Migration and Citizenship: Rights and Exclusions (in Social Protection and Migration, 2011)

Rethinking International Development Series

Series Editors:

Andy Sumner, Fellow of the Vulnerability and Poverty Research Team, Instituteof Development Studies, UK.Ray Kiely, Professor of International Politics, Queen Mary University ofLondon, UK.

Palgrave Macmillan is delighted to announce a new series dedicated to publishingcutting-edge titles that focus on the broad area of ‘development’.

The core aims of the series are to present critical work that:

– is cross disciplinary;– challenges orthodoxies;– reconciles theoretical depth with empirical research;– explores the frontiers of development studies in terms of ‘development’ in

both North and South and global inter-connectedness;– reflects on claims to knowledge and intervening in other people’s lives.

Titles include:

Simon Feeny and Matthew ClarkeTHE MILLENNIUM DEVELOPMENT GOALS AND BEYONDInternational assistance to the Asia-Pacific

Niamh GaynorTRANSFORMING PARTICIPATION?The politics of development in Malawi and Ireland

Sue Kenny and Matthew Clarke (editors)CHALLENGING CAPACITY BUILDINGComparative perspectives

Sara C. Motta and Alf Gunvald Nilsen (editors)SOCIAL MOVEMENTS IN THE GLOBAL SOUTHDispossession, development and resistance

Eric Rugraff, Diego Sánchez-Ancochea, Andy Sumner (editors)TRANSNATIONAL CORPORATIONS AND DEVELOPMENT POLICYCritical perspectives

Rachel Sabates-Wheeler and Rayah Feldman (editors)MIGRATION AND SOCIAL PROTECTIONClaiming social rights beyond borders

Jens Stilhoff Sörensen (editor)CHALLENGING THE AID PARADIGMWestern currents and Asian alternatives

Andy Sumner and Meera TiwariAFTER 2015: INTERNATIONAL DEVELOPMENT POLICY AT A CROSSROADS

Rethinking International Development SeriesSeries Standing Order ISBN 978–0230–53751–4 (hardback)(outside North America only)

You can receive future titles in this series as they are published by placing astanding order. Please contact your bookseller or, in case of difficulty, write tous at the address below with your name and address, the title of the series andthe ISBN quoted above.

Customer Services Department, Macmillan Distribution Ltd, Houndmills,Basingstoke, Hampshire RG21 6XS, England

Migration and SocialProtectionClaiming Social Rights Beyond Borders

Edited by

Rachel Sabates-WheelerResearch Fellow, Centre for Social Protection, Institute of Development Studies,Sussex, UK

Rayah FeldmanHonorary Visiting Research Fellow, Department of Arts and Human Sciences,London South Bank University, UK

Introduction, selection and editorial matter © Rachel Sabates-Wheelerand Rayah Feldman 2011Individual chapters © contributors 2011Preface © Don Flynn 2011

All rights reserved. No reproduction, copy or transmission of thispublication may be made without written permission.

No portion of this publication may be reproduced, copied or transmittedsave with written permission or in accordance with the provisions of theCopyright, Designs and Patents Act 1988, or under the terms of any licencepermitting limited copying issued by the Copyright Licensing Agency,Saffron House, 6-10 Kirby Street, London EC1N 8TS.

Any person who does any unauthorized act in relation to this publicationmay be liable to criminal prosecution and civil claims for damages.

The authors have asserted their rights to be identifiedas the authors of this work in accordance with the Copyright,Designs and Patents Act 1988.

First published 2011 byPALGRAVE MACMILLAN

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,registered in England, company number 785998, of Houndmills, Basingstoke,Hampshire RG21 6XS.

Palgrave Macmillan in the US is a division of St Martin’s Press LLC,175 Fifth Avenue, New York, NY 10010.

Palgrave Macmillan is the global academic imprint of the above companiesand has companies and representatives throughout the world.

Palgrave® and Macmillan® are registered trademarks in the United States,the United Kingdom, Europe and other countries.

ISBN 978–0–230–24591–4 hardback

This book is printed on paper suitable for recycling and made from fullymanaged and sustained forest sources. Logging, pulping and manufacturingprocesses are expected to conform to the environmental regulations of thecountry of origin.

A catalogue record for this book is available from the British Library.

Library of Congress Cataloging-in-Publication DataMigration and social protection : claiming social rights beyond borders /[edited by] Rachel Sabates-Wheeler and Rayah Feldman.

p. cm.Includes index.ISBN 978–0–230–24591–4 (hardback)1. Emigration and immigration. 2. Human rights. 3. Immigrants—Civil rights. 4. Foreign workers—Civil rights. 5. Refugees—Civil rights.6. Illegal aliens—Civil rights. I. Sabates-Wheeler, Rachel, 1971–II. Feldman, Rayah.JV6035.M5423 2011323.3′291—dc22 2011001484

10 9 8 7 6 5 4 3 2 120 19 18 17 16 15 14 13 12 11

Printed and bound in Great Britain byCPI Antony Rowe, Chippenham and Eastbourne

Contents

List of Tables, Figures, and Boxes vii

Preface by Don Flynn ix

Acknowledgements xv

Notes on Contributors xvi

Part 1 Frameworks

Introduction: Mapping Migrant Welfare onto SocialProvisioning 3Rachel Sabates-Wheeler and Rayah Feldman

1 Migration and Citizenship: Rights and Exclusions 36Tendayi Bloom and Rayah Feldman

2 Structures of Access to Social Provision for Migrants 61Ian MacAuslan and Rachel Sabates-Wheeler

Part 2 Policies and Policy Spaces

3 Social Security for Migrants: A Global Overview ofPortability Arrangements 91Rachel Sabates-Wheeler, Johannes Koettl and Johanna Avato

4 Political and Regulatory Dimensions of Access, Portabilityand Exclusion: Social Security for Migrants, with anEmphasis on Migrants in Southern Africa 117Marius Olivier

Part 3 Case Studies

5 ‘An Increasingly Uncomfortable Environment’: Access toHealth Care for Documented and UndocumentedMigrants in the UK 143Rosalind Bragg and Rayah Feldman

v

vi Contents

6 A Market Place for Migrants: Mobility, Settlement andSocial Protection in Kerala 164Mythri Prasad-Aleyamma

7 Crossing Internal Boundaries: Political and Physical Accessto the Public Distribution System in India 183Ian MacAuslan

8 Social Protection as a Multi-Actor Process in EcuadorianMigration: Towards a Transnationalism of Social Rights? 210Paolo Boccagni

9 Coping and Investment Strategies of Migrants in theSouth: Malawian Migrants in South Africa 232Rachel Sabates-Wheeler

10 Bangladeshi Labour Migrants in the Gulf: Coping with aLarge Unmet Need for Social Protection 262Chowdhury R. Abrar

Index 283

List of Tables, Figures, and Boxes

Tables

I.1 Determinant of disadvantage for migrants 122.1 Access constraints in different distribution systems 733.1 Global migrant stock estimates and bilateral portability

arrangement by origin and host income group (2000) 1043.2 Global emigrant stock estimates by origin region and

portability regime (2000) 1066.1 Number of contractors registered in different sectors

under Inter-state Migrant Workmen Act, Ernakulamdistrict 177

7.1 Categories of migrants’ disadvantage 1857.2 PDS households in Delhi 2007, various estimates 1877.3 Fair Price Shop prices, Delhi, 2007 1887.4 Rules for the ration card queue, Delhi, 2007 1897.5 Summary of migrant-related disadvantage in the PDS

in Delhi 2038.1 Emigrant policies as a source of social protection from

above: a typology of the realms of action and of the keystrategies 217

9.1 Demographic indicators by country of destination, means 2409.2 Level of education by country of destination 2419.3 Investments for the future of South–South labour

migrants in home locations, per cent indicating ‘yes’ 2499.4 Percentage of migrants having access to standard

employment-based SP at origin and at destination 2509.5 Variable description and mean values (standard

deviations in parentheses) 2559.6 Probit regression estimates for determinants of a positive

change in income due to migration 257

Figures

7.1 Venn diagram of relationship between migration, povertyand slum-dwelling in Delhi 192

vii

viii List of Tables, Figures, and Boxes

9.1 Pre-migration and post-migration poverty status, bycountry of destination, in percentages 245

9.2 Prior occupation in origin, by country of destination 2469.3 The impact of migration on physical assets in home

country (percentage saying yes) 248

Boxes

3.1 Differential access to welfare services at destination 969.1 Bribery and extortion 2439.2 ‘Parallel’ systems of social protection 253

Preface by Don Flynn

The new global economy, which emerged in phases from the drivetowards deregulation and the expansion of capital markets in the 1980s,firmly strapped prospects for high future growth rates to the ever morerapid movement of the factors of production and commerce acrossnational frontiers. The capacity of the nation state to act on its ownto stimulate the levels of demand needed to counter downturns in thebusiness cycle was reduced, as the levers which allowed for economicmanagement – control of exchange rates, direct state involvement insectors commanding the heights of the national economy, and exten-sive regulation of production and markets – were dismantled in favourof an economic liberalism which assumed that more rapid growth wouldbe assured by the unfettered operation of market forces.

This model, aggressively pursued by the governments which were atthe core of the Washington Consensus that had emerged in coherentideological form by the end of the 1980s, produced consequences whichare highly relevant to the issues discussed in this collection of essays onmigration and social protection. In developing countries it led to a sharpreduction in the levels of social provision made for welfare and protec-tion against hardship as the structural adjustment programmes imposedby the International Monetary Fund on fragile economies bit deep. Thepublic sector contracted, incomes fell and systems of governance cameunder huge strain as countries in the developing South struggled to finda place in the globalized economy.

In response to these changes, hundreds of thousands of people fromthe regions of Africa, Asia, Latin America and the Caribbean wereobliged to consider migration as a livelihood strategy to replace thework which had formerly been available on farms, in the education andhealth sector, and in government-sponsored infrastructure developmentprojects. For most of these people, the migration trail took them onjourneys measured in hundreds rather than thousands of miles, as theymoved from rural to city areas or across the immediate borders of theircountries to seek employment in neighbouring states. For this groupat least, being close to home meant that some semblance of the socialprotection structures that had existed in their home countries could berebuilt in the often informal networks of mutual aid and support whichare typical of transnational communities.

ix

x Preface by Don Flynn

Yet one of the notable effects of globalization was to open up optionsfor migratory journeys, which took some individuals far beyond theirhome regions to seek employment opportunities in the developedeconomies of the North. This group typically had occupational skillsor education qualifications which had derived from previous globalexpansions, reproducing the education systems of the metropolitanheartlands, and their forms of management and administration, in thecountries of their various peripheries.

For this group, migration had extended the lines of communica-tion with their home regions and thinned the capacity of networksof co-nationals to provide even basic levels of mutual aid and support.However, the reduced opportunities for these types of solidarities couldbe compensated for by access to social rights and protection provided bymature welfare systems in the host countries. The immediately preced-ing generation of migrants, whose entry into the metropolitan countrieshad been facilitated by the old relations of colony and empire, had ben-efited from the relative equality of access to health, education, welfareand social security systems they usually enjoyed with the citizens ofthose states.

But equality of access can less safely be assumed for the migrants whowere propelled into movement by the new economic order of the 1980sand onwards. For one thing, the same restructuring that had changedthe prospects for livelihoods in the developing countries was also rollingover the forms of welfare state which had been established in developedcountries on the presumption of social and political control over mar-kets during the immediate post-Second World War period. The greaterease of movement now enjoyed by capital rendered the revenue basisfor the welfare state markedly less secure. The emerging political elitesaccommodated themselves to this dilemma by rethinking the principlesof universalism which had underpinned the more radical versions ofwelfare societies, changing the rules governing entitlement and accessfor many categories of citizens and other residents.

With these political moods driving welfare reform in many coun-tries it is perhaps not surprising that migrants would be particularlyvulnerable to exclusion from the mainstream forms of social welfareand protection available in the developed countries. Yet if these mattersstopped at this point, the accounts of welfare dilemmas and migrants’lives would not be as intriguing and suggestive of vibrant political dis-course as those described in the chapters in this collection. For the factis that the story is not one of simple, straightforward migrant exclu-sion from mainstream welfare provision, but a much more complex and

Preface by Don Flynn xi

nuanced narrative, which takes into account regional integration, polit-ical contest, civil society activism and the perennial striving of migrantsthemselves for agency and influence within systems which appear to bedriving down exactly this possibility.

The authors of these studies invite the reader to consider actual expe-riences and the real practices of welfare and social protection systems,including the ways in which they interact with the strivings of migrantcommunities for inclusion. The example of the European Union is citedon several occasions as having singularly paradoxical features in thatit aims simultaneously for strict control of migrant inflow (‘FortressEurope’) but at the same time also for the rapid integration of thosewho do cross its frontiers. Among the best protected, in formal termsat least, are those migrants who hold citizenship of a member state andwho therefore benefit from the right of equality of treatment in respectof welfare and social security. But the EU also provides for the positionof long-term migrants, who reside lawfully in any one member statefor five or more years, and who are at this point assimilated into theprotection of the free movement regime.

But a more complete account of the story requires that we lookbeyond what is prescribed in formal law to examine the relations whichexist between host societies and people who have newly arrived in them.The studies here draw attention to asymmetries which exist even whenthe law provides for equality, in the form of the vulnerabilities whichderive from cultural uncertainty, language deficits and presumptions infavour of the power of employers over wage earners, which are criticalelements in determining the outcomes of social protection.

A further component of these dilemmas comes from the reassertion ofthe category of citizenship in deciding matters of access, which is a fea-ture of the ways in which migration is managed in many countries. Theholding of full citizenship is increasingly the condition of access for agrowing range of social welfare services. Even for some EU nationals, theimposition of transitional arrangements limiting access to social secu-rity and welfare arrangements for citizens of the ‘accession 8’ countrieswhich joined the EU in 2004, and the ‘accession 2’, who joined in 2007,has had a significant impact on their integration into the realm of legalrights and entitlements.1 In the case of the UK, limitations on inclu-sion were intended to expire after one year of registered employment,and thereafter the migrant acquired full equality with British citizens.But the experience of the recession from 2008 onwards has revealed thefact that the concentration of these workers in employment in casualand informal sectors has meant that many were not able to accrue the

xii Preface by Don Flynn

required record of continuous employment and are now experiencingreal hardship. These often young migrant workers remain excluded frommainstream social security systems even after contributing through taxand national insurance for four or more years.

Indeed, as the contributors to this book point out on many occa-sions, complexity is the dominant feature of modern social welfaresystems, producing not only the rules which include and exclude, butwhich also organize citizens and residents into multi-layered hierar-chies where nationality and immigration status open doors to accesssome types of service or benefits, but not others. Bilateral and multilat-eral arrangements which set out the availability of reciprocal relationsbetween citizens of specified states are a feature of the systems operat-ing in Europe and the Americas, and to a much lesser extent in Africaand Asia. This complexity might be considered an undesirable compo-nent of modern administration but for the fact that, in the immigrationcontext, it provides a justification for enrolling welfare agencies and ser-vices into the task of maintaining surveillance over the movements ofmigrant groups and a theoretically increased capacity to enforce rulesand regulations.

All of these factors provide a very difficult set of circumstances formigrants as they attempt to pursue their livelihood strategies in hostcountries. On the face of it, the question of whether they succeed ornot wholly depends on their good fortune in finding a decent employerwho is prepared to honour legal and moral obligations to pay the fairrate for the job and provide a safe and secure working environment.The absence of accessible mainstream welfare systems means that thereappears to be precious little scope for the migrant to endure any level ofadversity, such as sickness and unemployment, or to counter the actionsof an employer whose business strategy requires excessive exploitation.

In practice, the very worst of these experiences are mitigated by theactions of migrants in establishing networks which provide a degree ofmutual aid. The motivation for this activity can be, and often is, altruis-tic and springs from moral, ethical and sometimes political sentimentswhich mandate solidarity with co-nationals or other migrants involvedin the same area of work. Traditions of savings clubs allow people to poolresources and plan escape from particularly exploitative sectors, typi-cally by gaining access to business and self-employment opportunities.Migrant networks provide information resources which improve thechances of gaining better standard accommodation, jobs with decentemployers or general practice doctors indifferent to the immigrationstatus of their patients.

Preface by Don Flynn xiii

Migrant networks can also be the starting point for the accumulationof the social capital needed to open negotiations with representatives ofthe mainstream host society. Providers of public health services or localauthorities concerned with public health standards in the private rentedsector will often find aspects of their work facilitated by good work-ing relations with groups which represent migrant interests. Migrantcommunities in turn will see the benefits to be had from nurturingindividuals equipped with the language and personal skills needed tomake the case for better treatment. In time the process can advanceto the stage where formal alliances are made with organizations rootedin the civil mainstream, like trade unions, faith communities, tenantsand residence associations, and others working for integration and socialcohesion.

However, this process also produces its own tensions and dilemmas.Communities can be preyed on by individuals selling self-proclaimedskills as legal experts or go-betweens, often asking high prices for indif-ferent services. The trust required to maintain voluntary networks canbe abused, with one-time representative advocates abandoning theirassociations in pursuit of personal advantages. These uncertainties canmake migrant support networks unstable and unreliable mechanismsfor promoting welfare over longer periods of time, but they are never-theless an important feature of the social protection systems which arediscussed in this volume.

The theoretical perspectives set out here, and the detailed discus-sions of actual experience make this a valuable book that will hopefullyinspire more research and thinking into what is currently a poorly con-sidered aspect of migrant life. But one final feature of the approachestaken by virtually all the chapters which follow is the rigorous insistenceon addressing the situation of undocumented migration at the heart ofthe dilemmas which need to be considered in this area of social policy.

Undocumented, or irregular, migration is commonly presented bygovernment and immigration control agencies as a choice made on thepart of an individual to attempt to defraud a system in order to gain ben-efits and entitlements to which he or she has no legal or moral claim.From this perspective it is a criminal activity that fully deserves the pro-scription of ‘illegal migration’ and the heavy penalties it merits. This isan unworthy approach to understanding a phenomenon which prop-erly needs to be understood as the outcome of a prolonged social andeconomic process in which the formal and legal choices of the migrantare successively closed down to the point where only the options ofmarginality and undocumented survival are left. What pushes migrants

xiv Preface by Don Flynn

in the direction of irregular migration is seldom a cynical desire to gainan advantage at the expense of others, but rather the fact that it becomesthe only remaining way of pursing a livelihood strategy. For this reasonthe consideration of irregular migration as a set of circumstances whichtake place largely within the realm of action-promoting welfare, secu-rity and protection is wholly appropriate, and the insights gained fromsituating the subject in this context in the various chapters in this bookshould be thoroughly welcomed.

This book is a valuable addition to the currently small but growingliterature on the welfare dimension of migration and it will no doubttake its place as a resource used by academic researchers. But its perspec-tive and core arguments also need to be assimilated into the workingpractices of those dealing directly with the needs of migrant commu-nities, such as policy makers, public service administrators, civil societyactivists and the proponents of the rights of migrants. Current trendstowards the exclusion of migrants from systems providing adequate lev-els of protection from risk and hardship might become more entrenchedduring what is likely to be a long period of recovery from the financialcrisis of recent years. Nevertheless, arguments for placing limits on suchexclusion and returning to a logic of protection and inclusion will beavailable to those who follow closely the dynamics of migration in theyears ahead.

Note

1. The 2004 enlargement of the European Union concerned the followingcountries: Cyprus, Czech Republic, Estonia, Hungary, Latvia, Lithuania, Malta,Poland, Slovakia and Slovenia. Part of the same wave of enlargement was theaccession of Bulgaria and Romania in 2007.

Acknowledgements

This book arose from a conference on Social Protection and Migration atthe University of Sussex in November 2008. The conference representedthe culmination of a four-year piece of research on Social Protection forMigrants that was conducted as part of the DFID-funded DevelopmentResearch Centre (DRC) on Migration, Globalisation and Poverty at theUniversity of Sussex, England.

Many individuals and organizations have been helpful in encour-aging, funding and supporting the production of this book. We areextremely grateful to the DRC and the DFID Migration team for fund-ing much of the research contained here. In particular we would liketo thank Richard Black (the Director of the DRC) and Charlotte Heath(then Head of the DFID migration team) for their commitment tosupport this area of work, and in particular for supporting Rachelin developing her work on migration, poverty and social protection.Collaboration with colleagues at the World Bank, especially JohannesKoettl, Robert Holzmann and Johanna Avato, enabled this research tomove to a ‘global’ level.

We also wish to thank the many other collaborators who supportedthis work, through providing background papers, facilitating fieldwork,commenting on research findings, and attending conferences and meet-ings. These include: Colette Solomon, Sarah Coll-Black, Wendy Buta,Don Flynn, Bilsoni Itaye, Mythra Waite, Maxine Reitzes, and the manymigrants and their families who willingly gave up their time to answersurveys and recount their migration histories.

For helping with getting this manuscript into shape – editing, format-ting, indexing, chasing contributors and much more – we are extremelygrateful to Marion Clarke. She has provided consistent, professional andcalm support throughout.

xv

Notes on Contributors

Chowdhury R. Abrar is a Professor of International Relations and theExecutive Director of the Refugee and Migratory Movements ResearchUnit, University of Dhaka, Bangladesh. He has been working on issuesrelating to refugees, statelessness and labour migration. With JanetSeeley, he co-edited the book Social Protection: Marginalised Migrants ofSouth Asia. He is currently involved in research on labour recruitmentindustry in Bangladesh.

Johanna Avato is an Economist consulting for the World Bank’s HumanDevelopment Network for Social Protection and Labor (HDNSP) in thefield of international migration. She works primarily on migration andskills, and on the portability of social security benefits for migrants.Before joining the World Bank, she was a visiting scholar at the Institutefor the Study of International Migration at Georgetown University. Sheholds a PhD in economics from the University of Tübingen, Germany.

Paolo Boccagni, PhD, is a Research Fellow at the University ofTrento. His main areas of work are migration studies, social policy andtransnational studies. He has recently analysed patterns of immigrantintegration and transnationalism of Latin American migrants in Europethrough a combination of different methods, including policy analysis,surveys, and ethnographical and biographical analysis.

Tendayi Bloom is a PhD student in the School of Law at Queen MaryUniversity of London, and Research Associate in the Centre for the Studyof Migration, also at Queen Mary. Her research looks at ways of explor-ing state obligations towards territorially present non-citizens, and herpublished papers examine the interactions between theories of justiceand the real-world concerns of migrants.

Rosalind Bragg is Director of the UK charity Maternity Action, whichworks to challenge inequality and promote the well-being of all preg-nant women, new mothers and their families. She has worked on issuesof social exclusion and health inequality in voluntary organizationsin the UK and Australia. Rosalind has a particular interest in socialprotection for vulnerable migrant women.

xvi

Notes on Contributors xvii

Rayah Feldman is Honorary Visiting Fellow at London South BankUniversity, where she was formerly a Senior Lecturer in Social Sciences.She is now a freelance researcher with a special interest in refugee andwomen’s health. She has campaigned for women’s and migrants’ rightsfor many years and is the chair of Hackney Migrant Centre, a drop-in centre for asylum seekers and other new migrants in Hackney, eastLondon.

Johannes Koettl is an Economist at the Human Development Depart-ment of the World Bank’s Europe and Central Asia unit. He has beenworking on issues related to labour markets, migration, health andsocial protection at the World Bank since 2004. Previously, he com-pleted an MA in International Relations at Johns Hopkins University’sSchool of Advanced International Studies and a PhD in Economics atthe University of Vienna.

Ian MacAuslan has been a Consultant in Oxford Policy Management’sPoverty Reduction and Social Protection portfolio since 2007. Ian con-ducts research and gives policy advice on the design, implementation,evaluation and politics of social protection programmes for vulnerablepopulations. He works across Asia and Africa, but mainly in India andKenya, for governments, NGOs and international organizations.

Marius Olivier is Director of the International Institute for Social Lawand Policy (IISLP), an independent institute which has as its objectivethe rendering of in-depth and high-quality research, and advisory, pol-icy and capacity-building services in the areas of social law and policy.Marius has been specializing in labour law and social security for severalyears, and has, in particular, been doing extensive work on the socialsecurity position of non-citizens, also in South and southern Africa. Heholds adjunct professorial positions with universities in Australia andSouth Africa and is the chair of the SADC Core Group of Social SecuritySpecialists.

Mythri Prasad-Aleyamma is a PhD student at the Centre for Develop-ment Studies, Thiruvananthapuram, Kerala. Her areas of interest includemigration, city studies and labour geography. Her doctoral thesis is onmigration, settlement and urban transformation in Kerala, focusing onworkers who have migrated from north and north-eastern Indian states.Her publications include ‘Spatial Strategies and Sojourner Workers: Map-ping Roadwork in Kerala’ in C. R. Abrar and Janet Seeley (eds) Social

xviii Notes on Contributors

Protection and Livelihoods: Marginalised Migrants of South Asia and ‘Mobil-ity, Migrancy and Globalisation: City-spaces in Kerala’, in Antipode:A Radical Journal of Geography. She also dabbles in poetry and theatre.

Rachel Sabates-Wheeler is a Development Economist with over 15years of experience working on the themes of rural development, socialprotection and migration. Rachel is Research Fellow at the Institute ofDevelopment Studies (IDS), and has been the Director of the Centre forSocial Protection since 2006. She is also a Senior Researcher with theMigration, Globalization and Poverty Research Centre, Sussex, and hasled a large research programme on social protection and migration. Herkey publications in the field of social protection include: TransformativeSocial Protection: The Currency of Social Justice and IDS Bulletin DebatingSocial Protection. She has published other articles on this theme in AppliedEconomics and Food Policy. Rachel’s work on migration has informed theHuman Development Report 2009, the World Migration Report, andhas been published in journals including World Development, EuropeanJournal of Development Research and Development.

Part 1

Frameworks

This page intentionally left blank

Introduction: Mapping MigrantWelfare onto Social ProvisioningRachel Sabates-Wheeler and Rayah Feldman

1. Purpose of the Book

International migration of all kinds has grown significantly since themid-twentieth century. It has become an important element of glob-alization, with political, economic and social implications for bothsending and receiving countries. Inevitably there has been a concomi-tant increase in scholarly attempts to understand and theorize migratoryprocesses.

Recent studies of migration have recognized the complexity andvariability of migratory flows as well as the political and economicfactors which shape patterns of migration. There has also been a grow-ing interest in how social networks of family and community operateamong migrants, linking both migrants and others in and across send-ing and receiving countries (Castles and Miller, 2003, p. 27; Gardnerand Ahmed, 2006). A relatively new feature of migration studies is aninterest in the ways in which migrants, through their own agency, areconstructing new forms of ideological, cultural, economic and political‘transnationalism’, which shape their relationships with both sendingand receiving countries (Vertovec, 2004). For some commentators thisincludes the creation and development of parallel institutions, such asunregistered labour recruitment and travel agencies, which constitutepart of the contemporary ‘migration industry’, and are responses to gov-ernment’s migrant control policies and migrants’ strategies to counterthem (Broeders and Engbersen, 2007).

These recent studies focus on migrants’ agency in contrast with otherswhich draw attention to the social inequality characterizing relation-ships between migrants and indigenous or settled populations, andwhich focus primarily on migrants’ vulnerabilities (Kothari, 2002; Abrar

3

4 Frameworks

and Seeley, 2009). Discussions of irregular migration and welfare, inparticular, draw on a tradition of urban studies which stresses the polar-ization of ‘global’ cities. This approach emphasizes the exclusion ofcertain groups, notably migrants and some settled ethnic minorities,from social benefits and other rights normally available to citizens inmodern advanced capitalist states (e.g. Freeman, 1986 and Schierupet al., 2006). Interwoven with these concerns is a recognition of thediversity of forms and goals of migration both between and within givenmigrant populations, and a more intermittent acknowledgment of thegender dimensions of migration.

Taken together, all these considerations present a formidable task forresearch and analysis. Analysis is made more difficult by the cat-and-mouse nature of contemporary migration policy continually pittingmore stringent immigration controls against migrants’ efforts to bypassthem. This process is especially characteristic of Europe but is also, assome of the chapters in this book show, a feature of migration in otherregions, and even internal migration in federal states such as India (forexample MacAuslan, Chapter 7). One writer has suggested that certainforms of illegal migration are ‘unanalysable’ as they represent constantlymutating situational responses to unpredictable economic and politicalconstraints imposed by states and markets (Bade, 2004, p. 355).

The plethora of rules and restrictions surrounding entitlements tosocial rights means that migrants often find themselves excluded fromor in ill-defined relationships with states, markets and mainstreamsocial institutions. This results in political and economic forms of dis-advantage, including restricted access to ‘normal’ welfare provision.These formal processes reinforce a range of barriers to welfare, suchas entitlement confusion, migrant-specific vulnerabilities, informationalblockages and bureaucratic hurdles. Migrants’ access to formal systemsof social protection is, therefore, dependent on political decisions aboutthe social rights of non-citizens, and on how far welfare is used as ameans of controlling migration. Migrants may choose or be forced notto engage with formal provisioning systems; instead they may go with-out social protection or may attempt to access it through alternativeroutes and institutions.

A number of the case studies in this book explore the tension betweenthe idea of social provisioning as a risk insurance and poverty reductionstrategy and the reality of migrants’ situations and experience (for exam-ple Bragg and Feldman, Chapter 5; MacAuslan, Chapter 7; Boccagni,Chapter 8). This involves both examining how governments use welfareas a means of regulating migration, and exploring what specific forms

Rachel Sabates-Wheeler and Rayah Feldman 5

and arrangements for social protection might improve the welfare ofmigrants. In compiling this volume we recognized that the contestednature of migrants’ rights means that migrant-specific needs for socialprotection cannot be considered without a parallel examination of therelevant state-specific policies and practices.

In particular, various contributions to this book:

• identify failures and inequalities of social protection and welfareprovision in relation to migrants;

• explore problems of migrants’ access to social protection;• consider what constitutes appropriate social protection for migrants;• illustrate social protection in practice in different migration situa-

tions;• examine the contradictions between ideas of global rights to social

protection and exclusion from welfare as a strategy for immigrationcontrol.

In this Introduction we briefly discuss the main conceptual and policyconcerns that arise in the varied contributions to this book. They aredifficult to disentangle in practice, but for the purpose of exposition wedistinguish between: (1) forms and processes of migration; (2) migrants’vulnerabilities and increasing insecurities; (3) approaches to welfare, cit-izenship and social protection for migrants; and (4) issues of accessand the politics of migration policies. We then go on to outline theproceeding chapters of the book.

2. Migrant classification and migratory flows

At the outset we must distinguish between different types of migrantsand different types of migration. It is important to question simplis-tic definitions of migrants because different categories can carry verydifferent rights, entitlements and associated exclusions.

2.1. Classifying migrants

Distinguishing between different types of migrants and migration isnot as simple as it seems. An apparently simple distinction betweenforced and voluntary migration may founder on a change of personal orgeo-political circumstances. For example, a student or visitor may findthey cannot return to their home country because of political changeswhich have taken place during their absence. A woman who voluntar-ily migrated to get married may have had to leave a violent partner but

6 Frameworks

be unable to return to her home country. On the other hand, membersof minority groups such as the Roma, fleeing ethnic discrimination andpersecution in some of the A10 European Union accession countries,may have been asylum seekers in an EU country before 2004. On 1 Maythat year, when their own country became a full member of EU, thesame people would have lost their asylum-seeking status and becomevoluntary migrants with full rights of work, residence and movement,as well as limited welfare rights in the receiving EU country.

Typically, voluntary migration is perceived as economic or labourmigration, but forced migrants also have to work, unless they are enti-tled to receive state support. Depending on their stage in the process ofhaving a claim for protection accepted, and on the rules of the countryin which they find themselves, they may be working legally or illegally,but either way they are by definition migrant workers as well as refugeesor asylum seekers.

Further complicating any distinction between forced and voluntarymigration is the definition of what constitutes ‘forced’. People whomove in response to violent conflict or ethnic cleansing would appear tosit relatively securely in the ‘forced’ category. However, whether the cat-egory should include migration as a response to grinding poverty or fearof hunger and family food security has been the subject of recent debate(Gibney, 2009). Many poor people migrate as a risk reduction or liveli-hood promotion strategy due to limited or inadequate opportunitiesat home. In reality there may be multiple reasons for migration; how-ever, migrants typically become categorized within a forced–voluntaryspectrum for immigration and entitlement purposes.

Migration is often classified by whether settlement is permanent ortemporary. However, again, the distinction may be spurious. People orig-inally seeking to work for a period in the receiving country and to savefor a specific purpose before returning to their home country often staylonger than planned. This may be because they do not earn enoughto go back, or because they find partners in the receiving country, orarrange for spouses and families at home to join them. The question ofpermanent or temporary residence is as much one of subjective mean-ing for the migrant as of conceptual accuracy for the researcher. Theconfusion and construction of that meaning is expressed in the con-cept of the ‘myth of return’ (for example Zetter, 1999). With improvedtelecommunications and cheap transport, the categories of temporaryor permanent settlement in any case may have become too narrow. Inan attempt to expand them, Vertovec (2004) explores the implicationsof transnational migration in terms of dual orientation, identities, social


practices and material exchange, which may be fuelled by visits in bothdirections.

There has been much attention in recent years to the ‘legality’ ofdifferent categories of migrants, given the increasing range of legalrestrictions placed on non-citizens and legal meanings of ‘residence’. Inpractice legal and illegal migrants are not clearly distinguishable fromeach other, as people may be both ‘legal’ in one respect and ‘illegal’in another (Düvell, 2006). According to Migration Work and Migrants’Rights Network (2009, p. 3), people are characterized as ‘irregular’ orundocumented if they ‘are not complying with some aspect of immigra-tion law and rules’. These authors prefer the term ‘irregular’ to ‘illegal’,on the grounds that the migrant’s offence is administrative rather thancriminal. Irregular migrants commonly enter the country legally as stu-dents, tourists or workers, and then either stay beyond the authorizedlength of stay, work without a permit, or breach other conditions oftheir visa. Some enter illegally using false documents or by making falseclaims.

Many irregular migrants are hidden from the rest of the populationbecause they fear being made known to the authorities and because theirtwilight existence keeps them apart from the formal agencies and insti-tutions of the state. In addition they are frequently very poor, workinglong hours in unregulated conditions for very low wages. Their livesare often mediated by parallel institutions through which they can findwork, housing, advice, and even health services. Depending on the par-ticular nature of their legal ‘irregularity’ they may engage with some orno state agencies and institutions and receive some state services withor without formal entitlement. In the UK it is relatively easy for childrenof irregular migrants to become enrolled in school whatever the immi-gration status of their parents, but less easy for the family to receive freehealth care. Whether they can engage with the tax and benefit systemis variable and may depend on whether they have National Insurancenumbers, or are working using false documents or without documents.They are, therefore, often without the safety-nets that are regarded asthe basic minimum of state welfare systems in industrial societies.

It is partly in response to a desire to understand the nature of theseapparently ‘hidden’ communities of irregular migrants, that ‘micro-structures’ of migratory processes and migrant experience are beingincreasingly studied and acknowledged. Earlier studies of migrationdemonstrated the importance of social ties and networks for helpingboth international and rural-urban migrants within countries to orga-nize their journeys, find accommodation and work, and adapt to new

8 Frameworks

environments (for example Boyd, 1989). Such networks are often pre-sented as undifferentiated ‘informal’ arrangements. Yet, within thesenetworks, class, ethnicity, education and gender are among the keyfactors determining how people can differentially utilize and benefitfrom them. Some recent studies have explored migrant networks indetail, showing, for example, the way that migrant residence patternsin a big European city are shaped by networks that vary according todifferent patterns of ‘ethnic incorporation’ or by interaction withindifferent ethnic communities (Engbersen et al., 2006). Gardner andAhmed show how highly structured hierarchies of relationships betweentransnational and non-migrant individuals and households in a vil-lage in Bangladesh determine the unequal distribution and conditionsof social protection for the local poor. The amount, if any, of assis-tance that the local poor may receive from wealthier transnationalmigrants is highly dependent on the closeness of the kinship tie betweenthem, or on other specific personal patronage (Gardner and Ahmed,2006).

The specificities of kin, community and transnational networks needto be further explored, taking into account how they reshape themselvesin the face of the constantly changing context of migration policiesand migrant lives (Broeders and Engbersen, 2007). Migrant communitiesand networks, rather than being seen as simply informal forms of wel-fare and protection, should be understood as social systems embodyingrules and power relations, and responsively and creatively reconstitutingthemselves. Such systems may indeed operate outside the framework ofstate or even international laws, and are more likely to do so as laws andrules in developed countries increasingly restrict the rights of interna-tional migrants to enter, work and utilize services in them. Furthermore,existing studies suggest that migrants’ dependence on extra-legal sys-tems are likely to increase the precariousness of their situation and mayindeed force more migrants into criminal activity.

2.2. Migratory flows

The most recent Human Development Report (HDR) on ‘Human Mobil-ity and Development’ (2009) estimates the total global migrant stock at214 million people, approximately 3 per cent of the world’s population.This represents a substantial increase since 1960, where the stock was 75million, and even more recently 2005, where estimates indicated thatthe total migrant stock was 191 million. Furthermore, it is expected thatif the migrant population continues to increase at a similar pace then


by 2050 there will be a stock of around 235–415 million (40 per centhigher than now1). In part this reflects the increasing world population,estimated to increase to between 7.8 and 11.9 billion by 2050 – morepeople translate to more migrants and non-migrants. However, projec-tions of increasing migration flows also reflect larger demographic shiftsdue to climate change, urbanization, conflict and opportunities.

In the last two decades migration literature has burgeoned but it hasmainly been focused on South–North and North–North migration flows.‘South-South migration is not new. It is just ignored by economists’and many others (Hatton and Williamson, 2002).2 The extent of migra-tion between developing countries, and issues surrounding it, remainpoorly understood, largely because data on migration is sparse and unre-liable. Ratha and Shaw (2007), using recent data to estimate bilateralmigration stocks, provide some hypotheses of the determinants andsocio-economic implications of South–South migration. By a rough esti-mate, in 2005 two out of every five migrants on the globe – some78 million out of 191 million migrants – were residing in a devel-oping country.3 Most of these migrants are likely to have come fromother developing countries. Ratha and Shaw estimate that South–Southmigration is nearly as large as South–North migration.

Almost 80 per cent of South–South migration is estimated to takeplace between countries with contiguous borders, and mostly appearsto occur between countries with relatively small differences in income.In Africa, for example, more emigration (relative to population) occursfrom low-income than middle-income countries, and most emigrantsfrom low-income countries go to neighbouring countries, unlike emi-grants from middle-income countries, more of whom go to industrialcountries (Lucas, 2005). Given the size of these estimates it is clear thatSouth–South migration is economically important.

World Bank research (2005) reports that income, proximity and net-works are the major drivers of migration from developing to industrialcountries.4 As South–South income differentials are relatively modest,proximity and networks are likely to have a proportionally greaterimpact. Unlike South–North migration, available data indicate that dif-ferences in average country income play a limited role in South–Southmigration. About 38 per cent of identified South–South migrants comefrom countries with higher incomes than their host country (Rathaand Shaw, 2007). And of the migrants from lower-income countries,most go to countries with incomes only slightly above that of theirhome country. Only about 20 per cent of identified South–South migra-tion from lower- to higher-income countries is to countries with very

10 Frameworks

large differences in income. Several factors are likely to explain themuted impact of country-income differences on South–South migra-tion. Income differences among developing countries are, of course,much smaller than between developing and industrial countries. Lackof resources, limited access to travel documents and the location ofnetworks may limit migrants’ ability to travel far beyond neighbouringcountries, where income differences are often low.

Ratha and Shaw (2007) further find that migrants who travel to otherdeveloping countries enjoy much lower increases in income, are morelikely to be irregular migrants, are subject to greater risks of exploita-tion and are more likely to be expelled than those who migrate fromdeveloping countries to industrial countries. While the economic ben-efits of South–South migration are limited in absolute terms, giventhat migration enables most migrants to move out of poverty, evensmall increases in income can have very substantial welfare implica-tions. Short-distance migration between developing countries has someimportant features, specifically, dense social networks, high prevalenceof irregular migration, repeat migration flows. Several chapters in thisbook explore what these differences imply for migrant welfare and socialprotection systems (Chapters 3 and 9).

3. Migrants’ vulnerabilities and increasing insecurities

Social protection and welfare provisioning are fundamentally abouthelping people manage the negative effects of risk as well as to facili-tate positive improvements in livelihoods. Clearly, migrants face manyof the same risks and hazards as anyone else in the population, how-ever, they also face migrant-specific risks, such as those related to theirdocumented status. Furthermore, some risks are intensified for migrantsdue to the fact that migrants may be disproportionally located in ghet-tos or in sectors where work practices are unsafe. Moreover, publicpolicy may create some special vulnerabilities for migrants, such asexclusion from access to welfare services. For instance, in the case ofmigrants living in urban informal settlements, the source of vulnera-bility stems from spatial factors,5 whereas predominantly socio-politicalfactors underlie the creation and maintenance of the informal sector.In this case vulnerabilities are intensified for migrants based on thefact that they are over-represented in these sectors as a proportionof the population. These types of vulnerabilities can be referred to asmigrant-intensified vulnerabilities as opposed to migrant-specific vulnerabil-ities. Migrant-specific disadvantage applies by virtue of having migrated.


Migrant-intensified disadvantage occurs when a negative outcome isintensified for a migrant.

Concerns for migrants’ welfare can emerge at all stages of a migrationprocess as different vulnerabilities characterize the ‘deciding migrant’,the ‘mobile migrant’, the ‘arrived migrant’, the ‘returned migrant’ andthe migrant’s family that may remain at home. Disadvantages or vul-nerabilities specific to migrant status define and shape the parametersof outcomes for migrants, such as social welfare, security, dignity, asso-ciational participation and inclusion.6 Furthermore, vulnerability andinsecurity has implications for downward spirals into poverty. In ananalysis of 27 studies from developed and developing countries, Yaqubfinds that economic insecurity, understood as the exposure to tran-sitory fluctuations in income, is one of three key factors explainingdownward mobility among households. Similarly, in a review of lit-erature on vulnerability, Dercon asserts that ‘persistence in poverty iscaused by the presence of risk and its consequences’ (Dercon, 2001,p. 11). It is clear then that from the perspective of planning forthe protection of migrants’ rights and poverty reduction, there is astrong need to implement social protection policies, to protect themigrants against the adverse economic and social consequences of theirvulnerability.

Locating migration within social protection literature is interesting asit can be conceptualized in a variety of ways depending upon the unit ofanalysis. Conceptualizing vulnerability as a characteristic of the migrantindividual or group we can focus on the migrant’s access to social pro-tection during and after the migration process.7 We can also think aboutthe changing social protection needs of the migrant’s family, who mayremain in the origin location. Alternatively we can think about the fun-damental causes of vulnerability that affect migrants and families ofmigrants. In this book we highlight four categories of vulnerabilitiesthat migrants are especially prone to: temporal, spatial, socio-culturaland socio-political.

Temporal determinants of disadvantage factor largely in migrants’lives. In a static sense the migrant faces different vulnerabilities asso-ciated with different points in the migration process (migrants intransit, migrants at destination, and the migrant’s family at source).In a dynamic sense the temporal vulnerabilities of a migrant and thefamily of the migrant at the source are nuanced by the length of migra-tion (temporary, seasonal, long-term, daily, temporary, lifetime). Wehave not elaborated on this second dimension of temporal disadvan-tage here (see Kothari, 2002 for a typology of migration time-spans).

12 Frameworks

Intergenerational vulnerability is another type of temporal vulnerabil-ity (for instance Zohry, 2002, for Egypt). Table I.1 below illustrates howspatial, socio-political and socio-cultural determinants of disadvantageinteract with temporal determinants of vulnerability for migrants.

Table I.1 Determinant of disadvantage for migrants

Determinants ofvulnerability

Temporal: stages of the migration process

Spatial/environmental

Transit:Environmental hazards, especially in the case ofillegal migrants. Risks associated with dangerousclimate/geography/remoteness from points of help;health risks (HIV/AIDS).

Destination:Relocation constraints: unfamiliarity withsurroundings (especially international migrants);problems in acquiring adequate housing. Forillegal migrants: constant need to remain hiddenand ‘unidentified’ leads to exclusion fromlivelihood-promoting possibilities. Hazards associatedwith work environment; dirty, dangerous, demeaningwork. Health risks. ‘Ghettoization’ of immigrantcommunities (e.g. urban informal settlements).

Family at Source:Remoteness from main income earner. Possibledecline into poverty over the longer term ifremittances are not forthcoming.

Socio-political Transit:Exploitation; lack of legal representation (illegal); lackof legal protection.

Destination:Lack of representation (illegal); lack of rights to formalinstitutions due to restrictive legislation; lack of accessbased on discrimination; exploitation; more prone toinjustices; discrimination and disadvantage of somegroups in relation to education, social services andeconomic opportunities; unequal distribution ofresources; denied participation in political life; lack ofrights due to illegal status.

Family at source:Lack of representation and voice in the absence ofthe main income earners (often male); limiteddecision-making power if resources are in migrant’sname.


Socio-cultural Transit:Social exclusion based on ethnicity or illegal status;cultural devaluation disadvantage (Kabeer, 2000, p. 6);isolation; marginalization; exclusion from participationin social life. Exclusion of certain groups more thanothers (women, children, elderly); harassment.

Destination:(as above); language and cultural barriers, especially inthe case of international migration.

Family at source:Children may drop out of school in response to the dropin family labour; the elderly may not obtain as muchcare; burden on family members’ time at source is likelyto increase.

A major factor causing vulnerability for migrants is spatial dislocationassociated with mobility. In transit, migrants may be ‘remote’ in termsof geography and in terms of access to basic services such as healthand education. A large number of illegal migrants are vulnerable tohealth problems because of inhospitable terrain and isolation duringtransit. They are also vulnerable to exploitation and poverty due to theirspatial dislocation from economic and social opportunities. This latterpoint also holds for many migrants at their destination, especially illegalmigrants. Spatial dislocation, and the remoteness of transit migrationsin particular, make it very difficult for the government to provide for-mal social protection schemes. It is more often the case that migrantsuse informal social protection mechanisms to help manage risk duringtransit. Many migrants rely on informal social protection mechanisms,such as migrant networks, to enable them to better manage the spa-tial dislocation from source to destination. Spatial vulnerability alsointeracts with vulnerabilities related to environmental hazards.

Socio-political determinants of vulnerability refer to the institutionalconstraints facing migrants and typically reflect the lack of politicalcommitment from the destination government/society to the migrant.This type of vulnerability is especially pronounced for illegal migrants,who by definition are excluded from participation in political life, fromaccess to legal institutions, and also from social and economic benefits.The exclusionary processes resulting from this determinant of vulner-ability often causes the illegal migrant to become further marginalizedover time and prone to exploitation and discrimination, leading to aspiral into poverty. Furthermore, governments may actively discourage

14 Frameworks

legalization of certain categories of migrants so that the economy is ableto benefit from access to cheap (exploited) labour.

Socio-cultural determinants of vulnerability for migrants reflect dif-ferences in the norms, values and customs which constitute localconstructions of the ‘migrant’. These constructions are often interwo-ven with culturally held notions of race, gender and illegality, which canconstrain the nature of migrants’ participation in labour markets (obvi-ously depending on different categories of migrants and demographiccharacteristics). That is, social constraints may militate against their tak-ing up certain occupations. This is likely to hamper migrants’ searchfor employment, especially in situations of recession and unemploy-ment. These socio-cultural constraints can lead migrants to be relativelyexcluded from access to public goods, such as health and education, andfrom civic engagement.

Clearly, at any point in the migration process a migrant can beaffected by one or more forms of disadvantage. Furthermore, the formsmay reinforce each other. All these features point to substantial unmetneeds for social provisioning, and at the same time pose significant chal-lenges to the effective provision of both formal and informal protectionto vulnerable migrants.

4. Approaches to welfare, citizenship and socialprotection for migrants

How we should think about migrants and social protection is the cen-tral concern of this book. From one point of view, we need to examinethe kinds of protection migrants require in relation to sickness, socialsecurity, education, pensions and so on, which may differ from thatof resident populations. We also need to examine who delivers andfunds such protection, and how migrants can claim eligibility outsidethe traditional framework of the nation-state as provider or regulatorof services to its citizens. From this point of view, social protection is aquestion of meeting individual welfare needs and the rights of people tohave these needs met, whatever their citizenship or residence status.

At the same time, social protection has become a key concept in devel-opment policy. In response to the negative social impact of structuraladjustment policies, the idea of social protection has changed frombeing simply an underlying concept of welfare to become a distinctpolicy agenda, driven by both national and international players (inter-national agencies, donors, NGOs and national governments) to addressboth welfare deficits and to facilitate growth (in terms of income and


livelihood security) for lower-income countries. Since the 1990s thescope and ambitions of social protection policies have expanded dra-matically, beyond the focus on safety nets, to include activities to helppoor people manage stresses and shocks, in order to indirectly stimulatelivelihood development and growth (by stimulating asset accumulationand reducing inefficiencies from risk-averse resource use) as well as todirectly protect these people’s welfare and livelihoods. Some welfare the-orists make a similar case for the evolution of welfare philosophy in theWest. Taylor-Gooby and colleagues (2004) make the case that Europehas seen a significant change in social welfare emphasis, with social wel-fare becoming increasingly framed as the handmaiden to growth ratherthan vice versa, as evidenced in initiatives such as welfare-to-work andwork-for-pay.

In the following pages we explore these different approaches to socialprotection.

4.1. Migrant statuses and welfare systems

Much of the literature about contemporary welfare systems in developedcountries pays little attention to the implications of increasing migra-tion on welfare provision and eligibilities (Esping-Andersen, 1990),with some exceptions (for example Ginsberg, 1992; Schierup et al.,2006; Brochmann and Hammar, 1999). At the same time, studies ofmigrants often accept as a given the limited social protection to whichthey are entitled, and explore the ways in which different groups ofmigrants negotiate their lives in the face of restrictions on a wholerange of social, economic and political rights (for example Castles andMiller, 2003, Düvell, 2006, among others). Yet the combination of pres-sures on welfare expenditure and growing inequalities between migrantand indigenous populations requires an exploration of the interactionof migration with welfare systems, and how, in this context, socialprotection for migrants can be achieved.

Inevitably, the scale of welfare provision, and its place in the poli-tics and political consciousness of different nation-states, means thatmigrants are necessarily positioned within an existing context in whichsocial protection or welfare addresses different groups or statuses in a dif-ferential way. We have already noted that migrants may have varied andvariable legal statuses. The other side of the coin is that they bring theirlegal statuses into situations in which these statuses are differentiallyregarded, depending on their particular host nation’s system of welfareprovision and rules for dealing with migrants of different statuses. These

16 Frameworks

systems are themselves dynamic, responding to particular political andsocial forces, and limited by the existing institutional structures whichframe them (Schierup et al., 2006; Esping-Andersen, 1990; Bommes andGeddes, 2000, among others).

Although each country is unique in how it treats migrants, it is pos-sible to use some of the conceptual apparatus of welfare regime analysis(Esping-Andersen, 1990) to explore the relationship between welfareand subordinated ethnic minorities, including migrants. We are not hereconcerned with typologies of such regimes, which in any case have beencriticized by numerous writers for failing to deal with more than a nar-row range of regime types (for example Guillén and Alvarez, 2002 andWood and Gough, 2006). Other writers have also sought to identify fea-tures of welfare regimes in developing countries in order to render thetypologies more meaningful (for example Davis, 2001 and Abu Sharkhand Gough, 2010). Here we simply use some of the key concepts whichunderlie welfare regime classification and explore their implications inrelation to migrants.

Firstly, the idea that welfare regimes or systems contribute to spe-cific forms of stratification can be used to examine the ways in whichdifferent categories of migrants may be included or excluded fromprovision (for example Rosenhek, 1999 and Schierup et al., 2006). Sec-ondly, like other writers on welfare systems, Esping-Andersen takes itfor granted that entitlements derive in various ways from citizenship.This obviously raises questions about the entitlements of non-citizens.More importantly though, his recognition that this can be the basis forimportant solidarities can help us understand how the exclusion of non-citizens from welfare may have popular and significant political reso-nance. Thirdly, Orloff (1993) has shown how Esping-Andersen’s originalfocus on decommodification can be extended to include, among otherthings, the role of the family in welfare provision, in order to understandhow welfare systems impact on gender relations. In the case of migrants,we might consider the balance of provision not just between the state,market and family, but also the social networks and parallel systems ofprovision, which both serve migrants, but also reinforce their exclusionfrom the mainstream, and may even create different class solidarities(see Mythri, this volume Chapter 6).

From the point of view of disadvantaged groups such as women andethnic minorities, including migrants, the organization of welfare iscrucial in shaping both their eligibility and access to decommodifiedbenefits, and also in defining and reinforcing their subordinate sta-tuses (Orloff, 1993; Rosenhek, 1999). Rosenhek’s revealing case study


shows how Palestinian Israeli citizens are structurally and institution-ally excluded from access to welfare benefits despite de jure entitlementas citizens. He argues that this is a product not only of Zionist exclusion-ary policies, but also of a tendency for the expansion of welfare statesto be accompanied by the increasing exclusion of subordinated groups,more commonly foreign workers (Freeman, 1986). The weaker welfareposition of Palestinian citizens has made them more dependent on themarket, helping to push them into a more unprotected secondary labourmarket. Rosenhek’s analysis is deliberately about ‘citizens’ rather thanforeign workers, and parallels studies and debates about the subordinatewelfare and labour market position of black Americans and the exclu-sion of ethnic minorities more generally (for example Quadagno, 1994and Lieberman, 1998).

However, if it is possible to impose exclusionary criteria on citizens,how much easier is it to do this with non-citizens? Bommes and Geddes(2000) have suggested that national welfare states act as ‘political fil-ters’ which mediate migrants’ efforts at social inclusion by excludingcertain types of unwanted migration and defining the legal conditionsfor immigration and residence, which are then combined with differen-tiated welfare entitlements. How this is done varies from state to state,but the centrality of how welfare provision can be used as both a toolfor inclusion, granting the rights of social citizenship, and for exclu-sion, indicates the contradictory pressures on contemporary migrants.It also highlights the difficulties facing analysts who seek either to iden-tify the ‘chances for migrant inclusion’ within national welfare statesor to devise systems of social protection addressing migrants’ particularneeds.

In addition, welfare has long been a site of political contestation inrelation to both the importance of decommodification in the protec-tion of workers and the social solidarities which underpin the specificorganization of any welfare system. ‘Social policy helps define the rele-vant boundaries of collective identity because, for workers, it constitutessuch a vital element in their livelihood. Thus the ways in which socialprotection schemes are organized comes to delineate social identities,status communities and solidarities’ (Esping-Andersen, 1990, p. 109).This means that the rules that determine who gets what are often under-pinned by shared histories of collective struggle and entitlement whichhave the function not only of defining eligible groups, but also ofexpressing and legitimizing the criteria of eligibility, and concomitantlyof exclusion. The insight that the organization as well as the quantity orterms of provision of welfare ‘help determine the articulation of social

18 Frameworks

solidarity, divisions of class and status differentiation’ (Esping-Andersen,1990, p. 55) can be used to explore both the popular pressures and therelative ease with which certain groups, including migrants, have beenmarginalized or excluded within systems of welfare.

Where groups are excluded or disadvantaged in relation to welfarerights, concomitant problems may lead to special provision being madefor them. However, such provision may itself be a mechanism for exclu-sion and is often stigmatized (Castles and Davidson, 2000, p. 119). Atthe same time, in response to lack of access, migrants may developalternative strategies for social support, for example through religious orhome-based affiliations, or through involvement in parallel labour mar-kets. Short-distance migrants frequently return to their home countryfor medical treatment, or use unregistered practitioners from their homecountry in the country of settlement. Some services may be obtainedfrom other migrants, often under very poor conditions, for instancelow-paid, unregistered childcare to enable women to take up employ-ment. Clearly, where migrants lack unemployment or health insurance,their situation in sickness or unemployment may be very precariousindeed.

4.2. Social protection

There is a wide variety of views on what factors gave rise to welfare statesand their variability in advanced industrial countries. Some writershave seen welfare as an outcome of national patterns of industrializa-tion, while others explain its development in terms of governments’attempts to pre-empt or subdue working-class struggle. Alternatively, itmay be seen as representing gains from class struggle, or resulting fromthe spread of social democratic ideology. Some explanations involvepermutations of these and other ideas.8

Since the late 1980s, the withdrawal of the state from welfareprovisioning and livelihood stabilization (that is, through market inter-ventions), followed by the massive failure of structural adjustment todeliver on growth in many developing countries, has effectively drawna line under the possibility of establishing comprehensive welfare statesin the short to medium term. Even in developing countries which hadhad very limited public welfare provision, structural adjustment poli-cies since the 1980s effectively cut these back, and limited the scopefor further public provision. Any likelihood of achieving comprehensivewelfare provision in many developing countries is further underminedby: stable patterns of clientalism in many countries; ‘kleptocratic states’;


tiny proportions of tax payers and limited ability to collect taxes; weakgovernance; conflict and insecurity; fragmented and weak civil soci-ety movements; and large numbers of extremely poor people in manyareas. At the same time the welfare vacuum has exposed a huge need forassistance and protection for large parts of the populations.

While concepts of social protection are implicit in conceptions of‘social welfare’ and ‘social policy’ in many Northern states, the termi-nology of ‘social protection’ is dominant in many developing countries,particularly in Latin America, Africa and Asia. A plethora of social pro-tection instruments has increasingly stepped into the near absence ofa welfare state in many countries. In other words, social protectionis largely an instrumentalist agenda to ‘mop-up’ vulnerability in theabsence of a widespread provision. Some commentators believe thatsocial protection activities will facilitate the evolution of established,permanent and predictable welfare regimes in many countries. In thisscenario, donors and governments envisage that the plethora of socialprotection activities will be subsumed under a nationally owned wel-fare system embedded in a social contract between civil society and thestate. Whether this will happen is yet to be seen. Furthermore, socialprotection is not necessarily a perfect ‘welfare’ substitute in the absenceof a comprehensive welfare regime. One reason for this is that the mainframing and implementation of the agenda emphasizes income growthand risk reduction to the neglect of building governance and civil soci-ety maturity. That is, economic protection at the expense of social andpolitical protection. As one of us has argued elsewhere, the dominantframing of social protection has been lacking a transformative element(Sabates-Wheeler and Waite, 2003). This element requires responsesto structural disadvantage to be included within any social protectioninitiative:

Social protection describes all public and private initiatives thatprovide income or consumption transfers to the poor, protect thevulnerable against livelihood risks, and enhance the social statusand rights of the marginalised; with the overall objective of reduc-ing the economic and social vulnerability of poor, vulnerable andmarginalised groups.

(Sabates-Wheeler and Devereux, 2008)

The main elements in this ‘transformative’ approach to social protectionmust be: a progressive agenda to move the vulnerable out of depend-ency and chronic disadvantage; social as well as economic goals for the

20 Frameworks

movement into productive spheres of activity; and a focus on needs,assets and rights as a complementary set of initiatives for supportingsocial protection initiatives. In particular, the approach specifies fourfunctions of social protection and provision: ‘provision’ measures, whichprovide relief from deprivation; preventive measures, which attempt toprevent deprivation; promotive measures, which aim to enhance incomesand capabilities; and transformative measures, which seek to address con-cerns of social justice and exclusion from access to social provision.The transformative element is particularly relevant for considerationsof migrant welfare (see Sabates-Wheeler and Devereux, 2008, for a fulldescription of transformative social protection).

This is a book with a global perspective, at times looking at migrantsliving in the UK under an established welfare state, at times migrantsin Malawi where the welfare state does not exist. For this reason wesometimes use the terminology of welfare states, and, at other times,the language of social protection. Critically, what we are concerned withhere is how migrants construct, negotiate and access welfare in relationto legal, physical, state and global structures. This welfare may at timesbe accessed or negotiated in relation to a welfare state, at times in rela-tion to a plethora of social protection mechanisms within and acrosscountries, at times in relation to clientelistic or altruistic relationshipsin which migrants and their families are involved.

4.3. Social protection and migration

While social protection as an agenda primarily for reducing the vul-nerability and risk of low-income households with regard to basicconsumption and services has become an important part of the devel-opment discourse at both national and international levels, there isvery little literature linking migration to social protection frameworksor policies. The need to manage risk and secure livelihoods can be themain driver of migration decisions. However, at the same time a deriveddemand for various forms of social protection, state and non-state, mayarise from the migration process.

Migrants often live and gain a living outside the parameters ofthe state. At times they strategically choose how to interact withstate provisioning and negotiate other regimes of provision that maybe transnational, cross-border, charity-based or non-formal. However,more often migrants are purposely excluded from welfare systemsand social protection initiatives. Neither welfare regime analysis norsocial protection frameworks attend to the plight of migrants andtheir relation to welfare. This is largely because these framings do not


incorporate institutional, social and political understandings of the bar-riers to welfare provision, which are characteristics of migrant-specificvulnerability.

In concrete terms, social protection for migrants consists of four com-ponents: (i) access to formal social protection – that is, social securityand social services – in host and origin countries; (ii) portability of vestedsocial security rights between host and origin countries; (iii) labourmarket conditions for migrants in host countries and the employmentrecruitment process for migrants in the origin country; and (iv) accessto informal networks to support migrants and their family members.

First, access to formal social protection is crucial for migrants inorder to reduce their vulnerabilities. This includes access to health carebenefits, long-term social security benefits like old-age and disabilitybenefits, and short-term benefits like social assistance, maternity andunemployment benefits, family and housing allowances, as well as pub-lic housing and education. Migrants often do not fully benefit fromthese provisions, either because access is only granted some time afterarrival, if at all, or because family members are spread across variouscountries.

Second, portability of social security rights is important not only tomigrants to avoid financial losses, but also to social security institutionsout of concerns of actuarial fairness. Portability ensures the ability topreserve, maintain and transfer vested social security rights or rights inthe process of being vested, independent of nationality and country ofresidence (Cruz, 2004; Holzman et al., 2005). Portability is particularlyimportant for long-term benefits that have an explicit pre-saving ele-ment (as in the case of old age pensions) or implicit pre-saving element(as in the case of health care). In the absence of portability, migrantsrun the risk of financial loss when leaving their host or home country.For example, migrants contributing to an old age pension scheme intheir host country could lose part of their contributions and benefitswhen returning to their home country. Similarly, migrants contribut-ing to health insurance in their host country could lose coverage whenmoving back to their origin country. They might find it difficult to findaffordable health insurance in their origin country after return, in par-ticular when close to or during retirement. By default, they could end upbenefiting from the origin country’s welfare or health system, althoughthey might have spent most of their productive life working abroad andcontributing to a scheme in another country. This could have importantfiscal implications for social systems in origin countries. Obviously thisscenario does not apply to the large number of non-contributors withincontributory systems.

22 Frameworks

Third, labour market conditions are an essential component of socialprotection for migrants. Migrants are often disadvantaged in foreignlabour markets due to information asymmetries and monopsony powerof employers. National labour market policies are often intended toovercome failures in the labour market, to strike a balance betweenthe needs of employers and the need for protection of workers. Thesepolicies, enacted and implemented on the national level, regulate theworkings of labour markets, that is, hiring and firing conditions, min-imum wages, benefits, and other rights for workers. For migrants,though, who are often recruited in the origin country to work in thelabour market of the host country, many of these regulations might beundermined because of substantial information gaps. These informationgaps can be exploited by employers, recruiters and middlemen, who usethe promise of high earnings to extract unjustified fees from migrants orto offer unfavourable work contracts. Finally, if immigration policies aretoo restrictive vis-à-vis the demand for foreign labour, employers andmigrants have incentives to meet in the informal labour market and thelabour market for undocumented migrants—the latter of which is notsubject to regulation and might provide weak or only informal socialprotection.

Finally, less formal systems of social protection refer to the way thatmigrants manage their risks by relying on a range of kinship or patron-age institutions and local or community-based networks that can act associal safety nets. Such networks and groups are an important comple-ment to formal social protection tools and are of particular importanceto undocumented migrants, for whom they are often the only source ofsupport in times of crisis. They may also be used as sources of informa-tion on how to access indispensable (social) services in the host country.These are more fully explored in Chapters 2 and 9 (MacAuslan andSabates-Wheeler and Sabates-Wheeler respectively).

5. Issues of access and the politics of migration policies

5.1. Physical, legal and political access

As discussed above, some attention has been devoted to the ‘portabil-ity’ of social protection for migrants, such that they are able to movewith their entitlements (see chapter by Avato et al., 2010). This atten-tion to portability necessarily leads to a focus on formal arrangementssuch as bilateral social security agreements, permanent labour contractsand regular migration. While this constitutes extremely important work,


the focus on formal, legal arrangements means that irregular and sea-sonal migrants are typically overlooked. These groups usually work onnon-permanent labour contracts and are largely comprised of the poor-est and most vulnerable migrants. Also overlooked are other types ofportability arrangements emerging from unofficial arrangements andinstitutions (such as burial society payments or hometown associationmembership). While irregular migrants and familial or network-basedarrangements pose great challenges for conducting research, we mustexplore them to understand the constraints for migrants and social pro-tection. Furthermore, where the focus is on international and bilateralarrangements, internal migration may also be overlooked. Of particularinterest is an exploration of how access to social protection provisionsfor migrants interacts with the shape of labour markets in destinationlocations, both internally and internationally.

As well as reflecting a limited view of the nature of migrant labour,current research only investigates one element of a portability story –that of the administrative procedures associated with portability. Furtherelements, related mainly to access, need also to be explored. Ignor-ing the issue of access implies a direct cost for government budgets,as employers and migrants are often able to circumvent tax and socialsecurity payments. Moreover, if access issues are not addressed this mayresult in a cost for the migrant as they are often exploited and unable tospread income and consumption over their life cycle.

Access to social protection for migrants is of crucial concern for allmigrants, regular and irregular. If we are concerned with poverty reduc-tion and inclusive development then access is a prior question to thatof portability. What is needed is an investigation into the constraintsand conditions of access to specific social protection instruments formigrants. How is access to advanced portability regimes in both sendingand receiving countries determined and negotiated? How is the ‘migrantvoice’ represented? Which migrants are represented and who deter-mines this? These questions pivot on issues of rights and empowerment,which in many ways overlap with a transformative social protectionagenda.

The right or entitlement of migrants to social provisions can beexamined through the rules that structure access – in other words thephysical and legal access structures (see MacAuslan and Sabates-Wheeler,Chapter 2). While the analytical focus on eligibility and data rules canbe mundane, the questions raised are crucial in determining access andportability. Specified entitlements and rights are only as useful as theaccess structures that surround them. Much of this turns on the ability

24 Frameworks

of individuals (formally entitled or not) to persuade employers andadministrators of their eligibility for the resource. This approach sug-gests, then, that getting the formal rules right is not enough. A rigorousunderstanding of access must also examine the negotiations, bargain-ing, threatening, pleading that take place around those rules, betweenapplicants, rule-setters, employers and administrators – in other wordsthe political access structures.

The issue of migrants’ access to social protection and its portabilityis irrevocably linked to politics. For instance, the government of Delhihas been wary of issuing permanent ration cards to migrants becausethey are expected to return home, and it was thought that formallyentitling migrants to ration cards would encourage migration, putting agreater strain on Delhi’s stretched public services or increasing migrants’competition with current residents for jobs (see Chapter 7). For similarreasons, governments of in-migration hubs such as the UK and SouthAfrica restrict social benefits to non-citizens (see Chapters 4 and 9). Theissue reflects political pressures within states and rule-setters’ interestsand perceptions about state sovereignty and the sanctity of territo-rial borders, as well as social spending objectives. Do states considermigrants as potential political constituencies? Can they build accessstructures to achieve political goals? On the other hand, do migrantshave an interest in subverting the system? What incentive do migrantshave to engage in ‘proper’ legal and political access, as compared to pur-suing any means to ensure physical access? The discourse around socialprotection entitlements and rights must incorporate these questions intaking the debate on access to welfare forward.

5.2. Regulating access and regulating migration

Migration has become an important and contested political issue inreceiving countries and among supranational organizations, notablythe European Union. This results in frequent policy changes whichaffect the conditions and behaviour of migrants and the institutionsof host countries as well as the attitudes of their citizens and otherexisting residents. Migration policies are a response to contradictorypolitical and economic pressures, with governments attempting to rec-oncile demand for cheap and flexible and/or highly skilled labour, or tomeet Refugee Convention obligations, while pacifying xenophobic hos-tility from some sections of the public and media and demonstratingthat governments are in control.

Recent flows of migration, whether in response to work opportunitiesin a much wider range of industries and countries, or to the upsurge


in flows of refugees following political conflicts and changes in the1980s and 1990s, have provoked increasing amounts of regulatory leg-islation, especially to control entry and to regulate the employmentof non-citizens. Such regulations, which create delays in processing orhave increased migrants’ attempts to evade them, have in turn resultedin the emergence of thousands of ‘undocumented’ or in some sense‘illegal’ migrants throughout Europe and in other migrant receivingcountries, as well as the growth of people smuggling and trafficking. Inthis process, in the ‘self-perpetuating logic behind . . . policing and secu-rity’ (Schierup et al., 2006, p. 67), immigrants have become constructedas existentially illegal. This construction affects all types of migrants,whether originally labour migrants, refugees or even tourists, or familymembers and dependants of earlier migrants.

Traditionally, border controls have served as the primary means ofimmigration control. However, as migrants find ways to counter exist-ing border controls, other strategies are being developed, includingthe use of digital surveillance and the incorporation of internal and‘remote’ bodies in operating control mechanisms at the behest of thestate (Broeders and Engbersen, 2007). These include attempts to blockaccess to travel, through penalties exerted on transporters, and to blockaccess to the labour market by policing employers and work premises,as well as exclusion from access to welfare (Düvell, 2006; MigrationWork and Migrants’ Rights Network, 2009). Hollifield has drawn atten-tion to liberal states’ preference for external border controls rather thanthe limitation of foreigners’ rights within states, which may representan infringement of their civil liberties and be resisted by courts andother civil society organizations (Hollifield, 2000). But the continueddevelopment of such internal controls, including detention and expul-sion, exclusion from work and welfare and redefinitions of citizenship,suggest that such resistance has not always been adequate to hold theground in retaining rights for migrants.

This is obviously of enormous significance in the discussion of socialprotection for migrants, where de facto social exclusion because of defi-ciencies of language, skills, income, and hostility or discrimination bylocal people may or may not be underpinned by legal and administra-tive exclusion. For example the guest worker model of labour migrationis an extreme form of what Schierup and colleagues (2006) call ‘differen-tial exclusion’, where an attempt is made to draw clear rights boundariesbetween the temporary migrants and the citizen population. This wasthe basis of South African apartheid and other temporary labour systemsin many countries until now. Moreover, there are numerous exam-ples of states in several world regions recognizing the illegal status of

26 Frameworks

undocumented migrant workers, for example by allowing employers totemporarily register their illegal workers for a limited period (see, forexample Chalamwong, 2006). This gives the government flexibility tocontrol the flow of cheap labour on terms very different from thosegranted to citizens.

On the other hand, experience in many countries has shown thatgovernment efforts to segregate migrant workers in this way may beineffective, in that settlement and family formation do take placewhatever the obstacles. In addition, such exclusionary measures clearlycontradict policies for social inclusion and cohesion on the part of thestate, and challenge the fundamental objectives of trades unions, healthproviders and minority groups, as well as other campaigners againstracism. They are in conflict with the social rights implied in modernconceptions of citizenship in which contemporary welfare regimes arerooted.

5.3. Social exclusion and citizenship

The identification of welfare with citizenship rights, most famouslynoted by Marshall in 1950, is also at the heart of the concept ofsocial exclusion, denoting the ‘negation of citizenship’ (Schierup et al.,2006, p. 1). Marshall (1950), followed later by Townsend (1979) in hisstudy of poverty in Britain, saw citizenship as including the right andability to participate as a full member of society. But citizenship inMarshall’s terms took for granted the conception of a homogeneousnational community and did not address the question of the rights ofnon-citizens.

Although Marshall attempted to deconstruct the philosophies inher-ent in the concept of citizenship in order to explore their relationshipwith social inequality, he did not problematize the notion of ‘member-ship of a community’ (our emphasis) which underpinned the concept.But it is precisely the question of membership that is problematic in therelationship of migrants to welfare, rather than whether or not citizen-ship implies social rights. In other words, if national citizenship doesimply social rights (and most commentators seem to accept, follow-ing Marshall, that it does), then does non-citizenship mean that suchrights must be denied? It should be remembered that social rights arenot always predicated on membership of a nation-state. One chapterin this volume shows that today such rights may be restricted to stateswithin a federal system as in India (MacAuslan, Chapter 7), and anotherreminds us that under the English Poor Laws, paupers had entitlements


deriving from their residence in a parish rather than, more generically,as fellow citizens (Bragg and Feldman, Chapter 5).

The particular forms of inclusion that states offer to citizens varywidely and constitute part of the basis of categorizations of welfareregimes. However, how far migrants are included depends not only onthe welfare regime, but also on the status accorded to them by the state(Bommes and Geddes, 2000). Schierup and colleagues argue that

welfare policies can lead both to inclusion and to exclusion of immi-grants. Where immigrants were citizens from the outset (such ascolonial immigrants in the UK), welfare systems initially played apositive part in their incorporation into society. Where immigrantshad a legal status that denied them normal welfare rights, they wereexcluded and marginalised.

(2006, p. 25)

But states have also recognized the use of welfare systems as a toolfor managing and controlling migration. How governments choose tomanipulate the welfare system in relation to migration is in princi-ple independent of the type of system operating and more a functionof the balance of political forces in the nation-state in question (seefor example Guiraudon, 2000 and Hollifield, 2000). The beginning ofrestrictions on migration in Western Europe in the early 1970s coin-cided with financial pressures on state spending, and many Europeangovernments found themselves struggling to limit the cost of welfareprovision. But, by then, entitlement to health, education, social secu-rity and public assistance had become a widespread public expectation.In this context the ‘crisis’ of the modern welfare state could be used tojustify the exclusion of migrants from some types of provision.

The position of refugees should be mentioned here. Refugees recog-nized under the 1951 Geneva Convention have the same citizenshiprights as nationals of their host country. This may in part explain thereluctance of governments in receiving countries to grant refugee statusto asylum seekers. The German Basic Law or constitution, following theSecond World War, offered a guarantee to the right to asylum, but as asy-lum applications grew from the late 1970s onwards, both administrativerestrictions and outright hostility to asylum seekers increased, culminat-ing in the enactment of significant legal restrictions on the granting ofasylum by 1993 (Bosswick, 2000). In other places the standard of proofrequired to justify the granting of refugee status also limits the probabil-ity of success of many asylum claims. As a result, ‘asylum seekers’ have

28 Frameworks

become another distinctive administrative category in most states, withminimal ability to claim social rights, and usually without the right towork. Refused asylum seekers, many of whom are unable to return totheir country of origin, often have still fewer rights.

