for International Development, CAER II Discus- wdrpoverty/report/index.htm sion Paper No. 17. http://www.cid.harvard.edu/caer2/htm/content/papers/ World Bank, 1990, World Development Report 1990: Poverty, New York: Ox- paper17/paper17.pdf ford University Press Tudawe, I. (2001a). Chronic poverty and development policy in Sri Lanka: over- Wu, W. (2001). Labor mobility in China: a review of the program redressing view study. Chronic Poverty Research Centre Working Paper 9. Manchester: discrimination against labor migrants 1997–2001. Report prepared for the Institute of Development Policy and Management/CPRC. http://www.chronic- Ford Foundation China Program. poverty.org/pdfs/srilanka.pdf Yaqub, S. (2000) ‘Poverty dynamics in developing countries, IDS Development Tudawe, I. (2001b). ‘Undernutrition and poverty’. Unpublished paper for Chronic Bibliography’ University of Sussex. Poverty Research Centre, Theory Working Group, December 18, 2001, Yarkova, T., Botoeva, G., Reeves, M., Konokbaev, K., Yarkova, N., Marcus, R. Birmingham. and Satybaldieva, E. (2004) Childhood Poverty in Kyrgyzstan, Initial Literature UNAIDS (2002) Epidemiological Fact Sheets on HIV/AIDS and Sexually Trans- Review. CHIP Report 1, CHIP, London. mitted Infections – 2002 Updates for India and China. http://www.unaids.org/ Yeo, R. and Moore, K. (2003). ‘Including disabled people in poverty reduction hivaidsinfo/statistics/ fact_sheets work: ‘nothing about us, without us’’, World development, Vol. 31, No. 3, pp. UNAIDS (2000) HIV and AIDS-related stigmatisation, discrimination and denial: 571–590. forms, contexts and determinants – Research studies from Uganda and India. ‘Zedillo Panel’ (2001) Recommendations of the High-level Panel on http://www.unaids.org Financing for Development. http://www.un.org/reports/financing/full_report.pdf UNAIDS/WHO. (2003). AIDS Epidemic Update: December 2003. http:// www.unaids.org Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 139 Black plate (139,1)

APPENDIX CPRC working papers and Chronic Poverty Report C 2004–05 background papers

Background Papers Investigating Chronic Poverty in West Africa by Abena D Oduro and Baulch, B. (2003) Aid for the Poorest Ivy Aryee. Baulch, B and A. McKay (2003) How Many Chronically Poor People Poverty Dynamics in Uganda: 1992 to 2000 by John A Okidi and Are There In The World? Some Preliminary Estimates. Andrew McKay. Bezemer, D. (2003) Poverty in Post-Socialist Transition Countries. Targeted Transfers in Poor Countries: revisiting the trade offs and poli- De Swardt, C. (2003). The Shadow of the Rainbow Nation: Chronic cy options by Martin Ravallion. Poverty after a Decade of Liberalisation Chronic Poverty and Older People in South Africa by Julian May. Hickey, S. with S. Bracking (2003) Chronic Poverty Report 1: Politics Towards a clearer Understanding of ‘Vulnerability’ in relation to Background Paper. Chronic Poverty by Martin Prowse. Kyegombe, N. (2003). Health and Chronic Poverty The Political Economy of Chronic Poverty by Sarah Bracking. Lebrun, N. (2003). Chronic Poverty in China Thinking ‘Small’ and the Understanding of Poverty: Maymana and Mo- Marcus, R. (2003). CHIP Contributions to CPRC Chronic Poverty Re- fizul’s Story by David Hulme. port 1 Chronic Poverty: scrutinizing estimates, patterns, correlates and explan- Masset E. and H. White. (2003). Are the chronic poor to be left out of ations by Shahin Yaqub progress towards the Millennium Development Goals? A quantitative Targeted Development Programmes for the Extreme Poor: experiences analysis of the elderly, disabled, orphans and unsupported females from BRAC experiments by Imran Matin. Piron, L.H. (2003). Chronic Poverty and Human Rights Background Whose Poverty Matters? Vulnerability, social Protection and PRSPs by Paper Rachel Marcus & John Wilkinson. CHIP Working Paper No 1. Wheeler, J. (2003). Background Paper on Chronic Poverty in Latin Chronic Poverty in Semi Arid Zimbabwe by Kate Bird and Andrew America Shepherd. Yaqub, S. (2003). Severe Poverty and Chronic Poverty Do Monetary and Non-Monetary Indicators tell the same story about chronic poverty? A study of Vietnam in the 1990s by Bob Baulch and Key Websites: Edoardo Massett. Conference Papers presented at ‘Staying Poor: Chronic Poverty and Chronic Poverty and Migration by Uma Kothari. Development Policy’, Manchester, 7–9 April 2003 can be viewed at: Chronic Poverty: A Review of Current Quantitative Evidence by An- http://idpm.man.ac.uk/cprc/Conference/conferencepapers.htm drew McKay and David Lawson. Chronic Poverty Research Centre Working Papers can be viewed at: Natural Resource Management and Chronic Poverty in Sub Saharan http://www.chronicpoverty.org Africa: An Overview Chronic Poverty and Remote Rural Areas by Kate Bird, David Hulme, Chronic Poverty Research Centre Working Papers Andrew Shepherd and Karen Moore. Thinking about Chronic Urban Poverty by Philip Amis Poverty Persistence and Transitions in Uganda: A Combined Qualita- An overview of Chronic Poverty and Development Policy in Uganda by tive and Quantitative Analysis by David Lawson, Andy McKay, John John Okidi and Gloria Mugambe. Okidi Chronic Poverty and Older People in the Developing World by Aman- The Politics of Staying Poor in Uganda by Sam Hickey da Heslop and Mark Gorman. Multiple Shocks and Downward Mobility: learning from life histories Chronic Poverty and Development Policy in Sri Lanka: Overview of rural Ugandans by Kate Bird and Isaac Shinyekwa Study by Indra Tudawe. Aid for the Poorest? The distribution and maldistribution of interna- Frameworks for Understanding the Inter-generational Transmission of tional development assistance by Bob Baulch poverty and Well-being in Developing Countries by Karen Moore. Annotated Bibliography: Poverty and Chronic Poverty in Uganda by Chronic Poverty in India: Overview Study by Aasha Kapur Mehta and Isaac Shinyekwa with Chris Taylor. CPRC Annotated Bibliographies Amita Shah. No 2 Violent Conflict, Poverty and Chronic Poverty by Jonathan Goodhand. What is the Impact of non-contributory Pensions on Poverty? Estimates Livelihoods Research: Some Conceptual and Methodological Issues by from Brazil and South Africa by Armando Barrientos. Colin Murray. The Intrahousehold Disadvantages Framework: A framework for the Chronic Poverty and Disability by Rebecca Yeo. analysis of intra-household difference and inequality by Vincent J. Study of the Incidence and Nature of Chronic Poverty and Develop- Bolt and Kate Bird. ment Policy in South Africa: An Overview by Michael Aliber. Bureaucratic Effects: ‘Weberian’ State Structures and Poverty Reduc- Chronic Poverty: Meanings and Analytical Frameworks by David tion by Jeffrey Henderson, David Hulme, Hossein Jalilian and Ri- Hulme, Karen Moore and Andrew Shepherd. chard Phillips. Chronic Poverty in Sub-Saharan Africa and South Asia: A select anno- Globalization, the International Poverty Trap and Chronic Poverty in tated bibliography with special reference to remote rural areas. by the Least Developed Countries by Charles Gore. Samuel Hickey. CPRC Annotated Bibliographies No 1 The Economic and Social Processes Influencing the Level and Nature of Chronic Poverty in Urban Areas by Diana Mitlin. All working papers can be downloaded free of charge from www.chronicpoverty.org Discript Postscript CRC Stage: Final page Date: 14/04/04 File: {CPR}3B2FILES/THIRD/3P312E-02.3D.3D Folio: 140 Black plate (140,1)