Restrictive changes in nationality laws, which have taken place in anumber of European countries since the 1960s, also reflect governments’attempts to prevent the acquisition of citizenship. The purpose of suchchanges is to enable governments to retain control over entry, and tospecify differential and more limited social rights to those excluded fromcitizenship. During the 1990s successive French governments restrictedand then partially expanded citizenship and the related social rights ofmigrants (Hollifield, 2000). At the time of writing the British govern-ment is proposing much more difficult procedures for the naturalizationof immigrants. These would both explicitly exclude non-citizens fromwelfare, housing and education services and benefits for longer peri-ods than currently, and involve very high costs in acquiring citizenship(Home Office, 2009; Symonds, 2009). This follows several pieces oflegislation which, since 1971, have restricted British nationality.

While citizenship has not always and uniformly been the basis ofaccess to social rights and social protection within states, it has increas-ingly come to be used in this way. As eligibility to welfare has becomemore restricted, so the rules of nationality and the recognition of refugeestatus have narrowed. It is in this context that this book seeks to explorethe conditions of social protection for migrants, including how theydeal with the reduced room for manoeuvre in receiving states by cre-ating alternative routes and methods of social protection, and howsupranational as well as national policies impact on different classes andgroups of migrants.

6. Outline of the book

The book is divided into three parts. The first part covers conceptual andethical debates and frameworks around notions of citizenship, rights,entitlements and access constraints, in relation to social provisioningand migrant welfare. Bloom and Feldman, in Chapter 1, explore someconceptual issues in past and contemporary understandings of citizen-ship, and their implications for non-citizens, especially migrants. Theirchapter presents examples of historical and cross-cultural notions ofstrangers’ rights to protection and welfare outside their own homeland.In different times and places, ideas of hospitality have supplemented theabsence of officially conferred entitlements to non-citizens. The chapter


discusses how the notion of social rights as a fundamental element incontemporary citizenship can serve to justify denying these rights tomigrants who are not citizens. It also explores the international instru-ments designed to protect the rights of refugees and migrant workers,but, nevertheless, the authors find that many migrants, as non-citizens,continue to suffer from a ‘non-status’ amplified by disadvantages of classand race, which serve to exclude them from legitimating grounds forprotection.

In Chapter 2, MacAuslan and Sabates-Wheeler start from the hypoth-esis that specified entitlements and rights are only as useful as the accessstructures that surround them. Successfully claiming social provisionsin practice turns on the way that individuals (formally entitled or not)are able to leverage opportunities for accessing resources. This hypoth-esis implies, then, that getting the formal rules right is not enough.A rigorous understanding of access must examine not only formal rulesof entitlement, but also the distributional mechanisms through whichthat entitlement is provided, as well as the negotiations and bargain-ing that take place around those rules, between claimants, rule-settersand providers (such as employers, administrators, friends and family).Migrants can find social protection from a range of sources: from themarket, from non-market public distribution, and from charity-basedor network-based relations. Using illustrative examples, MacAuslan andSabates-Wheeler develop a conceptual framework for understanding dis-tribution and access for each type of source. This framework is thenapplied to the specific access issues affecting migrants.

Part II of the book includes Chapters 3 and 4 and reviews the globalstatus of social security provisioning for migrants, as well as the regionalpolicy and legal space that defines, as well as obfuscates, migrants’access to welfare. In Chapter 3, Sabates-Wheeler, Koettl and Avato firstillustrate how social security access and provisioning have very specificremits for migrants. Second, they review the ways in which migrantsare able to access (or are constrained from accessing) social securityarrangements, and, in particular, the extent to which social securityis portable across international borders. Third, they provide some rec-ommendations as to how to extend and strengthen social protectionfor migrants. The chapter draws out best practice for providing socialsecurity coverage to migrant workers, but also identifies areas wherebest practice cannot be easily implemented. Olivier (Chapter 4) drawsout the implications of these different flows for regional coordinationefforts that may help to bring about change in the approach to socialprotection across borders. His chapter focuses on rights, regulation and

30 Frameworks

a normative understanding of migrants’ entitlements, in particular inrelation to social protection, and with specific reference to migrants inSouthern Africa-intra-SADC (Southern African Development Commu-nity) migrants. It explores the importance of adopting a rights-based andregulatory approach, addressing the need for appropriate internationalstandards, regional instruments, constitutional frameworks and statu-tory contexts to be aligned, in order to ensure access to social protectionand portability of social security entitlements. The reality is often differ-ent. Limitations in social security law effectively restrict social securityentitlements. In addition, immigration law regulation and status ofmigrant categories is often used to effectively restrict access to and porta-bility of social security entitlements, impacting in particular on vulner-able migrants such as asylum seekers and undocumented migrants.

The last part of the book presents a range of case studies from acrossthe globe that serve to illuminate themes presented in the first two partsof the book. Bragg and Feldman, in Chapter 5, discuss the politicizationof migrants’ access to state health services in Britain, resulting in theexclusion of some of the most vulnerable migrants, despite a historicrhetoric of universalism. They show that this politicization continuesto be a highly contested process, involving complex legal challenges.Complementing this, in Chapter 7, MacAuslan draws on the concep-tual framework presented in Chapter 2 – yet with evidence from theother side of India – to analyse the physical and political access of inter-nal migrants to formal social protection. The case study is that of Indianmigrants’ access to the government of India’s Public Distribution System(PDS), a programme distributing subsidized commodities to all Indianhouseholds. Using fieldwork conducted in Delhi in 2007, the chapterexplores the actual access of migrants and their ability to change the sys-tem to improve their access. The novelty of the work is that it examinesthe interactions between internal migrants and a formal social protec-tion programme. This is also shown in Prasad-Aleyamma’s chapter oninternal migrant workers in Kerala. The parallels between the situationof international and internal migrants described in all the chapters inthis section are striking.

Taking an anthropological approach, Prasad-Aleyamma, in Chapter 6,uses evidence from a Sunday Bengali market for migrant workers inPerumbavoor in Kerala to critique the dichotomous presentation of for-mal and informal social protection. This market, which sprang up inresponse to the specific needs of migrant workers, serves also as a ‘meet-ing point’ of workers. Prasad-Aleyamma posits that the market could be


construed as a site of interaction which ensures a form of social protec-tion. It is a site that is used for accessing both formal and informal socialprotection. In fact, the very existence of a sociality that is deemed ‘infor-mal’ facilitates the protection of more formal rights. At the same time,the ever-changing nature of the migrant population puts limits on theemergence of a right-based formal trade union. Prasad-Aleyamma arguesthat understanding the ‘social’ in welfare provisioning is important forpolicy-making which is required to go beyond the formal and informalcategories.

Widening out the geographical remit of social provisioning, Boccagniprovides a comparative analysis of the sources of social protectionwithin a long-distance migration flow, connecting Ecuador to Europe,with a particular focus on the role of the motherland. In Chapter 8, herehearses the discourse and presents policies emerging in Ecuador withrespect to the transnational protection of the citizens. Drawing on anempirical study of Ecuadorian migration to Italy, Boccagni explores theinteractions and trade-offs between migrant strategies for protection andresources available to them for protection. He asks, to what extent canthe integration between institutional and migrant-led initiatives resultin a ‘transnationalism of social rights’?

In Chapter 9, Sabates-Wheeler explores some hypotheses and asser-tions related to the characteristics of South–South migration using newsurvey data collected from Malawian migrants in and returned fromSouth Africa. The chapter details how migrants use migration as a socialprotection strategy in terms of consumption security for families backhome and income creation, as well as future investment strategies. Italso describes how Malawians in South Africa are able to access for-mal and informal social protection mechanisms, through legitimate andillegitimate routes. The role of networks and the use of networks toachieve a positive change in income from the migration process arehighlighted. The conclusion reports that social networks are vital forobtaining positive outcomes from migration. The final chapter of thebook, by C. R. Abrar, similarly focuses on the informal provisioning andlack of formal social protection to which Bangladeshi migrants to theGulf states have access. The primary fieldwork contained therein illumi-nates the range of vulnerabilities facing migrants at all stages of theirmigration journey – the deciding migrant, the transiting migrants, thearrived migrant and the return migrant. This brings the reader backto the fundamental problems that we have highlighted, here, in theIntroduction.

32 Frameworks

Public debates about international migration in developed countrieshave traditionally failed to address the lacunae in social protection formigrants, often reconstructing social protection as a ‘privilege’ to whichmigrants can only aspire. At the same time, social protection has cometo be seen in development circles as a mechanism to meet the basicneeds of the poorest. The chapters in this book explore different aspectsof the tension between migrants’ need for protection and the practicesand policies which often lead to such protection being denied or, insome cases, made available selectively to certain groups. By taking aglobal perspective, the book hopes to raise awareness of the multifacetedways in which questions of social protection for migrants need to beaddressed.

Notes

This book is both the product of and inspired by a four-year research projecton social protection for migrants at the Development Research Centre on Migra-tion, Globalisation and Poverty at the University of Sussex. The driving visionof the DRC is to promote policies to maximize the benefits and minimize the risksof migration for poor people. All the contributions to this book, whether or nottheir authors were involved in the DRC project, also strive towards this vision.Our hope is that the messages from this book will help clarify issues and encour-age academics, policy-makers and activists working with and for migrants andtheir families, to intensify their efforts to assist migrants to claim their socialrights.

1. See Black et al. (2008).2. Important recent exceptions are papers by Gindling (2009), estimating the

impact of migrants from Nicaragua on labour market outcomes in neigh-bouring Costa Rica, and by Ratha and Shaw (2007), estimating the size ofSouth–South migration and remittances, and discussing key issues related toSouth–South migration.

3. Avato et al. (2010) provide similar estimates.4. Motivations for South–South migration also include seasonal patterns and

flight from ecological disasters or civil conflict. Other motivations, notrecorded in the migration stock data given above, include transit to the Northand petty trade.

5. A further study, which is not detailed here, examining spatial segregation andthe vulnerability of international migrants, is Iosifides and King (1999).

6. Much of this conceptual framework draws on Sabates-Wheeler and Waite(2003).

7. Here ‘migrant’ can refer to the individual or the migrant family.8. For summaries of theories of welfare state development, see, for example,

Ginsburg (1992), Manning (1999) and Esping-Andersen (1990).


References

Abu Sharkh, M. and Gough I. (2010) ‘Global welfare regimes a cluster analysis’,Global Social Policy, 10(1), 27–58.

Abrar, C. R. and Janet Seeley (eds) (2009) Social Protection and Livelihoods:Marginalised Migrants of South Asia (Dhaka: University Press Limited).

Avato, J., J. Koettl and R. Sabates-Wheeler (2010) ‘Social security regimes, globalestimates and good practices: the status of social protection for internationalmigrants’, World Development, 38(4), 455–466.

Bade, K. (2004) ‘Legal and illegal immigration into Europe: experiences andchallenges’, European Review, 12(3), 339–375.

Black, R. et al. (2008) Demographics and Climate Change: Future Trends and their Pol-icy Implications for Migration. Working Paper T-27, Migration Centre, Universityof Sussex.

Bommes, M. and A. Geddes (eds) (2000) Immigration and Welfare: Challenging theBorders of the Welfare State (London: Routledge).

Bosswick, W. (2000) ‘Development of ASYLUM POLICY in Germany’, Journal ofRefugee Studies, 13(I), 43–60.

Boyd, M. (1989) ‘Family and personal networks in international migration:recent developments and new agendas’, International Migration Review, 23(3),638–670.

Brochmann, G. and T. Hammar (eds) (1999) Mechanisms of Immigration Control:A Comparative Analysis of European Regulation Policies (Oxford: Berg).

Broeders, D. and G. Engbersen (2007) ‘The fight against illegal migration: identifi-cation policies and immigrants’ counterstrategies’, American Behavioral Scientist,50(7), 1592–1609.

Castles, S. and A. Davidson (2000) Citizenship and Migration: Globalisation and thePolitics of Belonging (London: Palgrave Macmillan).

Castles S. and M. Miller (2003) The Age of Migration (Basingstoke: PalgraveMacmillan).

Chalamwong, Y. (2006) ‘Country report: Thailand’, paper prepared for the Work-shop on International Migration and Labour Market in Asia, organized by theJapan Institute for Labour Policy and Training, Japan Institute of Labour,Tokyo, 17 February, available at: http://www.jil.go.jp/foreign/event_r/event/documents/2006sopemi/countryreport11.pdf.

Cruz, A. T. (2004) ‘Portability of Benefit Rights in Response to External andInternal Labour Mobility: The Philippine Experience’, paper presented at theInternational Social Security Association (ISSA), Thirteenth Regional Conferencefor Asia and the Pacific in Kuwait, 8–10 March, available at: http://www.issa.int/pdf/kuwait04/2cruz.pdf (accessed 22 October 2008).

Davis, P. (2001) ‘Rethinking the welfare regime approach: the case of Bangladesh’,Global Social Policy, 1(1), 79–107.

Dercon, S. (2001) Assessing Vulnerability, Jesus College and CSAE, Department ofEconomics, Oxford University.

Düvell, F. (2006) ‘Undocumented migration in Europe: a comparative per-spective’, in Düvell, F. (ed.) Illegal Immigration in Europe: Beyond Control?(Basingstoke: Palgrave Macmillan).

Engbersen, G., M. van San and A. Leerkes (2006) ‘A room with a view: irregularimmigrants in the legal capital of the world’, Ethnography, 7(2), 209–242.

34 Frameworks

Esping-Andersen, G. (1990) The Three Worlds of Welfare Capitalism (Princeton, NJ:Princeton University Press).

Freeman, G. P. (1986) ‘Migration and the political economy of the welfare state’,Annals of the American Academy of Political and Social Science, 485, 51–63.

Gardner, K. and Z. Ahmed (2006) ‘Place, Social Protection and Migration inBangladesh: A Londoni Village in Biswanath’, working paper T18, SussexCentre for Migration Research and Jahangirnagar University, Dhaka.

Gibney, M. (2009) ‘Precarious residents: migration control, membership and therights of non-citizens’, Human Development Reports Research Paper 2009/10,April 2009, United Nations Development Programme, available at: http://hdr.undp.org/en/reports/global/hdr2009/papers/ (accessed 12 November 2010).

Gindling, T. H. (2009) ‘South-South migration: the impact of Nicaraguan immi-grants on earnings, inequality and poverty in Costa Rica’, World Development,37(1), 116–126.

Ginsburg, N. (1992) Divisions of Welfare (London: Sage).Guillén, A. and S. Alvarez (2002) ‘Southern European welfare states facing globali-

sation, is there social dumping?’, in Sigg, R. and C. Behrendt (eds) Social Securityin the Global Village (London and Piscataway, NJ: Transaction Publishers),pp. 67–84.

Guiraudon, V. (2000) ‘The Marshallian tryptich reordered: the role of courtsand bureaucracies in furthering migrants’ social rights’, in Bommes, M. andA. Geddes (eds) Immigration and Welfare: Challenging the Borders of the WelfareState (London: Routledge), pp. 71–88.

Hatton, T. and J. Williamson (2002) ‘What fundamentals drive world migration?’,NBER Working Paper Series No. 9159, National Bureau of Economic Research,Cambridge, MA, available at: http://www.nber.org/papers/w9159 (accessed 12November 2010).

Hollifield, J. (2000) ‘Immigration and the politics of rights: the French case incomparative perspective’, in Bommes, M. and A. Geddes (eds) Immigrationand Welfare: Challenging the Borders of the Welfare State (London: Routledge),pp. 109–133.

Holzman, R., J. Koettl and T. Chernetsky (2005) ‘Portability Regimes of Pensionand Health Care Benefits for International Migrants: An Analysis of Issues andGood Practices’, Geneva, Switzerland: Global Commission on InternationalMigration.

Home Office UK Border Agency (2009) Earning the Right to Stay: A New Points Testfor Citizenship (London: UK Border Agency).

Iosifides, T. and King, R. (1999) ‘Socio-spatial dynamics and exclusion of threeimmigrant groups in the Athens conurbation’, in Baldwin-Edwards, M. andArango, J. (eds) Immigrants and the Informal Economy in Southern Europe (UK:Frank Cass Publishers).

Kabeer, N. (2000) ‘Intergenerational contracts, demographic transition and thequantity-quality trade-off: parents, children and investments in the future’,Journal of International Development, 12(4), 463–482.

Kothari, U. (2002) ‘Migration and chronic poverty’, Chronic Poverty ResearchCentre, Working Paper No 16, Institute for Development Policy and Manage-ment, University of Manchester.

Lieberman, R. (1998) Shifting the Color Line: Race and the American Welfare State(London: Harvard University Press).


Lucas, Robert E. B. (2005) ‘Migration and economic development in Africa:A review of the evidence’, paper prepared for the African Economic ResearchConsortium Biannual Research Workshop, Nairobi, May 28–2 June.

Manning, N. (1999) ‘The politics of welfare’, in Baldock, J., N. Manning, S. Miller,and S. Vickerstaff (eds) Social Policy (Oxford: Oxford University Press).

Marshall, T. (1950) Citizenship and Social Class (Cambridge: Cambridge UniversityPress).

Migration Work and Migrants’ Rights Network (2009) Irregular Migrants: TheUrgent Need for a New Approach, London, available at: http://www.migrantsrights.org.uk/downloads/policy_reports/irregularmigrants_fullbooklet.pdf.

Orloff, A. (1993) ‘Gender and the social rights of citizenship: the comparativeanalysis of gender relations and welfare states’, American Sociological Review,58(3), 303–328.

Quadagno, J. (1994) The Color of Welfare (Oxford: Oxford University Press).Ratha, D. and Shaw, W. (2007) ‘South-South remittances and migration’, World

Bank Working Paper No. 102, Washington, D.C.Rosenhek, Z. (1999) ‘The exclusionary logic of the welfare State: Palestinian

citizens in the Israeli welfare state’, International Sociology, 14(2), 195–215.Sabates-Wheeler, R. and S. Devereux (2008) ‘Transformative social protection: the

currency of social justice’, in A. Barrientos and D. Hulme (eds) Social Protectionfor the Poor and Poorest. Risk, Needs and Rights (London: Palgrave Macmillan).

Sabates-Wheeler, R. and M. Waite (2003) ‘Migration and social protection: aconcept paper’, DRC Migration Working Paper T2, University of Sussex, Brighton.

Schierup C.-U., P. Hansen and S. Castles (2006) Migration, Citizenship, and theEuropean Welfare State (Oxford: Oxford University Press).

Symonds, S. (2009) ‘Borders, Citizenship and Immigration Act 2009 – the NewProvisions for Naturalisation’, paper presented to a meeting of Kalungan atPraxis, London, 12 September 2009.

Taylor-Gooby, P. (ed.) (2004) New Risks, New Welfare: The Transformation of theEuropean Welfare State (Oxford: Oxford University Press).

Townsend, P. (1979) Poverty in the United Kingdom (Harmondsworth: Penguin).Vertovec, S. (2004) ‘Migrant transnationalism and modes of transformation’,

International Migration Review, 38(3), 970–1001.Wood, G. and I. Gough (2006) ‘A comparative welfare regime approach to global

social policy’, World Development, 34(10), 1696–1712.World Bank (2005) Global Economic Prospects 2006: Economic Implications of

Remittances and Migration (Washington, D.C.: World Bank).Zetter, R. (1999) ‘Reconceptualizing the myth of return: continuity and transi-

tion amongst the Greek-Cypriot refugees of 1974’, Journal of Refugee Studies,12(1), 1–2.

Zohry, A. (2002) ‘Rural-Urban Migration in Egypt and Other Developing Coun-tries: A Statistical and Literature Review’ from Rural to Urban Labour Migration:A Study of Upper Egyptian Labourers in Cairo, DPhil Thesis, University ofSussex, Brighton.

1Migration and Citizenship: Rightsand ExclusionsTendayi Bloom and Rayah Feldman

1. Introduction

The dominant idea has long been that rights were connected tonationality and citizenship, thereby granting aliens with very lim-ited protection.

—De Guchteneire and Pécoud (2009, p. 6)

In a time of increasing migration, citizenship as a form of classificationhas come to assume the kind of importance once reserved for otherkinds of discriminatory and exclusionary classifications of status. Dis-tinctions in ancient times or in ante-bellum United States between freemen and slaves, in French and Portuguese colonial empires betweenévolués or assimilados and other colonial subjects, in Nazi-occupiedEurope between Aryans and Jews and Roma, or racial classifications inApartheid South Africa, were all means of granting or denying socialand political rights. Although citizenship has many other aspects, formigrants its primary significance is the extent to which it enables themto gain access to a territory and to rights within it.

In the contemporary world, having one’s human rights protected andenforced is usually dependent upon one’s status in a state. The rightsof non-citizens sometimes appear to be legitimately overlooked whenno particular body or state is assigned obligations towards them in theplace where they are living. This can be seen most clearly in the case ofmigrants, both those moving to a country in which they will be non-citizens and those who, in being described as newcomers, are deprivedof full citizenship. Migrants’ rights vary according to the state withinwhich they find themselves, and how they are categorized or classified in

36

Tendayi Bloom and Rayah Feldman 37

those states: for example, whether they are refugees or migrant workers,and according to the state’s current policies about these groups.

International conventions on refugees make it clear that people rec-ognized as refugees should enjoy broadly the same rights as citizens intheir country of refuge. However, the threshold of a grant of refugeestatus is extremely high, and governments often try to avoid theseresponsibilities by requiring unrealistic levels of proof of persecutionor danger. Many more international migrants, however, do not seekrefugee status but travel for the purpose of finding employment. Of theestimated 214 million migrants in the world today, about 16 millionare refugees (United Nations, 2009). Rights of labour migrants havebeen even more contested than those of refugees, and despite attemptsto produce international agreements on their rights, little has beenratified.

This chapter explores the rights of different kinds of migrants in thecontext of contemporary and historical understandings of citizenship.Its purpose is to explore some key assumptions contained in contem-porary uses of the term, especially as they pertain to welfare rights. Byfocusing on these, we try to show that many rights that are taken forgranted are linked to, though not necessarily dependent on, citizenship,and that the absence of citizenship status can enable governments tolimit the rights of non-citizens, particularly migrants.

2. Presence, residence and citizenship

The three terms presence, residence and citizenship are relevant tomigrants. Although they are often taken to be distinct, it is not obviousexactly what the distinctions between them are, nor what these dis-tinctions imply. Prima facie, presence is a momentary descriptor whileresidence happens over time. However, a more precise distinction isneeded because different rights can come with the different statuses.In the UK, presence is necessary and sufficient to receive emergencyhealth care, while residence is needed to obtain education and votingrights. Moreover, residence, particularly ‘permanent residence’, is oftena necessary precursor for obtaining citizenship.

The state can exert coercive power over physically present individu-als in a way it does not and cannot exert over someone, even a citizen,who is not territorially present. Thus a state generally cannot prosecutecitizens for crimes committed in other states or for actions carried outabroad that are not legal in the state of which the person concerned isa citizen. Common examples of such actions include the use of drugs

38 Frameworks

deemed illegal at home, abortion or polygamy. By contrast, in the fieldof international human rights law, the doctrine of universal jurisdictionallows national courts to try cases of the gravest crimes against human-ity, even if these crimes are not committed in the national territory andeven if they are committed by government leaders of other states. Thisprinciple was used in the attempted prosecution of the former Chileandictator, Augusto Pinochet, and campaigners have attempted to invokeit against other alleged war criminals, but such violations of humanrights are normally the province of the International Criminal Courtin The Hague.

Although the state where one has presence is the best placed to pro-vide health and welfare services, in practice legal residence, and evencitizenship, may be required to obtain them, and even then there maybe many restrictions placed on access. States with universal welfare sys-tems, including both Scandinavia and the UK, have sought to limitaccess by restricting eligibility to citizens or those with very specificresidence statuses (Schierup et al., 2006). Linda Bosniak suggests thatborders can be conceptual as well as geographical. ‘The “border” –conceived as regulatory sphere – follows the immigrant into the nationalgeographic space and shapes her experience there’ (Bosniak, 2007,p. 397). Other writers, similarly, describe the welfare state itself as beingused as an aspect of border control (e.g. Cole, 2010, p. 6; Crowley, 2005).Many debates about immigrant rights can be characterized as debatesabout how far into the territory the border can come (Bosniak, 2007,p. 397, n. 25). Access to territory cannot be controlled in the same wayas access to services inside the territory, although the increasing digital-ization of borders makes this less and less true (Broeders and Engbersen,2007).

Residence might indicate a stronger stake in a state than citizenshipin relation to political rights, since citizens may have been absent, evenfor their whole lives, from the state, while residents pay taxes and par-ticipate in society. While this is sometimes recognized in local issues, itis not considered to be the case for national ones. While residency ina local jurisdiction may be necessary for a person to have voting rightsthere, general elections are different. For example, Ecuador has createda ‘Fifth Region’ in its parliament of expatriate citizens, acknowledginga membership separated from any particular geographical area of itsterritory (see Boccagni, Chapter 8). Thus, continued residence, or evenpresence, is not required in order to retain crucial citizenship privileges.Furthermore, although residence may carry with it protections of statusand rights of abode much like citizenship, citizenship is not a merely


ceremonial status. For example, with citizenship come travel documentsand, usually, the assurance that protection will not be removed in thefuture.

Another crucial difference between presence, residence and citizen-ship is that the former two can be irregular. This reinforces them asa non-status, as a mere descriptor of physical location, which in turnreinforces the frail rights of a person whose very physical location isdescribed as ‘irregular’ or ‘illegal’. Taran describes what he calls the‘illegalization’ of migrants, which, he argues,

directly contradicts two fundamental human rights enshrined in theUniversal Declaration of Human Rights: article six which states thatevery human being has a right to recognition before the law; andarticle seven that every person has right to due process.

(Taran, 2000, p. 23)

People may be present against the state’s wishes in two ways: first,through irregular entry or stay past their agreed date of departure; sec-ond, through facts of history and regime change. Some commentatorsargue that giving anything more than the most basic of rights to irreg-ular immigrants undermines the legitimate immigration regime, andit disadvantages those who have sought immigration by legal routes(Miller, 2008, p. 197). By extension, if controlling such immigrationis seen as an important aspect of sovereignty, allocating such rights tomigrants could seem to undermine the sovereignty of the state. How-ever, since the arguments for rights based on presence and on residencesimply refer to physical location, they need not be undermined by theirregular status of the migrant.

Joseph Carens contends that being present without authorizationdoes not have any relevance to the entitlement to basic human rights.

There is a wide range of legal rights that people ought to possess (andnormally do possess) simply by virtue of being within the jurisdic-tion of the state, whether they have permission to be there or notand whether they are obeying the laws or not. We can lump theserights under the heading of basic human rights. These rights are pos-sessed not only by citizens and legal residents but also by tourists andtemporary visitors. Irregular migrants are entitled to them as well.The right to security of one’s person and property is a good example.After all, the police are supposed to protect even irregular migrantsfrom being robbed or killed. People do not forfeit their right to be

40 Frameworks

secure in their persons and their possessions simply by virtue of beingpresent without authorization.

(Carens, 2008, pp. 165–166)

Irregular entry and presence, he argues, are normally not treated ascrimes, but as administrative misdemeanours, and so should not evengive rise to the minimal rights loss of convicted criminals. For thisreason, Carens advocates the institution of a ‘firewall’ around welfareservices, so that basic human rights, and indeed the much wider rangeof rights he would bestow, can be obtained without fear of apprehensionby immigration services.1 However this is to be resolved, it is clear thatof all the groups discussed, irregularly present non-citizens are the mostvulnerable in a wealthy, stable state, since as non-citizens they lack anyautomatic status and as ‘irregulars’ they often lack the ability to accessany legitimizing status for protection.

3. Citizens and non-citizens in the pre-modern world

While one must be careful not to draw unwarranted conclusions fromselective historical and geographical examples, they can show that theuse of citizenship as a political tool has far-reaching antecedents andthat its meaning can change according to circumstances. Even a cursoryhistorical examination of the status of migrants in relation to citizensindicates that both political and ethical considerations have long playeda part in shaping the treatment of migrants and may help illuminatecontemporary understanding of the concept of citizenship. A very briefexploration of citizenship and migration in the classical and medievalworlds also shows the erratic past of a concept which can sometimesseem timeless.

Greek city states started to emerge in around the eighth century BCE,likely influenced by the practices of the Phoenicians, among other fac-tors (Ferguson, 1991, p. 177). In the polis (city state), the institutionof citizenship came to be important, though many inhabitants of citystates and surrounding lands were not citizens. Aristotle explained that,in an ideal state, agricultural workers should be ‘slaves, or non-Greeksdwelling in the country roundabout’ (Aristotle, 1992). This was intendedto enable citizens to be virtuous and to fulfil their citizenly role, whileleaving labour to others. This persists today, where the high standards ofliving in wealthy states are often supported by the work of the poorest,including marginalized foreign labourers. Taran notes:


Migrant labour has long been utilized in developed and under-developed economies as a low cost means of sustaining economicenterprises, and sometimes entire sectors that are only marginallyviable or competitive. For example, migrant labour has been usedin many countries to ensure low cost provision of agricultural pro-duce, to provide domestic services, and to provide services in the ‘sexindustry’.

(Taran, 2000, p. 14)

In ancient Greece and in ancient Rome, some non-citizens receivedspecial protections: those seeking refuge and those who held religiousoffice, as well as traders, heralds and envoys (Phillipson, 1911, pp. 269,271–272; Ferguson, 1991, p. 189). However, in both ancient Greece andRome, non-citizens did not receive legal protection. This has parallelstoday, when good treatment of non-citizens depends on them being ableto show entitlement to some special status, or membership of a state thatis feared or is willing to intervene on their behalf. In ancient times, thiswas partially mitigated by the strong Roman and Greek cultural beliefthat protecting foreigners was required by the gods (Lillich, 1984, p. 21).Weissbrodt argues that this strong cultural protection arose because, inthe absence of official respect for strangers, if it were not for hospitalityforeigners would have been without any protection (Weissbrodt, 2008,p. 19).

In ancient Arab society, mustajir ( ) meant someone who ‘fleeshis land seeking protection and welfare’, regardless of the reason. Inthe early Muslim state, this term came to refer to both Muslim andnon-Muslim asylum seekers and refugees (Shoukri, 2008, p. 101), andit retains a similar meaning today (Faruqi, 1972). The two main under-standings of migrant in Islamic Law, muhajir ( ) and musta’min( ), provide an interesting categorization of migrants. Muhajir refersto anyone that has performed hijrah ( ), that is, left one land foranother, or cut ties with someone or somewhere (Elmadmad, 2008,p. 53). Given its connection to the Mohammedan Hijra, muhajir is oftenused to refer specifically to Muslim migrants, but hijrah seems in theliterature to refer to refugees more widely. Musta’min literally meanssomeone that has received aman ( ), that is, ‘safety, protection, safeconduct’ (Shoukri, 2008, p. 107). Historically, this was used for non-Muslim migrants entering the Muslim world. Today, aman is presentedby one theorist ‘as the official or legal instrument by means of whichprotection is extended, not only to refugees and asylum seekers, but

42 Frameworks

also to foreign traders, diplomats and visiting scholars’ (Al-Rahim, 2008,p. 20). This historically differentiated use of muhajir and musta’min primafacie implies an inherent distinction between Muslim and non-Muslimmigrants akin to the initial exclusively European focus of the 1951Refugee Convention. However, both groups were afforded at least theprotections currently articulated in the 1951 Convention with its 1967Protocol.

A tension between three means of citizenship acquisition can beobserved throughout historical discussions of citizenship. These are:

• ius soli – citizenship acquired through attachment to a place• subjecthood – citizenship acquired through attachment to a ruling

authority• ius sanguinis – citizenship acquired through descent or membership

of particular ethnic groups.

The priority given to different bases or criteria of citizenship has var-ied over time and place. It becomes an issue, as in the contemporaryworld, where there is mass migration, or as empires collapse or arereconstituted.

The colonization and settlement patterns following the First Crusadehighlight the complex relationship between tribal, imperial and local-ity affiliations. For example, the twelfth-century chronicler, Fulcher ofChartres, took part in the First Crusade as a member of the domus(or immediate retinue) of one of the crusade’s leaders, Godfrey ofBouillon. After the crusade Fulcher transferred his loyalties to Baldwinof Boulogne and thereby received a fief in Edessa (overlapping mod-ern Syria, Turkey and Iraq) (Riley-Smith, 1983, 1988). Fulcher wrote,‘we who were Occidentals now have been made Orientals, he who wasa Roman or a Frank is now a Galilean or an inhabitant of Palestine.One who was an inhabitant of Rheims or of Chartres has now beenmade a citizen of Tyre or Antioch’ (Phillipson, 1911, p. 51). This sug-gests that the apparently historical and permanent connection of apeople to a place is not certain or fixed. The affiliations to place thatFulcher describes derived from people whose initial allegiance was toa nobleman who led his followers to set up settlements in the HolyLand. These settlements, in the course of time, became places to whichthey developed an allegiance and from which they gained an identity ascitizens.

Geary has argued that a view of a people based on place stems fromthe Romans, and that it is their rejection of Herodotus’ description of a


Europe of changing and evolving peoples in favour of a world of distinctpeoples tied to geographical place that roots many modern national-ist problems in Europe. The alternative picture Geary paints of ancientEurope in the time of the Greek and Roman empires involves manypeoples moving, jostling, forming alliances, conquering and disappear-ing, with changing customs and dress, speaking many languages andbelonging to a variety of groupings (Geary, 2002, ch. 2).

4. Issues in contemporary citizenship

In the last two decades, interest in citizenship has resurfaced in the faceof large-scale international migration, increasing ethnic diversity withinnation states and the growth of transnational institutions. Citizenshiphas increasingly come to represent social as well as legal and politicalrights, following the classic analysis by Marshall (1950). Yet it is preciselyin the area of social rights that migrants often find themselves deficient.At the same time, the relevance of citizenship as the basis of rights hasbeen called into question as dual nationality increases and internationalinstitutions set standards which have global referents.

4.1. Social rights and migration

In 1950, T.H. Marshall criticized earlier work that held that ‘the inequal-ity of the social class system may be acceptable provided the equalityof citizenship is recognised’ (Marshall, 1950, p. 2). Marshall contendedthat the equality of citizenship does not mean much if other inequal-ities preclude an equal assumption of full citizenship (Marshall, 1950,p. 7). Taking England as his locus, he traced a three-part development ofcitizenship from the seventeenth to the twentieth centuries. While civilrights developed in the eighteenth century, representation and welfarewere not given importance until the nineteenth century when politicalrights gradually increased. However, these were not fully realized untilthe next century. In England, voting for all men over 21 and all womenover 30 in was instituted 1918, and women between the ages of 21 and30 were only enfranchized in 1928.

Marshall charts a progression in social rights in the twentieth cen-tury, so that eventually all three categories of rights came to hold thesame status. Thus, while the apparent movements from subjecthood tocitizenship occurred in England much earlier, it is only later that citizen-ship in its modern sense developed. Marshall describes how the usuallydescribed ‘second generation rights’ of social protection and welfare arein reality necessary for the full realization of ‘primary rights’ of political

44 Frameworks

participation, so that until the full triumvirate of civil, political andsocial rights were available, any individual category of rights was of lessvalue.

Following Marshall, citizenship in developed countries today hascertain core ‘hallmarks’, including numerous social as well as legaland political rights. Medjouba and colleagues (2008) describe these asincluding rights to freedom of movement, to a passport, to vote, to standas a candidate in national and local elections, of access to public officeand public service, to protection, to welfare benefits, to health care, toeducation, to employment, to housing, and to linguistic freedom andnon-discrimination. They also include duties of allegiance: undertak-ing military or alternative service, paying taxes and national insurancecontributions.

One of Marshall’s most significant contributions to our understand-ing of citizenship was his perception that, without social rights, it wasnot possible for poorer citizens to benefit from civil and political rights.Subsequent writers have added other social divisions, especially thoseof race, ethnicity and gender as involving further inequalities whichmight lead to such exclusion (for example Lister et al., 2007; Castles andDavidson, 2000). The focus on exclusion is less concerned with whetheror not a person is a citizen, but whether the social rights availableactually address disadvantages produced by unequal social relation-ships of class, ethnicity, gender, or by other conditions such as age ordisability.

Although such social rights are not exclusively the preserve of citizens,their identification with citizenship makes migrants who lack citizen-ship in the country of destination potentially vulnerable, particularly asthey are especially susceptible to social exclusion. This is a consequenceof economic, social and political marginalization, including the effectsof racism, but can also be due to state policies of ‘differential exclusion’(Castles and Miller, 2003). This occurred formally in rules concern-ing guest workers in pre-unification West Germany, where immigrantswere incorporated into the labour market but excluded from welfare,citizenship and political participation. However, the term can also beused to describe the position of undocumented migrants more generally(Schierup et al., 2006). In such cases the denial of citizenship is only partof a larger cluster of mechanisms which promote exclusion. The spe-cific forms of exclusion faced by migrants can be seen as a consequenceof how welfare and immigration regimes interact (Sainsbury, 2006),though this is not a given but itself a product of particular historicalcircumstances and accompanying social forces.


During the 1990s much literature on migration in Europe focusedon the coercive measures taken by states to segregate refugees not yetgranted status, often denying them the right to work or to claim thesame benefits as citizens (see for example Bloch et al., 2000). With adecline in numbers of asylum seekers but an increase in migrant work-ers, other measures involve excluding migrant workers who do not haveparticular skills and accompanying permits, restricting conditions forfamily formation or reunion in order to prevent permanent settlement,and strictly regulating the numbers of migrants entering, in terms ofparticular professions and skills. All this creates distinctions between‘respectable’ (regular) migrants with work permits who are in work,refugees with status and those awaiting asylum decisions, and ‘illegal’(irregular or undocumented) migrants – refused asylum seekers, over-stayers, people who breach their conditions of stay by working morehours than permitted or illegally changing employers.

In order to survive, undocumented migrants are thus forced to seekout work in unregulated sectors of the economy (Clandestino Project,2009). Schierup and colleagues combine different categories of migrantsand minority ethnic citizens in contemporary Europe: ‘people of immi-grant and ethnic minority background who become trapped in thefragmented occupational ghettos of casualised labour in the post-Fordistservice industries and in deregulated municipal services’ and others,‘citizens of immigrant background subsisting on shrinking welfarein ethnically or racially segregated neighbourhoods’ (Schierup et al.,2006, p. 2). Others included undocumented migrant workers, contractlabourers, asylum seekers and other kinds of working poor. Schierupand colleagues argue that these groups suffer a common ‘racialisedexclusion’, which has been exacerbated by a retrenchment of publicwelfare.

It is clearly not exclusion from citizenship per se which results inthe conditions that Schierup and colleagues describe, but unequalsocial and economic spaces, which are the only ones available tothe most vulnerable and marginalized people, among whom undoc-umented migrants figure prominently but by no means exclusively.Deregulation of economic activity has helped to create increasinglysegregated labour markets, where the least regulated segment is mostattractive to those workers who command the least rights. As migrationbecomes more and more restricted and controlled, irregular migrantswith little chance of regularization are likely to occupy precarious jobsin the most unregulated sectors of the economy (Clandestino Project,2009).

46 Frameworks

Migrants are often further excluded by language. Until a critical massof migrants can effectively create a language of equal or almost equalsocial and political importance, migrants who do not know the languageof the destination country inevitably remain at a disadvantage. Lan-guage is one of the most potent markers of cultural difference, but bi- ormultilingualism is entirely compatible with a pluralist framework. Yetthe requirements to learn the language of the country of destination as aprecondition for acquiring citizenship, without concomitant provisionand funding, can be exclusionary rather than integrative, particularlyto women in some communities (Lister et al., 2007). A study by Gilland colleagues, of three minority groups in England (Indian, Pakistani,Chinese), found that women and older people are particularly disadvan-taged regarding local language competency. This necessarily limits themin their access to social rights and also serves to disenfranchise them, asindividuals from these groups cannot participate politically (Gill et al.,2009). One Bangladeshi academic has described the ‘shelter’ of the fam-ily and home as an ‘enforced dependence on men both for protectionand provision’ (Huq, 2005, p. 169). A lack of language skills, where thenew language is essential for daily tasks, reinforces this dependence.

Suppressing the languages of subordinate groups has historically oftenbeen used as a strategy of domination, as can be seen at differenttimes with the British suppression of Irish-medium education in thenineteenth century or the Turkish suppression of the Kurdish lan-guage. In Latvia, after the break-up of the Soviet Union, citizenshiplaws contained stringent language requirements for naturalization, thusexcluding much of the large ethnic Russian population who had beenforcibly relocated there under Stalin (Schmid, 2008).

4.2. Increasing transnationalism

Changing relations within and between states since the mid-twentiethcentury have prompted a great deal of debate about changes in themeanings of citizenship. Kivisto and Faist (2007), reviewing recentdebates on citizenship, discuss contrasting views that the ‘efficacy andsalience’ of citizenship is either expanding or eroding. Some argue thatneo-liberal forces are shrinking the welfare state and the social rights ofcitizenship contained in it, while others see citizenship as becoming pro-gressively more inclusive in terms of the rights granted to women andminorities. It has also been suggested that the nation state has becomeless significant with the development of transnational bodies such as the


United Nations and World Trade Organization or the European Union,and the development of international human rights standards.

Several international and transnational groups of states have devel-oped since the Second World War. The League of Arab States, formedin 1945, offers freer movement for citizens of member states: so, forexample, the Syrian government classifies people entering the countryas nationals, Arab non-nationals and non-Arab non-nationals. This hasenabled discrimination against non-Arab groups, such as the Kurdishminority, though facilitating freer movement of Arabs. An advantagefor Arabs of this freer movement can be seen in Syria’s response to theIraqi displacement crisis after the hostilities in that country began in2003. Syria hosts over a million new Iraqi migrants and has offeredschooling and health care to large numbers. The African Union hasdiscussed the possibility of a form of African citizenship that could sitalongside national citizenships (African Union, 2004). There have alsobeen suggestions of some sort of secondary citizenship arrangement forthe African diaspora (African Union, 2005).

The European Union also has special arrangements for citizens ofother European member states, described as ‘second country nationals’,as opposed to ‘third country nationals’. The EU is widely consideredto be the most integrated and extensive regional international body(for example Medjouba et al., 2008, p. 8). The 1991 Maastricht Treatydiscussed the notion of a new EU citizenship, which was more clearlyelaborated in the 2007 Lisbon Treaty. Consequently, many of the hall-marks traditionally associated with citizenship, like health care, freedomof movement and settlement, employment and the right to vote, arenow available to citizens across EU member states. Sabates-Wheeler andcolleagues highlight the EU as a model for the portability of socialprotection across national boundaries (see Chapter 3).

While the benefits for citizens of such groupings may be huge, theconsequences for non-citizens are less clear. Some commentators seedevelopments like the EU, combined with an increased emphasis onindividual human rights, as making national citizenship less impor-tant in the contemporary world, and so enabling foreigners to acquiresimilar rights to those of citizens (Soysal, 1994). On the other hand,the increased integration of states within regional associations like theEU has been accompanied by greater protection of the external border.This can be seen, for example, in the strong border controls around theEU, leading to what is often termed ‘Fortress Europe’, and a denial ofcitizenship and other rights to people who are outside it.

48 Frameworks

4.3. Dual citizenship

Widespread dual citizenship is recent and increasing dramatically(Sejersen, 2008, p. 531). The 1930 Hague Convention stated that ‘everyperson should have a nationality and one nationality only’. From then,and throughout the Cold War, dual citizenship was perceived as a pos-sible threat to national security (Dahlin and Hironaka, 2008, p. 54).However, by 2000, one study showed that 89 countries recognized someform of dual citizenship (quoted in Dahlin and Hironaka, 2008), and inSerjensen’s 2008 study, only 28 of the 115 countries studied never allowdual citizenship.

Historically, opposition to dual citizenship centred mainly on fears ofdivided loyalties. The increased acceptance of dual citizenship has beenattributed to varied factors including the dissolution of old empires,and, particularly, increased immigration and the concomitant dualitiesof migrants’ orientation (Vertovec, 2004). Moreover, governments inmigrants’ countries of origin have an interest in retaining their citizenswho have naturalized elsewhere, to maintain remittances and invest-ments (Kivisto and Faist, 2007). Dual nationality may have also becomemore common as a result of growing gender equality, which has increas-ingly equalized the terms for women of acquiring citizenship, and hasenabled children to take the nationality of both parents (Kivisto andFaist, 2007, p. 108).

As international human rights have become more significant, states’treatment of citizens and non-citizens has had to become increasinglysimilar, making dual nationality less threatening to individual states(Kivisto and Faist, 2007). If this is true, it suggests that, paradoxically,states are being forced to treat residents more equally, which maypartially explain the increased efforts on the part of many states toexclude non-citizens, and greater differentiation between citizens andnon-citizens. Recent changes in British nationality and immigrationlaws, for example, represent harsher exclusionary measures against somemigrants while others are constructed as ‘probationary citizens’ allowingthem limited yet growing rights.

At the same time, some states have refused to accept dual citizen-ship for historic political reasons, primarily to resist the real or fearedpersistence of colonial influence (Dahlin and Hironaka, 2008). This canbe seen in states of the Baltic. For example, in Estonia, many ethnicRussians have been stateless since its independence in 1991, and inLatvia over 365,000 people remain stateless as a result of nationalitylaws which excluded ethnic Russians. Many of them had lived in Latviaall their lives but had not obtained Russian citizenship, and after the


break up of the Soviet Union could not get Latvian citizenship either(Forced Migration Review, 2009, p. 27; Goris et al., 2009).

5. The rights of non-citizens

in a scenario where borders are taken as given, membership isassumed, and outsiders are invisible.

– Goodwin-Gill (2007, p. 170)

Hannah Arendt defined citizenship as ‘the right to have rights’,acknowledging that in the twentieth century certain protections andbenefits are only available to someone with the status of citizenship.This is seen most clearly when we consider those without such a right.Thus, ‘although national rights theories granted every individual rightsat birth, these rights could only be recognized and enforced in a prac-tical way through membership in a state, which alone could prescribethe criteria for membership’ (Weissbrodt, 2008, p. 25). Although thereare many rights-conferring statuses that can be held by non-citizens,where citizenship is used as a basis for eligibility to public services andother benefits, non-citizens need to demonstrate other claims to rightsor protection.

In situations where the central state lacks power, citizenship maybe less significant for accessing political or social rights, giving non-citizens power. In some cases, individuals are able to bypass citizen-ship requirements for benefits. This occurs in Kamal Sadiq’s discussionof irregular migrants in Malaysia who obtain the ability to vote byway of criminal networks and financial influence, while many poorMalaysian citizens remain unable to vote. This skews electoral politicsin the country, as government policies become increasingly favourableto irregular migrants, further consolidating the power of the irregularmigrant vote and in turn the criminal networks that enfranchised them(Sadiq, 2005). John Wheeler shows how some Brazilian favelas func-tion almost as if independent states, controlled by the ‘poder paralelo’(‘parallel power’) of drug trafficking groups (Wheeler, 2005, p. 101).This has led to increased irrelevance of formal political rights for thepoor. Wheeler’s interviews with favela residents indicate that they donot associate citizenship with national identity, individual rights or thedemocratic process, but focused on their family or community andaccess to services needed for daily life. Dignity was also important,with 74 per cent of his interviewees reporting this as ‘citizenship’s mostimportant characteristic’ (Wheeler, 2005, p. 109).

50 Frameworks

International obligations do not require a state in which a non-citizenis present to ensure the full range of his or her human rights. To claimprotections, non-citizens must prove eligibility or ‘desert’ (merit) byvirtue of some other rights-conferring status, such as worker, womanor refugee (or, sometimes, citizen elsewhere). They can then claim pro-tections of rights in virtue of this status. This can be shown through theexample of refugee status.

The 1951 Convention relating to the Status of Refugees (the GenevaConvention) allocates someone a status they can use to claim pre-existing human rights against a particular state or body, supporting theview that a person is presumed rights-less in a state unless he or she canprove otherwise. Accompanying eligibility is desert. While citizens arepresumed deserving of services in virtue of their status, non-citizens mayhave to prove this. For example, the right to work is usually only avail-able to non-citizens who can prove they meet certain criteria applicableonly to them, while it is available to citizens on the basis of more univer-sal criteria such as age. Similarly, while citizen criminals are released afterserving their sentence, non-citizen criminals are often either deportedor re-detained as it is commonly presumed that non-citizen criminalsought to be elsewhere (London Detainee Support Group, 2009). Prin-ciples of double jeopardy or of completing a punishment seem to bethought not to apply in their case.

David Weissbrodt (2008) offers an important and sustained analysis ofthe international position of non-citizens. He notes the absence of inter-national agreements, treaties or conventions relating to non-citizensper se. Weissbrodt is concerned at their resulting vulnerability, arguing:‘It is useful to see the human rights of non-citizens not as an amalga-mation of the rights of various non-citizen subgroups (refugees, asylumseekers, migrant workers, and so on), but rather as a unified domain’(Weissbrodt, 2008, p. 5). He argues that it is preferable to consider therights of non-citizens as a distinctive group, rather than obscuring theircommon situation by focusing discussion on non-citizen subgroups.

Although the general franchise is usually not available to non-citizens,a range of benefits are bestowed on many non-citizens in many states,although there are important variations as to their universality andaccessibility. Many restrictions are imposed on the basis of residenceand other criteria (see Sabates-Wheeler, Koettl and Avato, Chapter 3).Hammar (1990) has suggested an additional status of denizen to charac-terize foreigners with legal and permanent residence status. Owen (2010,pp. 55–58) lists countries where even voting rights are available to differ-ent degrees to non-citizens. However, there are also numerous examples


of non-citizens denied basic benefits. Indeed, ‘virtually all categoriesof non-citizens encounter racism and xenophobia; difficulty realizingall their economic, social, and cultural rights, particularly the right towork, the right to education, and the right to health care; difficultyobtaining identity documents; and the lack of means to challenge effec-tively or have remedied violations of their human rights’ (Weissbrodt,2008, p. 17). In the absence of a meaningful status applicable to all non-citizens, there is no official protection of these rights. In this sense it ispossible to describe non-citizenship as a non-status: as a condition ofhaving no status, of being particularly vulnerable.

International human rights declarations take the rights of citizens asa reference point. Statelessness, by being the most extreme type of ‘non-citizenship’, provides a useful indication of the rights which citizensenjoy and take for granted. Stateless people cannot vote or participate inthe political process. They are likely to be excluded from employment,education, social services, and public health services or health insur-ance. They cannot travel and may live in constant fear of expulsion. Insome states they may be subjected to indeterminate periods of deten-tion if they have been refused asylum because they cannot be removedto any country (Gibney, 2009).

The effects of statelessness can be mitigated either by the grant ofcitizenship in a new country or by acquiring denizen rights associatedwith residence and permanence. However, stateless people may actuallyfind it hard to obtain these statuses because their stateless status has pre-vented them from meeting the required criteria. Similar examples can befound in the UK where people who have lived in the UK legally or ille-gally for 14 years or more and who have not been convicted of a criminaloffence may be granted indefinite leave to remain or settlement. Suchlong-term residents often find it practically or psychologically impossi-ble to return to their country of origin. However, because they have nothad legal residence status (and therefore many rights) for a long time, ifthey become ill or become unemployed, they are ineligible for benefits.They are thus frequently destitute by the time they are able to applyfor settlement and cannot then afford the fees required by the HomeOffice.2

These kinds of situations illustrate how Weissbrodt’s concept of non-citizens’ absence of status can express itself in practice. Migrant workersand refugees are in principle protected by international conventions,but, as we see below, refugee status is hard to achieve, and interna-tional conventions on migrant workers’ rights have little clout and littlesupport among migrant-receiving states.

52 Frameworks

5.1. The rights of refugees

The idea of respecting universal human rights has, in recent years,become increasingly accepted as a basic principle of conduct and ofcitizenship rights, although, how human rights should be interpretedand legislated for in practice in individual states has been highly con-tested. In relation to non-citizens and migrants, nevertheless, humanrights instruments have sometimes been successfully invoked to protectmigrants either from expulsion from their current country of residenceor from exclusion from social protection there. However, the incorpo-ration of human rights into international instruments only began inthe twentieth century as wars and revolutions led to the dismantling ofEuropean and global empires and created population movements andstatelessness on a huge scale.

The League of Nations, set up in the aftermath of the First WorldWar, had drafted treaties offering protections and rights of entryto various specific groups of ‘refugees’ – Russians (1926), Armenians(1928), Germans (1933), perpetuating the ideal of nation state thathad entered at the time of the Enlightenment. After the Second WorldWar, however, the scale of displacement was of a new magnitude.Europe itself had more than 10 million displaced persons and the par-tition of India and Pakistan in 1947 left swathes more people suddenlydisplaced.

The United Nations General Assembly adopted the Universal Decla-ration of Human Rights (UDHR) in 1948. The UDHR gave rights, notonly to states, but to individuals within sovereign states, which madetwo important new claims in relation to non-citizens. First, that thereare rights that individuals should have, irrespective of which state theyhappen to live in. This means that outside agencies can legitimatelyjudge a state’s behaviour towards its own citizens. Second, it recognizedthe right of people to freedom of movement, and to asylum in othercountries. This was an acknowledgement of a moral requirement thatstates accept people fleeing persecution and observe certain obligationstowards them. However, it was still vague as to how this was to bemanifested.

The Universal Declaration of Human Rights included a right tonationality in response to the massive population movements and‘denationalisations’ stemming from the Second World War (Goris et al.,2009). Hundreds of thousands of Jews who survived the war fled fromtheir home countries, millions of ethnic Germans were expelled fromEastern European states, and millions more members of minority popu-lations fled or were expelled from the Soviet Union. While some of these


were accepted by other countries as refugees, others remained statelessfor many years.

The Geneva Convention laid out a new definition of refugee statuswhich referred to events before 1951. This was extended by a Proto-col in 1967, which removed earlier spatial and temporal constraintson claiming refugee status. The definition of a refugee in the GenevaConvention is:

Someone who, owing to a well-founded fear of being persecutedfor reasons of race, religion, nationality, membership of a particularsocial group or political opinion, is outside the country of his nation-ality and is unable or, owing to such fear, is unwilling to avail himselfof the protection of that country, or who, not having a nationalityand being outside the country of his former habitual residence as aresult of such events, is unable, or owing to such fear, is unwilling toreturn to it.

(United Nations, 1951, 1967)

The Geneva Convention assured certain protections to refugees inrelation to issues such as housing, education, social security and employ-ment rights. These were to be equivalent to those of citizens. There isno equivalent international instrument covering protections for othernon-citizen groups. However, in 1966, two further human rights instru-ments were adopted by the UN General Assembly – the InternationalCovenants on Civil and Political Rights and on Economic, Social andCultural Rights were adopted (United Nations, 1966a, 1966b). Theseintroduced broader basic rights than the UDHR and, more importantly,had a stronger legal status. Both reiterated the nation state ideal, yetthe International Covenants on Civil and Political Rights proclaimedthat a state must respect the rights of ‘all individuals within its terri-tory’, irrespective of ‘race, colour, sex, language, religion, political orother opinion, national or social origin, property, birth or other status’(Part II, Article 2 (1)). Reference to national origin and birth status, inparticular, indicate that the rights in these conventions are intended fornon-citizens as much as for citizens.

The Geneva Convention is necessarily limited to people who havefled human rights abuses and who fall within the categories identifiedin the Convention, or within the definition of ‘a particular social group’.Increasing attention to issues such as rape and violence against womenhave raised further questions of how far women can be construed as asocial group, and the value of adding a concept of gender persecution

54 Frameworks

to the Convention (Crawley, 2000; Anker, 2002). Women and othersfleeing persecution on grounds which are not interpreted in the des-tination state as falling within the remit of the Convention are likelyto be denied protection and the concomitant rights it offers. Neverthe-less, despite these and other limitations, the Convention is a crucialinstrument recognizing the rights of states as they relate to a particularcategory of non-citizens.

5.2. The rights of migrant workers

It was estimated that approximately 86 million of the world’s175 million international migrants in 2004 were economically active.Women, most of whom now migrate as primary income earners, com-prise about half of all migrants (International Labour Conference, 2004).Yet international Convention rights for migrant workers are much lessrecognized than those pertaining to refugees, despite several majorconcerns being widely recognized by many authorities including inter-national human rights, legal and church organizations and the Inter-national Labour Organization, as well as academic authors and NGOs(Taran, 2000; Clandestino Project, 2009). Such concerns include, par-ticularly, the many hardships faced by migrant workers, especially lowwages, poor working conditions, virtual absence of social protection,denial of freedom of association and workers’ rights, discrimination,xenophobia and social exclusion. There are gaps between national andregular migrant workers in terms of working conditions and treatmentat work as well as in wages, job security and unemployment rates(International Labour Conference, 2004).

Both the International Labour Organization (ILO) and otherauthorities have drawn attention to the particular problems of irreg-ular migrants, partly encompassed in the epithet ‘three-D’ jobs –dirty, demeaning and dangerous (Taran and Geronimi, 2003). TheClandestino Project argued that irregular migrants have the least accessto social rights and are thus particularly vulnerable and most at riskof poverty and social exclusion. They also find themselves in the mostexploitative work situations (Clandestino Project, 2009, pp. 18–19).Irregular migrants often have few opportunities to regularize their status.It is also suggested that fee-charging recruitment agencies profit fromthe prevalence of irregular migration, with many engaging in unethicalpractices (International Labour Conference, 2004).

Recognition of migrant workers as a distinct category goes back to thefounding of the ILO in 1919, with ‘the protection of the interests of


workers when in employed in countries other than their own’ explicitlyset out in the preamble to its Constitution (ILO, 1919, Constitution). Itpassed two conventions relevant to migrant workers in 1949 and 1975,the first, the Migration for Employment Convention, 1949 (No. 97),demanding equal treatment between national and regular migrantswith respect to legal issues concerning employment practice, proce-dures and conditions. It also included reference to living standards,and access to justice, tax and social security regulations. The second,the Migrant Workers (Supplementary Provisions) Convention (No. 143),was adopted in response to increased concern about irregular migration(Taran and Geronimi, 2003).

These conventions formed the basis of the 1990 UN InternationalConvention on the Protection of the Rights of All Migrant Workersand Members of their Families (United Nations General Assembly, 1990)This is regarded as one of the fundamental human rights instrumentsdefining basic universal human rights (Taran, 2000). Taran (2000, p. 17)attributes the importance of this Convention to seven key points –that it:

1. sees migrant workers as people with rights2. recognizes migrant workers’ lack of protection either in their own

state or in the state of residence3. provides an international definition of migrant workers and interna-

tional standards of treatment of the human rights of migrant workersand their families

4. extends fundamental human rights to both documented and undoc-umented migrant workers, with additional rights accruing to docu-mented workers

5. seeks to prevent and eliminate exploitation of all migrants by endingirregular movements and situations

6. seeks to establish universally acknowledged minimum standards ofprotection for migrant workers and their families.

7. would, if fully implemented, provide significant protection for allmigrants, especially those with irregular status.

In spite of its importance and despite a Global Campaign launched in1998 to obtain the requisite ratification by 20 countries, it took over12 years from the Convention’s adoption to its becoming legally bind-ing. This was finally achieved in 2003, and the Convention has nowbeen ratified by 42 states, though there is not one Western state amongthem (UN Treaty Collection, 2010).

56 Frameworks

In 1998 the ILO adopted an alternative declaration, ‘FundamentalPrinciples and Rights at Work’ (ILO, 1998), committing all memberstates to “respect and promote principles and rights in four cate-gories, whether or not they have ratified the relevant convention” (ILO,2010, our emphasis). These categories are: freedom of association andrecognition of the right to collective bargaining, elimination of com-pulsory and forced labour, abolition of child labour, and elimination ofdiscrimination in respect of employment or occupation.

Reviewing studies which considered why so few states had signed the1990 Convention, Taran (2000) suggested that the most important rea-sons were a lack of awareness and knowledge of the Convention, partlyconsequent on a failure to promote it, but more significantly becauseof a lack of political will on the part of governments to protect therights of migrants. One study he discusses suggested that the Conven-tion was adopted in a time of growing unemployment and a relativelyhostile climate towards foreigners, where governments felt that the Con-vention was incompatible with their policies on immigration.3 Anotherauthor Taran discusses, Richard Cholewinski, argued that despite theConvention’s underlying aim to end illegal migration, governmentssaw it as granting rights to illegal migrants, and that in a climate ofgrowing economic disparity between rich and poor nations, and withmigrants of different racial and cultural backgrounds, there was littlepolitical sympathy on their behalf.4 In other words, governments ofreceiving countries were inclined to pander to racism rather than toprotect migrants’ rights.

Although both the Convention and other ILO documents encourageregularized migration, this creates contradictions for those, includingthe ILO itself, who are seeking to promote the rights of irregular workers.Vollmer has argued that the discourse on irregular migration focuseson the ‘threat’ it poses, and subsumes it under trafficking and humansmuggling, thus both stigmatizing and criminalizing irregular migrants(Vollmer, 2009). At the same time migrants’ rights advocates continuallystress the ‘marginalisation and vulnerability’ of irregular migrants andargue for more protection. The problem is how to achieve it.

For example, a recent ILO report calls for discussion on actions to‘be taken against unscrupulous employers who employ migrant work-ers in irregular status in the informal economy’ (International LabourOffice, 2008). However, a study of migrant workers affected by the ‘civilpenalty regime’ in which the UK government has increased the penaltiesfor employers of undocumented migrants suggests that the system iscontributing to increasing vulnerability of undocumented migrants tomistreatment by unscrupulous employers. In addition the policy is


being enforced in a discriminatory manner, especially affecting minor-ity ethnic employers with small businesses (Migrants’ Rights Network,2008). The authors argue for regularization rather than punitive mea-sures to reduce the numbers of undocumented migrants and to protectthem from over-exploitation.

6. Conclusion

Nadine Gordimer’s Nobel Prize-winning novel, The Pickup, describeshow Julie Summers, a wealthy white South African and Ibrahim ibnMusa, an irregular migrant from an undisclosed Arab state, fall in love(Gordimer, 2001). Unable to remain in South Africa because Ibrahim isto be removed, and unable to gain access elsewhere, they travel to hishome country, where Julie now becomes the non-citizen. Throughoutthe novel, despite her efforts, Julie is advantaged by her citizenship,her social class and her race, while Ibrahim, despite his efforts, isdisadvantaged by his citizenship, his social class and his race.

The meaning of citizenship and its relationship to human rights isdifferent depending on country of origin, country of presence, class,employment status and many other factors. This chapter has shownthat, while the status of citizen does not protect all citizens equally, andminority groups can be excluded from significant rights of citizenship,the non-status of non-citizen does not currently offer any protectionat all until some other status can be proven – and for those irregularlypresent proving such status is particularly difficult.

Notes

1. Discussions around California’s Proposition 187 give an interesting insightinto this.

2. This example is based on personal communication from Hackney MigrantCentre.

3. S. Hune and J. Niessen (1994) ‘Ratifying the UN Migrant Workers Convention:Current Difficulties and Prospects’, Netherlands Quarterly of Human Rights (citedin Taran, 2000).

4. Cholewinski, R. (1997) Migrant Workers in International Human Rights Law:Their Protection in Countries of Employment (Oxford: Clarendon Press) (citedin Taran, 2000).

References

African Union (2004) Strategic Plan of the Commission of the African UnionVolume 3: 2004–2007 Plan of Action:Programmes to Speed up Integration of theContinent, (Addis Ababa: African Union), May 2004.

58 Frameworks

African Union (2005) ‘Report of the Meeting of Experts from Member States onthe Definition of the African Diaspora’, (Addis Ababa: African Union), Ethiopia,April.

Al-Rahim, M. (2008) ‘Asylum: a moral and legal right in Islam’, Refugee SurveyQuarterly, 27, 15–23.

Anker, D. (2002) ‘Refugee law, gender, and the human rights paradigm’, HarvardHuman Rights Journal, 15, 133–154.

Aristotle (1992) The Politics in Aristotle the Politics, trans by Sinclair, T. and revisedby Saunders, T. (London: Penguin Books).

Bloch, A., Galvin, T. and L. Schuster (2000) ‘Editorial introduction’, Journal ofRefugee Studies, 13(1), 1–10.

Bosniak, L. (2007) ‘Being here’, Theoretical Inquiries in Law, 8(2), 389–410.Broeders, D. and G. Engbersen (2007) ‘The fight against illegal migration: identifi-

cation policies and immigrants’ counterstrategies’, American Behavioral Scientist,50(7), 1592–1609.

Carens, J. (2008) ‘The rights of irregular migrants’, Ethics and International Affairs,22, 163–186.

Castles, S. and A. Davidson (2000) Citizenship and Migration (New York: PalgraveMacmillan).

Castles S. and M. Miller (2003) The Age of Migration (Basingstoke: PalgraveMacmillan).

Clandestino Project (2009) ‘Undocumented Migration: Counting the Uncount-able. Data and Trends Across Europe’, Final Report, available at: http://clandestino.eliamep.gr/ (accessed 3 April 2010).

Cole, P. (2010) ‘Border crossings – the dimensions of membership’, in Calder, G.,P. Cole and J. Seglow (eds) Citizenship Acquisition and National Belonging(Basingstoke: Palgrave, 2009).

Crawley, H. (2000) ‘Gender, persecution and the concept of politics in the asylumdetermination process’, Forced Migration Review, 9, 17–20.

Crowley, J. (2005) ‘Where does the state actually start? The contemporary gover-nance of work and migration’, in Bigo D. and E. Guild (eds) Controlling Frontiers:Free Movement into and Within Europe, 140–160, (London: Ashgate).

Dahlin, E. and A. Hironaka (2008) ‘Citizenship beyond borders’, SociologicalInquiry, 78(1), 54–83.

De Guchteneire, P. and A. Pécoud (2009) ‘The UN convention on migrantworkers’ rights’, in De Gucheteniere, P., A. Pécoud and R. Cholewinski (eds)Migration and Human Rights (Cambridge: Cambridge University Press, withUNESCO), pp. 1–44.

Elmadmad, K. (2008) ‘Asylum in Islam and in modern refugee law’, Refugee SurveyQuarterly, 27, 51–63.

Faruqi, Harith S. (1972) Faruqi’s Law Dictionary Arabic-English (Beiruth: Librairiedu Liban).

Ferguson, Y. (1991) ‘Chiefdoms to city-states: the Greek experience’, in Earle, T.(ed.) Chiefdoms: Power, Economy and Ideology (Cambridge: Cambridge UniversityPress), pp. 169–192.

Forced Migration Review (2009) ‘Statelessness around the world’, Forced MigrationReview, 32.

Geary, P. (2002) The Myth of Nations: The Medieval Origins of Europe (Princeton, NJ:Princeton University Press).


Gibney (2009) ‘Statelessness and the right to citizenship’, Forced Migration Review,32, 50–51.

Gill, P., A. Shankar, T. Quirke and N. Freemantle (2009) ‘Access to interpretingservices in England: secondary analysis of national data’, Public Health, 9, 12.

Goodwin-Gill, G. (2007) ‘Migrant Rights and “Managed Migration” ’, inChetail, V. (ed.) Mondialisation, Migration et droits de l’Homme: le droit interna-tional en question II (Bruxelles: Bruylant), pp. 161–187.

Gordimer, N. (2001) The Pickup (London: Bloomsbury).Goris, I., Harringron, J. and S. Köhn (2009) ‘Statelessness, what it is and why it

matters’, Forced Migration Review, 32, 4–6.Hammar, T. (1990) Democracy and the Nation-State: Aliens, Denizens and Citizens in

a World of International Migration (Aldershot: Gower).Huq, S. (2005) ‘Bodies as sites of struggle’, in Kabeer, N. (ed.) Inclusive Citizenship

(London: Zed Books), pp. 164–180.International Labour Organisation (1919) Constitution, available at: http://www.

ilo.org/ilolex/english/constq.htm (accessed 30 April 2010).International Labour Organisation (1998) Declaration on Fundamental Principles

and Rights at Work, available at: http://www.ilo.org/declaration/thedeclaration/textdeclaration/lang--en/index.htm (accessed 30 April 2010).

International Labour Organisation (2010) Programme for the Promotion of the ILODeclaration on Fundamental Principles and Rights at Work available at: http://www.ilo.org/declaration/lang--en/index.htm (accessed 30 April 2010).

International Labour Conference (2004) Provisional Record 22 Ninety-second Ses-sion, Geneva, available at: http://www.ilo.org/public/english/standards/relm/ilc/ilc92/pdf/pr-22.pdf.

International Labour Organisation (2008) Migration, Human Rights and Devel-opment: Protecting the Rights of Migrant Workers – A Shared Responsibility,Global Forum on Migration and Development, Manila, 27–30 October 2008.

Kivisto, P. and T. Faist (2007) Citizenship: Discourse, Theory, and TransnationalProspects (Oxford: Blackwell Publishing).

Lillich, R. (1984) The Human Rights of Aliens in Contemporary International Law(Manchester: Manchester University Press).

Lister, R., and F. Williams et al. (2007) Gendering Citizenship in Western Europe(Bristol: Policy Press).

London Detainee Support Group (2009) Detained Lives: The Real Cost of IndefiniteDetention (London: London Detainee Support Group).

Marshall, T. H. (1950) ‘Citizenship and Social Class’, in Marshall, T. H. andT. Bottomore (eds) (1992) Citizenship and Social Class (London: Pluto Press),pp. 1–51.

Medjouba, F., J. Stefanelli and M. Sanchez (2008) The Rights and Responsibilities ofCitizenship (London: British Institute of International and Comparative Law).

Migrants’ Rights Network (2008) Papers Please (London: Migrants’ Rights Net-work).

Miller, D. (2008) ‘Irregular migrants: an alternative perspective’, Ethics andInternational Affairs, 22, 193–197.

Owen, D. (2009) ‘Resident aliens, non-resident citizens and voting rights: towardsa pluralist theory of transnational political equality and modes of politicalbelonging’, in Calder, G., P. Cole and J. Seglow (eds) Citizenship Acquisition andNational Belonging (Basingstoke: Palgrave), pp. 52–73.

60 Frameworks

Phillipson, C. (1911) The International Law and Custom of Ancient Greece and Romevol.II (London: Palgrave Macmillan).

Riley-Smith, J. (1983) ‘The motives of the earliest crusades and the settlement ofLatin Palestine, 1095–1100’, English Historical Review, 98, 721–736.

Riley-Smith, J. (1988) ‘The Latin clergy and the settlement in Palestine and Syria,1098–1100’, Catholic Historical Review, 74(4), 539–557.

Sadiq, K. (2005) ‘When states prefer non-citizens over citizens: conflict over illegalimmigration into Malaysia’, International Studies Quarterly, 49, 101–122.

Sainsbury, D. (2006) ‘Immigrants’ social rights in comparative perspective: wel-fare regimes, forms of immigration and immigration policy regimes’, Journal ofEuropean Social Policy, 16, 229–244.

Schierup C. U., P. Hansen and S. Castles (2006) Migration, Citizenship, and theEuropean Welfare State (Oxford: Oxford University Press).

Schmid, C. (2008) ‘Ethnicity and language tensions in Latvia’, Language Policy,7, 3–19.

Sejersen, T. (2008) ‘I vow to thee my countries’, International Migration Review,42(3), 523–549.

Shoukri, Arafat M. (2008) ‘Refugee status in the Arab and Islamic tradition’, PhDThesis, SOAS.

Soysal, Y. (1994) Limits of Citizenship: Migrants and Postnational Membership inEurope (Chicago, IL and London: University of Chicago Press).

Taran, P. (2000) ‘Human Rights of Migrants: Challenges of the New Decade’,International Migration, 38(6), 7–51.

Taran, P. and E. Geronimi (2003) Globalization, Labour and Migration: Protectionis Paramount (Geneva: International Labour Office, International MigrationProgramme).

United Nations High Commission for Refugees (UNHCR) (1951, 1967) Conventionand Protocol Relating to the Status of Refugees, available at: http://www.unhcr.org/3b66c2aa10.html (accessed 27 April 2010).

United Nations (1966a) International Covenant on Civil and Political Rights, avail-able at: http://www.hrweb.org/legal/cpr.html (accessed 27 April 2010).

United Nations (1966b) International Covenant on Economic, Social and CulturalRights, available at: http://www2.ohchr.org/english/law/cescr.htm (accessed 27April 2010).

United Nations (2009) Department of Economic and Social Affairs, PopulationDivision Trends in International Migrant Stock: The 2008 Revision (UnitedNations database, POP/DB/MIG/Stock/Rev.2008), available at http://esa.un.org/migration/ (accessed 27 April 2010).

United Nations General Assembly (1990) International Convention on the Protectionof the Rights of All Migrant Workers and Members of Their Families, available at:http://www2.ohchr.org/english/law/cmw.htm (accessed 27 April 2010).

Vertovec, S. (2004) ‘Migrant transnationalism and modes of transformation’,International Migration Review, 38(3), 970–1001.

Vollmer, B. (2009) Political Discourses on Irregular Migration in the EU, CLAND-ESTINO Policy Brief, available at: http://clandestino.eliamep.gr/policy-briefs/(accessed 3 April 2010).

Weissbrodt, D. (2008) The Human Rights of Non-Citizens (Oxford: Oxford Univer-sity Press).

Wheeler, J. (2005) ‘Rights without citizenship?’, in Kabeer, N. (ed.) InclusiveCitizenship (London: Zed Books), pp. 99–113.

2Structures of Access to SocialProvision for Migrants∗Ian MacAuslan and Rachel Sabates-Wheeler

1. Introduction

Low-income migrants are often more vulnerable and have poorer accessto social provisioning than low-income non-migrants. The ethical andlegal arguments for the right or entitlement of migrants to social pro-vision have been discussed in Chapter 1. In this chapter we start fromthe hypothesis that being entitled or having a right to social provisiondoes not guarantee that it is actually received. Of course, receiving wel-fare provision depends on formal entitlements specified in internationaland national laws, regulations and policies. Provision can be granted,denied, controlled and obscured at this level. However, successfullyclaiming social provision in practice turns on the way that individuals(formally entitled or not) are able to leverage opportunities for accessingresources. This hypothesis implies that getting the formal rules right isnot enough.

A rigorous understanding of access must examine not only formalrules of entitlement, but also the distributional mechanisms throughwhich that entitlement is provided, as well as the negotiations andbargaining that take place around those rules between claimants, rule-setters and providers (such as employers, administrators, friends andfamily). Even where migrants are formally entitled to social provisioning(and often they are not), they have unequal access for various reasons.Unequal access leads migrants to secure social provision from a range ofsources: from the market, and from non-market distributions, includingpublic distributions, and charity or network-based relations. Using illus-trative examples specific to migrants, this chapter develops a frameworkfor analysing access to each type of source.

61

62 Frameworks

2. The basic components of an access framework

This chapter proposes a theoretical framework for analysing migrants’access to social provisioning. The foundation of this framework is arecognition that in practice people lack equal access to productiveopportunities and social provisioning. Differential endowments andimperfections in markets for goods, services and labour combine, inalmost every society, to regularly and repeatedly exclude some groupsand individuals (often including migrants) from both productive oppor-tunities and social provision, despite their formal entitlements to both.Delivering effective social provisioning for vulnerable groups and indi-viduals, such as migrants, requires a deeper understanding of thestructures that underlie this exclusion.

Our framework begins from two claims. First, individuals everywhereface threats (or risks) from various sources which affect their abilityto provide adequately for themselves and those who depend on them.Second, inequalities in endowments and in access to human, financialand physical capital – and the associated inequalities in confidence andself-esteem – produce unequal outcomes that increase the frequencyand negative consequences of these threats for those individuals andgroups with poorer endowments and access. These threats to livelihoodand the injustices of unequal opportunity and outcome motivate socialprovisioning, in order to protect people from threats and to reduceinequality.

Social provisioning, or what is (often pejoratively) referred to aswelfare, relates to the provision of goods, services to protect peoplefrom, and prevent, the negative livelihood impacts of risk, and toreduce structural inequalities in opportunity. Consider three approachesto understanding social provisioning. Social provisioning can be seenas motivated by an ethics that condemns the unfairness of inequal-ity (Rawls, 1971) and the injustice of some individuals and groupshaving significantly different levels of insecurity of livelihood thanothers (Sen, 2009). In another way, social provisioning that producessecurity and welfare can be seen as a ‘necessary supplement’ of thevery social-political-economic world that generates these threats andinequalities: insecurity and ‘ill-fare’ (Bevan, 2004; Gough et al., 2004).By distributing some of the fruits of that world and maintaining aworkforce in reasonable health, skill and contentment (at least con-tent to the point of not radically challenging the status quo), socialprovisioning sustains the world that makes social provisioning neces-sary in the first place (Polanyi, 1944). Third, social provisioning can be

Ian MacAuslan and Rachel Sabates-Wheeler 63

analysed as another type of resource potentially available in this inse-cure world – susceptible to the same inequalities and problems of accessand requiring the same negotiation and contestation.

These three ways of understanding social provisioning are notmutually exclusive, but suggest different approaches to the analysis ofthe distribution and receipt of provisioning. The ethical understandingtypically analyses recipients as members of a group that confers certainentitlements, whether as a citizen of a nation state with legal rights toshared state resources or as a human with universal rights to a minimumlevel of well-being.

If, however, social provisioning is seen within a world that generatesinequality and threat, and particularly if social provisioning is seen as aresource that must be negotiated, why should we expect access to socialprovisioning to be unproblematic, equal or fair? Even if we can showthat all citizens or humans should be eligible for provisioning, this doesnot mean that all citizens or humans actually receive provisioning, in anequal way, or receive provisioning according to their needs (rather thanthe needs of the dominant economic system). To the contrary, thereis a mass of evidence suggesting that social provisioning is not equallydistributed and that it is in fact quite poorly distributed according toneed. For example, it is found that the actual acquirers from non-market distribution systems are often not the intended beneficiaries:the ‘rich’ embezzle food distributed for the ‘poor’, homeless individu-als are rarely included in cash distribution programmes (Dorward et al.,2006), and children who do not attend school are not provided withnutritional supplements (Britto, 2005). Case studies throughout thisbook give examples of migrants often receiving worse provisioning thannon-migrants, despite greater specific need. Our theoretical frameworktherefore focuses on the difference between eligibility and actual accessto social provisioning.

Our framework accommodates the wide range of sources and dis-tribution systems of social provisioning. Social provisioning includestax-funded benefits, emergency food aid, private insurance, individ-ual charity, and support to home production through subsidized inputmarkets. Social provisioning can be provided through a spectrum ofmarket and non-market distribution systems. At one extreme, per-fect markets have no formal restrictions on access, price or quantity(this is non-existent in practice, but consider an open market withunlimited quantities available), and at the other, non-market systemsrestrict access to certain individuals, and provide a fixed quantity ata fixed price (consider food aid provided in fixed quantities for free

64 Frameworks

to all registered households in an internally displaced persons camp).Provisioning may be formal (consider governments or organizationswho distribute to those who meet a fixed criterion, such as chronicillness) or non-formal (consider individuals distributing alms out-side a temple or church to those that appear needy). Furthermore,provision and distribution may be ‘legitimate’ (consider a regularmigrant receiving free health care under the NHS) or ‘illegitimate’(consider, for instance, the practice whereby irregular migrants ‘borrow’national insurance numbers from regular migrants in order to enterthe labour force). In between, there are many combinations ofthese extremes and many variations on these features (consider, forexample, market-based pensions that are topped up by state non-marketdistribution).

As we are specifically concerned with access from the point of view ofthe migrant at the point of need, we suggest analysing the underpinningnorms, rules and procedures that govern access with respect to theirrelationship with those seeking to participate in the system. We refer tothese participants as ‘claimants’ (who may be applicants to non-marketdistributions, buyers of products in a market, workers, or recipients ofinformal or network-based assistance) and providers (who may be rule-makers or rule-administrators of non-market systems, sellers, employersor patrons).

These concerns suggest a range of questions. For instance, what arethe physical, financial, relational and institutional constraints to welfareprovisioning at any given point? What are the structures of access at thepoint at which a migrant interfaces with a provider? And, how doesthe migrant or patron strategize to overcome constraints? These ques-tions are useful for analysing any system of provisioning. However, theanswers play out differently according to the relational or distributionalbasis underpinning each system. Below we investigate these relation-ships in order to build a conceptual framework for understandingaccess.

3. Access under different distribution systems

Let us consider the two extremes of market and non-market access. Ina perfect market, all constraints of knowledge, market existence andsupply are assumed away: the only requirement is that the buyer pro-vides sufficient exchange value to obtain the good or service required.Acquiring goods from non-market distribution systems requires the


applicant to have knowledge of the provision, perhaps to provide anexchange value, and to meet some eligibility/targeting criteria (such asnationality, poverty or age). This often requires providing data (such asa passport or an income or birth certificate), or having the right socialrelations (such as being part of a church or community) or being part ofthe right networks (such as a savings group).

3.1. Market systems

It is evident that it is not straightforward to obtain resources fromeither perfect markets or non-market systems. Under the strict assump-tion of perfect markets, information, services, goods and supply wouldall adjust and re-adjust to clear the market at any given price. Theresulting allocation would be Pareto efficient.1 For instance, if an insur-ance product was made available, people would have full informationon the product, would demand the amount of insurance that wouldreflect their risk preference, and supply would respond perfectly as therewould be no moral hazard constraints (such that no one would engagein riskier behaviour because they were insured). The only restrictionto an equitable market outcome would be the unequal distribution ofresources that would exclude some people from entering the market ata given price. Because of this exclusion, a role exists for the state (evenwithin neo-liberal welfare economics) in ensuring that some wealth isredistributed to those at the bottom.

However, in the real world of imperfect markets, while the marketwill be able to provide many welfare needs around obtaining and sus-taining an adequate income (from labour markets, markets for privateinsurance, house insurance, health insurance, childcare provision andso on), the availability of these welfare provisions will depend on a rangeof formal rules, institutional constraints, missing markets, informationalasymmetries, access to resources and access to political capital. Specifi-cally, in imperfect markets many markets may not exist (for instance,credit, labour, finance and insurance). Those seeking to acquire goodsmay not know about the markets, or may not be able to engage withthem (for instance if they do not speak the right language). Sellers maybe unwilling to sell to certain buyers because they discriminate againstthem, or because they lack the basic trust required for any exchange totake place (for example, they may not believe that the buyer will actu-ally provide the exchange value for the good or service provided, or viceversa).

66 Frameworks

The brief discussion above has illustrated that simply relaxing the‘perfect’ market assumption (while staying within the parameters ofmainstay economic theory), creates space for inequities in access beyondthose determined by income. However, more fundamental access prob-lems arise once we problematize the notion of the market as a neutralplace of exchange. The vibrant critique of the notion of ‘neutral’ marketscomes from the feminist economics literature (Folbre, 1994; Elson, 1999)and can be summarized as follows. The market, as defined by economics,is a neutral arena in which buyers and sellers interact. These agentsmay be differentiated by a number of characteristics, such as gender,class, ethnicity and nationality, and they will have a range of endow-ments and preferences. ‘Discrimination’ in the market is acknowledgedby economists as the existence of significant pay differentials that areunexplainable by education, training and work experience. This dis-crimination is interpreted as a residual in economic models, reflectingemployers’ and employees’ preferences that are not formalized. Butwithin market theory this remains a puzzle as it does not reflect aprofit-maximizing strategy.

The problem with this ‘preference’ approach to explain ‘discrimina-tion’ is that it ignores broader structures of disadvantage and maldistri-bution. These structures translate into certain groups being less educatedand less able to accumulate the same experience as the more dominantor powerful group in any given society. Elson (1999) provides a usefuldiscussion of these processes and structures of disadvantage, embeddedin labour market institutions, which serve to maintain and entrenchgender inequalities and exclusion. Building on Whitehead (1979), Elsonbegins from the idea that markets are institutions which are ‘bearers ofgender’. This refers to the notion that there are social stereotypes aboutwhat constitutes men’s work and what constitutes women’s work.2 Thisanalysis readily applies to the case of migrants and their position relativeto native populations at the destination. If markets as social institutionsare ‘bearers of gender’ or class or nationalism, this means that marketsystems and rules, such as payment systems, job evaluation criteria,trade unions, policies and so on, always have scope for discretion intheir application.

It is often assumed that free or perfect markets equate with economicefficiency in terms of resource use. However, it is clear that the notionof a ‘free’ market is only an ideal and that anything less than a ‘free’market will mean that distribution will be discretionary based. This isparticularly important for the case of migrants as it implies that accessto provision is not based on merit or need but on discretion and possibly


discrimination. Markets will always need to be regulated to some extent,which necessarily requires a governing authority (Sawyer, 1995). Thisauthority is typically overseen and vested in the hands of the state. Thequestions then become: whose interests does the regulation serve? Whatkinds of norms and perceptions does it promote? The fact is that thosein power are not only interested in economic relations but in perpetu-ating their advantage in the social and political spheres – whether theybe based on gender, nationality, class or ethnicity. Furthermore, once weacknowledge that not everyone in markets is equally well informed andhas complete information about opportunities, and that these asym-metries can be created and entrenched by discretion-based distributionsystems, it is clear that some less empowered groups will have moredifficulties accessing markets than other more powerful groups.

3.2. Non-market systems

In non-market systems, we find a slightly different mix of access con-straints. Non-market distribution systems are not set up directly foreconomic gain, but are deliberate attempts to achieve some welfareor political objectives. Non-market distributions therefore differ frommarkets by deliberately restricting eligibility for the resource distributedand requiring evidence to prove eligibility. Usually an exchange value isnot required, but such markets retain a similar set of access constraintsas imperfect markets: claimants may lack information about the dis-tribution, and may suffer resource constraints (such as time, moneyand so on). All social provisioning that provides resources throughnon-market distribution systems restricts eligibility, whether explic-itly (for example, through published programme eligibility criteria) orimplicitly (for example, by personal charitable choice). These restric-tions have the intention of ‘targeting’ a particular group. Targetinghelps, for instance, to improve administrative efficiency or reduce pro-gramme costs (Sharp, 2001); improve equity outcomes (Schaffer andLamb, 1981), exclude some undesirable or undeserving group (Wood,1985), or solve a problem of distributing insufficient resources. A seconddifference from markets is that non-market systems are often public.This means that claimants and non-claimants can influence the eligibil-ity rules, access structures and eventual distribution through advocacyand protest (‘voice’). Voice usually aims to improve allocations forthe group as a whole (Schaffer, 1986). Finally, non-market distribu-tion systems, like imperfect markets, are open to manipulation whereclaimants seek to improve their individual allocations using social ties,

68 Frameworks

authority and influence or payments (Schaffer and Wen-Hsien, 1975).Likewise, individual administrators negotiate around the formal rules(von Benda-Beckmann et al., 1989) and interpret the rules unfavourablyfor some claimants and favourably for others, and may invent their owninformal, changeable, unpredictable rules (as patrons do in personalcharity).

Non-market systems are operated by governments, civil society orga-nizations (such as non-government organizations (NGOs) or faith-basedorganizations (FBOs)), or by individuals seeking to distribute charity. Ineach case, actual access to the resources distributed depends on nego-tiations between claimants and providers. The formal rules and accessstructures merely provide a framework for this negotiation. Each type ofsystem, however, implies rather different objectives, rules and structuresfor restricting access, and different opportunities for expressing ‘voice’and for negotiation. The question is whether migrants are well equippedfor these negotiations, relative to the rest of the population.

Government non-market distribution systems (re)distribute resourcesto influential or vulnerable groups (or both), reflecting political and eco-nomic strategy or a concern for equity and welfare, depending on eachcontext. Some groups, such as non-citizens or the non-poor, are sim-ply excluded by the rules. For example, the distribution of governmentfood aid to all citizens may require that citizens provide some form ofidentification, such as a passport, which they may not have or whichadministrators may not trust. This requirement makes it very difficultfor non-citizens to obtain the resource. Other groups may be excludedby administrators who bend rules because they are motivated by prej-udice or time pressure (Lipsky, 1980). The first options for redress forthose excluded might be negotiation with providers or administrativelysanctioned appeal. These low-cost attempts to improve individual allo-cations can be seen as ‘competitive’, since improvements in individualallocations come potentially at the expense of others’ allocations. How-ever, eligibility and selection rules and the rules for allocating resourcescan also be influenced by ‘collective’ voice. Collective voice influencesthe distribution of power within a polity, through democratic debateor by direct access to rule-makers such as politicians and civil servants.Particularly in the case of government provision, (excluded) claimantscan therefore resort to mobilization: ‘the forming of like minded groups,of parties and rallies, and of . . . riots’ (Schaffer and Lamb, 1974, p. 83).Mobilization is more likely when allocation rules are simpler and there-fore more difficult to manipulate (Schaffer, 1986). This collective voicemay lead to better improvements for migrants than competitive voice.


There are various examples of immigrant political participation, as partof wider groups or alone, leading to improvements in their provisioning.For instance, immigrant workers may join domestic colleagues in mobi-lizing for improved labour conditions (see Però and Solomos, 2010, foran overview of migrant mobilization).

The rules of civil society distributions (through NGOs or FBOs,for example) also deliberately exclude groups. An NGO may decideto provide support to people living with HIV and AIDS (PLWHA),which requires identifying PLWHA and excluding those not identified.Actual allocations are also subject to negotiation between claimants andproviders (and there is plenty of evidence showing this, as is clear fromany analysis of programmes’ targeting performance that finds inclusionand exclusion errors). However, these rules may be harder to influencethrough collective action, unless the organization invites participationin the definition of eligible groups or selection rules. Consider, forinstance, programmes that consult the community when designing tar-geting criteria or rules – a consultation that often reflects migrants’interests poorly if they are not considered ‘part of the community’.Usually, civil society distributions are based on a decision made indepen-dently by the organization, and eligibility criteria and rules are designedon a basis of need (for example, poverty targeting) or group membership(for example, being Catholic). This contrast with government distri-butions arises from their different relationships with the groups withwhom they work. Civil society organizations need the permission oflocal authorities and informal acceptance by members of local popu-lations to work. However, they are typically not directly and formallyaccountable to these populations and there are rarely formal channelsfor collective voice around civil society organizations. Governments,by contrast, are usually subject to a formal social contract (such as aConstitution), elections and the threat of expulsion.

Individual distributions restrict access not through formal rules orrequirements for payment, but through decisions made by the indi-vidual who is distributing. Sometimes, these decisions have a system(consider, for instance, the distribution of alms after church each day),and sometimes these decisions are more random (such as giving moneyto a beggar on the street). The eventual allocation of resources willdepend on knowing about the distribution, being noticed and notsuffering discrimination, and luck.

In each type of non-market distribution, claimants attempting tonegotiate better access or rule changes typically use brokerage throughthree types of intermediaries. ‘Representatives’ have better knowledge

70 Frameworks

of the rules or better ways of circumventing them, ‘Fixers’ may changethe rules or secure particular application by intervention higher upin the institutional hierarchy and ‘Agents’ find alternative sources ofprovisioning (Schaffer, 1986, p. 374). Migrants exercising voice fre-quently associate with these intermediaries (Però and Solomos, 2010).They may indeed rely on them, given migrants’ own often precar-ious relationship with the state. In India, for instance, middle-classactivists advocate to policy-makers on migrants’ behalf for better socialprovisioning (see Chapter 7). The risk is that migrants are fragmented,poorly incorporated politically, and that professional advocates can actpaternalistically, or arrogantly (Self, 1986) or unsystematically.

3.3. Network-based systems

Social provisioning can also be found in informal non-market systems,such as networks. Social networks and relationships act as a form ofclub good. In the market framework entry conditions to club goods aredetermined by financial, status and knowledge barriers. Formal non-market systems have explicit rules. In network systems, by contrast,entry is based on relations of trust and reciprocity, necessarily built upthrough repeated interactions between the people involved. Financialconditions for entry also play an important role (for instance, in thecase of burial societies, rotating savings funds or hometown associationmembership). However, the primary condition for entry is through rela-tional and status-based affiliation (family relations or ethnic solidarity).Stewart’s theory of group behaviour helps to inform our understand-ing of how these types of groups work to ensure positive outcomes. Shelabels Trust/Reciprocity (T/R) groups as those characterized by members’high degree of commitment to the group, which leads to trust, so thatactions are carried out in the belief that they will be reciprocated at sometime. This mode of operation generally evolves as a result of repeatedinteractions of long duration. Societal norms, as well as individual val-ues, are relevant to the strength and nature of the T/R norms. T/R groupsmay also be efficiency enhancing because T/R minimizes governmentinvolvement and monitoring and makes the most of local knowledge(Baland and Platteau, 1996; Stewart, 1996; Heyer et al., 2002).

Network-based affiliation and distribution provide important func-tions under high levels of uncertainty. Subsistence assurance and riskmitigation are the most valuable functions (see work by Sabates-Wheeler(2005), in the context of transition agriculture). To the extent that peo-ple are cut off or excluded from formal insurance and social service


provision, they need to insure downside risk through other avenues.Similarly, in places with thin state-sponsored and market-providedsocial protection, migrants face the full risk and uncertainty of marketfluctuations and institutional/legal changes. In these cases the groupsubstitutes for imperfect markets. Various social provisioning systemsare rooted in cultural norms, for instance payment into hometownassociations. Pooling of resources and mutual assistance (for example,savings groups, burial societies, fostering and childcare) can lower therisk for a particular individual or household from livelihood crises. Inthis way the network is seen as an insurance mechanism for maintain-ing a certain level of livelihood. Furthermore, different risks may beminimized through different strategies. That is, a regular migrant maychoose to opt into formal health insurance but not into a formal pen-sion scheme. This could be because the migrant’s time preference maymean she or he would rather invest in property or children’s educationat source location than a pension at destination. Or it may be that pen-sion payments are difficult to claim upon leaving the country. The pointis that migrants will choose and invest in a portfolio of risk managementmechanisms; some labour market based, some non-market based andsome network based. Clearly, some migrants will be more constrainedthan others in this choice, depending on their migration status and theresources available to them. However, this is not a trade-off betweendistribution systems. Instead, distribution systems are mixed to achieveoptimal risk insurance under the constraints given.

While social networks built on repeated personal interaction andtrust provide, in many cases, a comfortable fall-back position, acquiringgoods and services through these networks is not straightforward. First,there is the issue of covariate risk and shocks that hit the entire network.Group cooperation gives some protection against idiosyncratic risk andmild to medium covariate risk. For instance, in the case of unemploy-ment of one member, the group can provide subsistence support (foodand housing) while that member finds another job. However, if all themembers of the group are made redundant at the same time then theoutcome may be devastating. Second, and in the same vein as a clubgood provided by market-based systems, network-based systems typi-cally use exclusion-based eligibility criteria, where certain individualsand households living in traditional sending localities may be discrim-inated against because of issues of culture and identity such as gender,ethnicity, religion, family association and caste. Certain groups maylack access to migrant networks, where these develop within culturalboundaries.

72 Frameworks

4. A framework for access

Drawing the above discussions together, Table 2.1 sets out the range ofaccess constraints in different distribution systems. Very approximately,financial constraints are more likely to be significant in markets, andother constraints (aside from formal eligibility rules) tend to increasein magnitude as distribution systems become more personalized, ad hocand discretionary.

In this table, the darker shades represent ‘harder’ access constraints.In market systems, for instance, financial constraints (insufficient finan-cial endowments) present serious constraints as claimants need tobuy provisioning (such as insurance). Financial constraints are rarerin most non-market distribution (and non-contributory) systems. Inperfect markets there are only financial constraints, but in imper-fect markets, and especially in command economies, buyers maylack information or be discriminated against, or the market maysimply not exist. In command economies, some claimants may beexcluded by the rules of the economy or may lack the networksneeded to obtain the resource. In non-market distributions, informa-tion can also be a constraint, and the absence of networks is a moreserious restriction than for most markets, particularly for the morepersonal and network-based distributions where individual patrons’relationships affect their distributions. Group affiliation is relevantto both market and non-market distributions, but it is differentiatedby the nature of the exchange. In non-market, network-based sys-tems the affiliation is based on reciprocity and trust, while in marketsystems rules and contracts are required to uphold exchange relation-ships. Non-market distributions may also involve more discrimina-tion than found in most markets, especially where individual patronsdistribute their goods/services personally. Finally, most non-market dis-tributions operate according to fixed rules that may exclude manyclaimants, and often require demonstrations of legality that can excludemigrants.

4.1. Migrants in market-based distribution systems

Even without elaborate theorizing, it is easy to see that markets neces-sarily find their modus operandi within political-economic structures.In any national labour market, for instance, the ability of peopleto participate is defined by immigration laws and labour laws. Entry

73

Table 2.1 Access constraints in different distribution systems

Access constraints

Financial Information Network-based(Trust)

Targeted/Discretion

Legality

Perfectmarkets

Market Imperfectmarkets

Commandeconomy

Non-market

Governmentdistribution

Charitabledistribution

Personaldistribution

Group/networkaffiliation

74 Frameworks

is prohibited for certain categories of people due to their legal rela-tion to the state in which the market functions. This equally appliesto financial markets. For example, in South Africa and increasinglyin the UK, it is very difficult to open a bank account and accessa range of private markets without a permanent residency status asascribed by the state and, in South Africa, reflected in a specific13-digit ID number. Furthermore, many financial systems for remittingincome safely require a form of ‘legitimate’ ID. In this way mar-kets, through their subordination to states, become discretionary andexclusionary, translating into access constraints for some and not forothers.

It is obvious that in the context of labour markets and financial mar-kets, exclusion is intensified for migrants. It is not merely the case thatthe state imposes demands on how the market functions, but that themarket requires the state for smooth functioning. In order for the multi-tude of arms-length contracts that comprise any market to be upheld, aregulatory framework must be in place that enables the smooth func-tioning of the market. This regulation, by definition, excludes andincludes on a discretionary basis. As early as 1967, Rex and Moore arguedthat exclusion of immigrants in the Sparkbrook area of Birmingham, UK,from both council housing and from mortgages drove them into own-ership and tenancy of poor quality, overcrowded and high rent lodginghouses.

Immigration status is an important component of racial and eth-nic disparities in health insurance coverage and access to care. In theUnited States, non-citizen immigrants and their children have largegaps in their health insurance coverage and access to health care, evenwhen the children are citizens. The disparity in access has two com-ponents. First, non-citizens and their children are much more likelyto be uninsured. Since insurance strongly increases access to care,lack of insurance reduces immigrants’ ability to get care. Second, eveninsured non-citizens and their children have less access to medicalcare than insured native-born citizens have (Ku and Matani, 2001).This is suggestive of the discretionary and value-laden workings of themarket.

From a policy perspective, the insurance gaps for citizen childrenin immigrant families are distressing, since they are eligible for healthinsurance through Children’s Medicaid and the Children’s Health Insur-ance Program (CHIP) and are a major target of outreach campaigns.In fact the insurance coverage of the US-born children of immigrants


has fallen in recent years. Furthermore, non-citizen families have pooraccess to both ambulatory medical and emergency room care. The gapin emergency care is particularly relevant because federal policy letsnon-citizen immigrants, including undocumented aliens, receive emer-gency Medicaid services, even if they are ineligible for full coverage.Non-citizens and their children also have worse access to both reg-ular ambulatory and emergency care, even when insured. It is thusimportant to consider strategies to foster private job-based health insur-ance for immigrant workers and their families. A recent study foundthat non-citizen workers in California were offered health insuranceless often than citizens were, but that their take-up of insurance offerswas similar. Immigrants encounter other non-financial health care bar-riers that relate to information. Language problems were the leadingbarrier to child health services cited by Latino parents; they may alsoincrease medical errors because of misdiagnosis and misunderstandingof physicians’ orders.

Another example comes from the UK. Legal status, defined by thenature of residency within the UK, defines the rights of any categoryof migrant to different types of social protection. Migrants to the UKwho hold entry visas but not work permits are not permitted to obtainemployment within the UK. They are entitled to public health care incases of emergency, but no other social benefits. However, evidenceshows that a substantial number of Ghanaians and Malawians visit-ing the UK on entry/tourist visas obtain employment without workpermits. Many of them work in the informal sector, however, manyalso work in the health sector, as nurses and doctors, and otherswork in business. Within these groups, many incur ‘deductions’ fromtheir salaries (for national insurance or other taxes) yet are unable toclaim standard social security benefits from their place of employment(such as maternity and sick leave or redundancy packages). However,in cases of health emergencies some employers provide informal pay-ments to illegal workers. Some employers also provide housing (typicallyhighly inadequate) to attract cheap workers. Access to these benefitsis discretionary and clientelistic. While employers may be willing toemploy ‘illegal workers’, thus enabling irregular workers to acquirean income, employers may have incentives to illegally ‘tax’ the irreg-ular worker’s income. This arrangement is clearly beneficial to bothparties. However, it represents forgone income for the national gov-ernment budget and often the migrant becomes severely exploited astheir legal status does not privilege them with the right to question their

76 Frameworks

employment situation. The extent of infringement on human rights ver-sus labour rights/entitlements becomes highly blurred in the contextof irregular migrant labour. Many times migrants are ignorant of theirrights or are scared to pursue these rights, making access highly polit-ical. This example highlights the complexity of access within marketenvironments.

‘Free markets’ work well once one has fulfilled a range of criteriarelated to residency and employment as defined by a ruling body.In this way the market acts as a club good. But if one’s relation tothe state is ambiguous or ill-defined, as these examples of migrantshave shown, then markets will never produce effective outcomes formigrants.

4.2. Migrants in non-market systems

Migrants may be relatively disadvantaged in their access to non-marketdistributions because (1) the formal rules exclude them, (2) eligiblemigrants often face higher costs of proving eligibility, (3) they farepoorly in negotiations with administrators, and (4) they are less ablethan non-migrants to engage with the rule-makers.

Formal rules render migrants ineligible for social provisioning whengovernments are trying to deter migration (usually of low-incomegroups in the informal economy) or are unwilling or unable to adjustdistribution systems to mobile populations, or because migrants arenot seen as a core part of a constituency or risk pool. Research onIndia’s Public Distribution System (PDS, a national food subsidy pro-gramme) provides examples of the first two kinds (see Sabates-Wheelerand MacAuslan, 2007; MacAuslan, Chapter 7 in this volume). The Delhigovernment decreed that households migrating to slum areas between1998 and 2007 would be ineligible for the programme, partly in orderto deter migration (as membership of the programme also implies res-idence rights). Interviews with officials and some official documentsbetray a concern about additional migration. Moreover, migrants are defacto rendered ineligible because local quotas for programme member-ship are fixed for several years, and are not updated to account for newmigrants, irrespective of need. To take another example, many Europeansocial security programmes de jure or de facto exclude illegal immigrantsfrom contributing to or receiving from social insurance on the basisof their illegal status.3 Despite the value of illegal immigrants’ partici-pation to both the social security system and members of the system,immigration law trumps social security rationales.


Migration comes with non-trivial costs of establishing eligibility.For many programmes, migrants need to acquire new documenta-tion proving their eligibility every time they cross an administrativeboundary. The exclusionary effects of high time costs imposed onmigrants to reapply for their entitlements after moving is evident inthe case of the PDS, where many temporary migrants to Delhi areunable to meet these costs and so do not benefit from subsidizedgrain. Moreover, many migrants simply cannot obtain the requiredidentity documents (Smita, 2006), particularly if they are irregular (seechapter 5 in this volume on migrants’ problems accessing health care inthe UK).

Migrants’ ability to negotiate with administrators is constrainedby their precarious legal status or lack of personal connections. Forinstance, while claimants can often use administratively sanctionedmodes of appeal, Schwartz (2001) suggests immigrants may avoid thisin claiming food stamps to avoid disclosing their immigrant status.

For similar reasons, migrants’ political access is often restricted.MacAuslan (Chapter 7 this volume) argues that temporary migrants’political access to alter Public Distribution System rules is limited bytheir disorganization and transience, which together mean they donot form a visible political constituency. Their ability to work throughintermediaries is limited by their social dislocation.

4.3. Migrants in network-based systems

Network-based sources of welfare and provision are especially criticalfor people who are ineligible for, or excluded from, formal or market-based provisions. As illustrated in Table 2.1, ineligibility and exclusionfrom more formal systems can be due to a lack of finance, information,voice and representation, but can also be due to the legal structuresand social norms that may purposely (or not) serve to exclude. Formany poor people, and poor migrants in particular, access and dis-tribution through social ties is often the only form that their safetynet provision takes.4 For instance, social networks and affiliations withinstitutions such as trade unions, community-based organizations, andother relatives or friends facilitate labour migration by providing ini-tial accommodation and employment at destination or informationon employment opportunities, thus facilitating job searches. Anotherexample is childcare provision for working female migrants. This is oftenaccessed reciprocally through network-based relationships. Health carenetworks are also vital. Qualitative research with Ghanaians in the UK

78 Frameworks

found that elaborate network-based systems exist within the organizedGhanaian community that enable irregular migrants to access a ‘parallelhealth service’ to that provided by the British public sector. For exam-ple, a pregnant irregular migrant can be assigned a Ghanaian midwifefor most antenatal care. She can access this ‘parallel system’ until thepoint of delivery, at which point she will present herself to public sectoremergency care (Sabates-Wheeler, 2008).

While network-based systems can offer a range of provisioning tomigrants, they may also constrain the migrant. In a similar way tothe possibility described by Wood (2003) of adverse incorporation intomarket and formal provisioning systems, people may be unable toexit from a network once they become embedded into it. In network-based provisioning systems, moral economy factors exist to ensurecooperation in times of hardship. While network-based forms of socialprovision can provide a stable safety net, they may also act as aconstraint, in the sense that people engaged in these networks arenot always free to leave due to possible recriminations from relativesand other family members. In other words, there is a social lumpi-ness that comes with grouping around family and friends that canbe both constraint alleviating and choice constraining. Hence, riskmanagement in the present involves loyalty to institutions and organi-zations that presently work and deliver livelihoods, whatever the longerterm cost.

For instance, on the latter point, domestic and familial responsibilitiesas well as traditional divisions of labour limit migration opportunitiesfor women, particularly in sub-Saharan Africa. For example, Hampshire(2002) found that Fulani women have inequitable access to migrationnetworks used by men. Furthermore, relatively low levels of educationand lower wages at destination discourage female migration, as Agesaand Agesa (1999) found in their study of rural to urban migration inKenya. The impact of widespread male labour migration can lead tolong-term disempowerment of women who become dependent on menfor cash, as evidenced in Bihar in India as well as parts of eastern andsouthern Africa (Francis, 2002).

So, in a similar way to market- and state-based distribution systems,network-based systems represent both opportunities and constrains towelfare provision. The ‘trust’ pillars upon which they build are quali-tatively different to other distribution systems, and in this sense theycould be argued to be more stable and reliable over time. However,the level of provision is constrained by the homogeneity of the groupcharacteristics as well as by eligibility criteria.


5. Rules of inclusion and exclusion in a relationalframework

How can we analyse the relational content of these different distributionsystems? In our framework, distribution is carried out by organiza-tions composed of ‘rule-makers’ and ‘providers’ that admit or exclude‘claimants’ through certain criteria and processes (Schaffer, 1973, 1986;Schaffer and Lamb, 1981). Each system may be different in the way thatthe relation between claimant and provider presents itself. For instance,in a market-based system the rules for participation may be hidden andcovert (hiring and employment practices), whereas in a government-based system the criteria are more likely to be explicitly stated. Anyinstitution that provides goods, services or jobs sets up an access struc-ture that regulates how resources are obtained. This may be explicit orimplicit and through the market, or through non-market distributionsincluding personal distribution, or through networks. Each access struc-ture has three basic components described by metaphors (sometimesreal) of the ‘gate’, the ‘line’ and the ‘counter’ (Schaffer, 1973; Schafferand Lamb, 1974). Interactions at these points follow different types ofrules (Schaffer and Wen-Hsien, 1975). The ‘gate’ requires the claimant toclaim eligibility: admission rules. The ‘line’ is a mechanism that placeseligible claimants in some order: line rules. The ‘counter’ is the office ormeeting point where the actual allocation or refusal takes place, dictatedby rules concerning what types of data will be considered to prove eligi-bility: counter rules. Together, these components and their rules dictatethe structure of the ‘queue’.

Schaffer and colleagues initially developed this metaphor for non-market formal distribution systems, but we suggest that this applies toinformal systems and market-based systems of distribution as well. Themetaphor of the queue can be applied to a market system, where buy-ers access the market subject to certain admission criteria are ordered insome way, and then purchase the resource when they provide certainexchange values. In perfect markets, there are no admission criteria andthere is no ordering or waiting, and the only thing that buyers need topresent is an exchange value. In imperfect markets, however, admissionand line rules may apply and counter rules can become more complex.For example, in credit or insurance markets buyers often need to pro-vide collateral or evidence of health. In informal distribution systems,eligibility rules are often based on subjectivity (being known to the dis-tributors) and counter rules require merely the appearance of need (tothose distributing).

80 Frameworks

In a perfect market, or in a Weberian ideal type non-market distribu-tion system, success in the queue requires meeting the eligibility criteria,waiting in line and producing the correct data at the counter. Sinceclaimants know the rules (and therefore the cost of applying) and thevalue of the resource that is offered (distributed, sold or given away),they can calculate whether or not to enter the market or apply forthe resource. However, the ideal type queue, like the perfect market,never obtains in reality. Instead, we find ‘complex queues’ that candeviate from the ideal type (Schaffer and Wen-Hsien, 1975) in termsof structure, claimants’ behaviour and providers’ behaviour. These devi-ations may disadvantage migrants for a number of reasons, as alreadydescribed.

First, complex queues have different structures with supplementarygates of language, social class or ethnicity that can exclude migrants veryeasily. Queues may be contiguous: the resource sought in one queue maybe the eligibility criterion for joining another. For instance, acquiringcash under Mexico’s PROGRESA system required enrolment and atten-dance at schools and health clinics. Migrant households whose childrendo not attend school are formally excluded from this cash distributionsystem, despite their poverty (Stecklov et al., 2005).

Second, claimants in complex queues deviate from their ideal type.They may not have the basic ‘cultural resource’ of knowing whichqueue to join (Schaffer, 1971, p. 14), as in distributions from patronswhose habits are known to few recipients, and rarely to migrants. Theymay not know or understand their relationship to the admission rules(see also de Haan and Maxwell, 1998; and Sen, 2000), and line rulescan be ‘strange, unfamiliar and frightening to applicants’ (Schaffer,1973, p. 8). Even if the migrant does know the requirements, at ‘thecounter . . . [claimants] may not know how to present the information.The needier the [claimant] the greater this difficulty tends to be. Thisis a problem of literacy, time, research, legal processes, entitlementand registration conflicts, indebtedness, disputes, contests, and arrears’(Schaffer, 1973, p. 8). Studies indicate that migrants lack the knowl-edge, language skills and confidence required to access health care(Stephenson and Matthews, 2004) or food stamps (Schwartz, 2001).Migrants seeking to benefit from charity-based distributions may notknow where individuals distribute their donations, or may not be ableto gain access to a church to receive alms if they look different fromother churchgoers (and may be assumed not to be the relevant religiousdenomination) or cannot communicate effectively with church author-ities. Claimants may also feel put off by admission, line or counter rules


that carry additional connotations that suggest unfairness or stigma (seealso Sen, 2000).5

Third, providers exercise discretion over gate, line and counterinteractions, either because the rules require it (they are ambiguousor conflicting, see Schaffer, 1973) or because they have objectivesthat will not be met by following the rules (see Lipsky, 1980, fora more detailed treatment of discretion in the provision of publicservices). In the queuing process, claimants present themselves as sim-plified and compartmentalized ‘cases’, detached from their personalities,histories and stories (Lamb, 1975; Wood, 1986). In these circum-stances, migrants are often not able to demonstrate that they are theintended recipients (the income-poor, HIV/AIDS sufferers, the elderlyand so on).

Faced with these challenges, rule-makers may clarify access criteria(often under pressure from claimants or their representatives). Wherethis does not happen (perhaps because complex access structures limitclaims in a way that benefits rule-makers) providers resort to non-formal criteria in admitting claimants, deciding in which order to servepeople and accepting data. Very often, these criteria take the form ofexchanges, often pejoratively called corruption (Schaffer, 1986). Othercriteria include affinity, friendship, social ties and subjective percep-tions of influence, or of suitability for the resource (most commonlyin personal charity but present also in formal schemes).

5.1. Reconciling the relationship between claimant and provider

Queues for social provisioning are complex. The features of complexqueues suggest that claimants’ endowments and needs will not nec-essarily be reconciled easily with providers’ access structures (Schaffer,1971, 1973, and 1986). ‘Reconciliation’ will not involve a simple pair-ing of applicant characteristics with bureaucratic rules, or exchangevalues with resources, or need with availability, but will depend onclaimants’ power, interests, attitudes and strategies (see also Keeley andScoones, 2003, for parallels with complex policy processes). What strate-gies, then, can claimants deploy? And are any particular groups, such asmigrants, who are disadvantaged in deploying these strategies?

Claimants’ attempts to access distribution from which they areexcluded will tend to produce three types of action. They may exit fromthe system where the costs of deploying successful strategies are too high(or en masse as a form of boycott). They may remain loyal to the sys-tem that might involve adverse incorporation and dependency. They

82 Frameworks

may use voice, negotiation and protest (Hirschman, 1970; Schaffer andLamb, 1974).

As discussed in Section 4.2 above, loyalty can involve uncriticallyaccepting whatever investments (of time, money and so on) are requiredin order to obtain the resource. Claimants may do this because they feelthat protesting against the implicit or explicit rules of the distributionwill have longer term consequences or consequences in other queues.For instance, migrants may not protest against non-receipt of healthbenefits because they fear deportation. Alternatively, group membersmay not protest against unfair allocations of benefits for fear of beingforced out of the group. Or it may be because claimants feel they lackthe right to protest, as when they receive charity from individuals onan ad hoc basis. Wood (2003) proposes that poor people are dominatedby dysfunctional time preference behaviour (in the sense that pursuit ofcurrent consumption may outweigh future well-being), in which thepursuit of immediately needed security places them in relationshipsand structures which then displace the longer term prospects of a sus-tained improvement in livelihoods. This is likely intensified for poormigrants. For instance, irregular labour migrants in the domestic servicesector may be ‘bonded’ to their employers through positive dependencyor threat. Similar to the possibility of adverse incorporation into mar-ket and formal provisioning systems (described above), people may beunable to exit from a network-based system once they are embeddedinto it (also discussed in Section 4.3 above).

Under adverse circumstances, many claimants will choose the moreactivist, complex and pervasive option of voice (described in Section 3.2above). Schaffer and Lamb (1974, p. 79) argue that ‘[claimants] them-selves will be forced to employ voice so as to handle their opportunities,to defend themselves, to succeed, to refute stigma, etc., at all pointsof the access experience’. When denied opportunities to purchase ina market, buyers can protest at the high price of goods (this is espe-cially common where the government plays a role in setting prices),or organize in groups to boycott sellers or buyers (consider a strike) orengage in barter with the seller. Recipients of charity, such as beggars,may protest at small allocations or at being ignored in order to increasetheir gift. In networks, claimants may strategically emphasize their needto increase their share by advertising their poverty or bemoaning theirill luck. In terms of non-market systems, the recognition that claimants’voice pervades all access situations could complement social protectionliterature that refers to programmes’ ‘inclusion’ and ‘exclusion’ errors.Claimants use different forms of voice depending on their resources,


the rules of the distribution system and their relationship to them, andthe importance of the resource sought.

6. Conclusions

We believe that migrant welfare can best be secured by exploring andresponding to access constraints to social provisioning. The right orentitlement of migrants to social provision can be examined throughthe rules that structure access. While the analytical focus on eligibil-ity and data rules can be mundane, the questions raised are crucial indetermining access. Specified entitlements and rights are only as usefulas the access structures that surround them. Much of this turns on theability of individuals (formally entitled or not) to persuade employersand administrators of their eligibility for the resource. This approachsuggests, then, that getting the formal rules right is not enough. A rig-orous study of access must also examine the negotiations, bargaining,threatening and pleading that take place around those rules, betweenclaimants and providers.

With the above framework in mind it is possible to check whetheraccess structures enable those in need of social provisioning to obtainit and, in our case, whether migrants are disadvantaged in terms oftheir ability to acquire resources. To do this, one can examine the con-straints at various points of access to resources in markets and throughnon-market systems. For example, an imperfect market for insurancewill not yield effective social provisioning for low-income individuals ifthis requires providing an exchange value (a premium) that is beyondthe means of applicants, or if they are required to present identifica-tion documents (such as utility bills) that they do not have. Queuerules may disadvantage the vulnerable, and migrants specifically, as inthe USA where legal immigrants have to meet additional criteria (suchas employment duration) to qualify for food stamps, despite higheraverage poverty rates (Schwartz, 2001). Or in India where Smita (2006,p. 66) reports that poor seasonal migrants are excluded from most pro-grammes, despite being the ‘most disadvantaged’. Another example isfood distribution systems that require complex processes of registrationthat may exclude those with little time or who are illiterate, who againmay be those most in need of the distribution (see Chapter 7 on thePublic Distribution System in India). Or, we could ask, are some of thosewho gather outside a church for alms better able to reach the attentionof the alms-giver, perhaps because they are able to take their place at thefront of the queue through strength or know-how?

84 Frameworks

This chapter has set out a framework for analysing access to socialprovisioning through a range of distribution structures (we applythis framework explicitly in Chapter 7). Recognizing the connectionsbetween the sources of risk and threat that demand provisioning, andsocial provisioning itself, research must take a critical approach to pro-vision from markets, formal schemes, networks and personal charity.Given that migrants are often more exposed to threats, it is unsurpris-ing that they are less well incorporated into social provisioning. Just asin the job market, formal rules may exclude migrants from provisioning,while informal rules, discrimination or negotiations may disadvantagethem, or they may be adversely incorporated and dependent, lackingthe political, social and economic certainty to protest.

Indeed, the issue of migrants’ access to social provisioning is irre-vocably linked to politics. The government of Delhi has been wary ofissuing permanent ration cards to migrants because they are expectedto return home, and it was thought that formally entitling migrantsto ration cards would encourage migration, putting a greater strain onDelhi’s stretched public services or increasing competition with currentresidents for jobs. For similar reasons, in-migration hubs such as the UKand South Africa are concerned about extending social benefits to ‘for-eigners’. The issue concerns rule-setters’ interests and perceptions aboutstate sovereignty and the sanctity of territorial borders, as well as socialspending targets. Do they consider migration worth encouraging? Dothey desire migrants as potential political constituencies? Can they buildaccess structures to achieve these goals? Or do they, alternatively, per-ceive a value in a subdued labour force, and therefore design access struc-tures to disorganize, confuse and subdue migrants? On the other hand,do migrants have an interest in subverting the system and what incen-tives do migrants have to engage in ‘proper’ legal and political access ascompared to pursuing any means to ensure physical access? The vibrantdiscourse around social protection entitlements and rights could use-fully incorporate these questions in taking the debate on access forward.

Notes

∗ While overlapping, ‘social provision’ is a broader concept than ‘social protec-tion.’ The latter typically has a remit defined by instruments of provision and/ortargeted populations. In this chapter we use the term ‘social provision’ as we areinterested in the conceptual basis of access and distribution to all types of welfareprovision and all types of groups.

1. Pareto efficiency describes a situation where it is not possible to make anyonebetter off without making someone else worse off.


2. For a range of evidence supporting this theory, see Elson (1999) and Hodgson(1984).

3. We are grateful to Klaus Kapuy for pointing this out.4. See de Haan et al. (2002) on the importance of kinship networks in Mali for

access to food and housing, and Zohry (2002) on the Egyptian experience ofnetwork-based help with finding residence, employment and new commu-nity. Other examples are in King (2003), Mosse et al. (2002) and Rogaly et al.(2002).

5. The resource itself can also be stigmatized. In some African communities,accepting food marks poverty, where accepting cash does not (Dorward et al.,2006).

References

Agesa, J. and R. Agesa (1999) ‘Gender differences in the incidence of rural tourban migration: evidence from Kenya’, Journal of Development Studies, 35(6),36–58.

Baland, J. M. and J. P. Platteau (1996) Halting the Degradation of Natural Resources(Oxford: FAO and Clarendon Press).

Bevan, P. (2004) ‘Conceptualising in/security regimes’, in Gough et al. (ed.)Insecurity and Welfare Regimes in Asia, Africa and Latin America (Cambridge:Cambridge University Press).

Britto, T. (2005) Recent Trends in the Development of Latin America: An Analysis ofConditional Cash Transfers (Brazil: Ministry of Social Development).

de Haan, A. and Maxwell, S. (1998) ‘Poverty and social exclusion in north andsouth’, IDS Bulletin, 29(1), 1–9.

de Haan, A. et al. (2002) ‘Migration, livelihoods and institutions: contrastingpatterns of migration in Mali’, Journal of Development Studies, 38(5), 37–58.

Dorward, A., Sabates-Wheeler, R., MacAuslan, I., Penrose Buckley, C., Kydd, J. andChirwa, A. (2006) ‘Promoting Agriculture for Social Protection or Social Pro-tection for Agriculture: Strategic Policy and Research Issues Discussion Paper’,paper prepared for the Future Agricultures Consortium.

Elson, D. (1999) ‘Labor markets as gendered institutions: equality, efficiency andempowerment issues’, World Development, 27(3), 611–627.

Folbre, N. (1994) Who pays for the Kids?: Gender and the Structures of Constraint(London: Routledge).

Francis, E. (2002) ‘Gender, migration and multiple livelihoods: cases from easternand southern Africa’, Journal of Development Studies, 38(5), 167–190.

Gough, I. and Wood, G. (2004) ‘Insecurity and welfare regimes in Asia, Africaand Latin America’, in I. Gough, G. Wood, A. Barrientos, P. Bevan, P. Davisand G. Room (eds) Social Policy in Development Contexts (Cambridge: CambridgeUniversity Press).

Hampshire, K. (2002) ‘Fulani on the move: seasonal economic migration in theSahel as a social process’, Journal of Development Studies, 38(5), 15–36.

Heyer, J., F. Stewart and R. Thorp (2002) Group Behaviour and Development: Is theMarket Destroying Cooperation? (Oxford: Oxford University Press).

Hirschman, A. O. (1970) Exit, Voice, and Loyalty: Responses to Decline in Firms,Organisations, and States (Cambridge, MA: Harvard University Press).

Hodgson, G. M. (1984) The Democratic Economy (Harmondsworth: Pelican Books).

86 Frameworks

Keeley, J. and Scoones, I. (2003) Understanding Environmental Policy Processes: Casesfrom Africa (London: Earthscan).

King, R. (2003) ‘Across the sea and over the mountains: documenting Albanianmigration’, Scottish Geographical Journal, 119(3), 283–309.

Ku and Matani (2001) ‘Left out: immigrants’ access to health care and insurance’,Health Affairs, 20(1), 247.

Lamb, G. (1975) ‘Marxism, access and the state’, Development and Change, 6(2),119–135.

Lipsky, M. (1980) Street-level Bureaucracy: Dilemmas of the Individual in PublicServices (New York: Russell Sage Foundation).

Mosse, D., S. Gupta, M. Mehta, V. Shah, J. Rees and the Kribp Project Team(2002) ‘Brokered livelihoods: debt, labour migration and development in tribalwestern India’, Journal of Development Studies, 38(5), 59–88.

Però, D. and Solomos, J. (2010) ‘Introduction: migrant politics and mobilisation:exclusion, engagements, incorporation’, Ethnic and Racial Studies, 33(1), 1–18.

Polanyi, K. (1944) The Great Transformation (Boston, MA: Beacon Press).Rawls, J. (1971) A Theory of Justice (Boston, MA: Belkap).Rex, J. and R. Moore (1967) Race, Community and Conflict: A Study of Sparkbrook

(London: Oxford University Press).Rogaly, B., Coppard, D., Rafique, A., Rana, K., Sengupta, A. and Biswas, J. (2002)

‘Seasonal migration and welfare/illfare in eastern India: a social analysis’, inA. De Haan and B. Rogaly (eds) ‘Labour Migration and Rural Society’, Journal ofDevelopment Studies, 38(5), 89–114.

Sabates-Wheeler, R. (2005) Co-operation in the Romanian Countryside: An Insightinto Post-Soviet Agriculture (Lanham, Maryland: Lexington Books).

Sabates-Wheeler, R. (2008) report prepared for DFID, available from author.Sabates-Wheeler, R. and MacAuslan, I. (2007) ‘Migration and social protection:

exposing issues of access’, Development, 50(4), 26–32.Sawyer, M. (1995) ‘The operation of labour markets and the economics of equal

opportunities’, in J. Humphries and J. Rubery (eds) The Economics of EqualOpportunities (Manchester: Equal Opportunities Commission).

Schaffer, B. (1971) ‘Easiness of Access: A Concept of Queues’, IDS CommunicationPaper 104 (Brighton: Institute of Development Studies).

Schaffer, B. (1973) ‘Improving Access to Public Services’, IDS Discussion Paper 23(Brighton: Institute of Development Studies).

Schaffer, B. (1986) ‘Access: A Theory of Corruption and Bureaucracy’, PublicAdministration and Development, 6(4), 357–376.

Schaffer, B. and Lamb, G. (1974) ‘Exit, Voice and Access’, Social Science Informa-tion, 13(6), 73–90.

Schaffer, B. and Lamb, G. (1981) Can Equity be Organised? (Farnborough: GowerPublishing Company).

Schaffer, B. and Wen-Hsien, H. (1975) ‘Distribution and the theory of access’,Development and Change, 6(2), 13–36.

Schwartz, S. (2001) ‘Immigrant access to food stamps: overcoming barriersto participation’, Clearinghouse Review Journal of Poverty, Law, and Policy,September–October 2001, 260–275.

Self, P. (1986) ‘What’s gone wrong with public administration?’, Public Adminis-tration and Development, 6(4), 329–338.


Sen, A. (1981) Poverty and Famines: An Essay on Entitlement and Deprivation(Oxford: Clarendon Press).

Sen, A. (2000) ‘Social Exclusion: Concept, Application, and Scrutiny’, Social Devel-opment Papers No. 1, Office of Environment and Social Development, AsianDevelopment Bank.

Sen, A. (2009) The Idea of Justice (Cambridge, MA: Harvard University Press).Sharp, K. (2001) ‘An Overview of Targeting Approaches For Food-Assisted Pro-

gramming’, report prepared for CARE’s PHLS Unit (Atlanta, Georgia: CARE).Smita (2006) Locked Homes, Empty Schools: The Impact of Distress Seasonal Migration

on the Rural Poor (New Delhi: America India Foundation).Stecklov, G., Winters, P., Stampini, M. and Davis, B. (2005) ‘Do conditional cash

transfers affect migration? a study using experimental data from the MexicanPROGRESA program’, Demography, 42(2), 769–790.

Stephenson, R. and Matthews, Z. (2004) ‘Maternal health-care service use amongrural-urban migrants in Mumbai, India’, Asia-Pacific Population Journal, 19(1),39–60.

Stewart, F. (1996) ‘Groups for good or ill’, Oxford Development Studies, 24, 9–25.von Benda-Beckmann, K., van Eldijk, A., Spiertz, J. and Huber, F. (1989) ‘Interfaces

and Janus-Faces: A critical appraisal of the interface approach in developmentsociology from a social-legal studies perspective’, in N. Long (ed.) Encoun-ters at the Interface: A Perspective on Social Discontinuities in Rural Development(Wageningen: Agricultural University of Wageningen).

Whitehead, A. (1979) ‘Some preliminary notes on the subordination of women’,IDS Bulletin, 10(3), 10–13.

Wood, G. (ed.) (1985) Labelling in Development Policy (London: Sage).Wood, G. (1986) ‘ “Don’t give them my telephone number” – Applicants and

clients: Limits to Public Responsibility’, Public Administration and Development,6(4), 475–484.

Wood, G. (2003) ‘Staying secure, staying poor: the “Faustian Bargain” ’, WorldDevelopment, 31(3), 455–471.

Zohry, A. (2002) ‘Rural-urban Migration in Egypt and other Developing Coun-tries: A Statistical and Literature review’, in Rural to Urban Labour Migration:A Study of Upper Egyptian Labourers in Cairo, DPhil Thesis, University of Sussex,Chapter 3.

Part 2

Policies and Policy Spaces

3Social Security for Migrants:A Global Overview of PortabilityArrangementsRachel Sabates-Wheeler, Johannes Koettl and Johanna Avato

1. Introduction

The most recent Human Development Report (HDR) on Human Mobilityand Development (2009) estimates the total global migrant stock at214 million people, approximately 3 per cent of the world’s popula-tion. This represents a substantial increase since 1960, where the stockwas 75 million, and even more recently 2005, where estimates indicatedthat the total migrant stock was 191 million. Due to the underlying eco-nomic and demographic global imbalances, this trend is likely to persistand calls for policies that effectively manage migration to the benefitof all – migrants, origin countries and host countries (Holzmann andMuenz, 2004).

These substantial and increasing migration flows raise questions aboutsocial security for international migrants. The atypical life cycle ofmigrants requires special provisions for their social security to ensurethat they can adequately manage their risks. Migrants move betweencountries and hence between distinctively regulated labour marketsand social security systems, which creates specific vulnerabilities. Newlyarrived migrants are in a particularly vulnerable position as they areaway from their home community and have no access to importantinformal social networks and safety nets. In addition, access to for-mal social services in the new host country is often restricted formany reasons, such as informal labour market involvement, employermonopsony and delayed access until some months or years after arrival.Exacerbating this, many richer receiving countries have tightened theprovisions of their social security systems so that migrants cannot gainaccess.

91

92 Policies and Policy Spaces

At the same time, migrants might have contributed to formal socialsecurity systems in their country of origin or former host countries, yetany rights to benefits from these systems might cease to exist or sub-stantially diminish with their arrival in the new host country. Similarly,any contributions made to the social security system of the new hostcountry might be lost after the migrant departs because the associatedsocial rights and benefits might not be portable across internationalborders. Finally, migrants – in particular low-skilled, undocumentedmigrants – face challenging labour market conditions leading to exclu-sions from national security systems in host countries related to cross-border recruitment, information asymmetries between employers andmigrants, and visa requirements tied to employment.

Trends in international labour migration such as the increasing domi-nance of migrant workers, particularly women and children, in domesticwork, give rise to concerns that there is a trade-off between migrantnumbers and rights (International Labour Organization (ILO, 2008a).Various sectors in receiving countries are growing dependent uponmigrant labour – notably the service industry – and upon domesticwork in particular. The ILO reports that over the last 30 years regionsincluding Europe, the Gulf countries and the Middle East have seenmigrant women become the majority of domestic labourers. Workerssuch as these often find themselves at the intersection of various typesof informality relating to age, gender and sector, and thus tend to bebarred from contributing to social security systems in their host coun-try, and subsequently unable to claim any benefits when they returnhome (ILO, 2010). Out of the 63 countries surveyed by the ILO in 2009,23 considered domestic employees ineligible for basic protections suchas the minimum wage. In the resolution on migrant workers’ protec-tion passed at the 92nd Session of the International Labour Conference,2004, the ILO noted the challenge raised by these especially vulnera-ble groups. The resolution singled out female, irregular and temporarymigrant workers as especially in need of improved regimes for protectionand access to rights (ILO, 2004a).

An increasing proportion of migrant workers move as part of tem-porary migration schemes, designed by developed countries to remedylabour market gaps. These schemes pose difficult questions with regardto migrants’ integration in receiving countries, of which social securityis an important dimension. One current trend is the search for newarrangements to mediate the international mobility of such workers,such as Mode 4 of the General Agreement on Trade in Services (GATS),which governs service provision. However, there is concern (ILO, 2008a)

Rachel Sabates-Wheeler et al. 93

that treating migration as a trade issue leads inevitably to the devaluingof such workers’ protection. The ILO warns particularly of the impor-tance of keeping such negotiations transparent with regard to workers’rights, since the low-skilled workers who might become subject to suchagreements are those least able to safeguard their own rights.

Work by Avato and colleagues (2010) shows that it is predominantlyNorth–North migrants who enjoy access to and portability of socialbenefits, which translates into 23 per cent of all migrants worldwide.The most disadvantaged migrants are those moving within low-incomeregions. In these regions, formal social security provisions are less devel-oped, and migration is characterized by high numbers of undocumentedmigrants. The lack of access to social services and lack of portability ofsocial rights for migrants not only raises concerns about the vulnerabil-ities of migrants, but also creates distortions in labour markets and inmigration decisions. If migrants do not fully benefit from social secu-rity contributions or tax contributions because the associated benefitsare not accessible or not portable, they might decide to avoid contri-butions and work informally or understate earnings. If migrants havemade considerable contributions, but the acquired social rights are notportable, migrants’ decisions to return to the home country or to stayin the host country might be biased towards the latter because of theexpected income loss due to, for example, forgone pension benefits.Lack of portability of social rights could, therefore, undermine returnmigration and deprive origin countries – many of them developingcountries – of important beneficial development effects.

The rest of this chapter is organized into six sections. First we reviewwhat is meant by social protection and how it relates specifically tomigration. We then focus on the ways in which migrants are able toaccess (or are constrained from accessing) social security arrangements.This is followed by two related sections dealing with the coverage ofexisting bilateral and multilateral social security agreements. The finaltwo sections draw conclusions, first for the special case of South–Southmigrants and finally for more general policy implications.

2. Social protection for migrants

Social protection for international migrants consists of four compo-nents:

(i) access to formal social protection – that is, social security and socialservices – in host and origin countries;


(ii) portability of vested social security rights between host and origincountries;

(iii) labour market conditions for migrants in host countries and therecruitment process for migrants in the origin country; and

(iv) access to informal networks to support migrants and their familymembers.

For the purposes of this chapter we focus only on the first threecomponents.

First, access to social services is crucial for migrants as it impactstheir level of vulnerability. Social services include health-care benefits,long-term social security benefits such as old-age and disability bene-fits, and short-term benefits such as social assistance, maternity andunemployment benefits, family allowances, as well as public housingand education. Migrants often do not benefit fully from these social ser-vices, either because access is only granted some time after arrival, orbecause family members are spread across various countries. If migrantsfail to generate sufficient income to cover all their needs – and in manycases the needs of their family left behind – their situation worsens sig-nificantly if they have no access to safety nets (that is, social assistance).At the same time, according to Borjas’ selection model, countries withgenerous social security systems could attract low-skilled immigrantsthrough unintended self-selection, which is why many host countriesfollow policies of limited or delayed access (see Borjas, 1987).

Second, portability of social security rights is important to migrants toavoid financial losses, but also to social security institutions out of con-cerns for actuarial fairness. Portability is the ability to preserve, maintainand transfer vested social security rights or rights in the process of beingvested, independent of nationality and country of residence (Cruz,2004; Holzmann et al., 2005). Portability is particularly important forlong-term benefits with pre-saving elements that are either explicit (asin the case of old age pensions) or implicit (as in the case of health care).In the absence of portability, migrants run the risk of financial loss whenleaving their host or home country. For example, migrants contributingto an old age pension scheme in their host country could lose part oftheir contributions and benefits when returning to their home country.Similarly, migrants contributing to health insurance in their host coun-try could lose coverage when moving back to their origin country. Theymight find it difficult to get affordable health insurance in their origincountry after return, particularly when close to or during retirement. By


default, they could end up benefiting from the origin country’s socialsystem or health system, although they might have spent most of theirproductive life working abroad and contributing to a foreign social sys-tem. This could have important fiscal implications for social systems inorigin countries.

It is important to distinguish portability from exportability of socialbenefits. In order to achieve full portability, some cooperation betweenthe social security institutions of the origin and the host country isrequired in order to ensure a joint determination of benefit levels fora particular migrant. In the case of pensions, for example, this is donevia a totalization of periods of contribution in both countries. That is,social security contributions from both countries are taken into accountwhen determining eligibility for a pension and the pension amount.1

Third, labour market conditions are an essential component of thesocial protection for migrants. Migrants are often disadvantaged in for-eign labour markets due to information asymmetries and monopsonypower of employers. Countries have created labour market policies toovercome failures in the labour market in order to strike a balancebetween the needs of employers and the need for protection of workers.These policies – enacted and implemented on the national level – regu-late the workings of labour markets, that is, hiring and firing conditions,minimum wages, benefits and other rights for workers. For migrants,though, who are often recruited in the origin country to work in thelabour market of the host country, many of these regulations might beundermined because of substantial information gaps. These informationgaps can be exploited by employers, recruiters and middlemen who usethe promise of high earnings to extract unjustified fees from migrants orto offer them unfavourable work contracts. In addition, if work permitsand visas are tied to work contracts and migrants are facing limitationsto choose their employer freely, the bargaining power is tilted towardsthe employer, creating possibilities for exploitation and abuse. Finally, ifimmigration policies are too restrictive vis-à-vis the demand for foreignlabour, employers and migrants face incentives to meet in the infor-mal labour market and the labour market for undocumented migrants,which is not subject to any regulations and might provide weak or onlyinformal social security.

In principle, access to social services, such as health and education,is governed by the UN International Convention on the Protectionand Rights of All Migrant Workers, which in 2007 was ratified by37 (mainly low- and middle-income) countries. High-income countries


were reluctant to sign the Convention, perhaps because it provides(too) many entitlements for migrant workers. Clearly, the issue ofmigrants’ access to social security is irrevocably linked to politics andthe way that political opinion affects the interpretation of rules andrights. For instance, in-migration hubs such as the United Kingdomand South Africa are concerned about extending social benefits to ‘for-eigners’ because it is thought that formally entitling migrants to benefitscould encourage migration, putting a greater strain on stretched pub-lic services or increasing competition with current residents for jobs.This issue involves rule-setters’ interests and perceptions about statesovereignty and the sanctity of territorial borders, as well as social spend-ing targets. Also, the way the migrant is ‘constructed’ within the localculture can facilitate or constrain access for migrants to a range of socialprovisioning. Drawing on qualitative surveys described in Chapter 9,Box 3.1 illustrates the way that access to social assistance can differ bycountry of destination.

Box 3.1 Differential access to welfare services at destination

Access to social welfare is highly dependent upon the welfare stateand supporting legislation and structures within each country.For instance, many migrants were able to access state health andeducation services in the United Kingdom (UK), but migrants inSouth Africa were largely excluded from these services. In bothcases, however, access to transnational social networks arguablyserved as the migrants’ main ‘safety net’ against the risks posedby international migration. This difference is in large part dueto the fundamentally different welfare systems of the receivingcountries. In the United Kingdom, free public health care andschooling based on universalistic principles has led to limitedpolicing of ‘foreigner’ abuse of the system. To be enrolled inschool all that is required is evidence of permanent residence(and address). The immigration department does not regularlyshare information with the education and health departments. Onthe other hand, in South Africa health and education are largelywithin the private sphere. Furthermore, identity (ID) cards arerequired for registration for all social services. Where ID numbersare less that 13 digits long (indicating non-citizen status), rightsand entitlements are severely restricted. Irregular migrants haveextremely limited opportunities to access services.


Speaking about the UK health and education systems, aMalawian migrant claims:

It is not difficult for the children of illegal migrants to get into aschool or even for illegal migrants to get seen by an NHS doctorbecause schools and surgeries don’t ask questions about the statusof people who come to them. There is no communication betweenthe Home Office and Education and Health. So, even if you are anillegal, the school or surgery won’t ask for any documentation. Youare free to register.

The past lax immigration laws have also allowed otherwise non-residents access to the education system.

Others have made arrangements like my wife’s. She brought her sis-ter’s children to the UK as her own. No questions were asked at theairport: she simply claimed that they were her own children. Nowthey are both in school, living with their mother. It is common amongmigrants to do this for each other.

The South African situation is very different. An irregular femalemigrant recounted that:

I came to South Africa because my husband was here and I waspregnant with our second child. We didn’t know that you needed aSouth African ID number to go to hospital. Luckily, when the timecame to have the child, I used the ID of a Malawian woman whohas managed to get an ID. We look nearly the same, so I was ableto deliver the baby in hospital.

Source: Sabates-Wheeler (2008): transcripts available from author.

3. Access to national social security schemes

Existing social security measures for workers are often a problem formigrants in developed countries. A Canadian report (Elgersma, 2007)notes that migrant workers who are legal derive no benefit from basicemployment insurance and often work in non-unionized sectors, reduc-ing their recourse in cases of employment or other economic shocksand generating systemic instability in certain sectors. Even in a nation


as wealthy as the United States (US), the comparative financial disadvan-tage of being undocumented, and the concurrent lack of access to basicsocial security programmes that cover poor families, leads to concernssurrounding health and nutrition for this group in particular. A 2008study (Hadley et al.) indicates that among undocumented migrants liv-ing in the United States, uncertain and unpredictable work schedulesand limited access to public assistance seem to contribute to high lev-els of food insecurity and hunger, which in turn may also negativelyaffect mental and physical health. The study also finds that the longeran undocumented migrant lives in the United States, the poorer theirhealth indicators.

Formal social security for international migrants is essentially a mat-ter of national legislation. The host country regulates what benefitsmigrants have access to and under what conditions. However, evenwhere national laws dictate that the law must apply equally, protec-tion for migrant workers is limited. This has to do with the applicationof the law but also with the high proportion of migrants locatedwithin informal labour markets. An ILO Migration Survey (ILO, 2004b)showed that one-third of countries surveyed did not apply their anti-discrimination laws to migrant workers, meaning that minimum wagelegislation and access to social services could not be enforced by them,or on their behalf. Furthermore, according to the survey, many statesdictate that service providers must call the immigration authoritieswhen migrants try to appeal for the provisions for equal treatment tobe enforced, making it unlikely that irregular workers, in particular,would be able to access their rights. An ILO report (2003) estimatedthat only around 10 per cent of agricultural workers globally wereorganized, something especially problematic considering that agricul-ture has a disproportionately high number of workplace hazards butvery few sources of compensation for non-citizen workers. Exacerbat-ing this is the lack of unionization characterizing informal migrantworkers.

Where formal legislation is effective, it defines what benefits can bereceived after leaving the country. The European Union (EU) has themost advanced and complex system of portability of social benefits.EU nationals enjoy full non-discriminatory access to all and portabilityof most social benefits. With respect to third-country nationals, equal-ity of treatment is granted after a certain period of residence (no laterthan after five years according to EU Directive 109/2003). This meansthat even third-country nationals enjoy full access to and portability ofsocial benefits within the EU no later than after five years of residence.


Additionally, EU nationals can export their pensions to literally anycountry in the world. The coverage of health care outside the EU is muchless developed.

Canadian and Australasian social security systems also include pro-visions for international migrants. Canada allows access to the tax-financed universal pension, health-care benefits and the earnings-basedpension to all residents, including most migrants.2 All migrants inNew Zealand have immediate access to education, accident compen-sation, public health services, and in some cases emergency benefitsfor special hardship. After two years of residence, migrants gain accessto services such as housing assistance, unemployment benefits, sick-ness benefits and interest-free student loans for tertiary education. InAustralia, access to social services differs for temporary and permanentmigrants. Temporary migrants have no immediate access to social secu-rity benefits and public health services. However, upon leaving Australia,they are refunded for the contribution they were required to pay. Per-manent migrants have access to some benefits immediately and to thefull range of social services after 104 weeks of residence. Similarly toCanada, Australia has a dual social security system, including a means-tested national pension and a mandatory, earnings-based occupationalpension.

Two categories of social security systems exist in the main EastAsian host countries and economies. Hong Kong (China), Japan andthe Republic of Korea have multi-tiered schemes, consisting of a basicpart covering all residents and an occupational scheme. Malaysia andSingapore have provident funds that collect resources not only for retire-ment, but also for financing health care, education and housing. Thesocial security provisions in these countries allow some limited porta-bility of long-term benefits, either in the form of pensions paid abroad(Republic of Korea) or in the form of lump sum payments when leavingthe country for good (Malaysia).

The Gulf Cooperation Council (GCC) countries in the Middle East arespecial in the sense that they do not grant immigrants any access to thesocial system.3 Migrants (and their employers) do not have to contributeto the system (at least not directly), so the issue of portability of bene-fits and lost contributions is not relevant. This leaves the migrant withthe responsibility to provide for certain benefits like old-age or disabilitypensions on his or her own. If the migrant’s origin country is sufficientlydeveloped, this could happen in the form of voluntary contributions tothe public system of the origin country or by buying financial prod-ucts from the private insurance market. Regarding health care, the visa


sponsor of the worker is obliged by law to provide health insurance forthe migrant, but it is unclear how well these provisions work.

In middle-income countries, much less is known about provisions innational law regarding the access to social services and exportabilityof benefits for migrants. The middle-income regions of the Caribbean,Eastern Europe, Latin America and North Africa all have well-developedsocial security systems with relatively good coverage of the native labourforce. Some of the countries in these regions host significant numbersof immigrants, for example, Argentina, Belarus, Russia, Turkey, Ukraineand Venezuela. The social security provisions for immigrants in thesecountries are likely to be similar to those of high-income countries withcontribution-based social security systems.

The low-income regions of Central Asia, South Asia, South-East Asiaand sub-Saharan Africa include various major host countries – such asBurkina Faso, Côte d’Ivoire, India, Kazakhstan, Pakistan, South Africa,Tanzania, Thailand and Uzbekistan – yet even less is known about thesocial protection status of their immigrants. Many of these host coun-tries have weak social security systems that cover only a small portionof the labour force. Large numbers of immigrants are undocumentedand participate in the informal sector, which raises concerns about theirsocial protection status. Most formal social security is provided throughtax-financed social assistance and migrants seem, by and large, to beexcluded from these benefits.4

An important regional effort affirming the significant contribution ofmigrant workers to the society and economy of both host and send-ing states has been made by the Association of Southeast Asian Nations(ASEAN). In 2007 the Cebu Declaration on the Protection and Promo-tion of the Rights of Migrant Workers acknowledged the difficultiesthat migrant workers and their families often encounter in exercisingtheir rights. The Declaration committed all the ASEAN countries tostrengthen the protection afforded to migrant workers (see Tamagno,2008).

At the global level, legal provisions relating to social security forinternational migrants are scarce, with the exception of bilateral (andmultilateral) social security agreements.5 These currently constitute thebest practice on how to coordinate access to and portability of socialbenefits for migrants.

4. Bilateral and multilateral social security agreements

Bilateral social security agreements usually include provisions on non-discrimination between nationals and migrants with respect to social


security and rules of cooperation between the social security institu-tions of the signatory countries. The latter coordinate the totalizationof periods of contribution that migrants accrue in the two countriesand regulate the transfer and payment of acquired social security enti-tlements. Most agreements refer to long-term benefits like old age,disability and survivors’ pensions and other annuities. Health-care ben-efits are subject to social security agreements to a much lesser extent.Also, purely tax-funded – as opposed to contributory – benefits such associal assistance or maternity allowances are usually explicitly exemptfrom portability.

Social security agreements are also arranged on the multilateral level,as shown by the EU, CARICOM (Caribbean Community), MERCOSUR(Mercado Común del Sur) and, in the future, the Ibero-AmericanSocial Security Convention. The EU is also leading efforts to enhancesocial security cooperation within the Euro-Mediterranean Partnership(EMP).6 Social security agreements with Morocco, Tunisia and Algeriahave been concluded under this initiative.

The ILO has noted its concern that such multilateral agreements,where they do not involve countries with comparable rights regimes,should give attention to workers’ protection. Its ‘Multilateral Frame-work on Labour Migration’ provides guidelines on the equal treatmentof migrants in temporary worker schemes, focused on establishing andenforcing international standards (ILO, 2006). While several processesexist for establishing and discussing migrant workers’ rights, forums forrealizing and enforcing them may still be lacking. The ILO (2008a) sug-gests that some candidates include the Global Migration Group, formedin 2006 to coordinate responses within the United Nations system, andthe Global Forum on Migration and Development, a multinational con-sultative process first held in 2007. The ILO also notes that RegionalConsultative Processes such as the Bali, Budapest and Puebla Processesmay offer a framework within which to address these issues. However, itwarns that so far they have been insufficiently transparent and inclusiveto serve as forums for discussions of workers’ rights, and furthermorehave been mainly devoted to security issues (ILO, 2008a).

Outside this multilateral framework, many EU member states havealso concluded bilateral social security agreements with non-EU coun-tries and have created an extensive global network of portabilityarrangements. The United Kingdom, which receives and sends largenumbers of migrants, is a good example of an EU country havingextensive national, bilateral and multilateral legislation in place.

In the Latin American and Caribbean region (LAC), migrants cantake advantage of social security provisions that have been established


in the multilateral frameworks of CARICOM and MERCOSUR. To theextent that these countries have social security provisions, accrued ben-efits are made portable within the regions, though limitations apply.The MERCOSUR agreement came into force only in 2004, while theCARICOM Agreement on Social Security (CASS) was concluded in 1996.7

Forteza (2008), however, concludes that its impact has been limiteddue to factors such as differing social security systems and the lack ofawareness among the populations.

More recently, the Ibero-American Social Security Convention hasbeen signed; it includes 19 LAC countries8 as well as Andorra, Portugaland Spain. The arrangement will mostly apply to contributory systemsof social security, including disability benefits, old age pensions, widows’pensions and workers’ compensation.9 The degree to which countriescoordinate the portability of social security benefits via bilateral agree-ments varies greatly across regions. For example, the EU and otherWestern European countries have concluded 1628 bilateral social secu-rity arrangements – either through bilateral or multilateral agreements –of which 1034 are intra-EU arrangements. East Asian and Pacific (EAP)countries, on the other hand, have concluded only 181 such arrange-ments, although they provided the highest share (22 per cent) of allmigrants worldwide as of 2000. South Asian countries only concludedthree arrangements, and even though sub-Saharan African countrieshave concluded 177 arrangements it should be noted that a large num-ber (75) have been created by Reunion, which is counted as part ofFrance in all French agreements.10

5. Migrants potentially covered by international socialsecurity agreements

This section elaborates on the data regarding bilateral and multilateralsocial security agreements and provides global estimates on how manymigrants are potentially covered by such agreements and how manymigrants have access to social services in host countries.