APPENDIX The Chronic Poverty Report D 2006–07

The Chronic Poverty Research Centre is continuing to develop 5. Reducing levels of inequality through redistributive policies, qualitative and quantitative datasets, research methodologies including resource allocations for public expenditure which and theoretical frameworks for the analysis of chronic poverty. promote services for the poor and progressive social change. The Chronic Poverty Report 2004–05 looks at current global chronic poverty. The Chronic Poverty Report 2006–07 will Empowerment: look forward to 2015 to examine potential changes in chronic 6. Enhancing the rights of the chronically poor. poverty levels under different scenarios around rates of eco- 7. Improving the quality of governance, notably the quality of nomic growth and changes in levels of inequality; rising inci- institutions and policy implementation. dence and impacts of HIV/AIDS; and changes in conflict and 8. Reducing discrimination and working toward enhanced so- insecurity. cial integration. The Chronic Poverty Report 2004–05 will be directly focused 9. Addressing the key cultural maintainers of chronic poverty on policy and policy-making processes. It will examine possible in the household and community. futures not only in terms of changes in US$1/day chronic poverty, but by also looking at nutritional-based measures, changes Obligations to provide resources: in assets and other multidimensional indicators. 10. Monitoring international commitments to increase aid vol- The second Chronic Poverty Report will specifically focus on umes, and implications for recipient sectors and countries. the policies required to tackle chronic poverty globally. This 11. Improving financial transfers to chronically poor people, means addressing the complex social, economic and political promoting effective management of universal and targeted processes that enable individuals, households, communities, schemes. and key national and international development actors to respond to chronic poverty. The global challenge: This report will draw concrete conclusions not only on what 12. Identifying weaknesses in the international system that in- policies are important but also on how to promote their imple- hibit chronic poverty reduction, and generate performance mentation. This will involve engaging with issues and institu- indicators for international institutions. tions from the local to the global level. It will mean examining ways to encourage the political will necessary to drive the implementation of a chronic poverty reduction agenda and to secure the necessary financial resources. The key policy issues that will be the main focus for The Chronic Poverty Report 2004–05 are:

Prioritising livelihood security: 1. Addressing the chronic insecurity that affects chronically poor people, in ways that protects them against vulnerability and promotes their ability to take advantage of opportunities. 2. Improving the nutritional levels and key capabilities of chronically poor people, particularly children. 3. Providing social protection for the poorest to inhibit the inter-generational transmission of poverty. Regional chronic poverty reports and reviews Growth, inequality and redistribution: Chronic Poverty in India, CPRC-IIPA, New Delhi, 2003 4. Promoting pro-poor growth, enabling the chronically poor State of the Poorest in Bangladesh 2004/2005 (forthcoming, to participate more effectively in broad based economic June 2004) growth that increases demand for their labour, goods and Uganda Chronic Poverty Report 2004/2005 (forthcoming, services. June 2004)