Following Holzmann and colleagues (2005), bilateral migrant stocksare categorized into one of four regimes:

Regime I includes all legal migrants enjoying indiscriminate accessto social security in their host country. In addition, home andhost country have concluded a bilateral or multilateral social secu-rity arrangement guaranteeing that benefits are payable overseas(exportability), and also the social security institutions of both


countries jointly determine eligibility for and level of the benefit.Regime I is the most favourable regime in terms of formal socialprotection for migrants.

Regime II includes all legal migrants who have access to social servicesand social security in their host country in the absence of a bilateralarrangement between their host and origin country. In such cases,the national social law of the host country alone determines to whatextent benefits are payable overseas, which might result in limitedexportability of benefits. Totalization of periods of contribution is notpossible, so acquired social security rights are not fully portable.

Regime III includes all legal migrants who do not have access to socialsecurity in their host country – either because they are excludedor because there is no social security system in their host country.This is the case for most of the large migrant-receiving countries ofthe Gulf Cooperation Council (GCC) and some African countries.Access to other social services, like education and health care, mightbe granted. Despite the limited access to social security, an advan-tage of Regime III is that it does not require migrants to contributeto long-term benefits like old age pensions, and hence they do notrun the risk of losing benefits and rights associated with mandatorycontributions.

Regime IV includes all undocumented migrants who arguably face thegreatest challenge regarding their social protection. They have lim-ited access to social services and social security and work in a largelyunregulated labour market.

The tables below illustrate global coverage of social security for migrantsin different regimes. The global estimates on the status of social secu-rity for international migrants are based on data from the DevelopmentResearch Centre on Migration, Globalisation, and Poverty (MigrationDRC).11 For details of these estimates, see Avato and colleagues study(2010) which details the methodology and estimates of bilateral socialsecurity provisioning for different groups of migrants.

The main migration flows are from lower to higher income countries,but there is also considerable migration between lower income coun-tries. About 130 million migrants (70 per cent) originate from low- orlower middle-income countries, while they host only 40 per cent ofmigrants (see Table 3.1). In contrast, high-income countries host about93 million migrants (50 per cent) while sending only 19 per cent of

104Table 3.1 Global migrant stock estimates and bilateral portability arrangement by origin and host income group (2000)

Origin countryincome group

Host country income group

Low-incomecountries

Lowermiddle-incomecountries

Uppermiddle-incomecountries

Non-OECDhigh-incomecountries

OECDhigh-incomecountries

Total % globalstock

Low-incomecountries

23,339,921 10,989,418 4,832,109 3,332,615 11,128,178 53,622,241 29

Lowermiddle-incomecountries

4,817,600 27,815,316 9,099,077 5,592,653 29,162,504 76,487,150 41

Uppermiddle-incomecountries

691,741 2,095,454 1,447,944 1,110,490 16,293,585 21,639,214 12

Non-OECDhigh-incomecountries

195,520 1,270,164 332,871 205,867 3,844,703 5,849,125 3

OECDhigh-incomecountries

1,147,634 2,689,451 1,721,117 949,142 22,410,626 28,917,970 16

Total 30,192,416 44,859,803 17,433,118 11,190,767 82,839,596 186,515,700 100

% global stock 16 24 9 6 44 100

Number of bilateral portability arrangements concluded by countries in each income group:∗

92 404 823 461 1157 2937

∗Each bilateral portability arrangement is counted twice, once for both countries involved. Also, some bilateral agreements cover more than twocountries. This is the case if one or both of the contracting countries have overseas territories, which are usually included as national territory in theircontracts (for example, France, United Kingdom, United States). Some bilateral arrangements are part of a multilateral treaty, as in the EU.Source: Migration DRC (2007) and Avato et al. (2009).


migrants. Yet, when exploring the data in more detail, it also becomesclear that global migration is not a one-way street from lower to higherincome countries. Almost 67 million migrants from low- and lowermiddle-income countries live in another low- or lower middle-incomecountry. This is no surprise due to high intra-regional migration.

What is somewhat surprising, though, is that the large majorityof migrants from low-income countries live in another low-incomecountry – while for all other country-income groups, the majority ofmigrants go to high-income Organisation for Economic Co-operationand Development (OECD) countries. It seems that migrants from thepoorest countries tend to go to other, similarly poor countries.12 About23 per cent of global migrants move under the favourable Regime I thatpotentially provides them with full access and full portability of socialbenefits (see Table 3.2). The share of migrants under Regime I varies sig-nificantly among the regions. The EU-27 countries and other Europeancountries have 80 per cent of their migrants covered, followed by NorthAmerica with 68 per cent. Migrants from poorer regions are much worseoff: practically no migrants from South Asia, and only 4 per cent ofmigrants from sub-Saharan Africa move under Regime I.

Most migrants under Regime I, though, are from and go to high-income countries. More than half of the 43 million Regime I migrantsoriginate in high-income OECD countries. These countries cover86 per cent of their emigrants under this favourable regime. Also, thesocial security status of migrants seems to increase with migrants’ ori-gin income-country group. Moreover, almost all migrants (98 per cent)moving among high-income OECD countries – so-called ‘North–North’migration – are covered by bilateral agreements. Unlike migrants fromupper middle-income countries, migrants from developing countriesare very poorly covered by Regime I. In fact, the top migrant-sendingcountries with emigrant stocks between 6 and 13 million – Russia,Mexico, India, Bangladesh, Ukraine and China – have concluded nextto no bilateral portability arrangements.13 Thus, protecting their emi-grants through bilateral agreements seems indeed to be a practice that isprimarily common in high-income countries – migrants from develop-ing countries mostly fall into Regime II. Consequently, migrants frompoorer regions depend much more on national legislation regarding theprovision of social security. In fact, this results in another disadvan-tage because within Regime II, poorer migrants are more likely to moveshort distances and to other poor regions. While migrants moving tohigh-income countries have, by and large, access to better social securitysystems and services, those migrants moving to low- or middle-income

106Table 3.2 Global emigrant stock estimates by origin region and portability regime (2000)

Origin region Regime I Regime II Regime III Regime IV Total % global stock

East Asia and Pacific 3,189,217 15,588,651 825,255 3,609,755 23,212,878 12% total 14 67 4 16 100

Eastern Europe andCentral Asia

5,231,252 27,484,317 358,591 8,442,567 41,516,727 22

% total 13 66 1 20 100

EU-27 and otherEuropean countries

23,101,222 4,214,004 281,310 1,284,324 28,880,860 15

% total 80 15 1 4 100

Latin America andCaribbean

4,117,978 16,137,106 167,538 7,921,363 28,343,985 15

% total 15 57 1 28 100

Middle East and NorthAfrica

3,713,448 6,751,815 2,713,785 2,342,802 15,521,850 8

% total 24 43 17 15 100

North America 2,439,139 1,054,736 55,805 35,773 3,585,452 2% total 68 29 2 1 100

South Asia 20,105 16,528,148 4,413,451 4,220,898 25,182,602 14% total 0 66 18 17 100

Sub-Saharan Africa 714,570 14,104,664 627,117 4,824,994 20,271,345 11% total 4 70 3 24 100

Total 42,526,931 101,863,440 9,442,851 32,682,476 186,515,698 100% total 23 55 5 18 100

Source: Migration DRC (2007) and Avato et al. (2009).


countries will have access to only weakly developed social security sys-tems. In many lower and lower middle-income countries, migrants (andmost likely many nationals too) are often left in a situation where theycannot rely on the provision of public social services in times of need.

Nevertheless, some low and lower middle-income countries have suc-cessfully managed to protect their migrants by concluding bilateralsocial security agreements. For example, countries such as Morocco,Algeria and Turkey have managed to cover 89, 87 and 68 per centof their emigrants, respectively, under Regime I. In addition, there arealso the regional agreements of CARICOM and MERCOSUR countries,which have made efforts to grant non-discriminatory access to socialservices and make benefits portable for intra-regional migrants. Thesecases prove to be the exception rather than the norm. What then arethe options facing South–South migrants?

6. Options for coverage for migrants from the South

It is no surprise that southern countries have little to offer with respectto portability of benefits. Most benefits are tax-financed social assistancebenefits and social security often relies primarily on provident funds, ifthere are any. The weak development of social security provisions inlow-income countries and the lack of administrative capacity are likelyto be the reasons why poorer countries are not in a position to engagein bilateral or multilateral negotiations regarding the social securityfor their emigrants. Since the large majority of emigrants from low-income countries go to other low- or lower middle-income countries,the concern is not so much how developing countries can coordinatewith high-income countries in order to enhance the social security oftheir emigrants, but how low-income countries can coordinate amongthemselves to enhance the social protection of South–South migrants.

Regional efforts may help to bring about change in the approach tosocial protection across borders; however, their impact may be limitedfor the same reasons that deter the implementation of comprehensivesocial security systems. The Southern African Development Commu-nity (SADC), for example, has agreed on a Social Code which touchesupon migrants’ rights; it encourages members to protect their immi-grants, to give them equal access to the social security system, andto offer at least basic protection to undocumented migrants. Further-more, member states are encouraged to introduce, by way of nationallegislation and bilateral or multilateral arrangements, cross-border coor-dination principles, such as maintenance of acquired rights, aggregation


of insurance periods and exportability of benefits. However, the SocialCode is not a legally binding agreement and, given the status quo ofthe welfare systems in SADC, it may constitute wishful thinking. SouthAfrica has also entered into so-called labour agreements with a rangeof SADC countries, but most of the agreements do not cover publicsocial security schemes, they constitute merely employer-based occu-pational arrangements, and the enforcement of employers’ compliancewith such regulations is questionable. Also, it should be noted that allthese arrangements obviously do not provide the desired features ofcoordinating regimes, such as maintenance of acquired rights, aggre-gation of insurance periods and equality of treatment with nationals ofthe receiving country in social security matters.

Although protection for less-educated migrants is frequently lacking,the example of Sri Lanka demonstrates how origin countries can takeincreased responsibility for their migrants’ social protection, even in theabsence of receiving country commitments. The country responded tothe fact that migration for work often leaves migrants and their familiescut off from origin-country insurance systems by setting up an Over-seas Workers Welfare Fund to provide social insurance for migrants andfamilies left behind (ILO, 2008c). The fund covers payments to migrantsand their families in the case of death, disability or a need to cover travelexpenses.

Overall, the agenda to facilitate formal social protection for South–South migrants is not very well developed. Consequently, one questionis whether there are lessons to be learned from other regional agree-ments. The EU has probably been the most successful region in con-cluding multilateral and bilateral agreements. However, the EU is a fairlyhomogeneous region with similarly well-developed social systems. Onold age pensions, for example, most countries in the EU rely on definedbenefit systems. Equally well-developed systems are easier to connectand to coordinate, in particular with regard to complex actuarial trans-actions. Also, the capacity to administer such agreements, includingrecord-keeping and tracing contributions, as well as executing compli-cated actuarial operations like totalization of periods of contribution, iswell developed.

These observations about EU social security arrangements and the pre-vious analysis which mainly gave examples from SADC make clear thatmany low-income countries might be far from being ready to concludesuch agreements. Their welfare systems are too unprepared to engage inthese complex issues. Social security is not even ensured as a social rightfor citizens in many countries.


Hence, it seems that there is no strong model of how to move for-ward on social security provisioning for South–South migrants. Thisis not to say, however, that formal social protection for South–Southmigrants is irrelevant. Yet, the priorities are different to those in the caseof South–North or North–North migration and one has to differentiatewhen looking into policy implications. Given the limited provisions andlow overall coverage by social security of the labour force in low-incomeregions, it seems that concerns about the lack of portability of benefitsare premature. Instead, it seems that a more effective policy direction toimprove the social protection position of international migrants fromand within poorer regions would be to:

i) create a proper policy framework to manage migration in the region,in particular undocumented migration;

ii) focus on the social protection for the most vulnerable groups –women, children, refugees and undocumented migrants – byimproving their legal status in the host country and ensuring thattheir most basic rights are respected; and

iii) develop standards on how to coordinate social systems in the futureto ensure portability of acquired social security rights.

In other words, the critical issue is to find ways to bring the ‘uncovered’under a formal social security umbrella within the countries that theyreside. This may or may not require immediate attention to portability;it will depend on the country of interest and the state of social securityprovisioning within that country.

7. Policy conclusions

When the member states of the Global Forum on Migration and Devel-opment discussed migrant social protection at their 2008 meeting(GFMD, 2008), they did so within a framework that apportioned differ-ent responsibilities to states according to the migrants’ location and thestage of the migration process. For sending states, the discussion cen-tred around the provision of information, holding all those involved inthe chain of recruitment responsible where migrants are not awardedprotection, and both simplifying and making affordable the administra-tive processes migrants have to go through. Once migrants are abroad,their home countries are able to contribute to their protection by nego-tiating bilateral agreements with receiving countries (for example, theagreement between Spain and Ecuador, which incorporates guarantees


of social protection and access to benefits), and by establishing labourattachés at their consular services to advise migrant workers on theirrights. Some sending countries (Sri Lanka, Pakistan and the Philippines)have established welfare funds offering death and disability insurancefor migrants and their families. Receiving countries’ responsibilities wereoutlined as consisting mainly of ensuring effective provision and mon-itoring of migrants’ access to rights and services, and labour standards.The main avenues for provision include migrant resource centres, helplines and information in the migrants’ native languages.

The GFMD response to social provisioning for migrants is refreshingin its holistic response to the multiple vulnerabilities facing migrantsand their families. Importantly, it recognizes the different needs ofdifferent migrants at the various stages in their migration journey.Irregular migrants working in the informal sector have specific socialsecurity needs that are likely to be met through international con-ventions, national legislation, regulation and increased awareness ofrights. As noted in the introduction, the resolution on migrant work-ers’ protection (2004) argued that protection for less skilled migrantworkers has a high potential for poverty reduction via both earningsand remittances. Regular migrants working as professionals also benefitfrom legal provisions but are likely also to influence and be influencedby receiving country employment-related benefits, certain types of stateassistance and bilateral arrangements setting the parameters for transfersof benefits upon return.

Policy measures for northern countries primarily concern reforms ofcurrent defined benefit systems into more defined contribution-typewelfare systems, and move towards more actuarial structures. The goalwould be to improve individual and fiscal fairness as well as the adminis-trative burden associated with the coordination between social securityauthorities. More importantly, northern countries need to look intomodels of how to include health-care benefits into bilateral arrange-ments, a task that is certainly challenging due to the complexity ofhealth care systems. While bilateral social security agreements can cur-rently be seen as the best practice to enhance social security of migrantsfrom and to high-income countries, this might be insufficient in thecase of developing countries that do not have very well-developed socialsecurity systems. The analysis shows various examples of how differ-ent regional blocs and countries deal with formal social protection andportability. There seems to be no ideal model yet for regional economicblocs of mainly low-income countries on how to enhance the formalsocial protection for their emigrants and immigrants. Thus, the above


discussion leads to the conclusion that migration among developingcountries needs to be looked at from an entirely different perspective. Infact, migration itself provides one of the main social protection instru-ments for migrants and their families. The policy challenge is to makeSouth–South migration safer for migrants in order to maximize benefitsfrom this important livelihood strategy.

Some general policy conclusions can be drawn, which relate to allmigrants, with regard to improving their access to social security:

7.1. Ensure portability of those occupational benefits that arelegally also available to migrants, in particular workers’compensation, severance payments and benefits fromprovident funds

One opportunity to enhance the formal social protection for migrantswould be to improve the portability of some of the few occupationalbenefits that are available to formal workers, for example workers’ com-pensation benefits, severance payments and payments from providentfunds. These benefits are present in most low-income countries, andmigrants – legal and undocumented – frequently seem to make contri-butions to these occupational benefits. Although most of these benefitsare legally portable – in most cases benefits are simply paid out as alump sum – the provisions on cross-border payments seem to be poorlyimplemented, so that more often than not payouts from these ben-efits never reach migrants or their survivors in the migrants’ homecountries. Hence, it is crucial to ensure the proper implementation ofsuch provisions as a first step to improving the formal social security ofmigrants.

Italy has been innovative in offering pension portability to non-EUworkers: since 2002, employers have paid social security contributionsfor migrant workers separately from their citizen employees, throughthe National Social Insurance Institute (INPS), which then transfers theircontributions to the social security authorities in the migrants’ coun-tries of origin. The law also covers the families of permanent residentforeign workers, and offers survivorship or transferred pensions, whichare portable across national borders and can be claimed at the Italianretirement age of 65, regardless of national laws.

7.2. Ensure basic human and social rights to all migrants,independent of their status

Given the large number of undocumented migrants and their weak legaland social position, it appears that the most effective way to enhance


the social security of migrants is to ensure basic human and social rightsfor them. Due to their undocumented status, many migrants easily fallvictim to exploitation, particularly vulnerable groups like women. Otherproblems migrants may face are lack of freedom of association, isolation,exclusion and xenophobia. These problems erode the benefits of migra-tion for all parties and seriously undermine its development impact,including productive employment and decent work. Governments needto be serious about upholding human rights laws such that migrantsare entitled to basic services, freedom from exploitation and the rightto a fair trial. Taking the Decent Work Agenda seriously must trans-late into more labour market monitoring and ensuring workplace safety.Granting basic human and social rights is also in the interest of nativeworkers because without such provisions employers have an incentiveto favour more vulnerable and thus cheaper labour over native workers.In other words, discrimination against migrants in terms of social rightsalso makes migrants cheaper for employers – at the expense of nativeworkers.

Where necessary, countries of origin and destination need to put poli-cies, legislation and mechanisms in place that provide migrant workersprotection and support from any abuses in the labour migration process.Perhaps the best way to move forward on a basic human and social pro-tection rights agenda is to establish a country-specific minimum ‘socialfloor’ which all people have a right to regardless of residency status orlegal status. Such a floor would necessarily include access for migrantsto emergency health care and access to education for children.

Canada’s Seasonal Agricultural Worker Program offers a best prac-tice example for a number of reasons. First, the rules surroundingit give migrants rights to social protection that are similar to thoseof Canadian workers. Second, the government involves employers indesigning and implementing the programme, and gives administeringagencies discretion in implementing the rules (Martin, 2007).

7.3. Support migrant networks and associations

Access to informal social protection is critical for achieving positive out-comes from migration, particularly for poorer migrants. Encouragingand facilitating social networks and informal social security will helpmigrants reap the benefits from migration. For example, governmentscould actively support and provide funding for migrant associationsas well as migrant rights non-governmental organizations (NGOs) as away of supporting current migrants – or at the minimum should notundermine informal migrant networks and instead provide them with


an environment where they can duly develop. Informal networks couldalso be used as a channel to improve safety during the migration pro-cess. They could be used to inform migrants about benefits, costs andrisks of migration during and ideally prior to migration. Governmentsin both sending and destination countries may also attempt to connectwith migrant networks to learn details about exploitation incidents –for example, at the workplace – and feed this information into theirmonitoring process of labour market conditions. One particular con-cern about informal networks is that their usability for migrants couldcrucially depend on the migrants’ duration of stay: a sufficient lengthof stay is important so that migrants have to time to build strong tiesamong themselves.

Clearly it is politically impossible for governments to openly encour-age and support the activities of undocumented migrants. However,by supporting registered and recognized migrant groups, localities anddiaspora networks, policies will have the indirect effect also of strength-ening avenues by which undocumented migrants can report instancesof abuse and exploitation as well as claim basic rights, and at leastfacilitate spaces where migrants can access networks.

Many migrants are primarily concerned with the welfare of theirfamilies back home and remittances are the means by which fami-lies are supported, homes are built, basic needs are met, children areschooled and investments are made. Migrants, especially the undocu-mented ones, are often denied access to the financial system (be it abank, a post office account or a financial intermediary), which is of cru-cial importance for all migrants who want to safely transfer remittancesto their families. Governments and donors must think creatively aboutsecure and efficient ways of encouraging and facilitating access to remit-ting services for all migrants. This would be an obvious advantage for themigrant and for the destination economy. It may likely have a longerterm spin-off effect whereby migrants begin to save and access creditlines.

7.4. Develop a migration policy framework

Given that migration is, first, not a unidirectional, permanent phe-nomenon, and, second, it is, to a large extent, informal and tempo-rary, demand-driven migration programmes might offer the best policyframework for migration. Many migrants do not want to stay in thehost country for very long but rather come to earn money for cer-tain purposes and leave after having earned enough. Thus, a temporarymigration scheme that would grant work permits to migrants to work


in seasonal jobs and specific sectors for a limited time would be one wayto move forward. Additionally, a migration framework should considerthe significant number of regular cross-border traders. These circularmigrants, who contribute significantly to the destination economy,should be equipped with trade visas to enter the country legally. Thesemeasures would lift many migrants to the status of legal migrants, whichmay make it easier for them to improve their livelihoods (at destination,in transit and at source) and eventually to improve their access to formalsocial security.

Notes

1. Exportability, on the other hand, requires no such cooperation as eligibilityand the level of benefit are determined by the social security institution ofone country alone.

2. Seasonal agricultural migrant workers are excluded. See United States SocialSecurity Administration (2007).

3. The GCC countries are Bahrain, Kuwait, Oman, Qatar, Saudi Arabia and theUnited Arab Emirates.

4. The shortcomings of social protection for migrants in low-income regionsare discussed in more detail below.

5. The ILO and the UN have adopted a series of conventions concerning socialprotection for international migrants, most notably the International Con-vention on the Protection of the Rights of Migrant Workers and Membersof their Families. This Convention, however, had limited success; it wasadopted by the UN General Assembly in 1990, but came into force only after13 years, with just 22 ratifying states. Similarly, the ILO adopted a numberof conventions dealing with non-discrimination and equal opportunity formigrants in their host countries, all of which suffered from weak support interms of ratification by member states. ILO Convention 157, for example,aims at establishing a global regime of portability of benefits, but has onlybeen ratified by three countries.

6. Algeria, Egypt, Israel, Jordan, Lebanon, Morocco, Syria, Tunisia, Turkey andthe Palestine.

7. Suriname is not included in the agreement as it has an incomparable socialsecurity system.

8. Argentina, Bolivia, Brazil, Chile, Colombia, Costa Rica, Cuba, DominicanRepublic, Ecuador, El Salvador, Guatemala, Honduras, Mexico, Nicaragua,Panama, Peru, Paraguay, Uruguay and Venezuela.

9. See Online Pioneer (2007). For a copy of the Convention in Spanish, seeOrganización Iberoamericana de Seguridad Social (OISS, 2007).

10. Numbers are based on data presented in the next section.11. See Migration DRC (2007), Global Migrant Origin Database. Available at

http://www.migrationdrc.org/research/typesofmigration/Global_Migrant_Origin_Database_Version_4.xls (accessed 29 March 2008).

12. The largest hotspots of migration between low-income countries are SouthAsia, West Africa and Central Asia.


13. Nevertheless, the efforts of Mexico and India to negotiate an agreement withthe United States, and India also with various European countries should beacknowledged.

References

Avato, J., J. Koettl and R. Sabates-Wheeler (2010) ‘Social Security Regimes, GlobalEstimates and Good Practices: The Status of Social Protection for InternationalMigrants’, World Development, 38(4).

Borjas G. J. (1987) ‘Self selection and the earnings of immigrants’, AmericanEconomic Review, 77(4), 531–553.

CARICOM (1996) CARICOM Agreement on Social Security, available at:http://www.sice.oas.org/agreements_e.asp (accessed 14 October 2008).

Cruz, A. T. (2004) ‘Portability of Benefit Rights in Response to External andInternal Labour Mobility: The Philippine Experience’, paper presented at theInternational Social Security Association (ISSA), Thirteenth Regional Con-ference for Asia and the Pacific in Kuwait, 8–10 March 2004, available at:http://www.issa.int/pdf/kuwait04/2cruz.pdf (accessed 22 October 2008).

Elgersma, S. (2007) ‘Temporary Foreign Workers’, report prepared for Parliamen-tary Information and Research Service, Canada, 2007, available at: http://www.parl.gc.ca/information/library/PRBpubs/prb0711-e.pdf.

Forteza, A. (2008) ‘The Portability of Pension Rights: General Principles and theCaribbean Case’, Social Protection Discussion Paper No. 0825, Washington, D.C.:The World Bank.

Hadley, C. et al. (2008) ‘Hunger and health among undocumented Mexicanmigrants in a US urban area.’, Public Health Nutrition, 11:151–158.

Holzmann, R. and R. Muenz (2004) Challenges and Opportunities of InternationalMigration for the EU, Its Member States, Neighboring Countries and Regions: A PolicyNote, Washington, D.C.: World Bank; Stockholm: Institute for Futures Studies.

Holzmann, R., J. Koettl, and T. Chernetsky (2005) Portability Regimes of Pen-sion and Health Care Benefits for International Migrants: An Analysis of Issuesand Good Practices (Geneva, Switzerland: Global Commission on InternationalMigration).

International Labour Office (ILO) (2004a) Towards a fair deal for migrant workers inthe global economy, Report VI, Sixth item on the agenda, Ninety-second Sessionof the International Labour Conference, 2004, Geneva.

——. (2004b) Migration Survey 2003: Country summaries (Geneva).——. (2006) ILO Multilateral Framework on Labour Migration: Non-binding

principles and guidelines for a rights-based approach to labour migration(Geneva). http://www.ilo.org/public/english/protection/migrant/download/multilat_fwk_en.pdf (accessed 24 August 2010).

——. (2008a) International labour migration and development: The ILO perspective,International Migration Brief, International Migration Programme (Geneva).Available at http://www.ilo.org/public/english/protection/migrant/download/mig_brief_development.pdf (accessed May 2009).

——. (2008b) Best practices in social insurance for migrant workers: The case of SriLanka, Working Paper 12, ILO Asian Regional Programme on Governance ofLabour Migration (Geneva).


——. (2009) Decent work for domestic workers, Report IV(1), Fourth item on theagenda, Ninety-ninth Session prepared for the International Labour Confer-ence, 2010, Geneva.

——. (2010) Decent Work for Domestic Workers. Report IV(1), Fourth itemon the agenda, Ninety-ninth Session of the International Labour Confer-ence, 2010, Geneva. International Labour Organization (ILO) (Geneva).http://www.ilo.org/wcmsp5/groups/public/—ed_norm/—relconf/documents/meetingdocument/wcms_104700.pdf (accessed 24 August 2010).

Martin, J. (2007) Towards Effective Temporary Worker Programs: Issues and Challengesin Industrial Countries, International Migration Paper No. 89 (Geneva, ILO).

Sabates-Wheeler, R. (2008) ‘Migration Outcomes, Legal Status and Social Protec-tion: A Quantitative Study of Malawian Returnees from South Africa and theUK’, report prepared for DFID, 2008, available from author.

Tamagno. E. (2008) Strengthening social protection for ASEAN migrant workersthrough social security agreements, Working Paper No. 10, ILO Asian RegionalProgramme on Governance of Labour Migration (Bangkok).

United States Social Security Administration (U.S. SSA) (2007) Social SecurityPrograms Throughout the World: Africa (Baltimore, MD: U.S. SSA).

4Political and RegulatoryDimensions of Access, Portabilityand Exclusion: Social Securityfor Migrants, with an Emphasison Migrants in Southern AfricaMarius Olivier

1. Introduction

This chapter focuses on rights, regulation and a normative understand-ing of migrants’ entitlements, in particular in relation to social pro-tection, and with specific reference to Southern African intra-SADC(Southern African Development Community) migrants. It explores theimportance of adopting a rights-based and regulatory approach, address-ing the need for appropriate international standards, regional instru-ments, constitutional frameworks and statutory contexts to be aligned,in order to ensure access to social protection and portability of socialsecurity entitlements.

The reality is often different. Limitations in social security law effec-tively restrict migrants’ social security entitlements. In addition, immi-gration law regulation and the status of migrant categories is often usedto effectively restrict access to and portability of social security entitle-ments, impacting in particular on vulnerable migrants such as asylumseekers and undocumented migrants.

2. Contextual framework: migration data, nature and trends

Worldwide, there are 175 million people – about 3 per cent of the world’spopulation – currently residing outside of their countries of origin. Ofthese, 90 million are migrant workers. For a range of reasons, reliabledata on the extent and volume of migration within and to SADC is hard

117


to obtain, even in the major migrant-receiving country in the SouthernAfrican region, South Africa. In Southern Africa, political migration haslargely been the result of instability in countries such as the DRC and,earlier, Angola and Mozambique. Economic migration appears to beby far the most prevalent form of migration. The majority of SADCmigrants target countries with better economies. Therefore, ‘the migra-tion flow is towards Botswana, Namibia and South Africa because thesecountries have stronger economies and also experience skills shortages.These countries, therefore, offer migrants better prospects for improv-ing their quality of life’ (Kaseke, 2007, p. 2). South Africa in particularattracts by far the majority of intra-SADC migrants.

From the available evidence, subject to some exceptions, it appearsthat most of the migration from SADC is actually to other SADC coun-tries: intra-SADC movement is therefore the prevailing characteristic ofmigration from SADC countries. In fact, migration has been a long-standing feature of the labour market framework in Southern Africa, inparticular as far as work on the mines and in agriculture is concerned.Apart from informal cross-border trade-related migration, work on themines, in particular in South Africa, served as a magnet for both internaland external migrants. From a historical perspective, as is supported bydata on modern-day migration movements within SADC (Crush et al.,2005, pp. 5–6), it could be said that systems of labour migration inSouthern Africa are deeply entrenched and have become part and par-cel of the generations-long movements of people, primarily in search ofbetter living and working conditions (Kanyenze, 2004, pp. 2–10).

While a large number of cross-border migrants in Southern Africaremain circular migrants and their visits to the host country are gen-erally seen as temporary, contrary to the increasingly temporary natureof international labour migration, migration patterns within SADC havelargely been characterized by their permanent or ongoing nature. Onceimmigration linkages are established, they are very difficult to break,and migration flows are almost impossible to reverse. This is particu-larly true of the mining and agricultural industries in Southern Africa.In fact, a recent five-country migration study in SADC indicated thatmigration is now clearly regarded as a career rather than a passingphase in the working lives of most people who have been migrating,despite the fact that they maintain strong links with the home country(Pendleton et al., 2006, pp. 2–3, 15). This also flows from the fact thatmore migrants from the respective countries are older, married and, inmost cases, heads of households. In addition, the same study indicatesthat many migrant-sending households have a migration ‘tradition’

Marius Olivier 119

which is passed on from one generation to the next – parents and evengrandparents worked outside the home country.

It is generally accepted that SADC-related migration is characterizedby several elements (see Kanyenze, 2004, pp. 1–2):

• Labour contract migration – in particular on South African mines;• Declining levels of legal migration to and within the region and the

increase in clandestine and undocumented (irregular) migration;• Substantial brain drain migration to countries outside SADC;• While most intra-SADC migrants continue to be male, there is an

increased feminization of cross-border migration;• Growth in intra-regional informal cross-border trade;• Despite the largely voluntary nature of migration within SADC, there

are two broad exceptions: increased human trafficking and massinternal and external refugee movements;1

• Remittances are a primary source of household income in migrant-sending SADC countries;2

• And, generally, growth in the volume and complexity of cross-bordermovements.

3. The precarious position of intra-SADC migrantsand prevailing policy approaches

Cross-border migration in SADC is characterized by the precarious posi-tion of those who migrate and their dependants. There are severalreasons for this. The inchoate immigration, social security and labourmarket frameworks applicable to migrants are major contributing fac-tors. For those who migrate, working and living conditions are often,and often have been, inadequate. Cross-border migrants are mostlyunskilled or semi-skilled and are typically found at the lower end ofthe labour market in receiving countries. Irregular migrants in particu-lar suffer exploitation of their workers’ and human rights. Migrants areespecially affected by the restructuring and conditions prevailing at theenvironments where they are usually employed. In the mining indus-try in South Africa, for example, between 1989–2000 a large numberof regular jobs were shed, causing a drop in mining jobs from almost422,000 to about 231,000, and there was little effort made, either bythe employers or on the part of the state, to ameliorate the effects ofretrenchments (Kanyenze, 2004, p. 16). Also, it has been noted thatthe mining sector in particular has a stubbornly high rate of disable-ment and deaths (Kanyenze, 2004, p. 17). In addition to their precarious


position in SADC labour markets, migrants also suffer from negativeofficial and community responses, which flow from severely restric-tive policy and legislative approaches and the widespread prevalenceof xenophobia, particularly in South Africa.

From a policy perspective, the SADC member states have no clearcommon approach towards immigration. Few, if any, countries haveproactive immigration policies. In fact, security concerns in the formof control and deportation appear to characterize their migration lawsand policies. Migration of people within the region is viewed as a ‘prob-lem’ rather than an opportunity. Governments in Southern Africa, andin other parts of Africa, do not as a rule understand migration as a liveli-hood strategy which is crucial for the welfare of migrants, as well ascontributing to the development of the host country.

The position of migrants is further complicated by the fact that thestatus they enjoy is based on which of the different categories of non-citizens, including non-citizen workers, they are designated as belongingto. National laws, and often also international instruments, differ-entiate between permanent residents, temporary residents, migrantworkers, refugees, asylum seekers and illegal/irregular/undocumentednon-citizens. The interplay between immigration and social securitylaws is crucial to the understanding of the status of non-citizen work-ers, and yet it is evident that these laws often espouse conflicting policyobjectives.

3.1. Social security law limitations

National social security laws often contain provisions which adverselyaffect the social security rights of non-citizen workers. Sometimesnationality conditions apply. Such conditions may exclude foreignersfrom the scope of application of the social security schemes. Alterna-tively, these conditions may restrict access to social security benefits, inparticular upon the departure of the non-citizen worker from the hostcountry.

In many of the systems in SADC, non-citizens are wholly or par-tially excluded from the sphere of social protection. However there areexceptions, notably Mauritius, where permanent residence status and anumber of years actual residence are sufficient for social security cover-age, or in SADC countries where non-citizens are entitled to participatein social insurance-based schemes, such as public pension schemes. InSouth Africa, non-citizens who have acquired permanent residence sta-tus are eligible for social protection on the same basis as South Africans,for purposes of both social assistance and social insurance. However,

Marius Olivier 121

other categories of non-citizens do not necessarily qualify for equaltreatment with South African citizens and permanent residents, forexample in the area of unemployment insurance. Workers’ compen-sation legislation to a certain extent restricts the right of a returningnon-citizen resident to claim benefits, but simultaneously provides forthe portability of benefits, albeit not on a regular basis. However, lit-tle use has been made of it, partly because of the mismanagement offunds transferred to Mozambique for this purpose and partly because ofmigrants’ unfamiliarity with this arrangement.

In accordance with the provisions of refugee legislation, refugees inprinciple enjoy full legal protection, which includes the rights set out inthe bill of rights chapter of the South African Constitution. Persons whohave obtained refugee status therefore qualify for the constitutionallyentrenched right to access to social security, including appropriate socialassistance, if they are unable to support themselves and their depen-dants. However, the social assistance legislation does not fully extendprotection to refugees. An apparent conflict thus exists between thetwo laws.

3.2. Immigration law restrictions

Entitlement to benefits is sometimes linked to the immigration statusof the worker. Typically, legal residence may be required and may besubject to the requirement that the non-citizen worker does not becomea burden on the state. In South Africa, for example, the Immigration Actof 2002 regulates non-citizens’ entry into and residence in South Africa.Given the specific legal requirements that have to be met, the majorityof SADC migrants to South Africa are employed in low-skilled jobs anddo not qualify for a general work permit under the available categories.

The Act distinguishes between permanent and temporary residence.Permanent residents are not regarded as ‘foreigners’ as they are grantedall the rights of a citizen, except for those that a law or the Constitu-tion explicitly ascribes to citizenship (for instance, the right to vote).As such they are the elite of non-South Africans. The second categoryconsists of temporary residents. The crucial provision in this regard isthat, in accordance with Section 10 of the Act, a temporary residencepermit is issued on condition that the non-citizen is not, or does notbecome a ‘prohibited or undesirable person’. In terms of Section 30, anundesirable person includes anyone who is likely to become a publiccharge. This may imply that a non-citizen is deemed to be undesirableand denied entry if they lack financial resources and are in need of socialassistance or welfare.


The picture described above generally applies to other SADC mem-ber states also. Economic grounds constitute grounds for exclusion. Asremarked by Klaaren and Rutinwa (2004, p. 55), ‘[N]early universally,the status of being likely to become a public charge leads to prohib-ited immigrant status.’ The same applies when a person has insufficientfunds.

In SADC countries, the employment of illegal/irregular/undocumentedforeign workers is also prohibited and may imply that such a workeris not entitled to the protection of the various labour and social secu-rity laws. However, a recent Labour Court judgement in South Africa(Discovery Health Limited v. CCMA & others, 2008) held that a foreignerwhose work permit has expired still has a valid employment con-tract and is entitled to the unfair dismissal protection provided forin South African labour laws. It is submitted that this judgement hasimportant consequences not only for labour law purposes, but also for(employment-based) social insurance arrangements.

Special provision is also made for refugees and asylum seekers inmost national systems. Unlike most countries with refugee camps, SouthAfrica’s policy promotes local integration, which allows refugees to settleanywhere in the country. As a result of this policy, refugees are requiredto survive without any assistance from government. They are allowedto work, but are sometimes by law excluded from working in particu-lar industries. Attempts to prevent asylum seekers awaiting processionof their applications from working or studying were not upheld: theseprovisions were held to be ultra vires the Constitution (Minister of HomeAffairs and Others v. Watchenuka and Another, 2004).

3.3. Marginalizing impact

The prevailing evidence is clear: through available legislative mecha-nisms SADC governments are able to give precedence to immigrationlaws over labour and social security laws. Secondly, the legislative pol-icy in both South Africa and the other SADC member states is gearedtowards restricting access, controlling movement and regulating pres-ence in the host country, and not towards honouring a human rightsapproach and encouraging and supporting migration. Immigration lawsand policy in SADC countries generally focus on the effects, rather thanthe underlying causes of migration. A recent study (Crush et al., 2005,pp. 10, 24) remarked that ‘[N]o country, with the possible exception ofBotswana, has migrant or immigrant-friendly legislation on the books.’

Thirdly, essential compliance with human rights standards set out ininternational law instruments may be absent, despite the fact that a

Marius Olivier 123

country may have ratified these instruments. For example, the UnitedNations Committee against Torture (2006, Articles 3, 2, 12, 13 and 16)recently made the following remark, with reference to South Africa’snon-compliance with the treaty provisions of the Convention againstTorture and other Cruel, Inhuman or Degrading Treatment of 1984, inrelation to migrants:

The Committee is concerned with the difficulties affecting docu-mented and undocumented non-citizens detained under the immi-gration law and awaiting deportation in repatriation centers, who areunable to contest the validity of their detention or claim asylum orrefugee status and without access to legal aid. The Committee is alsoconcerned about allegations of ill-treatment, harassment and extor-tion of non-citizens by law enforcement personnel as well as withthe absence of an oversight mechanism for those centres and withthe lack of investigation of those allegations.

4. Regulatory responses

4.1. International standards

4.1.1. General

International norms and standards have already had a significant influ-ence on the regulation of labour markets and, to a more limited extent,on the social security frameworks in Southern African countries. It issuggested that much more could be achieved to further strengthen socialsecurity systems in SADC and to improve the position of non-citizens.There is indeed a range of relevant norms available in the interna-tional realm, in addition to extensive and specialized technical supportprovided by international agencies, which could be of assistance.

Constitutions in the region, such as the constitutions of Malawi,Namibia and South Africa, do sometimes make provision for the roleof international law, which potentially impacts on the terrain of socialsecurity for migrants. These constitutions contain explicit provisions,which permit the entering into of bi- and multilateral agreementsand regulate the potential incorporation thereof in the domestic legalsystems of the countries concerned.

4.1.2. The impact of international standards on migrants and their socialsecurity status3

The general principle contained in international human rights instru-ments pertaining to non-citizens is that all persons, by virtue of their


essential humanity, should enjoy all human rights unless exceptionaldistinctions, for example between citizens and non-citizens, serve alegitimate state objective and are proportional to the achievement ofthat objective (Weissbrodt, 2003, p. 4).

Various UN and ILO instruments also regulate the rights of non-citizens in the area of social security. The International Covenant onEconomic, Social and Cultural Rights of 1966, ratified by most SouthernAfrican countries, provides in Article 9 for the right to social secu-rity for everyone. In essence this implies that countries may not drawdistinctions between citizens and non-citizens as to social rights, includ-ing the right to social security. Developing countries, including thoseSADC member states that ratified the Covenant, are bound by thisprovision.

The recently adopted UN General Comment No. 19 (2008) on theright to social security enshrined in Article 9 of the InternationalCovenant on Economic, Social and Cultural Rights (1966) contains prin-ciples which are important for the understanding of the scope andcontent of the right to social security, in relation to non-citizens. Someof the salient principles which appear in the General Comment in thisregard can be summarized as follows:

• With reference to the prohibition on grounds of nationality, careshould be taken to ensure that migrants should have access tosocial security in law and practice/implementation. Special attentionshould be paid to specific vulnerable individuals and groups whotraditionally face difficulties in exercising the right to social secu-rity, including refugees, asylum seekers, returnees and non-nationalsgenerally.

• Non-nationals should be able to access non-contributory schemes forincome support, and should have affordable access to health careand family support. Any restrictions, including a qualification period,must be proportionate and reasonable.

• Also, all persons, irrespective of their nationality, residency or immi-gration status, are entitled to primary and emergency medical care. Itwould appear that this is an unqualified entitlement, given the vul-nerable status of those in need of, in particular, emergency medicalcare.

• Special categories of vulnerable migrants, in particular refugees,stateless persons and asylum seekers, should enjoy equal treatmentin access to non-contributory social security schemes, including rea-sonable access to health care and family support, consistent with

Marius Olivier 125

international standards. For these groups, the bar of entitlement issignificantly lowered – they are not only entitled to access non-contributory schemes, but also to enjoy this entitlement on the basisof equal access.

• In the case where non-citizens, including migrant workers, have con-tributed to a social security scheme, they should be able to benefitfrom that contribution or retrieve their contributions if they leavethe country.

• Also, a migrant worker’s entitlement should not be affected by achange in workplace.

• Where persons work temporarily in another country, they should becovered by the social security scheme of their home country.

• Finally, the General Comment notes the importance of establishingbi- and multilateral agreements or other instruments for coordinat-ing or harmonizing contributory social security schemes for migrantworkers, and stresses the importance to take into account the obliga-tions of a state party under the Covenant when entering into a bi- ormultilateral agreement.

The International Convention on the Protection of All Migrant Workersand Members of Their Families of 1990 protects all migrant workers andtheir families, except particular categories of workers, most of whomare protected in terms of other specific international instruments. ThisConvention, among others, provides for non-discrimination, equality oftreatment between nationals and migrant workers as to work conditionsand pay, equal access to social security, and the right to repatriate earn-ings, savings and belongings. The Convention confers specific rights ondocumented workers, and sets out core rights for both documented andundocumented/irregular migrant workers.

Other United Nations human rights instruments, treaties and con-ventions, such as the Universal Declaration of Human Rights (1948),the International Covenant on Civil and Political Rights (1966), theConvention on the Rights of the Child (1989), the Convention on theElimination of all Forms of Racial Discrimination (1965), the Conven-tion against Torture and Other Cruel, Inhuman or Degrading Treatmentor Punishment (1984) and the Convention on the Elimination of AllForms of Discrimination against Women (1981), confer protection onmigrants as well. These instruments are of significant importance to thedebate in Southern Africa on the position of non-nationals generallyand in social security specifically, as these UN agreements have beenratified by most SADC countries.


Targeted international human rights instruments, widely ratified bySADC countries, deal with the position of refugees and asylum seek-ers. For example, the Convention and Protocol relating to the Status ofRefugees (1951 and 1966) provide that refugees should be entitled totreatment at least as favourable as that accorded to citizens of the coun-try concerned with respect to, among other things, labour legislationand social security. Asylum seekers should also be granted the right towork, while employment and social assistance should not be denied torecognized refugees.

Relevant ILO Conventions and Recommendations protect the rightsof all workers, irrespective of citizenship. As Baruah and Cholewinskiremarked (2006, pp. 154–155), ‘[A]ll current ILO social security standardsdefine personal scope of coverage irrespective of nationality and almostall contain similar clauses on equality of treatment between nationalsand foreign workers in the host country, and most of them contain spe-cial non-discrimination clauses, such as, for example, Convention 102of 1952.’ Specific ILO instruments protect migrant workers and theirfamilies, although only a few SADC countries have ratified these instru-ments. Usually certain rights, including the right to social security, areextended only to those lawfully within a territory.

Finally, a range of regional human rights instruments would similarlyextend protection to migrant workers and their families, also in thearea of social security. However, regional adjudicative bodies, such asthe European Court of Human Rights and the Inter-American Courtof Human Rights, have held that instruments that draw a distinctionbetween nationals of particular countries bound together in a regionalframework (such as the European Union) are in principle permissible,on the basis, for example, that member states of a particular regional orsupranational entity (such as the European Union) form a special legalorder, which has established its own citizenship.

The position in terms of the above-mentioned human rights instru-ments is, in general terms, clear. Human rights protection is, in princi-ple, extended to everyone, regardless of nationality. In social securityterms, discrimination against lawfully residing non-national workersand their families is not tolerated. However, coordinating measuresadopted at either the bilateral or multilateral level may be needed toensure the maintenance of acquired rights and the exportability of ben-efits of migrating workers. The relevant principles are contained in awide range of international, regional and supranational instruments.These principles are incrementally applied to workers engaged in aself-employed capacity as well as to stateless persons and refugees.

Marius Olivier 127

Special protection is at times extended in an attempt to deal withirregular migration, as is evident from the provisions of the UnitedNations Migrant Workers Convention of 1990. One of the relevant ILOConventions that also extends protection, both indirectly and indi-rectly, is ILO Convention 143 of 1975, which provides in Article 8 thatthe ‘migrant worker shall not be regarded as in an illegal or irregu-lar situation by the mere fact of the loss of his employment, whichshall not in itself imply the withdrawal of his authorisation or resi-dence or, as the case may be, work permit’. It further stipulates (inArticle 9) that irregular migrant workers shall have the same rights asregular migrant workers concerning social security benefits arising outof past employment.

Many of the important UN instruments which impact on the socialsecurity rights of migrants, including the specialized UN refugee con-ventions, have been ratified relatively widely by SADC countries. How-ever, only two SADC countries (Lesotho and Seychelles) have ratifiedthe UN Migrant Workers Convention of 1990. As far as the ILO isconcerned, the position is that SADC countries have generally failedto ratify social security conventions, in particular the main socialsecurity Convention, Convention 102 of 1952 on Minimum Stan-dards in Social Security, and most of the other post-Second WorldWar social security conventions. However, certain pre-Second WorldWar social security conventions have been ratified by some SADCcountries. These conventions require equal treatment of citizens andnon-citizens who are citizens of other countries that ratified theseconventions. A few SADC countries have also ratified ILO migration-related conventions, which affect the area of social security as well.The poor ratification of ILO and to some extent UN conventionswithin SADC has been deplored, in view of the seriousness of migrationissues.4

4.2. Regional normative framework

4.2.1. Regional standards

In the foundational instruments of SADC there are several indicationstowards the creation of a special regime of SADC-wide social securitycoverage for citizens and residents of the different member states. Recall-ing the objectives of the SADC Treaty (1992) contained in Article 5 ofthe Treaty, including regional integration, poverty alleviation and thesupport of the socially disadvantaged, Article 10 of the Charter of Fun-damental Social Rights in SADC (‘Social Charter’) of 2003 imposes on


SADC member states the obligation to create an enabling environmentso that every worker in the region shall have a right to adequate socialprotection. Persons who have been unable to either enter or re-enterthe labour market and have no means of subsistence shall be entitled toreceive sufficient resources and social assistance. No distinction is drawnbetween citizens and non-citizens; the Treaty emphasizes regional inte-gration and the Charter focuses on harmonization of social securityschemes.

Even more explicit are the provisions of Article 17 of the Codeon Social Security in SADC of 2007, which do not allow disparatetreatment of foreigners, and encourage member states to ensure thatall lawfully employed immigrants are protected through the promo-tion of certain core principles. According to two of these princi-ples migrant workers should, firstly, be able to participate in thesocial security schemes of the host country and, secondly, enjoyequal treatment alongside citizens within the social security systemof the host country. Member states are further encouraged to intro-duce, by way of national legislation and bi- or multilateral arrange-ments, cross-border coordination principles such as maintenance ofacquired rights, aggregation of insurance periods and exportability ofbenefits.

The Code further suggests that illegal residents and undocumentedmigrants should be provided with basic minimum protection andshould enjoy coverage according to the laws of the host country. Asregards refugees, it stipulates that the social protection extended tothem should be in accordance with the provisions of international andregional instruments.

It could, therefore, be said that there is a clear tendency in both theCharter and the Code to create a regime within SADC which ensuresminimum levels of social protection on the basis of equality, regardlessof citizenship.

In addition, there are important provisions in the adopted (in 2006)but not yet implemented SADC Draft Protocol on the Facilitation ofMovement of Persons (2005), in view of the absence of a right tofreedom of movement in other SADC foundational documents. Thefreedom of movement principle in SADC is couched in much weakerterms than in the EU counterpart. Article 5(2) (d) of the SADC Treaty(1992) does not regulate the matter conclusively, but requires SADC to‘develop policies aimed at the progressive elimination of obstacles tothe free movement of capital and labour, goods and services, and ofthe people of the Region generally, among Member States’. However,

Marius Olivier 129

it is suggested that the Draft Protocol on the Facilitation of Movementof Persons does not guarantee freedom of movement in a way that issignificant for the purposes of enhancing the social security position ofintra-SADC migrants. The reason is that while the Draft Protocol recog-nizes visa free travel for up to 90 days, it nevertheless subjects the rightto residence and establishment (in the occupational sense of the word)to restrictions contained in national laws.

Finally, the SADC Treaty and, with the exception of the Code onSocial Security in the SADC, other SADC multilateral instruments donot display strong incentives for the development of social securitycoordination measures. The principle of non-discrimination containedin Article 6(2) of the Treaty does not include the prohibition ofdiscrimination based on nationality/citizenship.

4.2.2. Social security cross-border agreements

In the absence of a multilateral arrangement regulating coordinationand portability issues in SADC, SADC countries have had to rely onbilateral arrangements. However, only a few such agreements are in exis-tence, mostly involving South Africa as one of the parties. The scopeof these agreements is limited and, with some exceptions, does notcover public social security schemes, only employer-based occupationalarrangements. The obligations contained in the labour agreementsbetween South Africa and several SADC countries are primarily imposedupon employers, not the South African government. There are severalother reasons why these agreements, to the extent that they may stillbe operational, must be seen as limited in scope and effect, and asinsufficient from a coordination and portability perspective:

(a) The agreements are not reciprocal in nature, as they stand to regu-late the position of citizens of only one of the respective countries –that is, the citizens of the country that entered into the agreementwith South Africa, and not South African citizens.

(b) Repatriation regulation is accorded specific emphasis, alongsidelabour migration.

(c) In view of the above, and given the control and restriction orienta-tion and purpose of the agreements, apart from providing for somemeasure of portability these agreements do not provide for otherarrangements typical of coordination regimes, such as maintenanceof acquired rights, aggregation of insurance periods and equality oftreatment with nationals of the receiving country in social securitymatters.


(d) Institutional capacity appears to be weak. For example, the agree-ment between South Africa and Mozambique, which allows forpayments in respect of employment injuries and diseases to bemade in Mozambique has been widely criticized in view of the factthat payments seldom reach beneficiaries (Fultz and Pieris, 1998,pp. 19–20).

In fact, these agreements have the effect of excluding most citizensof the sending country who are migrant workers in South Africa fromentitlement to benefit from unemployment insurance in South Africa:migrant workers who have to return to their home country as a result ofthe agreements are not regarded as contributors to, and could not there-fore benefit from, the Unemployment Insurance Fund. While this maybe seen as an arrangement which benefits employers of such migrantworkers, it may leave returning migrants in a precarious position oncethey are home.

However, it has to be noted that recent bilateral agreements andarrangements between some of the SADC countries provide for the reg-ulation of movement across country borders. To some extent this is areflection of the impact of the considerable migration flows between, forexample, South Africa and some of its neighbouring countries, notablyLesotho, Mozambique and Zimbabwe, as well as between Namibia andAngola. Simultaneously, it underscores the need for arrangements whichqualify the strict provisions of immigration legislation, particularly inSouth Africa.

Finally, it is important to recognize that the scope for entering intobi- and multilateral social security agreements does exist in SADC. Forexample, in Tanzania the recently promulgated Social Security (Reg-ulatory Authority) Act (2008) provides for the making of regulationsin relation to portability of benefit rights and international recipro-cal agreements for the transfer of benefits. In South Africa, similarprovisions relating to entering into reciprocal bilateral agreements arecontained in the workers’ compensation and social assistance legisla-tion. As mentioned above, the constitutions of countries such as Malawi,Namibia and South Africa recognize both incorporation and transforma-tion as legitimate methods of deriving domestic effects from the state’sinternational obligations, including obligations flowing from multi-and bilateral arrangements – as indicated, these constitutions containexplicit provisions, which permit the entering into of bi- and multilat-eral agreements and regulate the potential incorporation thereof in thedomestic legal systems of the countries concerned.

Marius Olivier 131

4.3. Constitutional rights and jurisprudential perspectives

4.3.1. Constitutional frameworks

There are two reasons why SADC constitutional frameworks extend littlesocial security protection to non-citizens. The first reason is that someSADC country constitutions refer to social security in general terms anddo not create directly enforceable social security or broader social pro-tection rights. The second reason is that anti-discrimination provisionsin these constitutions are rarely made applicable to non-citizens.

One notable exception is the South African Constitution. In thechapter dealing with the Bill of Rights, the Constitution (1996, Section27) grants to everyone ‘[t]he right to have access to social secu-rity, including, if they are unable to support themselves and theirdependants, appropriate social assistance’. It also obliges the stateto implement appropriate measures: ‘[t]he state must take reason-able legislative and other measures, within its available resources, toachieve the progressive realization of each of these rights’. Furthermore,Section 9 of the Constitution prohibits unfair discrimination, whetherdirectly or indirectly. Discrimination again non-citizens is includedin this framework, although nationality is an unlisted discriminationground.

4.3.2. Progressive judicial pronouncements

As far as social security for non-citizens is concerned, the ConstitutionalCourt in South Africa remarked, when considering the purpose of pro-viding access to social security to those in need (Khosa v. The Minister ofSocial Development; Mahlaule v. The Minister of Social Development, 573A,case headnote summary) that:

A society had to attempt to ensure that the basic necessities of lifewere accessible to all if it was to be a society in which human dig-nity, freedom and equality were foundational. The right of access tosocial security, including social assistance, for those unable to supportthemselves and their dependants was entrenched because society inthe RSA valued human beings and wanted to ensure that people wereafforded their basic needs.

It has to be noted that there is a specific constitutional focus in SouthAfrica on addressing the plight of the most vulnerable and desper-ate in society. In fact, in the South African constitutional context ithas been accepted that non-citizens constitute a vulnerable group in


society. For example, the Constitutional Court held that excludingpermanent residents from access to employment in the public educationsector, and excluding permanent residents from the social assistancesystem would amount to unfair discrimination (Larbi-Odam v. Memberof the Executive Council for Education (North-West Province) v. The Minis-ter of Education). As a result of Constitutional Court and High Courtintervention, social assistance support has effectively been extendedto certain categories of non-citizens, in particular permanent residentsand their children (Khosa judgement) and unaccompanied foreign chil-dren (Centre for Child Law (Lawyers for Human Rights) v. Minister of HomeAffairs).

Several of the court cases dealing with the protection of non-citizensrelate to the constitutional protection, including social security pro-tection, available to irregular/undocumented non-citizens and refugees.These cases include instances where the judgement ordered the releaseof unlawfully detained refugees (Centre for Child Law judgement), theprocessing of applications for refugee status within reasonable time andin accordance with national and international refugee law (South AfricanHuman Rights Commission & Forty Others v. Minister of Home Affairs andDyambu (Pty) Ltd), the removal of unlawful conditions from the asylum-seeking process (Kiliko and Others v. Minister of Home Affairs), and theincorporation of refugees who are disabled within the framework ofthe social assistance grants (Somali Refugee Forum v. The Minister of HomeAffairs).

Access to employment for non-citizens is an important tool to ensuresocial insurance-based entitlements for migrants. The court judgementsin South Africa, which confirmed access to private and public employ-ment for at least certain categories of foreigners and the right of refugeesand asylum seekers to work (see the Larbi-Odam and Watchenuka judge-ments), effectively imply access to unemployment insurance (at least forpermanent residents not employed in government), compensation foroccupational injuries and diseases, occupational-based retirement andhealth insurance. This access is potentially also available to irregularmigrants whose work permits may have expired. According to recentcase law the expiry of their work permits does not affect the validityof the employment contract between such a migrant and his or heremployer (Discovery Health Limited v CCMA).

From these judgements it is clear that in South African jurisprudencenon-citizens are generally regarded as a vulnerable group which, fromthe perspective of the constitutional framework, is entitled to legal pro-tection. Save for those rights in the Constitution which are reserved

Marius Olivier 133

for citizens only, the whole gamut of fundamental rights, including theright to access to social security and appropriate social assistance, as wellas the right to fair labour practices, is therefore in principle available tonon-citizens migrating to South Africa.

5. Conclusion: the rationale for regulation

The legal position of non-citizens in social security is generally a weakone. Compliance with standards set at the international and regionallevel, even if in some respects incomplete, could go a long way to ensurethat migrants and their families are extended essential human rightsprotection in the area of social security. Integrating these standards inSADC domestic legal systems, in particular via the adoption of a funda-mental rights framework in country constitutions, would further helpto cement the protective framework. With such a constellation in place,as the experience in several jurisdictions around the world, including inSADC (for example, the South African jurisdiction), has shown, courtsand other adjudicating institutions through their jurisprudential activ-ity are quick to give legal effect to the embedded rights. In addition,there is a need to adopt targeted statutory provisions to which immi-gration law provisions are appropriately aligned. Coherence is required,not only between the legal and policy frameworks, but also between lawand practice: too often the protective rights framework pertaining tomigrating non-citizens and social security, even in SADC countries suchas South Africa, has been undermined by inconsistent policy regimesand oppressive treatment of non-citizens. Finally, the value of mean-ingful bilateral social security arrangements between affected countries,supported by a framework-setting multilateral edifice at the regional orsupranational level, has proved to be indispensable, as it clarifies thelegal position of the migrants and expresses the joint commitment ofthe countries concerned.

It is evident that the picture in Southern Africa reveals major short-comings. Ratification of and compliance with relevant internationalstandards are unsatisfactory; underlying and supportive measures, suchas the principle of freedom of movement of workers and other cat-egories of persons, are weakly developed; and regional standards inrelation to the coordination of social security, among other things, arestill being concretized. With the exception of the South African Con-stitution, and possibly the constitutions of a few other SADC countries,such as Tanzania, fundamental rights frameworks are not in place. Fur-thermore, across the Southern African region, immigration law and


policy regimes emphasize selection and control, undergirded byhostile treatment and immigration-unfriendly perceptions. There is nomultilateral social security framework agreement in place, and bilateralarrangements are few and weak in content, and hardly cover publicsocial security schemes.

The reality is that, apart from the provisions of the Code on SocialSecurity in the SADC (2007), none of the other SADC instrumentsprovide comprehensively for the vision expressed in this chapter inrelation to migration and social security. This vision shows that a rights-based framework and appropriate regulatory responses are crucial for themeaningful extension of social security coverage to migrants. It would,therefore, be necessary to adopt special policy interventions and mea-sures, based on the framework provided for in Article 17 of the Code,to ensure the protection and enhancement of the position of migrantsin relation to social security. This should also imply that the said policyand legal interventions and measures should develop minimum levelsof social protection for different categories of beneficiaries, on the basisof equality within SADC and (at least as far as social insurance coverageis concerned) with reference to cross-border coordination arrangements,to be adopted by way of national legislation and bi- and/or multilateralagreements.

Notes

1. Internally displaced persons (IDPs) in SADC – the result of, among other prob-lems, political and military instability in some of the countries – make up 2.9million of the approximately 13 million IDPs in Africa – more than half of theglobal total of IDPs, dwarfing the number of refugees (Brookings Institution,2005, pp. 4–6).

2. For example, in 2001 in Lesotho, remittances were estimated to contribute asmuch as 26.5 per cent of GDP (Crush et al., 2005, p. 18).

3. See also Olivier (2008, pp. 31–35), on which this part is based.4. At the ILO Tripartite Forum on Labour Migration in Southern Africa held

in Pretoria, 26–29 November 2002. The Forum identified migration issues asserious. See Kanyenze (2004, p. 19).

Bibliography

Books, articles, reports and internet resourcesBaruah, N. and Cholewinski, R. (2006) Handbook on Establishing Effective Labour

Migration Policies in Countries of Origin and Destination (OSCE (Organisationfor Security and Co-operation in Europe), IOM (International Organisation forMigration) and ILO (International Labour Office)).

Marius Olivier 135

Black, R. (2004) ‘Migration and Pro-Poor Policy in Africa’, Working PaperC6, Development Research Centre on Migration, Globalisation and Poverty,University of Sussex.

Brookings Institution – University of Bern Project on Internal Displacement(2005) Regional Seminar on Internal Displacement in the Southern AfricanDevelopment Community (SADC) Region, Gaborone, Botswana, 24–26 August2005.

Cholewinski, R. (2008) ‘The human and labor rights of migrants: visions ofequality’, Georgetown Immigration Law Journal, 22(2), 177–220.

Committee against Torture (2006) Consideration of Reports Submitted by StateParties under Article 19 of the Convention: Conclusions and Recommendations of theCommittee against Torture – South Africa, 37th Session, 6–24 November 2006,CAT/C/ZAF/CO/1, available at: http://www.unhchr.ch/tbs/doc.nsf/898586b1dc7b4043c1256a450044f331/9f410d04620f0197c125729800544666/$FILE/G0740324.pdf.

Crush, J., Williams, V. and Peberdy, S. (2005) ‘Migration in Southern Africa’, paperprepared for the Policy Analysis and Research Programme of the Global Commis-sion on International Migration, Global Commission on International Migration2005.

Dodson, B. (1998) ‘Women on the move: Gender and cross-border migrationto South Africa’, Southern African Migration Project, Migration Policy SeriesNo. 9, available at: http://queensu.ca/samp/sampresources/samppublications/policyseries/policy9.htm.

Fenwick, C. and Kalula, E. (2004) ‘Law and Labour Market Regulation inEast Asia and Southern Africa: Comparative Perspectives’, Development andLabour Monograph Series, Occasional Paper 2/2004, also published as (2005)International Journal of Comparative Labour Law and Industrial Relations, 21,193–226.

FIDH (International Federation of Human Rights) (2008) Surplus People? Undocu-mented and other Vulnerable Migrants in South Africa.

Fultz, E. and Pieris, B. (1998) ‘Employment Injury Schemes in Southern Africa: anOverview of Proposals for Future Directions’, ILO/SAMAT Policy Paper No. 7,ILO Harare, Zimbabwe.

Olivier, M.P. (2009) ‘Regional Overview of Social Protection for Non-Citizensin the South African Development Community (SADC)’, SP Discussion PaperNo. 0908, The World Bank, available at: http://siteresources.worldbank.org/SOCIALPROTECTION/Resources/SP-Discussion-papers/Labor-Market-DP/0908.pdf.

Global Migrant Origin Database – Version 4 (2007) available at: http://www.migrationdrc.org/research/typesofmigration/Global_Migrant_Origin_Database_Version_4.xls.

Holzmann, R., Koettl, J. and Chernetsky, T. (2005) ‘Portability Regimes of Pen-sion and Health Care Benefits for International Migrants: An Analysis of Issuesand Good Practices’, Social Protection Discussion Paper No. 0519, World Bank(paper prepared for the Global Commission on International Migration).

ILO (2006) ‘Changing Patterns in the World of Work’, report of the Director-General, ILC, 95th Session, ILO, Geneva.

IOM (2005) Current Migration Themes in Southern Africa: An IOM Perspective (IOMRegional Office for Southern Africa).


Jansen van Rensburg, L. and Olivier, M. P. (2003) ‘International and supra-national law’, in Olivier, M. P. et al. (ed.) Social Security: A Legal Analysis(LexisNexis Butterworths).

Kanyenze, G. (2004) ‘African Migrant Labour Situation in Southern Africa’, paperpresented at the ICFTU-AFRO Conference on ‘Migrant Labour’, Nairobi, 15–17March 2004.

Kaseke, E. (2007) ‘The Social Security Context of Migration’, paper presented at anInternational Conference on Migration and Social Security in SADC: Prospectsand Challenges, University of Johannesburg, 3–4 October 2007.

Klaaren, J. and Rutinwa, B. (2004) ‘Towards the harmonisation of immigrationand refugee law in SADC’, in Crush, J. (ed.) MIDSA Report No 1 (IDASA & QueensUniversity).

Landau, L. (2004) ‘Democracy and discrimination: Black African migrants inSouth Africa’, Global Migration Perspectives No. 5, GCIM (Global Commissionon International Migration), Geneva.

Maharaj, B. (2004) ‘Immigration to post-apartheid South Africa’, Global Migra-tion Perspectives No. 1, GCIM (Global Commission on International Migra-tion), Geneva.

Olivier, M. (2008) ‘The end of labour law in the global workplace context?A South and Southern African response’, in Blanpain, R. and Tiraboschi, M.(eds) Global Labor Market: From Globalization to Flexicurity (Alphen aan den Rijn:Kluwer), pp. 13–52.

Olivier, M. (2009) ‘Regional Overview of Social Protection for Non-Citizens inthe Southern African Development Community (SADC)’, report prepared forthe World Bank.

Palmary, I., (2002) ‘Refugees, Safety and Xenophobia in South African Cities:The Role of Local Government’, a research report for the Centre for theStudy of Violence and Reconciliation, available at http://www.csvr.org.za/docs/foreigners/refugeessafteyand.pdf.

Peberdy, S. (2005) ‘Hurdles to trade? South Africa’s immigration policy and infor-mal sector cross-border traders in the SADC’, , presented to a workshop ofthe Southern African Poverty Network (HSRC), Lawyers for Human Rights andthe Southern African Migration Project “Regional Integration, Migration andPoverty,” Pretoria, April 2002.

Pendleton, W., Crush, J., Campbell, E., Green, T., Simelane, H., Tevera, D. and deVletter, F. (2006) ‘Migration, Remittances and Development in SouthernAfrica’, Southern African Migration Project, Migration Policy Series No. 44,IDASA.

Reitzes, M. (2003) ‘Regionalizing International Migration: Lessons for SADC’,Migration Policy Brief No. 11, SAMP (Southern African Migration Project).

SAMP and IOM (2005) HIV/AIDS, Population Mobility and Migration in SouthernAfrica: Defining a Research and Policy Agenda (Southern African Migration Project(SAMP) & International Organisation for Migration (IOM)).

Standing, G. et al. (1996) Restructuring the Labour Market: The South AfricanChallenge (An ILO Country Review), ILO.

UN (Office of the United Nations High Commissioner for Human Rights) (2006)The Rights of Non-Citizens (New York: UN).

Weissbrodt, D. (2003) Prevention of Discrimination: The rights of non-citizens(UN Doc. E/CN.4/Sub.2/2003/23) available at: http://burmalibrary.org/docs4/Rights_of-non-citizens.pdf.

Marius Olivier 137

Williams, V. (2002) ‘An Overview of Migration in the SADC Region’, paper pre-sented at SAMP/LHR/HSRC workshop on Regional Integration, Poverty andSouth Africa’s Proposed Migration Policy, Pretoria, 23 April 2002.

International and regional instrumentsCaricom Agreement on Social Security of 1996 (Caribbean).Charter of Fundamental Social Rights in SADC of 2003 (Social Charter) (SADC).Charter of Civil Society for the Caribbean Community (Caribbean).Code on Social Security in the SADC of 2007 (SADC).Convention against Torture and Other Cruel, Inhuman or Degrading Treatment

or Punishment of 1984 (CAT) (UN) – see also Committee against Torture Con-sideration of reports submitted by State Parties under Article 19 of the Convention:Conclusions and Recommendations of the Committee against Torture – South Africa(37th Session, 6–24 November 2006, CAT/C/ZAF/CO/1).

Convention on the Elimination of All Forms of Discrimination against Womenof 1981 (CEDAW) (UN).

Convention on the Elimination of all Forms of Racial Discrimination (CEARD) of1965 (UN).

Convention on the Rights of the Child of 1989 (UN).Convention relating to the Status of Refugees of 1951 (UN).Council Directive 79/7/EEC on the progressive implementation of the principle

of equal treatment for men and women in matters of social security (EU).Council Directive 86/378/EEC of 24 July 1986 on the implementation of the prin-

ciple of equal treatment for men and women in occupational social securityschemes (EU).

Council Recommendation 92/441/EEC of 24 June 1992 on common crite-ria concerning sufficient resources and social assistance in social protectionsystems (EU).

Council Recommendation 92/442/EEC of 27 July 1992 on the convergence ofsocial protection objectives and policies (EU).

Draft Protocol on the Facilitation of Movement of Persons of 2005 (SADC).Economic and Social Council, Committee on Economic, Social and Cultural

Rights General Comment No. 19: Right to social security (article 9) E/C.12/GC/19of 4 February 2008 (UN).

Equality of Treatment (Accident Compensation) Convention 19 of 1925 (ILO).Equality of Treatment (Social Security) 118 of 1962 (ILO).European Convention on Social Security of 1972 (European Treaty Series 78)

(Council of Europe).International Convention on the Protection of All Migrant Workers and Members

of Their Families (the UN Migrant Workers Convention) of 1990 (UN).International Covenant on Civil and Political Rights of 1966 (UN).International Covenant on Economic, Social and Cultural Rights of 1966 (UN).Maintenance of Social Security Rights Convention 157 of 1982 (ILO).Malawi/Zambia Bilateral Social Security Agreement (2003).Migrant Workers (Supplementary Provisions) Convention 143 of 1975 (ILO).Migration for Employment (Revised Convention) 97 of 1949 (ILO).Protocol relating to the Status of Refugees of 1966 (UN).Regulation (EEC) 1612/68 of the Council of 15 October 1968 on freedom of

movement for workers within the Community (EU).


Regulation (EEC) 1408/71 of the Council of 14 June 1971 on the application ofsocial security schemes to employed persons and their families moving withinthe Community (EU).

Regulation 883/2004 of the European Parliament and of the Council of 29 April2004 on the coordination of social security systems (EU).

Social Security (Minimum Standards) Convention 102 of 1952 (ILO).South Africa/Botswana bilateral agreement of 1973.South Africa/Lesotho bilateral agreement of 1973.South Africa/Malawi bilateral agreement of 1967.South Africa/Swaziland bilateral agreement of 1986.South Africa/Portugal bilateral agreement (re Mozambique) of 1964.Treaty establishing SADC of 17 August 1992 (SADC).Unemployment Convention 2 of 1919 (ILO).Universal Declaration of Human Rights of 1948 (UN).

Laws/statutory instruments and constitutionsCompensation for Occupational Injuries and Diseases Act 130 of 1993 (South

Africa).Constitution of the Republic of Malawi.Constitution of the Republic of Namibia.Constitution of the Republic of South Africa 108 of 1996.Constitution of the United Republic of Tanzania of 1977.Constitution of Zambia Act 18 of 1996.Immigration Act 13 of 2002 (South Africa).Refugees Act 130 of 1998 (South Africa).Social Assistance Act 13 of 2004 (South Africa).Social Security (Regulatory Authority) Act, 2008 (Tanzania).South African Citizenship Act 88 of 1995 (South Africa).Unemployment Insurance Act 63 of 2001 (South Africa).

Case lawsAdvisory Opinion on the Proposed Amendments to the Naturalization Provisions

of the Political Constitution of Costa Rica (OC-4/84) (Advisory Opinion of theInter-American Court on Human Rights).

Belgian Linguistic Case, 6 Eur. Ct. H.R. (ser. A) (1968) (European Court of HumanRights).

C. v. Belgium, Eur. Ct. H.R., Reports 1996-III (1996) (European Court of HumanRights).

Centre for Child Law & Another (Lawyers for Human Rights) v. Minister of Home Affairs& Others 2005 (6) SA 50 (T) (South Africa).

Discovery Health Limited v. CCMA & others [2008] 7 BLLR 633 (LC) (South Africa).Khosa and Others v. The Minister of Social Development and Others; Mahlaule and

Others v The Minister of Social Development and Others 2004 (6) BCLR 569 (CC)(South Africa).

Kiliko and Others v. Minister of Home Affairs and Others 2006 (4) SA 114 (CC); [2007]1 All SA 97 (C) (South Africa).

Larbi-Odam v. Member of the Executive Council for Education (North-West Province) v.The Minister of Education 1997 BCLR 1655 (CC) (South Africa).

Marius Olivier 139

Lawyers for Human Rights & Another v. Minister of Home Affairs 2004 (4) SA 125 (CC)(South Africa).

Minister of Health & others v. Treatment Action Campaign & others 2002 BCLR1033 (CC) (South Africa).

Minister of Home Affairs and Others v. Watchenuka and Another 2004 (2) BCLR120 (SCA) (South Africa).

Somali Refugee Forum and Another v. The Minister of Home Affairs and Others (Caseno. 32849/2005, High Court) (South Africa).

South African Human Rights Commission & Forty Others v. Minister of Home Affairsand Dyambu (Pty) Ltd (Case no. 28367/99, Witwatersrand High Court) (SouthAfrica).

The Union of Refugee Women & others v. The Director: The Private Security IndustryRegulatory Authority 2007 (4) BCLR 339 (CC) (South Africa).

Watchenuka and Another v. Minister of Home Affairs and Others 2003 (1) BCLR 62 (C)(South Africa).

Part 3

Case Studies

5‘An Increasingly UncomfortableEnvironment’: Access to HealthCare for Documented andUndocumented Migrants in the UKRosalind Bragg and Rayah Feldman

1. Introduction

In spite of the much vaunted claim to universalism of the BritishNational Health Service (NHS), entitlements to treatment of people fromoutside Britain have been open to challenge since its inception in 1949.The 1949 Act, which established the NHS as a service free at the pointof delivery, included powers for a Minister of Health to draw up regu-lations imposing charges for people not ‘ordinarily resident’ in the UK.These powers were also included in the NHS Act of 1977, although werenot implemented until 1982.

However, long before such charges were actually imposed, there wereattempts to restrict access to services. For example, in 1963, the Min-istry of Health had issued a circular to hospitals instructing them torestrict NHS treatment to those ‘ordinarily resident’. And in 1976, theLiberal peer Lord Avebury reported that 185 Asian women attendingLeicester General Hospital antenatal clinic were asked to produce theirpassports. One woman, who refused to do so as she had already hadone confinement at the hospital, was refused antenatal care (Cohen andHayes, 1998).

Some writers have viewed attempts to restrict NHS care to foreignersas directly bound up with increasing immigration controls (ManchesterLaw Centre, 1982; Cohen and Hayes, 1998). This is certainly suggestedin the way that NHS regulation in this area has proceeded in line withimmigration legislation: the 1963 Ministry of Health circular was issuedone year after the first Commonwealth Immigration Act of 1962, and

143

144 Case Studies

the 1982 NHS (Overseas Visitors) charging regulations followed the 1981Nationality Act with its dramatic changes to conditions of citizenshipfor Commonwealth citizens.

Whatever the recent connection between immigration control andthe provision of public welfare, the relationship between welfare andmobility has very long historic roots in Britain, certainly going back tomedieval vagrancy legislation and the Elizabethan and Victorian PoorLaws. Such legislation restricted poor relief to ‘local’ people, and localauthorities of various kinds were empowered to remove people whomight be a burden on the parish back to their parish of origin. Indeed‘parish officers kept a “lookout” for those newly arrived in a parish “whomight apply to it for poor relief” ’ (Gordon and Newnham, 1985, cit-ing Sharma, ‘Second Class Claimants’: Race and Social Security, Leicester,CPAG 1981).

Immigration rules have operated in similar ways, primarily to con-trol international movements of poor people. Since the 1905 Aliens Act,migrants to Britain could be denied entry if it was believed they wouldnot be able to support themselves or their dependants, and they couldbe deported for the same reason. Supposedly, exceptions were made forbona fide refugees from persecution, but in practice this was not alwaysthe case (Cohen, 2003). Immigration and nationality legislation from1962 onwards redefined the rights to settlement of British citizens soas to exclude Commonwealth citizens, and routinely invoked the rulethat migrants should have ‘no recourse to public funds’ (Gordon andNewnham, 1985, pp. 7ff.; Southall Black Sisters, 2007).

As international mobility has grown in the wake of colonialism,post-colonialism and globalization, concerns about ‘undesirables’ haveacquired more explicitly xenophobic and racist overtones. These includethe widely held claims that foreigners bring with them disease andcrime, and steal not only ‘our’ money but also ‘our’ jobs, and in anera of increasingly state provided public services, ‘our’ houses and ‘our’welfare. What changes is not the content of the argument but thecategory of foreigner invoked – Irish, alien, Jew, immigrant, coloured,black, bogus asylum seeker, failed asylum seeker, health tourist – and thespecificity of the policy solutions proposed at any particular time. Theterminology may suggest something new, but the ideas are definitely oldand unremittingly xenophobic.

This background is important because in recent years ‘abuse’ of freeNHS services has been presented by significant sections of the media,politicians and government in terms of a novel crisis attributable to

Rosalind Bragg and Rayah Feldman 145

increases in rates of migration in the last two decades (Departmentof Health, 2003; Kyambi, 2005; Cooley and Sriskandarajah, 2007). Webelieve it is important to note the historical precedents for such percep-tions of crisis, at least as a reminder to retain a degree of critical distancefrom claims made on this issue.

This chapter explores recent changes and attempts to further restrictentitlements to NHS services. We will explore both the issues associ-ated with the legal term ‘ordinarily resident’ and the way in whichattempts to exclude undocumented migrants, especially refused asylumseekers, has become conflated with the new and confusing bogeymanof ‘health tourist’. After setting out the trajectory leading to currentchanges in entitlement to NHS health care, and disputes about thecurrent regulations and further proposed changes, we will discuss theimplications of existing and proposed restrictions on both documentedand undocumented migrants.

2. Definitions

There is considerable debate about the most appropriate terminologyto use when referring to individuals without regular immigration status(IPPR, 2006). The international NGO, Platform for International Coop-eration on Undocumented Migrants (PICUM), uses the term ‘undocu-mented migrant’ to refer to migrants without a permit authorizing themto regularly stay in the country of destination (PICUM, 2007). The UKthink tank, Institute for Public Policy Research, uses the term ‘irregu-lar migrant’, which it defines as people who are liable to be deported forissues related to immigration (IPPR, 2006). This refers to a broader groupthan undocumented migrants as it includes those who have a currentpermit but are in breach of their immigration conditions, for examplelegal migrants working without permission. The UK government, theEuropean Union and some media outlets use the term ‘illegal immi-grant’, and this term is often used synonymously with ‘asylum seeker’in public debate (IPPR, 2006), resulting in much confusion.

This chapter uses the PICUM term ‘undocumented migrant’. Whenusing this terminology, it must be recognized that the distinctionbetween the two categories ‘documented’ and ‘undocumented’ isblurred, as people may shift from ‘legal’ (documented) to ‘illegal’(undocumented) status for numerous reasons, or indeed shift the otherway, for example if they manage to get asylum refusals overturned inthe courts. For this reason we are concerned with both groups. More

146 Case Studies

importantly, we seek to show that whether ‘undocumented’ or not, allmigrants are adversely affected by entitlement restrictions ostensiblydirected at only the undocumented.

Data on undocumented migrants is limited. Estimates of the num-ber of undocumented migrants in the UK range from 310,000 to570,000, with the Home Office adopting a median estimate of 430,000(Woodbridge, 2005). The Home Office calculates that undocumentedmigrants make up less than 0.7 per cent of the UK population andcompares this favourably with the US where the figure is 2.5 per cent,particularly in view of amnesties for illegal migrants in the US whichhave served to reduce the numbers of such migrants there. A morerecent estimate suggested that the irregular migrant population inthe UK in 2007 was 725,000 (range 524,000 and 947,000), a lowerproportion of all immigrants than in the EU as a whole (Vollmer, 2009).

Undocumented migrants are a diverse group whose social and eco-nomic circumstances can vary widely. The group includes individualswho have overstayed their visas, refused asylum seekers and individualswho have been trafficked or have otherwise entered the country withouta visa. It also includes women (and men) who have entered the countryon a two-year spouse visa but whose partner has refused to regularizetheir status.

The number of undocumented migrants in the UK is likely to increase,since the recent introduction of a ‘points-based’ immigration systemin the UK means that many low-skilled workers from outside the EUwill not be able to renew their work visas (Wintour, 2007). In addi-tion, the number of asylum seekers whose claims are refused continuesto exceed the number returned to their home country (National AuditOffice, 2005).

3. Context

During the 1990s and over the millennium there was widespread mediahostility to people seeking asylum in the UK and frequently levelledaccusations that asylum seekers were fraudulently trying to accessBritain’s welfare system. There were repeated changes in legislation tomake asylum seekers ‘less eligible’, in effect to make the conditionsof asylum seekers worse than they might face in their country of ori-gin. People seeking asylum lost the right to work and to claim generalstate benefits. Instead, the government established a dedicated and morerestrictive benefits scheme for asylum seekers (Cohen, 2003; Somerville,2007).


3.1. Health tourism

In this section we explore the background and context of recent changesto NHS entitlements. In particular we show how changes to the NationalHealth Services (Charges to Overseas Visitors) Regulations 1989 were ini-tially justified in terms of preventing ‘abuse of the NHS’, and only laterhave become incorporated into a broader policy of ‘shutting down theprivileges of the UK to those here illegally’ (Department of Health, 2003;Home Office, 2007).

In the contemporary debates about NHS charges, the terms used havebecome significant tools in shaping government agendas as well as influ-encing popular attitudes to migrants and migration. In particular, theterm ‘health tourist’ and the concept of ‘health tourism’, concepts hith-erto barely known in 2003, have dominated discussions about NHScharging. With their connotations of both fraud and greed, these termshave struck a chord with fears of spiralling health costs fuelled by ‘failed’and thus fraudulent asylum seekers.

We conducted a cursory search using Google and selected websites(BBC, Daily Mail, Guardian and Hansard) to try to find when the termhealth tourism began to be associated with fraudulent use of the NHS,and how it became identified with ‘failed’ asylum seekers. In Hansard in2002, we found only two references to health tourism, with two quitedifferent meanings, neither of which related to migrants, the NHS or anykind of abuse of services. The first one extolled the market opportunitiesoffered by health tourism, citing the English Tourism Council’s claimthat health tourism is worth £6 billion a year (Hansard, 2002a). Thesecond used it to scoff at MPs who try to implement policy ideas gainedfrom brief trips abroad (Hansard, 2002b). However, ‘health’ or ‘medical’tourism has for some time been widely used to refer to overseas travelto purchase health care, a sector which has been promoted as a keysource of export earnings by a range of countries including India, SouthAfrica and Cuba (Oswald and Henthorne, 1999; Chinai and Goswami,2007).

A wholly new usage seems to have been coined by Shadow HealthSecretary Liam Fox in January 2003 when he ‘warned of the impact thatthe asylum seeker crisis [sic] is having on the National Health Service’in a story posted on the Conservative Party website. He claimed that‘the government’s long-term failure to control the number of asylumseekers is impacting hard on the ability of the NHS to treat UK citi-zens’. Dr Fox was cited as saying: ‘It is reasonable to treat the healthdemands of genuine asylum seekers. However, it is not acceptable togive them preferential treatment ahead of UK taxpayers. And it is

148 Case Studies

certainly not the duty of the NHS to act a health tourism destinationfor economic migrants’ (Conservative Party, 2003). In June of that yearDr Fox enlarged on the theme, calling for a health entitlement cardon the grounds that ‘the NHS is becoming the health equivalent ofDisneyland for tourists . . . Hundreds of thousands of people who havefailed the asylum process are trying to use our health care system. Thismeans British citizens already waiting for treatment have to wait longer’(BBC, 2003).

Health tourism quickly became a euphemism for abuse of the NHSby ‘foreign visitors’. The government freely admitted that it had nevercollected data on the numbers of ‘health tourists’ who used the NHS(Hansard, 2004a). Indeed, the Secretary of State for Health stated veryclearly that they had no data because ‘trusts are not required to submitdata on the numbers of overseas visitors treated or the costs of doingso, not least because overseas visitors, as defined by the charging reg-ulations, are not automatically “health tourists” ’ (Hansard, 2003). Thedistinction between ‘health tourists’ and other ‘overseas visitors’ (notethe terminology here too) was recognized and sometimes even com-mented on, but not without constantly re-emphasizing health tourism.For example, the government’s 2003 Consultation on amending the1989 (Charges to Overseas Visitors) Regulations began its justificationfor making new proposals with reference to the ‘considerable publicconcern about the abuse of the NHS by health tourists’ (Departmentof Health, 2003, p. 7). It followed this by adding that much mediacoverage was ‘confused and inaccurate’, but throughout it repeatedunsubstantiated suggestions of ‘abuse’ due to ‘loopholes’ in the reg-ulations (Department of Health, 2003, p. 7; Kelley and Stevenson,2006).

Not all government ministers were equally circumspect in differen-tiating ‘health tourists’ from other overseas users of health services.Beverley Hughes, then Minister of State for Immigration, Citizenshipand Counter Terrorism, wrote that the purpose of the proposals toamend the regulations was ‘to remove certain loopholes that have beenexploited by health tourists. In particular anyone who has no legal basisto remain in the United Kingdom will be liable to be charged for NHStreatment, except for emergencies or conditions for which they werealready receiving treatment before the final determination of their sta-tus’ (Hansard, 2004b, our emphasis). If the Home Office itself wasconflating this issue, small wonder that others were picking it up. Dis-tinctions between the groups being targeted by the proposed policieswere easily eroded in both the press and in parliament. This could be


done simply by linking or identifying health tourists with other ‘sus-pect’ foreigners. For example, one contributor to a House of Lords debatestated that: ‘A report last May by the Centre for Policy Studies suggestedthat some health tourists run up bills of over £50,000. In that report, onesenior consultant estimated that 20 per cent of patients on his inner-cityward were asylum seekers, refugees and foreign nationals who were notentitled to NHS treatment’ (Hansard, 2004c).

So ‘health tourists’ became a catch-all concept suggesting the mostmercenary type of abuse of the NHS which all foreigners could be per-ceived to be party to, and which in some vague way also implicated‘failed’ asylum seekers, the group in the public mind with the leastlegal claim to services. Indeed, even sympathetic commentators remainconfused. An article in the social work magazine Community Care saidof later proposals to charge for primary care, that ‘They are designedto tackle so-called health tourism, where people come to the UK andpresent false asylum claims to receive health care’ (Ahmed, 2007).

3.2. Enforcing the rules

Increased restriction of NHS entitlements serves also to provide a basisfor new government policies on surveillance and identity cards. Theless comprehensive and universal a public service, the more possibili-ties exist for fraud or perceived fraud, and the more procedures needto be implemented to distinguish the eligible from the ineligible. Thepresumption of a widespread risk of fraud helps to justify increasedsurveillance, in particular the creation of ‘entitlement cards’ whoseobject is to deny access to many basic rights, including that of accessto health care. But even without legislation for identity or entitlementcards, complex regulations have had to be developed on how to imple-ment charges. Within the NHS, a good part of the current regulationsaddress themselves to the statutory role of hospital trusts to determine theeligibility of patients to free care (NHS, 2004, p. 5). This represents ashift from the earlier 1989 regulations, which merely expected the rel-evant Health Authority to make ‘reasonable’ enquiries as to a person’sliability to charges ‘including the state of health of that overseas visitor’(NHS, 1989).

What has alarmed many people, including health workers at all levels,as well as those working with migrants, is both the non-clinical cri-teria for eligibility that are now applied and the close and deliberateintegration of this process with attempts to curb ‘illegal’ immigration.Thus responsibility for ensuring that patients who are not ‘ordinarilyresident’ in the UK are identified has now been devolved to agencies

150 Case Studies

not normally charged with law enforcement. This procedure forms anexplicit part of the government’s new enforcement policy as set out inthe Home Office’s strategy document Enforcing the Rules: A Strategy toPromote or Enforce Compliance Without Immigration Rules (Home Office,2007, p. 19).

This strategy sought to justify a more open policy towards migrantworkers at the expense of those it was prepared to deem illegal, to forcethem into the category which we have called ‘undocumented’. To dothis it sought ways increasingly to criminalize the undocumented and torestrict their ‘privileges’, many of which, including the rights to shelterand health care, are regarded by many as basic human rights. It went stillfurther by proposing the development of mechanisms to involve civilsociety agencies such as health and other public service providers, andprivate agencies such as banks, in accessing information and determin-ing themselves whether or not ‘migrants are here legally and entitledto services’ as a means of applying sanctions to those living ‘illegally’(Home Office, 2007, p. 5).

It is this Home Office strategy, Enforcing the Rules, that helps us under-stand that restrictions to access are not about cost. The documentacknowledged that illegal migrants ‘were unlikely to place a great strainon the NHS as most are thought to be young and therefore relativelyhealthy’ (Home Office, 2007, p. 13) but it planned ‘a review of accessrules for NHS care for foreign nationals to simplify the process of applyingcontrols’ (Home Office, 2007, p. 14, our emphasis).

Enforcing the Rules was not about costs or cost recovery to the NHS oreven about preventing specifically defined ‘abuses’. It sought to redefinethe idea of comprehensive welfare, in order to ‘shut down inappropri-ate access that are privileges for people who are in the UK legitimately’(Home Office, 2007, p. 24). This is the clearest re-statement of theVictorian distinction between the deserving and undeserving poor. Pre-senting a continuum of ‘harm’ caused by illegal immigration, it assignedculpability to migrants for anything from involvement in terrorismto responsibility for ‘reduced community cohesion’ and ‘undermin-ing public confidence in the immigration system’ (Home Office, 2007,p. 11). Such harm must be challenged by sanctions, and it is here thatwe find a shift in the underlying basis of NHS charging and restrictionson entitlement:

For those that cause the most harm for example criminal activity, thiswill mean prosecution, detention and removal or deportation . . . Forothers, the sanctions will increase with non-compliance to ensure


that living illegally becomes ever more uncomfortable and con-strained until they leave or are removed.

(Home Office, 2007, pp. 23–24)

In this scenario the object of NHS charging was to be part of a wholesaleplan to drive people away. This remains the premise behind continuinggovernment attempts to sustain and define eligibility for free treatmentin the face of legal challenges. Thus, in the face of criticism that denial ofaccess can damage physical or mental health, the assumption is that itdoesn’t matter – they shouldn’t be here anyway. We have moved a longway from Aneurin Bevan’s affirmation that ‘One of the consequencesof the universality of the British health service is the free treatment ofoverseas visitors’ (Bevan, In Place of Fear, 1952, cited in Manchester LawCentre, 1982).

4. Current entitlements to NHS health care

The rules governing entitlement to free NHS care in the UK areextremely complex. Cross-cutting rules relate to nationality, immigra-tion status, type of illness and type of care. Some foreign nationals areentitled to care as a result of health care agreements between their homecountry and the UK or through EU arrangements. Some visa categoriescarry with them immediate entitlement to free care on arrival in theUK, while other visa categories disentitle the individual to free NHScare throughout their stay in the country. For many migrants, theirentitlement depends on whether or not they are ‘ordinarily resident’in the UK.

To be ‘ordinarily resident’, an individual must be lawfully in the coun-try and demonstrate that they are ‘settled’ in the UK. There are no hardand fast rules about what is required to establish that an individual is‘settled’ in the UK. Department of Health guidance states that an indi-vidual who has been in the country for less than six months is unlikelyto be able to establish that they are ‘ordinarily resident’, however, this isnot a rigid timeframe (NHS, 2004).

There are exemptions from this entitlement system for some medicalconditions and some specific types of care which apply to all migrants.Free NHS care is available for compulsory mental health treatment andtreatment for communicable diseases including tuberculosis (TB) andsexually transmitted infections. HIV/AIDS testing is free; however treat-ment is not. Emergency care provided in Accident and Emergency isavailable free of charge, however subsequent care is charged for. All care

152 Case Studies

which is classed as ‘immediately necessary’ must be provided withoutdelay irrespective of the individual’s ability to pay, but the person willsubsequently be charged for their care. The definition of ‘immediatelynecessary’ is unclear and dependent on clinician judgement, but doesinclude all maternity care.

Anyone who is not entitled to free NHS care and whose condition doesnot fall within the exemptions is required to pay for NHS secondary carein advance. Secondary care consists of hospital-based and other special-ist care. For migrants who are destitute or on low incomes, this canbe an insurmountable barrier to accessing hospital care. Primary care,consisting of GP and other direct access health services, is subject todifferent rules. Access to primary care is dependent on GPs, who retainthe discretion to provide free NHS care to anyone, irrespective of theirimmigration status.

The complex rules governing charging for secondary care for refusedasylum seekers and other migrants were introduced in 2004. In that year,the government also consulted on a proposal to introduce charging forGP and other primary care although this did not proceed. The followingsections focus on the situation of asylum seekers, refugees and undoc-umented migrants as prime examples of migrants whose entitlementsto free NHS care have become unclear or have been called into ques-tion since the 2004 changes to hospital charging regulations and theproposals to charge for primary health care.

4.1. Entitled to free NHS secondary care

Individuals in the UK who are escaping persecution in their home coun-try are classified in a number of ways by the UK government. Thosewho have made a formal claim for refugee status are classed as ‘asylumseekers’. Those whose claim has succeeded are described as ‘refugees’ oras ‘having leave to remain’, depending on the specific status grantedto them. Those whose claim has been refused are described as ‘failed’asylum seekers.1 These terms are constructions which relate not tothe person’s experience of persecution and flight, but to their locationwithin the UK immigration system.

Asylum seekers have leave to remain in the country while they seekto prove their claim. They have full entitlement to free NHS care duringthis period. If their claim is successful, they have the same entitlementto health care as any other UK resident. If their claim is unsuccess-ful then their circumstances become more difficult. The entitlement ofrefused asylum seekers to free NHS care is contested. This is discussedbelow.


Asylum seekers and refused asylum seekers may be subject to incar-ceration in detention centres prior to deportation or, in some cases,throughout the consideration of their claim. Health care in deten-tion centres is provided free of charge but is not provided by theNHS. A recent official report criticized the quality of care provided anddescribed it as unsuitable for individuals with serious health problemsand for those in detention for long periods (HM Chief Inspector ofPrisons, 2006).

Refugees and asylum seekers often have very poor health. They facehealth problems related to poverty and overcrowding, in common withother deprived or excluded groups. Many have experienced the psycho-logical and medical effects of fleeing war and conflict in their homecountry. They may have undergone torture and or other traumaticexperiences. They may have unrecognized or untreated medical con-ditions such as congenital heart disease or TB (BMA, 2002; Burnett andFassil, 2002).

This group faces significant barriers to accessing health care. Languageand cultural barriers may affect attendance for health care and effectivecommunication with their clinician. They may have difficulty obtaininginformation about how the UK health system works. In addition, thepractice of dispersing asylum seekers to new locations can contributeto poor continuity of care. Confusion about entitlement to care can alsoaffect access to care, with evidence of hospital trusts requesting paymentfrom asylum seekers who are entitled to free care (Joint Committee onHuman Rights, 2007).

Case 1

A Chinese woman asylum seeker sought maternity care from ahospital shortly after her arrival in the country. The hospitalgave her a scan as emergency treatment, but refused anotherappointment. Instead, the midwife asked her to go to the Over-seas Visitor Manager. When she saw the Overseas Visitor Manager,there was no interpreter and she did not know what was going on.Midwifery staff refused subsequent appointments and referred herback to the Overseas Visitor Manager. She sought assistance froma voluntary organization. An advocate advised the hospital that,as an asylum seeker, the woman was entitled to care and she wasgiven an appointment (Medact, unpublished research).

154 Case Studies

The experiences of the asylum process in the UK can also impact neg-atively on health. There is evidence that the health status of refugeesand asylum seekers worsens in the two to three years after entry tothe UK (BMA, 2002). A small-scale study of female asylum seekers inScotland found that 54 per cent reported worse health in the UK than intheir home country (London School of Hygiene and Tropical Medicineand Scottish Refugee Council, 2009). Of these women, 57 per cent metthe criteria for Post Traumatic Stress Disorder and 20 per cent reportedsuicidal thoughts in the past week.

The ultimate effect of these policies is to exacerbate the very poorhealth of the individual affected. In maternity care, for example, refugeeand asylum seekers made up 12 per cent of all maternal deaths in2003–2005 (G. Lewis, 2007), despite this group constituting less than0.5 per cent of the population.

4.2. Not entitled to free NHS secondary care

Undocumented migrants, including refused asylum seekers, are notentitled to free NHS secondary care. As with all migrants, they mayaccess primary care at the discretion of the GP. There are numerousreports of undocumented migrants who have been unable to obtainprimary care. This can result in individuals approaching Accident andEmergency services for primary care (Project London, 2006; Feldman,2008).

Case 2

J is a Roma woman with diabetes and high blood pressure. Shewas referred to the Refugee Council after being turned away byevery local GP in her area. When she went to her local Accidentand Emergency department, they too turned her away, saying hercondition wasn’t life threatening or an emergency, that if shewanted health care, she would have to pay. J is destitute. (Kelleyand Stevenson, 2006)

One anomaly in the rules governing entitlement to free NHS care is theset of primary care services covered by the HC2 form. The NHS chargesfor NHS prescriptions, dental care, sight tests, wigs, and spinal and othersupports. These services are free of charge to people on low incomeswho hold a HC2 certificate. This certificate also entitles the holder to


vouchers to assist them with the costs of glasses and contact lenses,and some assistance with transport costs. The HC2 certificate is unusualin that it is available to all who meet the income test irrespective oftheir immigration status. Many undocumented migrants who are notentitled to free NHS hospital care can obtain a HC2 certificate. Indeed,current entitlement rules from the Department of Health recommendthat practitioners who come into contact with asylum seekers shouldencourage them to apply (our emphasis). The rules also state specifi-cally that ‘failed asylum seekers can also apply’ (Department of Health,2008).

The denial of free NHS care to refused asylum seekers has been con-tested. New rules introduced in 2004 left refused asylum seekers withoutentitlement to free NHS care even if they receive ‘Section 4’ support,discussed below. If they were receiving care at the time that they hadexhausted all their appeal rights, that care could continue free of charge.However, new care which did not fall within the exemptions was onlyavailable if they could pay in advance. A 2008 court case reviewed therelevant regulations and guidance and decided that many refused asy-lum seekers were entitled to free care. This decision was overturned in2009 (see ‘The Court Challenge’ below).

For refused asylum seekers, restrictions to NHS entitlements are com-pounded by the denial of the right to work and restricted access to statebenefits and services. If their claim is refused, asylum seekers will havetheir regular support from the UK Borders Agency (UKBA) terminated.But if they find themselves destitute, they may apply for ‘Section 4’support, which is granted subject to very stringent conditions. This sup-port consists of accommodation and a weekly supermarket voucher for£35 per individual in a family, redeemable at a specified supermarketchain. The value of this voucher (less than 60 per cent of the adultincome support rate) is extremely low, leaving refused asylum seekersin poverty. There is no access to cash to cover costs such as transport,nor is there any choice in the location of the accommodation.

Section 4 support is not available to all refused asylum seekers, leavingmany refused asylum seekers in destitution. In order to claim Section 4support, refused asylum seekers must either sign up to return to theirhome country, which many are not prepared to do, or seek support onother grounds, mostly following the submission of a fresh asylum claimor the inability to travel for physical or medical reasons. However, sup-port for medical reasons may be for short periods only. For example,pregnant women can only claim this form of Section 4 support from‘late pregnancy’ (approximately 34 weeks gestation) till six weeks after

156 Case Studies

the birth, unless there are significant complications earlier in the preg-nancy. All applicants for Section 4 support must also provide evidencethat they are destitute. It is not sufficient to demonstrate that they arenot entitled to work or to receive other benefits.

Destitute refused asylum seekers not granted Section 4 support arealso denied access to other forms of social housing and to most govern-ment services. Local authorities have child protection responsibilities inrelation to children and young people who are dependents of refusedasylum seekers, but have no obligations in relation to adults unless theyhave care needs covered by the National Assistance Act 1948 (NationalAIDS Trust and Hackney Community Law Centre, 2008). Many refusedasylum seekers rely on support from church groups and friends. Manyare homeless and live in extremely difficult circumstances (H. Lewis,2007; The Children’s Society, 2008).

5. The court challenge

Denial of free NHS hospital treatment to refused asylum seekers waschallenged in a recent judicial review, (R (A) v. Secretary of State forHealth). In April 2008, the High Court judgement resulted in changes inentitlement to free NHS care for refused asylum seekers. The case con-cerned A, a Palestinian man, who suffered from chronic liver disease. Hewas a refused asylum seeker living on UKBA Section 4 voucher paymentsof £35 per week. The hospital trust asked him to pay for care and ini-tially refused care when he was unable to pay. After A commenced legalaction, the trust agreed to provide free treatment. The claimant soughtpermission to continue the case against the Secretary of State and thiswas granted.

The case addressed the narrow legal question of the meaning of‘ordinarily resident’. In a departure from previous decisions, the courtconcluded that an individual did not need leave to remain in the UKin order to be ‘ordinarily resident’ but did need to be lawfully present.Refused asylum seekers who applied for asylum at their port of entryand received temporary admission to the UK are lawfully present in thecountry. Refused asylum seekers who entered the country illegally andthen applied for asylum, referred to as ‘in country application’, are notlawfully in the country.

The judge considered the implications of the practice of drawing adistinction between ‘port of entry’ and ‘in country’ applicants, and con-cluded that it was not workable for clinicians. He decided that all asylumseekers granted temporary admission should be considered ‘ordinarilyresident’ so long as they can demonstrate an intention to remain here.


In practical terms, this meant that individuals who could show evidencethat they had been granted temporary admission were entitled to accessfree NHS care unless they had been issued with removal directions.

The Department of Health overturned this decision on appeal in 2009(The Queen (on the application of YA) v. Secretary of State for Health).The court decided that an individual was not ‘ordinarily resident’ unlessthere was some kind of positive authorization to be in the UK. It wasnot enough that an individual had been granted temporary admission.This meant that refused asylum seekers are not entitled to free NHS care.The applicant’s inability to obtain Legal Aid funding has meant that ithas not been possible to mount a challenge to this ruling in the newSupreme Court.

The judicial review also examined the Department of Health guidanceon charging for secondary care and concluded that this was unlawful. Itruled that the guidance on the interpretation of ‘urgent’ and ‘immedi-ately necessary’ treatment was too restrictive as the guidance does notspecify what hospitals should do if a patient is unable to pay and cannotreturn to their home country for the time being.

6. New legislation

Meanwhile, at the same time as this appeal was underway, the govern-ment was pursuing other legislation to address entitlement to healthcare for new migrants. The consultation document on the proposedImmigration and Citizenship Bill 2008 stated that the governmentwould simplify the complex legislation on access to benefits and services(Home Office UK Border Agency, 2008). They would review terms suchas ‘ordinarily resident’ and ‘lawfully present’ to ensure that the termsmet the policy objective of restricting access to benefits and services tothose migrants who have ‘earned the right to them’.

The Border, Citizenship and Immigration Act was passed in 2009. Itdid not address entitlement to health care or to other benefits and ser-vices. A proposed second bill in 2010 to address outstanding issues wasnot completed because of the General Election. At the time of writingthe new government’s intention is still not clear.

This legislation follows a series of policy reviews and public consul-tations on the question of overseas visitors’ entitlements to NHS care.In 2004, the government consulted on proposals to restrict entitlementto free NHS primary care (Department of Health, 2004). In 2007, theHome Office policy document, Enforcing the Rules, announced a reviewof entitlement to health care for foreign nationals (Home Office, 2007).This was expected to consider further restrictions on entitlement to

158 Case Studies

primary care for refused asylum seekers. Neither the outcome of the2004 consultation nor the review promised in Enforcing the Rules hasbeen released.

In 2010, the government consulted again on proposed changes toentitlement to health care for foreign nationals. Proposals includedentitlement to free healthcare for some refused asylum seekers and forunaccompanied minors, and the introduction of new obligations to dis-close personal information which would be used in recovering debt. Theconsultation began prior to a change of government and the fate ofthese proposals remains unclear.

7. Emerging restrictions on access to NHS primary care

While entitlement to primary care is based on different rules fromentitlement to secondary care and is subject to the discretion of GPs,there are emerging practices within the NHS which are restricting accessto primary care for people currently ineligible for free secondary carebut who hitherto have been entitled to free primary care. Primary CareTrusts (PCT), which fund primary care at a local level, play an importantrole in this process.

Voluntary organizations and lawyers working with undocumentedmigrants have reported that some GPs have been wrongly informedby their PCTs that they are not permitted to register undocumentedmigrants (Maternity Action, unpublished research, 2008–10). In someareas, the PCT Counter Fraud Teams have sought to prevent GPs provid-ing care to undocumented migrants. Counter Fraud Teams usually focuson theft, fraud or other criminal activity.

There is evidence that confusion about entitlement and the interven-tion of PCTs is affecting access to primary care services. Project London,which provides medical care to undocumented migrants, contacted150 GPs in London asking about registering a fictitious pregnant womanwith their practice (Project London, unpublished research, 2008). Ofthe practices contacted, 6.5 per cent said that they would register thewoman, 56 per cent refused to register the woman, 6.5 per cent said thedecision would be at the GP’s discretion, and the remainder asked foradditional documentation.

8. Implications of current policies

Many commentators have drawn attention to public health issues inresponse to proposals to change entitlements to hospital care and to


proposals to introduce charges for ‘overseas visitors’ in primary care.Particular concerns have been raised about the impact on public healthof delayed detection of TB or HIV, leading to failure to treat these con-ditions (House of Commons, 2005; Wood et al., 2007; Medsin, 2008).While TB treatment is exempted from charges, TB is only treatable if itis detected. Fears of being reported to the authorities, threats of charg-ing, experience of billing, a decline in trust between patient and doctor,recourse to parallel and even illegal markets in drugs all act as barriersto speedy and effective access to health care. Altogether, public healthstrategies involving mass childhood immunization, TB reduction andmaternal and infant mortality are likely to be undermined (Hargreaveset al., 2006, pp. 33–58).

The health of individual migrants, often presented as a ‘humani-tarian’ as opposed to a public health issue, is, of course, also at riskof being compromised if problems are not identified and treated. Ina 2001 critique of government policies which adversely affected thehealth of asylum seekers, when there was not yet a debate about theirentitlements, the British Medical Association and Medical Foundationpublished a dossier of 22 cases reflecting ‘the suffering of refugees’ (BMAand Medical Foundation for the Care of Victims of Torture, 2001). Thecases described delayed child development and failure to thrive, skinconditions, severe mental health problems, and ongoing effects of tor-ture in countries of origin, as well as health problems stemming fromthe physical and social conditions in which asylum seekers were living.

More recent studies have documented examples both of the impactof charging in secondary care and confusion about eligibility to bothprimary and secondary care. Cases include descriptions of unattendedhome birth delivery, refusal of post-natal follow-up, denial of cancertreatment, problems obtaining essential insulin for diabetics, lack oftreatment for prior injuries, and mental health problems, sometimesexacerbated by failure to treat serious conditions, leading to physi-cal pain and ill-health (Cheedella, 2006; Kelley and Stevenson, 2006;Project London, 2006; Feldman, 2008).

Moreover, the consequences of health policy go beyond healthimpacts alone, whether on individuals or on groups. Invidious distinc-tions between different groups in relation to their rights of access tohealth services are likely to increase barriers to access for all whosestatus is uncertain to the gatekeepers. Legitimizing discriminatory prac-tice against foreigners – a form of racism based on xenophobia ratherthan on particular physical or cultural characteristics – feeds further dis-crimination (Hargreaves et al., 2006). Studies in two London Primary

160 Case Studies

Care Trusts found evidence of lack of access to GPs for people whowere legally quite entitled to primary care, including more stringentdemands for documentation to prove eligibility (Hargreaves et al.,2006; Feldman, 2008). While anti-discrimination legislation requires allpotential patients to be treated in the same way, in practice demandsto prove eligibility are more likely to be made of foreigners than ofapparently native born people, whatever the entitlements of each group.

9. Conclusion

The complex charging regulations which the government is still in theprocess of developing have been justified in terms of costs to the NHSand fraudulent ‘health tourism’. Closer scrutiny reveals that they arepart of a long tradition of attempts to exclude foreigners and ‘outsiders’from health care and public welfare more broadly, despite failure to pro-vide any empirical justification in terms of cost, or consideration of theirimpact on other policy goals such as public health. At present the impo-sition of charges is part of a broader attempt to integrate immigrationcontrols with the administration of public services. This is itself costlybut is also invidious and leads to contradictory outcomes in terms ofhealth policy. It results in increased use of emergency services whenthe policy is to reduce inappropriate use of such services, poorer mater-nal health in particular, and increased risks to public health. The NHScharging policy also risks undermining other social policies such asrace equality, refugee integration and programmes to integrate skilledmigrant workers, and is in danger of creating a more intolerant andunequal society.

Note

1. When not discussing government-defined categories, we use the term ‘refused’rather than ‘failed’ asylum seekers to refer to those whose claim for asylum hasbeen refused by the Home Office.

References

Ahmed, M. (2007) ‘Primary health care charges to apply if asylum claim fails’,Community Care, 15 November, available at: http://www.communitycare.co.uk/Articles/2007/11/15/106461/Primary-health-care-charges-to-apply-if-asylum-claim.htm

BBC (2003) ‘Tories target “health tourism” ’, 1 June, available at: http://news.bbc.co.uk/1/hi/uk_politics/2954438.stm.


British Medical Association (BMA) (2002) Asylum Seekers: Meeting their HealthcareNeeds (London: British Medical Association).

British Medical Association and Medical Foundation for the Care of Victims ofTorture (2001) Asylum Seekers and Health, available at: http://www.bma.org.uk/ap.nsf/Content/Asylumseekershealthdossier.

Burnett, A. and Fassil, Y. (2002) Meeting the Health Needs of Refugees and AsylumSeekers in the Uk: An Information and Resource Pack for Health Workers (London:Department of Health).

Cheedella, K. (2006) Barriers for Failed Asylum Seekers to Accessing NHS ServicesIncluding Charging for Healthcare, and the Consequences of Destitution: A Studyin Leeds, United Kingdom, unpublished B.Sc. dissertation, University of Leeds,available at: http://www.medact.org/article_refugee.php?articleID=592.

The Children’s Society (2008) Living on the Edge of Despair: Destitution amongstAsylum Seeking and Refugee Children (UK: The Children’s Society).

Chinai, R. and R. Goswami (2007) ‘Medical visas mark growth of Indian medicaltourism’, Bulletin of the World Health Organisation, 85(3), 164–165.

Cohen, S. (2003) No One is Illegal (Stoke on Trent, UK and Sterling USA: TrenthamBooks).

Cohen, S. and D. Hayes (1998) They Make You Sick: Essays on Immigration Con-trols and Health (Manchester Metropolitan University: Greater ManchesterImmigration Aid Unit and Department of Applied Community Studies).

Conservative Party (2003) Asylum Crisis Undermining NHS Care, availableat: http://www.conservatives.com/tile.do?def=news.story.page&obj_id=50595(accessed 3 October 2008).

Cooley, L. and D. Sriskandarajah (2007) ‘Facts and figures: a context for under-standing the issues’, in D. Flynn and Z. Williams (eds) Towards a Progres-sive Immigration Policy (London: Compass, Migrants Rights Network, BarrowCadbury Trust).

Department of Health (2003) Proposed Amendments to the National Health Ser-vice (Charges to Overseas Visitors) Regulations 1989: A consultation (London:Department of Health publications).

Department of Health (2004) Proposals to Exclude Overseas Visitors from Eligibility toFree Nhs Primary Medical Services (London: Department of Health publications).

Department of Health (2008) ‘Table of entitlement to NHS treatment (cor-rect as of May 2008)’, http://www.dh.gov.uk/en/Healthcare/International/asylumseekersandrefugees/index.htm (accessed 14 September 2008).

Feldman, R. (2008) ‘Evaluation of Hounslow Health Advocacy Projectfor Refugees and Asylum Seekers’, Report to Hounslow Primary CareTrust.

Gordon, P. and A. Newnham (1985) Passport to Benefits? Racism in Social Security(London: Child Poverty Action Group).

Hansard (2002a) Baroness Blackstone, 20 March.Hansard (2002b) Lord Clement-Jones, 19 November.Hansard (2003) Hutton, 30 October.Hansard (2004a) Ladyman, 12 January.Hansard (2004b) Hughes, 2 March.Hansard (2004c) Earl Howe, 5 March.Hargreaves, S., J. Friedland, A. Holmes and S. Saxena (2006) ‘The Identification

and Charging of Overseas Visitors at NHS Services in Newham: a Consultation’,

162 Case Studies

Final Report (London: Newham Project Board, Newham Primary Care Trust,London Borough of Newham).

HM Chief Inspector of Prisons (2006) Inquiry into the Quality of Healthcare at Yarl’sWood Immigration Removal Centre (London: HM Inspectorate of Prisons).

Home Office (2007) Enforcing the Rules: A Strategy to Ensure and Enforce Compliancewith Our Immigration Laws (London: COI).

Home Office UK Border Agency (2008) Making Change Stick: An Introduction to theImmigration and Citizenship Bill (London: HMSO).

House of Commons Health Committee (2005) New Developments in Sexual Healthand HIV/AIDS Policy, Third Report of Session 2004–2005, Volume I, HC 252-I.

Institute for public policy research (IPPR) (2006) Irregular Migration in the UK(London: Institute of Public Policy Research).

Joint Committee on Human Rights (2007) The Treatment of Asylum Seekers: TenthReport of Session 2006–7, Volume 1 (London: The Stationery Office).

Kelley, N. and Stevenson, J. (2006) First Do No Harm: Denying Health Care to PeopleWhose Asylum Claims Have Failed (London: Refugee Council).

Kyambi, S. (2005) Beyond Black and White: Mapping New Immigrant Communities(London: Institute of Public Policy Research).

Lewis, G. (ed.) (2007) ‘The Confidential Enquiry into Maternity and Child Health(CEMACH) Saving Mother’s Lives: Reviewing Maternal Deaths to Make Moth-erhood Safer 2003–2005’, the seventh report on Confidential Enquiries intoMaternal Deaths in the United Kingdom (London: CEMACH).

Lewis, H. (2007) Destitution in Leeds: The Experiences of People Seeking Asylum andSupporting Agencies (Leeds: Joseph Rowntree Charitable Trust).

London School of Hygiene and Tropical Medicine, and the Scottish RefugeeCouncil (2009) Asylum-Seeking Women, Violence and Health: Results from a PilotStudy in Scotland and Belgium (London: London School of Tropical Medicine &Scottish Refugee Council).

Manchester Law Centre (1982) From Ill Treatment to No Treatment (Manchester:Manchester Law Centre Immigration Handbook No. 8).

Medsin (2008) Proposals to Exclude Overseas Visitors from Eligibility to Free NHSPrimary Medical Services (Cambridge: Global Health Advocacy Project), availableat: www.wherestheconsultation.org.

National AIDS Trust and Hackney Community Law Centre (2008) Advice forAsylum Seekers in Receipt of Section 21 Support (London: National AIDS Trust),available at: http://www.nat.org.uk/document/478.

National Audit Office (2005) Returning Failed Asylum Applicants (London: TheStationery Office).

National Health Service (1989) (Charges to Overseas Visitors) Regulations 1989,available at: http://www.opsi.gov.uk/si/si1989/Uksi_19890306_en_1.htm.

National Health Service (2004) Implementing the Overseas Visitors Hospital Charg-ing Regulations: Guidance for Nhs Trust Hospitals in England, (as amended2007).

Oswald, S. and T. Henthorne (1999) ‘Health tourism: A Niching strategy formarketplace survival in Cuba’, Global Development Studies, 1(3–4), 220–233.

Platform for International Cooperation on Undocumented Migrants (PICUM)(2007) Access to Health Care for Undocumented Migrants in Europe (Brussels:PICUM).


Project London (2006) Helping Vulnerable People to Access Healthcare (London:Médecins du Monde).

The Queen (on the application of YA) v. Secretary of State for Health [2009] EWCACiv 225.

R (A) v. Secretary of State for Health and West Middlesex University Hospital NHS Trust(interested party) [2008] EWHC 855 (Admin).Somerville, W. (2007) Immigrationunder New Labour (Bristol: Policy Press).

Southall Black Sisters (2007) No Recourse Campaign Leaflet, available at: http://www.southallblacksisters.org.uk/publications.html.

Vollmer, B. (2009) Irregular migration in the United Kingdom: Counting the Uncount-able, Policy brief CLANDESTINO Project, available at: http://clandestino.eliamep.gr/policy-briefs/#more-105.

Wintour, P. (2007) ‘Smith seeks to restore trust in immigration system,’ TheGuardian, 6 December.

Wood, C. et al. (2007) ‘Global health and foreign policy’ (letter), The Lancet, 369,1688–1689.

Woodbridge, J. (2005) Sizing the Unauthorised (Illegal) Migrant Population in theUnited Kingdom in 2001 (Home Office).

6A Market Place for Migrants:Mobility, Settlement and SocialProtection in KeralaMythri Prasad-Aleyamma

1. Introduction

This chapter focuses on a Sunday Bengali market for migrant workers inPerumbavoor, a small town situated close to the city of Kochi in Kerala,South India, to interrogate the concept of social protection. The market,which sprang up in response to the specific needs of migrant workersfrom West Bengal, Bihar and Nepal, serves also as a ‘meeting point’of workers. Such markets, by acting as a point of contact, facilitate anetwork of migrant workers, contractors, traders and trade unions. TheSunday market is a place where workers seek leisure, make friends, sendmoney home through agents and settle disputes. The migrant workerswho visit the market are mainly from north and north-eastern states ofIndia, namely West Bengal, Orissa and Assam and, to a lesser extent,from Jharkhand and Bihar and from the neighbouring country, Nepal.They work in stone quarries, construction sites, and plywood, chemicaland plastic factories in Angamaly, Aluva and Perumbavoor and othersmall towns in Ernakulam district. Migrant labour is involved in allsectors which involve manual labour.

The conditions that underpin these migrations are varied, includ-ing low wages, lack of employment, natural calamities and feudal landrelations in places of origin, and the development of an elaboraterecruitment network especially for construction companies. Exploitativeconditions of work prevail in these work sites, including discriminatorywages, long working hours, cheating by contractors regarding wages andlack of safety measures. I do not intend to take up these issues for thepresent chapter unless they are relevant to or are helpful in addressing

164

Mythri Prasad-Aleyamma 165

the central concerns of the chapter, which could be summarized interms of the following questions: what are the connections betweencommunity formation and social protection? Is a dichotomous under-standing of social protection as formal and informal analytically useful?What, if any, are the contradictions between migrant mobility and con-cepts of social protection, or, do concepts of social protection carry asedentary bias within them?

In this chapter I use the Sunday market in two ways: firstly as animportant social institution in its own right, and secondly as a vehicleto explore these questions concerning migrant labour and social pro-tection. The fieldwork was carried out in 2008 and 2009 and involvedinterviews with migrant workers, their employers and officials of thegovernment labour department. The fieldwork in the market involvedparticipant observation as well as interviews with workers and hawk-ers to understand the nature of activities and actors in the market andhow these contribute to an understanding of the nature of social pro-tection for these migrants. The chapter also uses archival material fromthe government labour department.

2. Making of a market

This section presents a brief ethnographic description of the market,showing the range and complexity of the activities that the migrantworkers engage in there. It first introduces the concept of ‘place making’and discusses how migrant workers use or ‘make place’ in the market, asa means of community formation and political mobilization.

Migrant labour makes place in two different but inseparable ways: bycontributing labour to build physical space like roads, buildings andbridges, and by using and interpreting and extracting meaning out ofspaces like Sunday markets. While these city forms – roads and bridges –displace local communities, their very building engenders new spatialpractices and communities like Sunday markets in another part of thecity or its periphery. I use the term ‘place making’ in the sense of Gieryn(2000) who said, drawing on the seminal work of de Certeau (1984),that ‘places are endlessly made, not just when the powerful pursue theirambition through brick and mortar, not just when design professionalsgive form to function, but also when ordinary people extract from con-tinuous and abstract space a bounded, identified, meaningful, named,and significant place’. A sense of place and geographical proximity arecrucial to community formation and the ability to make claims on thestate as a community.

166 Case Studies

Perumbavoor, where the market is situated, is a municipality inErnakulam district that has a population of 26,547 and an area of13.6 square kilometres as per the 2001 census. Agriculture remains themain occupation and economic activity that provides the economicbase of the town. However, the town has seen a spurt in industrialproduction as small and medium-sized industrial units started mush-rooming in the adjacent areas and towns. Some of these units aresituated in the industrial estates in Aluva, another town in the district.These units produce chemicals, plywood and garments as well as pro-cess food, rubber and agricultural products. Migrant workers form theprimary source of manual labour in these units. The local populationof Malayalis also work in these factories, but mostly in managementand supervisory positions.1 There has also been a rise in constructionand quarrying activities in the area, which has led to an increase inconstruction and stone quarry workers.

Getting off at Gandhi junction and walking downtown to Jyothitheatre junction in Perumbavoor, one can see the crowds of workershaggling with roadside vendors selling various goods, mainly second-hand clothes, plastic utensils and CDs of Oriya and Bengali songs andmovies. Bidis (cigarettes) are brought in from as far as Murshidabadin West Bengal since Bengali workers prefer them to bidis producedlocally. Besides construction workers, there are men working in cloth-making units, spice factories, plywood factories, plastic factories andso on.

The workers in plywood factories started the practice of coming totown on Sundays, as Sundays are holidays. As quarrying and buildingwork increased as the building boom set in, the number of workers wentup. Workers from Aluva, Angamaly and other towns and industrial cen-tres in the district, as well as from nearby districts like Kottayam andThrissur, started joining them. Shereef, a Malayali shop owner in themarket, remembers that the market came into its present form gradu-ally over the past five years, though migrant workers used to come tothe market even ten years back, which is when large scale migration tothe district from north and north-eastern India started. The big shopsremain closed on Sundays while the space in front of them becomesthe Sunday market, with traders coming in from Tamilnadu to sell theirwares exclusively to the migrant workers. The market is divided into‘Oriya and Bengali sections’ with a less apparent ‘Assamese section’ afew yards away on the other side of the road.

On other days of the week the market acts as a ‘pick-up point’ forworkers. Contractors come in trucks to hire workers for construction


sites and quarrying units. Siddique, a shopkeeper who acts as ago-between explains the hiring process at the market:

I do business here. I inherited the shop from my father. Workers tellme to inform them if there is work in any company. The employersalso tell me to inform them when workers come down from Bengal orOrissa. They leave their phone numbers with me. I call them up andtell them that workers have come down. This is all I do. The employ-ers [muthalalimar, literally meaning capitalists] give something inreturn as I spend money on making calls. These workers smoke andchew pan and so they come here early in the morning. Once I startedputting the workers and contractors/employers in touch, workersstarted coming in search of Siddikkakeda kada [Brother Siddique’sshop]. I also give tools like pick axes, shovels etc on loan to work-ers especially to Tamil workers as Bhayeemar [a slang word used torefer to North Indian workers, originating from the Hindi word bhai]do not do earthwork.

Manoj who sits in AM Street, the Oriya section of the market, sellsrings blessed by a holy man in Orissa. He has been coming to the mar-ket for the past eight years. He said that the market started attractingpeople in 2000 when the plywood factories began employing migrantworkers from Assam and Nepal. Manoj lives near Aluva. He said thathe had been to many cities in the country. Another man chipped in,introducing himself as Ajay Mallick from Cuttack in Orissa. He worksin a plywood company Palikkavila near Perumbavoor. He described theprocess as ‘people and things attracting more people and things’.

The market has further social and political functions. It operates asa meeting point for workers from the same village in Orissa or WestBengal who have settled in different parts of the district of Ernakulam.Ajay, from Rayagada in Orissa, said that all meetings between friendsand gaavvalen (people from the same village) occurred in the market. Itwas a place where they borrowed and lent money. Transfer of money tovillages also happens in the market. Disputes and quarrels are discussedand solved in the market. The market is, furthermore, a place wheretrade unions attempt to address the workers. The Trade Union Centre ofIndia (TUCI), a trade union owing allegiance to the Communist Partyof India (Marxist Leninist) organizes meetings, bringing comrades fromOrissa and West Bengal to speak in the market.2 The trade union hadsuccessfully campaigned for compensation for workers who died in anaccident in a labour quarter.3 However, the attempts by the trade union

168 Case Studies

to organize the construction and quarry workers in the area have notbeen successful.

The highly unionized Malayali construction workers in the districtare recruited through trade unions in a system known as ‘site pidikkal’(capturing the site) whereby unions negotiate with the builder regard-ing the number of workers, wages and the amount of work. HiringMalayali construction workers is done either through unions or throughcontractors; Malayali construction workers do not stand in markets orintersections waiting to be picked up. Various trade unions affiliated topolitical parties have organized Malayali construction workers since the1980s. However, the unions do not give membership to migrant work-ers and have not attempted to organize them. In the plywood industry,the migrant workers have replaced local Malayali male workers, andwomen workers from neighbouring districts in Kerala and Tamilnadu.However, this replacement has happened gradually as new units sprangup and existing ones increased in scale. Malayali workers in the plywoodindustry have remained largely non-unionized.

Workers have acted in solidarity with their employers in the Sundaymarket. For example, in September 2008, the plywood factory ownersorganized a meeting against a move by the police to introduce identitycards for the workers. These were to be obtained after receiving a policeclearance certificate from the police station in the migrant’s place oforigin stating that he/she has no criminal record. Many of the workersreturned home because they did not have these identity cards. Salim, aplywood factory owner who employs migrant workers, estimated that10,000 went back. Plywood factories had to close down temporarily orslow down operations. Large numbers of workers participated in themeeting organized by the factory owners in protest against the police.The police did not pursue the issue of clearance cards. Other meetingswere organized by Communist Party of India (Marxist) and the tradeunion affiliated to it, Centre of Indian Trade Unions (CITU), demandingmore government intervention and the issuing of identity cards. AbdulMajeed, the president of the plywood manufacturers’ association and anemployer of migrant workers says:

They [migrant workers] are harassed here. The police asked them tobring police clearance certificates from their native places. We toldthe police that they cannot harass them. They are also Indian cit-izens. You can wait for a truck in Perumbavoor market, right? Ifsomebody tells you that you cannot, what will you do? We were withthe workers. Or rather, they were with us. We took out a procession.


I, P K Abdul Majeed, organized the procession and spoke in the pub-lic meeting. They are like migratory birds. Should we destroy theecosystem and drive away the migratory birds? The police snatchedthe voter’s ID card of a worker and tore it away when he was stand-ing in the Perumbavoor market. That is the only identification cardthese workers have. When they were making trouble to our workers,we acted. They were harassing them unnecessarily. They [the police]thought that we will stop the industry if they threaten us. Due tothe harassment, many workers started leaving and that’s when weorganized the procession and meeting. We took them to streets. Whydo they want to harass these poor people? We said it is not possibleto harass them. We will take out a procession again if needed. Manyleft Perumbavoor to work in the construction industry in Bangalorewhere wages were higher than what we paid. So we had to stop workfor some days. This crisis coincided with the onset of recession. It wasa terrible crisis.

This example shows how distinctive solidarities are forged betweensmall capital and migrant labour in the face of state power and surveil-lance. However, this solidarity is not between traditional unionizedlabour and capital but between a particular section of labour (migrantworkers) whose interests here coincide with those of small local capital-ists. This brief exploration of the range of social and political relation-ships in the Sunday market has shown how migrant workers attempt toreduce the precariousness of their position within Perumbavoor. Obser-vation of activities and relationships in the market helps to understandthe strategies migrants use to protect their position, many of whichdo not follow formal channels. Following from this, the next sectionexplores how far we can characterize such strategies as ‘informal’, andwhether the formal–informal dichotomy can hold together the concretereality of the Sunday market in its empirical complexity.

3. Market as a space of informality

Informal economies have been shown to be the dominant rather thanthe marginal form of economy in most parts of the world (Portes, 1996),and are often backed by political might. ‘Informality’ is characterized asthe realm of the personal, tacit, casual and often the chaotic. The ‘for-mal’ is often defined as the realm of the legal, and more importantlythat of the state. Misztal (2000) points out that while the latter is nec-essary for liberal democracy’s sense of universal rights and contractual

170 Case Studies

justice, the former is important for communication, trust and emotionalcommitment.

Migration facilitates ‘informalization’ of the economy in many ways.By providing cheap labour, it might erode the bargaining power of orga-nized local labour. It introduces labour practices which rely on informalnetworks. Such networks shape recruitment and settlement by usingethnicity and kinship for the mobilization of labour. The ethnic andcaste identities of the workers, supervisors and recruiters play an impor-tant role in the work process and in labour disciplining (Chakrabarty,1989).

Social relations and networks, which are often characterized as infor-mal, play an important role in shaping the Sunday market. Recruitmentnetworks, the casual and temporary nature of employment, the pave-ment traders, the credit networks, and even the pre-used clothes on sale,are all markers of a perceived informality in practice. The segregation ofthe market into Bengali and Oriya sections also points to the natureof solidarities forged here. These solidarities are based on affinities oflanguage, village and kinship. They are also usually categorized as con-stitutive of informal social and economic relations. These relationshipsprovide a moral and practicable basis for social protection.

What are the forms of social protection offered in the market? Infor-mation is exchanged about employment opportunities, wages in varioussectors and factories, and about the practices of contractors. When con-tractors or employers cheat on wages or on the nature of work, theyare summoned up before the owner of the Bengali restaurant, who isa Malayali, and a settlement is reached regarding the defaulted wages.When a migrant dies, his friends from the same village or his co-workerscontribute towards the expenses of transporting the dead body to thevillage and do the bureaucratic work necessary for such transportation.The market is the place from which such efforts are coordinated.

The market is also a site of resistance against the intervention of thestate to monitor and police the workers by insisting on police clear-ance cards. As already mentioned, the workers and the employers joinedhands in this resistance against the state. On the other hand, the marketis also a site where the trade union has made attempts to organize theworkers. The claims made by trade unions are formal in the sense thattrade unions are formally constituted organizations and operate withinformally established rules. The process of organization of a trade unionis also viewed as formal. Does this process thus fall outside the scope ofinformality?

We therefore need to ask how useful the concept of informality is asa way of characterizing the multiple and complex relationships forged


in the market. The very tangibility and contextual embeddedness ofinformality defies definitions. Defining informality as a residual cat-egory, something that remains after definitions, makes it a nebulousterm. Benjamin and Bhuvaneswari (2007) point out that the dualisticviews ‘formal and informal’ have been shown long ago to be conceptu-ally faulty and operationally fuzzy. For them, the persistent use of suchconcepts can only be explained from an ideological viewpoint whichprioritizes the needs of capital over that of poor groups in cities byperpetuating a situation of forced illegality. In the context of cities likeBangalore and Chennai, such ideologies are harnessed to rework insti-tutional structures and the policy environment to further claims of bigbusiness on locations in these cities. Papola (1980) also questions theanalytical significance of the idea of the informal sector and notes thata dichotomous frame might not be adequate to analyse the multiple andvaried elements of urban economies.

Both formal and informal strategies and relationships shape the spaceof the market. As Linda J. Seligmann (2004) notes in her book onPeruvian street vendors, the boundaries between the formal and theinformal are as porous and permeable as national borders, as peopleand things flow across them. She notes in the context of a market inCuzco in Peru that formal and informal markets are intertwined, andconsiders the market and the activities of the street vendors as capableof dissolving dualities such as formal and informal. The Sunday mar-ket, by letting people and things flow across these borders, acts as a fluidand at times a convivial space. In the face of exclusion from mainstreameconomic and social institutions like trade unions, political parties andlocal governing bodies like panchayats, the Sunday market is where cit-izenship is established, however briefly and albeit in a limited manner.Seeing the market as only informal, and hence residual, precludes seeingits empirical complexity as a space offering social protection in crucialsituations involving state oppression and surveillance. These forms ofsocial protection do not fall within the boundaries of formality or infor-mality. Moreover, the varied social relations that form the basis for socialprotection are not captured by the term informality.

4. Settlement and social protection

Social protection should not be viewed merely as a set of given poli-cies and legal formulations. Rather, it is activated through a politicalprocess of negotiation and struggle. Any discussion on social protectionfor migrant workers needs to take into account the parameters in whichsuch a negotiation and struggle is framed and the political and economic

172 Case Studies

context of the migration. For example, settlement, or at least an inten-tion to settle down, is seen by the state as an important prerequisite indetermining the eligibility of a person or a community to be welfarerecipients. In fact, most of the time, the moral basis for a communityis forged only if the families or persons in question are settled in closegeographic proximity.

Partha Chatterjee (2004), while discussing the political practices ofmarginalized communities in West Bengal, describes this process as onethat ‘gives the empirical form of a population group the moral attributesof a community’. He notes that marginalized communities access wel-fare and resist state power by opening up sites of negotiation andcontestation with governmental agencies. He calls these sites of nego-tiation ‘political society’, and thereby opens up a critique of the conceptof civil society which he feels is too sanitized to accommodate such com-plex forms of politics. This political society is in active relationship withthe procedures of governmentality. One of the cases Chatterjee takes upto substantiate his point is that of a heterogenous migrant colony inCalcutta (where migrants come from East and West Bengal). The empir-ical population group of the colony is given the moral attributes of acommunity by defining the neighbourhood in terms of kinship, by call-ing it a ‘family’. However, he notes that the geographical boundariesof the ‘colony’ are important in determining the neighbourhood andcommunity. Settlement or an intention for settlement is important indeciding the kind of claims a migrant is able to/wants to make. A rationcard, often, is a crucial document which could activate any mechanismof social security. A proof of residence, which is required to obtain aration card, is also a proof of prolonged settlement and a relative lack ofmobility.

Settlement is no longer, if it ever was, the central feature of muchof the movement occurring around the globe (Papastergiadis, 2000).Authors writing on migration in rural India (Breman, 1996; De Haan,1997) note that the predominant pattern of labour mobility is notmigration, but circulation. Deshingkar and Farrington (2009) also notethat livelihoods are multi-locational for many rural households, withimportant implications for settlement. My own previous work (Prasad-Aleyamma, 2009) suggests that journeys, rather than settlement, charac-terize the life of many of the workers, especially those who are engagedin construction work since large construction companies keep shiftingtheir worksites as projects move. The workers who had come to Keralafrom Saharsa district in Bihar to build a road had worked in Delhi,Belgaum (Karnataka), Ludhiana (Punjab) and Guwahati (Assam) before


coming to Kerala. Labour migration in the post-liberalization era, par-ticularly to sites of infrastructure building, has been characterized bydisparate journeys rather than circular movements to a specific place,such as a factory or a mine. Most of the migrant workers to construc-tion sites in Kerala are men. Many of the young workers frequentingthe Sunday market tell me of their intention to go back to their vil-lage and get married and ‘settle down’. These temporary movementsshape agency and determine the potential ways in which workers nego-tiate their everyday lives and work. Rogaly (2009) points out that muchof the labour geography on migration is concerned with ‘settledness’and that the spatial embeddedness of the migrant worker, who is rel-atively mobile for short periods of time, is neglected. He argues thattemporary migrants spend shorter periods of time working togetherthan settled workers, and live in temporary accommodation, whichmeans fewer possibilities for developing class-based antagonism andconsciousness.

Mobility/lack of settlement have to be considered in any discussion ofsocial protection, which is at present heavily biased in favour of settledcommunities. Policy research on migrant workers needs to incorporatehuman mobility as an active element in evaluating and formulatingsocial protection policy. Deshingkar and Farrington (2009) note thatpro-poor schemes are based on the assumption that populations aremore or less static. Castles (2010) points out that a certain ‘seden-tary bias’ cripples policy and research, which assumes away continuousmobility, and that this sedentary bias practised by developmental agen-cies has its roots in colonial policy, which viewed mobility of poorpeople as a threat to public order and prosperity.

Social protection does not begin or end in policy but is a continu-ous process of negotiation and contestation which uses both formal andinformal strategies. In Perumbavoor, although the idea behind the iden-tity cards was to monitor and police the workers and employers, it wasalso supposed to eventually grant the workers access to some means ofsocial security and rights. However, the workers resented this act of for-malization since it effectively restricts their mobility across employersand places. This problematizes social protection, especially the attemptsto formalize it, as efforts to ‘protect the migrant’ could very well turninto those of ‘protection from the migrant’. Invariably, procedures ofgovernmentality and social security are bound together. As in the caseof identity cards, the ‘formal’ is also the realm of the state, the motivesof which are often at variance with the interests of the migrant and theemployer, who share a space of informality.

174 Case Studies

Political society, in the sense used by Partha Chatterjee, works ina complex way in the case of the migrant. It could be said that thelocal is not easily negotiated by the unsettled migrant who considershimself/herself as transient. Even those who have stayed for as manyas eight years in Perumbavoor (and never visited their village since)do not consider themselves to be residents of Perumbavoor. This per-ceived transience has important implications for the formation of arights-based trade union. This could be one reason why, even after theearnest and persistent efforts of activists like T. C. Subrahmanian, theseworkers remain unorganized even in the face of wage discriminationand workplace accidents. Other trade unions have been hostile to theorganizational work that TUCI has undertaken among migrant workers.

5. Threats to citizenship

T. C. Subrahmanian remembers a strike he organized in a constructionsite after a worker died in an accident:

It was a lightning strike. I had organised it along with other comrades.The workers were really enraged by the death of a fellow worker. The work-ers used to stay in the construction site itself. Suddenly a group of menbelonging to the CITU head load workers’ union4 led a few goons into thesite. They carried big lathis (canes) and started beating the workers. Theworkers got scattered and were thrown out from work. Many went back totheir villages of origin. The construction companies pay local CITU men acommission for protection.

The hostility of the head load workers’ union, composed entirely ofMalayalis, is not necessarily because they perceive the migrant workersas a threat to their jobs, since Malayalis do not engage in construc-tion work in an extensive manner. The trade unions are embedded instructures of power which are not always shaped by class consciousness.Rather, they collude with small capital (as in the example above) andthe state (as in the case of the identity card) and form part of a politicalsociety composed of contingent interest groups and makeshift alliances.In this sense, the formality associated with a trade union withers away asits relationship with capital and local power structures comes to the fore.

The lone entry point to political society for migrants in Perumbavoorhitherto had been via their employers. It is easy to dismiss this as aninstance of the demonstration of strength by the plywood factory own-ers and lose sight of the state power and oppression against which they


were protesting, bound up as it is with the lack of emergence of class-based action. At times, migrants’ protests against the state gain moreurgency than those against exploitation within factories. This is not toobscure such exploitation, which is itself predicated upon the refusal ofthe state to accept the migrants as citizens.

The social relations in the market which could be construed as ‘infor-mal’ facilitate the articulation of more formal rights, as in the case of theidentity cards and the resultant resistance based at the market. But, asthe example of the strike in the construction site suggests, this resistanceis made tenuous by local power structures, which include trade unions.

The erosion of citizenship as people move within national bound-aries is made clear by the police clearance certificates which the policeinsisted upon in Perumbavoor. This also makes it clear that, in a multi-ethnic society, citizenship does not just relate to territorial borders. Inother words, substantive citizenship rights and entitlements which actu-alize citizenship (Levitt and Schiller, 2007) may erode even as nominalcitizenship (formal membership of a political unit) continues whenpeople, especially the poor, move within a country. This erosion isintensified by local power structures, including political society.

Organized labour forms part of this political society in Kerala and isanother axis of the local which threatens citizenship along with police.Local, here, is constructed out of the intersections and interactions ofsocial relations and social processes (Massey, 1991). Migration is a pro-cess which is underpinned by a host of social relations: class, ethnicity,gender and kinship. The market is a locality constructed out of variousinteractions between these social relations, the nature of which couldbe collaborative, conflictual or contesting. Clearly, as has been demon-strated by authors writing on class and ethnicity, citizenship is seriouslyquestioned in the context of ethnic tensions.5 But what is importantin the context of relatively recent migration of north and north-easternIndian workers to Kerala is that the lack of permanent settlement hasthwarted community formation and the processes of making a claim,which could lead to acquisition of citizenship.

Social protection for workers has been an important element of thedevelopment experience of Kerala and is very well known in devel-opment literature. Policy-making was the result of political action byworkers (Duvvury and Sabu, 1997), and trade union membership hasbecome the primary criterion for occupational identity for the purposesof eligibility in social security arrangements, such as receiving statepension and membership in welfare fund schemes (John, 2004). How-ever, such social protection has not been extended to migrant workers.

176 Case Studies

For example, the construction workers’ welfare board, which operatesthrough trade unions, does not consider migrant construction work-ers in its various schemes for social protection. This lack of inclusionbecomes institutionalized as only unionized workers are considered forwelfare benefits.

Social protection policy thus has to be evaluated, keeping in mind thehistorical context and the political actions and struggles that shapedit. The highly influential mainstream trade unions in Kerala have notmade any attempt to unionize migrant workers or to critically examinethe implications of migration for local labour. Union membership alsothrows up interesting questions on the formation of class-consciousnessin Kerala, which perhaps recognizes only a Malayali as a thozhilali(worker). The ethnic bias of welfare systems comes to the forefront whenwelfare systems come in contact with migrant populations. The studyof the Perumbavoor Sunday market also demonstrates that social pro-tection does not just relate to policy but is continuously claimed andaccessed through contestations. In the absence of such struggles, socialprotection can easily bypass migrants, even in the presence of the mostprogressive policies and welfare systems in practice.

Migration in Perumbavoor, especially its unsettled nature, preventsthe development of a rights-based trade union. This unsettledness is notmerely about an all too easily attributed informality, , but is embed-ded in the varied and multiple relationships forged in the market. Thetrade unions of Malayali workers constitute ‘the local’, which is at timeshostile and discriminatory. The working class is a category which isbroken up at the local level when it comes into contact with eth-nicity, gender or caste. The ability of migrants to negotiate the localis seriously undermined by local power structures and is affected bythe intersection of different identities and other social divisions withinclass. These factors are also crucial in attaining social protection andwelfare.

A visit to the district labour office in Ernakulam reveals that from2006 to 2009, only 8257 workers figure in the archives of registrationunder the Inter-state Migrant Workmen (Regulation of Employment andConditions of Service) Act (1979), which requires contractors to obtainrecruitment licenses specifying the number of workers to be employedfrom t he District Labour Office. The labour officers admit that this num-ber is only a fraction of the actual number of workers who came to workin the district. Looking closely through the archives of the labour depart-ment, I found that only some of the contractors who recruited migrantworkers for factories and establishments have been registered, whereas


the construction sector seems to have been totally outside the purviewof the act until 2007.6

Two construction workers from Orissa died as the dilapidated buildingin which they lived collapsed in September 2007. The TUCI marchedto the labour office with the dead bodies of the workers demandingcompensation. It was a moment in public when migrants came outand participated in confrontational political activity. The constructioncompany paid Rs. 360,000 as compensation to the family membersof the workers after negotiations with the labour department and thetrade union. The accidents in construction sites which happened beforethis were suppressed by companies, either by paying meagre compensa-tion or by using violence. After this event the labour department wokeup and started registering contractors in the construction sector (seeTable 6.1).

The district labour officer explained that the Inter-state Migrant Work-men Act stipulates that the worker shall be under a contractor. However,according to him, the factories recruit workers directly. He said thatthese workers come ‘independently’ with their relatives. Since there isno contractor involved, they are not covered under the act. Though fac-tories mushroomed in different parts of the district since 2004, theywere not registered with the labour department. In fact, the namesof factories that employed migrants that I visited in Perumbavoor didnot figure in the archives. The definition of contractor by the labour

Table 6.1 Number of contractors registered in different sectorsunder Inter-state Migrant Workmen Act, Ernakulam district

Year Factories and establishments Construction

1998 10 21999 8 12000 8 02001 5 12002 3 −2003 2 −2004 4 12005 6 22006 4 −2007 7 262008 9 902009 7 58

Source: Data collected by author from the District Labour Office, Ernakulam.

178 Case Studies

department did not include small time contractors or workers whobrought in relatives to these factories. The Inter-state Migrant Work-men Act, 1979, is an important piece of legislation to protect migrantworkers. It stipulates:

• Provision for payment of a displacement allowance equivalent to 50per cent of monthly wages or Rs. 75/- whichever is higher.

• Provision for payment of a journey allowance including payment ofwages during the period of a journey

• Provision for suitable residential accommodation, medical facilitiesand protective clothing as prescribed.

However, the workers have to be under a registered contractor to beeligible for these benefits. Due to a narrow definition of the act andlack of understanding of the role of the contractor/agent, the act doesnot seem to have any teeth. This brings us back to the questions ofinformality and citizenship.

These workers are clearly outside the gaze of the state in terms ofaccess to social protection. It is in the interest of capital to keep theworkers outside the realm of the legal and that of the state. However,registration of contractors alone does not ensure that the workers underthem get the above-mentioned benefits. Moreover, workers continu-ously come and go and contractors or agents bring in more workers asand when the need arises.

The contractors need to spend Rs. 1000 per worker to get them-selves registered. This is a caution deposit, which is refunded whenthe registration ends. Most of the time, the number of workers theyreport to the labour department are far fewer than the actual num-bers. Inspections are conducted by the labour department regardingworking and living conditions. Since most of the workers in factoriesare considered outside the purview of the act because they are notrecruited through licensed contractors, these inspections do not ensurethat the workers get the above-mentioned allowances and benefits. Asthe registration of contractors in the labour department represents onlya fraction of the migrant population in the district, it would be easy tomistake this as a question of lack of implementation. However, a closerlook reveals that it is again a question of the citizenship of migrantworkers.

The nominal citizenship of these workers does not translate into thepolitical agency necessary for substantive citizenship. Just as local powerstructures erode the migrants’ citizenship, being outside the state’s gaze


can also lead to such erosion, for being enumerated or counted is alsopart of the complexities of citizenship. Here, we are forced to complicateour ideas about the role of the state as an actor which shapes migrant’slives.

6. Conclusion: are limits to politics also limits to socialprotection?

The above discussion points to the limits and possibilities of politicsfor migrant workers. The market is a site where negotiations for socialprotection in the form of kinship and friendship networks operateto facilitate money lending, saving and exchange of information onemployment opportunities. It also acts as a space where attempts aremade to organize workers by trade unions, and also as one of protestagainst state oppression.

The attainment of social protection in the case of migrant workers inKochi is all the more complicated due to the interplay of state powerand surveillance through the issue of police clearance cards, which isthought to be a prerequisite for social protection by formalizing theexistence and identity of the migrant and the desirability of his/herpresence. A dichotomous understanding of social protection as eitherformal or informal misses these contestations and conflicts around citi-zenship. The local can become an important locus in these struggles forsocial protection and citizenship, even though it is fraught with threatsto citizenship. That the migrant perceives himself/herself as transienthas crucial implications for citizenship and social protection. The unset-tled migrant develops a stake in the local that is different from that of asettled migrant.

A possible way to think about social protection would be to prob-lematize the ‘social’ as the terrain of contestations and conflicts throughwhich citizenship and rights are negotiated and established. However,to understand the social in social protection, it is important to thinkbeyond policy-making as mere law making and implementation or asbenevolent civil society initiatives. It is important to recognize the socialfield in which each stream of migration occurs, including relationshipsin recruitment, settlement and work. Such a reframing of the idea ofsocial protection can reflectively assist policy-making by understandingmobility for work as one of the contemporary human conditions, andby emphasizing that social protection also might have to become asportable as capital and labour have turned out to be. However, portablesocial protection is usually enabled and activated through elaborate

180 Case Studies

identification and tracking systems which are potentially oppressive andexclusionary and can restrict mobility of workers even though they arenot crossing any national border.

The question of citizenship and portable social protection raises ques-tions about the limits and possibilities of politics for migrant workers.As elaborated in this chapter, settlement is an important element inshaping the nature of migrants’ political engagement. Many times,the precariousness of short duration migration as a livelihood strategyis made even more so by state oppression. These significant limita-tions to politics can haunt social protection policy by destabilizing andemptying the constituency it attempts to target.

Acknowledgements

This chapter draws on research carried out for my doctoral disserta-tion at Center for Development Studies, Thiruvananthapuram. I thankDr K. N. Harilal, Dr Praveena Kodoth, Dr Ben Rogaly and Zainab Bawafor comments and fruitful discussions on earlier drafts of the chapter.I also thank Sunitha A S for early inspiration and for alerting me tothe existence of the Sunday market as I was beginning my fieldwork inKochi. I am also grateful to Dr Rachel Sabates-Wheeler and Dr RayahFeldman for helpful questions and comments.

Notes

1. It is not clear why Malayalis do not engage in the manual work whichmigrants do. Different explanations have been put forward by the Malayaliinformants I have talked to. A factory owner who employs migrant workers inAluva industrial area said: ‘It is difficult to get labour here. Local labour willnot do this work, whatever wage you offer. Who will do manual work? Sup-pose you want to clean toilets?’ CITU Perumbavoor area committee secretaryBeeras, who has organized Malayali workers who do loading and unloadingwork in the area said that Malayalis would not be able to withstand the chem-ical odour in the factories even if paid high wages. More research is requiredto account for the absence or reduced number of Malayali workers in thesefactories and construction sites.

2. Interview with T. C. Subrahmanian, office bearer of TUCI dated 11 June 2008.3. Labour quarters are temporary accommodation for migrant workers that are

either rented houses which are typically old and dilapidated or tin shedspropped up in the vicinity of construction sites. In this case, it was adilapidated old building on rent that collapsed causing the accident.

4. The human development report (2005) published by the government ofKerala describes the headload worker in these words: ‘A phenomenon, whichhas been unique to Kerala, is the category of casual worker, who is referred to


as a “headload” worker, almost exclusively male. The Kerala Headload Work-ers’ Act, 1978 (Act 20 of 1980) defined a headload worker as “a person engageddirectly or through a contractor in or for an establishment whether for wagesor not for loading or carrying on head or person or in a trolley any articleor articles for wages” . . . What drew wide attention to these workers was theemergence of a strong trade union movement among them, which was usedto extract exorbitant rates for loading/unloading and the “attached” work-ers became powerful enough to procure a wage even without performing anywork’.

5. There have been no instances of large-scale ethnic tensions in the areabetween migrants and Malayalis. However, there are instances of Malayalisand police rounding up large number of workers when there are murders orthefts, and some protests from the migrants.

6. The archival research was carried out during August–October 2009. Data for2009 in Table 6.1 represents only units registered till the month of Septemberthat year.

References

Benjamin, S. and Bhuvaneswari, R. (2007) ‘Urban futures of poor groups inChennai and Bangalore: how these are shaped by the relationships betweenParastatals and local bodies’, in Jayal, N. G., Prakash, A. and Sharma, P. K. (eds)Local Governance in India: Decentralization and Beyond (Delhi: Oxford UniversityPress).

Breman, J. (1996) Footloose Labour: Working in India’s Informal Economy(Cambridge: Cambridge University Press).

Castles, S. (2010) ‘Understanding global migration: a social transformation per-spective’, Journal of Ethnic and Migration Studies, Volume 36, Issue 10, pages1565–1586.

Chakrabarty, D. (1989) Rethinking Working Class History: Bengal 1890–1940 (Delhi:Oxford University Press).

Chatterjee, P. (2004) Politics of the Governed: Reflections on Popular Politics in Mostof the World (Delhi: Permanent Black).

de Certeau, M. (1984) The Practice of Everyday Life (Berkeley, CA: University ofCalifornia Press).

De Haan, A. (1997) ‘Unsettled settlers: migrant workers and industrial capitalismin Calcuttta’, Modern Asian Studies, 31(4), 919–949.

Deshingkar, P. and Farrington J. (2009) Circular Migration and MultilocationalLivelihood strategies in Rural India (Delhi: Oxford University Press).

Duvvury, N. and Sabu M. George (1997) Social Security in the Informal Sector:A study of Labour Welfare Funds in Kerala (Trivandrum: Centre for Developmentof Imaging Technology).

Gieryn, T. F. (2000) ‘A space for place in sociology’, Annual Review of Sociology, 26,463–496.

Government of Kerala. 2006. Human Development Report 2005.(Thiruvananthapuram: Centre for Development Studies).

John, C. P. (2004) ‘Social Security and Labour Welfare with Special Ref-erence to Construction Workers in Kerala’, Discussion paper No. 65,Thiruvananthapuram, Centre for Development Studies.

182 Case Studies

Levitt, P. and Glick Schiller, N. (2007) ‘Conceptualising simultaneity:A transnational social field perspective on society’, in Sahoo, A. K. andMaharaj, B. (eds) Sociology of Diaspora: A Reader (Jaipur: Rawat Publications).

Massey, D. (1991) ‘The political place of locality studies’, Environment & PlanningA., 23, 277–281.

Misztal, B. A. (2000) Informality: Social Theory and Contemporary Practice (London:Routledge).

Papastergiadis, N. (2000) The Turbulence of Migration: Globalisation, Deterri-torialisation and Hybridity (Cambridge: Polity Press).

Papola, T. S. (1980) ‘Informal sector: concept and policy’, Economic and PoliticalWeekly, 3 May, 817–824.

Portes, A. (1996) ‘The informal economy’, in Pozo, S. (ed.) Exploring the Under-ground Economy (Kalamazoo, MI: W. E. Upjohn Institute for EmploymentResearch).

Prasad-Aleyamma, Mythri (2009) ‘Spatial strategies and Sojourner workers: map-ping roadwork in Kerala’, in Abrar, C.R. and Seeley J. (eds) Social Protectionand Livelihoods: Marginalised Migrants of South Asia (Dhaka: University PressLimited).

Rogaly, B. (2009) ‘Spaces of work and everyday life: labour geographies and theagency of unorganised temporary migrant workers’, Geography Compass, 3(6),1975–1987.

Seligmann, L. J. (2004) Peruvian Street Lives: Culture, Power, and Economy amongMarket Women of Cuzco (Champaign, IL: University of Illinois Press).

7Crossing Internal Boundaries:Political and Physical Access to thePublic Distribution System in IndiaIan MacAuslan

1. Introduction

Although migration is a social protection strategy that allows actorsto diversify risk and escape hazards (Stark, 1991), migrants also havemigration-related vulnerabilities that imply migrant-related needs forsocial protection (Rafique and Rogaly, 2003; Sabates-Wheeler and Waite,2003). As many chapters in this book demonstrate, moreover, migrantsoften face disadvantages in accessing social protection from govern-ments, non-governmental organizations, markets or individuals. Veryoften, therefore, migration makes people more vulnerable and less wellprotected.

Until recently, migrants’ vulnerability and lack of access to socialprotection schemes had not received much policy or academicattention.1 While there has been a welcome increase in the attentiongiven to this topic, most analyses of social provisioning for migrantsfocus on policy design and the legal provisions for migrants’ accessto formal social protection schemes, without examining actual prac-tice. For instance, as discussed in Chapter 4, Holzmann and colleagues(2005) focus solely on laws regarding migrant access, rather than onactual access that reflects both laws and deviations from them. How-ever, broader analyses of social provisioning make clear that, while legalprovisions and policy design certainly influence who receives benefitsfrom social protection schemes, who actually secures benefits or not(physical access) is the result of complex interactions between serviceproviders and service (non-)receivers (Schaffer, 1973; Lipsky, 1980; Sen,2000; Keeley and Scoones, 2003; Sabates-Wheeler and MacAuslan, 2007;

183

184 Case Studies

MacAuslan and Sabates-Wheeler, this volume, Chapter 2). Furthermore,service provision is dynamic: legal provisions and policy change, oftenas a result of negotiations by (non-)receivers for changes that improvetheir ability to access services (political access). These negotiations oftentake the form of ‘voice’. Schaffer and Lamb (1974, p. 79) argue that‘applicants themselves will be forced to employ voice so as to handletheir opportunities, to defend themselves, to succeed, to refute stigma,etc., at all points of the access experience’. An unanswered research ques-tion is implied: do migrants have better or worse physical and politicalaccess to social protection than non-migrants?

There is, of course, no simple answer to this question, because thereare many different contexts in which we should ask it. There are manypossible interactions between different types of social protection mea-sures (informal, formal, employment-based, citizenship-based and soon) and different types of migration (international, internal, undocu-mented, legal and so on), and each interaction has a different set ofcomplexities around legal provisions and actual access. For instance,internal migrants may be eligible for formal social protection schemesbut international migrants may not.

This chapter analyses the physical and political access of internalmigrants to formal social protection by offering a case study of Indianmigrants’ access to the government of India’s Public Distribution Sys-tem (PDS), a programme distributing subsidized commodities to allIndian households. Using fieldwork conducted in Delhi in 2007,2 it goesbeyond an analysis of the legal provisions and policy design of the PDSto explore the actual access of migrants and their ability to change thesystem to improve their access. As this chapter examines interactionsbetween internal migrants and a formal social protection programme, itcomplements other chapters of this volume that examine internationalmigrants’ access to formal and informal social protection (Avato; Olivier;Bragg and Feldman; Reitzes and Sabates-Wheeler) and internal migrants’access to informal social protection programmes (Prasad-Aleyamma).

Before turning to the case study, three conceptual points introducethe approach of this chapter in relation to the present volume. First isthe Schafferian access framework (of the gate, line and counter form-ing a queue) for examining migrants’ access to social protection, whichhas been set out elsewhere in this volume (see Chapters 1 and 3) andwill not be repeated in depth here, other than to outline some of thebasic terms used throughout the chapter. Table 7.1 presents a matrix ofthe categories of disadvantage and different determinants of vulnerabil-ity. Migrant-specific disadvantage applies by virtue of having migrated.

185

Table 7.1 Categories of migrants’ disadvantage

Determinant ofvulnerability

Examples ofmanifestations

Intensified (e.g.low-income actors)

Bureaucraticallyimposed

Over-representation

Migrant-specific

Spatial/Environmental Unfamiliarity withsurroundings(knowing which queueto join)

Lack of knowledge (ofe.g. where the gate is)

Public informationin local language(can’t understandaccess rules)

Health risks associatedwith informalsettlements (difficult tospend time in line)

Socio-political Lack ofrepresentation(illegal) (ineligible tojoin queue)

Uncertainty interactingwith government(difficult to engage ininstitutional complaintsmechanisms at counter)

Discrimination inaccess to services;‘citizens withoutrights’, (ineligible tojoin queue)

Lack of political accessfor slum dwellers(inability to express voicefor change to queue rules)

Socio-cultural Xenophobia(discrimination inqueue, at counter)

Social discriminationbased on ethnicity,language, illegal status(discrimination inqueues, at counter)

Additionalstigmatizingrequirements toaccess services(required to showadditional eligibilitydocuments at gate,counter)

Social perceptions of‘criminal poor’(additional scrutiny ofaccess documents atgate, counter)

Source: Adapted from Sabates-Wheeler and Waite (2003, p. 14).

186 Case Studies

Migrant-intensified disadvantage occurs when migration exacerbatesa disadvantaged caused by, for example, poverty. Bureaucraticallyimposed disadvantage refers to officials’ attitudes and ideas that exac-erbate customary discrimination. Migrants suffer ‘over-representation’disadvantage when they are over-represented in a disadvantaged group.

Second, we are concerned here with individuals and householdsthat move within India (internal migrants) rather than internationalmigrants. Other chapters in this volume emphasize the vulnerabilitiesthat international migrants face as they travel between social protectionand welfare regimes with different legal bases and citizenship rights.While internal migrants in theory retain their citizenship rights tonational social protection programmes, they often face parallel prob-lems to international migrants in accessing social protection as theymove between different administrative structures.

Third, while Chapter 2 presented a framework for studying accessto market and non-market distribution systems, this chapter studiesa social protection measure that transfers goods through public non-market distribution systems. All non-market distribution systems sharethe characteristic that they deliberately impose restrictions on thosewho can receive the good or service, usually in order to target thedistribution.

The next section describes the access structures of India’s PDS inDelhi, with particular reference to migrants, and the use of the PDS.The third section presents a deeper analysis of migrants’ political andphysical access to the PDS. The chapter closes with some brief policyrecommendations.

2. Access structures of the public distribution system

The PDS is a public non-market system of procuring, transporting, stor-ing and selling ‘essential’ goods, currently operated by central and stategovernments and the parastatal Food Corporation of India (FCI). It is thelargest formal social protection measure in India, for which India’s 1.1billion people are eligible as citizens. Over 300 million of these are inter-nal migrants (Census of India, 2001). India contains more income-poorindividuals than any other country, and some of the worst nutritionindicators in the world. The PDS is also the most expensive social pro-tection measure in India. The government of India’s expenditure on thePDS in 2003–2004 amounted to over 5 per cent of total central govern-ment expenditure and was over seven times greater than all other socialsecurity spending in India (Zhou and Wan, 2006).3 The analysis in this

Ian MacAuslan 187

chapter focuses only on the distribution of food grains (wheat and rice)4

in Delhi, a city containing at least 5.6 million reported migrants (40 percent of its total population).

The ‘targeted’ PDS offers social provisioning by selling 35 kilograms offood grains per month to households presenting a ration card and suffi-cient cash at the Fair Price Shop (specially licensed PDS outlet) to whichtheir ration card is registered. There are three types of ration card issuedto households on the basis of income poverty and social characteristics:Above Poverty Line (APL, or ‘non-poor’), Below Poverty Line (BPL, or‘poor’) and Antyodaya Anna Yojana (AAY, ‘very poor’).5 Although thegrain is the same, the price per kilogram differs by card type, increasingfor ‘poor’ and again for ‘non-poor’ households, although all prices arebelow the market price for similar quality grain. In terms of the accessframework in Chapter 2, households must complete two queues: one toobtain their ration card, and the second (at the Fair Price Shop) to obtaintheir food grains.

Central government administrative procedures designed to reducecorruption and increase efficiency reduce the amount available for eachcard. The central government allocates PDS wheat and rice for only 35per cent of non-poor cards in Delhi, because many non-poor cardhold-ers do not purchase their allotment, leading to reported diversions ofnon-poor grains. The central wheat and rice allocation for poor/verypoor cardholders is based on the 35 kilogram entitlement multiplied bythe number of ration cards issued by the state, or the (controversial)Planning Commission poverty estimates, whichever is lower.

Table 7.2 illustrates the discrepancies between the number of non-poor, poor and very poor households counted by the Planning

Table 7.2 PDS households in Delhi 2007, various estimates

Household estimate Non-poor Poor Very poor

Planning Commission – 252,200 156,800Government of Delhi ration card

figures2,411,514 378,507 62,012

Implied by 2007/2008 allocation offood grains to Delhi by the centralgovernment

807,929 294,571 46,262

Source: Planning Commission figures obtained from official documents, viewed 16 August2007. Government figures obtained from the Department of Food and Supplies, governmentof Delhi, 16 August 2007. Ration card figures given directly. Implied allocation computedfrom figures for the 2007/2008 allocation of food grains to Delhi by the central government.Implied households = annual grain allocation/(35 kilograms × 12 months).

188 Case Studies

Commission estimates, the number of ration cards issued to each groupby the government of Delhi, and the proposed 2007/2008 allocations bythe central government.

The central government’s low non-poor allocation means that theDelhi Food and Supplies Department limits distribution to 14 kilogramsper non-poor household, and shops are ordered to sell only 14 kilo-grams to each non-poor cardholder, though exceptions are made forslum dwellers with non-poor cards who allegedly receive 35 kilograms.The Delhi government allocates 35 kilograms per poor/very poor card-holder. Non-poor cardholders may therefore purchase 14 kilograms offood grains per month and poor/very poor cardholder may purchase35 kilograms of food grains per month, from the shop to which theirration card is assigned. The counter rules require the presentation of aration card to prove that the applicant is in one of the three categoriesof household, as well as the required cash.

PDS prices are fixed by the state government for ‘non-poor’ and ‘poor’households, and by the central government for ‘very poor’ households.Table 7.3 presents the shop prices charged in Delhi.

The ration card queue has more complicated rules. It is divided intoseveral sub-queues for each type of ration card, which are summarizedin Table 7.4. The general admission rule, according to the PDS (Control)Order, 2001, and the Delhi government, is that an eligible applicant is‘any individual who is resident of a State’ and is ‘permanently residingor intending so to do in . . . Delhi . . . a person can be deemed to be perma-nently residing in . . . Delhi for purpose of issue of supplies from publicdistribution system if he is ordinarily living in . . . Delhi and is not absentthere from for a period exceeding one month at a time’(Government ofDelhi, 2007).

In addition to the permanent residence criterion, Table 7.4 indicatesthat the ration card queue is not always open. Rather, applications forBPL, AAY and slum cards remain open only for temporary periods oruntil quotas have been filled. Households arriving in Delhi outside these

Table 7.3 Fair Price Shop prices, Delhi, 2007

Non-poor Poor Very poor

Wheat (Rs/kg) 6.8 4.65 2Rice (Rs/kg) 9 6.15 3

Source: http://www.delhigovt.nic.in/dept/food/fpds3.asp, accessedAugust 2007.

189

Table 7.4 Rules for the ration card queue, Delhi, 20076

Cardtype

Admissionrules

Applyanytime?

Counter rules Issuecost

APL,‘Non-poor’

Permanentresidence inDelhi

Yes Completedapplication form.Signature must beattested by a‘gazetted officer,area MP, MLA ormunicipalcouncillor.’1 Formavailable fromCircle Offices forRs. 0.50 or for freefrom internet.

3 attestedpassport sizephotos of headof family,attested.

Proof ofresidence.

For owner:registration deed,allotment letter,or house taxreceipt.

For tenant: NoObjectionCertificate (NOC)from landlord or2 witnesses ofcontinuousresidence.

Surrendercertificate forprevious rationcard or affidavitstating nopreviouscard, attestedby DistrictMagistrate, OathCommissioner, orNotary Public.

Free

BPL,‘Poor’

As APL; andtotal annualfamilyincome lessthan Rs.24,200.

No – closedsince 2002.

As APL, except:Form costs Rs. 1.Income certificate verified by the District Magistrate and Food Inspector.

Rs. 10

190

Table 7.4 (Continued)

Cardtype

Admissionrules

Applyanytime?

Counter rules Issuecost

AAY,‘Verypoor’

As BPL; andwidowed,disabled,elderly, orsingle earnerin largehousehold.

No – openuntil quotafilled.

As BPL, except:Require BPL card.Meet criteria specified by government and verified by Food Inspector.

Variable

Slum‘Non-poor’RationCard

Permanentresidence inslum inDelhi.Present onvoter listJanuary2007.

No – openJune–Sept2007 only.

Completedapplication form.Form collectionand submission atCircle Office 9.30am to 1.30 pmevery workingday.

2 photos, 1attested by areaMLA/councillor,or gazettedofficer.

Affidavit that noother rationcard, attestedby DistrictMagistrate, OathCommissioner, orNotary Public.

Photocopy ofvoter list, January2007. Residenceverified by FoodInspector.

Free

1Government of Delhi, n.d., Guidelines for applying for ration card. MP: Member of Parliament. MLA: Member of Legislative Assembly (State Parliament).

Ian MacAuslan 191

periods therefore find no queue to join. Counter rules involve a rangeof documentation verified by many different administrators and politi-cians, including an application form signed by ‘a gazetted officer, areaMP [Member of Parliament], MLA [Member of Legislative Assembly, stateparliament] or municipal councillor’ – an exacting requirement – and aproof of residence requiring a certificate from a landlord or two wit-nesses. Those wishing to apply for BPL or AAY cards must, in addition,present income certificates verified by the District Magistrate and FoodInspector. Aside from the other practical difficulties of obtaining, fill-ing out and submitting forms (discussed more below), the counter rules’reliance on the signatures and verifications of officials who may be busyor inaccessible presents a formidable barrier to many applicants. As dis-cussed in Chapter 2, those individuals and households most in need ofsocial protection are typically those least likely to have good access tothese powerful individuals. The formal requirement to obtain these sig-natures therefore intensifies inequality of access to social provisioningeven before applicants’ ability to obtain goods is considered.

How much is the PDS used in Delhi? Analysis of data from the 61stround of the National Sample Survey Organisation (NSSO, 2007)7 inurban Delhi (2004–2005) shows that if households purchased their full35 kilogram allocation of rice and wheat from the PDS, this would coverbetween 65 per cent and 100 per cent of most households’ consumptionof rice and wheat, and the lower prices would save households betweenRs. 250 and Rs. 300 per month – or between 60 per cent and 90 per centof cereal expenditure. However, data show that very few householdspurchase from the PDS, or purchase fully from the PDS, and thereforehouseholds actually save far less than they could if they purchased theirfull PDS allocation (even though they do seem to buy PDS rations atapproximately the official PDS prices).8

What is the reason for this low take-up? While some households donot purchase grain at all (market or PDS) because of taste or sufficiencyof production, and some households’ wealth and taste lead them tochoose never to purchase from the PDS, there remain many householdswho do prefer to purchase PDS grains (Department of AdministrativeReforms and Public Grievances, 1993; Parivartan, 2004; interviews). Weassume that, other things equal, poor households prefer to purchasefrom the PDS in order to save money (Kochar, 2005). The next sectionslook the fact that, in Delhi, most poor households are likely also to bemigrants, and explores why poor migrant households are not able topurchase PDS grains (physical access), and why they cannot negotiatebetter access to them (political access).

192 Case Studies

3. Migrants’ access to the Public Distribution System

Nearly half of Delhi’s population are migrants, and many of these arepoor, but there are no comprehensive studies that detail the migrationstatus of poor households in Delhi. Based on data from studies and inter-views analysed in MacAuslan 2007, Figure 7.1 indicates that, within thepopulation of Delhi, a large proportion are migrants, and a sizeable pro-portion of migrants live in slums, resettlement colonies (areas to whichthe government relocated slum dwellers when slums were classed asunauthorized and destroyed), construction sites, or are homeless.9 Byno means all of these migrant and slum-dwelling households are poor,however, but most poor households are migrants and live in slums (orresettlement colonies, construction sites, or are homeless).10

3.1. Migrant-related disadvantage in the PDS rules

As indicated above, the PDS rules do not explicitly exclude migrants.However, analysing the PDS access structure reveals four migrant-specific disadvantages. These constitute bureaucratically imposed formsof disadvantage on poor migrants, reflecting perceptions that poorhouseholds should not migrate and put pressure on Delhi’s publicservices.

First, households who migrate to slums are periodically rendered inel-igible for a ration card by the admission rules. Until 1990, admissionrules made slum residents ineligible for ration cards because they wereconsidered to be ‘trespassers on public land and in violation of thelaw’ (Department of Administrative Reforms, 1993, p. 301). The Delhi

Poor

Slum/resettlement,construction site,homeless

Migrants

Delhi

Migrant poor inslums, etc.

Figure 7.1 Venn diagram of relationship between migration, poverty and slum-dwelling in Delhi.

Ian MacAuslan 193

administration reversed this decision in 1990 with a one-time rationcard issue to slums, and then fully liberalized applications in February1992. However, a December 1998 cabinet order decreed that ration cardswould be granted only to slum residents whose names were on theOctober 1998 voters’ list or who applied by 31 March 1999.11 This deci-sion was in turn superseded in May 2007, when the cabinet permittedthe one-time issue (between May and September 2007) of non-poorration cards to slum dwellers registered on the slum’s January 2007voters’ list.12 Households migrating to slum clusters between 1998 and2007 were thus formally ineligible for any PDS benefits until May 2007.Households migrating to slums after 1998 are ineligible for queues for‘poor’ and ‘very poor’ cards, irrespective of income. Recent migrants,despite greater need, face greater disadvantage specifically because theyrecently migrated.

Second, the ‘permanent’ residence criterion creates migrant-specificdisadvantages for migrants living temporarily on construction sites,who are formally ineligible for ration cards. Since these workers arenot considered permanent residents of any part of Delhi (even if theystay within the Union Territory limits), they cannot apply for rationcards. Organizations working with these construction workers in Delhiconfirmed that none have ration cards or use the PDS.

Third, poor migrants have migrant-specific disadvantages because thefixed quotas for ‘poor’ and ‘very poor’ cards, and the related non-continuous application periods, exclude households arriving after thequotas are full. Households migrating to Delhi after the governmentof Delhi closed applications for ‘poor’ cards in 2002 are excluded fromapplying for ‘poor’ or ‘very poor’ cards irrespective of their income(since, in Delhi, only households with ‘poor’ cards can apply for ‘verypoor’ cards). Households excluded from ‘poor’ or ‘very poor’ cards losesubstantial subsidies. Government officials do not consider ‘non-poor’rations an important part of PDS. In practice, interviewees suggest thatgiven the cost and inconvenience of buying PDS rations (see below), it isbarely profitable to make use of the subsidy offered by ‘non-poor’ cardsand many do not use them: ‘we have a “non-poor” card but what is theuse when non-poor rations are expensive and often unavailable? If weget a “poor” card, then we can use the PDS.’13

Finally, all migrants face intensified disadvantages because the accessstructures require card holders to reapply for a ration card with everychange of residence across Fair Price Shop boundaries. The applicationprocess includes direct costs for forms, cards, certificates, photographsand attestations up to Rs. 70, including notary fees. Applicants must

194 Case Studies

also travel four times to government offices, incurring substantial travelcosts and opportunity costs (i.e. costs of opportunities missed). In addi-tion, applicants lose at least a month’s PDS ration during the process.Structural inequalities (see Chapter 2) in India’s labour markets andurban planning mean not only that poorer households migrate moreto informal settlements, but also that urban planning often involvesdemolishing these settlements and forcibly relocating residents to reset-tlement colonies. Since March 2007 a formal exception has been madefor resettled slum households, but prior to this order households wouldhave to apply for a change of residence in the ordinary way. Atul’sstory indicates what this involved, even for a relatively connected andresourceful household:14

Atul works for an organization helping to organize constructionworkers, and is literate. He has a ‘poor’ card, and is applying for a‘very poor’ card because he is the only earning member of his fam-ily, which contains three children. He is originally from Bihar, andmoved to Delhi in 1988, living in a slum. It took three years or so tounderstand the system of acquiring ration cards, and then he got onewhen the government allowed their issue around 1990. In 2002 heacquired a ‘poor’ card under the targeted PDS. In 2003 the slum wascleared and the household moved to Bawana, where they were allo-cated a 12 metre square plot because of their post-1990 ration card.Transferring the card was relatively easy and no extra-legal costs wereinvolved, although of course two months of ration had to be forgoneduring the process, and there were a significant number of journeysto offices, costing money and preventing work.

These migrant-related disadvantages are deeply entrenched in the accessstructure of the PDS, and in service provision more generally in India(Stephenson and Matthews, 2004; Smita, 2006) and abroad (Schlosberg,1999; Schwartz, 2001; Holzmann et al., 2005; Sabates-Wheeler, personalcommunication). These constitute bureaucratically imposed disadvan-tages. In India, most programmes and rights are conditional on proofof (usually legal) residence, since there is no widely used or avail-able national identity card. However, the PDS is particularly structuredagainst migrants. The residency criterion is required in order to deter-mine the allocation of food grains to states, districts and shops, and tominimize the possibility of households claiming subsidies on multipleration cards. This is not the case with, for instance, education, wheretangible resources are not transferred, although both education and the

Ian MacAuslan 195

PDS have fixed location distribution outlets (schools or shops). More-over, the central government allocates grain to states annually everyMarch: if households move between states after March, states have toreallocate their existing ration accordingly. States experiencing net in-migration (such as Delhi) will therefore have to reduce the per cardallocation.

3.2. Migrant-related disadvantage in PDS queues

Evidence from studies of the PDS suggests that the PDS does not func-tion according to its ideal type (Radhakrishna et al., 1997; Swaminathan,2000; Kochar, 2005; MacAuslan, 2007). The next three sections explainthis underperformance using evidence from fieldwork conducted in2007 and grounded in access theory, which predicts that administratorsand applicants ‘reconcile’ their different positions, resources, under-standings and interests by negotiating around access structures (Schafferand Wen-Hsien, 1975).

Fieldwork reveals significant deviations from the ideal type in boththe Fair Price Shop and ration card queues. Deviation in Fair Price Shopqueues was not found to produce significant migrant-related disadvan-tage in terms of their physical access, but being a migrant affectedapplicants’ ability to obtain political access – to negotiate improvementsin Fair Price Shop provision. The ration card queue displays migrant-related disadvantages far more clearly as they have to apply morefrequently, lack knowledge of local bureaucracies, and because rationcards also function as identity documents, which migrants are unwill-ing to risk losing. Migrants are less likely to exercise an effective politicalvoice to change the rules of the PDS because of their uncertainty, andperceived tenuous legal status and biases in rule-makers.

3.2.1. Migrant-related disadvantage in the Fair Price Shop queue

Fieldwork indicated significant deviations from the PDS rules in FairPrice Shop queues, but did not reveal much migrant-specific disad-vantage in these deviations, although there were various disadvantagesfaced by poor, illiterate and daily labourer households, among whommigrants are over-represented. However, migrant status did affect thechoice of strategy (acceptance (loyalty), exit, or protest (voice)) for politi-cal access when faced with these discriminations (see Chapter 2 for moredetails on these types of strategies).

In slums and resettlement colonies fieldwork revealed severalinstances of shop-level protest. This can take the form of individualprotests, attempting to use personal influence or authority to cajole or

196 Case Studies

compel the shopkeeper to improve the individual’s allotment (ratherthan improving the functioning of the system for everyone). Thisprotest requires knowledge and effective social networks, and risks alien-ating the shopkeeper and encouraging the manipulation of the rulesagainst the protester. More recent migrants, therefore, generally do notengage in this strategy.

Protests involving migrants were more usually and more successfullyundertaken by groups containing some knowledgeable individuals whofunction as ‘representatives’ (brokers with a better knowledge of therules), such as NGOs, ‘people’s movements’ or unions. Union actiontends to produce improvements for members only: non-members do notbenefit from the protest and, actually, given the shortage of rations, loseout; however, migrants are free to join. Temporary construction workermigrants are recruited to join the Construction Workers’ Welfare Board,but those living on construction sites are more difficult to unionize thanslum or resettlement residents by virtue of their impermanent status.People’s movements may secure benefits for those associated with them(and have no formal barriers to entry as unions do), but they work inquite localized ways to secure improvements, for example, to the func-tioning of a single shop. A common technique is to organize jan sunwais(public hearings) that function to inform applicants about their entitle-ments and ‘shame’ shopkeepers into providing them, and are open toall.15 There is some evidence that these protests improve the function-ing at the level of individual shops (Parivartan, 2004; SNS, 2006; Pande,2007), at least temporarily.

Field interviews provided evidence of local-level protest. The follow-ing was reported in a resettlement colony:

A group of us construction workers spontaneously organised – with-out outside help – to exert pressure on the shopkeeper to allo-cate rations according to the entitlement. We exerted this pressurethrough both crowd pressure and political intervention. Since wecame together as a group, we were able to appeal to political inter-ests to compel politicians to exert pressure on the shopkeeper – thispolitical intervention made the difference. However, this was donewithin the boundaries of our construction union, which had been setup with help from an NGO. We also don’t know whether the goodbehaviour will last. We don’t vote on party lines – whoever helps usget services gets our vote.

Migrants also benefited from direct complaints (outside the normalchannels of administrative complaint) by ‘representatives’ to the Food

Ian MacAuslan 197

and Supplies bureaucracy that affected the provision of rations. Forinstance, during field research members of an NGO (a ‘representative’in a slum in Delhi) complained about the late provision of rationsto the shop to the local government official, and then to the Assis-tant Commissioner. This organization has some notoriety in the FoodDepartment following high-profile meetings and interactions with offi-cials and Supreme Court Commissioners. Building on this, slum dwellersin the organization were able autonomously to complain to FoodDepartment officials, who expedited the provision of rations to the localFair Price Shops.

3.2.2. Migrant-related disadvantage in the ration card queue

Deviations from the ration card queue create migrant-intensified disad-vantages because of migrants’ more frequent applications. Additionalcosts are added to the application process because administrators (officestaff) do not function as their ideal type (see also Mooij, 1999 and2000). Their offices may be closed unpredictably and forms may beunavailable. Some administrators impose additional costs for issuing,attesting and accepting forms, and even for dealing with the applica-tion (Gonsalves et al., 2005, p. 131). Counter servers are often unawareof the types of document that count as proofs of residence, or whetherthese documents are required (Parivartan, 2004). For instance, adminis-trators often require a proof of residence for applications for the slumration card (as required for normal ration cards) when in fact only acopy of the electoral roll is formally required. This additional require-ment imposes disadvantage on (particularly recent) migrants who donot have documentation.

Many applicants, but particularly new arrivals (such as Atul, intro-duced above, when he arrived in 1988), lack basic knowledge of theprocess of applying for ration cards, the location of the offices, where toobtain the relevant forms and documents (and what these documentsare), of how to fill out the forms and so on (see also Schwartz, 2001and Stephenson and Matthews, 2004 for similar findings). These fea-tures suggest that the longer the stay, the more likely migrants are tohave access to the PDS. This is confirmed in a survey of 200 slum house-holds that recorded duration of migration and ration card possession(Edelman and Mitra, 2006, p. 32).

In slums and resettlement colonies for more permanent residents,however, applicants can induce administrators to manipulate the rulesto facilitate their access to ration cards. Interviews with organizationsand applicants suggest that this is not the norm, but it certainly occurs.Edelman and Mitra (2006, p. 11) find that some short-term migrants to

198 Case Studies

slums have ration cards, which they claim are due to some migrants’‘political contacts’. Fieldwork findings indicate three types of devia-tions that positively help applicants. First, many politicians attest andendorse ineligible applications in order to build constituencies. Second,‘Janus-faced’ (Long, 1989) administrators and politicians assist ineligi-ble applicants to whom they are socially tied. Third, administratorsaccept bribes to process ineligible applications. These advantages are notmigration-related, except that available evidence suggests that politicalcontact, voter rights and social ties increase with the duration of migra-tion (Stephenson and Matthews, 2004; Jha et al., 2005; Edelman andMitra, 2006), so newer arrivals may be less advantaged. Intervieweesdid not report any pattern to these ‘positive deviations’ in terms ofhouseholds being from particular states, religions and so on.

A crucial aspect for migrants regarding the complexity of the rationcard queue is that it offers a resource that is used to enter multiple otherqueues. Despite Supreme Court orders and government directives,16 theration card has a dual function as PDS access card and general identitydocument. Ration cards are closely linked to identity, legality and citi-zenship; so that ownership of a ration card confers legality that makeseviction complicated and requires the provision of services. Govern-ment officials recognize that government departments treat ration cardsas general identity documents, despite orders to the contrary. For exam-ple, only slum households with ration cards are entitled to resettlementplots when slums are destroyed. The government of Delhi thereforefaces conflicting incentives over issuing ration cards (see Lipsky, 1980;Schaffer, 1986), especially where there is any confusion over the legit-imacy of residence or if services are stretched (as is the case in muchof Delhi). In their analysis of the PDS in slums, the Department ofAdministrative Reforms and Public Grievances (1993) noted that thegovernment had not issued ration cards to slum dwellers because itwould legitimize their perceived dubious legal status.

While discussing previous and current issuances of slum ration cards,several past and present Food and Supplies officials pejoratively referredto what they see as incentives – such as being given ration cards and sub-sequently plots and other services – created for households to migrate toDelhi and encroach on land. In order to minimize incentives to migrateto Delhi, the issuance of non-poor cards to slum dwellers closed on31 August 2007, and is accompanied by specific instructions that thesecards are not valid for plot allocation purposes.

The dual function of ration cards underlies the reported decisionof many temporary migrant households to exit the Delhi ration card

Ian MacAuslan 199

queue, because queuing involves surrendering their previous rationcard. In a city where migrants’ legitimacy and legality are tenuous, manyare understandably reluctant to give up the card that confers a senseof legitimate identity (even if not as a resident of Delhi) and enablesmigrant households to retain links with their home village, to whichmany hope to – or actually do – return. This partially explains organi-zations’ reports that no construction site dwellers (temporary migrants)have or seek Delhi ration cards.

The bureaucratic impositions on migrant PDS applicants created bythe de facto identity of the ration card reflect a broader theme, run-ning through this book, of citizenship rights for migrants (Rogaly et al.,2001; Kabeer, 2002, Uvin, 2004; Smita, 2006). The 42.3 million inter-state migrants in India, and 5 million in Delhi, have constitutionalrights as Indian citizens that the government must protect, including,according to Supreme Court judgements, the right to food. In 2001,the Right to Food case was brought by human rights groups against thecentral and state governments of India for not protecting citizens fromhunger (see Gonsalves et al., 2005). As a result, the Supreme Court hasissued a set of orders on the PDS and the Commissioners have raisedthe issue of migrant access with the Delhi government (see below).17

The PDS raises legal issues around migrants because, although the PDSis a national entitlement, its access structures stipulate that eligibleapplicants are residents of Delhi, rather than India. This means thatinter-state migrants must be considered ‘permanent residents’ of theirnew state to qualify for protection, and that temporary migrants donot qualify (Rogaly et al., 2001). Right to Food advocates argue thatthis access structure is ‘patently illegal’, since transient migrant pop-ulations fall into ‘an entitlement black-hole’. This perceived illegalityforms the basis for Supreme Court advocacy around migrants and thePDS. The articulation of social protection schemes as citizenship rightscould, therefore, contribute to support for bureaucratic reorientations toinclude migrants.

3.3. Migrant-related voice to change the PDS rules – politicalaccess

The access approach indicates that applicants will seek interventionsat higher levels to the extent that they cannot resolve access problemsthrough voice exercised at a local level. Since, as shown in the sectionsabove, many migrants (especially recent migrants) are disadvantagedby the PDS access structures, what opportunities do migrants have to

200 Case Studies

negotiate changes in the rules? We explore four types of voice used inthese negotiations: the voice that arises when many people exit from aprogramme, formal administrative appeal, organized local protest andinformation gathering, and direct communication to rule-makers.

First, the access literature suggests that when many people exit a pro-gramme, the effect is a form of voice, as in a boycott (Schaffer andLamb, 1974). NSSO data and field research indicates that many house-holds exit from both the ration card queue and shop queue, procuringgrains from alternative market sources – which are often (ironically)more reliable in terms of grain availability because Fair Price Shops are,in practice, not always well supplied with food grains or sell them infor-mally. The total number of migrant households who exit is thereforeprobably large, although this has not, in Delhi, aggregated to voice inthe sense of a boycott or protest. Exits from shop queues (by migrantsor anyone else) rarely translate into voice since shopkeepers have incen-tives not to report this and the checking system is not very rigorous(so no one notices the exits and no message is conveyed). Similarly,temporary construction workers’ exits from ration card queues do nottranslate into voice because these populations are normally not regis-tered as living in Delhi (due to their transience), so their exclusion fromration cards is not noted. There are no reliable data to check Food andSupplies officers’ contention that the number of temporary construc-tion workers is small, and therefore that full allocations can be foundfor them within the existing allocations from the central government.Some government officials argue that the formal exclusion of tempo-rary construction workers is not a problem of sufficient magnitude andtherefore not worth addressing.

Second, applicants use administratively sanctioned appeal structures.The PDS (Control) Order, 2001, for instance, provides for VigilanceCommittees at the shop-level. However, these do not always functionparticularly effectively (Parivartan, 2004), and migrants with dubiouslegal status or experience of harassment or discrimination are less likelyto resort to administrative sanctions, particularly where they involvecontact with the police.18 Much of the problem derives from a compli-cated relationship between criminal and administrative law in the PDS(see Mooij, 1999). In Delhi, the police argue that they cannot accepta First Information Report (the document required to begin a crimi-nal investigation) about diversion of grains from citizens (Parivartan,2004). They will only accept a Report from officials in the Food and CivilSupplies Department, who prefer to deal with misdemeanours adminis-tratively, often because they may be criminally implicated (Mooij, 1999;

Ian MacAuslan 201

Parivartan, 2004). This reduces shopkeepers’ incentives to abide by rules,and applicants – particularly migrants with dubious legal status – can dolittle to enforce complaints.

Third, voice for rule-change emerges from local protest and informa-tion gathering. In Delhi, however, there is little evidence that shop-levelprotests aggregate to overall improvements in the governance of shops(MacAuslan, 2007). As with aggregate exit, these protests can reveal torule-makers and brokers the extent of the malfunctioning of shops. Forinstance, an NGO assisted individuals to submit, simultaneously, 100applications for information on ration cards and ration distributionunder the recently passed Right to Information [RTI] Act,19 bringingmalfunctioning to the attention of Food and Supplies officials. However,this mode of voice does not create any accountability in the govern-ment: the RTI Act requires that they provide information, but no more.Similarly, evidence of poor functioning is communicated to rule-makersthrough the media, journals or invitations to hearings, but includes nocompulsion to act. Migrants are included in these protests and repre-sentations by virtue of their over-representation in poor or slum groups,but rarely specifically because they are migrants.20

Fourth, there is direct voice for rule-change. The multiplicity andeminence of rule-making organizations in the PDS (state and centralgovernments, including the Planning Commission), and the require-ments of knowledge, time and cost, complicate the exercise of directvoice by poor migrants, unless they have ‘fixers’ (brokers who canchange the rules). However, although in interviews many potentialfixers recognize that poor migrant populations both need social protec-tion from the PDS and are excluded from it, there are three constraintsto voice exercised by these fixers for migrants’ access to PDS. First, manypotential fixers do not work in the PDS, because they consider its rulestoo degraded, and prefer to focus on other, newer programmes, such asthe National Rural Employment Guarantee Act, where they perceive thepotential for positive change to be greater. Second, the variety of accessproblems in the PDS means that fixers have to be quite organized andmethodical in addressing them, lest they alienate rule-makers. This isproblematic because, third, fixers disagree about the ideal design of thePDS (whether it should be universal or targeted, for instance). Most orga-nizations working on the PDS are trying to save the system as a whole,rather than improve access for one excluded group. These limitationsmean that fixers are not generally advocating specifically for improve-ments to migrants’ access to the PDS, although general improvementswill improve migrants’ access.21

202 Case Studies

Only one fixer, the Supreme Court Commissioners, has made specificrepresentations to the central and Delhi governments about migrants’access to the PDS. The Right to Food case was brought against thecentral and state governments of India for depriving citizens of theirconstitutional right to food (see Gonsalves et al., 2005). As a result, theSupreme Court has issued a set of orders on the PDS and the Commis-sioners have raised the issue of migrant access (Gonsalves et al., 2005;Supreme Court Commissioners, 2006). For instance, the Commissionersrecommended that ‘migrant workers shall be entitled to draw grain ontheir existing [‘poor’] cards notwithstanding the fact they have movedoutside the area where they reside’ (Gonsalves et al., 2005, p. 134).The Fourth Report highlighted urban destitution and homelessness, andnoted that ‘migrant labourers . . . are cut off from the public distributionsystem as their ration cards are attached to place of residence . . . theyshould have “mobile ration cards” that can be used in the destinationarea’ (Gonsalves et al., 2005, p. 211).

Although the government of Delhi acted in 2007 both to open slumcard applications and to approach the Supreme Court about survey-ing homeless people (including construction site dwellers) in Delhi forration cards, it seems probable that it did so partly for reasons uncon-nected to the Supreme Court, such as longer term plans or officials’independent recognition that there was a problem. This interpretationis supported by interviews with government officials, and the surprisewith which fixers greeted both decisions.

Moreover, it is clear that Supreme Court fixing is not a sufficientcondition for rule changes. For instance, although the Supreme Courtordered on 20 April 2004 that the criteria for ‘very poor’ cards shouldnot include possession of a ‘poor’ card, because the ‘poor’ surveys wereflawed (Gonsalves et al., 2005, p. 55), the Delhi Food and SuppliesDepartment maintains this requirement. Moreover, the government’s2007 decision is only a temporary and partial implementation of theCommissioners’ recommendations. And it has not adopted the pro-posal for mobile ration cards, despite initiatives in other states to allowtemporary migrant workers to access PDS rations at their migrationdestination.

Interviewees frequently referred to politicians as key brokers inimproved distribution. They intervene in the queues for rations andration cards or exert pressure for rule manipulation to improve accessfor their constituents or for applicants with connections, money, orwho form groups. This may disadvantage recent migrants whose shortstay means they lack connections to politicians and seek help insteadfrom informal ‘big men’ (Harriss, 2005; Jha et al., 2005; Edelman and

Ian MacAuslan 203

Mitra, 2006; MacAuslan, 2007). This kind of intervention is limited tothe individual level and scarcely manifests itself in pressure for improve-ments to the rules, while the informal channels of communication thatarise to lubricate migrants’ access to services (including ration cards) arenot accountable, and so are potentially fragile.

The overwhelming model of politics around the PDS is of politicalpatronage to the electorate, rather than fighting for the systemic rightsof constituents. However, since short-term migrants are less likely to bevoters (Deshingkar and Start, 2003), their gains from this patronage aremore limited.

Interviewees expressed strong dissatisfaction with politicians’ integrityin terms of group improvement and perceived a disjunction betweenpoliticians’ expansive pre-election promises and post-election inactiv-ity. As Edelman and Mitra (2006, p. 32) point out, ‘politics in slums isnot rocket science. Slum dwellers want ration cards, water, and a guaran-tee that their shacks won’t be demolished’. Field interviews confirm thatslum and resettlement colony dwellers do not vote on any other basisthan the provision of amenities. Almost all Members of Parliament inDelhi and Members of the Delhi Legislative Assembly include promisesabout improving services to slum dwellers in their election literature(Jha et al., 2005). But interviewees were pessimistic about the likelihoodof politicians delivering on their promises.

Table 7.5 summarizes these findings by presenting migrant-relateddisadvantages of different types (see Table 7.1) in the shop and rationcard queues, and in terms of voice. It also adds some disadvantages

Table 7.5 Summary of migrant-related disadvantage in the PDS in Delhi

Shop Ration Card Voice

Specific Post-1998 slumarrivals; temporaryresidents; post-2002arrivals excludedfrom ‘poor’ queue

Lower likelihoodof politicalcontact for recentarrivals

Intensified Recent migrantshave fragilenetworks and lessconfidence toovercomediscriminationagainstpoverty/illiteracy

More frequentapplications (regularand deviation costs).Demolition ofsettlements/informal labourmarkets

Fear of policelimits sanctionedcomplaints

204 Case Studies

Table 7.5 (Continued)

Shop Ration Card Voice

Bureaucratic Fixed location ofshops; stablepopulation basis ofallocation system

Identity function(migrants lesswilling to apply;officials lesswilling to grant);within-state rightsonly

Limitedperception orminimization ofmigrant problem

Over-represented

Shop deviations inopening time andprice (illiteratesand daily workers);reduced socialprotection (poor)

Slum dwellers onlyreceive ‘non-poor’card

Exercise localvoice throughslumrepresentativesand SupremeCourt

not discussed above (see MacAuslan, 2007, for more details), such aspoor networks or lack of confidence that cause problems in the shopqueue for low-income households that are intensified by them havingmigrated (as migrants have even poorer networks).

4. Conclusions and discussion

This case study has shown that low-income internal migrants suffer par-ticular disadvantages in accessing social protection in Delhi. Structuralinequalities that contribute to the insecurity of their housing (in infor-mal settlements) and jobs (in the informal economy), and push themtowards livelihoods and lifestyles that either require migration or makemigration a very attractive strategy, also contribute to poor access tosocial protection. Despite retaining citizenship rights to the PDS, and sobeing eligible for the PDS according to the Constitution and law, internalmigrants do not in practice receive full social protection from the PDS.Focusing on the PDS access structures in detail, rather than only the PDSpolicy documents, shows clearly how they disadvantage migrants in var-ious ways. In order to secure physical and political access to the PDS,applicants require documentation, time and knowledge that migrantsdo not have, and contacts that they cannot hope to make.

What can be done? ‘Migrant-aware’ research, planning and trainingis required. Changing the PDS to reduce migrant disadvantage is clearlydifficult: its allocation systems cater to stable populations and it does

Ian MacAuslan 205

not distribute a pure public good, so some eligibility criteria are nec-essary. Yet current policies are designed to discourage, not facilitate,the migration of low-income households. This reflects negative percep-tions of migrants, and their exclusion from rights. Establishing thatinternal migrants retain rights to social protection would undoubtedlyreinforce calls to reduce their formal exclusion, as in the PDS. However,rule-makers must still design access structures to implement mobilerights (to check claimants are eligible). The critical task, therefore, isto train, assist and inform rule-makers to become ‘migrant-aware’, toreduce as much as possible the disadvantages imposed on vulnerablemigrants by access structures.

Four specific policy changes are also suggested by these findings. First,it appears that ineligibility on the basis of slum dwelling or tempo-rary residence excludes a large number of households most in needof social protection from the PDS. The government should issue rationcards to these households and publicize the fact that ration cards do notconstitute identity documents. It would be useful, to facilitate this pro-cess, to expedite the distribution of a separate single national identitydocument, such as a passport.22

Second, ration cards need to be made mobile within and betweenstates, such that they can be used at any Fair Price Shop. Other states,such as Gujarat and Rajasthan, are experimenting with ‘roaming’ rationcards for intra-state mobile populations, and Orissa and Madhya Pradeshhad considered an agreement to provide rations for inter-state mobilepopulations. These initiatives could be reproduced in the NationalCapital Region of Delhi, comprising three states.

Third, in order to ameliorate the problems involved in allocatingfood grains to mobile populations, various methods could be piloted.Computerization of ration cards and allocations would probably assistthe process, and this is underway in Delhi. Alternative methods ofprovisioning for mobile populations could be experimented with. Forinstance, mobile populations could be given cash transfers or foodstamps that allow holders to purchase market grains rather thanattempting to redistribute PDS grains to their new locations. This wouldretain the expenditure subsidy social protection function, but not theexposure reduction function. This should not be seen, however, as anendorsement of suggestions for a replacement of the PDS with thesesystems, which clearly have their own problems (see, for example,Swaminathan, 2000). However, for the subset of the population thatis regularly mobile and has particularly poor (or no) access to the PDS,they may be very useful.

206 Case Studies

Fourth, the current system of ‘poor’ quotas formally excludes house-holds that move after the quotas are met. In this sense, the pre-1997‘universal’ system disadvantaged migrants less, and a return to a sin-gle ration card system would tend to equalize their access to rations,other things being equal. It is certainly true, moreover, that the ‘poor’and ‘very poor’ queues involve significant deviations from the rules,such that new programmes, such as the National Rural EmploymentGuarantee Act, do not use them. If the current ‘targeted’ system wereto be retained in PDS, special provisions could be made for states,such as Delhi, with high in-migration, so that their ‘poor’ and ‘verypoor’ quotas could expand between the surveys conducted every fiveyears by the Planning Commission, and accommodate new low-incomemigrants.

Notes

1. Some exceptions: Schlosberg (1999) on migrants’ access to health in the US;Stephenson and Matthews (2004) and Smita (2006) on migrants’ access tohealth care and education in India; and Sabates-Wheeler and MacAuslan(2007) on migrants’ access to social protection.

2. Interviews with PDS users in two slum areas of Delhi, including governmentofficials, activists, NGO workers and union organizers. Individual interviewsare not cited (see MacAuslan (2007) for more details). Although changes havebeen made to the system of identity cards and to the PDS since fieldwork wasconducted, the analysis here remains valid, and many problems identifiedhere remain.

3. Prior to the enactment of the NREGA.4. It also distributes kerosene and sugar, through slightly different access

structures.5. These epithets should not imply that households holding ‘non-poor’ cards

are necessarily not poor.6. While the rules for admission have evolved slightly since this fieldwork was

undertaken, with further AAY applications in 2008, and the governmentaccepting community-based verification of residence under the Delhi gov-ernment’s Mission Convergence architecture, the basic queue rules remainas presented here.

7. This is an annual and nationally representative socio-economic survey.8. See MacAuslan (2007) for further details.9. In this chapter, any actor who moves their usual place of residence across a

Fair Price Shop administrative boundary is a migrant, irrespective of registra-tion and intention to return. ‘Usual’ residence refers to the residence mostfrequently used in the last 30 days.

10. See MacAuslan (2007) for further details.11. Cabinet Decision No. 396, 28/12/998.12. Cabinet Decision No. 1203, 08/05/2007.13. Interview, slum areas, Delhi, August 2007.

Ian MacAuslan 207

14. Names and places have been changed.15. See also Parivartan (2004) and Pande (2007).16. PDS (Control) Order, 2001, Annexure 2(9) states: ‘ration cards shall not be

used as documents of identity’.17. Gonsalves et al. (2005); Supreme Court Commissioners (2006).18. See also Schwartz (2001).19. See ‘Slum Dwellers file RTI plea over ration’, The Hindu, 6 August 2007.20. Exceptions include Rogaly et al. (2001), Deshingkar (2006) and Smita (2006).21. As of August 2009, this is changing as the Supreme Court Commissioners’

Office is beginning a survey on migrant access to the PDS.22. As of December 2010, there are debates and the beginnings of policy on the

Unique Identity document in India.

References

Census of India (2001) ‘Migration data, data highlights I’, Census of India, Delhi.Department of Administrative Reforms and Public Grievances (1993) ‘Public dis-

tribution system: access of the poor – a case study of a jhuggi jhompri cluster’,Management in Government, January/March, 295–317.

Deshingkar, P. (2006) ‘Internal Migration, Poverty and Development in Asia’,paper prepared for the Asia 2015 Conference, IDS and ODI, London.

Deshingkar, P. and Start, D. (2003) ‘Seasonal Migration for Livelihoods inIndia: Coping, Accumulation and Exclusion’, ODI Working Paper 220 (London:Overseas Development Institute).

Edelman, B. and Mitra, A. (2006) ‘Slum dwellers’ access to basic amenities: therole of political contact, its determinants, and adverse effects’, RURDS, 18(1),25–40.

Gonsalves, C., Ramesh Kumar, P. and Srivastava, A. (eds) (2005) Right to Food(Delhi: Human Rights Law Network).

Government of Delhi (2007) Department of Food Supplies, Frequently Asked Ques-tions, Ration card, available at: http://www.delhigovt.nic.in/dept/food/faq1.asp(accessed 11 May 2007).

Harriss, J. (2005) ‘Political participation, representation, and the urban poor:findings from research in Delhi’, Economic and Political Weekly, 12 March,1041–1054.

Holzmann, R., Koettl, J. and Chernetsky, T. (2005) ‘Portability Regimes of Pensionand Health Care Benefits for International Migrants: An Analysis of Issues andGood Practices’, Social Protection Discussion Paper 519 (Washington, D.C.: WorldBank).

Jha, S., Rao, V. and Woolcock, M. (2005) ‘Governance in the Gullies: Demo-cratic Responsiveness and Leadership in Delhi’s Slums’, Q-Squared Work-ing Paper 5 (Toronto: University of Toronto, Centre for InternationalStudies).

Kabeer, N. (2002) ‘Citizenship and the Boundaries of the Acknowledged Com-munity: Identity, Affiliation, and Exclusion’, IDS Working Paper 171 (Brighton:Institute of Development Studies).

Keeley, J. and Scoones, I. (2003) Understanding Environmental Policy Processes: Casesfrom Africa (London: Earthscan).

208 Case Studies

Kochar, A. (2005) ‘Can Targeted Food Programs Improve Nutrition? An Empir-ical Analysis of India’s Public Distribution System’, Economic Development andCultural Change, 54(1), 203–236.

Lipsky, M. (1980) Street-Level Bureaucracy: Dilemmas of the Individual in PublicServices (New York: Russell Sage Foundation).

Long, N. (1989) ‘Introduction: the Raison d’etre for studying rural developmentinterface’, in Long, N. et al. (eds) Encounters at the Interface: A Perspective onSocial Discontinuities in Rural Development (Wageningen: Agricultural Universityof Wageningen).

MacAuslan, I. (2007) ‘Migrants’ Access to Social Protection: A Case Study of thePublic Distribution System in Delhi’, unpublished MPhil Dissertation, Instituteof Development Studies, University of Sussex.

Mooij, J. (1999) Food Policy and the Indian State: The Public Distribution System inSouth India (New Delhi: Oxford University Press).

Mooij, J. (2000) ‘Food and Power: The Political Economy of the Public Distri-bution System in Bihar’, Indo-Dutch Programmes on Alternatives in Development,Occasional Papers and Reprints 2000–1 (New Delhi: IDPAD).

NSSO (2007) ‘Public Distribution System and Other Sources of HouseholdConsumption’, Report 510, Government of India National Sample SurveyOrganisation.

Pande, S. (2007) ‘Social Accountability and the PDS Campaign: The DelhiExperience’, March 2007 draft, cited with permission from author.

Parivartan (2004) ‘Report of Jan Sunwai on Public Distribution System’, unpub-lished report.

Radhakrishna, R., Subbarao K., Indrakant, S. and Ravi, C. (1997) ‘India’s Pub-lic Distribution System: A National and International Perspective’, World BankDiscussion Paper 380 (Washington, D.C.: The World Bank).

Rafique, A. and Rogaly, B. (2003) ‘Internal Seasonal Migration, Livelihoods andVulnerability in India: A Case Study’, paper prepared for the regional confer-ence on Migration, Development and Pro-Poor Policy choices in Asia, June2003, DFID/RMMRU, Dhaka.

Rogaly, B., Biswas, J., Coppard, D., Rafique, A., Rana, K. and Sengupta,A. (2001) ‘Seasonal migration, social change and migrants’ rights,lessons from West Bengal’, Economic and Political Weekly, 8 December2001.

Sabates-Wheeler, R. and MacAuslan, I. (2007) ‘Migration and social protection:exposing issues of access’, Development, 50(4): 26–32.

Sabates-Wheeler, R. and Waite, M. (2003) ‘Migration and Social Protection:A concept paper’, DRC Migration, Globalisation, and Poverty Working Paper T2(Brighton: University of Sussex).

Schaffer, B. (1973) ‘Improving Access to Public Services’, IDS Discussion Paper 23(Brighton: University of Sussex).

Schaffer, B. (1986) ‘Access: A Theory of Corruption and Bureaucracy,’ PublicAdministration and Development, 6(4), 357–376.

Schaffer, B. and Lamb, G. (1974) ‘Exit, Voice and Access’, Social Science Informa-tion, 13(6), 73–90.

Schaffer, B. and Wen-Hsien, H. (1975) ‘Distribution and the Theory of Access’,Development and Change, 6(2), 13–36.

Ian MacAuslan 209

Schlosberg, C. (1999) ‘Immigrant Access to Health Benefits: A Resource Manual’,prepared for The Access Project, Boston, MA and Washington, D.C., The AccessProject and the National Health Law Program.

Schwartz, S. (2001) ‘Immigrant Access to Food Stamps: Overcoming Barri-ers to Participation’, Journal of Poverty, Policy, and Law, September–October,260–275.

Sen, A. (2000) ‘Social Exclusion: Concept, Application, and Scrutiny’, Social Devel-opment Papers No. 1, Office of Environment and Social Development, AsianDevelopment Bank.

Smita (2006) Locked Homes, Empty Schools: The Impact of Distress Seasonal Migrationon the Rural Poor (Delhi: America India Foundation).

SNS (2006) Satark Nagrik Sangathan (Delhi: SNS).Stark, O. (1991) The Migration of Labour (Oxford: Blackwell).Stephenson, R. and Matthews, Z. (2004) ‘Maternal health-care service use among

rural-urban migrants in Mumbai, India’, Asia-Pacific Population Journal, 19(1),39–60.

Supreme Court Commissioners (2006) Sixth Report of the Commissioners, availableat: www.righttofoodindia.org.

Swaminathan, M. (2000) Weakening Welfare: The Public Distribution of Food in India(New Delhi: Left Word).

Uvin, P. (2004) Human Rights and Development (Bloomfield: Kumarian).Zhou, Z. Y. and Wan, G. (2006) ‘The Public Distribution Systems of Food grains

and Implications for Food Security: A Comparison of the Experiences of Indiaand China’, UNU-WIDER Research Paper 2006/98, UNU-WIDER.

8Social Protection as a Multi-ActorProcess in Ecuadorian Migration:Towards a Transnationalism ofSocial Rights?Paolo Boccagni

Every person is equal and shall possess the same rights, dutiesand opportunities.

Nobody shall be discriminated against on grounds of [ . . . ]migratory condition . . .

[The state of Ecuador] fights for the principle of universal citi-zenship, for the free mobility of every inhabitant of the planetearth, for the gradual demise of the condition of alien as atransforming element of the unequal relationships betweencountries, with especial respect to the North-South divide.It demands respect of human rights, with especial regard tomigrating persons . . .

—Constitution of the Republic of Ecuador(2008, Articles 11.2 and 416.6–7 [my translation])

1. Introduction

This chapter provides a comparative analysis of the sources of socialprotection within a long-distance migration flow, connecting Ecuadorto Europe, with a particular focus on the role of the motherland. Itapproaches first the discourse and policies emerging in Ecuador, withrespect to the transnational protection of the citizens – one and ahalf million at least – who have left the country in the last decades(Gratton, 2007; FLACSO, 2008). It then explores the relative weight of

210

Paolo Boccagni 211

migrants’ own practices and resources of social protection, drawing frommy empirical study of Ecuadorian migration to Italy.1 Finally it takesstock of the current and potential interactions between these two axes ofmigrant social protection, asking to what extent, and in which respects,can an integration between institutional and migrant-led initiativesresult in a ‘transnationalism of social rights’?

While migrant access to social protection and welfare benefits in hostsocieties has been extensively studied (for example by Bommes andGeddes, 2000; Sainsbury, 2006), the same does not apply to the influ-ence exerted by the countries of origin. The latter may affect theirexpatriates’ life conditions in many regards, through an increasinglywide range of policy provisions appealing to their ‘external citizen-ship’ – to their claimed patriotic attachment, and to their persistentset of rights (and even of duties) vis-à-vis the motherland (Barry,2006).

In the case of Ecuador, emigration, though decades old, has becomea significant public policy issue – rather than an object of displacementfrom the public discourse – in the last few years only. Apart from theintroduction of expatriates’ voting rights, an emigrant-addressed min-istry has been established, with a view to lobbying for expatriates’ rights,protecting their family members left behind and enhancing their home-ward attachment. Facilitations to emigrants’ investments back home arealso being developed. Even their voluntary participation in the nationalsocial security scheme is to be guaranteed soon.

The tiny Andean country hence provides a promising terrain forexploring the connections between migration and social protection:whether sending states’ increasing attention to their citizens abroad(Gamlen, 2008) may result, among other things, in greater investmentsin their social protection; or if, instead, a more traditional vision of emi-gration as a safety valve only, or anyway as a lever for remittances, ismore suitable to describe sending states’ agendas, whatever their publicrhetoric. Or indeed, to what extent do the two approaches interminglewith one another, and what is the aggregate role of emigrants’ socialprotection practices in accounting for this?

The chapter will complement a review of Ecuadorian policies withan analysis of migrants’ social protection resources and practices. Thisapplies both to remittances and to the other channels of risk protec-tion made available by migration, approached with special attention totheir impact on the home communities. Emigrants’ attitudes and attach-ments towards the motherland will be dealt with at last, as they alsoaffect the scope for social protection ‘from above’ in Ecuador.

212 Case Studies

2. ‘Protecting our brothers overseas’? On the new turn ofemigrant policies in Ecuador

The role of migration in social protection terms, whether resulting inan increase or depletion of welfare resources (or in both, from differ-ent angles), should ideally be approached in a threefold perspective:as a ‘triangle relationship’ involving the interactions between sendingand receiving countries, along with migrants’ informal initiatives. Asmuch literature has shown (for example Givens, 2007; Penninx et al.,2007), migrants’ social protection is also contingent on the structure ofopportunities they encounter overseas, which may be highly variableon a national, and even on a local basis (Alexander, 2007). Having saidthis, the chapter will rather explore the other, relatively neglected sideof the coin: the potential for social protection inherent in the ‘bondingmechanisms’ (Chander, 2006) enacted by an emigration state towards itscitizens abroad, in relation to the latter’s own social protection strategiesand practices.

My effort aims at shedding light on the ways and the degree to whicha transnational welfare framework is being created in progress, fromabove and below, within Ecuadorian migration. This will allow us toassess the potential for integration between state-led and migrant-ledsocial protection. I will finally argue for a framing of social protection(also) in relational terms, with regards to the range of transnationalattachments that may survive – in an affective, an emotional and anidentity sphere – between host and home societies.

2.1. The novel extent of external citizenship: new prospects forsocial protection?

A helpful framework for exploring emigrants’ social protection at a dis-tance is provided by transnationalism, as it emphasizes not only theongoing connections between migrants and their homelands (Porteset al., 1999; Levitt and Glick-Schiller, 2004), but also the influenceexerted by the latter on their expatriates (Bauböck, 2003). Two typi-cal indicators of these new ‘diaspora-reaching’ strategies are the con-cession of external voting rights and the acknowledgement of dualcitizenship.2

A country of origin, in principle, may have many instrumental rea-sons for enhancing ties with its expatriates, or in making their ‘externalcitizenship’ effective (Fitzgerald, 2006). Among them are: human cap-ital upgrading (the expectation to ‘re-import’ the professional skills,

Paolo Boccagni 213

expertise and savings migrants may accumulate overseas); incentivizingremittances; political lobbying, that is approaching expatriates as alobby, supposedly pursuing the motherland’s interests; even cultivat-ing consensus in domestic politics, with a view to reasserting citi-zens’ national identity or pride and to attract emigrants’ votes (eventhough electoral expectations, in themselves, are seldom a sufficientreason).

Still, for a state strategy of loyalty-maintenance to have some impact,the prospect of some distinctive opportunity for expatriates should benecessary, well beyond appeals to their patriotism and nostalgia. Accessto social security programmes, facilitation of savings and investments,new opportunities for self- or qualified employment, and dedicatedsocial support to those left behind may all be viable options.

Whatever the interests driving states’ efforts, however, the reper-cussions for migrant social protection are far from clear – hence therelevance of exploring them in a country which, though secondaryas a political actor in the Southern American scenario, has recentlybeen characterized by a massive emigration flow and by a significant‘diaspora-reaching’ endeavour. This will also help to assess whether atypical corollary of transnational arguments holds true with respect tosocial protection: that ‘increasing opportunities’ for migrants to com-bine ‘external and internal status and affiliations’ (Bauböck, 2003) resultin social rights and resources deriving from both the country of originand that of settlement.

2.2. Emigrant policy-building in Ecuador: potentials andlimitations

While international emigration from Ecuador is at least four decadesold (Jokisch and Kyle, 2008), its latest boost has taken place since thecountry’s financial collapse in the late 1990s (Jokisch & Pribilsky, 2002;Herrera et al., 2005). An active engagement of national institutions, interms of dedicated policy-making, is however relatively recent. Onlyafter the 2006 presidential elections – the first ever with absentee bal-lots – has the role of emigrants gained some salience in Ecuadorianpolitical life, well beyond their already crucial economic weight.3 In thewake of a public discourse developed in many Latin American regimesin the last decades (Berg and Tamagno, 2006), the freshly elected pres-ident began to approach expatriates as a common, collective subject inits own right, hence the discourse on Quinta Región as a metaphor forall Ecuadorian emigrants.4 Their persistent attachment to the country,

214 Case Studies

so the argument goes, justifies their definition as a single political actor,irrespective of territorial proximity.

The new Constitution of the country (Constitution of the Repub-lic of Ecuador, 2008, Article 40 [my translation]), which dedicates awhole section to ‘human mobility’, has accordingly ushered in anunprecedented ‘right to migrate’:

Every person is acknowledged the right to migrate. No humanbeing shall be identified, nor regarded as illegal, as a result of theirmigratory conditions.

The State, through its dedicated institutions, shall develop the fol-lowing actions – among others – for Ecuadorian persons’ rights to beexercised overseas, whatever their migratory conditions:

1. Shall provide assistance to them and their families, whetherresiding overseas or in the country.

2. Shall provide attention, counseling services and thorough protec-tion, for them to be able to exercise freely their rights.

3. Shall protect their rights if, for whatever reason, they are deprivedof their freedom overseas.

4. Shall promote their ties with Ecuador, facilitate family reunionand stimulate return on a voluntary basis.

5. Shall safeguard the privacy of their personal information withinthe archives of Ecuador institutions overseas.

6. Shall protect transnational families and their members’ rights.

The statements in this excerpt suggest a fundamental transformationin the conventional wisdom on migration, which is framed in termsof a personal right. The Ecuadorian state binds itself to facilitatingboth its expatriates’ integration abroad and their transnational ties withthe motherland. Underlying the new Constitution’s approach to expa-triates, and indeed to every citizen, is a ‘transformative’ expectation(Sabates-Wheeler and Waite, 2003), linking their social protection tothe pursuit of anti-discrimination and social equity principles. In a trulytransnational perspective, all Ecuadorian citizens should be entitled tothe ‘good life’ (buen vivir), in so far as the scope for their participationand protection, whether migrants or not, should be granted by the state.

In strictly social policy terms, however, the only major innovation liesin the possibility for emigrants to enrol in the national social insurancefund – although the Ecuadorian welfare arrangements are incomparably

Paolo Boccagni 215

poor and residual compared with those of most destination countries. AsHerrera (2008) reminds us, the public insurance fund ‘currently coversonly 20 per cent of the population and is in constant financial crisis’;more broadly, ‘Ecuador’s rate of social spending is one of the lowestin Latin America’, despite the increasing public investments in the lastfew years, applying, for instance, to basic cash transfers (Weisbrot andSandoval, 2009). Migrants’ access to the national security system maystill mark some progress in the field of portability, responding to theclaims of some expatriate civic associations overseas. It is, however, tooearly to assess the potential of this new provision in terms of welfareresources.

A far less symbolic turning point has been marked, in the terrainof policy-making, by the setting-up of a dedicated Migrant Ministry:Secretaría Nacional del Migrante (SENAMI). The new institution has themission of

defining, managing and implementing migration policies, insideand outside the Country, aiming at the human development ofall the actors involved, allowing for migrants’ care, protection anddevelopment, in accord with the goals of the State of Ecuador.5

Judging from its official agenda, SENAMI is concerned with a range ofnovel initiatives, including: a dedicated saving fund, expected to advisecompatriots when investing at home (and even to turn into a spe-cial ‘migrant bank’); a programme for the grassroots enhancement ofmigrant associations; and a residual line of action to highly vulnerableemigrants, in order to facilitate their repatriation. Among the initia-tives already implemented, however, three programmes are especiallyrelevant to the migration-social protection nexus:

1) Vinculos (Ties), which aims to strengthen the opportunities andmechanisms of emigrants’ involvement and communication withthe motherland, by investing in migrant associations, as well as intheir family members’ collective initiatives in Ecuador;

2) Cucayo, a public fund aiming to stimulate and orientate returnees’productive investments via technical and legal assistance, as wellas through micro-funds for the start-up of new entrepreneurialinitiatives;

3) Volver a casa (Returning home), which aims to support emigrants’physical return by facilitating, via tax exemption, the transport oftheir personal belongings and professional equipment.

216 Case Studies

The ‘Returning home’ initiative has encountered especially mixed reac-tions, both in the motherland and overseas (Boccagni, 2011a). Thisreveals limitations to the legitimacy, let alone the viability, of moreproactive state action towards expatriates. While returning to Ecuadoris often perceived, from emigrants’ viewpoints, as a key future aim, it isgenerally unlikely to take place in the short run, and is even less likelyfor those relatively ‘well integrated’ overseas, after a family reunifica-tion. These emigrants may even be looking at Ecuador, by now, with a‘myth of return’ perspective (Anwar, 1979).

Although promoted with no apparent intention to counter emigra-tion, this return plan may be regarded as of little use to those strivingto move ahead overseas. At the same time, from a public policy view-point, it may be prone to criticisms on equity grounds, were it to beimplemented in a more than symbolic way.6 As migrants are not a dis-advantaged group as such – the contrary may even be the case – therisk exists for a dedicated return plan to be perceived as unduly privileg-ing them over non-migrant citizens. The same applies, indeed, to any‘targeting’ of public provision which is addressed to migrants only.

Altogether the mission and the remit of the new ministry seem tobe still in question, both from an institutional point of view (as itdisrupts the traditional division of labour between government depart-ments) and in the face of the pressures from civil society. EcuadorianNGOs, which made a major contribution to its start-up, are typicallyoriented to claims-making and advocacy more than to migrant-relatedservice provision or policy-making. Within these constraints, and leav-ing the public rhetoric aside, what is the actual potential of state-ledinitiatives?

2.3. How could a country of origin protect emigrants?Here, there, in-between

What are the predictable impingements of the new emigrant policies onthe life conditions of expatriates, and of their kin left behind? Judgingfrom the initiatives announced (and, to a degree, developed) in Ecuadorso far, emigrant social protection ‘from above’ could be effective onthree scales of reference: within the motherland, in the intermediatespace (physical, symbolic and relational) of migrants’ homewards con-nectedness, and concerning the countries of destination, with a view toaffecting migrants’ integration there (see Table 8.1). Each of these realmsdeserves further discussion.

Paolo Boccagni 217

Table 8.1 Emigrant policies as a source of social protection from above: atypology of the realms of action and of the key strategies

Focus of relevance Major examples

Here: in themotherland’sdomestic sphere

• De-stigmatization of the public discourse onemigration and on children left behind

• Psycho-social and legal support to migrant familymembers

• Incentives to return and to social-economicreintegration back home

In-between: in thetransnationalrelationships withexpatriates

• Transnational channels of interaction andcommunication with emigrants

• Exercise of external voting and dual citizenship• Incentives to investments and collective remittances• ‘Virtual return’ (e.g. skill-transferring, training at a

distance)

There: in thecountries ofsettlement

• Pro-migrant advocacy and lobbying towards hostsociety governments

• Legal and social support to migrants in their contextof settlement

2.3.1. Here: in the motherland’s domestic sphere

A more positive and humanistic construction of migration may help,first of all, to de-stigmatize the public discourse on the matter. A com-monsense representation seems to exist, all across Ecuadorian society,portraying family disruption and the abandonment of children as a nec-essary consequence of parental emigration (especially by the mother).The voice of the people in Ecuador, when it comes to approachingemigration, goes somewhat along the lines described by Soruco andcolleagues (2008, pp. 4–5):

a) Emigration is a problem; b) Emigration is bad for the region . . . ,for the emigrants themselves, and especially for the families of emi-grants; c) Emigrants are irrational, failing to use their remittancesin productive and sustainable activities and therefore failing tocontribute to the national economy; d) Emigrants are irresponsi-ble because they abandon their families and thus deserve moralcondemnation; e) Emigrants’ children do worse in school than non-emigrant children, they are not integrated into society . . . and theywill probably try to leave the country as their parents did.

218 Case Studies

Much of this argument is unwarranted, and in the face of the signifi-cant but far more ambivalent effects of emigration on those left behind(Herrera et al., 2005; Camacho and Hernández, 2008), a more nuancedand less reproving approach is helpful. On the one hand, more system-atic programmes supporting emigrants’ children (and those taking careof them) would be necessary. This may apply to their full inclusion inschools and in peer groups, and perhaps even to the delicate issue ofremittance management. At the same time, the risk of accentuating their‘diversity’, hence their segregation, should be avoided. As Herrera (2008)puts it,

Educational services should adopt a view of migration that is lessvictimizing and more normalizing. That is, migration should betreated as a part of daily life for many children and adolescents,thus familiarizing children with the phenomenon while avoidingstigmatization.

2.3.2. In-between: cultivating emigrant connectedness

The space of ‘in-betweenness’ potentially bridging migrants with theirmotherland, mainly through ICTs, provides emigrant policies with newopportunities and challenges. With a view to facilitating transnationalcommunication with the motherland, a range of instruments hasbeen developed by the Ecuadorian government; among them, a webplatform supporting virtual social networks and online TV channels.Fuelling emigrants’ nostalgia, facilitating their interactions with thoseleft behind, providing new scope for celebrating their patriotism – eventhrough external voting (Boccagni and Ramírez, 2009) – are all partsof the same strategy: cultivating emigrants’ connectedness with themotherland, no matter the distance from it in space, and even in time.

Yet, while the aim of enhancing ties between Ecuador and its expa-triates is likely to be shared by the latter, the ways being developedto pursue it, or any potentially alternative way, are more dubious. Asmy fieldwork, and even SENAMI figures suggest, policies investing inthe ‘in-between’ space involve just a narrow, self-selected minority ofEcuadorian migrants.7 Have such ‘diaspora-reaching’ strategies, even so,had a distinctive impact on their life conditions and opportunities? Inother words, can they be regarded as a strand of social protection, albeitfor a minority of emigrants? I will return to this point, after comparinginstitutional initiatives with migrants’ own strategies and practices.

Whatever the case, stimulating emigrants’ remittances and invest-ments is obviously an equally relevant goal. Ecuadorian policies

Paolo Boccagni 219

declaredly aim to create viable opportunities for new micro-enterprisesto start up, and this resonates with migrants’ initial expectations(Boccagni, 2009). It should even pave the way for sustainable and saferlife conditions in the country. However, the conversion from family-addressed to productive remittances is a complex and highly selectiveprocess (Hall, 2007). So far, altogether, remittances in Ecuador haveonly enabled the maintenance of consumption standards, rather thana significant improvement in life conditions (Olivié and Ponce, 2008).

2.3.3. There: struggling for migrant integration

To some extent, the action of a migrant-sending country such asEcuador can affect, on a discursive level at least, migrants’ chances ofintegration overseas. This can be regarded, in fact, as a critical precondi-tion for migrants to be able to return some of their skills and resourcesto the motherland (Fitzgerald, 2006). Diplomatic pressures towards thecountries of settlement typically involve demands for labour-relatedamnesties, or the denunciation of any blatant discrimination (and evenracist assaults) migrants may be subject to. Informal pressure strate-gies can also build on transnational opinion campaigns, potentiallycombining Ecuadorian institutions and civil society. Less conspicuousstrategies include systematic overestimates – in the public stances ofnational authorities – of the figures of Ecuadorians abroad. This espe-cially applies to the undocumented, whose lack of entitlements overseasdoes not justify any differential treatment as far as the country of originis concerned.

Among its declared aims, SENAMI is to provide social and legal helpwithin migrant-receiving countries. This should result in an enhance-ment of the mandate of Ecuadorian consulates (which is often contestedby immigrant associations), and even in establishing Casas ecuatorianas(Houses of Ecuador) in overseas cities with large Ecuadorian popula-tions. The latter are in charge both of facilitating migrants’ sociabilityand common initiatives, and of giving them legal counsel.

While potentially effective, this overseas protection strategy mayprove highly expensive. Its substantial implementation would entail asignificant public investment in capacity building. It would also requirethorough negotiations with local Ecuadorian associations, whose stan-dards and credentials, in terms of representation, democracy andaccountability, may prove difficult to assess.

Whether any of these initiatives is actively supported by the majorityof emigrants is, again, far from obvious. Patriotism and homesick-ness may not be enough to counter a deep-seated scepticism towards

220 Case Studies

Ecuadorian political institutions, which have long been peculiarly weakand unstable, even by Latin American standards (World Bank, 2004;Burbano, 2008). La política is typically framed in expatriates’ accountsas the fundamental culprit for the collapse which accelerated emigrationfrom the country in the late 1990s (Hall, 2005; Boccagni, 2011b).

3. Transnational migration as a mixed source of socialprotection

Whatever the initiatives of a migrant-sending state, the transnationalresources mobilized through migrants’ personal networks – as socialprotection – deserve to be studied in their own right. Financial aswell as relational and emotional resources are at stake here. LeavingEcuador, judging from the migrants’ accounts I collected, may wellprovide greater social protection, at least concerning the daily socialreproduction of those left behind and, sometimes to a lesser extent, ofmigrants themselves. Transnational helping practices, while producingmixed (and often unintended) effects, are, however, quite disconnectedfrom the new emigrant policies in Ecuador (or, for that matter, from anypublic policy).

As my field research suggests,8 emigrants’ ‘need to construct diversi-fied, globalized livelihoods’ (Hall, 2005) comes with significant costs,along with benefits, for those who stay and for those who leave.A twofold perspective should thus be maintained, in order to under-stand the transnational interdependence of the life conditions ofmigrants and non-migrant kin: the crucial ‘migration-left behindnexus’, as Toyota and colleagues (2007) put it.

How is it that Ecuadorian emigrants contribute to the social welfare ofthemselves, of their family members left behind, and even, to a degree,of their home communities overall? An understanding of two majorissues is needed in order to assess the strengths and the limitations ofmigrant protection ‘from below’, within transnational family networks.The generation of income overseas, and its transfer back home, is butone side of the coin. Transnational care giving – in terms of communi-cation, affection relationships and of any practices attempting to recovera sense of proximity at a distance – is the other, less obvious side.9

3.1. Family remittances: underlying expectationsand (un)intended effects

Family-addressed remittances are by far the most visible effectsof Ecuadorian migration in the motherland. The acceleration of

Paolo Boccagni 221

overseas-bound mobility has resulted in expatriates’ transfers constitut-ing a significant weight in the overall national economy (Acosta et al.,2005; Calero et al., 2009). Although remittances have been decreasingin the last two years or so, related to the global crisis impact (Weisbrotand Sandoval, 2009), their economic and social significance is indis-putable in many respects (de Haas, 2007; Carling, 2008): as a sourceof direct external financing, in reducing the vulnerability of a dollar-ized economy; in providing emigrants’ households with resources fordaily consumption and social welfare expenses; in fuelling internal con-sumption; and in revitalizing some economic sectors, along with theirmultiplier effects. Less positive repercussions of remittances, however,should also be taken into account, well beyond the fact, applying allacross Latin America, that only a tiny percentage of remittances are usedin strictly ‘productive’ terms (World Bank, 2007).

Even from a social protection viewpoint remittances may have amixed impact. Their social distribution within Ecuadorian society isuneven, and their actual contribution to local development in themedium term is contentious (Olivé and Ponce, 2008). In a nutshell,‘just as emigrants are often not the poorest citizens, so remittancesdo not necessarily flow to the poorest households’ (Hall, 2007).While remittances remain the major source of social protection chan-nelled by migration, their actual transformative potential is typicallylimited.

This said, my fieldwork suggests a few key points on the socialstructure of expectations and obligations underlying the transnationalcirculation of remittances at both ends of a migration flow. To beginwith, sending money back home is generally perceived by immigrantsas a moral duty in itself, regardless of whether one fulfils it or not.When it comes to transnational families, it often represents the veryfundamental reason for migration. Remittances demonstrate the com-mitment of those who leave to those still in Ecuador. They may even actas a sort of justification, in the face of both the difficult life conditionsof immigrants, and of the distress created by the detachment from theirdear ones at home. In immigrants’ expectations, remittances are a leverfor coping, first of all with basic consumption and ordinary expenses,then for achieving social opportunities – for instance in health and inbetter education, or even in further migration – earlier unattainable.Ultimately immigrants expect remittances to pave the way towards bet-ter future life conditions back home, although the expectation to turnthem into productive investments – and the very option of returning –may prove misleading.

222 Case Studies

As a two-way process across a long distance, however, remittancesand their management are open to different expectations and interestsbetween senders and receivers, even within the same household or kingroup. The issue does not lie only in the structural gap between thepotential demand of those left behind and the help migrants may beable to give; nor, for that matter, does it lie in the asymmetries betweenthe perceptions of ‘migrant life’ being nourished in Ecuador and theactual plight of immigrants. Although most Ecuadorians I met in Italycomplained about the distorted views of their relatives in the mother-land (as if those overseas were ‘making big money’), while in EcuadorI frequently encountered greater awareness of the real situation and lessroom for migration myths than I would have expected.10

A less obvious, but not less significant, asymmetry also applies: emi-grants are often unable, whatever their wishes, to exert any substantialcontrol on the ways remittances are spent. The issue is particularly awk-ward when it comes to dependent children and those taking care ofthem (and living off the remittances as well). While migrants may crit-icize the recipients’ supposed expediencies, they often have their owninterests, even apart from affections and moral obligations, to maintaingood relationships with the communities they left behind. Keeping intouch with non-migrants may compensate them for the loss of socialstatus they typically suffer, as immigrants, abroad. It is also a necessarycondition in paving the way for their return, if – and when – this shouldoccur.

Remittance circulation is certainly the primary means of fuelling a‘transnational protection field’ (paraphrasing Levitt and Glick-Schiller,2004), at least with respect to family members. Still, the extent and thevery contents of the social protection a migrant should assure – providedthey are willing and able to – are by no means obvious and self-evident.They are rather the outcome of an ongoing, reciprocal negotiation. Therequirements of a migrant – seemingly in a relatively stronger position –are set against those of their dear ones at home, which may prove noless influential. While most migrants share the commitment to sendremittances, fewer would completely agree with those left behind as tothe priority options for spending their money back home.

3.2. Transnational caregiving: the ‘affective glue’ of familylife at a distance

Migrants, as Cohen (2005, p. 104) puts it, ‘remit more than money. Theybring goods, services, knowledge, and possibilities that nonmigrants

Paolo Boccagni 223

sometimes cannot imagine.’ In much the same way, migrants’ contri-bution to social protection – whether of themselves or of those leftbehind – cannot be reduced to the money they gain and send back.Indeed, ‘ties with family and friends in migration source areas arean important means for the social protection of migrants whilst indestination areas’ (Sabates-Wheeler and Waite, 2003, p. 40).

Of especial import are the relationships of nurturing at a distance,which develop between family members set apart by migration. Increas-ing attention in recent years has been devoted to the practices oftransnational parenthood (for example Hondagneu-Sotelo and Ávila,1997; Parreñas, 2005; Schmalzbauer, 2005); even more so in femalemigration flows such as the Ecuadorian one, at the beginning at least.11

These practices are the attempts of migrant parents to maintain, whileoverseas, a fundamental role in their children’s upbringing – in terms oflivelihood, affections, and even of everyday care. The tools they use forthis purpose include frequent communication (increasingly involvingnew communication technologies), journeys back home, and system-atic circulation of gifts, pictures and information in a wider sense, viatransnational compatriot networks.

The twofold, ambiguous impact of migration – as a source of familyvulnerability, and as a lever to alleviate that very vulnerability – is espe-cially manifest here, as the emerging debate on a care drain in homesocieties, specifically related to female emigration, shows (Isaksen et al.,2008; Kofman and Raghuram, 2009).12

In a long-distance migration flow, such as the Ecuador-Europe one,transnational caregiving is mostly an informal and spontaneous socialprocess, poorly supported by statutory welfare agencies or NGOs. More-over, it stands in the very centre of a vicious circle. Mothers attempt toprovide their families with social protection by leaving to work abroadand sending home remittances, which satisfies some needs, in terms ofbasic consumption, access to services and even investment in humancapital, which is hardly accessible without remittances. This means,however, sacrificing other elements of their role as caregivers, namelytheir everyday involvement in the education and care of their childrenand families.

As my fieldwork research on Ecuadorian migrants suggests,transnational parents tend to make significant attempts to remaininvolved in their children’s everyday life, to convey a feeling of emo-tional proximity – even to retain the ‘last word’ on any decisioninvolving them (which may prove increasingly difficult, the closer thechildren get to adolescence and an autonomous life). However, this can

224 Case Studies

hardly offset – as their personal accounts suggest – their loss of thepotential for personal communication, care and even control, inher-ent in an ordinary proximity parental relationship (Urry, 2002). Quitewidespread in transnational mothers’ narratives is the perception ofmissing ‘something important’ in their children’s development, alongwith concerns for the expedient use of their remittances, and some-times for the inadequacies of those in charge of caring for their children.For, however valuable their efforts to recreate proximity at a distance, atransnational stratification of opportunities for child rearing is clearlyat work here (Hochschild, 2000). Co-presence, while being perceived as‘natural’ for most families in receiving countries, is in fact a selectiveasset on which migrants cannot necessarily rely.

In the Ecuadorian communities of origin, the condition of childrenleft behind is generally marked by a deep-rooted stigma and often,though not necessarily, results in higher distress and social vulnerability.While ‘transnational children’ do convey a greater need for psycho-social support, their state should not be framed as one of inherentdisadvantage; rather it is contingent on the role played by those tak-ing care of them and on their parents’ ability to bridge, inasmuch aspossible, physical distance (Boccagni, 2009). Moreover, the frailty ofhousehold structures by far pre-exists parents’ emigration, which mayeven result in a potential channel for children’s social mobility – at leastin the terrain of better health and education. To be sure, the access tobetter (private) health services and schools is one of the more tangibleadvantages family members enjoy as a result of emigrant remittances(Hall, 2005; Herrera, 2008).

From both sides, transnational family life is generally regarded as astate relatively more tolerable and less likely to strain family ties theshorter it persists. In the perceptions of those involved, practising closefamily relationships at a great distance makes personal sense only as atransient, short-term condition – which, despite emigrants’ wishes, is infact not necessarily the case.

Judging from the accounts of the Ecuadorian migrants (and fam-ily members at home) I met, transnational caregiving is laden withambivalences. On the one hand, a migrant’s decision is grounded inthe perception of having no real alternatives, and in the expectationof building a better future for their children through their own commit-ment and self-sacrifices overseas. On the other hand, the outcome of thisparticular form of social protection is critically contingent on the roleof children’s caregivers and of broader support networks, no less thanon migrants’ efforts. As one delves into emigrant parents’ narratives, a

Paolo Boccagni 225

painful awareness emerges of the ultimate insufficiency of their efforts,in relation to their children’s emotional and educational development,unless a family reunion is made. This, ironically, may even reduce theeconomic opportunities that stemmed from earlier emigration. Aftera family reunification overseas, while intergenerational relationshipsneed an unprecedented renegotiation, the social and economic statusof migrants’ children is likely to be, at the beginning at least, far lessadvantaged than it was in Ecuador.

4. Social protection from above and below: an integratedframework?

As my research shows, the nexus between social protection from aboveand below remains weak. Poor (if any) connections exist betweenEcuadorian policies, which build on a groundbreaking public discoursebut result in declarations of intents more than in targeted and effectiveprovisions, and migrants’ informal initiatives, which lack communalvision and strategy beyond a family level, but often allow a significantdegree of social protection.

From both viewpoints, however, there is clearly a need to addressmigration-related vulnerability in a multidimensional perspective – thatis, in terms wider than simply economic ones. What is also at stake, inthe face of the isolation and the severances migration may induce, isa need for psycho-social and relational support, both for migrants andnon-migrants (Falicov, 2007). This calls for capacity-building initiativeswith a more ambitious aim: to enhance their social capital and abilityto elaborate a communal vision for the future, despite the distance, pro-vided their basic needs have been satisfied. The point is that migrants’informal networks alone, however valuable, can hardly provide suchsophisticated support.

At the same time, despite the new emphasis of emigrant policies on‘fuelling connectedness’, the actual scope for transnational communi-cation between Ecuador and its migrants seems relatively small. Alto-gether, distance from the motherland does matter. So do the precariouslife conditions of many Ecuadorians overseas – hence some scepticismarises as to how involved they can be, via ICTs, in the evolving socialand political life back home. If external voting is a good test for work-ing out their orientations to participate in Ecuador, then the prospectsfor their effective connectedness are poor. In the 2006 presidentialelections and in subsequent ones, migrant participation – though a sym-bolic turning point – has been irrelevant in terms of numbers (Ramírez

226 Case Studies

and Boccagni, 2010). Overall, it seemed to be dictated by contingentreasons, such as the fear of being forced to pay a fee when back athome, or patriotic/homesick motivations (Boccagni, 2011b). In terms ofa proper interest in supporting distinct political projects and candidates,emigrants’ involvement resulted instead, in line with key literature find-ings (Guarnizo et al., 2003), in a self-selective and numerically marginalphenomenon.

Whatever the main reason for this, there are good grounds for con-cluding that, in practical terms, disconnectedness – rather than inclusion ata distance – better describes emigrants’ structural position towards theirmotherland.

5. Conclusion

Warranting emigrants’ social protection, as far as Ecuador is concerned,is obviously a goal worth striving towards. It is questionable, how-ever, in its real impact and viability, unless it is understood also as arhetorical strategy (as a matter of domestic consensus building), how-ever risky, as it may fuel far greater expectations than it can live upto. The notion of a transnationalism of social rights – which evokes thecreation of ‘overlapping memberships, rights and practices’ (Bauböck,2003) thanks to migrants’ ‘double embeddedness’ in two nation states(Glick-Schiller and Levitt, 2006) – by and large overestimates the state-led social protection accessible to Ecuadorian migrants (and, indeed, tomost migrants).

A greater transnational reach applies to the mutual, horizontal help-ing practices that migrants tend to develop with those left behind. Thisis mostly, however, a form of social protection critically contingent onliving overseas, hence quite difficult to ‘import back’ to the country oforigin. The idea of a selective transnationalism of social opportunities istherefore a more suitable description of the unequal and fragmented setof social protection resources generated by migration.

The action of a country of origin such as Ecuador, vis-à-vis a ‘diaspora’settled in relatively few countries (the US, Spain and Italy – FLACSO,2008), can have a relevant impact in the terrain of public recognition.An emigration state can legitimately vindicate, as Ecuador increasinglydoes, the rights expatriates should be entitled to in host societies, byappealing to the latter’s liberal-democratic regimes (applying at least tonative citizens). This may even contribute to framing the internationaldiscourse on migration in a vision other than a narrowly ‘securitarian’one. That said, the levers of migrants’ social protection ultimately still

Paolo Boccagni 227

lie in their own personal stories – whether successful or not – on theone hand, and in receiving states’ autonomous policy options, on theother.

One may still wonder if an emigration state’s attempts at strength-ening symbolic ties with expatriates – via ICTs, and investing in theircommunal identification overseas (for example via communal gather-ings and celebrations) – could represent a source of social protection inits own right. In practical terms, this is unlikely to have much effecton migrants’ life chances. If, however, such efforts are driven by abroadly patriotic register (rather than a strictly political one), and aslong as migrants have some significant family member left behind, theymay help build that subtle sense of commonality, or of proximity-at-a-distance, which is vital to the perceived well-being of both.

Indeed, it is in this terrain of ‘soft’ social protection – involving thecultivation of patriotic and emotional attachments, and above all thefacilitation of co-national sociability, whether in situ or at a distance –that social protection from above is more likely to succeed in Ecuador.When it comes to ‘hard’ social protection measures (and to the resourcesallowing for it), the key variable at stake is migrants’ spontaneous action,though structurally vulnerable and productive of significant downsidesin the very social protection terrain. It is ultimately on migration itself –as an inherent multiplier of opportunities and, even earlier, of vulner-abilities ‘here’ and ‘there’ – that the social protection of an increasingnumber of people, whether migrants or not, has to rely. Looking atsocial protection both from a hard and a soft side is helpful, however, toshed better light on the new potential of emigrant policies, whether inEcuador or elsewhere.

Notes

1. The empirical bases of this chapter are provided by two streams of researchon Ecuadorian migration. First, my PhD multi-sited fieldwork (2005–2007)on the transnational ties within a migration flow spanning Ecuador andItaly, approached via ethnography and in-depth interviews (Boccagni, 2009,2010a). Second, a study on the social impact of emigration in Ecuador,and then on new emigrant policies and the prospects for enhancing themigration-development nexus, realized on behalf of the Italian think tankCeSPI (2008–2009). The latter step resulted in documentary analysis andinterviews to some 40 key informants (including academics, civil servants,policy-makers and ONG leaders) in Ecuador.

2. Such a trend has applied almost everywhere across Latin American countriesover the latest two decades (Escobar, 2007). In the case of Ecuador, exter-nal voting was acknowledged in the Constitution of 1998, though it was

228 Case Studies

only implemented eight years later. The legitimacy of dual citizenship wasacknowledged in 1994 (Ramírez and Boccagni, 2008).

3. Remittances from overseas are estimated to account for 6–7 per cent of GNP,being the second source of external funding in the country after oil revenues(and by far exceeding development assistance): see, among others, Acostaet al. (2005), FLACSO (2008) and Calero et al. (2009). See also further remarksin Section 3.1.

4. The construction of emigrants as a ‘Fifth Region’ builds on a geographicmetaphor, comparing them to the four geographical regions of the country.

5. Mission statement, available at www.senami.gov.ec – my translation.6. Two years after the Plan’s start, at the end of November 2009, Ecuadorians

voluntarily returned from overseas thanks to the programme amounted tono more than 4500, despite the severe effects of the current recession intheir main receiving countries (Spain, the US and Italy). Further informa-tion on the ‘Welcome Home’ Return programme is available on the websitewww.senami.gov.ec.

7. Systematic users of the ‘Migrante Ecuatoriano’ platform (those registeredto it) were, at the end of 2009, some 5300 overall. While the numberof occasional users may be far higher, the fact remains that this state-led‘virtual involvement’ of emigrants is for now a limited, if promising phe-nomenon. Migrants’ poor access to ICTs, as well as their deep-rooted distrustfor Ecuadorian political rhetoric, may account for this.

8. While the previous section involved a nation-level focus on emerging pro-cesses of institution- and policy-building, here I will draw on my study ofmigrants’ interpersonal relationships with those left behind. This resultedin an ethnography within a trans-local migration flow between a southernEcuador locality (Pasaje, El Oro province) and a local area in northern Italy(Trento province) (Boccagni, 2010a, 2010b).

9. Another facet of the migration-left behind nexus, potentially relevant tosocial protection but still underdeveloped in the Ecuadorian case, involvesmigrant ‘collective remittances’ (Goldring, 2004; Boccagni, 2010b).

10. See, for a similar point in a different context, Mazzucato (2009).11. In fact, a certain gender realignment has taken place in recent Ecuadorian

immigration to Europe, with especial respect to Spain (Gratton, 2007). Thepractice of transnational parenthood has, however, been studied, to a lesserdegree, even in cases involving wives (for example Pribilsky, 2004) or elderlyparents (for example Baldassar et al., 2007) left behind.

12. The emphasis in this perspective, well epitomized by Hochschild’s (2000)‘global care chains’ notion, is on the severe loss of human resources involvedin care labour – in physical and, even more, in emotional terms (Yeates,2004) – in the countries of origin, as a result of widespread emigration.

References

Acosta, A., López, S. and Villamar, D. (2005) ‘Las remesas y su aporte para laeconomía ecuatoriana’, in G. Herrera, M. C. Carrillo, and A. Torres (eds) Lamigración ecuatoriana: transnacionalismo, redes e identidades (Quito: FLACSO –PCMD).

Paolo Boccagni 229

Alexander, M. (2007) Cities and Labour Immigration (Aldershot: Ashgate).Anwar, M. (1979) The Myth of Return: Pakistanis in Britain (London: Heinemann).Baldassar, L., Baldock, C. and Wilding, R. (2007) Families Caring Across Borders

(London: Palgrave Macmillan).Barry, K. (2006) ‘Home and away: the construction of citizenship in an emigration

context’, New York University Law Review, 81(11), 11–59.Bauböck, R. (2003) ‘Towards a political theory of migrant transnationalism’,

International Migration Review, 37(3), 700–723.Berg, U. and Tamagno, C. (2006) ‘El quinto suyo from above and from below’,

Latino Studies, 4(3), 258–281.Boccagni, P. (2009) Tracce transnazionali. Vite in Italia e proiezioni verso casa tra i

migranti ecuadoriani (Milano: Angeli).Boccagni, P. (2010a) ‘Private, public or both? on the scope and impact of

transnationalism in immigrants’ everyday lives’, in R. Bauböck and T. Faist (eds)Diaspora and transnationalism (Amsterdam: AUP).

Boccagni, P. (2010b) ‘Whom should we help first? transnational helping practicesin Ecuadorian migration’, International Migration, forthcoming.

Boccagni, P. (2011a) ‘Returning home: a project, a myth or a political device?the framing of return from above and below in Ecuadorian migration’, GlobalNetworks, 11(4), forthcoming.

Boccagni, P. (2011b) ‘Reminiscences, patriotism, participation. Approachingexternal voting in Ecuadorian immigration to Italy’, International Migration,forthcoming.

Boccagni, P. and Ramírez, J. (2009) ‘Building Democracy or Reproducing“Ecuadorianness”? A Transnational Exploration of Ecuadorian Emigrants’External Voting’, paper given at the IMISCOE Annual Conference, Stockholm,September 2009.

Bommes, M. and Geddes, A. (eds.) (2000) Immigration and Welfare (London:Routledge).

Burbano, F. (2008) ‘Deinstitutionalized democracy’, in C. De la Torre andS. Striffler (eds) The Ecuador Reader (Durham: Duke University Press).

Calero, C., Bedi, A. S. and Sparrow, R. (2009) ‘Remittances, liquidity constraintsand human capital investments in Ecuador’, World Development, 37(6), 1143–1154.

Camacho, G. Z. and Hernández, K. (2008) Niñez y migración en Ecuador (Quito:Unicef-INNFA).

Carling, J. (2008) ‘Interrogating remittances’, in S. Castles and R. Delgado Wise(eds) Migration and Development (Geneva: IOM).

Chander, A. (2006) ‘Homeward bound’, New York University Law Review, 81, 60–89.Cohen, J. H. (2005) ‘Remittance outcomes and migration’, Studies in Comparative

International Development, 40(1), 88–112.Constitution of the Republic of Ecuador (2008), available at: www.

migranteecuatoriano.gov.ec (accessed 4 November 2010).de Haas, H. (2007) ‘Remittances, Migration and Social Development’, UNRISD,

Social Policy and Development Working Paper no. 34.Escobar, C. (2007) ‘Extraterritorial political rights and dual citizenship in Latin

America’, Latin American Research Review, 42, 3.Falicov, C. (2007) ‘Working with transnational immigrants: Expanding meanings

of family, community, and culture’, Family Process, 46(2), 157–171.

230 Case Studies

Fitzgerald, D. (2006) ‘Rethinking emigrant citizenship’, New York University LawReview, 1(1), 90–116.

FLACSO (2008) Ecuador: las cifras de las migraciones internacionales (Quito:UNFPA – FLACSO Ecuador).

Gamlen, A. (2008) ‘The emigration state and the modern geopolitical imagina-tion’, Political Geography, 27(8), 840–856.

Givens, T. E. (2007) ‘Immigrant integration in Europe: empirical research’, AnnualReview of Political Science, 10, 67–83.

Glick-Schiller, N. and Levitt, P. (2006) ‘Haven’t We Heard This Somewhere Before?A Substantive View of Transnational Migration Studies’, CMD Working Paperno. 06-01.

Goldring, L. (2004) ‘Family and collective remittances to Mexico’, Developmentand Change, 35(4), 799–840.

Gratton, B. (2007) ‘Ecuadorians in the United States and Spain: history, genderand niche formation’, Journal of Ethnic and Migration Studies, 33(4), 581–599.

Guarnizo, L. E., Portes, A. and Haller, W. (2003) ‘Assimilation and transnation-alism: Determinants of transnational political action among contemporarymigrants’, American Journal of Sociology, 108(6), 1211–1248.

Hall, A. (2005) ‘Globalized Livelihoods: International Migration and the Chal-lenges for Social Policy: The Case of Ecuador’, paper given at the World Bankconference New Frontiers of Social Policy, Arusha, December 2005.

Hall, A. (2007) ‘Moving away from poverty: Migrant remittances, livelihoods anddevelopment’ in D. Narayan and P. Petesch (eds) Moving Out of Poverty: Cross-Disciplinary Perspectives (Washington, D.C.: The World Bank).

Herrera, G. (2008) ‘Migration and Trends in the Field of Social Policy in Ecuador,1990–2005’, paper available at: www.cespi.it.

Herrera, G., Carrillo, M. C. and Torres, A. (eds) (2005) La migración ecuatoriana:transnacionalismo, redes e identidades (Quito: FLACSO – PCMD).

Hochschild, A. (2000) ‘Global care chains and emotional surplus value’, inW. Hutton and A. Giddens (eds) On the Edge: Living with Global Capitalism(London: Jonathan Cape).

Hondagneu-Sotelo, P. and Avila, E. (1997) ‘I’m here, but I’m there: The meaningsof Latina transnational motherhood’, Gender and Society, 11(5), 548–565.

Isaksen, L., Devi, U. and Hochschild, A. (2008) ‘Global care crisis: a problem ofcapital, care chain, or commons?’, American Behavioral Scientist, 52(3), 405–425.

Jokisch, B. and Pribilsky J. (2002) ‘The panic to leave: Economic crisis and the“new emigration” from Ecuador’, International Migration, 40(4), 75–101.

Jokisch, B. and Kyle, D. (2008) ‘Ecuadorian international migration’, in C. De laTorre and S. Striffler (eds) The Ecuador Reader (Durham: Duke University Press).

Kofman, E. and Raghuram, P. (2009) ‘The Implications of Migration for Genderand Care Regimes in the South’, UNRISD paper no. 41/09.

Levitt, P. and Glick-Schiller, N. (2004) ‘Conceptualizing simultaneity: a trans-national social field perspective on society’, International Migration Review,37(3), 1002–1039.

Mazzucato, V. (2009) ‘Bridging boundaries with a transnational researchapproach’, in A. M. Falzon (ed.) Multi-Sited Ethnography (Aldershot: Ashgate).

Olivié, I. and Ponce, J. (2008) Remittances and Development: Are we MissingSomething? (Quito: FLACSO, Documento de trabajo 06/305).

Parreñas, R. (2005) ‘Long distance intimacy’, Global Networks, 5(4), 317–336.

Paolo Boccagni 231

Penninx, R., Gerger, M. and Kraal, K. (eds) (2007) The Dynamics of InternationalMigration and Settlement in Europe: A State of the Art (Amsterdam: AmsterdamUniversity Press).

Portes, A., Guarnizo, L. E. and Landolt, P. (1999), ‘The study of transnationalism:Pitfalls and promises of an emergent research field’, Ethnic and Racial Studies,22(2), 217–237.

Pribilsky, J. (2004) ‘ “Aprendemos a convivir”: Conjugal relations, co-parentingand family life among Ecuadorian transnational migrants in New York Cityand the Ecuadorian Andes’, Global Networks, 4(3), 313–334.

Ramírez, J. and Boccagni, P. (2010), ‘Construyendo la Quinta Región: Voto a dis-tancia y analisis de la participación política de los ecuatorianos al exterior’,in C. Amescua and J. Luque (eds) Política en movimiento: estado, ciudadanía,exilio y migración en América, (Ciudad del Mexico, CRIM-UNAM), 14, 347–73,(forthcoming, 2010).

Sabates-Wheeler, R. and Waite, M. (2003) ‘Migration and Social Protection’,Working Paper T2, Migration DRC.

Sainsbury, D. (2006) ‘Immigrants’ social rights in comparative perspective’,Journal of European Social Policy, 16(3), 229–244.

Schmalzbauer, L. (2005) Striving and Surviving. A Daily Life Analysis of HonduranTransnational Families (New York: Routledge).

Soruco, X., Piani, G. and Rossi, M. (2008) ‘What emigration leaves behind:The situation of emigrants and their families in Ecuador’, Inter-AmericanDevelopment Bank WP # R-542.

Toyota, M., Yeoh, B. and Nguyen, L. (2007) ‘Bringing the “left behind” back intoview in Asia’, Population, Space and Place, 13, 157–161.

Urry, J. (2002) ‘Mobility and proximity’, Sociology, 36(2), 255–274.Yeates, N. (2004) ‘Global care chains’, International Feminist Journal of Politics, 6(3),

369–391.Weisbrot, M. and Sandoval, L. (2009) ‘Update on the Ecuadorian economy’,

policy brief, Washington, Center for Economic and Policy Research.World Bank (2004) Ecuador: Development Policy Review (Washington, D.C.: The

World Bank, Report no. 27443).World Bank (2007) Close to Home: The Impact of Remittances in Latin America

(Washington, D.C.: The World Bank).

9Coping and Investment Strategiesof Migrants in the South: MalawianMigrants in South AfricaRachel Sabates-Wheeler

1. South–South migration and social protection

In the last two decades, migration literature has burgeoned, but ithas been mainly focused on South–North and North–North migrationflows. The extent and issues surrounding migration between developingcountries remain poorly understood, largely because data on migra-tion is sparse and unreliable. ‘South-South migration is not new. It isjust ignored by economists’ and many others (Hatton and Williamson,2002).1 Ratha and Shaw (2007) use recent data to estimate bilateralmigration stocks and provide some hypotheses of the determinants andsocio-economic implications of South–South migration. By a rough esti-mate, in 2005 two out of every five migrants on the globe – some78 million out of 191 million migrants – were residing in a devel-oping country.2 Most of these migrants are likely to have come fromother developing countries. Ratha and Shaw estimate that South–Southmigration is nearly as large as South–North migration.

World Bank research (2005) reports that income, proximity and net-works are the major drivers of migration from developing to industrialcountries.3 As South–South income differentials are relatively modest,proximity and networks are likely to have a proportionally greaterimpact on migration flows. Unlike in the context of South–North migra-tion, available data indicate that differences in country income play alimited role in South–South migration. About 38 per cent of identifiedSouth–South migrants come from countries with higher incomes thantheir host country (Ratha and Shaw, 2007). And of the migrants fromlower income countries, most go to countries with incomes only slightlyabove that of their home country. Only about 20 per cent of identified

232

Rachel Sabates-Wheeler 233

South–South migration from lower to higher income countries is tocountries with very large differences in income.

Several factors are likely to explain the muted impact of country-income differences on South–South migration. Income differencesamong developing countries are, of course, much smaller than betweendeveloping and industrial countries. Lack of resources, limited accessto travel documents and the location of networks may limit migrants’ability to travel far beyond neighbouring countries where income differ-ences are often low. Ratha and Shaw (2007) further find that migrantswho travel to other developing countries experience much lowerincreases in income, are more likely to have irregular status, are subjectto greater risks of exploitation, and are more likely to be expelled thanthose who migrate from developing countries to industrial countries.

1.1. Characteristics and strategies for risk reduction

It can, then, be hypothesized that South–South migration differs fromSouth–North migration in several important aspects. Four particularlyimportant ones that have been noted are: proximity of sending andreceiving countries, the importance of networks, high prevalence ofirregular migration, and duration and frequency of visits.

1.1.1. Proximity of sending and receiving countries

Proximity and ease of crossing borders explain the large volume ofSouth–South migration despite relatively small wage gains from migra-tion. Ratha and Shaw (2007) hypothesize that South–South migrationis usually less expensive than South–North migration, because distancesare shorter (as a large majority of South–South migration is to nearbycountries) and air travel is often unnecessary. Because many South–South migrants lack adequate travel documents, they are restrictedto overland migration. Moreover, ethnic, family and religious tieslink communities across borders, particularly in Africa, where colonialboundaries straddle tribal groups, but also in other parts of the world.

1.1.2. Networks

The importance of networks within the study of migration is firmlyestablished (Boyd, 1989; Fafchamps, 1996; La Ferrara, 2003; Cassini,2005; Mazzucato et al., 2006). It is not our purpose to review this lit-erature here, but rather to draw attention to the likely importance ofnetworks for South–South migrants in terms of the characteristics oftheir migration. Ethnic, community and family ties reduce the costs anduncertainties involved in migration. In Africa, migrant networks play a

234 Case Studies

critical role in encouraging outflows once migration is underway (Lucas,2005). For instance, in some areas migrant diasporas are long-standing:the arrival of significant numbers of Swazis, Tswanas and Basothos inSouth Africa (as opposed to Swaziland, Botswana and Lesotho) stemsfrom movements of tribal groups in the nineteenth century (Solomon,1996).

1.1.3. High prevalence of irregular migration

For a variety of reasons, irregular migration is even more commonin South–South than in South–North migration. The demand for suchworkers is higher because law enforcement is weaker in developingcountries and so employers are more prone to hire migrants withoutproper documents so that they can pay lower wages. Moreover, infor-mal economies are larger in developing countries, and consequentlythese countries can accommodate higher shares of illegal workers intheir informal sectors. The supply of irregular migrants (the likelihoodof migrants entering into the receiving country illegally, or overstay-ing their visas) is also higher in developing countries. First, due to theporous borders, entering developing countries illegally is easier thanentering developed countries. Second, the restrictive, bureaucratic andcostly migration regimes of developing countries push immigrants tocircumvent them by entering or staying in host countries illegally, andby working without work permits (and without a work contract aswell). And third, very poor migrants face difficulties in obtaining traveldocuments in their countries, and find it easier to migrate to otherdeveloping countries without proper documents.

1.1.4. Duration and frequency of visits

Much of the economic literature considers migration as permanent.But temporary migrations are frequent in South–South flows, and oftenthe rule rather than the exception. Arguments as to why migrantsreturn home, even in the context of persistently higher wages in thehost country are: theories of relative deprivation and arguments ofrisk spreading (Stark, 1992), location-specific preference (Hill, 1987;Raffelhüschen, 1992), credit market rationing (Mesnard, 2000), highpurchasing power of the host country currency in the migrant’s homeeconomy, and higher returns to human capital and self employmentactivities (Dustmann, 1995, 2003).

Research suggests that South–South migration patterns have a higherprevalence of shorter, seasonal migration flows, as well as higher returnmigration and re-migration, as a direct consequence of the proximity


of sending and receiving countries characterizing South–South migra-tion, and the relative ease of crossing borders (Beals and Menezes,1970; World Bank, 1993; Peberdy, 1998; Nangulah and Nickanor, 2005;Lassailly-Jacob et al., 2006). Lucas (2005) reports that the durations ofstay depend in part upon the fixed costs of re-entry and the odds ofbeing able to return abroad again. For instance, intensive repeat migra-tion is prevalent in the border areas of Mexico and among seasonalworkers in the EU; whereas migrants from Asia to the Gulf stay onaverage for around four to five years.

1.2. Migration as social protection

While the economic benefits for migrants from South–South migrationappear to be limited in absolute terms (as discussed by Ratha and Shaw,2007), it is critical to note that, given that the majority of South–Southlabour migrants move due to poverty at home, even small increases inincome can have very substantial welfare implications for people in suchcircumstances. In this sense the action of migration itself is an informalsocial protection mechanism for migrants and their families who remainbehind. Within this understanding, migration as social protection over-laps substantially with the literature on coping strategies, such thatmigration becomes a coping strategy within a portfolio of livelihoodchoices (see Sabates-Wheeler and Waite, 2003, for a fuller exposition ifthis relationship).

Holzmann and Jørgensen (2000, p. 141) point out that migrationas an informal coping strategy may be conceptualized as a means ofreducing the probability of shocks before they happen, or an informalmeans of coping with shocks once they have occurred. Given the gener-ation of income through migration and the transfer of income throughremittances associated with migration, migration can be argued to act asa private or informal coping mechanism or even as a social protectionstrategy (in the sense that it may be a form of risk management), usu-ally at the individual or household level, that primarily provides incometransfers to disadvantaged households.

We can think of migration as social protection as fulfilling differ-ent functions. Some individuals may migrate in order to improve theirlife chances or incomes. Others use migration as an insurance or riskdiversification strategy. That is, a family strategy may be to send oneor more of its members to a city or abroad, but the whole familywill only move once the migrants have secure livelihoods. By diver-sifying their activities the family reduces vulnerability through both

236 Case Studies

income diversification (agricultural and non-agricultural, for instance)and informal insurance (retaining a livelihood at origin represents afall-back position for the migrants in case they are unsuccessful at thedestination location). Migration can also be used as a protective strategy,essentially a safety net for very vulnerable/poor households.

1.3. The social protection needs of South–South migrants

The substantial South–South migration flows, documented by Rathaand Shaw and highlighted in Chapter 3 by Sabates-Wheeler and col-leagues, raise concerns about social security for these migrants. As in anymigration flow, migrants move between countries and hence betweendistinctive labour markets and social security systems, which creates spe-cific vulnerabilities. Given that many of the developing countries’ socialsecurity systems are underdeveloped and underfunded, South–Southmigrants are particularly vulnerable in host labour markets. Further-more, migrants – in particular low-skilled, irregular migrants – face chal-lenging labour market conditions leading to exclusions from nationalsecurity systems in host countries related to cross-border recruitment,information asymmetries between employers and migrants, and visarequirements tied to employment. Moreover, access to formal socialservices in the new host country is often restricted for many reasons,such as informal labour market involvement, employer monopsony anddelayed access until some months or years after arrival.

In the absence of security, protection and rights for the majority ofmigrants, how do migrants seek, obtain and negotiate relationships andsystems in order to secure themselves against health and income shocksat destination, as well as providing services for dependents, such aschildcare, schooling and health care? At times these strategies requirethe use of parallel systems that function to serve those who are notserviced through the formal sphere (see Chapter 2 for a discussion ofa parallel system operating within the Ghanaian migrant communityin the UK). At other times, migrants may access security through ille-gitimate means, either by being employed illegally, or by engaging inpractices that enable them to circumvent and use the formal systemsto their advantage (for instance, the widespread ‘borrowing’ of ID cardsand national insurance cards). Later in the chapter we look into thesestrategies.

Based on the review above I am interested here to explore some of thehypotheses and assertions related to the characteristics of South–Southmigration using new survey data collected from Malawian migrants in


and returned from South Africa. I am particularly interested in explor-ing the poverty status of migrants and how this might drive them tomigrate. The chapter details how migrants use migration as a socialprotection strategy in terms of consumption security for families backhome, and income creation, as well as future investment strategies. Thelast section of the chapter describes how Malawians in South Africaare able to access formal and informal social protection mechanisms,through legitimate and illegitimate routes. I look at the role of net-works and the use of networks to achieve a positive change in income.The conclusion reports that social networks and past experience in thecountry (which I hypothesize are developed through repeat migration)are vital for obtaining positive outcomes from migration. I suggest thatefforts must be made to facilitate repeat migration as well as to fostermigrant networks if the migration experience is to be positive for thepoorest migrants.

2. Methodology and datasets

The data used to inform the findings reported here come primarily fromtwo new specialized migrant surveys, one quantitative, the other qual-itative.4 These surveys were designed to explore the different character-istics and experiences of migrants who had spent at least one migrationepisode in South Africa. The quantitative survey was conducted with 150Malawian return migrants from South Africa. The survey was conductedby the Centre for Social Research, Zomba, Malawi, between January 2007and August 2007. The South African returnee survey was complementedwith a qualitative survey of 21 undocumented Malawian migrants andillegal Malawian workers living in the township of Katlehong and inHillbrow, which is located in inner-city Johannesburg, where migrantslive in slum-like conditions. The qualitative interviews were conductedin order to better understand some of the motivations for return,the details and stories of how migrants manage risk and access socialprovisioning, and other issues around employment and legality.

A further quantitative survey was performed that collected informa-tion from 201 returnee Malawians to Malawi from the UK. I use thisdata insofar as it helps to illustrate the characteristics of South–Southmigrants.

2.1. Malawi and South Africa: migration and migrants

With the exception of studies examining the migration of Malawianmen to South African mines in the nineteenth and twentieth centuries,

238 Case Studies

only limited information is available on Malawi labour migration.5 InOctober 2005 the Malawi National Statistical Office carried out an Inte-grated Household Survey. The findings from the survey to a large extentconfirm previous statements regarding the one-way migration streamout of the country. Asked whether household members had always livedin their current location or if they had moved from elsewhere within thelast five years, less than 5 per cent reported having moved from outsideMalawi to rural or urban areas (MNSO, 2006).

Malawi has historically been a prominent labour source country tothe South African mines. Since the centrally organized migration toSouth Africa (for mining) ended in the 1980s, the regional labour migra-tion from Malawi continued to be predominantly circular in nature anddominated by men migrating towards ‘centres of production’. Thosemigrating clandestinely have been grouped into two categories: thosewho go to work in tertiary and service sectors, and those engaging incross-border trade, capitalizing on past migration experience and knowl-edge of local languages. Due to the history of widespread intra-regionalmigration, reinforced by shared languages and cultures, and also rein-forced by colonialism, there is widespread undocumented cross-bordermovement of frontier workers that daily commute to and from workacross national boundaries. Such informal cross-border movements havecontributed to the fragmentary and incomplete nature of data collectionon cross-border movements in Africa.

The fragmentary accounts of Malawian irregular labour migration thatexist suggest that some Malawians cross the border into South Africairregularly, look for work or work irregularly without work permits, getpicked up by the police and deported, and turn round and go backagain to South Africa. Some migrants are motivated to return due tothe fact that when they are picked up they are never allowed time towind up their personal business and carry home with them any acquiredasset. These deportees typically enter South Africa irregularly or overstaytheir visitor’s permit. They frequently access a destination country utiliz-ing century-old migration culture, colonial ties, connections and basicknowledge.

Itaye (2007) reports that Malawian irregular migrants in South Africaare mostly unskilled or semi-skilled. The study notes that the jobsof these irregular migrants consist largely of the temporary, hard andunpleasant tasks that few people want to do. In the host countries theyare rated as poorly paid jobs. In English these jobs are often referred toas the ‘three Ds’ – dirty, dangerous and difficult.6 In this job category,the study notes that Malawian migrants are characterized by evidence


of non-possession of work permits and are often overstayers on theirresidence permits. When required to travel back to Malawi, they visit theConsul Office to obtain an ‘Emergency Travel Document’, giving variousreasons for not having a Malawi Immigration Passport. This group ofMalawi migrants are mostly working adults, between the ages of 18 and30. Migration for these Malawians is an investment and insurance strat-egy to secure assets for their future, which include physical capital andeducation. Some acquire assets for their marriage as social investmentstrategy and for future livelihood.

Informal social security mechanisms (social networks, informal creditand information exchange) provide many Malawian migrants withsocial protection in both the destination country and Malawi. Thegovernment of Malawi does not have any formal policy guidelines toenhance migrants’ capacities and capabilities. Accumulation of assetsconstitutes the predominant social security. These assets, if not ‘crimi-nalized’ and the migrants sent home involuntarily, are transferred hometo Malawi where they are either liquidated and the cash used for invest-ments, or the asset is utilized on a day-to-day basis providing a liveli-hood for the immediate and extended family (Tsoka and Mvula, 2006).

3. Findings I: who migrates?

In this section I present the findings from the surveys to illustrate thecharacteristics of Malawian migrants to South Africa. In keeping withthe literature reviewed above we would expect the characteristics ofMalawian migrants to the South to be different from those of Malawianmigrants to the North. This is mainly due to the differential costs ofmigration (costs of flight, visa and information), distance of migration,and destination immigration and labour market policies (for instance,the ability to enter South Africa with a 30-day visitor permit). We wouldexpect that the distance travelled to destination would affect the oppor-tunity cost of return. For instance, if deported or returned from SouthAfrica, it is relatively easy and cheap for a migrant to return to SouthAfrica, compared to the long-distance, northern locations. Thus, thelength of stay is unlikely to be as important for Malawians in SouthAfrica due to the ability to easily and quickly return. Our discussionswith Malawians in Malawi and in Johannesburg reveal that migrantscan be deported one day and be back in their jobs in South Africa verysoon afterwards. A respondent in Malawi reported that he had beendeported from South Africa on a plane. It was the first time he hadbeen on a plane and it was full of other Malawian migrants. He revealed

240 Case Studies

that the whole experience had been quite pleasant, in fact very enter-taining, as he was able to experience a flight, visit his family for freeand then save to return again to South Africa a couple of months later.Other interviewees in our Johannesburg sample reported incidences ofdeportations and return within just a few months.

Furthermore, we would expect to see largely unskilled/semi-skilledentrants with lower levels of education going to South Africa, as com-pared to, say, the UK. Migrants to South Africa would also comefrom poorer families and communities. Descriptive statistics from thedatasets, presented below, confirm the suggested differences in migrantcharacteristics. In Table 9.1 we see that Malawian migrants returningfrom South Africa come from larger households with significantly morechildren (on average) and are more likely to be male and single. Migrantswho have returned from the UK are more likely to come from smallerfamilies, have less children and stay longer (due to the high cost ofmigration and the high opportunity cost of return).

Table 9.2 below illustrates the educational differences betweenMalawian migrants according to destination. We see, as expected, thatmigrants to South Africa had less education on average than those to theUK, with only 24 per cent having tertiary or technical qualifications andnone having graduate and professional training. At the other end of theeducational distribution we see that a massive 78.7 per cent of Malawianmigrants to South Africa have primary or no education as their highestlevel of educational attainment, whereas only 3 per cent of Malawians

Table 9.1 Demographic indicators by country of destination, means

Malawiansto SA

Malawiansto UK

Significant?

DemographicsAge at point of migration (yrs) 26.80 25.02 NoFemale migrant (per cent) 11.30 30.80 YesMarried (per cent) 72.00 98.00 YesHousehold Members 8.79 6.32 YesNo. of people residing with 5.99 5.01 NoNo. of workers here 4.56 3.59 YesNo. of Children in household 4.64 0.58 YesLength of stay at destination (yrs) 2.83 3.17 NoYears since return 1.33 1.08 No

Note: Number of observations = 150 for Malawians in South Africa, and 201 for Malawiansin the UK.


Table 9.2 Level of education by country of destination

Level of education UK SA Total

None/primary 6 118 1243% 78.7% 35.4%

Secondary 81 30 11140.5% 20% 31.7%

Tertiary/technical 82 2 8441% 1.33% 24%

Graduate/professional 31 0 3115.5% 0 8.9%

Total 200 150 350100% 100% 100%

to the UK have a similar educational profile. As discussed by Itaye (2007)and others, this is what we would expect from anecdotal and qualitativeevidence. That the less-educated Malawians (who are likely to be poorer)are less likely to see the North as a destination for migration probablyreflects the different labour market policies of the receiving countries.

The relationship between educational attainment and country of des-tination is likely to reflect underlying differences in poverty profiles ofmigrants rather than differences in the respective migration policies ofdifferent receiving countries.

South Africa represents a huge market for unskilled and semi-skilledlabour from neighbouring countries. Qualitative work tells of howMalawians are widely perceived to be reliable, honest and hardworking,to the extent that white South Africans advertise for domestic and gar-dening help in Malawian newspapers. They also advertise specifically forMalawians in South African suburban and national newspapers. Manyof the ‘wanted’ adverts for home help specify that the help should be‘Malawian’. White South Africans interviewed in Johannesburg reportthat Malawians are a reliable and cheap source of domestic help andservice sector labour. The Malawian respondents also appeared to dis-tinguish themselves, in terms of their work ethics, from black SouthAfricans:

We are friendly with whites, but our brothers kill us. They are makingapartheid among themselves; there is no unity; the Zulu and Sothoare fighting; this doesn’t solve any problems. They should remem-ber who helped South Africans when they were in exile. Malawianshaven’t got a problem with white people. White people defend

242 Case Studies

foreigners; we haven’t got a problem with white patrols of ‘Bandapeople’ (Zulus), but they warn us about black people. South Africanshate foreigners because they say we take their jobs. South Africansare lazy and want more money. It is now their turn to develop theircountry, otherwise the country will go down. South Africans are fail-ing even to make a garden to grow and sell vegetables. And thewomen are just bathing and looking for men. Whereas we work tobuild our lives.

(Interview with Smart Banda)

A focus group respondent revealed that:

I got a job working with my cousin in a gardening business. Theowner didn’t care that I didn’t have a permit. He preferred to hireMalawians instead of South Africans. He said we are hard workerswhile South Africans don’t like to work hard. I have been workingthere for 5 years with no problems. But I was arrested once and spent3 days in jail. My cousin told the owner and he gave him R700.00which my cousin used to pay a guard for my release.

3.1. Legal status

We would expect all Malawian migrants entering South Africa to enterlegally given the ability to enter on a 30-day visitor permit. The major-ity of Malawian labour migrants to South Africa are unskilled and haverelatively low levels of education, as illustrated above. These migrantschoose to extend or overstay their visitor’s permits as they engage in theSouth African labour market. Thus, over time we would expect to seea large number of migrants becoming completely undocumented. Thedata collected indicates the change in legal status of the survey respon-dents at arrival in the country of destination and upon departure. Allmigrants entered South Africa legally as expected. Over time, 51 per centof migrants became completely undocumented. The category of ‘legal’or ‘documented’ is not a particularly useful one when interrogatingmigrant labour patterns and employment rights. Disaggregating ‘legal’into its constituent parts shows that no migrants entered South Africawith the right to work full-time. From the 100 per cent who entered ona 30-day visitor permit, 98.7 per cent were working illegally. At point ofdeparture, 13.5 per cent had obtained the right to work full-time, 21.4per cent were residing legally but working without the correct docu-ments, 13.5 per cent were residing legally and not working. So, over timewe witness a huge increase in the number of workers who are residing


and working completely undocumented, as well as a large increase inthe number of migrants who have been able to obtain the right to worklegally and full-time.

Perhaps the primary way in which the problems of differences in legalstatus manifest themselves is through the labour market and migrants’economic opportunities. Workers’ rights generally cannot be enforcedfor the undocumented, they cannot be accurately counted or assessed,and thus are correspondingly more vulnerable to employer abuses thatlead to reduced earnings and the ability to remit.

Many of these effects are likely to be indirect. Migrant workers whoare brought by recruiters to work in developed countries may be eitherlegal or undocumented, but their rights are likely to be constrained bya lack of networks and lack of knowledge of the receiving country. Thisoccurs wherever recruited migrant labour exists in significant numbers.For the legal but bonded migrant (in the sense that they are tied to aparticular employer), dependence on a single employer/recruiter (whichmay be exacerbated for those present on temporary worker visas whichtie them to a single employer) leads to financial abuses such as finesfor not fulfilling quotas, forced withdrawal of wages for payment of‘debt’ for equipment or travel, excessive interest repayment on debt andnon-payment or extremely low payment of wages. Box 9.1 illustratesthis for the case of Malawians in Johannesburg. Anderson and Rogaly(2005) term this ‘super-exploitation’. They note that this problem is farworse for undocumented migrants since they are often smuggled intothe country and therefore have to pay off fees when they work.

Box 9.1. Bribery and extortion

Irregular migrants, especially South–South migrants, are freque-ntly required to pay substantial bribes to officials to allow them toremain in the destination country. The story recounted by Frans,a Malawian irregular migrant in Johannesburg, illustrates theinstitutionalized nature of this type of exploitation of migrants:

I used the bus to travel to South Africa from Malawi. I did have apassport and a visitor’s visa for one month. I knew I would never getlegal permission to work here. Like everybody else, I simply stayed onafter my visa expired. I was arrested 3 times – the first time was in1993 – for having an expired visa and no work permit. They sent meto Morningside Jail where I remained for 5 days. My brother-in-law

244 Case Studies

Box 9.1. (Continued)

bribed one of the guards with R700.00 to let me go. The other twotimes I was arrested, my brother-in-law did the same thing, althoughthe price went up each time: the last time was in 1994 when he hadto pay R850.00 to get me released.

It is common for Malawians (and all other foreigners in South Africa)to bribe officials for various things – it all depends on how muchmoney you have. If, like me, you are not rich, you’ll only affordto pay small money to be released if you are caught. If you haveenough money, you can pay an official to get you a South African ID(identity) book.

I used to visit Malawi during those years using my passport andexpired visa. I simply paid the border official R50.00 to allow me tocross. I did that each time I travelled to Malawi. Many others weredoing the same thing. Although R50.00 isn’t a lot, when you add allthe fifties from all the people crossing with expired documents, thoseguards made a lot of money each day. We are not happy bribingofficials, but if it is the only way of staying on and working, thenwe will have to do it. We all have to support our families, so we areprepared to do anything to remain here.

On coping with illegal status, many of the Malawian respondentsin Johannesburg reported that when stopped by police, they bribethem and are allowed to go, as long as they are carrying theirpassports. The amount of the bribe is random. The police do nottake it directly, but ask them to put the money on the dashboardof the car, or somewhere similar, so they do not handle the cashdirectly. The police also sometimes tear up documents in front ofmigrants.

The right to change employers is important in mitigating wage dis-parities between legal and undocumented workers. Even legal migrantsin many countries do not have the right to change jobs (Andersonet al., 2006). This type of restriction is generally focused on low-skilledworkers, as in Asia, where Taiwan, Thailand, Malaysia, Singapore, Japanand Korea have all passed laws designed to restrict workers’ mobilityand access to jobs. This leads to multiple inefficiencies: workers can-not access jobs, are punished where they work without authorization,


and are deported or imprisoned. Some states, such as Malaysia, allowfreelance companies acting as vigilantes to deport the undocumented(Xiang, 2008), while others, such as Korea, impose fines on apprehendedirregular migrants before deporting them, and subsequently make thempay the costs of deportation.

In these restrictive labour markets there are also hidden penaltiesfor remaining undocumented. Domestic workers’ lack of legal status,for instance, gives their employers power to withhold earnings andpay below minimum wage, and also means they cannot open a bankaccount to save their money (Punpuing et al., 2005).

3.2. Poverty and employment

Figure 9.1, below, displays pre-migration and post-migration subjectiveevaluations of poverty. Respondents were asked to classify themselvesaccording to the following four categories: (1) poor, unable to manage;(2) average, just able to manage; (3) above average; (4) well-off. We seethat for pre-migration poverty profiles, Malawians choosing the UK asa destination were much more likely to classify themselves as aboveaverage than Malawians going to South Africa. Approximately 60 percent of Malawians returning from South Africa classified themselves aspoor before departure, whereas only 7 per cent of Malawians returning

0

10

20

30

40

50

60

70

80

pre-migration

post-migration

pre-migration

post-migration

UK SA

poor average above average well-off

Figure 9.1 Pre-migration and post-migration poverty status, by country ofdestination, in percentages.

246 Case Studies

Prior Occupation in origin, by country of destination

05

101520253035404550

per

cent

UK SA

Prof

Clerica

l

Servic

e

Produ

ction

Other

Unem

ploye

dSale

s

Figure 9.2 Prior occupation in origin, by country of destination.

from the UK felt they were poor before departure. No respondent thatreturned from South Africa claimed to be well-off prior to migration,whereas nearly 10 per cent of migrants to the UK said they were well-off. As we hypothesized above, poverty status/income is likely to be asignificant determinant of choice of destination.

A similar pattern can be seen when we look at prior occupations(Figure 9.2). Migrants to the UK are more likely to have been in profes-sional, clerical and sales occupations, whereas migrants to South Africaare much more likely to have been in the production sector, holdingjobs as labourers, drivers and so on. A large percentage of migrants fromboth countries were unemployed prior to migrating: approximately 47per cent for those going to the UK and 27 per cent for those to SouthAfrica. The category ‘other’ does not reflect a low-status occupation. Infact the majority of ‘other’ for UK migrants are the self-employed, busi-nessmen and students, whereas the majority of ‘other’ for South Africamigrants are fishermen and subsistence farmers.

4. Findings II: migration as social protection

Given that most respondents migrated to improve their incomes, migra-tion itself can be seen as a social protection strategy, in that the purposeis to reduce vulnerability with regard to income (and obviously con-sumption) for the migrant and their family back home. In the lasttwo years, at destination 89 per cent of Malawians in South Africa still


provided support to relatives and family in Malawi and 45 per cent pro-vided regular financial support. Over the same period, 23 per cent ofmigrants in South Africa received regular financial support from Malawi.

4.1. Uses of financial support from the destination tothe home country

Regular financial support from South Africa to Malawi was used inmany ways. The vast majority of migrants said that it was used for‘basic needs’. Just 20 per cent of remittances were used for schooling.A significant amount of remittances from South Africa were used forfarming purposes, 35.3 per cent. This likely reflects the migrants’ rurallivelihoods back home. Health expenditure also represented a significantuse of remittances.

Poorer Malawian migrants prioritize family welfare, as indicated fromqualitative work with 17 irregular migrants living in two townships ofJohannesburg.7 All respondents cited poverty as a reason for coming toSouth Africa; they were often encouraged to go by either family mem-bers or friends at home or in South Africa. The majority of respondentsdid not want to stay in South Africa. Thus, for them, migration is, inand of itself, a means of social protection for themselves and their fam-ilies, both in South Africa and in Malawi. All but one of the migrantsinterviewed had a clear objective to their migration. Many times theobjective was to fund a very specific investment, such as ‘to build a roofon my house back in the village’, or ‘to cover the costs of secondaryschool for one of my daughters’, or ‘to purchase a fishing net’ or ‘tobuy a popcorn machine’. These plans were time bound and most of themigrants had performed calculations about the exact time it would take(given their current salary) to fund the investment and return home.

Frans, a Malawian migrant in Johannesburg, tells of the poverty thatcaused him to migrate:

I am 40 years old and first came to South Africa in 1991. My cousinwas working and living in Johannesburg. I was doing small-scalefarming in Malawi and not making much of a living for the fam-ily. The soil was poor and fertiliser expensive, so I was only growingenough for the family to eat – sometimes, we ran out before the nextharvest. There was no opportunity to find any work in Malawi. Mycousin in Johannesburg said I should come to Johannesburg and lookfor a job. He helped me find my first contract job here. But it wasn’tenough. So, I looked for work by standing on the street corners andwaiting for employers to stop and offer work. I was picked up by

248 Case Studies

a White man with a grass cutting business who needed additionallabour.

4.2. Impact of migration on physical assets

Only around 7 per cent of migrants accumulated physical assets in theplace of destination as a result of migration. In contrast, approximately45 per cent of migrants returning from South Africa accumulated phys-ical assets in their home country as a result of migration. However, thenature of investments differs quite highly between migrants. Figure 9.3,below, shows that significantly more UK migrants invested in land backhome (27 per cent as compared to 8 per cent of South African migrants),whereas South Africa migrants invested significantly more in housingand small livestock. This difference probably reflects the poverty andlivelihood profiles of the different groups of migrants, with migrants toSouth Africa being poorer and accumulating less in absolute terms asindicated through the investment choices.

4.3. Future provisions

Regular migrants from the South often contribute to social security pro-grammes while working abroad, but many have little to show for thesecontributions when they move on or return home. This is particularlyrelevant in the case of public pensions or health insurance, to which

0

5

10

15

20

25

30

35

40

UK SA

Land

Housin

g

Busine

ss

Buildin

g

Agr.M

ach

Larg

e Liv

esto

ck

Small

lives

tock

Figure 9.3 The impact of migration on physical assets in home country(percentage saying yes).


some migrants contribute significant sums of money while abroad.Limits to the ‘portability’ (or transferability) of pensions gives someshort-term migrants less incentive to work in jobs in the formal sec-tor, if this means contributing to social security schemes from whichthey will not benefit. Moreover, a lack of portability may underminereturn or circular migration, as migrants who have spent a consider-able amount of time in their host countries are likely to factor in theloss of benefits if they return to their countries of origin. More dras-tically, many short-term migrants will prefer to opt out of the formal,legal system altogether as they wish to avoid tax and social securitypayments. Furthermore, it is possible that the vulnerabilities that mayattract insurance in the global North, such as pensions and life insur-ance, may not be priorities for poorer migrants from some countries inthe South, where health insurance may be a priority.

Priorities for the future expressed by Malawian migrants are shown inTable 9.3, below. The most significant investments for the future weresavings, housing and land, and small businesses. Very few migrants werecontributing to private pensions or health in home country or at desti-nation. This is likely to reflect the preferences of migrants in general,their relative lack of attachment to a host labour market and also theirrelative poverty status. The table illustrates that, overall, migrants witha full-time right to work make more investments for the future, duein large part to their higher levels of initial and current wealth. How-ever, the social protection priorities are similar across all categories ofmigrants: savings, investment in property and small businesses.

Table 9.3 Investments for the future of South–South labour migrants in homelocations, per cent indicating ‘yes’

Legal Status of Malawian returnees from SA atpoint of return

Full-timeright to work

Residencyrights but noemployment rights

No residency oremploymentrights

Savings 53 23 22Pensions and health 0 0 1Housing and land 59 34 29Small business 47 30 45Farming 41 9 6

Source: Sabates-Wheeler (2008).

250 Case Studies

5. Findings III: social protection for migrants atdestination

5.1. Employment-based social protection

Migrants often have a plethora of alternative strategies to access inter-national labour markets and to achieve positive outcomes from theiroverseas migration. Many legal migrants saw little point in contributingto formal social security benefits, and this made working in the informalsector an attractive option for some migrants. Table 9.4, below, showsthe percentage of migrants that were entitled to a standard package ofemployment-based social protection in both their first and last jobs atdestination.8 It also shows the per cent of migrants that were entitled toa similar package in their jobs prior to migration (the number of obser-vations reduces to 162 as many of the migrants were unemployed). Wesee that the majority of Malawians who travelled to South Africa andwho had jobs prior to migration were not entitled to standard benefits inMalawi (only 13 per cent). This reflects the occupations that South Africamigrants had prior to migration (as discussed above). During the migra-tion period only 28 per cent had social protection across their first andlast jobs, probably reflecting the sectors that the migrants are workingwithin at destination.

In terms of labour contracts, the data shows that upon arrival only5 per cent of workers to South Africa claim to have formal writ-ten legal contracts. Surprisingly, we find that many irregular workersclaim to have access to formal social protection. For workers enteringSouth Africa – all irregular – 33 per cent claim to have entitlementsthrough their employment. This entitlement-belief might appear sur-prising given that the majority of workers are working without thecorrect documentation. Two conclusions can be drawn: (1) migrantsare ignorant of their inability to access benefits, or/and (2) migrantsare entitled to other employment-related benefits that are not state-funded. The later conclusion is likely to be true. While this may appearunusual, the supporting qualitative work from South Africa provides

Table 9.4 Percentage of migrants having access to standardemployment-based SP at origin and at destination

Entitled at origin (%) Entitled at destination (%)

Yes 13 28No 87 72


rich stories about many employers of illegal workers providing healthsupport, housing and other allowances for sickness.

5.2. Informal social protection

Many migrants were already part of pre-existing social networks beforethey arrived in South Africa, which afforded them accommodation, foodand, at times, job recommendations. Family networks are of vital impor-tance as people can make different types of claims on particular familymembers.

Of family support, Frans reported:

When I arrived, I stayed with my cousin for eight months. I didn’tpay any rent; I only contributed to the food. He was very good to me.We believe that we must help one another because tomorrow I couldalso need help from someone. Also, because you have been helped bysomeone, when you can, you help the next person because you knowhow important this kind of help is and you know you yourself bene-fited and God will punish you if you don’t help someone else in turn.

I have managed to bring two of my brothers to South Africa becausethey weren’t doing anything useful in Malawi. One returned toMalawi, but the other one is still working here as a gardener, with-out the legal papers for the last four years. I found him his first jobin the neighbourhoods where I work. We have had to bribe officialseach time that he has been caught – I think he has been caught threetimes already.

From the quantitative work, we find that most migrants (96 per cent)received support with accommodation and employment in their firstfew months at destination. A majority of migrants received financialsupport. Only 4 per cent received support with the immigration process.Some migrants received information on state benefits (approximately 15per cent). Very few (less than 5 per cent in both destinations) receivedsupport with childcare. During their migration period 73 per cent bothgave and received social support at destination. In other words, recip-rocal social support was the norm. Migrants were asked if they weremembers of one or more of five groups: (1) a workers’ (trade) union,(2) a migrants’ association, (3) a religious institution, (4) a ‘home’(Malawian) association, and (5) a burial/funeral society. Close to 37 percent of South Africa-based Malawians were members of at least oneof these groups. Approximately 20 per cent of South Africa migrants

252 Case Studies

claimed to be members of a religious institution and 5 per cent weremembers of a home association.

The majority in the qualitative study said that they were members ofchurch groups, some of which are all-inclusive, and some of which areexclusively for Malawians. In many instances, respondents had receivedhelp from the church in the form of food, clothing and monetary dona-tions. Some respondents contribute to this form of social protection bythemselves donating money to the church. Church membership alsoprovides a sense of community and belonging, and one respondent saidthat he is able to share his problems with church members, includingSouth Africans.

Burial societies are also an important form of identity and affiliation.A focus group held with men in Johannesburg revealed that

it is important for every Malawian to die at home in Malawi or at leastto be buried in Malawi. We believe that our ancestors will be able tofind you easily if you are buried there. If you are buried elsewhere,how will they locate you? So, even if we earn little, we think it isimportant to belong to the burial society so we know that whateverhappens, we will be buried at home in Malawi.

Another respondent said:

Many of us are also members of a burial society. We donate moneyevery month for the purpose of sending bodies home for burial inMalawi. Otherwise, if we see that someone is busy dying here, wecan also use the money to send them home to die in Malawi. It isalways better to die at home. That is what we all wish for: nobodywants to die far from home. We pay R50.00 per month. The soci-ety has three elected members who take care of the banking and thefinances: whenever any money is to be withdrawn, all three signa-tures are needed, so no person can cheat. If someone is not in theburial society, but needs to be returned to Malawi for burial, we alldonate whatever we can afford to make it possible. We also make suchdonations when a fellow Malawian is in need because of some crisis –maybe illness or death in Malawi. We don’t usually give individualsassistance or money here in Johannesburg while they are looking forwork. In that situation, it is up to that individual’s own relatives andfriends to assist them.

Other forms of informal protection take place through accessing parallelor illegitimate systems of provision (Box 9.2).


Box 9.2. ‘Parallel’ systems of social protection

Men’s group

Man 2:

Some Malawians get South African IDs illegally. I know ofMalawians who ‘bought’ or got a forged South African birth cer-tificate and then used it to get an ID. It is easier to get a false birthcertificate than a false ID. Some forgeries are good, but you can neverbe sure about how good the forgery is until you try and use it to getan ID.

Man 4:

I know of two Malawian men who have married South Africanwomen in order to get Permanent Residence. [Laughs] I can’t sayfor sure that they didn’t love these women, but I know that they saidprivately that they were marrying them in order to get the permits.I have also heard other men, not only Malawians, say that they areplanning to marry South African women in order to get ResidencePermits.

Women’s group

Woman 1:

It is nearly impossible for Malawian women to find work here ifthey don’t have the right papers. I came to South Africa because myhusband was here and I was pregnant with our second child. Wedidn’t know that you needed a South African ID number to go tohospital. Luckily, when the time came to have the child, I used theID of a Malawian woman who has managed to get an ID. We looknearly the same, so I was able to deliver the baby in hospital.

Woman 4:

I managed to get a job as a domestic worker with the help of a SouthAfrican lady who is my friend. She is also a domestic worker andheard of a job in the house next to the one where she was working.She told the Madam that I didn’t have papers but that I was a goodworker: I worked in a hotel in Malawi before, so had experience ofhousekeeping before. I get paid less than the South Africans (less than

254 Case Studies

Box 9.2. (Continued)

the minimum wage) but that’s because I’m an illegal. I understandthat and accept it. My husband is working as a mechanic’s assistantin a garage. He also gets paid less than he should because of being anillegal immigrant. We are supporting many, many people back home.He is supporting about eight people every couple of months. I am sup-porting my mother who is taking care of the four orphans left behindby my brother. His wife returned to her family. He died of AIDS.

Woman 5:

Without an ID, you are lost in South Africa. When you try to go toa hospital, they ask you for your ID number. The form that you fillin needs your ID number.

In summary we see that a substantial number of migrants rely onreciprocal support between home and destination. Furthermore, theyacquire few ‘immovable’ assets as a result of migration and accumulatemodest savings. Across all descriptive results we see that Malawians trav-elling to South Africa are relatively poor, not well educated and fromlower-skilled professions. Despite this, the majority of migrants in oursample reported an improvement in their financial situation as a resultof migration.

6. Improved incomes, social networks and time

From the descriptive results above we see that migration works as a formof social protection for the majority of migrants. That is, remittancesand asset accumulation enable a migrant and his family to insure againstcurrent and future shocks as well as future expenses. Migration alsoenables migrants to maintain and build assets at home. In terms of socialprotection for migrants (as opposed to migration as social protection)the findings above show that informal social networks are importantfor Malawian migrants abroad, whereas the results on formal social pro-tection are less clear. One of the main hypotheses here is that socialprotection (formal and informal), by defining a migrant’s fall-back posi-tion or insurance, enables migrants to benefit (or not) from a migrationspell, and is thus related to positive migration outcomes. I argue thatsocial networks necessarily require time and experience at destinationto build up and that, therefore, the length of stay at destination or


the familiarity with destination will also be important for determiningoutcomes.

It is difficult to fully understand the effects of networks and experi-ence at destination on migration outcomes, due mainly to the possibleendogeneity between these two factors. Social networks are likely to bebuilt through experience, while at the same time experience (durationor repeat migration) acts as proxies for the quality of the network, asover time networks become thicker as migrants invest more time inthem. While it is not possible to fully grasp the extent of this endoge-nous relationship given the data available, I attempt to understandbetter the relationships between networks and experiences at differentdestinations. In particular, I explore the factors that influence positiveoutcomes from migration experiences for returnees’ using multivariateregression analysis.

Table 9.5 provides a description of the variables included in the anal-ysis, as well as their means. The dependent variable being used is anindicator of the perception of income change due to the migration

Table 9.5 Variable description and mean values (standard deviations in paren-theses)

Variable Variable description Mean

Change income 1 if income has improved due to migration;0 otherwise

0.49

Age 1 if age between 30–40; 0 if age 20–29 0.58

EducationNone/primary 1 if none or primary; 0 otherwise 0.78Secondary 1 if secondary; 0 otherwise 0.22

initial_occupUnemployed 1 if unemployed; 0 otherwise 0.29Other 1 if ‘other’; 0 otherwise 0.28

Service/labour 1 if service sector or labourer; 0 if unemployed 0.43

members number of household members 8.79 (3.51)

jobnum number of jobs held during migration (0–14) 1.91 (0.99)

Gender 1 if female; 0 if male 0.11

num_visits Number of previous migration episodes1 if visits >1; 0<= 1

0.33

length number of years at destination (0–14) 2.84 (2.42)

SPindex1 informal SP index (0.5–5.5) 2.36 (1.05)

entitled_SP 1 if entitled to employment based SP;0 otherwise

0.28

256 Case Studies

episode. This variable is not constructed from income, but from askingthe respondent to evaluate their change in income due to the migrationexperience.

We would expect social networks to be of prime importance in achiev-ing a return to migration for migrants to South Africa, as the majorityare unskilled, of low education, and many will go only for acquiringjobs and income. The frequency of visits will also be important as this isan indicator of the duration of migration and familiarity/connectednesswith the destination labour market. In order to test the relevance of net-works to migration outcomes I constructed an index by combining twotypes of informal social protection into an index of informal social pro-tection. More specifically, the index was constructed from the followingtwo items:

1. (a) Number of types of social support given in South Africa9

(b) Number of types of social support received in South Africa2. Number of groups in South Africa

For items (1a) and (1b), migrants in the bottom quartile are scored zero,those in the second bottom quartile are scored one, those in the sec-ond top quartile are scored two, and those in the top quartile are scoredthree. Migrants are scored their mean score on the two sub-items. Foritem two, migrants in the bottom tercile are scored 0.5, those in the mid-dle tercile are scored 1.5 and those in the top tercile are scored 2.5. Theindex is the sum of the scores on the two items.10 The average (mean)value of the informal SP index is 2.63 with a standard deviation of 1.15.For UK migrants the SPI is 2.84, while for the South Africa migrantsthe value is 2.36. The difference in mean in statistically insignificant(t = 3.17).

From a total sample of 150 I ran a limited dependent variable analysis.The marginal effects from the models are used to measure the impact ofdifferent factors on two different outcomes. The outcome is whether themigrant experienced an improvement in their income due to the migra-tion experience. This is a subjective indicator but specifically requiresthe migrant to reflect on whether the migration spell enabled them toincrease their financial income.11

6.1. Findings of the model

Table 9.6, below, provides the results of a probit regression estimation.The dependent variable takes the value of one for a positive change inincome due to the migration experience, and zero otherwise. In keeping


Table 9.6 Probit regression estimates for determinants of a positive change inincome due to migration

Change in income South Africa returnees

Model 1 Model 2

Demographics Duration and SP

dF/dx S. Err. dF/dx S. Err.

Demographic characteristicsAge (base: 20–29)

30–39 0.045 0.092 0.010 0.102Education obtainedbase: none/primary

secondary 0.080 0.112 0.099 0.125≥secondary

Initial occupationbase: unemployed

Other −0.047 0.118 −0.123 0.123Service −0.021∗ 0.105 −0.275∗ 0.114Professional

Gender −0.312∗ 0.114 −0.263∗ 0.129Members_HH 0.003 0.013 0.092 0.015Num of jobs 0.117∗ 0.044 −0.007 0.056

Duration and social protectionNum of visits 0.285∗ 0.111Duration −0.006 0.026Informal SP index 0.109∗ 0.048Formal SP 0.082 0.105obs. 143 141Pseudo R2 0.084 0.164

∗Indicates significance of the parameter at the 5 per cent level.

with the broader literature I include a number of variables to measuredemographic and respondent characteristics, such as age, education,prior income and prior occupation. Model 1 presents the results ofthis estimation for South Africa migrants. Prior occupation and genderare significant. Due to the different characteristics of the group noneof these migrants held professional, clerical or sales positions prior tomigration. We see that migrants who were labourers and service sec-tor workers in previous occupations are less likely to achieve a positivechange in income compared to those who are unemployed. This isprobably due to the fact that any job, after being unemployed, will pro-vide a positive change in income. We also see that female migrants are

258 Case Studies

less likely to achieve a positive outcome compared to males. Also, thenumber of jobs that a migrant holds during the migration episode isrelated to positive outcomes.

Model 2 shows point estimates of an expanded regression thatincludes the informal social protection index and length of stay as sepa-rate variables. As expected, we see that informal social protection turnsout to be highly significant while length of stay is not significant. Thenumber of prior visits (repeat migration) is also of critical importance.For each extra visit, migrants are 28 per cent more likely to achieve animprovement in income. Again, this fits well with what we expect fromthe literature and our qualitative work.

7. Conclusions

As described by Sabates-Wheeler and colleagues in Chapter 3, there is nostrong model of how to move forward on social security provisioningfor South–South migrants. This is not to say, though, that formal socialprotection for South–South migrants is irrelevant. Yet, the priorities aredifferent to those in the case of South–North or North–North migra-tion and one has to differentiate when looking into policy implications.Given the limited provisions and low overall coverage by social secu-rity of the labour force in low-income regions, it seems that concernsabout the lack of portability of benefits are premature. Instead, a moreeffective policy direction to improve the social protection position ofinternational migrants from and within poorer regions would be to (i)create a proper policy framework to manage migration in the region, inparticular undocumented migration; (ii) focus on social protection forthe most vulnerable groups – women, children, refugees and undocu-mented migrants – by improving their legal status in the host countryand ensuring that their most basic rights are respected; and (iii) developstandards on how to coordinate social systems in the future, to ensureportability of acquired social security rights. In other words, the criti-cal issue is to find ways to bring the ‘uncovered’ under a formal socialsecurity umbrella within the countries that they reside.

In this chapter I have used new datasets to explore the characteristicsand social protection profiles of South–South migrants – in particu-lar, Malawians moving between Malawi and South Africa. In keepingwith recent suggestions about the differences between South–South andSouth–North migration, I confirm that the two groups are differentacross a range of characteristics. In particular I show that migrants to theSouth are a poorer group of people and come from ‘poorer’ families than


migrants to the North. This is not unsurprising given the literature thatposits a relationship between costs of migration and relative wealth ofmigrants. In other words, there is a clear selection effect on destinationand for this reason alone it is important not to pool data on migrantsacross a range of destinations if we are interested in understandingmigration and migrant characteristics.

I was also interested to understand how migrants use migration as aform as social protection, as well as how they access social protection atdestination in order to obtain security and higher income for themselvesand their families. Migration represents a very important safety net aswell as an asset-building strategy for most migrants. Social networks andrepeat migration, in particular, appear to be critical to achieving higherincome and secure livelihoods.

Notes

1. Important recent exceptions are papers by Gindling (2009), estimating theimpact of migrants from Nicaragua on labour market outcomes in neigh-bouring Costa Rica, and by Ratha and Shaw (2007), estimating the size ofSouth–South migration and remittances, and discussing key issues related toSouth–South migration.

2. Avato et al. (2010) provide similar estimates.3. Motivations for South–South migration also include seasonal patterns and

flight from ecological disasters or civil conflict. Other motivations, notrecorded in the migration stock data given above, include transit to theNorth and petty trade.

4. The research was funded through the Development Research Centre onMigration, Globalisation and Poverty, University of Sussex, Brighton. Thisis a DFID-funded research consortium.

5. While scholars affiliated to the Southern African Migration Project (SAMP)have over the last 14 years worked extensively on labour migration insouthern Africa, this work has largely focused on migration betweenBotswana, Lesotho, Namibia, South Africa and Swaziland. This observationwas confirmed when scholars and employees at large bilateral organiza-tions working on Malawi were approached to inform this study. All agreedthat there was no or only limited literature on labour migration fromMalawi.

6. Translated into the Malawian Chinyanja vernacular language as ‘three Zs’–zonyansa, zoopysa ndiponso, zobvuta.

7. A qualitative, micro-study was undertaken of a group of 17 Malawians livingin Johannesburg, South Africa, in 2008, the majority of whom were ‘undocu-mented’, ‘irregular’ or, in terms of official interpretations of the ImmigrationAct, ‘illegal’. Four of the respondents were women, and 13 men.

8. The question was a general question that asked: ‘Were you entitled to stan-dard social benefits such as paid sick leave, paid holiday leave; maternityleave, pay for overtime and redundancy?’.

260 Case Studies

9. This is created based on the five types of networks identified in the ‘informalsocial protection’ section.

10. Its range is 0.5–5.5. Its mean is 3.0 and its standard deviation is 1.3.11. The second dependent variable is a measure of actual current income. As

some migrants in the sample returned home up to 20 years ago, I restrictedthe sample to those migrants who have returned within the last five years.I used this to check whether the significance of explanatory variables (thatinclude length of migration spell and social networks at migrant destination)are robust across different measurements of income.

References

Anderson, B. and Rogaly, B. (2005) ‘Forced labour and migration to the UK’,COMPAS.

Anderson, B., Ruhs, M., Rogaly, B. and Spencer, S. (2006) ‘Fair enough? Cen-tral and East European migrants in low-wage employment in the UK’, JosephRowntree Foundation.

Avato, J., J. Koettl and R. Sabates-Wheeler (2010) ‘Social Security Regimes, GlobalEstimates and Good Practices: The Status of Social Protection for InternationalMigrants’, World Development, 38(4.)

Beals, Ralph E. and Menezes, Carmen F. (1970) ‘Migrant labour and agriculturaloutput in Ghana’, Oxford Economic Papers, 22(1), 109–127.

Boyd, M. (1989) ‘Family and personal networks in international migration:recent developments and new agendas’, International Migration Review, 23(3),638–670.

Cassini, S. (2005) ‘Negotiating Personal Success and Social Responsibility: Assess-ing the Developmental Impact of Ghanaian Migrants’, Business Enterprisesin Ghana’, Master’s thesis, International School for Humanities and SocialSciences, University of Amsterdam.

Dustmann, C. (1995) ‘Savings behavior of migrant workers: a life-cycle analysis’,Zeitschrift für Wirtschafts- und Sozialwissenschaften, 115, 511–533.

Dustmann, C. (2003) ‘Return migration, wage differentials, and the optimalmigration duration’, European Economic Review, 47(2), 353–369.

Fafchamps, M. (1996) ‘The enforcement of commercial contracts in Ghana’,World Development, 24, 427–448.

Gindling, T. H. (2009) ‘South-south migration: the impact of Nicaraguan immi-grants on earnings, inequality and poverty in Costa Rica’, World Development,37(1), 116–126.

Hatton, T. and Williamson, J. (2002) ‘What Fundamentals Drive World Migra-tion?’, NBER Working Paper 9159 (Cambridge, MA: National Bureau of Eco-nomic Research).

Hill, J. K. (1987) ‘Immigrant decisions concerning duration of stay and migrationfrequency’, Journal of Development Economics, 25, 221–234.

Holzmann, R. and Jørgensen, S. (2000) Social Protection Sector Strategy Paper: FromSafety Net to Trampoline (Washington, D.C.: World Bank).

Itaye, B. (2007) ‘Social Protection and Migration: Draft Report of a Prelimi-nary Desk-Based Study of Malawian Migrants in South Africa and the UnitedKingdom’, Background paper prepared for this research, available from Author.


La Ferrara, E. (2003) ‘Kin groups and reciprocity: a model of credit transactionsin Ghana’, American Economic Review, 93, 1730–1751.

Lassailly-Jacob, V., Boyer, F. and Brachet, J. (2006) South-South Migration: Exampleof Sub-Saharan Africa (Strasbourg: European Parliament).

Lucas, R. E. B. (2005) International Migration and Economic Development: Lessonsfrom Low-Income Countries (Cheltenham: Edward Elgar).

Mazzucato, V., Kabki, M. and Smith, L. (2006) ‘Transnational migration and theeconomy of funerals: changing practices in Ghana’, Development and Change,37, 1047–1072.

Mesnard, A. (2000) ‘Temporary Migration and Capital Market Imperfections’,mimeo, University of Toulouse.

MNSO (Malawi National Statistical Office)2006. ‘Migration Baseline Survey’,Draft Report. National Statistical Office, Zomba. Malawi.

Nangulah, Selma M. W. and Nickanor, Ndeyapo M. (2005) ‘Northern Gate-way: Cross-Border Migration between Namibia and Angola’, SAMP MigrationPolicy Series 38, available at: http://www.queensu.ca/samp/forms/form.html.(Kingston, Ontario: Southern Africa Migration Project, Queens University).

Peberdy, S. (1998) ‘Trading places: cross-border traders and the South Africaninformal sector’, SAMP Migration Policy Series 6, available at: http://www.queensu.ca/samp/forms/form.html. (Kingston, Ontario: Southern AfricanMigration Project, Queens University).

Punpuing, S., Caouette, T., Panam, A. and Zaw, K. M. K. (2005) Migrant DomesticWorkers: From Burma to Thailand, available at: http://iussp2005.princeton.edu/download.aspx?submissionId=52058.

Raffelhüschen, B. (1992) ‘Labor migration in Europe: experiences from Germanyafter unification’, European Economic Review, 36, 1453–1473.

Ratha, D. and Shaw, W. (2007) ‘South-South Migration and Remittances’, WorldBank Working Paper No. 102 (Washington, D.C.: World Bank).

Sabates-Wheeler (2008) ‘Migration Outcomes, Legal status and Social Protection:a quantitative study of Malawian returnees from South Africa and the UK’,report prepared for DFID, available from Author.

Sabates-Wheeler, R. and M. Waite (2003) ‘Migration and Social Protection: AConcept Paper’, DRC Migration Working Paper T2 (Brighton: University ofSussex).

Solomon, H. (1996) ‘Strategic perspectives on illegal immigration into SouthAfrica’, African Security Review, 5(4), 12–14.

Stark, O. (1992) The Migration of Labour (Oxford: Blackwell).Tsoka, M. and Mvula, P. (2006) ‘Migration and Poverty Study’, Draft Report,

Centre for Social Research (Zomba: University of Malawi).World Bank (1993) Mali: Assessment of Living Conditions (Washington, D.C.: World

Bank).World Bank (2005) Global Economic Prospects 2006: Economic Implications of

Remittances and Migration (Washington, D.C.: World Bank).Xiang, B. (2008) ‘Return Migration in Asia’, Presentation at Metropolis Confer-

ence on Migration and Development, Bonn, 28–31 October 2008.

10Bangladeshi Labour Migrants inthe Gulf: Coping with a LargeUnmet Need for Social ProtectionChowdhury R. Abrar1

1. Introduction

Over the last few decades international migration has become morecomplex and diverse with temporary migration becoming increasinglypopular. Migrant workers (those who migrate for employment) and theirfamilies account for about 90 per cent of the estimated total of 214million international migrants in 2010. After the oil price hike in thefirst half of the 1970s, the Gulf states required various kinds of workersas they embarked on large-scale development projects. This created con-ditions for the migration of workers for temporary employment fromdifferent parts of Asia, including South Asia. The Gulf countries con-tinue to remain major destinations for migrants from South Asia, whoare mostly employed in the construction, domestic work and serviceindustries.

While labour migration has significantly contributed to alleviatingpoverty and stabilizing some of the less-developed economies of theregion, it has posed enormous challenges for the migrants and mem-bers of their families, having significant implications on their welfareand protection. Many are well treated and succeed in attaining theirgoal of earning a better livelihood. A section, however, is subjected to‘low wages, unsafe working environments, non-payment of wages, a vir-tual absence of social protection, denial of freedom of association andworkers’ rights, discrimination and xenophobia’(International LabourOrganization, 2010).

In recent years the plight and the problems of migrant workers inthe Gulf states have received a fair degree of attention from humanrights groups. In the following paragraphs, summary reports of the New

262

Chowdhury R. Abrar 263

York – based Human Rights Watch (HRW) on the treatment of migrantworkers in some of the Gulf states are presented:

Based on extensive interviews with workers, government officials andbusiness representatives, the HRW report titled Building Workers andCheating Workers documents serious abuses of construction workersby employers in the UAE. These abuses include unpaid or extremelylow wages, several years of indebtedness to recruitment agencies forfees that UAE law says only employers should pay, the withholding ofemployees’ passports, and hazardous working conditions that resultin apparently high rates of death and injury.

(HRW Report on UAE, 11 November 2006)

In early 2008, 55 Pakistani, Indian, Sri Lankan and Bangladeshi work-ers, including medical professionals and cleaners, complained to theSaudi Ministry of Labour that Nukhba House of Medical Servicesrequired them to work an extra year after their contracts expired,failed to pay their return tickets home, retained up to eight months’wages and withheld approval for exit visas required under Saudi law.On receipt of the complaint the Ministry contacted the company andadvised the aggrieved migrants to wait for two weeks for the companyto respond. When the company failed to respond, the matter wasreferred to the Labour Court, a tribunal administered by the Ministryof Labour. The Court agreed to hear the case only after four months,leaving the workers without legal status and pay.

(HRW Report on Saudi Arabia, 8 May 2008)

In Saudi Arabia HIV testing is mandatory. A positive test result maylead to detention for up to a year without access to medicationand subsequent deportation. In 2008 the UAE deported 1518 non-citizen residents infected with HIV, hepatitis types B and C andtuberculosis. Migrants living with HIV are often explicitly excludedfrom treatment. If they are detained, they are often denied accessto antiretroviral drugs, and then if deported they cannot get thenecessary care.

(HRW Report on HIV Positive Migrants, 23 September 2009)

Withholding the wages and confiscating passports appears to be ram-pant, but the authorities do nothing to stop it. There is no systemto make sure these vulnerable migrant workers can actually recoverboth their passports and wages, let alone to punish the abusiveemployers. Instead of enforcing the law, as it is authorized to, the

264 Case Studies

government steers workers into an arbitration system that favours theemployers.

(HRW Report on Bahrain, 4 November 2009)

Labour courts in some of these countries are notoriously slow, oftentaking years to hear and decide a case, which can then be appealed,further delaying a final decision. Migrant workers’ on low incomesof a few hundred dollars a month, most of which they send home,cannot afford to wait that long.

(HRW Report on Saudi Arabia, 28 May 2010)

This chapter focuses on the vulnerability of Bangladeshi short-termmigrant workers working in the Gulf states within the broad frameworkof social protection. In the introductory chapter of this volume Sabates-Wheeler and Feldman identify four components of social protectionthat have important implications for the migrants. These are:

(i) access to formal social protection – that is, social security and socialservices – in host and origin countries; (ii) portability of vested socialsecurity rights between host and origin countries; (iii) labour marketconditions for migrants in host countries and the recruitment pro-cess for migrants in the origin country; and (iv) access to informalnetworks to support migrants and their family members.

They argue that access to formal social protection mechanisms reducesthe vulnerability of the migrants.2 It may be pertinent to mentionhere that neither the Bangladesh state nor any of the host states ofBangladeshi labour in the Gulf have any formal social protection mech-anism in place for the migrant workers. In such situations where thesocial security provisions are virtually non-existent, the issue of porta-bility of rights necessarily becomes redundant. The third and the fourthcomponents of social protection (labour market conditions and infor-mal support networks) identified by Sabates-Wheeler and Feldman arevery relevant for the group of migrants with which this chapter isconcerned.

As will be seen from the field findings, labour migrants are often dis-advantaged in foreign labour markets ‘due to information asymmetriesand monopsony power of employers’. In some instances the policiesenacted and implemented at the national level by the host countries areinherently skewed against the foreign workers, while in other instancesweak enforcement of the laws create conditions in which recruiters,


employers and the intermediaries exercise arbitrary leverage over therecruitment and treatment of workers. Limited capacity and inadequatetraining, coupled with the lack of motivation of the labour attachés ofBangladesh in the host countries, contribute little to migrants’ effortsto seek redress through official channels. Under such circumstanceslabour migrants rely on ‘a range of kinship or patronage institutions andlocal or community based networks that can act as social safety nets’.The informal networks and groups provide crucial support to migrants,particularly those in irregular status, in the absence of formal socialprotection mechanisms.

2. Methodology

This chapter is based on primary information. Qualitative data was gen-erated through focus group discussions (FGD) and in-depth interviews.Ten FGDs with 15–20 returnee migrants were held in six villages andtwo municipal towns of the two study districts. In-depth interviews of57 migrants were conducted over a period of two months. Both cate-gories of migrants, regular and irregular, were covered under the study.Migrants who had served at least 12 months in the countries of des-tination constituted the bulk of the interviewees. Some were deportedby the authorities for their irregular status, while a few others were onvacation when they were interviewed.

2.1. Study sites

The study sites were located in the migration intensity districts ofComilla and Tangail. The study team noted that some specific charac-teristics of each of the study sites had some bearing on the migrationpatterns and practices of the area.

The people of Dhakargaon village, located in the Comilla district,have a long and diversified migration experience. One of them, along-term migrant in Saudi Arabia, plays a critical role in facilitatingmigration from the locality. The connectivity of the other two studysites, Baher Chor and Hasnabad, with the urban centres is very poor.Adverse work conditions in the villages push people to opt for migrationboth to urban areas within the country and abroad.

The study sites in Tangail were located in Kaloha village and KorotiaBazar. Kaloha village was noted for its handloom industry. Increasedcompetition from synthetic fibres and decreases in state subsidy sub-stantially weakened the industry. This led many in the trade to opt foroverseas migration. A special feature of the other study site, Korotia

266 Case Studies

Bazar, is that it hosts as many as 58 offices of informal facilitatorsof international migration. This overt presence of informal facilita-tors in the bazar is distinctively different from other regions, whereinformal facilitators prefer to work in secrecy and have little institu-tional face.

3. Deciding migrants

For deciding migrants, accessing information may become a major prob-lem. Securing basic information about the countries where jobs areavailable, the opportunities that exist in different sectors in respectivecountries and the skills that are in demand sometimes proves to bea very difficult task. Although government rules oblige the recruitingagencies to advertise availability of overseas jobs in the national dailies,this does not make much of an impact due to the prevalence of lowliteracy.

Field findings from the study show that none of the participants of theten FGDs or the 57 interviewees secured information through the formalchannels. In other words, at the initial stages of migration none of themhad any contact with formal recruiting agencies or had any knowledgeof the work opportunities abroad that were posted in newspapers. Whenasked why they were not in contact with government offices such as Dis-trict Employment and Manpower Offices for information, all expressedignorance about the existence of such a source.

It is in this context that one can explain the near universal depen-dence of deciding migrants on informal sources of migration informa-tion. The informal sources include dalals (informal facilitators), relativesand friends. The informal facilitators were returnee migrants, migrantswho come periodically on holidays, subagents of recruiting agencies orformer employees of recruiting agencies. In many instances these infor-mal facilitators are relatives of or known persons to aspirant migrants ortheir immediate family members. In both the study sites it was foundthat local Union Council members themselves were involved in facili-tating migration. One interviewee (54) candidly stated, ‘I had no optionbut to take the assistance of a dalal.’ His is a typical case where aspi-rant migrants are unable to secure information about migration at thegrassroots level.

I have lived abroad for more than ten years and have known manymigrant workers of Bangladesh. Their work permits were eithersecured from relatives or friends working overseas or through dalals.


Not a single person could migrate directly through recruiting agen-cies. One of my friends went to an office of a recruiting agency inDhaka. Before he could enter the office he was asked to name hisdalal. When he replied he did not have any, he was advised to contactthe agency through a dalal.

(Shukur Ali, 41, Dhakargaon, Comilla)

When interviewees were asked what made them trust the informal inter-mediaries, many said that their success in facilitating the migrationof others from the locality was the most important factor in trustingthem. A few others in Korotia Bazar of Tangail thought that the factthat the dalal was working for a reputed agency was a good enough rea-son to trust him. Although aspirant migrants lay their trust on informalsources, they have very little opportunity to cross-check informationprovided by them. When this issue was raised, Selina Akter (41) ofDaudkandi, Comilla, a female migrant, replied that she knew she wastaking a major risk and was prepared for the consequences. She statedthat she did not have any option to validate the information providedby her facilitator.

However, some of the deciding migrants were cautious in trustingdalals. Through their social network they tried to verify if the infor-mation provided by the dalals was genuine. To make sure that thedalals were providing correct information they contacted the familiesof persons whose migrations were facilitated by the dalals.

A major source of vulnerability for migrants has been the near absenceof any written contract with the employer/recruiter. Except for six casesthere was no such contractual arrangement. The terms and conditionsof employment (wage, overtime, health care, holiday, accommodation,food and other entitlements) were verbally communicated.3

After waiting for about a year I was told by the dalal that he securedan office assistant’s job for me with the Dubai municipal authoritieson a good salary. In order to be sure of the arrangement I asked himto give me a copy of the contract. He agreed to give me the same, butcame up with various pretexts and got away by not giving me a copy.When I insisted he gave me the choice that either I believe what hewas saying and go to Dubai or forget about going. I went to Dubaiand landed with the job of a cleaner of a hospital. My salary was 30per cent less than what was promised by the dalal.

(Abu Bakr, 41, Tangail town)

268 Case Studies

In a few instances where contracts were signed they were done at the lastmoment, just before the migrants boarded their flights. Most of thesemigrants reported that they could not look into the details of contractdue to shortage of time, nor could they get the contracts verified bypersons whom they could trust and who had some understanding aboutthese issues.

The respondents noted that they were deeply concerned about thehigh cost of migration. Cases were also found where migrant householdshad to sell assets (livestock, jewellery and land) to finance migra-tion. Some households became further steeped in poverty when theirmigration experiences turned out to be unsuccessful.

I suppose that the dalal understood that I was desperate to go abroad.He demanded money in excess of the amount agreed and I continuedto meet his demand by selling whatever things I had. The day he wasdue to provide me with work visa he disappeared. I could never tracehim again. I also heard that at least three others were cheated by him.

(Monu Sheikh, 37, Pakulla)

In the FGDs many participants noted that in their locality there werescores of examples where aspirant migrants fell victim to unscrupulousagents. In such cases those victims did not have any recourse to theformal protection of the state and as such their failure to migrate hadsevere implications for downward spirals into poverty.

In several instances dalals/recruiters demanded in excess of the agreedamount. The interviewees stated that they did not have any choicebut to agree to high migration costs. Otherwise, they would have toforego their migration dream. When migrants were asked why they didnot seek redress when excess amounts were demanded by the dalals orrecruiters most said that such action was unlikely to yield any desiredresult. Conversely, it was likely to result in the loss of amount they hadalready invested. Migrants resisted increased demand by agents in onlya few instances, by mustering the support of influential relatives.

It is noteworthy that there was not a single instance where a receiptwas issued by the recruiters or dalals for money transactions. Most of themigrants interviewed stated that they did not ask for such receipt as thedalal or recruiter was known to them. In other instances, when migrantsdemanded a receipt the dalal or recruiter declined to issue one. MofizMia of Pakulla, Tangail, said that his dalal asked to be relieved of theresponsibility to secure a work permit for him, and to find a replacementwho would agree to issue a receipt.


Failure to secure a receipt against the transacted money by decidingmigrants has severe implications for their protection. If the dalal or theagent does not succeed in sending the migrant abroad, in the absence ofany documentary evidence he or she cannot be held legally responsible.

3.1. Migrants’ initiative for their own protection

In the above section negative experiences of interviewee migrants werenoted. This section presents how some migrants worked out their ownways to meet the challenges posed in the pre-migration phase.

With regard to the absence of a written contract the only protectionthat some migrants secured was having one or two relatives, friends orneighbours as witnesses. Idris Khan (36) and Naren Sarker (41), bothof Dhakargaon of Comilla, viewed such protection to be more effec-tive if the local Union Council chairperson or members consented to bethe witnesses. Three interviewees reported that public representativesof their respective areas agreed to be witnesses to the verbal agreementwith the dalals.

There is a virtual absence of any formal source to finance migrationcosts. It is in this context that a variety of informal support sourceswere reported by the aspirant migrants. In several instances siblings,fathers and in-laws of aspirant migrants working overseas bore most ofthe costs. In other instances, migrants raised a good section of the costfrom local sources, and family members living overseas took care of par-tial costs. In at least nine instances friends and relatives arranged the visaand paid for it with the understanding that the migrants would pay backthe amount from their wages after they took up employment. In thesecases migrants arranged the cost of the airfare and facilitation expenses.In situations where dalals demanded more than the agreed amount,some migrants could secure the good offices of the village elders orthe local public representatives to offset the pressure of dalals. In mostsuch cases, however, the aggrieved party was encouraged by the medi-ators to pay at least a part of the additional amount as a compromisearrangement.

4. Arrived migrants

In many instances newly arrived migrants did not face any problems atthe airport. After clearing immigration either their relatives or friendsor the representatives of the sponsoring company or kafeel met themat the airport. However, in several instances migrants faced major hard-ships soon after their arrival in the country of destination. There have

270 Case Studies

been a number of instances in which they had to wait for long hours foremployers to meet them at the airport. In the case of Manu Munshi (33)of Baher Chor, Comilla, it took three days. Lack of support at the air-port in contacting the employers was identified to be a major problem.The absence of any interpreter’s service caused the migrants difficulty incommunicating with the immigration officials. Some reported that thelack of this service contributed to them facing unnecessary harassmentwhen they arrived in the destination country.

An overwhelming majority of the interviewees reported that theirpassports were taken away by their employers soon after their arrival inthe destination country, in most instances at the airport. None of themhad any photocopy of the relevant pages of their passports, and theywere thus were not left with any document to prove that their stay in thecountry was bona fide. Migrants who worked for established companiesstated that they were issued with photo identity cards instead. Manyreported that, as well as having their passport taken away, they werenot allowed to keep the copy of their iqama (work permit). Some wereexplicitly warned that without those documents they might be arrestedby members of law enforcement agencies and thus should restrict them-selves to the company premises. When asked what they thought of thepractice, many noted that initially they did not find anything wrongin it, but subsequently realized it was a method to keep them tied tothe employer. They thought the documents were seized so that theemployers can have a free hand in treating them the way they wishedwithout having to worry about migrants reporting abusive practices tolaw enforcement agencies. A few stated that they disliked the practicefrom the beginning but did not want to raise the issue with the employ-ers for fear of upsetting them. Asgar Ali (22) of Hasnabad, Comilla, wasrefused a copy of the iqama even though he asked for it.

Seizure of passports and work-related documents by the employersseverely compromises migrants’ protection. Despite having the nec-essary papers, their inability to produce documentary evidence whendemanded to do so by law enforcement agencies subject them to pros-ecution and even deportation. The inability to retain control over theirpapers makes even the regular and the bona fide migrants insecure. Italso restricts their freedom to move to alternative work or seek redressagainst errant employers and recruiters.

In only a few instances did the interviewee migrants report that theyreceived pay and other entitlements according to their contract. Threeof the six migrants who had signed contracts before their departure saidthat they were provided with fresh contracts with lower pay and longer


working hours than stipulated in their original contracts. Bakul Begum(33) of Dhakargaon of Daudkandi stated that under the revised contractalmost half of her salary was deducted for accommodation and food.She insisted that, in the original contract, the company was to pay forthose expenses. She claimed that she tried to convince others in thecompany to report this violation of contract to higher authorities, butothers advised her to accept the new terms, which she did reluctantly.The original contract of Saiful Islam (45) of Hasnabad, Comilla, stipu-lated that he would get 1400 dirhams per month. In the fresh contracthis salary was fixed at only 520 dirhams. He signed the new contractbut within months quit his job and became an irregular worker. In fiveinstances migrants all signed the substitute contract against their wishesas they did not have any choice. The lone migrant who refused to signthe revised contract was threatened with an exit visa. Subsequently, hecomplied.

This was another major problem that the interviewee migrantsreported. In all instances, the jobs they were placed in were physicallymore demanding and financially less well paid than the jobs they werepromised in the first place. Afaz Miah (29) of Korotia Bazar, Tangail, waspromised a government job but was made to work as an ‘iron boy’ ina laundry. Rafiq Sheikh (34) of the same locality was promised a clean-ing position at the airport, but was placed in the agricultural field. AltuKarim (34) of Kaloha village, Comilla, said that he was promised thejob of a loader, with a salary 600 riyals per month, but ended up askitchen helper in a restaurant on a salary of half of the agreed amount.Shona Mia (44) of Dhakargaon, Comilla, was promised a regular job as aconstruction worker. On arrival he was made to work for an outsourcingcompany without any regular wage. He said that he barely had anythingto eat for the first 19 days before being sent to work at a constructionsite. Only in one instance did an interviewee, Hossain Ali (24), end upwith a better job. He was recruited as paint worker but, as he had a driv-ing license, within a year he was able to get a job driving a construction-related vehicle. The new job secured him a much higher salary.

Not receiving due wages was a major problem for many migrants.Although quite a few interviewees secured their iqama within days oftheir arrival, for many the delay in issuance of iqama made them sit idlefor weeks and months. During this period they were denied any wages.In most instances the kafeel bore only their food expenses.

Amir Mian (29) of Korotia, Tangail, stated that he worked for threemonths for the kafeel who sponsored him, but did not get any paymentfor his labour. Shapon Rashid (34) of Hasnabad, Comilla, stated that his

272 Case Studies

kafeel virtually treated him as a bonded labourer. He claimed to havereceived only a month’s salary after six months of work and thus hadto flee his job. At least 17 others reported that they experienced non-payment of wages at some stage of their employment overseas. Experi-ences of non-payment for overtime work were more rampant. Fourteenof the 57 migrants claimed that instead of eight hours a day they hadto routinely work for 10–12 hours without being paid any additionalamount. One migrant reported that the company did not pay the foodallowance that was promised to him. Retention of wages was common.One interviewee stated that every six months his company deducted 200riyals from his salary for uniform and accessories. Many migrant workersreported that they did not get their wages at the due time. MohammadNazrul (40) of Pakulla, Tangail, stated that instead of paying him everymonth, the kafeel paid him every two months, retaining a month’s wage.In the cases of three other interviewees as much as three months’ wageswere retained. Sharif Sarkar (29) of Dhakargaon, Comilla, claimed thatthe factory foreman demanded part of his wages every month. He fur-ther claimed that he was threatened with an exit visa when he refusedmake such payments to the foreman. Mohammad Shahjahan (28) ofPakulla, Tangail, was paid 600 riyals as salary, although he was made tosign receipts to say his salary was 1000 riyals.

Restriction of movement was a major problem for many migrants.Most of the interviewees reported that delay in securing iqama severelycurtailed their freedom of movement within the country. Lack of accessto their own passports and work documents made them vulnerableto arrest. If an alien fails to produce proper documents to membersof law enforcement agencies they are deemed to be irregular migrantsand thus are subjected to arrest and prosecution. Shafiq Rahman (33)of Baher Chor, Comilla, stated that he was a cleaner at a restaurant.He was strictly advised by his employer not to venture out of hisworkplace and living quarters. He felt that he was living the life of aprisoner.

As many as 17 interviewees reported that their kafeels did not havework at their disposal and thus they were sent to other employers ona temporary basis. Before migration they were all promised secure jobsand only found out about the temporary nature of their employmentafter arrival.

A good number of the interviewees stated that they had to work forlong hours. The ‘iron boy’ of the laundry, Afaz Miah (29) of KorotiaBazar, Tangail, stated that he suffered from severe backache as he hadto keep standing for 8–10 hours at a stretch. Lorry driver Nazrul Islam


(41) of Tangail town said that he had to work from dawn to midnightwithout a break, even on weekly holidays. The migrant who worked asa loader claimed that he was contracted to handle light packages but inreality he had to physically lift items weighing 40–50 kilograms quite anumber of times each day.

The lack of adequate air conditioning in work premises was alsoreported by a few migrants. They stated that they had to work for longhours in the heat as the cooling system of their factories often mal-functioned. Verbal abuse by employers or supervisors was reported by anumber of migrants and two claimed to have been physically abused.It is interesting to note that two of the interviewees (Monir (33) ofGoshaibari, Korotia, and Parvez Ali (28) of Kaluha, Tangail) were partic-ularly critical of their Bangladeshi supervisors. They felt the supervisorsconcerned did so to prove that they were ‘not biased’ to their fellowBangladeshis.

I put up with the maltreatment and the long hours only becauseI knew I could not earn the quarter of the amount at home. Thatkept me going.

(Nazrul Islam, 42, lorry driver in Saudi Arabia)

Living conditions of the migrants varied significantly. Workers who wereemployed by established companies generally expressed satisfactionwith their living arrangements. In these cases the number of residentsper room was low, there were air conditioners and the resident-toiletratio was good. In a few instances they expressed their dissatisfactionover the quality of food supplied. However, many others expressed theirgrievances about living arrangements. Abdul Baten (23) of Dhakargaon,Comilla, stated that 700 workers were staying in company-providedaccommodation which had only six baths and six toilets. He furtherstated that there was no air-conditioning system and that the numberof fans provided in the rooms was quite inadequate. Likewise, SamadMia (35) of Pakulla, Tangail, resented that he had to share a smallroom with 17 others. A similar experience was noted by Habibullah (37)of Korotia, Tangail. He informed that his 21 square metre room hadto accommodate 12 people. ‘There was no privacy’, he noted. Livingconditions were particularly harsh for the migrants who worked as agri-cultural labourers or who tended livestock, observed Shapan Khan (34)of Dhakargaon, Comilla, and also a few other interviewees. They saidthey had to live in tents which were not good enough quality to pro-tect them from the heat during the day or the cold at night. The sedan

274 Case Studies

driver migrant reported that he was asked to sleep in the car when hisemployer took him to outstations for work.

Sometimes migrants collaborated with each other to mitigate theirliving problems. Parvez Ali (43) of Kaluha stated that when they askedfor a refrigerator his company arranged for one. In another instance theresidents of a dwelling contributed to a fund to buy a refrigerator whenthe company turned down their request.

The level of access to health care also varied among the intervieweemigrants. Migrants working for big companies, particularly the onesrun on a corporate ethos, received a good degree of health coveragethat included free hospital coverage and medicines. Workers were alsoentitled to sick leave. The health care facilities of other employers weregenerally poor. Suruj Kabir (41) of Baher Chor, Comilla, who worked ata packing plant, stated that when he cut his hand his company refusedto take him to the hospital as he was not covered by a health insur-ance plan. Another worker, Mohammad Shakeel (34) of Daudkandi,Comilla, sustained injury to his head while working at a constructionsite. Though he had to bear the cost of the treatment, he felt thatthe company was generous in granting him 15 days leave for recupera-tion. Aminul Islam (32) of Kaluha, Kalihati, claimed that his employerdeducted three days of leave when he failed to report for duty for oneday due to illness. He added that it was the supervisors who were mostlyresponsible for such harsh treatment of the workers. Those in the goodbooks of the supervisors got paid sick leave, he stated.

Enjoyment of the weekly holiday was a rarity for many intervieweemigrants. In most cases, migrants who worked for individual kafeels as adriver, gardener, agricultural worker or sheep rearer, hardly enjoyed anyweekly holidays. Others working in the formal sector (in restaurants,clinics or hospitals) as a waiter, helper or cleaner, were also deprived ofsuch holidays. They said that it was only during the Eid holidays thattheir employers granted them reprieve from work for a day or two.

In several instances workers were not granted the annual leave thatwas due to them. Aminur Khan (24) of Selimpur, Tangail, informedthat he was upset when his employer denied his application for twomonths’ leave. Later, he gained courage and threatened to resign ifhe was not granted the leave that fell due. As he was a good worker,his kafeel could not risk losing him and eventually granted him sixweeks’ holiday. Habibullah (44) of Korotia, Tangail, quit his job afterthree and a half years when he was refused his entitled long leave.The employer of Mohammad Nazrul (34) of Tangail town retainedpart of his outstanding salary and half the cost of the airfare to make


sure that he returned. As his father did not allow him to go back tohis country of employment he lost the amount retained by his formeremployer.

In most of the Gulf states the work visas issued to migrants are tied toa specific kafeel. The migrant needs clearance from his or her sponsoringkafeel if he or she wishes to move on to another kafeel. However, thisclearance may not be readily available. Even in cases where sponsoringkafeels are unable to provide the worker with a job, they would onlyconsent to such requests against the payment of a transfer fee. FarukHossain (41) of Chor Kagmara, Tangail, changed his work permit fromvegetable packer to salesman in a departmental store by paying a sumof 1000 riyals to his sponsor. In another instance Idris Khan (23) ofDhakagaon, Comilla, paid as much as 5000 riyals to change his spon-sor. Sharif Sarkar (35) of the same locality claimed that he had to payas much as five months’ wages to change his kafeel. In another case thesponsor demanded a fee of 4000 riyals to grant his worker permissionto change employer. The worker refused to pay that amount, quit hisemployer and became an irregular migrant, as he did not have an iqamaand passport.

A major barrier to the redress of labour abuses is the sponsorship(kafala) system by which a migrant worker’s legal residency istied to his or her employer, who serves as a ‘sponsor’. Migrantworkers can only transfer employment with their sponsor’s con-sent. . . . Sponsorship traps workers to abusive situations, includingin situations of forced labour and blocks their access to means ofredress. If an employer withdraws sponsorship, workers who flee abu-sive workplaces can be arrested and deported for being out of statusin the country.

(HRW World Report 2010)

Most of the interviewee migrants stated that there was no grievanceredress mechanism available to them. The kafeels’ tendency to rely onthe manager or the supervisor thwarted any move for in-house settle-ment of disputes. Not having access to iqama and other work-relateddocuments to support claims is the single most important obstacle inthe way of meeting grievances at the official level. Shahjahan (26) ofPakulla, Tangail, stated that that he would have preferred to approachthe labour court to make a claim against his employer for non-paymentof his due wages. But he could not do so as he did not have his iqamawith him to produce before the court. Only in one instance when an

276 Case Studies

interviewee complained to the police that his employer had failed toprovide him with a job, did the police intervene. Though the employersubsequently agreed to fully refund his migration cost in a few instal-ments, in reality he paid back about 60 per cent of the amount. It wasa widely held view by the migrants that the law enforcement author-ities in many of these destination countries are corrupt. The practiceof apprehending irregular migrants and releasing them on payment ofcash was another commonly shared perception. Amir Shikder (33) ofKorotia, Tangail, stated he was caught by the police four times and paidon average 100 riyals to secure his release. It is interesting to note that ina couple of instances irregular migrant interviewees claimed that whenthey decided to return to their home country, they bribed the policeto ensure that they were caught.4 The arrested irregular migrants arereturned to their home countries by the Saudi authorities at the latter’sexpense.

4.1. Networking of migrants

A number of interviewees stated that in the absence of formal supportservices they relied on each other. Many stated that without the sup-port of fellow Bangladeshi migrants it would be almost impossible toadjust in the alien environment of the destination countries when theyfirst arrived. Such support was particularly crucial for those migrantswho had to wait for their iqama to be processed and did not have anyresource to maintain subsistence. Fellow migrants extended their sup-port through providing shelter to those whose kafeels failed to providethe same.

Many interviewee migrants worked out alternative strategies to sup-plement their income. Alam (31) of Tangail town stated that he knewsewing, bought a sewing machine with his savings and worked as a tai-lor in his free time. He made shirts and trousers. His fellow migrantswere his clients. Another interviewee, Monir (34) of Goshaibari, Korotia,originally an unskilled migrant, picked up electrician’s skills. He boughta bicycle to move around the neighbourhood and provided services as ahandyman, earning an extra amount averaging 1000 riyals per month.He said that fellow Bangladeshi workers introduced him to many of hisclients. Abdul Hakim (42) of the same locality stated that after he fin-ished his regular work for his kafeel, he worked as a shop assistant in alocal market. In all these cases, their respective kafeels gave their con-sent. Otherwise, they ran the risk of being reported to the police bythe kafeels, which could lead to prosecution, loss of iqama and eventual


deportation. In the case of Monir Hossain of Kaluha, Kalihati, the kafeelencouraged him to learn driving and helped him secure his licence. Thisled to a major increase in his salary.

A large number of interviewees acknowledged receiving care fromfellow migrants when they were ill. Many reported that in adverse situ-ations (such as a dispute at the workplace or maltreatment by kafeels orsupervisors) the counselling of other migrants restrained them from tak-ing any drastic action. Financial support by fellow migrants was crucialfor many. In situations where migrants were unemployed or were notpaid their due wages others provided them with loans to send home. Inall instances those were interest-free loans. Aminul Islam (30) of Kaluha,Tangail, said that a fellow migrant stood as a guarantor for the kafeelof another migrant when the latter asked for leave to visit his family.Another interviewee stated that four of them pooled resources to buy amobile phone so that they could get in touch with their families.

Most interviewees responded negatively when asked if they had anycontact with the highly skilled migrant Bangladeshis. They felt thatthe latter group preferred to maintain a distance from migrant workers.A few migrants stated that they were active supporters of certain polit-ical parties at home. When their party organized an event, generallylinked to a visit of senior members of the party from Bangladesh, theyattended those events. Others, however, felt strongly against such parti-san activity in a foreign land. They felt that it divides the Bangladeshicommunity and does not achieve anything fruitful for the migrantcommunity.

A few of the interviewee migrants stated that they belonged to com-munity associations. One claimed that he belonged to a group of24 Bangladeshi workers who paid 20 riyals per month to a commonfund. Any member could take a loan from the common fund with theconsent of other members. Borrowers did not have to pay interest ifthe loan amount was under a stipulated sum and paid back within cer-tain duration. He stated that quite a few members have secured loansfrom the fund and repaid the amount, mostly within the due time.A few of the other migrants said that they are involved in informalassociations. One reported that members of his informal associationcontributed 5 riyals per month to a common fund. That fund is usedto celebrate religious festivals or national events (Victory Day, Inde-pendence Day and Mother Language Day). Not all migrants have hadgood experiences of such collective initiatives. Mohammad Kamal (34)of Hasnabad, Comilla, stated that he, along with a few friends, tried cre-ate a common fund to meet emergency expenses and began depositing

278 Case Studies

5 dinars a month. However, that initiative failed as those in charge ofmaintaining the account embezzled the entire amount. Having beeninvolved in several such associations for a long time, Alam Shaikh (45)of Pakulla, Tangail, observed that these organizations are set up with alot of enthusiasm and good faith, but in most cases things go wrong andthe organizations wither away.

4.2. Migrants in irregular status

It took Shakeel (33) of Dhakargaon, Comilla, as many as three years’earnings to repay the cost of migration. After his contract expired hedecided to stay, albeit with irregular status, to save enough so that hecould come back home and do something meaningful. Shakeel workedfor three more years and was cautious in his movements so that hedid not get arrested. When he felt he had enough savings he reportedhimself to the authorities and was returned home.

There have been instances where the desire to take on fresh challengesinduced migrants to have irregular status. Idris Khan of Hasnabad,Comilla, was doing well and had a good kafeel. After listening to othershe became excited with the idea of going to Europe. While attemptingto cross the Turkish border with Syria, Idris was arrested by the Turkishauthorities. He was subsequently sent back to Saudi Arabia, served hissentence for nine months and was deported to Bangladesh.

The migrants with irregular status were likely to be subjected to vari-ous kinds of ill treatment. They were particularly hard hit in getting theirdue wages. One irregular migrant noted that he was not paid a singledime, even though he worked for nine months, and another was deniedwages of five months. Both noted that there was very little they could doas they did not have proper iqama. The kafeels ‘took full advantage of thefact that we did not have valid documents’, they noted. Another inter-viewee, Ahsan Madbor (35) of Hasnabad, Comilla, worked in a companyfor six months and was only paid for two months. If such pattern of pay-ment continued, Ahsan said he would explore the option of leaving thecompany.

For irregular migrants, accessing health care services become particu-larly difficult. Even if they can pay for the treatment their inability toproduce the iqama during registration at hospitals or private practition-ers deter them from being able to avail themselves of such services.

4.3. Role of Bangladesh missions

There was a general impression among the interviewee migrants that theBangladeshi missions in the destination countries were not providing


desirable services to the migrant workers. A good number of migrantsclaimed that under the local laws of destination countries, foreign work-ers could only lodge complaints against their employers through theirrespective embassies. They felt that Bangladesh missions were not asproactive as their Filipino or Sri Lankan counterparts. One migrantworker stated that there was a general reluctance among embassy staffto take labour matters up with the officials and this discouraged work-ers from reporting cases of irregularities. Others believed that if theembassy officials were sincere in pursuing cases against errant work-ers they could secure redress for the workers. Sometimes that interestis generated through financial inducements. Mustafa Kamal (33) ofKaluha, Tangail, claimed that he paid a lower functionary of the embassy200 dirhams and secured a compensation of 1100 dirhams from hisemployer. Aminul Islam of the same locality stated that he, alongwith 15 other Bangladeshi workers, registered a complaint with theembassy but that the embassy did not take any action against the errantemployers.

Some migrants reported that the embassy took more time than othermissions in providing routine consular services such as renewal of pass-ports. Interviewees observed that many migrant workers were forced tochoose the expensive one-day service, although they could not affordit. They felt that regular service was even more expensive as it entailedat least a second visit, costing time and money. A few others stated thatinformal facilitators linked to the embassy sometimes took advantage ofthe migrants’ lack of time. They provided a one-day service against thepayment of fee that was added to the regular charges. Three migrantsnoted that although they had put in specific complaints about embassyservices in the complaint boxes that are kept at the entry of Bangladeshmissions, these were not acted upon.

5. Return migrants

A good number of interviewees stated that the absence of investmentopportunities had led to the loss of savings that they once had. In manyinstances they had to pay to meet the various demands of family mem-bers, relatives, friends and neighbours. These ranged from requests forfinancial support for distress situations, such as medical treatment andfuneral expenses, to payment of annual fees for children’s education,dowry and marriages. Migrants also felt the pressure to provide loansor invest in others’ enterprises against their wishes. Appeal for contri-butions for the building and renovation of schools, mosques, roads orculverts were also common experiences of many interviewee migrants.

280 Case Studies

Yousuf Molla (34) a migrant from Daudkandi, Comilla, stated that hefaced tremendous pressure from his brother-in-law to invest in the lat-ter’s rice trade. Rahim Shaikh (45) of Korotia Bazar, Tangail, said thathis family was very upset with him as he did not spend enough moneyduring the Eid celebrations that took place soon after his arrival backhome.

I sent all my earnings to my brother believing that he was makingprudent use of my hard earned money. On return home after fiveyears I found most of the amount was wasted and the small plot ofland that he purchased in the local bazar was recorded in his name.I have nothing left for my family. Although before coming home Iwas planning to retire and spend time with my family, now I have nooption but to look for another job overseas.

(Tota Mia, 45, Tangail Sadar)

In a socio-economic milieu where social and political contacts play amajor role in enterprise development, lack of knowledge of local powerdynamics and lack of contact with the locally influential economic andpolitical elite turn out to be major hindrances for migrants’ integration.Lack of knowledge about opportunities for investment is another majorproblem that migrants face. A recent survey5 of enterprise initiativesof returnee migrants found that some of the returnee migrants haveformed agricultural and fish farming cooperatives, others have set upmarket places. They reported that the banks have not yet developedfinancial instruments and programmes to cater for their investmentneeds, nor were there any private initiatives to finance such schemes.

In many instances migrant workers learn important skills when work-ing overseas and wish to utilize those skills upon return home. But manytimes these workers find it difficult to get suitable jobs. This problemcan be mitigated by creating a database of the returnee migrants withinformation about their skills. That would help prospective employ-ers recruit the workers by accessing the database. Three intervieweesstated that they learnt new skills when they were abroad and wantedto work in the related field. But they had no idea where they could findsuch jobs.

6. Conclusion

This chapter noted that the Bangladeshi workers who migrate to theGulf states encounter a plethora of problems at every step of their


migration process. Lack of credible sources of information on workopportunities abroad make the migrants dependent on informal inter-mediaries. The role of these intermediaries is further augmented by themigrants’ lack of access to formal recruiting agencies. The simplificationin procedures for issuance of passports has reduced migrants’ relianceon intermediaries, although this remains quite high for securing workpermits, leading to increases in the cost of migration. The absence offormal sources of loans to finance migration, the refusal of agents to pro-vide copies of written contracts in advance and delay and uncertaintyin flight schedules are some of the other problems that Bangladeshimigrants face once they decide to migrate.

The vulnerability of migrants was found to be quite acute in thecountries of deployment. Migrants experienced the seizure of personaldocuments by the employers or recruiters, non-placement against jobspromised, non-payment and irregular payment of wages, poor workingand living conditions, lack of access to adequate health care facilitiesand prohibition on changing employer. The absence of effective redressmechanisms in the form of accessing courts, and the poor services pro-vided by the Bangladeshi missions further weaken migrants’ position inalien lands. The chapter also documented that despite all odds a num-ber of self and collective protection mechanisms have been developedby the migrants.

In the final section, the problems of returnee migrants were docu-mented, including loss of savings, lack of investment opportunities andlack of information for utilization of skills. The study also notes thatthere is no worthwhile government programme or non-governmentalinitiative for returnee migrants in Bangladesh. Most of the govern-ment activities are geared towards the pre-migration stage. However, oflate, in policy-making circles there has been some recognition of theimportance of reintegration issues.

In this chapter the problems and vulnerabilities of Bangladeshimigrant workers serving in the Gulf states were examined. It was notedthat, while social protection as an agenda is being increasingly espousedfor reducing the vulnerability of weaker sections of the communities,and has become an important part of the development discourse, themigrant workers of Bangladesh, particularly those working in the Gulfstates, continue to be deprived of any such protection. Although man-agement of risk and improvement of livelihood options are often theprincipal guiding factors that shape their migration decisions, theseworkers are in dire need of both formal and informal forms of socialprotection.

282 Case Studies

Notes

1. The author acknowledges the support of Wahidur Rahman, Makfie Farah,Motasim Billah, Niaz Ahmed and Rigan Chakma, members of the YoungResearchers’ Forum of Refugee and Migratory Movements Research Unit ofDhaka University, in conducting the fieldwork of the study.

2. The formal protection mechanisms include access to health care benefits,long-term social security benefits like old-age and disability benefits, andshort-term benefits like social assistance, maternity and unemployment bene-fits, family and housing allowances, as well as public housing and education.

3. In many instances the terms of the verbal contract were completely ignoredwhen the migrants arrived in the destination countries. This issue will bediscussed in the next section.

4. One claimed to have paid 300 riyals, another 700.5. The survey was conducted by the Refugee and Migratory Movements Research

Unit on Migrants’ Enterprise Development Initiatives in the four districts ofBangladesh (2009–2010).

Reference

International Labour Organization (2010) International Labour Migration: A Rights-Based Approach (Geneva: International Labour Organization).

Index

accessconstraints, 28, 67, 72–4, 83structures, 23–4, 29, 67–8, 81, 83–4,

186–95, 199, 204–6data rules, 23, 83eligibility rules/criteria, 67, 72, 79equality of, 191legal, 23physical, 24, 84, 183–207policies for inclusion, 27political, 24, see also access‘queue’ rules (‘counter’, ‘gate’ and

‘line’), 79regulating, 24rules for exclusion, 54, 79–83rules for inclusion, 79–83to services, 38, 49, 99, 143,

149–50, 185, 203, 223discrimination/discretion, 73exclusion, 10, 25financial, 113information, 22, 65, 266language, 80, 153legal, 23network-based, 64relational, 79–81resource availability, 10, 29, 61,

65, 68accommodation, 7, 77, 155, 173, 178,

180, 251, 267, 271, 273actuarial fairness, 21, 94administration

procedures, 23, 287public service, 160

AfricaAfrican union, 47burial society, 23, 252hometown association, 23, 70–1SADC (Southern African

Development Community), 30,107–8, 118–31, 133–4

Sub-Saharan Africa, 78, 100, 102,105–6

see also South Africaagreements and arrangements

bilateral, 23, 102, 104–5, 108–10,129–30, 134

cross-border, 128–9multilateral, 101–2, 107, 123, 125,

128–9, 134Arab states

Arab society, 41Islamic law, 41Kurdish minority, 47The League of Arab States, 47see also Gulf states

Asia, 19, 99–100, 102, 105–6, 114,143, 235, 244, 262

ASEAN (Association of SoutheastAsian Nations), 100

asset accumulation, 15, 254asylum

claims, 27, 149refused asylum seekers, 28, 45,

145–6, 152–8seekers, 6, 27–8, 30, 41, 45, 50, 117,

120, 122, 124, 126, 132, 144–7,149, 152–60

see also refugees

Bangladesh, 8, 105, 264–6,278–82

Bangladeshi migrants to the Gulfstates, 31

bank account, 74, 245benefits and allowances

disability, 21, 94, 99, 101, 282family, 94health care, 21, 94, 99, 101, 110,

149, 282housing, 21, 282maternity, 101occupational, 111

283

284 Index

benefits and allowances – continuedold age pensions, 14, 21, 64, 94–5,

99, 101–3, 108, 111, 248–9social assistance, 21, 94, 101, 107,

282unemployment, 21, 94, 99, 282

bilateral migration stocks, 9, 232Bommes, M., 16–17, 27, 211borders

control, 25, 38, 47as regulatory sphere, 38territorial, 24, 86, 96, 175

Caribbean, 100–1, 106CARICOM (Caribbean Community),

101–2, 107Castles, S., 3, 15, 18, 44, 173children

childcare, 18, 65, 71, 77, 236, 251‘transnational caregiving’, 222–5

Chile, 38, 114citizenship

in the ancient world, 41citizens, 14, 16–17contemporary, 29, 43–9denial of, 44, 47equality of, 43exclusions, 36–57expatriate citizens, 38full, 36, 43historical discussions of, 42for migrants, 5, 25, 36, 199national, 26, 47non-citizens, 4, 7, 16–17, 24–6,

28–9, 36–7, 40–3, 49–57, 68,74–5, 96, 98, 120–1, 123–5,127–8, 131–3

as a political tool, 40in the pre-modern world, 40–3presence, residence and, 37–43‘probationary citizens’, 48rights, 26–8, 36–57, 175, 186, 199,

204, see also rightscivil liberties, 25civil society

activism, 70activists, 32, 174civil society organizations (CSOs),

25, 68–9

class, 8, 16, 18, 29, 43–4, 57, 66–7, 70,80, 144, 173–6

class solidarities, 16clientalism, 18clientelistic relationships, 20, 75collective struggle, 17collective voice, 68–9community, 3, 26, 30, 49, 65, 69,

77–8, 85, 91, 101, 107, 117, 120,149–50, 156, 165, 172, 175, 206,233, 236, 252, 265, 277, 822

co-nationals, see nationalityconflict, 6, 9, 19, 25–6, 32, 80–1,

120–1, 153, 175, 179, 198, 259violent, 6

credit, 65, 79, 113, 170, 234, 239criminal activity, 8, 150, 158cross-border

movements, 119, 238recruitment, 236traders, 114see also agreements and

arrangements, cross-border;migrant(s)

data on migration, 232decommodification, 16–17de Haan, A., 80, 85, 172demographic shifts, 9development discourse, 20, 281development effects, 93Development Research Centre (DRC)

on Migration, Globalisation andPoverty, 32, 103, 259

disadvantagechronic, 19disadvantaged/subordinated

groups, 17ethnic minorities, 4, 16–17foreign workers, 17, 111, 122,

126, 264, 279women, 16

economic, 4, 19–20migrant-intensified, 10–11,

186, 197political, 4structural, 19see also vulnerability/vulnerabilities

(of migrants)

Index 285

discriminationracism, 26, 44, 51, 56,

125, 159stigma, 81–2, 184, 224xenophobia, 51, 54, 112, 120,

159, 185, 262distribution systems

market-based, 72–6network-based, 70, 72non market-based, 63–5, 67–8, 72,

76, 80, 186duties of allegiance, 44

economicderegulation, 45disadvantage, see disadvantage,

economicglobal economy, see globalisationgrowth, 14, 18–19liberalism, 25, 143, 169–70,

193management, 121recession, 14, 169

EcuadorConstitution, 210, 214diaspora, 218SENAMI, 215, 218–19, 228

educationqualifications, 240systems, 97

elderly, the, 13, 81, 190, 228Elson, A., 66, 84employers

decent, 112exploitation by, 119monopsony power of, 22,

95, 264obligations (legal and moral), 24,

36, 50, 52, 125, 129–30, 156,158, 221–2

policing, 25employment

casual sector, 170, 180–1informal sector, 10, 75, 110opportunities, 77, 170, 179registered, 176–8self-, 234see also labour; work

empowerment, 23, 78

entitlementconfusion, 4formal, 7, 61–2see also benefits and allowances

environmental hazards, 12–13equality of treatment, 98, 108, 125–6,

129see also access, structures, equality of

Esping-Andersen, G., 15–18, 32ethnic

‘incorporation’, 8minorities, 4, 16–17persecution, 6

ethnicity, 8, 13, 44, 66–7, 71, 80, 170,175–6, 185

European Union (Europe/EU)EU nationals, 98–9European enlargement, 126Lisbon Treaty, 47Maastricht Treaty, 47

exclusionary processes, seediscrimination

exclusion, see access; socialexclusion

exploitation, 10, 12–13, 55, 57, 95,112–13, 119, 175, 233, 243

see also employers, exploitation by

faith-based organisations (FBOs),68–9

faith communities, 278family, the role of the, 16Feldman, R., 3–32, 36–57, 143–60,

180, 184, 264feminist economics, 66food aid, 63–4, 68

Geddes, A., 16–17, 27, 211gender relations, 16

dimension of migration, 4Geneva Convention, 27, 50, 53GFMD (Global Forum on Migration

and Development), 101, 109–10ghettos/ghettoisation, 10, 45‘global’ cities, 4globalisation, 32, 103, 259governance, 19, 201guestworker model of labour

migration, 25

286 Index

Gulf states, 31, 262–4, 275, 280–1GCC (Gulf Cooperation Council),

99, 103, 114see also Arab states

healthcare, 7, 21, 37, 44, 47, 51, 64, 74–5,

77, 80, 96, 99, 101, 103, 110,112, 124, 143–60

doctors, 75, 97, 159general practice doctors, 75insurance, 18, 21, 51, 65, 71, 74–5,

94, 100, 132, 248–9, 274midwife, 78, 153public, 51, 75, 96, 99, 158–60

services, 51, 99services, 7, 30, 51, 75, 99, 147–8,

152, 159, 224system, 21, 95, 153

hierarchies, 8Holzmann, R., 91, 94, 102, 183, 194,

235hometown association, 23, 70–1host countries, 21, 24, 91–2, 94,

99–100, 102, 114, 234, 236, 238,249, 264–5

see also migrant(s), receivingcountries

housing, 7, 12, 21, 28, 44, 53, 71,74–5, 85, 94, 99, 156, 204, 248–9,251, 282

public, 21, 94, 282see also accomodation

HRW (Human Rights Watch), 263–4,275

Human Development Report (HDR),8, 91, 180

human trafficking, 119hunger, 6, 98, 199

identities, 6, 17, 170, 176‘illegal’ migration, see migration,

‘illegal’ILO (International Labour

Organization), 54–6, 92–3,98, 101, 108, 114, 124,126–7, 135

IMF (International MonetaryFund), ix

immigrants, 25, 27–8, 39, 44, 74–7,83, 94, 99–100, 107, 110, 128,146, 221–2, 234

naturalization of, 28immigration

context, 117–19control agencies, 266–7control, 4–5, 25, 143–4, 160, see also

borders, controllaws, 48, 72–3, 97, 122rules, 144, 150status, 7, 74, 121, 124, 145, 151–2,

155inclusion, 11, 17, 26–7, 69, 79–83,

176, 218, 226see also social inclusion

inclusive development, 23incomes

growth, 14, 19loss, 93transfers, 19, 235

Indiacaste, 170, 176Government of Delhi, 24, 84,

187–8, 190, 193, 198, 202Kerala, 164Public Distribution System (PDS),

30, 76–7, 83, 183–207ration cards, 24, 84, 187–8, 192–5,

197–203, 205–6informal labour market, see labour,

marketinformal networks, see networksinformation

asymmetries, 22, 92, 95, 236,264

blockages, 4resources, 265–7

infrastructure development, 173institutions

international, 43structures, 16, 171transnational, 43

insurancehealth, 18, 21, 51, 65, 71, 74–5, 94,

100, 132, 248–9, 274unemployment, 121, 130, 132see also national insurance

Index 287

intermediaries‘agents’, 70‘fixers’, 70, 201‘representatives’, 69, 196

international courts, 38international laws, see lawinvestment, 31, 48, 82, 113, 211, 213,

215, 217–19, 221, 223, 232–60,279–81

IOM (International Organization forMigration), 134–6

irregular migrants/migration, seemigrant(s), irregular

IsraelPalestinian Israeli citizens, 17Zionist exclusionary

policies, 17

kinship, 8, 22, 85, 170, 172, 175,179, 265

see also networks, kinship

labourcontracts, 22–3, 250

non-permanent, 23permanent, 22–3

market, 14, 17–18, 21–3, 25, 32,44–5, 65–6, 71, 74, 91–5, 98,103, 112–13, 118–20, 123, 128,194, 203, 236, 239, 241–3, 245,249–50, 256, 259, 264

informal labour market, 22, 91,95, 98, 203, 236

national labour market, 22, 72,250

policies, 22, 95, 239, 241temporary, 25

language deficits, see access,language

Latin America, 19, 100–1, 106, 213,215, 220–1, 227

lawinternational, 8state, 8

legal rights, see rights, legallegal systems, 8, 123, 130, 133livelihood/s

strategy, 111, 120, 180local authorities, 69, 144, 156

low-income regions, 93, 100, 109,114, 258

Lucas, R., 9, 234–5

Malawi, 20, 123, 130, 237–9, 243–4,247, 250–3, 258–9

Malawian migrants in and returnedfrom South Africa, 31

Marginality, 13, 18, 40, 44, 45,172

marketscapital, 167deregulation of, 45‘free’, 66forces, 84, 100‘neutral’ markets, critique of, 66regulation of, 45, 123, 130, 176see also distribution systems,

market-based andnon-market-based

Marshall, T. H., 26, 43–4material exchange, 7media, 24, 144–6, 148, 201Mexico, 105, 114–15, 235

PROGRESA, 80middle-income regions, 100migrant(s)

children, 217, see also childrencircular, 114, 118classification (classifying migrants),

5–8communities, 8, 12entitlements, 30, 117, 132, 152experience, 7, 256female migrant workers, 77, 97, 240,

257, 267groups, 113‘illegalization’ of, 39‘illegal’, 25informal, 98, 112insecurities, 5, 10–14international, 8, 12, 32, 37, 54, 91,

93, 98–100, 103, 109, 114, 184,186, 258, 262

irregular, 7, 10, 23, 39, 45, 49, 54,56, 64, 78, 96, 110, 119, 132,236, 238, 243, 245, 247, 272,276, 278

legality of categories of, 7

288 Index

migrant(s) – continuedlegal statuses of, 15lives, 8, 98networks, see networks, migrantnewly arrived, 91, 269population, 4, 8, 31, 146, 176, 178,

199, 201receiving countries, 3, 23–5, 27,

56, 91–2, 96, 103, 109–10, 119,212, 219, 224, 228, 233, 235,241

returned/returning, 11, 130seasonal, 23, 83sending countries, 105, 110statuses, 15–18undocumented, 22, 30, 44–5, 56–7,

92–3, 95, 98, 107, 109, 111,113, 117, 128, 143–60, 243,258

workers, 6, 26, 29–30, 37, 45, 50–1,54–7, 69, 75, 92, 95–8, 100–1,110–12, 114, 117, 120, 125–9,130, 150, 160, 164–9, 171,173–6, 178–80, 202, 243, 262–4,266, 272, 275, 277, 279–81

young migrant workers, 173see also vulnerability/vulnerabilities

(of migrants); disadvantage,migrant-intensified

Migrants’ Rights Network, 7, 25, 57migration

controlling, 4, 27cross-border, 119decisions, 20, 93, 281forms and processes of, 5gender dimension of, see gender

relations, dimension ofmigration

‘illegal’, 25, 45internal, 4, 23international labour, 92, 118‘irregular’, 7, 39labour, 6, 25, 77–8, 92, 101,

118, 129, 134, 173, 238, 259,262

migratory flows, 3, 5–10North-North, 9, 105, 109, 232,

258South-North, 9, 232–4, 258

South-South, 9–10, 31–2, 111,232–7, 259

migratory journeys, 7, 31, 110policy/policies, 4–5, 8, 22–8, 95,

113–15, 215, 241regulating, 4, 24–6repeat, 10, 235, 237, 255,

258–9return, 93, 234studies of, 3, 7voluntary, 5–6

minimum social floor, 112minority groups, 6, 26, 46, 57multilateral arrangements, see

agreements and arrangementsmutual aid and support, 71‘myth of return’, the, 6, 216

national courts, 38national insurance, 7, 44, 64, 75,

236nationality

co-nationals, 227dual, 43, 48see also transnational

nation state, see state(s)networks

access to, 71, 92, 94, 96, 113community-based, 22family, 220, 251formal, 21, 94, 113, 170, 225,

264–5informal, 21, 94, 113, 170, 225,

264–5kinship, 85local, 113migrant, 8, 13, 71, 112–13,

233, 237patronage, 265social, 3, 10, 16, 31, 70–1, 77, 91,

96, 112, 196, 218, 237, 239,251, 254–60

transnational, 8voluntary, 158

non-citizens, see citizenship,non-citizens

non-governmental organisations(NGOs), 14, 54, 68–9, 112, 196,216, 223

Index 289

OECD (Organisation for EconomicCo-operation and Development),104–5

Olivier, M., 29, 117–34, 184origin countries, 21, 91, 93–5, 108,

264see also migrant(s), sending

countries

passports, 143, 244, 263, 270, 272,281

patronage, 8, 22, 203, 265pensions, see benefits and allowances,

old age pensionsPICUM (Platform for International

Cooperation on UndocumentedMigrants), 145

police, 39, 168–70, 173, 175, 179, 181,200, 203, 238, 244, 276

policies and instrumentsbilateral, 109international, 120, 122–3, 126, 128multilateral, 129national, 28–9, 52, 120, 125–6supranational, 28, 126

politicalconstituencies, 24, 84disadvantage, see disadvantage,

politicalelites, 280the politics of migration policies, 5,

22–8portability

definition of, 21–4see also social protection and

provisionpoverty, 4, 6, 10–13, 23, 26, 32, 54, 65,

69, 80, 82–3, 85, 103, 110, 127,153, 155, 186–7, 192, 203, 235,237, 241, 245–9, 259, 262, 268

poverty reduction, 4, 11, 23, 110pre-saving element, 21, 94public goods, 14, 205public sector, 78public service, 24, 44, 49, 81, 84, 96,

144, 149–50, 160, 192

racism, see discriminationRatha, D., 9–10, 32, 232–3, 235–6, 259

ration cards, 24, 84, 187–8, 192–5,197–203, 205–6

see also India, Public DistributionSystem (PDS)

Rawls, J., 62recession, see economic, recessionrefugees

international conventions on, 37status, 27–8, 37, 50–1, 53, 121, 123,

132, 152regional

coordination, 29integration, 127–8

remittances, 12, 32, 48, 110, 113, 119,134, 211, 213, 217–24, 228

representation, 12, 43, 77, 185–6,201–2, 217, 219

representative advocates, 70, 153residence

legal meanings of, 7permanent residency status, 74status, 14, 38, 50–1, 120temporary residence, 6, 121, 205see also citizenship, presence,

residence andresidents associations, 23, 70–1, 252

see also tenants associationsretirement, 21, 94, 99, 111, 132rights

-based approaches, 30, 117based on presence, residence or

citizenship, 39, see alsocitizenship, rights

civil, 43economic, 15, 51, 53, 124human rights law, 38, 57, 112international human rights law, 38,

57of labour migrants, 37legal, 39, 63migrants’, 5, 7, 11, 25, 36, 56–7, 107political, 15, 36, 38, 43–4, 49, 53,

125‘primary’, 43the right to social security, 124, 126‘second generation’, 43social, 4, 26, 28–9, 31–2, 43–6, 49,

54, 92–3, 111–12, 124, 127,210–28

290 Index

rights – continuedviolations of, 38, 51voting, 37–8, 50, 211–12welfare, 6, 18, 27, 37workers’, 51, 54, 93, 101, 243, 262

riskinsurance, 4, 71reduction, 6, 19, 233–5

Roma (minority group), 6, 36, 41–3,154

rules and regulations, enforcement of,108, 264

rural areas, 7, 78, 172, 238

Sabates-Wheeler, R., 3–32, 47, 50,61–85, 91–115, 180, 183–5, 194,206, 214, 223, 232–60, 264

safety nets (social), 7, 15, 22, 91, 94,265

savings, 65, 70–1, 125, 213, 249, 254,276, 278–9, 281

clubs, 70Scandinavia, 38Schaffer, B., 67–8, 70, 79–82, 183–4,

195, 198, 200school (enrolment in), 7, 13, 47, 63,

80, 96–7, 113, 154, 195, 217–18,224, 236, 247, 279

security, 4, 21, 91–2, 94, 99–100,102–3, 105, 107, 109–10, 120,123, 127, 236, 258, 264

see also social securitySen, A., 62–3, 80–1, 183service industry, 92settled populations, 3sickness, 14, 18, 99, 219, 251social benefits, 4, 24, 75, 84, 93,

95–6, 98social contract, 19, 69social exclusion, 13, 25, 26–8, 44, 54

see also discriminationsocial inclusion, 17, 26social inequality, 3, 26social institutions, 4, 66, 171social justice, 20social networks, see networks, socialsocial policy, 17, 19, 214social practices, 6–7, 211–12

social protection and provisionaccess to formal social protection,

see accesscharity-based, 20, 29cross-border, 20, 236, 238definition of, 10formal, 13, 21–2, 30–1, 93, 103,

108–12, 183–4, 186, 235, 237,250–4, 256, 258, 260, 264–5

four functions of, 20frameworks, 20for migrants, 5, 14–23, 25, 28–9,

61–85, 62, 103, 111, 114, 250–4policies and instruments, 11, 15, 19,

23, 173, 176, 180rights, 112, 131sources (market, non-market,

charity, etc), 61state, 63–4systems, 10, see also welfare, systemstransformative, 20transnational, 210–28, see also

transnationalsocial rights, see rights, socialsocial security

mainstream, 4rights, 21, 94, 103, 109, 120, 127,

258, 264systems, 91–2, 94, 99–100, 102, 105,

107, 110, 123, 236social spending, 24, 84, 96, 215social support, alternative strategies

forhome-based affiliations, 18religious affiliations, 18

social welfare, see welfaresolidarity, 18, 70, 168–9

see also class solidaritiesSouth Africa

Apartheid, 36mines, 119, 237–8

South-south migration, see migration,migratory flows

state(s)agencies and institutions, 7‘kleptocratic’, 18laws, 61, 98, 111, 120nation, 14–15, 26–7, 43, 46, 52–4,

63, 226

Index 291

services, 7sovereignty, 24, 84, 96-specific policies and practices, 5see also welfare, state

stigma, see discrimination, stigmastresses and shocks, 15structural adjustment policies, 14, 18students, 7, 246supranational organizations, 24surveillance, 25, 149, 169, 171, 179

targeting, 65, 67, 69, 216see also access, structures, eligibility

rules/criteriatax

contributions, 93payers, 19system, 7

Taylor-Gooby, P., 15telecommunications, 6temporary visitors, 39tenants associations, 189

see also residents associationstourists, 7, 25, 39, 148–9trade unions, 66, 77, 164, 167–8,

170–1, 174–6, 179–80trafficking, 25, 49, 56, 119

see also human traffickingtransformative measures, see social

protection and provision,transformative

transnationalcommunities, 8, 211, 220institutions, 43networks, see networks,

transnationaltransport, 6, 25, 155, 170, 186, 215travel agencies, 3

UKEnglish Poor Laws, 26‘health tourism’, 147, 160Migrants’ Rights Network (MRN), 7,

25, 57National Health Service (NHS), 143,

147unemployment, see employmentUN International Convention on the

Protection of the Rights of All

Migrant Workers and Members oftheir Families 1990, 55, 95

Universal Declaration of HumanRights, 39, 52, 125

see also rights, human rights lawuniversalism, 30, 143unregistered labour recruitment, 3urban areas, 238, 265urban informal settlements, 10, 12USA

California, 57, 75Children’s Medicaid, 74CHIP, 74, 167

visa requirements, 92, 236voluntary networks, see networks,

migrantvulnerability/vulnerabilities (of

migrants), 10–14

wage earners, 12, 54wages (low), 7, 54, 162–3, 164Washington Consensus, 213, 226Weber, M., 80welfare

arrangements, 214as a means of controlling

migration, 4for migrants, four components of,

21, 93provision, 4–5, 10, 15–18, 21, 27,

31, 61, 64–5, 78as a tool for exclusion, 17as a tool for inclusion, 17

regimes, 16, 19, 26–7services, 10, 38, 40, 96as a site of political contestation, 17systems, 7, 15–18, 20, 27, 38, 96,

108, 110, 176Wood, G., 16, 67, 78, 81–2, 159work

Decent Work agenda, 112domestic, 92, 245, 253, 262environment, 12opportunities, 24, 266, 281permits, 45, 75, 95, 113, 132, 234,

238–9, 266, 281seasonal jobs, 114work-for-pay, 15

292 Index

workersforeign, 17, 111, 122, 126, 264, 279‘guest’, 25, 44low-skilled, 93, 146, 244semi-skilled, 119, 238, 241

working class struggle, 18world population, 9World War Two, 47, 52, 127

post-World War Two, 52, 127WTO (World Trade Organization), 47

Migration and Citizenship: Rights and Exclusions (in Social Protection and Migration, 2011)

Documents

Transcript of Migration and Citizenship: Rights and Exclusions (in Social Protection and Migration, 2011)